University of Southern California Dissertations and Theses

Radium isotopes in San Pedro Bay, California: Constraint on inputs and use of nearshore distribution to compute horizontal eddy diffusion rates

(USC Thesis Other)

Radium isotopes in San Pedro Bay, California: Constraint on inputs and use of nearshore distribution to compute horizontal eddy diffusion rates

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content RA ISOTOPES IN SAN PEDRO BAY, CA: CONSTRAINT ON INPUTS AND USE OF NEARSHORE DISTRIBUTION TO COMPUTE HORIZONTAL EDDY DIFFUSION RATES By Steven Laurence Colbert A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (GEOLOGICAL SCIENCES) December 2004 Copyright 2004 Steven Laurence Colbert Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3155396 Copyright 2004 by Colbert, Steven Laurence All rights reserved. INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. ® UMI UMI Microform 3155396 Copyright 2005 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A CKNOWLEDGEMENTS This project was funded by USC Sea Grant Program, part of the National Sea Grant College Program, National Oceanic and Atmospheric Administration, U.S. Department of Commerce, under grants NA86RG0054 and NA16RG2256. In addition, this project received considerable support from the Wrigley Institute for Environmental Studies. Thanks to Captain Gerry Smith for the long days on the water and help in the lab. Thanks to Cody Walker, Carole Belamo, and Shelley Howard for help in the lab. Thanks to all the other field assistants: Erica, Kathy, Valerica, Brenda, Reni, Dewey, Geoff, Timur, and Altay. Thanks to Will Berelson for the continued moral support. Thanks to Heather for her love and understanding. Thanks to my committee members: Richard Ku, Donn Gorsline, Jed Fuhrman, and Dennis Williams. Your support helped me complete this dissertation, and your insights made this dissertation worth reading. I am deeply indebted to my advisor, Doug Hammond, who saw my potential many years ago. With Doug’s nurturing and leading by example, I have become a better scientist and a better person. Thank You. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS PAGE ACKNOWLEDGEMENTS........................................................................ ii LIST OF TABLES....................................................................................... v LIST OF FIGURES...................................................................................... vii ABSTRACT................................................................................................. x CHAPTER 1. INTRODUCTION................................................................ 1 General Statement............................................................................ 2 Regional Oceanography................................................................... 9 Santa Monica Bay and San Pedro Bay Geography........................ 10 Fresh Water Inputs............................................................................ 11 General Local Circulation................................................................... 24 Currents and Transport in Santa Monica Bay.................................... 27 Transport in San Pedro Bay.............................................................. 33 Summary of Circulation.................................................................. 35 Radium Geochemistry...................................................................... 35 Radium in the Hydrosphere.............................................................. 38 CHAPTER 2. A RADIUM BUDGET FOR SAN PEDRO BAY, CA 45 Abstract............................................................................................. 45 Introduction...................................................................................... 46 Study Site......................................................................................... 50 Methods........................................................................................... 53 Results.............................................................................................. 59 Adsorption............................................................................ 59 Emanation Rate.................................................................... 61 Production Rate.................................................................... 61 Equilibrium Pore Water Concentration............................... 65 Seafloor Inputs..................................................................... 68 Shoreline Inputs.................................................................... 84 Tidal Exchange at Estuarine................................................ 94 Surface Water Inventory...................................................... 96 Ra-228 Budget...................................................................... 104 Box Model........................................................................................... 105 Conclusions...................................................................................... 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3. CALCULATING CROSS-SHELF MIXING RATES USING SHORT-LIVED RA ISOTOPES................................................................ 118 Abstract............................................................................................. 118 Introduction...................................................................................... 119 Study Site......................................................................................... 122 Results.............................................................................................. 123 Shoreline............................................................................... 123 Offshore............................................................................... 134 2-D Mixing Model.......................................................................... 141 Model Results................................................................................... 148 Application to 228Ra........................................................................ 169 Discussion......................................................................................... 170 Conclusions....................................................................................... 181 CHAPTER 4. IDENTIFYING COASTAL SOURCES OF 2 2 6 RA IN SAN PEDRO BAY, CA....................................................................................................... 184 Abstract............................................................................................. 184 Introduction...................................................................................... 184 Study Site......................................................................................... 186 Methods........................................................................................... 189 Results.............................................................................................. 196 Discussion......................................................................................... 196 Conclusions...................................................................................... 206 CHAPTER 5. CONCLUSIONS.................................................................... 208 REFERENCES.............................................................................................. 216 APPENDIX A. METHODS............................................................................230 Mn Fiber Preparation...........................................................................230 The RaDeCC System.......................................................................... 230 RaDeCC Method................................................................................. 258 Radium Extraction from Mn Fibers................................................ 259 Gamma Method................................................................................... 261 226Ra Method..................................................................................... 262 Field Methods- The Radiator............................................................ 264 Shoreline Sampling Protocol.............................................................. 269 APPENDIX B. FORTRAN SOURCE CODE FOR 2-D BAY MODEL... 271 APPENDIX C. DATA TABLES.................................................................. 298 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES TABLE PAGE 2-1. Weight percent of sediments finer than 180 pim , Ra distribution coefficient (Kd ), Rn and Ra emanation rate, and the Rn and Ra predicted equilibrium pore water concentration measured in the laboratory..................................................... 60 2-2. Shoreline sand A) bulk sediment isotope production rates and B) equilibrium pore water equilibrium concentrations........................................................... 64 2-3. Huntington Beach A) shoreline and B) offshore Pore water measurements for 2 2 2 Rn, 2 2 3 Ra, and 2 2 4 Ra............................................................................... 66 2-4. Radium concentrations in nearshore, near bottom water samples.... 69 2-5. Pore water model input parameters and results................................ 77 2-6. Summary of Ra flux from wave and tide pumping through the beach 87 2-7. A) Locations, beach surface, and water table heights at each well measured relative to high water mark near well C. B) Individual water table measurements for each well on 8/28/03................................................................................ 88 2-8. A) Summary of samples collected near the mouth of each marsh. B) Summary of Ra flux from tidal exchange with rivers and marshes 95 2-9. Short-lived Ra isotope concentrations from A) sites > 20 km offshore in Borderland basins, B) shoreline and mixed layer off of Huntington Beach, and C) Sunset Beach............................................................................................ 97 2-10. San Pedro Bay box model summary................................................. 101 2-11. Estimate of inputs to the region south of San Pedro Bay................ 108 2-12. Computed fraction of water from Sunset Beach that is advected down the coast to Huntington Beach....................................................................... 112 3-1. Model derived values for average shoreline concentration and scale distance, along with the computed inventory, are presented........................ 135 v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3-2. Model parameters used in the two-dimensional finite difference model................................................................................................................ 142 4 1 2 2 6 Ra samples collected in Santa Monica Bay and San Pedro Bay: A) offshore samples, B) shoreline samples, and C) estuary samples 190 4-2. Groundwater 2 2 6 Ra concentrations........................................................192 A-l. Summary of drying fibers experiment............................................... 243 A-2. Efficiency of 2 2 4 Ra as a function of the fraction of air in RaDeCC .. 246 A-3. Half-lives and alpha particle energies for Ra, Rn, and Po................ 247 A-4. Efficiency of 2 2 4 Ra as a function of the flow rate............................. 250 A-5. RaDeCC Channel 1 2 2 4 Ra system efficiency computed for samples processed on both RaDeCC and by gamma spectroscopy............................. 251 A-6. Isotope efficiency in each channel..................................................... 255 A-7. Sensitivity of counting efficiency to changes in cartridge orientation and repacking...................................................................................................257 A-8. List of parts to construct the Radiator..................................................266 A-9. Stripping efficiency as a function of weight of fiber...........................268 A-10. Packing list for shoreline sampling.................................................... 270 C-l. Volume of water pumped at each station......................................... 298 C-2. Temperature and salinity data from each station............................... 306 C-3. Temperature and salinity at beach stations....................................... 316 C-4. Average station location, distance offshore, and water depth 317 C-5. Mixed layer thickness and surface temperature............................... 319 C-6. Tides for 3 days around each sampling date........................................ 321 C - l . Tide summary.................................................................................... 330 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LTST OF FIGURES FIGURE PAGE 1-1. General circulation in the Southern California Bight...................... 3 1-2. Physiographic map of Los Angeles and Orange County, CA.............12 1-3. General stratigraphic column for the Los Angeles Basin................... 17 1-4. North-south cross section of Talbert Gap along Bushard St...............21 1-5. Simplified geologic map of Santa Monica Bay and San Pedro Bay..23 1-6. Uranium and thorium decay series isotopes...................................... 37 2-1. Map of San Pedro Bay, CA............................................................... 49 2-2. Results of adsorption experiment.........................................................58 2-3. Pore water concentration composite profiles at 7.5 m and 8.3 m depth offshore of Huntington Beach..............................................................70 2-4. Schematic diagram of flow through sediments produced when waves and currents interact with seafloor ripples.................................................... 71 2-5. 2 2 4 Ra:2 2 3 Ra flux normalized to the emanation rate ratio as a function of the irrigation rate....................................................................................... 83 2-6. Schematic cross-section diagram of flow through the beach 86 2-7. Well hydrographs at Huntington Beach compared to the tide on 9/11/03........................................................................................................... 89 2-8. Cross section of wells at Huntington Beach shoreline........................ 90 2-9. Composite surface water data collected offshore from Huntington Beach.............................................................................................................. 102 2-10. Composite surface water data collected offshore from Sunset Beach...............................................................................................................103 2-11. Average summer circulation in San Pedro Bay, CA........................... 113 2-12. Surface water temperature measurements from September 2001... 114 3-1. Shoreline concentration of 2 2 3 Ra and 2 2 4 Ra at San Pedro Bay............ 124 V ll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3-2. Shoreline temperature as a function of time..................................... 129 3-3. Nearshore distribution of 2 2 3 Ra and 2 2 4 Ra......................................... 130 3-4. Short-lived Ra isotope inventory in surface water offshore of Huntington Beach on each sample date........................................................ 137 3-5. Correlations between inventory, scale distance, and shoreline concentration for 2 2 3 Ra and 2 2 4 Ra..................................................................... 138 3-6. Summer short-lived Ra isotope inventory as a function of various environmental conditions............................................................................. 139 3-7. Model bathymetry within the first 2 km offshore................................ 143 3-8. Comparison of 2-D model and 1-D boundary model with a constant mixed layer thickness of 10 m 1-D analytical solution.................................. 149 3-9. Longshore Ra distribution for 1-D analytical solution and 2-D model with constant mixed layer thickness of 10 m and no seafloor topography... 150 3-10. Longshore Ra distribution for 1-D analytical solution and 2-D model with constant mixed layer thickness of 10 m and no seafloor topography... 151 3-11. Comparison of offshore distribution of Ra between 2-D model with seafloor topography and 1-D boundary model............................................... 152 3-12. Composite summertime Ra profiles at Huntington Beach with model best-fit results for three advection rates.......................................................... 154 3-13. Chi squared contour plots at three different advection rates as a function of mixing rate (Kh ) and Sunset Beach inventory fraction...............................155 3-14. Best fit to data given various input fluxes at a longshore advection rate of 5 cm s'1 ........................................................................................................ 157 3-15. Model results for various Ra inventories at the upstream boundary as a fraction of the inventory at Sunset Beach............................................ 3-16. Best fit to data given various distribution of seafloor inputs........... 3-17. Model sensitivity to tidally-driven changes in the shoreline flux.... 3-18. Model results for various mixing rates............................................ 3-19. Shoreline concentration as a function of time after a change in the mixing rate...................................................................................................... viii 158 160 161 162 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3-20. Model response to a change in the eddy diffusivity........................... 166 3-21. Model sensitivity to changes in Ra distribution at the upstream boundary....................................................................................................... 167 3-22. Composite summertime 2 2 8 Ra at Huntington Beach with model best-fit results for three advection rates..................................................... 168 3-23. Model results for various advection rates........................................... 173 3-24. Model generated profiles for scale-dependant Kh ............................ 176 3-25. Composite summertime 228Ra at Huntington Beach with model best- fit results for constant mixing rate and a scale-dependant mixing rate 177 3-26. Diffusivity as a function of distance from a source......................... 180 4-1. All 2 2 6 Ra data as a function of distance offshore.............................. 193 4-2. 2 2 6 Ra concentration collected offshore from Huntington Beach as a function of depth.............................................................................................. 194 4-3. Groundwater 226Ra concentration as a function of salinity 195 4-4. Estimate of 2 2 6 Ra concentration at the shoreline derived from a 2D advection-diffusion model............................................................................ 204 A -l. Flowmeter reading as a function of the helium flow rate................. 232 A-2. Daughter ingrowth vs. time for various 224Ra:228Th ratios.............237 A-3. Daughter ingrowth vs. time for various 228Ra:228Th ratios.............238 A-4. Count rate of 220Rn and 219Rn as a function of voltage............... 245 A-5. Count rate of 220Rn and 219Rn from samples counted on both channels.......................................................................................................... 253 A-6. Schematic diagram of the radium extraction system- the Radiator.. 265 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT The short-lived radium isotopes Ra-223 (11 days half life) and Ra-224 (3.5 day half- life) are potentially useful for evaluating cross-shelf dispersion rates in the coastal ocean. A requirement for this application is that their source function and its variability in time and space must be defined. The primary mechanisms for introducing radium into coastal surface waters include: (1) wave and tide-driven circulation of water through permeable beach sands, (2) input from the seafloor due to molecular diffusion and circulation of bottom water through surficial sands, (3) flow of water rich in Ra from marshes and estuaries, and (4) net advection of groundwater. The importance of these inputs to San Pedro Bay, CA, was determined from concentrations in waters collected from each of these potential sources. In most of the region, mechanism (2) supplies 90 percent of the input, although mechanisms (1) and (2) may become dominant locally as the coastal morphology varies in the longshore direction. Longshore variations in the composition of beach sand and the presence of persistent coastal eddies create longshore gradients in Ra concentration that are significant in this region. Temporal variations in shoreline concentrations on time scales 6-8 hours reflect variations in mechanism (1) as tides rise and fall, with drainage of water from the beach face creating higher concentrations during the falling tide. Despite these complications in characterizing the source function, the distribution of short-lived Ra isotopes is useful in constraining the rate of horizontal mixing. A two-dimensional advection-diffusion x Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. model was best fit with an eddy diffusivity of 1.3±0.2 m2 s'1 over length scales of several km offshore, with a value about 50% smaller in the littoral zone. The scale dependence of eddy diffusivity is also apparent in the distribution of Ra-228, which requires lower eddy diffusivities in the nearshore than in the offshore region. A budget for Ra-226 indicates that little groundwater directly enters the ocean in this region, although some groundwater may enter marshes and estuaries that are adjacent to the coast. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 1. Introduction The goal of this dissertation is to quantify the sources of Ra isotopes to the coastal ocean, identify mechanisms that control the distribution of Ra isotopes in the coastal ocean, and use the distribution of Ra isotopes to constrain the rates of processes occurring in the coastal ocean. Chapter 1 is the introduction, and contains the general background on the study site, San Pedro Bay, CA, along with a brief introduction to Ra geochemistry. In the second chapter, the input fluxes of 2 2 3 Ra, 2 2 4 Ra, and 2 2 8 Ra to coastal surface water are measured from the shoreline, benthic inputs, and estuaries. In addition, the inventory of Ra in surface water at Sunset Beach and Huntington Beach is estimated. Using this information, a mass-balance budget for 2 2 3 Ra and 2 2 4 Ra is constructed. Based on this budget, some general aspects of circulation in San Pedro Bay can be assessed. In the third chapter, a two- dimensional model of San Pedro Bay surface water is developed and used to determine the horizontal eddy diffusivity required to generate the observed distribution of 2 2 3 Ra and 2 2 4 Ra, given the inputs predicted in Chapter 2. In addition, the model is used to explore the temporal variability in the 2 2 3 Ra and 2 2 4 Ra distribution at various time scales. In Chapter 4, the 2 2 6 Ra concentration of groundwater, estuaries, shoreline, and nearshore samples is presented, and the source of high 2 2 6 Ra concentrations is identified. In the final chapter, the results of this dissertation are summarized. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. General Statement: Quantifying nearshore cross shelf mixing rates is essential to develop predictive solute transport models needed to improve beach and coastal ocean management. Most current ocean circulation models stop at the 10 m isobath, ignoring inner shelf and surf zone processes. In this region, rivers, tidal marshes, storm drains, and shallow groundwater aquifers discharge, supplying the coastal ocean with nutrients and contaminants. With accurate models, the ability of the ocean to dilute domestic and industrial wastes can be assessed for water quality control. As population grows globally, our need to manage the coastal oceans as a resource is also growing. Radium, a naturally-occurring element with four radioactive isotopes, has previously been used to measure rates of dispersion on different scales. These include vertical dispersion in the deep ocean (Sarmiento et al., 1976), ocean basin-scale dispersion (Cochran, 1992; Hammond et al., 1990; Huh and Ku, 1998; Kaufman, 1973; Knauss et al., 1978; Moore, 1987; Sarmiento et al., 1982), and cross-shelf dispersion on the continental shelf (Moore, 2000). In this dissertation, a budget for Ra isotopes in San Pedro Bay is constructed, and their nearshore distribution is used to calculate the cross-shelf rate of dispersion, as well as the temporal variability of dispersion. The study sites for this project are the two bays that lie west of the Los Angeles megalopolis: Santa Monica Bay and San Pedro Bay (Figure 1-1). These bays are near the middle of the Southern California Bight (SCB), which extends from Point Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1-1. General circulation in the Southern California Bight. Modified from Hickey (1992). 121°W 35°N 120° 119° 118° 117° 34° 33° 32° C onception B b Point - D um e I os A ngeles Anacapa S a n ta ^ Monica' Santa fv San\<g Pedro Ba Santa Barbara Santa Catalina San Nicolas San Clem ente 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Conception to the US- Mexico border. Measuring the movement of currents, waves, and sediments has been the focus of hundreds of studies ranging in complexity from individual theses to large, multi-agency studies. These studies have covered a broad range of issues, including concerns over water quality, mineral resources, and the needs of the military, industry and shipping. Some of these studies are included in the summary presented below. Before addressing the specifics of circulation in each bay, the general features of circulation in the coastal ocean are reviewed. The coastal ocean extends from the shoreline to the shelf break. Tides, winds, and regional- to basin-scale circulation all influence circulation on the shelf. Unlike the open ocean, coastal ocean circulation is confined by two boundaries, the seafloor and the shoreline. These no-flow boundaries influence circulation in several ways. First, the shoreline blocks flow perpendicular to the coast and is a fundamental constraint that any flow must adjust. Second, seafloor frictional dissipation is enhanced in the shallow nearshore water column. Third, changes in water column depth result in vortex stretching in association with cross-isobath flow. Fourth, the seasonal occurrence of strong stratification in shallow water leads to dramatic cross-shore variability (upwelling, downwelling, and coastal jets) over length scales of the local baroclinic deformation radius (r), which is about 25 km for southern California (r=c/f; f=Coriolis parameter, c= baroclinic gravity-wave phase speed~2.1, or the barotropic c=(gH)0 ,5 where, g is gravity, and H is water depth). Finally, runoff and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. preferential diurnal heating nearshore dramatically alters the hydrographic structure of the coastal ocean and results in thermohaline flow features. The coastal ocean can be divided into 4 regions based on the water column depth: littoral zone, inner shelf, middle shelf, and outer shelf or shelf break. Coastal ocean circulation, whether driven by local winds or by regional circulation, is influenced by the characteristics of each of these regions. Relative to the open ocean, dynamical variability is pronounced over small distances on the continental shelf. Given this variability, sub-mesoscale (10 km) to mesoscale (100 km) physical processes are expected to play important roles in the coastal circulation. The littoral zone extends less than 1 km from the shoreline. In this region currents are wave dominated, and mixing is rapid. The littoral zone morphology can be characterized based on Dean's dimensionless fall velocity (Q) (Dean, 1973): Where Hb is breaker height, T is the wave period and Ws is the mean particle fall velocity. Beaches can then be categorized as reflective (Q <1), intermediate (1< Q <6), or dispersive (Q >6) (Wright and Short, 1984). In most of Santa Monica Bay and San Pedro Bay, beaches are intermediate (e.g. for Huntington Beach, Q =4.5). Beaches at the south end of Santa Monica Bay, south of King Harbor are more reflective. At intermediate beaches, a bar-trough system generally forms. From the 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. beach, water deepens into a trough, shallows at a sand bar, and deepens beyond the bar. Waves approaching the beach first break at the sand bar, dissipating their energy throughout the surf zone. Waves breaking over the bar carry water into the trough. Rip currents form at breaks in the bar, where water flows out of the trough. During periods with strong winds and sea waves, exchange between the surf zone and offshore waters is more diffusive (Inman et al., 1968). Water expelled from the littoral zone may be carried back into the surf zone with the next set of waves or mixed away from the coast, depending in part on the cross-shelf transport rate of inner shelf waters. Onshore winds and the flood tide keep water near shore, increasing the residence time in the littoral zone, while offshore winds and the ebb tide preferentially transport water away from the shoreline. Longshore transport in the littoral zone is driven by the momentum flux of shoaling waves approaching the shoreline at an angle to the shoreline. Waves rarely approach parallel to the shoreline, and in Southern California, the swell often has both a NW and a SW component (Emery, 1959). Wave-generated longshore currents may also produce a significant current outside the littoral zone, which are superimpose upon the wind-driven, thermohaline, and baroclinic flows in the coastal ocean. Longshore currents may also be opposite to currents outside of the littoral zone, creating shear at the littoral zone- inner shelf boundary (Lentz et al., 2003). The inner shelf extends from the littoral zone to about 10-20 m depth. In this region, the water column remains well mixed. Horizontal density contrasts may arise due to 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. intensive nearshore heating of the shallow water column. On a coast that faces west, thermal expansion would produce a baroclinic longshore current in the poleward direction, which is maximum at the surface and zero at the bottom (Pettigrew and Murray, 1986). The net transport of bottom material is enhanced typically during strong wind events and large swell events, the inner-shelf and surf zone overlap, and the water column is well mixed (Wright 1995). Local wind stress can overwhelm baroclinic forces in the inner shelf, and eventually a force balance between wind stress and dissipation is expected. These wind-driven currents are characterized by the strong influence of bottom friction and a reduced effect of Coriolis acceleration, which causes surface currents to flow parallel to the underwater topographical variations or the surface wind stress (Swift and Niedoroda, 1985). In the SCB, winds from the west or north-west contribute to an equatorward nearshore current. Under steady conditions, a coastal boundary jet, with current velocities that are greatest at the shoreline and decrease offshore, may form (Csanady, 1977). However, the diurnal nature of winds in Santa Monica and San Pedro Bay most likely prohibit the development of a stable coastal boundary jet. The mid-shelf extends from the inner shelf to the shelf break. Deeper waters of the mid-shelf are often in geostrophic balance, with a balance between lateral pressure gradients and the Coriolis force, but bottom and wind stress remain important (Lentz et al., 1999). This region is also influenced by the open-ocean circulation. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Meandering boundary currents that follow the continental slope can impact mid-shelf circulation both at the surface and at depth (Gawarkiewicz et al., 1996). The shelf break is characterized by a dramatic increase in bottom slope, and represents the end of the continental shelf. This is a dynamic region, dominated by exchange between the open ocean and shelf waters (Biscaye et al., 1994). At the seafloor of Santa Monica and San Pedro Bays, coarse-grained sediments and rocky outcrops at the shelf break are evidence of high energy bottom currents that scour the seafloor. Eddy diffusion, or dispersion, is the transport of a solute produced by the random motion of eddies of varying size and intensity. At low mean current velocities, such as cross shelf advection rates, eddy diffusion can dominate transport. Many different solutes have been used to quantify eddy diffusion. The most common is the use of dyes (e.g. (Okubo, 1971). A packet of dye is deployed and then tracked during the time it is measurable, generally on the order of hours. Concentrations are either measured on water samples, or remote sensing can be used to map the dye. The change in distribution of dye with time can be modeled to calculate diffusivity. Alternatively, naturally occurring, radioactive solutes provide another tool for measuring these processes. For example, 2 2 2 Rn, produced by the decay of 2 2 6 Ra in sediments, is a radioactive noble gas with a 3.8 day half-life. In the deep ocean, the vertical distribution of 2 2 2 Rn is a function of the rate of radioactive decay, which is 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. known, and the rate of transport from the seafloor, which can then be calculated (e.g. Sarmiento et al., 1976). Similarly, the radium isotopes have been used individually or in pairs to explore vertical and horizontal eddy diffusion, which is further explored in this dissertation. As the tracer spreads, its distribution is influenced by larger scale processes. This causes the tracer to spread more quickly, resulting in an increase in eddy diffusivity. The scale dependence of eddy diffusion is most apparent for horizontal transport and ranges over several orders of magnitude (Okubo, 1971). The relationship is between scale length and eddy diffusivity follows approximately a 4/3 power law (Okubo, 1971; Stommel, 1949). However, the absolute value of the relationship varies because the apparent diffusivity and scale of diffusion are arbitrarily defined (Okubo, 1976). With this background on the coastal ocean structure and the processes that influence circulation, we can take a closer look at the currents and mixing in Santa Monica Bay and San Pedro Bay. Regional Oceanography The California Current (CC) is a broad, slowly meandering, cold current found up to 1000 km offshore from Oregon to Baja California. South of Point Conception, the shoreline of the SCB cuts eastward, while the CC remains far offshore (Figure 1-1). Near the US-Mexico border, the CC turns east and bifurcates. This produces the 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. northward flowing Southern California Countercurrent, which follows the continental shelf. Beneath the Southern California Countercurrent, the poleward- flowing California Undercurrent is concentrated, distinguished as a subsurface maxima in the current speed (Hickey, 1993). In the winter, during periods of low stratification, these currents can merge (Miller et al., 1999) These currents reach a seasonal speed maximum in late summer (Hickey, 1993). As a result, warmer water moves up the coast during the summer, while cooler water, along with branches of the CC, flow into the SCB in the winter (Lynn and Simpson, 1987). Decadal and longer time scale current fluctuations in the SCB occur in association with El Nino events (Chelton and Davis, 1982) and basin-wide or global events (i.e. Pacific Interdecadal Oscillation, North Atlantic Oscillation) (Casey and Adamec, 2002). Santa Monica Bay and San Pedro Bay Geography Santa Monica Bay and San Pedro Bay are located in the middle of the SCB. Santa Monica Bay is bounded by the Transverse Ranges to the north and the Palos Verdes Peninsula in the south. San Pedro Bay stretches from the south end of Palos Verdes Peninsula to Newport Beach, where the Newport submarine canyon impinges on the shoreline. Uplift along the Palos Verdes fault near the center of these bays has produced a broad continental shelf, relative to the rest of California (Nardin and Henyey, 1978). Both bays are similar in areal extent (Santa Monica Bay = 650 km2 vs. San Pedro Bay = 775 km2 ), and have similar shoreline orientations, with south- and west- facing beaches (Figure 1-1). The shorelines of these two bays are the 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. western and southern border of the Los Angeles basin, a extensively urbanized region that has retained few of its natural features. Fresh Water Inputs Surface Water: The Southern California climate is Mediterranean, characterized by warm, dry summers and mild, wet winters. On average, 36 cm of rain fall each year, with higher amounts falling in the local mountains. The average rainfall is misleading because of high inter-annual variability. Generally, there are a couple rainy years, often associated with an El Nino event, followed by several years of drought. These major storm events are most important for transferring large volumes of sediment into the SCB (Brownlie and Taylor, 1981; Schwalbach and Gorsline, 1985). The Los Angeles Basin is a flood plain for the three major rivers that drain the surrounding mountains: Los Angeles River, San Gabriel River, and Santa Ana River (Figure 1-2). In the past, these rivers have changed their courses across the LA Basin. For example, before 1815, the Los Angeles River turned westward from downtown LA and followed the course of present day Ballona Creek, and the Santa Ana River outlet was in the vicinity of Seal Beach, 18 km northwest of the present river mouth (Stevenson, 1954). Both rivers had major changes in their course after the floods of 1824-1825. After the floods, the Los Angeles River drained south into San Pedro Bay and the Santa Ana River moved to approximately its present position, 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 1-2. Physiographic map of Los Angeles and Orange County, California. Gray regions are rock outcrops. Dotted lines represent the location of injection well fields. Prado Df.n t o I Matfou Santa Monica Los Angeles Co. Orange Co. nta Monica Bay Central W e str \ # Coast'-j 1 Basin Domingup^ LA/LB Harbor Sunset Beach San Pedro Huntington Bea Bay Newport but with the outlet close to the present day Newport Beach Pier (Sherman, 1931). A still greater flood in 1861 moved the Santa Ana River to its current location. Since the Pleistocene, at least four remnant channels of the Santa Ana River have formed: Sunset Gap, Bolsa Gap, Newport Bay, and Talbert Gap, where it currently flows (CalDWR, 1967). Today, these three rivers are confined by cement channels designed to prevent flooding from 100-year storms. These channeled rivers reduce the risk of flooding by rapidly transporting rain water out to sea. However, installing these channels has destroyed the natural estuaries and coastal marshes, and the swift-flowing water prevents fish and animals from living in the river. In addition, upstream dams divert the water for agriculture and groundwater recharge before reaching the ocean. These dams and diversions also prevent sediments from reaching the shoreline. Assessing the decrease in sand delivery to the shoreline is difficult for several reasons. First, there is little information on sediment concentrations before the dams were installed. Second, the bulk of the sediments are delivered during episodic floods with recurrence intervals of at least several years. During the summer, most of the base flow in these rivers is captured behind dams at the east end of the Los Angeles Basin and used to recharge groundwater. The lower reaches of these rivers then acquire dry weather runoff from excess irrigation, car washing, etc., as well as treated domestic and industrial sewage. A low salinity river plume can be tracked as it travels south and remains close to shore (B. Jones, pers. 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. comm.). A major source of freshwater to the Los Angeles coastal ocean is effluent from waste water treatment. The largest three producers of effluent are Los Angeles City's Hyperion outfall, which discharges into Santa Monica Bay (15 m3 s'1 ) ; Los Angeles County's Whites Point Outfall off of Palos Verdes (14 m3 s'1 ); and the Orange County Sanitation District's outfall offshore from Huntington Beach (11 m3 s' ’). Each of these outfalls is located several km offshore in water at least 60 m deep. The treated water is generally still rich in organics, and its density is high enough that the plume remains below the mixed layer. For example, the plume from the Orange County Sanitation District's plume is generally found between 12 and 14°C (Noble et al., 2003). Plume water is then transported with local currents and diluted with open ocean water. Groundwater: Groundwater is another potential freshwater source to the coastal ocean. Fortunately, the geohydrology of the Los Angeles Basin has been well characterized. All deposits, except the most recent, were formed as either submarine fan deposits or alluvial deposits, both originating in the Santa Ana Mountains and Santa Monica Mountains. This depositional environment creates a grading of sediments from coarse-grained material that is deposited closest to the source and finer-grained material at the furthest reaches of the deposits. Whether the sediment being deposited is gravel that grades to silt or sand to clay depends on the energy of the water moving the material. In general, the coarser material is transported during times of extensive uplift, lower sea levels, and/or times of increased precipitation, 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. while the finer material is deposited during more tectonically quiet times, higher sea levels, and/or times of lower rainfall. The eastern part of the basin is known as the Forebay, where coarser material was deposited and the aquifers can be characterized as unconfined or semi-confined to depths of 600 m. This area is characterized by a stratigraphic sequence of relatively coarse-grained deposits of sand and gravel with occasional lenses of clay and silt (Herndon, 1992). While the clays and silts generally do not impede groundwater flow from one aquifer to another, the lack of continuous aquitards in the Forebay makes aquifer delineation and correlation extremely difficult (Herndon, 1992). Extending across the central and coastal portions of the basin is the Pressure area, generally defined as the area in the basin where surface water and shallow groundwater are prevented from percolating in large quantities into the major production aquifers by clay and silt layers at shallow depths (Herndon, 1992). Aquifers in this area are characterized as being confined, but drops in water table elevations can make these aquifers act unconfined because they are not completely filled. Comparisons made between hydrographs provide evidence that there are some areas in the Pressure area where deeper aquifers can exchange with surface waters (CalDWR, 1967). The current conceptual model of the basin is based on studies by the California Department of Water Resources in the mid 1960's that described the existence of 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. three major aquifer systems- the Upper, Middle, and Lower aquifer systems (Herndon, 1992). These aquifers are found in sediments ranging from late Pliocene to Recent in age and in a range of depositional environments (Figure 1-3). The Lower Aquifer is found in the late Pliocene Pico Formation. Since this aquifer system is below the realm of general groundwater production, most of the information on these deep aquifers are from oil well loggings. The Lower aquifer contains the largest aquifers in the Orange County ground water basin, 350 to 500 feet thick (CalDWR, 1967), but their rate of recharge is unknown. Uplift rates slowed at the end of the Pliocene, resulting in the deposition of an extensive low permeability cap that marks the boundary between the Lower and Middle aquifer systems (CalDWR, 1967). The Middle aquifer system consists of sediments deposited during the lower Pleistocene, the lower San Pedro Formation. The top of the Middle aquifer system, named the Silverado in Los Angeles County and the Main in Orange County, consists of a sequence of coarse sand and gravel, with interbedded lenses of silt and clay and is currently the primary production aquifer in the basin (Callison, 1992). Deeper aquifers in the Middle system are probably hydraulically connected, but this has not been extensively mapped. The Upper aquifer system was deposited during the Upper Pleistocene to Holocene and represents the transition from inner neritic to non-marine deposition (Blake, 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1-3. General stratigraphic column for the Los Angeles Basin. Aquifers found in each formation are listed along with their location. In general, aquifers from the same formation are believed to be hydraulically connected. Formation Age Recent/ Aluvium Quaternary Aquifer (Location) Semi-perched Aquifers Talbert (Talbert Gap) Bolsa (Bolsa Chica Gap) Recent (Alamitos Gap) Gaspur (Ballona and Dominguez Gaps) Pleistocene Palos Verdes/ San Pedro Sands Lakewood Fm. Upper San Pedro Sunnyside/ Lower San Pedro Pliocene Upper Pico Lower Pico Repetto Local Importance Alpha, Beta, Lambda (Talbert Gap) 200 ft. Sand (West Coast Basin) Omicrom, Rho (Talbert Gap) C, B, A, I Aquifers(Alamitos Gap) Lynwood Aquifer (Orange Co. Basin) 400 Ft. Gravel (West Coast Basin) Main (Orange County Basin) (A, I Aquifers?) (Alamitos Gap) Silverado (LA County Basin) Lower Aquifer Zone (Currently not tapped) ~Base of Fresh W ater- Miocene Puente El Modeno Volcanics Topanga Sespe Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1991). Currently lying between 300 and 2000 feet below land surface are Pleistocene deposits of unconsolidated sands and gravels, including the marine San Pedro and Lakewood formations, the terrestrial La Habra formation, and to a lesser extent the Coyote Hills formations and Palos Verdes sands (Herndon, 1992). The Upper San Pedro and Lakewood deposits in the Orange County groundwater basin are correlated with the 400 Foot Gravel and 200 Foot Sand of Los Angeles County, respectively (Callison, 1992). (These units are named not for their thickness, but for their average depth below land surface.) Variations in sedimentation patterns has created a complex aquifer system in the eastern and central parts of the basin, but near the coast, a more regular series of sands and gravels divided by silts and clays has created a more delineated system. Based on water chemistry, there is possible circulation between the Main and the Upper aquifer system near the center of the basin (CalDWR, 1967). The final aquifer forming event occurred during the Wisconsin glacial stage (15-60 ka), when a continued uplift on the margins, a drop in sea level, and a wetter climate allowed erosion to occur across the Los Angeles Basin. At the center of the basin, however, deposition appears to have continued uninterrupted (CalDWR, 1967). As sea level rose during the early Holocene, the incised channels filled with sediments. Sediments deposited during this time were fluvial deposits of the Los Angeles River, San Gabriel River, and the Santa Ana River, consisting of mostly coarse sands and gravels. 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. With the end of the glacial epoch came a drier climate and a rise in sea level. These coarse grained, shallow aquifers were overlain by finer grained sediments and organic matter accumulating in lagoons and marshes that developed as the sea approached current levels. The coarse-grained sediments were left behind in the channels that had been incised during glacial stages, becoming the Gaspur aquifer in Los Angeles County and the Alamitos, Bolsa, and Talbert aquifers in Orange County. These shallow aquifers were originally the primary source of groundwater across the Los Angeles Basin, but because of salt water intrusion due to over pumping, most current wells are drilled into the deeper Main and Silverado aquifers. At the aquifer seafloor outcrops, red, iron oxide stained sands are found (McCurdy, 1964; Moore, 1951; Stevenson et al., 1956). Across this relatively low energy, marshy environment, floods would periodically deposit stringers of coarse debris along southwesterly flow alignments, forming perched water tables. Approximately 50 to 70 feet of sediments including fluvial, marsh, lake, and overbank deposits accumulated, along with organic sediments and peaty alluvia (Moran and Wiebe, 1992). Semi-perched aquifers at the surface overlie much of the central and coastal portions of the basin (Herndon, 1992). In Los Angeles County, the delineation between shallow coarse- and fine-grained sediments was widespread enough that the Department of Water Resources divided similar Recent-aged deposits in that area into two units, the Semi-perched aquifer and the underlying Bellflower aquiclude (Herndon, 1992). Groundwater flow in semi 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perched aquifers throughout the basin tends to be site-specific and often is controlled by localized surficial recharge sources, such as over-irrigated areas, or unlined drainage channels that intersect the shallow water table (Herndon, 1992). Because the Semi-perched aquifer has been contaminated from underground storage tanks, increased total dissolved solids and nitrates from historical fertilizer application, as well as by various industrial chemicals, it is of minimal use from a water supply standpoint (Herndon, 1992). Cleanup of industrial pollutants within the this aquifer is considered a necessary effort since no conclusive evidence has been published about the potential for eventual migration of these contaminants from the Semi perched aquifer to underlying aquifers (Herndon, 1992). There is one major fault that slices longitudinally through the Los Angeles Basin: the Newport-Inglewood fault zone (NIFZ). In Orange County, uplift along the NIFZ at the coast has caused older aquifers to generally dip inland towards the northeast and then emerge at shallower depths along the eastern side of the basin syncline (Anaheim area). Deep aquifers of lower Pleistocene age that would ordinarily be in direct hydraulic conductivity with the ocean have been effectively sealed by either being offset against impermeable sediments, or the fault's damage zone has created an impermeable barrier (Figure 1-4). More recent deposits have either been isolated from the ocean in the uplifted pressure ridges or have been truncated by the ancestral rivers, allowing for recharge of these deeper aquifers from the recent aquifers at the south-western edge of the basin (CalDWR, 1967). These rivers carved six distinct 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1-4. North-south cross section of Talbert Gap along Bushard St. Injection wells are located along Ellis Ave. Major cross streets are shown at the top. (Modified from Orange County Water District Water Reclamation and Seawater Intrusion Barrier Project Engineering Report, 1990.) Pacific Coast Highway Garfield Ave. Yorktown Ave. Adams Ave. Atlanta Ave. S ea- Level PACIFIC OCEAN TALBERT AQUIFER valleys, or gaps, across the coastal plain: Ballona Gap in Santa Monica Bay, and Dominguez Gap, Alamitos Gap, Bolsa Chica Gap, Sunset Gap, and Talbert Gap in San Pedro Bay, where inland aquifers are in direct contact with the ocean and can feed deeper aquifers. In Los Angeles County, however, the NIFZ moves inland and forms a low permeability zone that retards groundwater exchange between the Central Basin from the West Coast Basin. Along the south end of Santa Monica Bay, the Middle and Upper Aquifer systems are in contact with the coastal ocean. Until about 1947, the Los Angeles Basin appeared to be self-regulating, with inputs from rainfall matching outputs. After this, water tables began to drop as pumping began to exceed recharge. To increase the rate of recharge, the first recharge basin in the Forebay was opened in 1949. To prevent seawater intrusion into the aquifers at the coast, four series of injection wells have been established (Figure 1-5). At the southern end of Santa Monica Bay is the West Coast Barrier. The other three are located across the major gaps around San Pedro Bay: Dominguez Gap, Alamitos Gap, and Talbert Gap. The water injected into these wells today is generally a mixture of imported water, deep well water, and treated wastewater and seawater that has undergone reverse osmosis. As a result of local heterogeneities and difficulties in designing injection wells, these barriers have helped reduce the rate of seawater intrusion. They are not 100% effective, and some intrusion around and through the barriers does occur. 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1-5. Simplified geologic map of Santa Monica Bay and San Pedro Bay. Dotted lines are location of injection well fields. Circles represent regions where aquifers outcrop at the seafloor. 1) Canyons draining Santa Monica Mountains, 2) West arm of the Gaspur Aquifer, 3) 200 ft. Sand, Silverado, and Pico Formation, 4) South arm of the Gaspur Aquifer, 5) Recent Aquifer, and 6) Talbert Aquifer. Santa Monica Fault ▲ Dominguez [farrier Alamitos Barrier x> Talbert .Barrier General Local Circulation Since fresh water inputs to the coastal ocean are minimal during the summer and episodic during the winter, most of the water column structure is defined by the thermal structure. During the winter, there are two water masses on the shelf. The surface mixed layer is about 14°C and extends to depths of 20 to 30 m. Below this is colder, 11°C slope water. Between these two is a thermocline of variable thickness. During the summer, a shallow, warm mixed layer, from 5-15 m thick and 18- 21°C, develops. Reduced winds during the summer and an increase in solar radiation help maintain this profile. However, the mixed layer development often occurs more rapidly than can be explained by solar radiation, implying the importance of the advection of warm water masses onto the shelf (Noble et al., 2003). While the primary direction of wind and swell on the west coast of the United States are from the north and west, Santa Monica Bay and San Pedro Bay are naturally sheltered from these directions. Both bays are protected to the north by the shape of the coastline and the Transverse Ranges. Santa Monica Bay is also protected from the south by the Palos Verdes peninsula. To the southwest, Santa Catalina Island provides shelter for both bays. The southern swell, produced by storms in the South Pacific, is strongest in the late summer. San Pedro Bay is less protected, resulting in a higher energy environment. As a result, Santa Monica Shelf sediments are finer and better sorted than those on the San Pedro Shelf (Orys-Brink, 1983). Waves approach the Bays through corridors between the offshore Channel Islands from the 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. west and south. The Channel Islands and associated banks tend to attenuate long period (15-20 sec) swell from the west, leaving short (7-10 sec) period waves to pass relatively free of interference. Long period waves propagating from the south reach the shelf without interruption (Karl et al., 1980). Tides in Southern California are mixed, with two high tides and two low tides of unequal height. Mean tides vary with location and coastal configuration, but are approximately 1.1m; spring tides are about 2.2 m. Tidal currents have strong oscillatory motions that result in little net movement over a tidal cycle (Noble et al., 2003). Across much of the shelf, the major tidal currents flow parallel to the shoreline, with a poleward flow occurring during flood tides and equatorward flow during ebb tides. Tidal currents on the inner shelf may be influenced by the shelf width, with stronger tidal currents on when the shelf is narrow (Hickey, 1993). Submarine canyons in Santa Monica Bay and San Pedro Bay are too small to have an effect on regional circulation. Current measurements over Redondo Canyon at a depth of 30 m suggest that the flow passes directly over the canyon with no apparent modification (Hickey, 1992). SCCWRP measurements at a mooring within Santa Monica Canyon and San Gabriel Canyon showed no significant correlation between canyon and shelf flows (Hendricks, 1980; Hendricks and Stubbs, 1984). However, the canyons may influence local circulation. Newport Submarine Canyon at the southern terminus of San Pedro Bay extends to within 0.5 km of the shoreline. 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Under almost all conditions, the presence of the canyon influences surface waves, causing them to fan out at the canyon head (Emery, 1960). This produces north flowing longshore currents to the north of the canyon and south flowing currents south of the canyon (Pipkin, 1985). These divergent currents at the head of the canyon lead to sediment accumulation at the shoreline. However, about 1 km north of the canyon, the north flowing limb intersects a southward current traveling down San Pedro Bay, producing an offshore flow (Pipkin, 1985). In the past, these conditions have produced severe erosion at the shoreline. An effective solution to prevent erosion was the installation of a series of groins. Redondo Canyon and Newport Canyon represent the southern terminus of the Santa Monica Bay and San Pedro Bay littoral cells (Emery, 1960; Inman and Bush, 1975; Inman and Frautschy, 1966). In general, there is a flux of sand along the shoreline from north to south. However, Redondo Canyon, Palos Verdes and the Southern California Counter Current prevent any exchange of beach sediments between these two bays (Handin, 1951; Rice et al., 1976). However, DDT-contaminated sediments, which originate offshore from White's Point near the south end of Palos Verdes, have migrated northward into Santa Monica Bay (Zeng et al., 2000). With a widespread distribution of DDT-contaminated sediments in Santa Monica Bay, Redondo Canyon does not appear to be effective at removing these sediments from the shelf. Since the late 19th century, there have been significant man-made modifications to the San Pedro Bay. Local navigation hazards have been removed and many 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. structures have been built in these bays, including breakwalls, oil drilling platforms, piers and jetties. The greatest change was the installation of the Los Angeles-Long Beach Harbor at the north end of San Pedro Bay. Here, a breakwall several miles long was installed, and several small islands were removed to improve navigation. These modifications have made the flow of commerce easier, but significantly altered the original ecosystem. For example, In the mid-1890's the net longshore transport near the San Gabriel River was from the northwest to the southeast, as indicated by observed erosion of what is present day West Beach and deposition on East Beach (Magnusen, 1995). Presently, the net longshore transport has reversed, with poleward flow resulting in an average recession of East Beach between 0.5 and 0.9 m/yr (Everts, 1993). This thesis is focused on San Pedro Bay, but a few samples were collected in Santa Monica Bay as a pilot program (See Appendix). Consequently, a discussion of Santa Monica Bay circulation is pertinent. Currents and Transport in Santa Monica Bay Geostrophic currents form as a balance between gravitational and Coriolis forces and require several hours to reach steady-state. The characteristic feature of a geostrophic current is that it flows parallel to isotherms, with cold water to the left of the direction of flow in the northern hemisphere. These currents are slow to develop, taking several hours to reach steady-state. Conversely, their presence in the coastal 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ocean implies stability on this time scale. To test this, a few drogue deployments were made along with mapping of the temperature structure of Santa Monica Bay to test if thermal slopes did accurately represent the direction of flow (Stevenson et al., 1956). There was a positive correlation between the two, lending support to the presence of geostrophic currents. Both surface and mid-depth drogues were deployed. Both shallow and deep drogues traveled in approximately the same direction during most of the deployments. However, they usually had different velocities, with the deeper drogue moving slower than the shallow drogue. During 1955-56, the temperature structure of Santa Monica Bay was extensively mapped (Stevenson et al., 1956). In general, shelf waters displayed a summer-winter thermal pattern. During the winter, November through May, the water 5 km or more from shore was warmer than inshore. The reverse was true in the summer months. Exceptions to this general distribution did occur. The location of the major portion of warm and cold water determines the basic direction of water motion that is consequently established. This is especially true within three miles of shore where temperature slopes are better developed than farther offshore. Thus, there is a general southerly flow in the winter and a northerly flow in the summer (Stevenson et al., 1956). In all months, there are parts of the bay with horizontal isotherms, which generally occur > 5 km offshore. In areas where the surface layer exhibits no horizontal gradients, wind is the dominant force contributing to water motion. Prevailing 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. westerly winds would move this water towards shore with a slight southerly component. Thus, surface water in the outer portion of the bay can move shoreward to the southwest until it reaches a point 3 to 5 km from the coast where it would be carried either to the north or south prior to reaching the beach (Stevenson et al., 1956). During this study, two persistent cold water units were identified: one along the Malibu coast and another adjacent to the Palos Verdes Hills (Stevenson et al., 1956). At Point Dume in Malibu, The shape of the shoreline and direction of winds transport water offshore, which leads to the upwelling of colder water. This is also a region of formation for small (<10 km) eddies (DiGiacomo and Holt, 2001). The cold unit in the southern part of the bay is usually most intense along the western shore of the Palos Verdes Hills. Frequently, it is close to shore with a steep thermal gradient separating it from the offshore water. At other times, the unit spreads into the bay and along the coast as far north as Redondo Beach (Stevenson et al., 1956). The existence of cold water may be from the more or less constant flow of water out of the bay to the south. Varying conditions within the bay, causing varying current flows to the south, would result in considerable variation in the cold water unit (Stevenson et al., 1956). Current motion based on thermal distribution in the coastal ocean can give the direction, but not the velocity, of flow. The boundaries of these currents are difficult to identify since energy dissipates by spinning off eddies at the boarders. One useful 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. way to get at this information is to use Lagrangian drifters or drogues. Seaward of the shelf, drifter experiments have shown that the movement of water is both spatially and temporally variable. Some drifters barely moved at all over several days, whereas others nearby moved large distances over the same period (Hickey, 1992). During 1955-56, over 5000 drift cards were released in Santa Monica Bay, of which 34% were recovered (Stevenson et al., 1956). Drift cards float just below the water surface and are expected to track the movement of the upper meter of water, which is strongly influenced by the wind. Flow within the bay was found to be variable. During the autumn and winter, the bay appeared to be well flushed by currents directed either to the north or to the south. During the spring and summer, the predominant drift was toward the shores of the bay, but frequently superimposed on this was a moderate southerly or northerly component. The maximum velocities were usually found in the southern inshore portion of the bay. During these periods, water flowed in near the center of the bay and diverged a few miles offshore. To balance the hydrologic budget for the bay, water must leave the bay either at the northern and southern limits or along the seafloor (i.e. downwelling). A flow out of the bay in the southeastern half of the bay has been documented with drifters, direct current measurements, shipboard surveys, and satellite-derived sea surface temperature maps (Hickey, 1992). A third method is to deploy current meters, which observe the flow characteristics (i.e. velocity) in the vicinity of a given point. This Eulerian or flow method can 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. provide high-resolution data on the changes in ocean currents at a specific location and water depth. An accurate analysis of the areal and temporal variability of currents in a region requires many deployments spanning all seasons during several years. However, such a thorough analysis is yet to be done for Santa Monica Bay. Instead, much work has focused on the currents surrounding Santa Monica Canyon in the north-central part of the bay and Redondo Canyon in the south. From these studies we can gain valuable insight into current behavior within Santa Monica Bay. In general, the Southern California Counter Current is present as a poleward flow through the San Pedro Channel (Hickey, 1992). Fluctuations over the outer shelf tend to be aligned with the slope axis and vary more slowly than the diurnal tidal periods (Hendricks, 1980; Hickey, 1992). The large amplitude, relatively long period (20-30d) fluctuations are strongly coherent over the mainland shelf edge and slope (Hickey, 1992). During the fall/winter period, a significant amount of variance is related to local wind forcing (Hickey, 1992). Over the mid-shelf and inner shelf, mean currents are equatorward at a velocity between 2 and 5 cm s'1 ; median currents at 41 m water depth were between 9-11 cm s'1 (Hendricks, 1980; Hickey, 1992). On the inner shelf, current variability was found to occur over several different time scales. Two important periods for fluctuations in both the longshore and cross-shore directions were semi-diurnal and diurnal (Hendricks, 1980). The semi-diurnal and diurnal fluctuations are the result of internal waves of approximately semi-diurnal periodicity (Hendricks, 1980). 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Flow over the inner Santa Monica shelf also has significant subtidal variability. The velocity fluctuations over the inner shelf do not appear to be strongly related to those over the shelf edge, nor to those over the continental slope. Also, the dominant time scales are markedly less than those over the shelf edge and basin slope (5-10 d vs. 20-30 d) (Hickey, 1992). Despite the similar time scales between these time scales and the wind field along the US west coast, there is only a weak correlation between the two (Hickey, 1992). To further investigate driving mechanisms for the inner shelf flow, estimates of terms in the along-shelf momentum equation were compared with the observed forcing mechanisms. These results suggest that the inner shelf along-isobath circulation is primarily driven by pressure gradient forcing of cross-isobath flow over the shelf. Most frequently the relationship is such that the geostrophic, poleward flowing Southern California Countercurrent has an associated sea level gradient perpendicular to the current with a sea level maximum on the nearshore side of the current. In the San Pedro Channel, the sea level maximum extends to the shoreline against the Palos Verdes peninsula. As the current flows north, this boundary disappears in Santa Monica Bay, producing a sea level topographic high that drives cross-shelf flow and an equatorward flow on the inner shelf (Hickey, 1992). 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Transport in San Pedro Bay There has been significant work done in the Huntington Beach region associated with the Orange County Waste Management District's waste treatment effluent that discharges offshore. During summer 2001, a study using current meters, dye studies, and shoreline sampling at Huntington Beach was conducted (Noble et al., 2003). The results of this study are an excellent summary of the local oceanography. Some of their results along with the work of other projects is summarized below. During the summer 2001, there was a substantial down-coast mean current over the San Pedro shelf. Surface currents increased from 5 cm s ’ at 10 m water depth to 10 cm s'1 at the shelf break (Noble et al., 2003). At depths below about 70 m, an undercurrent flow was predominantly up-coast, that occasionally rose to depths of 30 m for periods of a few days (Hendricks, 1994; Noble et al., 2003). Similar results have been observed just to the south (Winant and Bratkovich, 1981). During the winter, the poleward undercurrent may extend to the surface (Hickey, 1993). The surface mean current is generally unrelated to local winds over the shelf (Noble et al., 2003). However, there is some indication that large-scale fluctuations in shelf currents may be generated by coastal-trapped waves propagating from Baja California (Hickey, 1992; Noble et al., 2003). Fluctuations with periods of 7 to 20 days are observed in the outer shelf mean flow, with a magnitude similar to the mean flow (Noble et al., 2003). Similarly, coastal 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sea level in the SCB fluctuates with the same period (Hickey, 1992). These fluctuations occur on the time scale for weather systems passing through the region (Halliwell and Allen, 1987). However, coastal sea level fluctuations in the Bight in this weather band (7-15d) are not strongly related to those north of the Bight (Hickey, 1992). On the inner shelf, similar to Santa Monica Bay, shorter period current fluctuations dominate (Noble et al., 2003). In this region, winds show a strong diurnal pattern, generated by differential heating of the land that produce strong onshore winds in the afternoon. Afternoon winds force surface waters directly and drive an onshore flow that piles up water at the shoreline (Noble et al., 2003). As the winds calm in the evening, nearshore waters relax, driving an offshore flow. This vertical flow generates diurnal shoreline water temperature fluctuations of up to three degrees (Noble et al., 2003). This conveyor belt process transports water from near the base of the mixed layer to the surface and back. Sustained local winds from the south-west may also generate upwelling at Point Fermin, at the northern end of San Pedro Bay (Karl et al., 1980). Significant diurnal cross shelf transport events have been observed at Huntington Beach. For example, isotherm displacement of up to 4 km have been observed (Noble et al., 2003). The onshore excursion only lasts briefly, and is then followed by an equally strong offshore transport. These oscillations have a diurnal period and can repeat for several days (Noble et al., 2003). The occurrence of these oscillations 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. depends on the proper phasing of a combination of subtidal, diurnal sea breeze and tidal movements. Summary of Circulation Some conclusions can be drawn from the published studies on the current dynamics of the Southern California continental shelf. First, mean flow on the shelf tends to follow the direction of local, relatively shallow isobaths (Hickey, 1993)). Second, currents in the upper part of the water column (0-15 m) on the shelf appear to be predominantly equatorward. Since the axis of the eddy occurs at some distance offshore, southward flow can occur inshore of the northward flow, especially south of Palos Verdes from January to May (Crowe and Schwartzlose, 1972; Hickey, 1979; Sverdrup and Fleming, 1941). In all cases, however, the equatorward flow on both narrow and wide shelves (approximately 5-10 cm/s) is much less than that of the poleward flow over the slope (Hickey, 1993). Third, variability in flow is expected. Radium Geochemistry Radium was first identified in 1898 by Pierre and Marie Curie and Gustave Beaumont, and added to the periodic table in 1902. Almost immediately, medical uses for radium were identified. In oceanography, by the late 1930's, radium received considerable attention because of the unusually high content of pelagic sediments and because of the possibility of using the concentration changes with 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. depth in cores as a means of establishing the rates of deposition (Sverdrup et al., 1942). New applications for naturally-occurring radium in the hydrosphere continue to be identified. Radium has four naturally-occurring isotopes, with a broad range of half-lives: 2 2 3 Ra (11.4 days), 2 2 4 Ra (3.66 days), 2 2 6 Ra (1600 years), and 2 2 8 Ra (5.75 years). Each isotope is an intermediate in the uranium- thorium decay series (Figure 1-6). In a closed system at secular equilibrium, the Ra activity equals the activity of its parent. The activity ratio for 2 3 5 U:2 3 8 U: 2 3 2 Th in average crustal rocks is 1:20:30 (Faure, 1986). Thus, at equilibrium, the isotope ratio for 2 2 3 Ra:2 2 6 Ra:2 2 4 Ra:2 2 8 Ra is expected to be 1:20:30:30. Disequilibrium, or when the isotope ratio differs from the equilibrium value, indicates the preferential transport of one or more Ra isotope or their parent isotopes. For this project, naturally-occurring radium isotopes are measured in rivers and tidal marshes, at the shoreline, and in the coastal ocean. From a human health point of view, the Ra concentration in this environment is very low. Based on the EPA drinking water standards, the activity concentration of 2 2 6 Ra and 2 2 8 Ra must be less than 11.1 dpm L'1 . However, the surface water concentration of each is typically 0.1 dpm L'1 . 2 2 4 Ra and 2 2 3 Ra usually have the highest and lowest concentrations of any Ra isotope at 0.4 and 0.06 dpm L'1 . Including all isotopes, the total Ra activity concentration in water at the shoreline is 0.66 dpm L"1 . This is more than an order of magnitude below the EPA drinking water standard. [Note: instead of using the mass 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1-6. Uranium and thorium decay series isotopes. Figure modified from Ku (1976). E lem ent U -2 3 8 S eries T h -2 3 2 S eries U -2 3 5 S eries Neptunium Uranium U-238 4.5 x lO 9 y U-234 2 5 0 ky U-2 35 7 x 1 0 s y Protactinium P a - 2 3 4 ' 1 min i f 1 Pa-2 31 3 3 ky f Thorium Th-23 4 ' 2 4 day T T h-230 75 ky Th-23 2 1 .6x1 0 ’° y T h-228 2 yrs Th-23 1 25 hrs \ Th-22 7 1 9 d f Actinium \ A c -2 2 8 ' 6 hrs i 1 A c - 1 1 7 ' 11 y \ Radium R a-226 1 6 00 y Ra-228 5.8 y Ra-224 3.6 d Ra-223 1 1.4 d Francium \ \ \ Radon Rn-222 3.8 day Rn-220 1 min Rn-219 4 sec As tatine 1 Polonium Po-218 3 min Po-21 4 0.1 ms f Po-210 1 38 d Po-21 6 0.1 5 s Po-21 2 0.3 i j s f P o -2 1 5 2 ms Bismuth Bi-21 4 2 0 min -f I Bi-21 O' 5 d < 1 \ Bi-21 2 1 hr 1 \ Bi-21 1 2 min Lead Pb-214 27 min Pb-21 0 22 y Pb-206 Stable Pb-21 2 1 1 hrs \ P b-208 Stable 4 P b -2 11 ‘ 36 min Pb-2 07 Stable f Thallium TI-20 8 ' 3 min 7 TI-20 Y 5 min of radium to calculate concentrations, activity units of disintegrations per minute (dpm) have been used. The activity of radium (A) is the number of atoms (N) times the isotope's decay constant (A .): A= A .N . The decay constant is ln(2)/(half-life). The advantage to this nomenclature is that at equilibrium in a closed system, like the crustal rock described above, the activity of each U-Th decay series isotope, including the radium isotopes, is equal to the isotope activity of their parent. Radium in the Hydrosphere At the atomic level, there are three processes that release radium to the hydrosphere: mineral weathering, in situ production, and alpha recoil (Osmond and Ivanovich, 1992). Since weathering occurs at such a slow rate, it may be an important source for the longest-lived isotope, 2 2 6 Ra, but is inefficient for the release of the other isotopes (Hammond et al., 1988). In situ production is small in most environments, because only a small fraction of thorium, the parent isotope for Ra, is dissolved. Dissolved parent isotopes have been measured in the Southern California Bight. For 2 2 3 Ra, the grandparent, 2 2 7 Ac, one sample at 0.11 ± 0.02 dpm m'3 was collected in the SCB, and was at the low end of the average of 0.20 + 0.11 dpm m "3 for Southern and Central California stations (Kim, 1993). Concentrations of 2 2 7 Th (t1 / 2 = 19 d), the daughter of 2 2 7 Ac, should be even lower because it is particle reactive. For 2 2 6 Ra and 2 2 8 Ra, the dissolved concentration in surface waters at Santa Monica Basin of their parents, 2 3 0 Th and 2 3 2 Th, respectively, were 0.16 ±0.03 dpm m'3 and 0.049 ±0.001 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dpm m "3 (Huh and Beasley, 1987). Measurements of 2 2 8 Th in the Borderlands made in this dissertation were 0.6 ± 0.3 dpm m'3 . Alpha recoil is the most important mechanism for the release of Ra from solid phases. When thorium decays, it ejects an alpha particle (a positively-charged helium atom). To conserve momentum, the produced radium atom recoils in the opposite direction of the alpha particle and travels roughly 30 nm in minerals of normal rock density. Depending on the orientation of the recoil relative to the grain surface, the atom produced may be pushed deeper into the grain, pushed closer to the grain surface, or even ejected from the grain. The daughter atoms that can reach the pore space represent the mobile fraction. For each isotope, the ratio of the mobile fraction production rate to the total sediment activity is the emanation efficiency. There are two major geometric factors that influence the emanation efficiency: the grain size and the distribution of the parent isotope. The emanation efficiency increases as the grain size decreases, because of the increase in the probability that a parent atom is close to the surface. To assess the distribution of each parent isotope, the transition from igneous rock to sediments must be considered. Beginning with a metamorphic or igneous rock, U and Th isotopes are preferentially incorporated into various minerals, and within each mineral, ideally, the distribution of U and Th is homogeneous. Chemical weathering dissolves some minerals. Since Th is insoluble, when its host mineral dissolves, Th is not transported far before adsorbing onto another surface. Thus, the long-lived, 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. particle-reactive isotopes from each decay series, 2 3 0 Th, 2 3 2 Th and 2 3 1 Pa, are concentrated on the surface of sediments. For 2 3 2 Th, the total activity in the coating may be an integration of the mineral dissolution that has occurred proximate to that grain. In seawater, 2 2 8 Th may also accumulate on the surface of particulates falling through the water column or at the seafloor. This produces near-surface sediments that are enriched in 2 2 8 Th (Cochran, 1979; Torgersen et al., 1996). However, deeper sediments are depleted in 2 2 8 Th relative to 2 3 2 Th because of the mobility of 2 2 8 Ra. In fresh groundwater systems, 2 2 6 Ra may also concentrate at the mineral surface (Tanner, 1964). While Ra is more soluble than Th, it is particle-reactive. In pore water, the dissolved Ra concentration is buffered by adsorption-desorption. Sorption occurs rapidly, on a time scale of minutes (Krishnaswami et al., 1982). The distribution coefficient, Kd , defined as the ratio of adsorbed to dissolved radium (units: VOLUME MASS'1 ), primarily depends on two variables. First, as the ionic strength of a solution increases, there is greater competition for the sorption sites on minerals and Kd decreases. Thus, the Kd decreases when moving from fresh water to seawater environment. As a result, sediments transported into an estuary release adsorbed Ra (Elsinger and Moore, 1980; Key et al., 1985; Li and Chan, 1979; Moore and Scott, 1986). Desorption from suspended sediments occurs at relatively low salinities (<4 PSU) (Burnett et al., 1990). Desorption is also enhanced if dissolved Ra forms complexes (Langmuir and Riese, 1985). In fresh water, the concentration of 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. complexing anions is too low. In seawater, as much as 45% may be complexed withS04 2 '. In some environments, such as saline lakes and geothermal waters, complexes with OH', Cl", and C 03 2 ' may also be important (Zukin et al., 1987). Second, the adsorption coefficient is also a function of the sediment surface area. Fine grained material has a greater surface area per unit volume than coarse grained material, and therefore can adsorb more radium, increasing Kd . Manganese oxides, common in ocean sediments, have a particularly high surface area, and their presence can greatly increase Kd . As a result, Kd varies for different sediment types. For Ra in muds and silts, typical Kd values arelOO- 1000 cm3 g"1 in fresh water, 10- 100 cm3 g'1 in oxic seawater, and 2- 50 cm3 g"1 in anoxic seawater (Kadko, 1980; Krishnaswami et al., 1982; Sun and Torgersen, 2001). There are two significant sources of radium to the ocean: river inputs and benthic inputs (Cochran, 1992). As discussed above, an important component of the 2 2 6 Ra budget is its release from suspended sediments as salinity in estuaries increases (Elsinger and Moore, 1980; Key et al., 1985; Li and Chan, 1979; Moore and Scott, 1986). Desorption contributes a significant fraction of the Ra brought in by rivers. Benthic inputs are governed by transport of Ra from pore fluids. Besides diffusion, there are a variety of processes occurring in the coastal ocean that enhance the exchange of solutes and interstitial waters in sandy sediments. First, pressure gradients that drive advection through surficial sediments are produced by current- obstruction interactions and pumping as waves propagate over a flat bed (see review 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. by (Huettel and Webster, 2001). Macroscopic organisms living on the seafloor irrigate their burrows, increasing the rate of exchange (see review by Aller, 2001). Bioturbation may also increase the flux of particle- reactive solutes, such as Ra (Sun and Torgersen, 2001). At the shoreline waves and tides pump water through beach sands (McLachlan, 1979; McLachlan, 1982; McLachlan, 1989; Riedl, 1971). Tide and wave pumping extend beyond the beach, and are also an important process releasing radium in estuaries and tidal marshes (Bollinger and Moore, 1993; Rama and Moore, 1996). Finally, terrestrially-derived submarine groundwater discharge may also increase the Ra flux from seafloor sediment (Burnett et al., 1990; Moore, 1996). In the coarse grained sediments of shallow marine environments, the rapid flushing of seawater through near-surface sediments impacts the concentration of mobile Ra. For example, when the water residence time in sediments is less than a few days, essentially all of the mobile 2 2 6 Ra, 2 2 8 Ra and 2 2 7 Ac produced in the sediments should be flushed out. This has consequences for the production of their daughter isotopes, 2 2 2 Rn, 2 2 4 Ra and 2 2 3 Ra. Instead of an emanation efficiency controlled by a surface coating of a longer-lived parent isotope, the emanation of short-lived daughter isotopes is largely dependent on the total sediment activity of the parent. Therefore, the production rates of 2 2 2 Rn, 2 2 3 Ra, and 2 2 4 Ra may be closely related to the non- mobile activity of each parent isotope. 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In general, the Ra flux into the coastal ocean is enhanced by processes that increases the rate of exchange between pore water and the water column. But because the accumulation of each Ra isotope depends on its half life, pore water concentrations of each isotope respond differently to transport by different processes. For example, molecular diffusion maintains a concentration gradient in the sediments between the pore water equilibrium concentration at depth and a low Ra concentration in the overlying water column. The distance an isotope can travel by diffusion before decay occurs depends on its half life. Since 2 2 6 Ra and 2 2 8 Ra have the longest half lives, they have the best chance of diffusing out of the sediments before radioactive decay occurs. Short lived Ra does not have as much time to escape from the seafloor. For these isotopes, diffusion is a less efficient removal mechanism. Instead, the flux of short-lived isotopes is more sensitive to processes that enhance the rate of seawater exchange with pore water in near-surface sediments, such as irrigation and wave pumping. The application of Ra isotopes as a tracer of sediment-water exchange and ocean mixing is demonstrated in the many recent papers published (Hancock et al., 2000; Kelly and Moran, 2002; Moore, 2000; Rama and Moore, 1996; Torgersen et al., 1996). This study contributes to the radium literature in several ways. First, the temporal variability of radium at the shoreline and in the nearshore are measured. Second, the relative importance of three different sources: shoreline, seafloor, and marshes and estuaries, of the short-lived Ra isotopes to the coastal ocean are 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. quantified. Finally, the importance of these different sources to the 2 2 8 Ra budget are estimated. Building on the previous work of Moore (Moore, 1987; Moore, 2000), the distribution of Ra isotopes are used to quantify the rate and variability of cross shelf eddy diffusion on the inner shelf. These rates are then incorporated into a model of surface water circulation in San Pedro Bay. 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 2. A Radium Budget for San Pedro Bay, California Abstract Constraining cross-shelf dispersion rates is necessary to develop accurate models that may improve beach and coastal ocean management. One potential method to calculate these rates in nearshore surface waters is based on the distribution of the short-lived radium isotopes, 2 2 3 Ra and 2 2 4 Ra. To apply this method in San Pedro Bay, CA, the sources and sinks for each isotopes must be constrained. Sediments are the primary source of Ra in the ocean. The total sediment activity, emanation rate, and Ra pore water equilibrium concentration were measured on beach sands and found to decrease along the shoreline from north to south. Offshore, decreases in grain size cause both emanation rate and adsorption coefficient to increase, and the Ra pore water equilibrium concentration remains constant. 2 2 2 Rn, 2 2 3 Ra and 2 2 4 Ra pore water profiles were measured at 8 m water depth. A two layer model was fit to the data, indicating water exchange rate in the top 28 cm was 27 L irf2 d"1 . At the shoreline, waves and tides combine to drive pore water circulation. Based on a box model for water, 2 2 2 Rn, and Ra in the beach, the water residence time in the beach is 2.0 ± 0.2 d. The total flux of water through the beach was 13 ± 2 m3 d"1 (m shoreline)'1 . In the Huntington Beach region, seafloor inputs account for 90% of the Ra inputs to the mixed layer. Further south, lower pore water equilibrium concentrations and a steeper seafloor reduce the importance of this source to less than 20%. The remainder of the inputs come from flushing of estuaries. The Ra input flux into San 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pedro Bay and the region south, as well as the Ra inventories in water north of San Pedro Bay, were all greater than the observed inventory in surface water at Huntington Beach. These results require that onshore advection of low-Ra water accounts for between 65% and 90% of the nearshore water at Huntington Beach, with the remainder supplied by longshore advection. Introduction Beach and coastal ocean management require accurate, predictive models that track the fate of contaminants and nutrients from the shoreline to the open ocean. However, most ocean circulation models stop at the 10 m isobath, ignoring inner shelf and surf zone processes. In this region, rivers, tidal marshes, storm drains, and shallow groundwater aquifers discharge, supplying the coastal ocean with nutrients and contaminants. Nearshore circulation models are needed to accurately connect these sources with regional circulation models. To do this, the nearshore transport of water and solutes must be quantified. Because of slow mean cross-shore flow rates, dispersion plays an important role in transporting contaminants and nutrients from the shoreline to the middle shelf. One potential method to calculate the rate of dispersion in nearshore surface waters is based on the distribution of short-lived radium isotopes (2 2 3 Ra, t1 / 2 = 11.4 d; 2 2 4 Ra, t1 / 2 = 3.6 d) (Moore, 2000; Torgersen et al., 1996). However, before these models can be developed, the sources and sinks for short-lived Ra in the coastal ocean must be 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. constrained. The objective of this article is to develop a mass balance for the short lived Ra isotopes in San Pedro Bay. In the next chapter, which builds on the conclusions presented in this chapter, a model to compute nearshore dispersion rates is developed. Along with the two short-lived Ra isotopes, there are two long-lived isotopes, 2 2 6 Ra (t1 / 2 = 1600 y) and 2 2 8 Ra (t1 / 2 = 5.6 y). Each Ra isotope is produced by a thorium isotope parent, which is either found within the mineral structure or adsorbed to sediments. A portion of the Ra produced in sediments escapes into pore fluids and may be transported into the water column by a variety of mechanisms. These include molecular diffusion, waves and tides that pump water through beach sands (Emery and Foster, 1948; McLachlan, 1979; McLachlan, 1982; McLachlan, 1989; Riedl, 1971), circulation through near-surface sediments driven by pressure gradients that develop when currents interact with topographic obstructions (i.e. seafloor ripples) and waves propagating over a flat bed [see review by (Huettel and Webster, 2001)], macroscopic organisms irrigating their burrows [see review by (Aller, 2001)], and groundwater advection (Burnett et al., 1990; Moore, 1996). As a result, the shoreline and seafloor are major sources of Ra to the coastal ocean. River estuaries and tidal marshes are another important source of Ra to the coastal ocean (Cochran, 1992). Where rivers intersect aquifers, terrestrially-derived groundwater may contribute significantly to the dissolved 2 2 6 Ra in an estuary (Burnett et al., 1990). Plus, an additional source of 2 2 6 Ra is its release from 47 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. riverborne suspended sediment as the salinity increases (Elsinger and Moore, 1980; Li and Chan, 1979). Wave and tide pumping at the banks and interactions of currents with seafloor obstructions are also important mechanisms for releasing radium from estuary and tidal marsh sediments (Bollinger and Moore, 1993; Rama and Moore, 1996). For the short-lived Ra isotopes, the river bed and banks seem to be the primary source (Hancock and Murray, 1996). Produced by the decay of 2 2 6 Ra in sediments, 2 2 2 Rn (t1 / 2 = 3.8 day) has a similar source and half-life as the short-lived Ra isotopes. In addition, as a noble gas, 2 2 2 Rn behaves conservatively in sediments. 2 2 2 Rn has long been recognized as a useful tracer of sediment-water exchange (Broecker, 1965). In this paper, the sources and sinks of short-lived Ra in the surface mixed layer of San Pedro Bay, CA, are quantified. To begin, the production rate, emanation rate, adsorption coefficient, and pore water equilibrium concentrations for 2 2 2 Rn, 2 2 3 Ra, and 2 2 4 Ra are computed. Next, the Ra fluxes from three sources are constrained: seafloor, shoreline, and tidal exchange with marshes and rivers. Finally, a mass balance budget for San Pedro Bay is constructed. With this model, the importance of longshore transport of each isotope can be constrained. This must be done in order to assess what type of model (i.e. 1-D or 2-D) is appropriate to compute mixing rates. A model to compute cross-shelf mixing rates is presented in Chapter 3. 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Figure 2-1: Map of San Pedro Bay, CA. Points represent sample stations offshore from Huntington Beach; triangles represent sample stations collected offshore from Sunset Beach. Dotted line represents the boundaries of the box used for the mass balance calculation. Open squares are bouy stations: NOAA Outer Los Angeles Harbor Station 9410660 and Scripps San Pedro Buoy 092. km C A L I F O R N - 'I - / : ■ _ . 43 118 R ed o n d o B each > C Q 33.8 °N 1 P ort of L os A n g e le s/ ~ Long B each | 33 120 °W P alo s V erdes P en n in su la 20 m. O ) 33.65 __ 4.5 km 118.3 °W 118.1 117.9 Study Site San Pedro Bay is a relatively broad continental shelf located near the middle of the Southern California Bight (Figure 2-1). It is bounded to the north by Palos Verdes Peninsula and to the south by Newport Canyon. The climate is Mediterranean, characterized by wet mild winters and warm, dry summers. Three rivers discharge into the Bay: the Los Angeles, San Gabriel, and Santa Ana Rivers. Most of the river water is diverted upstream to recharge groundwater. During summer, below these diversions, these rivers have a minor flow composed of dry-weather runoff, treated domestic sewage, and industrial discharges. These three rivers are the primary source of sediment to the San Pedro littoral cell, which encompasses San Pedro Bay (Emery, 1960). Coastal sands are primarily arkose, with heavy minerals dominated by hornblende (Emery, 1960). The mean grain size of San Pedro Bay beach sands is 250 ^m, and changes slightly with environmental conditions (Bascom, 1951; Magnusen, 1995). The beach slope ranges between 1/12 and 1/20. Dean's (Dean, 1973) dimensionless fall velocity is about 4.5, which is characteristic of an intermediate beach. While beach sands are rather homogeneous, by 1 km offshore, mean grain size decreases from > 250 p im in the north part of the bay to < 60 f i m offshore from the Santa Ana River (Grant, 1973; Moore, 1951). This fine grained material probably originated from the Santa Ana River and the Orange County Sanitation District's sewage outfall and settled out under low energy conditions offshore (Emery, 1960; Grant, 1973). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since fresh water inputs to the coastal ocean are minimal during the summer and episodic during the winter, temperature defines most of the water column structure. During the winter, there are two water masses on the shelf. The surface mixed layer is about 13°C and extends to depths of 20 to 30 m. Below this is colder, 11°C slope water. Between these two is a thermocline of variable thickness. During the summer, a shallow, warm mixed layer, from 5-15 m thick and 18- 21°C, develops. Reduced winds during the summer and high solar radiation help maintain this profile. However the mixed layer development often occurs more rapidly than can be explained by solar radiation, implying the importance of the advection of warm water masses into the region (Noble et al., 2003). While the primary direction of wind and swell on the west coast of the United States are from the north and west, San Pedro Bay is naturally sheltered from these directions by the Palos Verdes peninsula and Santa Catalina Island. Waves approach the Bays through corridors between the offshore Channel Islands from the west and south. The southern swell, produced by storms in the South Pacific, is strongest in the late summer. Tides in Southern California are mixed, with two high tides and two low tides of unequal height. The mean tide varies with location and coastal configuration, but is approximately 1.1m; the spring tides exceed 2 m. Tidal currents propagate to the north and have strong oscillatory motions that result in little net movement over a tidal cycle (Karl et al., 1980; Noble et al., 2003). 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Slope currents are focused through the San Pedro Channel at the north end of San Pedro Bay. At depth is the northward flowing Southern California Countercurrent, which generally extends to the surface during the winter (Hickey, 1993). During the summer, surface slope currents generally flow towards the south, with a northward flowing undercurrent (Hickey, 1993; Noble et al., 2003). Superimposed on the outer shelf mean flow are current fluctuations with periods between 7 and 20 days (Hickey, 1992; Noble et al., 2003). On the shelf, currents generally follow the direction of local isobaths and are geostrophic in character (Hendricks, 1994; Hickey, 1993; Karl et al., 1980; Noble et al., 2003; Stevenson et al., 1956). During summer, shelf surface currents (< 15 m) generally flow towards the south (Hickey, 1993; Noble et al., 2003; Winant and Bratkovich, 1981). Across much of the shelf, the major tidal currents flow parallel to the shoreline. At longer time scales, current fluctuations with a period of 5-10 days dominate (Noble et al., 2003). While this period is generally correlated with weather systems, the surface mean shelf current is generally unrelated to local winds. Instead, there is some indication that large-scale fluctuations in shelf currents may be generated by coastal-trapped waves propagating from Baja California (Hickey, 1992; Noble et al., 2003). Los Angeles-Long Beach Harbor is at the northern end of San Pedro Bay. The harbor complex is protected by a 14 km breakwall that extends east from Palos Verdes. Exchange between San Pedro Bay and the harbor complex occurs through 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. two 0.45 km openings in the breakwall, Angels Gate and Queens Gate, and at the east end, where there is 3.6 km between the breakwall and the Anaheim Bay jetty. Flow through these openings fluctuates with the tides. Current measurements during March showed a net inflow of water through Angels and Queens Gates, and outflow at the eastern end of Long Beach Harbor (Corps and LAHD, 1980). This flushing maintains good water quality in the Outer Harbor (Soule and Oguri, 1980). However, shallow water regions of the Outer Harbor have reduced flushing relative to deeper water areas (Vemulakonda et al., 1991). The Inner Harbor is flushed at a slower rate, with an estimated daily exchange of water due to tidal action between 8% and 25% (Corps and LAHD, 1980; Maloney and Chan, 1974). No published results have been found that describe flow patterns in the Harbor region during summer months. Methods Water Sampling: Nine transects were made extending offshore from Huntington Beach. Typical transects contained at least 3 stations and 3 depths per station. Salinity and temperature profiles were collected at each station using a hand-held CTD (Yellow Springs Instr. Model 85). The mixed layer thickness was defined as the depth from the surface to where the temperature decreased by 1°C. For 7 of these transects, as well as 4 other days, samples were also collected at the shoreline in knee-deep water every two hours for 10 hours- a total of 5 samples each day. 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Shoreline samples were collected with buckets in knee-deep water. Offshore samples were collected from the R/V Harmony with hoses deployed to the desired sample depth. About 100 L of seawater was pumped through a 5 p im filter to remove particulates, then through a cartridge containing acrylic fibers impregnated with manganese dioxide. Ra is extracted from sea water onto these fibers with high efficiency (Moore, 1976). Radium Analysis: After returning to the lab, 2 2 3 Ra and 2 2 4 Ra were analyzed using a flow-through system with a scintillation detector linked to a time-delayed coincidence counter (RaDeCC) that can distinguish the decays of the Ra daughters, 2 1 9 Rn and 2 2 0 Rn, which rapidly grow into equilibrium with their parents (Moore and Arnold, 1996). The measured activities were then corrected for decay and 2 2 4 Ra ingrowth from 2 2 8 Th that also adsorbed to the fibers (Appendix A). Samples from at least 20 km offshore were assumed to represent the supported activity for 2 2 3 Ra and 2 2 4 Ra. To compare methodologies, a few fiber samples were processed and measured by gamma-ray spectrometry. After measuring the sample on the RaDeCC, fibers were soaked in a 90°C 1 N NH2 OH-HCl solution, with 2 g of BaCl2 added to dissolve the Mn-oxide and release the Ra from the fibers. Once the fibers turned white, the solution was allowed to cool and filtered to remove small pieces of fiber. To precipitate BaS04 , which efficiently incorporates Ra into its structure, 25 ml o f a 20% H2 S04 solution was added. The precipitate was filtered from the solution, 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dried, and transferred to a test-tube. The gamma ray activity of 2 2 4 Ra and its daughter, 2 1 2 Pb, were measured using a HP-Ge well-type gamma spectrometer (Elsinger et al., 1982). Results of the two methodologies agreed within analytical uncertainties. Two different methods were used to measure 2 2 8 Ra. One was gamma spectrometry (Michel et al., 1981). Following the method described above, fibers were leached and BaS04 was precipitated. After waiting several weeks, the gamma-ray activity was measured in the region of 338 and 911 keV for 2 2 8 Ac and 295 and 351 keV for 2 1 4 Pb. Assuming equilibrium between 2 2 8 Ac and 2 2 8 Ra and between 2 1 4 Pb and 2 2 6 Ra, the 2 2 8 Ra:2 2 6 Ra ratio was computed. The radium activity ratios were multiplied by the absolute 2 2 6 Ra activity to yield absolute 2 2 4 Ra and 2 2 8 Ra activities. The 2 2 6 Ra activity was measured on unfiltered 20-L samples by 2 2 2 Rn ingrowth (Mathieu et al., 1988). When no concurrent 2 2 6 Ra sample was collected, the average 2 2 6 Ra concentration at Huntington Beach, 98 ± 10 dpm m'3 , was used. The second method for 2 2 8 Ra was to store fibers for over a year after collection and measure the 2 2 4 Ra ingrowth using the RaDeCC system, making appropriate corrections for 2 2 8 Th. Pore Water Sampling and Analysis: Pore water samples were collected at three stations at Huntington State Beach: the shoreline, about 175 m offshore in 7.3 m water depth, and about 225 m offshore in 8.5 m water depth. These were obtained using a piezometer with a 2 cm screened interval, with SCUBA divers sampling offshore sites. After inserting the piezometer, 100-200 ml of water was drawn to 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flush and develop the well. Samples were then drawn into evacuated 0.5 L glass bottles. Assuming laminar, radial flow, water should be extracted from about a 15 cm radius around each well, effectively averaging the concentration at each depth. Samples were analyzed for Rn following the procedure outlined by Key and co workers (Key et al., 1979). After the Rn analysis, samples were poured through cartridges containing acrylic fibers impregnated with manganese dioxide and analyzed for the short-lived Ra isotopes as described above. Tidal Wedge Volume: To measure the volume of water exchanged through the beach during a tidal cycle, 5 wells with 4 cm screened intervals were installed perpendicular to the shoreline at Huntington State Beach. The water table elevation was monitored during the transition from a high tide to a low tide. The distance between the top of the well and the water table was measured by lowering an electrode until it contacted water. The uncertainty in the water table measurements is 0.25 cm. Emanation and Adsorption Characteristics of Sands: In the laboratory, the 2 2 3 Ra and 2 2 4 Ra emanation rate and distribution coefficient of sands collected at the shoreline and 7.3 m water depth was measured by making a slurry with seawater. About 300 g of sand was collected and stored wet for a month or more to allow these isotopes to reach secular equilibrium. Then, 1000 g of seawater was added, and the slurry was mixed for 10 minutes at room temperature (22 °C). This should be sufficient time to reach adsorption equilibrium (Krishnaswami et al., 1982). Afterwards, the 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. supernatant water was decanted and filtered if fine-grained particles remained suspended. The water was then passed over Mn-impregnated fibers to extract the Ra. Water addition and removal was repeated a total of 4 times. Any ingrowth of Ra was assumed to be negligible during the 90 minutes required to perform this experiment. Sands were collected from several locations, from Long Beach down to San Clemente. To measure the pore water equilibrium concentration, about 12 L of wet sand was collected from the swash zone, down to a depth of < 10 cm. In the laboratory, the sample was homogenized and saturated with seawater, leaving about 3 cm of water at the surface. The sample was then left to incubate for at least 16 days. Using a pore water sampler described above, 520 ml of water was extracted and passed over Mn-impregnated fibers to extract the Ra. For 2 2 2 Rn analysis, 110 ml was extracted and analyzed using a rapid radon extraction system (Berelson et al., 1987). Splits of these samples and other, smaller samples collected from the swash zone, were used to measure the total sediment activity. These samples were dried, placed into test tubes, and counted using a HP-Ge well-type gamma spectrometer (Elsinger et al., 1982). Two splits from each site were measured to constrain sample heterogeneities. Differences between the splits were within the uncertainty based on the counting statistics. 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Total Activity Removed (dpm) Total Activity Removed (dpm) Figure 2-2. Results of adsorption experiment, A) 223Ra and B) 224Ra. Solid lines are for beach sands and the dashed lines for sand at 7.3 m depth. THr 0.8 - 0.6 0.4 0.2 m = -0.856 ± 0.370 L V = 0.7 L Ms = 563 g Kd = 0.00 ± 0.02 cm3 g'1 0 0.1 m = -1.572 ± 0.182 L V = 0.67 L Ms = 227 g = 3-?5 ± 0.46 cm3 g~1 ^ A) 223Ra 0.2 0.3 0.4 0.5 223Ra Cone, (dpm L'1) 0.6 0.7 12 10 * m = -1.118 +0.156 L V = 0.7 L Ms = 563 g Kd = 0.41 ± 0.06 cm3 g'1 i-in ' 6 4 2 0 H ® m = -1.379 ± 0.246 L V = .67 L Ms = 227 g Kd = 3.-)0±0.55pm 3 g-1 B) 224Ra o 2 3 4 5 224Ra Cone, (dpm L"1) 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results A d s o r p t i o n : There are two reservoirs of mobile Ra in sediments: dissolved and adsorbed. The dissolved (Cw ; atoms cm'3 ) and adsorbed (Cs; atoms g"1 ) concentrations are related by a distribution coefficient: Kd (cm3 g'1 ): Kd = Cs /Cw (2-1) For solutes in low concentration, such as Ra, the relationship between these two pools is linear (Krishnaswami et al., 1982; Langmuir, 1997). Kd was measured in the seawater-sediment slurry experiment. After the sediment in the slurry settled, the Ra concentration of the supernatent water was measured. This was repeated several times with new aliquots of water. The total Ra isotope activity removed after each step is plotted against the Ra concentration in Figure 2-2. The Y intercept of a straight line fit through these points is the total mobile Ra isotope activity for the sample, which is further discussed below. The slope of this line, m , is a function of the distribution coefficient: m = V + K dM s (2-1) where V is the volume of water in contact with the sediments and M s is the mass of sand. 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-1. Weight percent of sediments finer than 180 /on, Ra distribution coefficient (Kd ), Rn and Ra emanation rate, and the Rn and Ra predicted equilibrium pore water concentration measured in the laboratory. Sands collected on 9/11/03. Values and uncertainty represent average of two measurements. Slurry effect (Hammond & Fuller, 1979) is assumed to be negligable. Beach Sand 7.3 m Station 8.5 m Station Fraction of Fines (<180 pm): 11% 70% 76% Porosity: 0.40 + 0.02 0.40 ± 0.02 Bulk Sediment Production Rate (dpm kg'1 ): 226Ra 1090 ± 60 1300 + 50 2470 + 290 224Ra 1840 ± 80 1760 ± 70 4120 ± 90 228Ra 1970 ± 180 1930 ± 140 4380 ± 190 Emanation Rate (dpm kg'1 ): 222Rn 46 ± 2 57 ± 2 83 ± 2 223Ra 1.7 ± 0.2 3.5 ± 0.5 224Ra 21 ± 1 34 ± 1 Emanation Power (Emanation Rate/Production Rate): 222Rn 4.2 ± 0.2% 4.4 + 0.2% 3.4 + 0.1% 223Ra* 3.3 ± 0.2% 5.7 ± 0.2% 224Ra 1.1 ± 0.1% 2.0 + 0.1% Adsorption (Based on data In Figure 2): Distribution Coefficient (cm3 g'1): 0.4 + 0.1 3.1 ± 1.0 Partition Coefficient: 1.5 ± 0.4 11.6 ± 3.8 Predicted P W Concentration** (dpm L -1 ): 222Rn 172.4 + 6.4 214.4 ± 7.0 310.3 ± 9.2 223Ra 2.5 + 0.7 1.0 ± 0.4 224Ra 31.0 ± 7.9 10.2 ± 3.3 * Estimated assuming 238u / 235U = 222Rn/223Ra = 2 1 . ** Computed using Equation 2 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The distribution coefficient was 0.4 ±0.1 cm3 g'1 for beach sands and 3 ± 1 cm3 g'1 for surficial sands collected at 7.3 m water depth (Table 2-1). The increase in adsorption coefficient from the shoreline to the 7.3 m site is most likely a manifestation of an offshore decrease in grain size (Table 2-1). At Huntington Beach, the mean grain size of surficial sediments decreases from 250 p im at the shoreline to < 60 /im at 1 km offshore (Grant, 1973). E m a n a t i o n R a t e : The emanation rate, E (dpm g'1 ), is the rate mobile Ra is produced per gram of sediments. The Y-intercept from Figure 2-2 is the total mobile 2 2 3 Ra and 2 2 4 Ra. Dividing this by the mass of sediment used in the experiment, E for each isotope can be computed (Table 2-1). The emanation rate for 2 2 2 Rn was also measured in the laboratory (Table 2-1). For each isotope, the emanation rate increases with water depth. The emanation rate and its relation to the total sediment activity is further discussed in the next section. P r o d u c tio n R a t e : The total sediment activity of 2 2 8 Ra (via 2 2 8 Ac), 2 2 4 Ra, and 2 2 6 Ra (via 2 1 4 Pb) in sands from beaches and two offshore stations was measured with gamma spectrometry (Table 2-1). For each sample, 2 2 8 Ra and 2 2 4 Ra were in secular equilibrium. This suggests there is no enrichment of 2 2 8 Th on sediments, which is observed in some environments (Cochran, 1979; Torgersen et al., 1996). In most samples, 2 2 6 Ra activity was greater than 2 2 8 Ra and 2 2 4 Ra. This must reflect higher activities of 2 3 0 Th relative to 2 3 2 Th. 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. For shoreline sands collected between Long Beach and San Clemente, changes in the total sediment activities were observed (Table 2-2). Moving from north to south, there is a slight decrease in activity from Long Beach to Huntington Beach. The activity at Sunset Beach was anomalously low. This may represent imported sand, which is used to combat erosion. At Laguna Beach and Dana Point, the activity of 2 2 8 Ra and 2 2 4 Ra is half that observed at Huntington Beach. At San Clemente, the total sediment activities increase again. There are two mechanisms responsible for changes in the total sediment activities. First is the source of the sands on the beach. North of Newport Canyon, the primary source of sediments are the local rivers (Emery, 1960). South of Newport Canyon, much of the sediment is from erosion of cliffs at the back of the beach (Rice et al., 1976). One of the rivers, most likely near the north end of San Pedro Bay, must drain a region with relatively high U- and Th- bearing rocks. Second, Newport Canyon appears to inhibit exchange of beach sands to the north and south. However, there is evidence that Newport Canyon is not currently an active sink for sands (Felix and Gorsline, 1971). Regardless, sediments to the north and south of Newport Canyon have differing 2 2 6 Ra, 2 2 4 Ra, and 2 2 8 Ra signatures. At Huntington Beach, total sediment activities were identical between the shoreline and the 7.3 m station (Table 2-1). The fraction of fines apparently made no difference in the total sediment activities. The activities doubled at the 8.5 m station. It is unlikely that this was a function of the grain size, since the fraction of fines at 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the two offshore stations are similar. This may represent a change in mineralogy with distance offshore. For example, the fraction of epidote in surficial sediments increases with distance offshore (Grant, 1973). However, since these samples were collected only about 50 m apart, it is also likely that there are small scale heterogeneities in sands. Emanation rates are less than the total sediment activities. The factor relating these two values, emanation power, is presented in Table 2-1. These values are comparable to previously published values for 2 2 2 Rn (Tanner, 1964). Between the shoreline and the 7.3 m stations, the emanation power increases. One possibility is that the 2 2 2 Rn and Ra activities are not controlled by the total sediment activity of the parent isotope, but instead by the activity in a secondary crust at the grain surface. This has previously been observed in groundwater systems, where the production of 2 2 2 Rn is controlled by adsorbed 2 2 6 Ra (Tanner, 1964). However, this is unlikely in this environment since the flushing of sediments with seawater should effectively remove the soluble, long-lived parent and grand parent isotopes, 2 2 6 Ra, 2 2 7 Ac, and 2 2 8 Ra. In addition, no excess accumulation of 2 2 8 Th was observed as shown by the consistency of 2 2 4 Ra and 2 2 8 Ac results (Table 2-2). Instead, it is likely that the emanation power is a function of grain size, where finer particles have a greater surface area, which increases the probability for 2 2 2 Rn and Ra to escape from a sediment grain. 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-2. Shoreline sand A) bulk sediment isotope production rates and B) equilibrium porewater concentrations. Stations are listed from north to south. A ) B u t e S e d i m e n t Bulk Sediment Collection Production Rate (dpm kg'1 ): Station Date 226Ra 224Ra 228Ra Long Beach 9 /2 3 /0 3 1810 + 96 5 0 8 0 + 140 5 4 7 0 ± 3 1 0 N. Seal Beach 9 /2 3 /0 3 6 5 0 ± 77 1140 ± 105 1340 ± 230 S unset Beach 5 /2 6 /0 4 1420 ± 62 3 1 5 0 ± 80 3 2 9 0 ± 182 N. Huntington Beach 9 /2 3 /0 3 892 ± 47 17 5 0 ± 64 1830 ± 139 S. Huntington Beach 9 /1 1 /0 3 1090 ± 60 18 4 0 ± 82 1970 ± 177 N. Newport Beach 5 /2 6 /0 4 775 ± 47 1 2 8 0 ± 58 1230 ± 134 S. Newport Beach 9 /2 3 /0 3 561 ± 62 9 0 2 ± 83 1020 ± 178 Laguna Beach 9 /2 3 /0 3 581 ± 50 532 ± 65 507 ± 144 Laguna Beach 5 /2 6 /0 4 4 5 2 ± 51 4 3 7 ± 62 4 9 8 ± 141 Salt Creek 9 /2 3 /0 3 4 7 6 ± 86 8 1 8 ± 115 6 3 8 ± 258 Dana Point 5 /2 6 /0 4 4 2 2 ± 65 393 ± 80 5 2 0 ± 177 San Clem ente 5 /2 6 /0 4 798 ± 49 781 ± 60 9 4 0 ± 136 B) Pore W ater Equilibrium Pore Water Collection Concentration (dpm L'1 ) Date 222Rn* 223Ra 224Ra S unset Beach 5 /2 6 /0 4 (7 7 .8 ± 1.5) 2 .2 0 ± 0.12 2 5 .6 ± 0.3 S. Huntington Beach 9 /1 1 /0 3 (5 7 .9 ± 1.2) 1.59 ± 0.1 2 14.4 ± 0.3 Laguna Beach 5 /2 6 /0 4 (30.1 ± 0.9) 0 .8 7 ± 0.08 8.0 ± 0.2 Dana Point 5 /2 6 /0 4 (2 6 .0 ± 0.7) 0 .5 7 ± 0.09 5.1 ± 0.2 San Clem ente 5 /2 6 /0 4 (9 8 .9 ± 1.3) 1.69 ± 0.17 14.3 ± 0.4 * C oncentrations in p aren th eses are a lower limit estim ate because of atm ospheric exchange of sam ples. 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E q u i l ib r i u m P o r e W a te r A c t i v i t y : Three methods were used to assess the equilibrium pore water activity (XCe q ) for each isotope (Table 2-3). First, with a known emanation rate and partition coefficient, the equilibrium pore water activity can be computed as: X .r ( l ~ * ) P E (n 2 ) eq 0 + ( l + tf> )p £ rf where ( j ) is the porosity (0.40 ± 0.02) p is the sediment density (2.5 g cm'3 ). The values of E and Kp are presented in Table 2-1 for the Huntington Beach station and the 7.3 m station. Both E and Kp should depend on grain surface area and increase with decreasing grain size. This is observed between the Huntington Beach shoreline station and the 7.3 m offshore station, where the emanation rate increases by a factor of 2 and the partition coefficient increases by a factor of 6 (Table 2-1). These increases largely compensate for each other when XCe q is computed (Table 2- 3). Second, relatively deep pore waters (>0.48 m) were collected at three stations. Deep pore waters are beyond the region where pore waters exchange solutes with bottom waters, or at least the rate of exchange is slow relative to the isotope half-life. In these pore waters, a balance between the emanation rate and the pore water concentration, 7 * C e q , should be nearly achieved. Similar deep pore water 65 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Table 2-3. Huntington Beach pore water measurements for 2 2 2 Rn, 2 2 3 Ra, and 2 2 4 Ra: A) shoreline samples and B) offshore samples. The residence time of water in the beach was computed using the box model presented in the text. Depth M ow Collection Dtot From Concentration In dpm L'1 Residence Time In Days Os Os Surface (cm) Date Shore (km) 2 2 2 Rn 223 Ra 2 2 4 Ra 22 2 Rn 2 2 3 Ra 2 2 4 R a A) SH O R E U N E S A M P L E S 8 9 /1 1 /0 3 28 9 /1 1 /0 3 48 9 /1 1 /0 3 78 9 /1 1 /0 3 180 8 /2 8 /0 3 0 0 0 0 0 20.5 ± 1.0 0.3 ± 0.3 61.4 ± 1.3 134.2 ± 0.6 0.70 + 0.10 0.17 + 0.05 0.64 ± 0.16 1.20 ± 0.19 1.16 ± 0.09 11.8 ± 0.4 3.6 ± 0.3 9.2 ± 0.6 12.9 ± 0.6 12.2 + 0.3 1.9 ± 0.1 0.0 ± 0.0 7.6 ± 0.2 48.3 + 2 3.3 ± 0.5 0.4 ± 0.1 2.7 ± 0.7 76.3 ± 12.0 35.8 ± 2.7 26.2 ± 0.9 2.4 ± 0.2 11.9 ± 0.7 42.2 + 2.1 30.6 ± 0.7 Average of 0-48 cm: 27.4 ± 31.1 0.50 + 0.28 8.2 ± 4.2 2.6 ± 3.0 1.7 ± 1.0 9.1 + 4.7 Beach Sand Incubation Experiment: Predicted Equilibrium Concentrations: (57.9 + 1.2) 172.4 ± 6.4 1.59 ± 0.12 2.5 ± 0.7 14.4 ± 0.3 31.0 ± 7.9 B) O F F S H O R E SA M P L E S Z O N E 1 (< 28 cm) 8 9 /1 1 /0 3 8 9 /1 1 /0 3 0.18 0.23 67.9 ± 0.8 68.9 ± 0.9 1.06 ± 0.24 0.64 ± 0.14 11.4 ± 0.7 8.5 ± 0.5 ZOtE 1-Z O N E 2 Transition (28 28 9 /1 1 /0 3 28 9 /1 1 /0 3 cm) 0.18 0.23 57.6 ± 0.4 145 ± 1 0.74 ± 0.25 0.89 ± 0.16 15.2 ± 1.1 10.7 ± 0.5 ZOtE 2 (> 28 cm) 48 9 /1 1 /0 3 78 9 /1 1 /0 3 0.18 0.18 153 ± 1 1.22 ± 0.26 1.21 ± 0.17 15.9 ± 0.9 14.3 ± 0.5 Predcted EqdRbrium Concentrations h Surface Sands: 0-4 9 /1 1 /0 3 0.18 214.4 ± 7.0 0-4 9 /1 1 /0 3 0.23 310.3 + 9.2 1.0 ± 0.4 10.2 ± 3.3 concentrations were observed at the shoreline and at both offshore stations (Table 2- 3). Third, in the laboratory, beach sands from several other locations were incubated for to allow the isotopes to grow into equilibrium with pore waters. For Huntington Beach, the incubation experiment pore water concentration was equivalent to the deep pore water concentrations observed both at the beach and offshore. These results were also in agreement with the A .C e q calculated with Equation 2-2, considering the large uncertainty associated with this calculation. Taken together, at Huntington Beach, XCe q of Ra changes little with distance offshore. Longshore variations in LCc q and the total sediment activity were similar for each isotope (Table 2-2). Sunset Beach has the highest activity, decreasing to Huntington Beach. Laguna Beach and Dana Point have significantly lower concentrations. The activities increase again at San Clemente. However, the ratio between XCe q and the total sediment activities is greater south of Newport Canyon. Assuming that these differences result from differences in adsorption, the emanation rates appear to be a function of the total sediment activity, not an external mineral coating as observed for 2 2 2 Rn in fresh groundwater systems (Tanner, 1964). For 2 2 2 Rn, the activity measured in incubation experiment is significantly less than that measured using the other two methods. The most likely cause is a loss of Rn to the atmosphere during the incubation process. Samples were stored in 20 L plastic 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. buckets exposed to the atmosphere. Despite being stored indoors, convection may have occurred by either temperature changes or because of evaporation of water at the surface. Based on the model presented below, this would require an advective flow rate of about 4 cm d'1 . For Ra, mixing would prevent local heterogeneities, and minimize boundary effects in the bucket. S e a flo o r . Pore water 2 2 2 Rn, 2 2 3 Ra and 2 2 4 Ra data is presented in Table 2-3, and a composite pore water profile for each isotope in 7-8 m water depth is presented in Figure 2-3. Bottom water Ra concentrations were measured on several occasions and found to be about 0.014 dpm L"1 for 2 2 3 Ra and 0.15 dpm L’1 for 2 2 4 Ra (Table 2-4). Bottom water Rn concentrations have not been measured, but are expected to be about 1 dpm Lf1 . All three profiles show similar patterns, with a large increase in concentration between the surface and the first sample depth. Samples collected at 28 ± 2 cm depth showed the greatest variability, suggesting a significant change in concentration near this depth. Below this depth, concentrations remain relatively constant. These profiles can be interpreted using a two zone model. In the upper zone, waves and currents interact with ripples on the seafloor, driving a vertical flow of pore water with some regions having net downward flow and others having net upward flow (Figure 2-4) (Huettel et al., 1996; Ziebis et al., 1996). This increases the depth of exchange for solutes, relative to a diffusive system. For example, oxygen will be carried deeper into the sediments, and Rn and Ra produced at depth will be 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-4. Radium concentrations in nearshore, near bottom water samples. Collection Diet. Above Diet. From Concentration In dpm nrf3 Date Seafloor (m) Shore (km) 223Ra 224Ra 9 /1 3 /0 1 0.4 1.35 4.5 ± 2.3 98 ± 9 4 /3 /0 2 1.4 1.67 9 .4 ± 1.4 106 ± 4 6 /7 /0 2 2.1 1.08 11.5 ± 1.2 138 ± 4 6 /1 2 /0 2 0.6 0.99 11.0 ± 1.1 142 ± 4 8 /2 3 /0 2 0.6 0.5 15.5 ± 1.2 155 ± 3 6 /1 8 /0 3 0.5 0 .9 4 19.8 ± 2.1 159 ± 5 6 /2 5 /0 3 0.9 1.10 2 3 .4 ± 1.8 2 7 0 ± 6 Average: 13.6 ± 6 .5 153 ± 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-3. Pore water concentration composite profiles at 7.5 m and 8.3 m depth offshore of Huntington Beach for A) 222Rn, B) 224Ra and C) 223Ra. Depth uncertainty is equal to the size of the points. Solid line represents the equilibrium pore water activity. Arrow indicates bottom water concentration. E o a a > a o 10 - E o 20 - SZ 30 - a a ) 40 - a 50 - 60 - 70 80 "A) 222 Rn 0 20 40 60 80 100 120 140 160 0 10 20 30 40 50 60 70 80 B) 224Ra 60 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Concentration (dpm L-1) 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-4. Schematic diagram of flow through sediments produced when waves and currents interact with seafloor ripples. (Modified from Ziebis et al. [1996]) Reproduced with Mean Current — > Low Low Low Water Zone ^ Base of Influence of Flow Zone 71 permission of the copyriph, owner. Further reproduction p r o v e d witnou, p e n s io n . transported out of the sediments before radioactive decay occurs. This also produces both vertical and horizontal gradients for pore water solute. In this study, the relatively large samples collected (0.5 L) extract water from a radius of several cm, integrating these regions. The depth of wave- and current-driven advection through the seabed may be limited and appears to extend to 28 cm. Below this, transport should be dominated by diffusion. However, some deeper advection may occur, but must be slow enough that the short-lived isotopes decay faster than they can be transported. Deeper advection may impact the 2 2 8 Ra distribution. Along with the different pore water transport rates for the two zones, there may also be a difference in adsorption. Adsorption can be influenced by mineral coatings on sand grains. The presence of manganese oxide, which has a large surface area, increases the Ra distribution coefficient. Under anoxic conditions, found deep in the sediments, manganese oxide is reduced and becomes soluble. This decreases the sediment adsorption coefficient. Grain size may also increase with depth in the sediments. The fine grained sands at the surface may represent recent deposits that overlie relic beach and dune sands. Late Pleistocene sands are known to outcrop near the center of San Pedro Bay, which correlate with uplifted sediments found in the Palos Verdes Hills (Moore, 1951). Coarser sediments would adsorb less Ra. Therefore, lower adsorption rate may be more representative at depth. 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A numerical model of this two zone system was developed to use the distribution of Ra and Rn isotopes to compute the water exchange rate and isotope fluxes. In Zone 1, extending from the surface to depth L, exchange is accomplished by diffusion and non-local exchange with bottom water (Imboden, 1981). While non-local exchange may not accurately characterize the physical mechanism of exchange, it should be appropriate when the concentration is averaged at a given depth. In Zone 2, solute transport is attributed exclusively to molecular diffusion. Two processes, which may be important in other environments, are not included in this model. First, bioturbation undoubtedly occurs in these sediments, and in some environments it may decrease Ra pore water concentrations and increase the Ra flux from sediments (Sun and Torgersen, 2001). However, at this site, irrigation rates are sufficiently fast that the influence of bioturbation is negligible. Second, an upward advection of terrestrially-derived groundwater water is not included. In this region, terrestrially-derived submarine groundwater discharge is negligible, and is further examined, along with the 2 2 6 Ra budget, in Chapter 4. The governing equations for each zone are: ZONE 1: For 0 < z < L, ^ _ o - ^ _ a ^ Cx - C 0) - c t> ( 1 + K x) X C x + p(l - < f)E (2-3a) o t a z ZONE 2: For z > L , 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. < ? c 2 dt = 0 = 4>DS - ^(1 + k 2) AC2 + p { 1 - <t>)E oz (2-3b) Where C: is the dissolved isotope concentration in Zone i, C0 is the isotope concentration in the overlying water, t is time, Ds is molecular diffusivity adjusted for tortuosity (cm min"2 ), z is sediment depth (cm), a is the non-local exchange rate (min"1 ), and K is the Ra isotope decay constant (m in1 ). Diffusion, porosity, emanation rate, and non-local exchange (in Zone 1) are all constant with depth. The partition coefficient, Ki; is defined as: Where different values of Kd can be chosen for each Zone. Four boundary conditions are required to solve these differential equations: Briefly, the first boundary condition states that the concentration at the sediment- water interface is equal to the concentration in the overlying water. The second and (2-4) <P Ci(z = 0) = C0 (2-4a) C1 (z = L) = C2(z' = 0); z ' - z - L (2-4b) dz )Z , L \ dz / z ._0 (2-4c) (j)(l + K 2)A .C 2(z '= ™ ) = p { l- ( t > ) E (2-4d) 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. third conditions produce continuity in total concentration and pore water flux, respectively, between Zone 1 and Zone 2. The last condition is that the total mobile concentration at depth is defined by the isotope emanation rate. The solution to these differential equations are in the form: ZONE 1: For 0 < z < L, C, = A ,e rz + A 2e ri + A3 (2-5a) ZONE 2: For L < z , C2 = B x e si + B. (2-5b) Where the scaling parameters are: r = a + <pk(Kx + 1 ) (2-6a) s =. m K 2 + 1 ) (2-6b) The constants are defined as: B 2 a 3 ( c 0 a 3) A - I , H » + * 2 )U , r(U K >) s ( U K , ) J { s(UK,)l (2-7a) A2 — C„ A { A 3 (2-7b) p ( l - < p ) E + a C 0 </)(a + (j>k(Kx + 1)) (2-7c) 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (2-7(1) p ( l - < P ) E 2 A 02(1 + if 2) (2-7e) Model parameters are presented in Table 2-5. The base of Zone 1 was defined as 28 cm because the pore water concentration showed the greatest variability at this depth. For each isotope, the emanation rate was computed with Equation 2-2 using the average XCe q for samples collected in Zone 2. For Ra, an intermediate distribution coefficient of 2.2 ± 0.3 cm3 g"1 was used. The resulting Ra emanation rates are within the uncertainty of that measured at the 7.3 m station (Table 2-5). For Rn, the emanation rate calculated based on the pore water concentrations in Zone 2 is 45% lower than the measured emanation rate. The model was first fit to the 2 2 2 Rn profile by adjusting the non-local exchange rate (Figure 2-4). The model profile predicts a rapid increase in concentration in the top few cm, then reaching a relatively constant concentration throughout most of Zone 1. At the base of Zone 1, the pore water concentration again increases to the equilibrium pore water activity. Based on this model fit, the non-local exchange rate is 11 x 10"7 s'1 , which is comparable to values for presumably biologically-driven exchange in muddy shelf sediments (Colbert, 2001; Emerson et al., 1984). However, this circulation extends much deeper, driving a pore water flux of 27 L m 2 d'1 through the top 28 cm, replacing 24% of the pore water in Zone 1 each day. This 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-5. Pore water model input parameters and results. PW Model 222 Rn 223 R* 224 R , Physical Parameters: Decay C onstant (xl 0"4 min"1): 1.26 0.42 1.32 Porosity: 0.4 0 0.4 0 0.4 0 Diffusivity* @18° (x10"s cm 2 s e c '1): 1.05 0.69 0.69 T ortuosity Adj. Diffusivity** (x IO '5 cm 2 min"1): 10.1 6.62 6.62 Defined Parameters: Zone 1 Thickness (cm ): 28 28 28 Overlying W ater Cone, (dpm L'1): 1 0 .0 1 4 0 .15 Ave. Irrigated Zone Cone, (dpm L"1): 65 0.83 10.2 Equilibrium Cone, (dpm L"1): 149 1.34 14.9 Mobile Phase Production Rate (dpm L"1): 59.5 1.4 16.1 Adjustable Parameters: Non-Local Exchange Rate (xIO"5 m in'1) 6.6 6.6 6.6 Zone 1 Distribution Coefficient (cm 3 g"1): 0 3.1 3.1 Zone 2 Distribution Coefficient (cm 3 g"1): 0 2.2 2.2 Results: Model Flux (atom s m"2 sec"1) (±50% ): 163 6.5 33.7 Seafloor in c o n tac t with mixed layer (m) (±15% ) 1200 1200 1200 Total Flux (xIOOO atom s m"1 s e c '1)A (±50% ): 196 8 40 * Diffusivity com puted based on algorithm presen ted by Li and Gregory (1 9 7 4 ) for Ra and Jahne e t al. (1 9 8 7 ) for 222Rn. ** The adjusted diffusivity w as com puted by dividing th e diffusivity by th e tortuosity, defined as (porosity)"2 (Ullman and Aller, 1982). A Total flux to mixed layer per m of shoreline. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. water flux is similar to previous estimates of exchange driven by wave pumping through sediments based on computed pressure gradients and dye studies (Mu et al., 1999; Precht and Huettel, 2003; Riedl et al., 1972). To fit the Ra profile, the non-local exchange rate estimated with the 2 2 2 Rn data was used and the value of Kd ), was increased to 3.0 cm3 g'1 (Figure 2-4). This value of Kd ), produced the best fit to the 2 2 4 Ra data and is within the uncertainty of the measurement for surface sands at 7.3 m water depth. These parameters also produce a satisfactory fit to the 2 2 3 Ra data. From the modeled pore water profiles, the Ra flux can be computed. The total seafloor isotope flux (JS F ) is the sum of two components, a diffusive flux (Jd ) and an irrigation flux (J,): J SF = h + I = ~ « f f a ~ C ° Y Z V - V \ "Z / Z = 0 where all terms are as defined above, and the negative sign represents a flux out of the sediments. In this system, the diffusive flux Jd accounts for less than 10% of the total flux. The total fluxes are presented in Table 2-5. The uncertainty in the flux may be greater when averaged across the seafloor in contact with the mixed layer. There are four main factors that control the flux: the emanation rate, adsorption coefficient, and the depth and rate of irrigation. As noted above, at Huntington Beach, adsorption and the emanation rate do change with 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. distance offshore (Table 2-3). However, these changes largely compensate, producing an equilibrium pore water activity that is relatively constant. When XCe q is held constant and the emanation rate and adsorption are varied within a reasonable range, the flux changes by less than 10%. The depth and rate of irrigation is not expected to change considerably across the sediment surface area in contact with the mixed layer. Of primary importance is the presence of ripples on the seafloor. Ripples are evidence that wave energy is sufficient to move sediments and drive a flow of water through sediments. In addition, ripples can interact with currents to produce pressure gradients at the seafloor that can maintain an advective flow through sediments. With minimum surface wave lengths of 30 m in San Pedro Bay, ripples are expected to occur on the seafloor to a depth of at least 15 m (Denny, 1988). Based on seafloor video and photographs, the transition from well-developed, physically generated ripples to biogenic structures occurs between the 40 and 45 m isobaths (Karl et al., 1980). Across this zone, stretching from the shoreline to the seafloor-mixed layer intersect, waves have their strongest impact on the seafloor, and steady advection leads to a permanent flushing of the upper sediment layers (Precht and Huettel, 2003). Therefore, while the water exchange rate may decrease with depth, the presence of ripples should prevent any dramatic decrease in the water exchange rate. Since the pore water profiles were collected near the mean water depth of the mixed layer, they should be representative of the mean. Taking this into consideration, the Ra flux 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. uncertainty is assumed to be 50%. The flux per unit length of shoreline is calculated as the product of flux per unit area computed with Equation 2-8 and the distance from the shoreline to the depth where the mixed layer base intersects the seafloor (Table 2-5). The seafloor flux to the mixed layer south of San Pedro Bay can be estimated. For beach sands at Laguna Beach and Dana Point, the equilibrium pore water activity was about 45% less than at Huntington Beach for both isotopes. Assuming offshore sands have a similar equilibrium pore water activity, porosity, irrigation rate and irrigation depth, the isotope flux per unit area of seafloor should be 45% less. The seafloor flux to the mixed layer is further reduced because of an decrease in the sediment area in contact with the mixed layer. South of Newport Canyon, the seafloor slope increases by a factor of 3, and the mixed layer intersects the seafloor only 350 m offshore. Accounting for both the reduced isotope activity and reduced contact area, the seafloor flux to the mixed layer south of Newport Canyon is only 13% of that at Huntington Beach. The relative flux of Ra isotopes is a useful parameter. The dynamics between the flux ratio and the model-predicted non-local exchange rate are not obvious and deserve further discussion. To better understand which mechanisms are important, the equations given above need to be simplified. First, since the short-lived Ra concentration in overlying water is orders of magnitude smaller than the equilibrium pore water activity, C0 = 0. Second, examining Equation 2-7a in this setting, the 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. product rL in the exponent is large, reducing this equation to A,=0. Applying these assumptions to Equation 2-7b, A2 =-A3 . The diffusive flux ratio and the irrigation ratio can be investigated individually. Solving for the diffusive flux in Equation 2-8: (2-9) where the variables are as defined above. After simplifying, the diffusive flux ratio can be computed: r 224 A ? ,224 \ u 223 / ^223^3,223 (2- 10) where the subscripts 223 and 224 are for 2 2 3 Ra and 2 2 4 Ra, respectively. This can be further simplified for two end-member cases. First, in an entirely diffusive system, where irrigation is negligable, Equation 2-10 simplifies to: / J N 224 ■A .224 12-224 \ 223 / d 3,223 V 223 (2-1 la) And, when the system is dominated by irrigation, then: 'j x 224 _ -^3.224 \ ^ 2 2 3 Jd -^3,223 (2-1 lb) 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. At intermediate irrigation rates, the diffusive flux ratio is between these two extremes. Solving for the irrigation flux in Equation 2-8: /,= a ■*i) '- ( e rL - 1 ) - ^ ( e ' rL - 1) + L ( A 3 - 0 (1 + K y) C 0) (2-12) where the variables are as defined above. Again, this equation can be simplified and the irrigation flux ratio computed: J 224 _ ^ 3 ,2 2 4 (2-13) 223 / j 3,223 This equation is the same as to equation 2-1 lb. Thus, at high irrigation rates, the total flux ratio can be approximated as the A3 ratio. Recalling Equation 2-7c, its important to note that A3 , is a function of the non-local exchange rate. Again, using the assumptions outlined above, the A3 ratio can be expressed: 3,224 _ 224 3,223 '2 2 3 a 0 (^ + 1 ) + A . ’ 223 a <P(Kl + 1 ) + A '224 (2-14) This equation can then be generalized by normalizing the A3 ratio to the emanation rate ratio. Then, at high irrigation rates, the normalized A3 ratio equals 1. 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-5. A) 224Ra:223Ra flux normalized to the emanation rate ratio as a function of the irrigation rate. The diffusive flux and irrigation flux ratios are as defined in the text. Except for the irrigation rate, values used to compute the model-derived flux ratio are presented in Table 2-5. B) 224Ra:223Ra flux normalized to the emanation rate ratio as a function of the irrigation rate for various thicknesses of Zone 1. ra x 3 ra O' CO C M CM _ ra O' C M C M ■ o o N ra E o z 1 Normalized Diffusive Flux Ratio ^ 0.9 0.8 Normalized Model- Derived Flux Ratio 0.7 0.6 0.5 Normalized Irrigation Flux Ratio' 0.4 0.3 10‘ 9 10’ 8 10‘ 7 10"6 10’ 5 0.0001 0.001 0.01 0.1 Irrigation Rate (m in1) ra 01 x 3 ra O' C O C M C M ra O' C M C M T3 0 ) N ra E V . O z L=4 cm L=14 cm L=28 c 0.0001 0.001 0.1 Irrigation Rate (min-1) 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This discussion is tied together in Figure 2-5a, where Equation 2-14, the irrigation flux ratio, Equation 2-10, the diffusive flux ratio, and the model-derived flux ratio are each plotted as a function of the non-local exchange rate. Each ratio was normalized to the emanation rate ratio. As anticipated, at very low irrigation rates, the model-derived flux ratio follows the diffusive flux ratio, and at high irrigation rates, it follows the irrigation flux ratio. But at intermediate irrigation rates, such as those measured in this study, the flux ratio lies between these two extremes. Interestingly, the normalized flux ratio may even be less than the diffusive flux ratio at modest irrigation rates. Therefore, care must be taken when using the flux ratio to interpret the mechanism of Ra removal from the seafloor or the production rate ratio in sediments. The model-derived flux ratio presented in Figure 2-5a is only valid for the model values presented in Table 2-5. The model-derived flux ratio as a function of the irrigation rate is also sensitive to the adsorption coefficient, porosity, molecular diffusivity, and the depth of irrigation. For example, in Figure 2-5b, the model- derived flux ratio computed for three irrigation depths is presented. Regardless of changes in the environment that would change these parameters, the relationship between the flux ratio and the irrigation rate maintains its general shape. S h o r e lin e : Waves and tides pump water through beach sands. These two processes occur at very different rates and length scales, so it is necessary to assess each mechanism individually (Figure 2-6). Seawater percolates into the beach as waves 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. run-up onto unsaturated sand. A fraction of this water rapidly drains back out of the beach, producing a shallow, rapid cycling of water that may strip all Ra produced in a layer of sand extending across the swash zone. The wave pumping Ra flux, JW P , is then equal to the production rate in a volume of bulk sediment (V): J WP= E p (l-< P )V (2-15) Each of these values is known except V, which can be estimated. The volume depends on the unit area of the swash zone, which is about 21.5 m2 (m shoreline)"1 , extending from the low tide seepage face to the mean high high water mark. On a similar beach, the thickness of sand filled and drained by wave activity was surveyed, and the depth of wave pumping was found to be about 16 cm (Riedl, 1971). Using these values, the flux from wave pumping is computed in Table 2-6. The primary source of uncertainty is the depth of water cycling, which may vary by 30% (McLachlan, 1979). Seawater can infiltrate the beach when the water table within the beach is lower than sea level or when wave run-up extends above the water table. This generally occurs during high tide, with a reciprocal flow out of the beach during low tide (Figure 2-6). This tidal pumping of water through the beach face was measured by recording changes in the water table height over a tidal cycle (Table 2-7). The hydrographs for 3 representative wells and the theoretical tidal height are plotted in Figure 2-7. The tidal wedge is defined as the volume of sediments filled and drained during a tidal 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Figure 2-6. Schematic cross-section diagram of flow through the beach. Solid line represents beach surface. Dotted line are the maximum and minimum water table elevations. In general, water enters the beach during high tide and drains out during low tide (solid arrows). There is also some mixing within the beach (see text). Shaded box at beach surface represents the volume that is affected by wave pumping (open arrows). 50 B u -50 -100 Tidal Pumping Wave Pumping Tidal . Range -150 -200 N / -250 30 20 10 0 -10 -20 -30 Distance (m) O O O s Table 2-6. Summary of Ra flux from wave and tide pumping through the beach. 2 2 4 R a /2 2 3 R* 2 2 3 Ra 2 24 Ra S H O R E L I N E FLUX: Length of swash zone (m): 22 ± 2 22 ± 2 Wave penetration depth (m): 0.16 + 0.08 0.16 + 0.08 Porosity: 0.40 ± 0.02 0.40 + 0.02 Pore Water Volume (m3 m'1): 1.41 ± 0.07 1.41 ± 0.07 Shoreline Emanation Rate (dpm g'1): 1.7 ± 0.3 21 ± 2 Wave Flux (atoms sac'1 (m shore)1 ): 60 ± 18 727 ± 218 Beach Water Concentration (dpm m'3): 502 ± 151 8175 ± 2452 Water Flux* (m3 day'1 (m shore)'1): 2.6 + 0.3 2.6 + 0.3 Tide Flux (atoms sac'1 (m shore)'1 ): 363 ± 113 1899 ± 494 * Flux from tidal pumping after subtracting volume of sediments affected by wave pumping 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-7: A) Locations, beach surface, and water table heights at each well measured relative to high water mark near well C. Water table elevation at each well has a precision of 0.25 cm. The absolute elevations have an uncertainty of ± 2 cm. B) Individual water table measurements for each well on 8/28/03. Well A Wei B WeIC Well D Seepage Wei E Face T ID E A) Well Location: Distance (m) 20 10 0 -7.3 -15.54 Beach Surface (cm) 32.7 25.7 1.7 -59.6 -127 Watertable Range (cm) 12 16 26 35.5 44.8 213 B) Individual water table measurements (cm) Time 9:35 -168.3 -151.5 -135 10:05 -166.8 -127 11:20 -145.5 -129 11:40 -163.3 12:18 -163.3 -159.3 -155.8 -144.5 -132.5 13:27 -164.3 -160.8 -159.3 -155 -153 14:22 -167.5 -165.3 -165.3 -162 -167.5 14:37 -201.5 15:22 -171.3 -169.3 -175.3 -168 -171.8 15:33 16:30 -175.3 -181.3 -173.5 <-179.5 16:37 -205 17:30 -178.5 <-179.5 18:00 -174.3 -173.8 -181.8 19:03 -175.3 -175.3 -180 < -179.5 4:02 -283 10:22 -122 15:37 -214 21:44 -70 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-7. Well hydrographs at Huntington Beach compared to the tide on 9/11/03. All water heights are relative to the beach surface at Well C. -100 24-Hour TIDE -100 -120 -200 -3001— 0:00 12:00 24:01 g* -140 o -160 O) '5 3 X -180 * » ‘ -nr— W ell A ■ *• » W ell D - r - W ell E — Tide -200 -220 18:00:00 10:00:00 14:00:00 Time 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-8. Cross section of wells at Huntington Beach shoreline. Vertical exaggeration is about 12:1. Well location demarked by letter across the top. Solid line is the beach surface. Maximum and minimum water table elevations at each well are plotted, and the dotted line outlines the tidal wedge defined by these points. The area between the wells is divided into sections, and the fractional volume of the tidal wedge in each section is indicated by the numbers. Open boxes represent the location of pore water samples. E o 4 - » .S> -1 0 0 C D X 15% 19% 25% 2 6 % " ° " - -200 10% 5% -10 30 20 10 0 -20 -30 Distance (m) 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cycle. The volume of the wedge was calculated by integrating between the maximum and minimum water table elevation (Figure 2-8). The volume of water that drains out of the sands relative to the total volume, the "effective porosity", was measured in the laboratory to be 0.10 ± 0.02. This value is much smaller than the true porosity (0.40 ± 0.02) because surface tension retains water in pore space. For the observed tidal cycle, the volume of water pumped through the beach was 1.4 ± 0.2 m3 (m shoreline)"1 . This value can then be normalized to the average tidal height of 1.1 m, assuming that the volume scales linearly with the tidal height. The volume can then be converted to a flux by multiplying by 1.95 tidal cycles d'1 . The average flux of water through the beach due to tides is 3.3 m3 d"1 (m shoreline)"1 . An estimate for the residence time of water in the tidal wedge, tb , can be made by constructing a simple mass balance for water in the beach: tb = QT p/VT P (2- 16) where QT P is the rate of water exchange and VT P volume of the sands flushed and the residence time in the beach, tb . The average tidal wedge volume is 6.9 m3. The tidal wedge volume is about a factor of 4 greater than the flux of water through the beach on a tidal cycle because of the difference between the effective porosity (0.1) and the total porosity (0.4). The estimated residence time for water in the beach is about 2.1 days. This estimate depends on two assumptions. First, the volume of water filled and drained during a tidal cycle is an isolated system (i.e., there is no mixing with 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. water below the low-tide water table). Second, an equal volume of water is exchanged on each tidal cycle. When the tides are mixed, the majority of the water drains out of the beach during the lowest-low tide. Violation of either of these assumptions would result in underestimating the residence time. Treating the beach water as a well-mixed system, a box model can be developed to estimate the residence time of water in the beach based on the Ra and 2 2 2 Rn pore water concentrations. Assuming the water entering the beach has a negligible concentration, then the only source of isotopes in the beach is from emanation. The loss of each isotope is by radioactive decay and removal from the beach. The mass- balance equation for each isotope (i) is then: Where (j)V b is the volume of pore space in the tidal wedge, and the other terms are as previously defined. The term on the left side computes the production rate of mobile Ra. On the right side, the first term is for decay, and the second term represents exchange. Equation 2-11 can then be rearranged to solve for tb : The location of beach water collected during the ebb tide at Well C and at the water table outcrop are plotted in Figure 2-8. The Ra concentration of these samples had a (2-17) 1 _________W jC p w 1 (2-18) A A ^ - ( ^ + ( l - 0 ) p ^ ) A ;C w 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. wide range of concentrations, and are presented, along with the computed age of the water, in Table 2-3. The equilibrium pore water activity was taken as the average pore water concentration collected deeper than 78 cm, and the measured Kd of 0.4 cm'3 g"1 was used. Samples had a wide range of computed ages, with the lowest ages for water collected within the top 48 cm. For 2 2 2 Rn, the low ages computed for the two upper-most samples may represent a loss of 2 2 2 Rn by gas exchange. Deeper samples have much longer residence times. The deep pore water concentrations used to estimate the equilibrium pore water activity may be less than equilibrium because of exchange, and thus these estimates represent an upper limit. The tidal pumping flux (JT P ) is the product of the volume of water exchanged and the average Ra concentration in that water (Table 2-6). For the volume of water exchange, the volume of sediments flushed by waves is subtracted from the total. For the concentration, the average pore water Ra data in the top 48 cm was used. The total Ra flux from the shoreline is the sum of the tidal pumping flux and the wave pumping flux: JS L =JT P +JW P . These values are computed and presented in Table 2-7. The total water exchange rate can also be computed by taking the sum of tidal pumping and wave pumping fluxes, and is about 13.3 m3 d'1 (m shoreline)'1 . This value is similar to an estimated water exchange rate between 7.8 and 11.6 m3 d'1 (m shoreline)'1 based on grain size and beach slope (McLachlan, 1982). 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The shoreline flux to the mixed layer depends on the equilibrium pore water activity, which was measured in beach sands south of San Pedro Bay (Table 2-2b). Relative to Huntington Beach, the equilibrium pore water activity at Dana Point was approximately 45% less for both isotopes. Assuming the shoreline water flux is the same, the shoreline flux is 45% less than at Huntington Beach. T id a l E x c h a n g e a t E s t u a r i e s : During the summer, surface freshwater inputs to San Pedro Bay are small, and exchange between rivers and the coastal ocean is dominated by tidal flushing. All of the processes described above that drive an advective flow through coastal ocean sediments also occur in estuaries and marshes. Instead of constraining each of these sources within each estuary, the net exchange of Ra at the mouth of each tidal inlet was computed. Three rivers and three tidal marshes exchange water with San Pedro Bay. At the south end of the bay is the Santa Ana River and the Talbert Marsh, and for each, the volume of water exchanged on a tidal cycle has been computed (Grant et al., 2003). In these systems, the surface area scaled with the same proportion as storage, and the ratio of storage to surface area reflects an average tide of about 1 m (Grant et al., 2003). Applying this surface area to volume ratio, the tidal exchange volume in Alamitos Bay, Anaheim Bay, Newport Bay, and the San Gabriel River was computed. The surface area of each of these was measured from USGS topographic maps. For Anaheim Bay, this estimate represents an upper limit because tidal flow 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-8. A) Summary of samples collected near the mouth of each marsh. B) Summary of Ra flux from tidal exchange with rivers and marshes. A) M areh Samples (Ebb TTdrt Station Collect Date and Tim e Sample Depth (m) “ 3 R a Concentration (dpm m'3 ) 3i4Ra 2I3Ra Anaheim Bay 5/23/03 7:45 0.2 55 ± 6 378 ± 11 San Gabriel River 5/23/03 8:30 0.2 36 ± 4 296 ± 9 61.6 ± 3.5 Talbert M areh 10/23/00 11:20* 0.2 550 ± 80 111.3 ± 29.0 Talbert M areh 1 2/3/00 1 6:05 0.2 215.01 24.0 Talbert M areh 3/28/02 10:00 0.2 58 ± 3 506 1 7 Talbert M areh 5/23/03 7:00 0.2 46 ± 6 333 ± 12 Santa Ana River 10/23/00 10:15 * 0.2 460 1 50 93.0 ± 20.6 Santa Ana River 12/3/00 15:20 0.2 183.5 ± 19.8 Average: 49 ± 12 421 * 101 132.9 * 64.1 B) Water Budget and R a Input Concentrations Exchange Rate Diet. Offehore Input Concentration (dpm m'3 ) Station (xIO8 m 3 day’) to Entrance (km) “ 3 R a “ < R a “ "Ra Alamltoe Bay 3.41 0.9 37 ± 4 247 1 25 6 0 1 6 San Gabriel River 0.52 0.3 47 ± 5 394 1 39 76 1 8 Anaheim Bey 8.02 0.9 37 ± 4 247 1 25 6 0 1 6 Talbert M areh 0.52 0 32 1 3 338 1 34 79 1 8 Santa Ana River 0.87 0 32 1 3 338 1 34 79+8 Newport Bay 10.1 0.35 22 1 2 206 1 21 66 1 7 C) R a Fluxea Station R a Flux (x108 atome aacf1 ): “ 3 Ra ««Ra "'Ra Alamltoe Bay 12 1 1 2 52 1 31 12451 1 11078 San Gabriel River 0.3 1 2 1 1 5 1489 1 1701 Anaheim Bay 28 1 28 122 1 73 29330 1 26096 Talbert M areh 2 1 2 4 1 5 1411 1 1702 Santa Ana River 4 1 3 6 1 8 2347 1 2832 Newport Bey 77 1 34 191 1 91 34160 1 32885 * For 224Ra, samples were processed and counted by gamma spectrometry following the method of Elsinger and coworkers (1982). 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. into the wildlife refuge is restricted by culverts and tide gates (CalDeptF&G, 1976). The total volume of water flushed through these estuaries is presented in Table 2-8. Water flowing into the estuary has an initial concentration, which can be estimated based on the average nearshore distribution of short-lived Ra isotopes computed in the next section. For estuaries that discharge into the surf zone, the average shoreline concentration was used (Table 2-8). For estuaries with discharge channeled through offshore jetties, the model presented in the next section was used to compute the concentration of water at the mouth of the jetty, up to 0.9 km offshore. The average concentration for samples collected at the mouth of each estuary during the ebb tide is in Table 2-8. The flux from each source was computed as the product of the water exchange rate and the difference in concentration between ebb and flood waters. The greatest source of uncertainty is the net concentration difference between the inflow and outflow, which was greatest for samples that discharge at the shoreline. S u r fa c e W a te r I n v e n t o r y : Ra at the shoreline and in the surface mixed layer was measured on 9 occasions at Huntington Beach, collected mostly during summer, and all during dry weather (Table 2-9B; Figure 2-9). Similarly, samples were also collected at Sunset Beach, at the north end of the bay, on 4 occasions, generally within 2 days of sampling at Huntington Beach (Table 2-9C; Figure 2-10). Sampling was done to cover a range of tide and wave conditions, as well as differences in local and regional circulation that may influence nearshore circulation. Assuming that 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-9. Short-lived Ra isotope concentrations from A) sites > 20 km offshore in Borderland basins, B) shoreline and mixed layer off Huntington Beach, and C) Sunset Beach. Values are organized based on collection date. Distance offshore is measured from the mean low low water mark. Uncertainty is 1 standard deviation based on the counting statistics. At BASIN STATIONS (>20 km ofhhorrt Mixed Layer Sample Concentration In dpm m's Collect Date. Tima Thick, Cm) Depth Cm) Station “ ■Ra M4Ra *«Ra 2 /2 6 /0 3 19:46 29.1 2.0 Santa Barbara Basin 0.6 ± 0.1 3.0 ± 0.2 7.7 ± 3.4 2 /2 5 /0 3 16:35 26.2 2.0 Santa Monica Basin 0.3 ± 0.1 1.9 ± 0.1 17.9 ± 6.1 3 /1 /0 3 0:15 2.0 Santa Monica Basin 0.3 ± 0.1 2.4 ± 0.2 8.7 ± 2.5 2 /2 4 /0 3 10:00 25.0 2.0 San Pedro Basin 0.9 ± 0.1 2.5 ± 0.2 18.2 ± 5.2 2 /2 4 /0 3 14:15 25.0 2.0 San Pedro Basin 0.7 ± 0.1 3.1 ± 0.1 19.5 ± 5.3 Average 0.6 ± 0.3 2.6 ± 0 .5 14^4 ± 5.7 Bt HUNTINGTON BEACH SHORELINE AND OFFSHORE TRANSECTS Collect Date, Time Mixed Layer Thick. On) Sample Depth Cm) Dletance Offehore flan) Concentration In dpm m'3 223*. M4Ra “ •Ra 1 0 /1 9 /0 0 16:05 0.2 0.00 89.5 ± 15.9 1 0 /1 9 /0 0 17:25 0.2 0.00 111.4 ± 23.7 1 0 /1 9 /0 0 18:45 0.2 0.00 197.1 ± 35.7 1 0 /1 9 /0 0 8:45 0.5 0.78 84.5 ± 8.4 1 0 /1 9 /0 0 10:00 0.5 0.71 107.7 ± 10.2 1 0 /1 9 /0 0 10:45 0.5 0.70 101.6 ± 9.5 1 2 /1 /0 0 14:10 0.2 0.00 168.9 ± 19.5 1 2 /1 /0 0 1 5:00 0.2 0.00 152.9 ± 17.9 1 2 /1 /0 0 15:55 0.2 0.00 278.6 ± 29.2 1 2 /1 /0 0 9:20 0.5 0.74 116.9 ± 14.9 1 2 /1 /0 0 10:20 0.5 0.65 108.5 ± 12.5 1 2 /1 /0 0 11:05 0.5 2.43 77.5 ±11.0 9 /1 3 /0 1 16:53 8.5 1.5 1.35 8.3 ± 2.0 83.5 ± 5.1 52.4 ± 6.6 9 /1 3 /0 1 14:50 >10 1.5 2.97 1.3 ± 0.5 13.9 ± 1.4 32.0 ± 8.0 9 /1 3 /0 1 12:49 >12.1 1.5 9.86 0.1 ± 0.1 0.6 ± 0.1 6.1 ± 4.7 9 /1 3 /0 1 9:16 >12.1 1.5 19.98 0.1 ± 0.1 0.4 ± 0.1 1.2 ± 6.6 1 0 /16/01 9:12 0 0 12.0 ± 0.9 150.1 ± 3.3 1 0/1 6 /0 1 11:10 0 0 12.5 ± 0.8 187.2 ± 3.6 1 0/1 6 /0 1 13:15 0 0 34.7 ± 4.7 290.2 ± 10.6 1 0/1 6 /0 1 15:15 0 0 48.6 ± 4.9 446.9 ±11.5 1 0/1 6 /0 1 17:20 0 0 45.6 ± 4.4 399.0 ± 10.2 4 /3 /0 2 9:45 1.5 1.67 22.9 ± 1.6 258.7 ± 5.4 80.7 ± 13.8 4 /3 /0 2 12:15 1.5 4.41 11.1 ± 1.4 86.5 ± 3.4 38.4 ± 9.9 4 /3 /0 2 14:15 1.5 7.48 0.9 ± 0.2 3.8 ± 0.3 10.0 ± 6.9 4 /3 /0 2 1 5:45 1.5 11.89 2.3 ± 0.3 9.6 ± 0.6 7.9 ± 7.0 6 /7 /0 2 9:00 0 0 32.3 ± 6.1 264.6 ± 15.4 152.8 ± 3.5 6 /7 /0 2 10:45 0 0 48.7 ± 6.5 423.7 ± 17.4 154.3 ± 3.6 6 /7 /0 2 12:40 0 0 50.7 ± 6.3 349.5 ± 14.5 140.3 ± 3.0 6 /7 /0 2 14:35 0 0 65.1 ± 9.4 635.8 ± 26.4 6 /7 /0 2 16:35 0 0 25.5 ± 5.4 324.4 ± 17.1 99.7 ± 2.4 6 /7 /0 2 10:05 7.0 1.5 1.08 13.5 ± 1.0 132.2 ± 3.0 6 /7 /0 2 12:30 >1 5 1.5 3.77 0.6 ± 0.2 5.4 ± 0.6 6 /7 /0 2 14:20 20.0 1.5 7.09 0.2 ± 0.1 0.7 ± 0.1 6 /7 /0 2 16:50 10.0 1.5 12.48 0.1 ± 0.1 0.4 ± 0.1 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-9 Continued. CoDtct Data, Tima Mbcad Layar Thick. On) Sampla Daoth On) Dlstanca Offshora (km) Concantratkm In dpm m‘* zzsRa 224Ra 2MRa 6 /1 2 /0 2 9:25 0 0 29.1 ± 3.9 369.9 ± 14.1 69.3 ± 2.0 6 /1 2 /0 2 11:05 0 0 36.6 ± 3.9 403.4 ± 13.6 78.2 ± 2.3 6 /1 2 /0 2 13:05 0 0 35.8 ± 3.5 360.9 ± 10.9 63.7 ± 2.1 6 /1 2 /0 2 15:05 0 0 38.0 ± 3.7 430.2 ± 13.0 79.5 ± 1.9 6 /1 2 /0 2 17:00 0 0 41.8 ± 4.0 444.8 ± 13.7 81.7 ± 2.0 6 /1 2 /0 2 10:00 9.1 1.5 0.99 12.7 ± 1.1 147.5 ± 3.6 6 /1 2 /0 2 12:00 >17 1.5 3.77 0.6 ± 0.1 5.0 ± 0.3 14.2 ± 6.6 6 /1 2 /0 2 14:20 22.0 1.5 6.12 0.2 ± 0.1 0.9 ± 0.1 6 /1 2 /0 2 16:45 >1 5 1.5 10.97 0.1 ± 0.3 0.9 ± 0.2 7 /2 6 /0 2 7:50 0 0 49.2 ± 3.8 553.8 ± 13.1 7 /2 6 /0 2 9:50 0 0 33.6 ± 2.9 389.8 ± 10.8 7 /2 6 /0 2 11:55 0 0 36.0 ± 3.0 361.0 ± 9.3 7 /2 6 /0 2 14:05 0 0 65.7 ± 6.3 726.4 ± 21.2 7 /2 6 /0 2 15:55 0 0 30.7 ± 2.4 394.0 ± 8.7 7 /2 6 /0 2 13:20 8.0 1.5 0.80 10.6 ± 0.9 78.1 ± 2.7 7 /2 6 /0 2 14:10 5.0 1.5 2.90 11.9 ± 0.9 80.2 ± 2.6 8 /1 9 /0 2 9:30 0 0 21.2 ± 2.8 202.8 ± 7.7 70.1 ± 16.2 8 /1 9 /0 2 11:20 0 0 22.8 ± 2.7 227.8 ± 7.6 77.0 ± 16.2 8 /1 9 /0 2 13:10 0 0 28.4 ± 4.1 241.0 ± 10.5 79.8 ± 17.1 8 /1 9 /0 2 15:00 0 0 36.6 ± 4.0 321.4 ± 10.6 69.9 ± 15.2 8 /1 9 /0 2 17:30 0 0 22.0 ± 2.3 210.9 ± 6.5 62.5 ± 15.7 8 /1 9 /0 2 14:30 10.0 1.5 0.68 12.6 ± 1.7 94.2 ± 3.9 57.9 ± 10.9 8 /1 9 /0 2 13:20 18.0 1.5 2.96 0.2 ± 0.1 2.4 ± 0.3 7.2 ± 5.6 8 /1 9 /0 2 11:40 14.0 1.5 6.09 0.1 ± 0.1 1.0 ± 0.3 13.8 ± 6.9 8 /1 9 /0 2 9:35 10.0 1.5 9.31 0.2 ± 0.1 0.8 ± 0.1 20.1 ± 7.5 8 /2 3 /0 2 8:00 0 0 26.5 ± 2.2 303.3 ± 7.0 73.6 ± 20.1 8 /2 3 /0 2 1 1:50 0 0 11.8 ± 1.4 183.4 ± 5.3 83.8 ± 21.9 8 /2 3 /0 2 14:00 0 0 23.6 ± 1.9 264.8 ± 6.5 90.4 ± 23.1 8 /2 3 /0 2 15:55 0 0 37.1 ± 3.2 380.0 ± 9.2 122.3 ± 25.2 8 /2 3 /0 2 17:50 0 0 22.7 ± 1.7 332.5 ± 6.9 109.7 ± 22.7 8 /2 3 /0 2 14:10 10.0 1.5 0.67 5.8 ± 0.8 51.9 ± 2.1 47.5 ±11.2 8 /2 3 /0 2 13:00 11.0 1.5 2.96 3.3 ± 0.3 19.1 ± 0.6 39.1 ± 10.0 8 /2 3 /0 2 1 1:20 13.0 1.5 6.09 0.2 ± 0.1 1.3 ± 0.2 9.4 ± 5.9 8 /2 3 /0 2 9:45 11.0 1.5 8.58 0.3 ± 0.1 0.8 ± 0.1 0.0 ± 0.5 9 /6 /0 2 7:35 0 0 30.9 ± 2.9 247.3 ± 6.4 9 /6 /0 2 9:30 0 0 28.9 ± 2.5 273.9 ± 6.0 9 /6 /0 2 11:30 0 0 27.9 ± 2.2 253.6 ± 5.4 9 /6 /0 2 13:30 0 0 30.0 ± 2.0 334.6 ± 5.6 9 /6 /0 2 15:30 0 0 25.6 ± 2.5 287.3 ± 7.9 6 /1 8 /0 3 12:00 0 0 19.6 ± 3.8 168.4 ± 9.4 60.9 ± 21.5 6 /1 8 /0 3 14:05 0 0 21.0 ± 2.7 270.4 ± 8.4 6 /1 8 /0 3 10:20 >10 1.5 0.94 9.2 ± 0.5 64.2 ± 1.1 53.3 ± 9.6 6 /1 8 /0 3 12:30 14.0 1.5 3.28 1.3 ± 0.1 3.9 ± 0.2 6 /1 8 /0 3 14:35 6.0 1.5 5.1 5 0.7 ± 0.1 2.0 ± 0.1 25.3 ± 5.6 6 /2 5 /0 3 10:30 0 0 26.0 ± 3.0 243.3 ± 7.5 108.2 ± 22.6 6 /2 5 /0 3 12:30 0 0 28.6 ± 2.7 373.3 ± 8.1 113.1 ± 23.6 6 /2 5 /0 3 14:15 0 0 25.3 ± 2.6 255.9 ± 6.8 6 /2 5 /0 3 16:15 0 0 31.1 ± 2.7 370.6 ± 7.9 6 /2 5 /0 3 17:35 0 0 24.7 ± 3.6 242.4 ± 9.7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-9 Continued. Mixed Layar Sampla Distance Concentration In dpm m's Collect Date. Tima Thick. On) Daoth Cm) Offshore flan) M8Ra M4Ra z“ Ra 6 /2 5 /0 3 9:30 8.0 1.5 1.10 10.5 ± 0.8 77.3 ± 1.7 59.6 ± 13.5 6 /2 5 /0 3 11:10 9.0 1.5 3.29 6.9 ± 0.6 33.6 ± 1.0 6 /2 5 /0 3 13:10 10.5 1.5 5.16 0.8 ± 0.1 4.5 ± 0.2 8 /2 8 /0 3 11:10 0 0 41.5 ± 4.1 416.6 ±11.8 8 /2 8 /0 3 15:45 0 0 18.4 ± 1.5 237.3 ± 4.5 8 /2 8 /0 3 18:15 0 0 34.1 ± 2.3 392.8 ± 7.3 Q OTHER SHORELINE AND OFFSHORE TRANSECTS Mixed Layer Sample Distance Concentration In dpm m'8 Collect Date. Time Thick, (m) Depth Oh) Offshore flan) «*Ra 224Ra 22® R a Sunset Beach 9 /1 1 /0 1 11:10 7 1.5 2.20 0.5 ± 0.6 6.0 ± 1.5 35.2 ± 1.7 9 /1 1 /0 1 13:53 7 1.5 5.50 11.6 ± 2.1 31.7 ± 2.6 34.9 ± 1.8 9/1 1 /0 1 17:47 6 1.5 9.10 0.8 ± 0.2 2.2 ± 0.3 25.9 ± 2.1 9 /1 3 /0 1 9:16 >12.1 1.5 14.00 0.1 ± 0.1 0.4 ± 0.1 Sunset Beach 9/2 0 /0 1 12:55 9 1.5 1.70 18.4 ± 2.7 114.1 ± 5.0 61.0 ± 1.4 9/2 0 /0 1 10:51 11 1.5 4.90 10.9 ± 2.7 42.6 ± 4.3 44.5 ± 1.6 9 /2 0 /0 1 8:44 >8.6 1.5 9.10 2.5 ± 0.3 11.6 ± 0.5 17.8 ± 0.7 Belmont Shore 9 /2 5 /0 1 8:35 0 0 23.3 ± 3.7 256.0 ± 10.1 65.3 ± 2.6 9/2 5 /0 1 10:40 0 0 52.0 ± 8.1 345.0 ± 16.3 90.2 ± 2.9 9 /2 5 /0 1 12:35 0 0 48.7 ± 4.2 327.1 ± 8.3 88.9 ± 2.1 9 /2 5 /0 1 14:35 0 0 61.9 ± 7.4 277.1 ± 1 1.7 86.4 ± 2.0 9 /2 5 /0 1 16:35 0 0 32.4 ± 4.1 278.1 ± 9.2 9 /2 5 /0 1 18:35 0 0 36.5 ± 4.5 290.7 ± 9.8 90.8 ± 4.3 Sunset Beach 6 /5 /0 2 8:40 0 0 55.0 ± 5.7 457.2 ± 12.9 6 /5 /0 2 10:45 0 0 58.7 ± 5.0 474.4 ±11.4 6 /5 /0 2 12:40 0 0 76.0 ± 10.5 485.0 ± 20.8 6 /5 /0 2 14:35 0 0 95.5 ± 10.7 671.3 ± 22.6 Sunset Beach 6 /5 /0 2 15:55 12 1.5 1.9 39.2 ± 6.0 244.6 ±11.8 6 /5 /0 2 13:45 >12 1.5 2.89 23.1 ± 1.0 52.1 ± 1.2 6 /5 /0 2 10:30 12 1.5 7.28 25.8 ± 1.4 259.2 ± 3.7 Sunset Beach 7 /2 6 /0 2 8:00 0 0 46.8 ± 3.4 805.4 ± 19.8 130.8 ± 5.3 7 /2 6 /0 2 9:50 0 0 49.5 ± 3.8 755.1 ± 21.3 122.2 ± 4.8 7 /2 6 /0 2 11:50 0 0 44.9 ± 2.9 632.0 ± 15.1 96.2 ± 3.7 7 /2 6 /0 2 13:50 0 0 64.8 ± 5.0 804.9 ± 23.0 140.8 ± 7.7 7 /2 6 /0 2 15:50 0 0 70.5 ± 3.3 637.4 ± 13.4 108.7 ± 5.9 7 /2 6 /0 2 15:30 4 1.5 0.72 23.3 ± 2.4 219.3 ± 7.0 66.9 ± 2.3 Newport Beech 7 /2 6 /0 2 8:30 0 0 20.5 ± 2.0 219.1 ± 6.7 7 /2 6 /0 2 10:20 0 0 19.4 ± 1.1 223.1 ± 5.0 7 /2 6 /0 2 12:15 0 0 17.5 ± 1.9 185.0 ± 5.8 7 /2 6 /0 2 14:10 0 0 21.7 ± 2.0 185.0 ± 5.9 7 /2 6 /0 2 1 5:55 0 0 23.9 ± 2.5 222.2 ± 8.3 7 /2 6 /0 2 11:20 1.5 0.4 10.5 ± 1.0 90.0 ± 2.9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. these profiles capture most of the variability in the system, then this data set can be used to compute the average nearshore distribution of short-lived Ra. To compute the inventory of Ra in the mixed layer, the average concentration was computed at three distances where sampling was clustered. Between these points, the concentration was linearly extrapolated. The short-lived Ra isotope inventory (I in atom s'1 (m shoreline)'1 ) in surface waters was calculated by integrating the concentration offshore, assuming the mixed layer concentration is independent of depth and taking into account the increasing depth of water with distance offshore: c o H m I f X C d z d x (2-19) 0 0 where Hm , the mixed layer thickness, is a function of x and constrained by water depth. An average Hr a of 11.8 m was used for both the Huntington Beach and Sunset Beach transects. The inventory for each isotope is presented in Table 2-10. Based on the standard deviation of the mean at the first station offshore, where the greatest change in concentration is observed, the uncertainty for this calculation is 25%. Near Sunset Beach, shoreline and seafloor sources of Ra are augmented by exchange with the Los Angeles/Long Beach Harbor and inputs from the Los Angeles and San Gabriel Rivers, Alamitos Bay and Anaheim Bay. These latter sources are difficult to evaluate but sufficient to produce an inventory in surface water at Sunset Beach that is 3 to 4 times greater than observed at Huntington Beach. 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without perm issio n . Table 2-10. San Pedro Bay box model summary. A) N P irrS F O R SAN P E D R O B A Y 2 2 3 Ra 224 Ra 2 28 Ra 224 Raf 223 Ra Flux Ratio Fraction o f Inputs 2 2 3 Ra 224 Ra 2 2 8 Ra SH ORBJNE F L U X : Wave Flux (x1000 atoms s '1 (m shore)"1): 0.06 + 0.02 0.7 ± 0.2 0.6 ± 0.2 12.2 1% 2% 0.1% Tide Flux (x 1000 atoms s '1 (m shore)"1 ): 0.36 ± 0.11 1.9 ± 0.5 2.5 ± 0.4* 5.2 5% 5% 0.4% S E A F L O O R F L U X : Ra Flux (xIOOO atoms s"1 (m shore)"1): 6.3 ± 3.1 31 ± 16 367 ± 184* 5.0 89% 91% 65% TVAL M A R S H E S AND R IV E R S F L U X : Talbert Marsh (xl 06 atoms s '1): 2.5 ± 1.8 4 ± 5 1411 ± 1702 Santa Ana River (xl 0 6 atoms s"1): 4.1 ± 3.0 6 ± 8 2347 ± 2832 Length of Coastline (m) 19000 19000 19000 Ra Flux (xIOOO atoms s"1 (m shore)'1): 0.34 ± 0.18 0.53 ± 0.50 198 ± 174 1.5 5% 2% 35% Box Model Input Fluxes Sum: 7.1 ± 3.2 35 + 16 568 ± 253 4.9 100% 100% 100% B) L O S S E S Aw . Measured tnwntory (xIOOO atoms sec’1 (m shore)'1 ) Sunset Beach Transects 26 ± 7 140 ± 35 5.4 Huntington Beach Transects 5.1 ± 1.3 42 ± 11 8.3 C) Total Inputs to Newport Bay Box** Input Flux (xl 000 atom s s '1 (m shore)"1 ): 7.5 ± 2.5 21 + 6 2896 ± 2740 2.8 * Lower limits ** Computed from box model in Table 2-11. o Figure 2-9. Composite surface water data collected offshore from Huntington Beach. Solid line represents linear extrapolation used to compute the surface water inventory. Dashed line represents exponential fit to the data. Exponential equation and fit parameters are shown in box. 100 223 C(x) = C0e_ x /a C0 = 30.5 dpm m a = 1.11 km 'a a a , -o 0 1 2 3 4 5 6 Distance Offshore (km) 800 224 700 C(x) = C0e'x/a C0 = 335 dpm m a = 0.847 km 600 \ 500 B £ 400 i 3 0 0 200 100 0 1 3 4 6 5 Distance Offshore (km) 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2-10. Composite surface water data collected offshore from Sunset Beach. Solid line represents linear extrapolation used to compute the surface water inventory. Dashed line represents exponential fit to the data. Exponential equation and fit parameters are shown in box. 100 223 C(x) = C0em X la CG = 53.6 dpm m a = 2.38 km « ? E E a T3 R > D C CO M CS 0 1 2 3 4 5 6 Distance Offshore (km) 800 224 700 C(x) = C0e'x/a Cc = 498 dpm m a = 1.29 km * ? E 600 a 500 2 400 a " 300 200 100 0 1 2 3 4 5 6 Distance Offshore (km) 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R a d i u m - 2 2 8 : For the short-lived Ra isotopes, the emanation rate appears to reflect the total sediment activity. Since 2 2 4 Ra and 2 2 8 Ra share a common radiogenic source, 2 3 2 Th, their emanation rate should also be similar. But while the 2 2 4 Ra and 2 2 8 Ra total sediment activities were equal, the emanation rate for 2 2 8 Ra is expected to be less than that for 2 2 4 Ra. First, with each successive alpha decay, an atom is dislodged from the mineral structure, which damages the crystal lattice and makes it easier for subsequent decays to escape from a mineral. Thus, it should be easier for 2 2 4 Ra to escape than 2 2 8 Ra. Second, with a smaller alpha decay energy for 2 3 2 Th than 2 2 8 Th (4.012 MeV vs. 5.423 MeV), the recoil distance for 2 2 8 Ra is proportionally less for 2 2 4 Ra. Therefore, the emanation efficiency for 2 2 8 Ra is expected to be less than that for 2 2 4 Ra. Assuming that the emanation rate is 20% lower for 2 2 8 Ra than for 2 2 4 Ra, the calculated Ra fluxes can be repeated for 2 2 8 Ra (Table 2-10). The 2 2 8 Ra estuary flux and surface water inventory at Sunset Beach and Huntington Beach were computed in the same way as the short-lived isotopes. At the seafloor, the exchange rate is so fast in Zone 1 that almost all of the Ra produced is flushed into the overlying water column. But the assumption of a diffusive system in Zone 2 may not be valid for 2 2 8 Ra. With a significantly longer half-life, 2 2 8 Ra is sensitive to non-local exchange or to advection rates much slower than those that impact the distribution of the short lived isotopes. As a result, the 2 2 8 Ra flux from the Zone 2 may be underestimated, and the total 2 2 8 Ra seafloor flux is assumed to be a lower limit. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Similarly, the residence time of water in the tidal wedge at the beach is sufficiently short that essentially all of the 2 2 8 Ra produced is flushed out. The zone below the tidal wedge is also more sensitive than the short-lived Ra isotopes to slow rates of advection, which may increase its flux. Again, ignoring these effects results in a lower limit flux estimate. The integrated activity of 2 2 8 Ra was not computed. Because of the long half-life for 2 2 8 Ra, it can be transported far beyond our sampling region before it decays. In addition, upwelling may be an important source of 2 2 8 Ra on the shelf (Moore, 1987). As a result, the mixed layer inventory may not originate solely from estuaries, the shoreline, and the seafloor in contact with the mixed layer. Box Model A box model for surface waters in San Pedro Bay was constructed to compute the short-lived Ra isotope mass balance. The box covers 19 km of shoreline, extending from Sunset Beach to Newport Beach (Figure 2-1). The model extends offshore far enough to capture all of the excess short-lived Ra in offshore transects, approximately 10 km. The mixed layer receives short-lived Ra from three sources: tidal exchange at rivers and marshes (JT E ), the shoreline (JS L ), and the seafloor (JS F ). Ra is lost from surface waters by radioactive decay (JD e c a y ). Longshore advection acts as a source (J,„) and a sink (J0 u t) for Ra. At steady-state, the mass balance equation for the box is: 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ^O ut JD eca y (2-14) Vertical mixing and particulate removal of Ra are two additional mechanisms that may lead to the loss of Ra from the mixed layer. However, relative to radioactive decay, their contribution to the overall budget is negligible. To show this, each rate is compared to the rate of radioactive decay. For a given box of water that is well mixed, the radioactive decay flux is: where Hm is the mixed layer depth, A is the map-view area, and C is the average concentration. The combined terms, XHm , represents the rate of radium loss, which for the short-lived isotopes is about 10'5 m s'1 . For vertical mixing, the flux out of the mixed layer (Jv e r t) can be expressed as: where Kz , the vertical mixing rate (m2 s'1 ), is a function of the buoyancy gradient, which is dominated by the water column thermal structure. Using a conservative estimate for the temperature gradient at the base of the mixed layer (2°C over 5 m), an upper bound for Kz is 10'5 m2 s'1 (Gargett, 1984). The concentration difference between the mixed layer and deep water is about 50% over a 5 m distance. Thus, the loss rate by vertical mixing for a given parcel of water is at most 10'6 m s'1 , an order of magnitude smaller than the radioactive decay flux. (2-15) (2-16) 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Particulate removal may strip Ra from the mixed layer. Using Equation 2-15, the loss rate by radioactive decay for the long-lived 2 2 6 Ra isotope is 10'1 1 m s'1 . With a decay rate that is negligible compared to the vertical mixing rate, the vertical distribution of 2 2 S Ra is expected to be homogeneous. If particulate removal is faster than the vertical mixing rate, a vertical 2 2 6 Ra gradient then would be generated. However, there is no 2 2 6 Ra concentration gradient, horizontally or vertically, in this region (Chapter 4). Therefore, the rate of Ra removal by particulates is less than the vertical mixing rate and is negligible for the short-lived Ra budget. Returning to Equation 2-14, three of the input fluxes have been computed at Huntington Beach: JT E , JS L , and JS F (Table 2-10A). For San Pedro Bay, the primary source of short-lived Ra is from the seafloor, while the shoreline and tidal exchange with the Talbert Marsh and Santa Ana River represent minor sources. A similar summary of inputs can be made for Newport Beach, just south of San Pedro Bay (Table 2-11). For this budget, estuarine inputs were averaged over 12 km, twice the distance between Newport Bay and the San Pedro Bay boundary at Newport Canyon. Ra inputs from the shoreline and seafloor are significantly reduced south of Newport Canyon because of the decrease in the equilibrium pore water activity, a function of the emanation rate, and a steeper seafloor. Whereas these sources are the primary sources of Ra to San Pedro Bay, they account for less than 25% of the budget in the region south of the bay. Instead, the primary source of Ra is Newport Bay. 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Table 2-11. Estimate of inputs to the region south of San Pedro Bay. 223 Ra 2 2 4 Ra 224 223 Ra R a /2 2 3 Ra Flux Ratio Fraction o f Inputs 223 Ra 2 24 Ra 22 8 Ra S H O R E L IN E FL UX: Wave Flux (atom s sec’1 (m sh o re )1): 27 ± 5 327 ± 35 262 ± 79 12.2 0% 2% 0.01% Tide Flux (atom s sec’1 (m shore)’1): 163 ± 49 855 ± 256 1145 ± 191 5.2 2% 4% 0.04% SEAF LOOR FLUX: Ra Flux (atoms sec’1 (m shore)’1): 826 ± 118 4126 ± 589 48212 ± 6887 5.0 11% 19% 2% TDAL MARSHES AND R IV E R S FLUX: Newport Bay Length of Shoreline (km): Ra Flux (106 atom s sec'1): Ra Flux (atoms sec’1 (m shore)’1): 12 77 ± 34 6435 ± 2539 12 191 ± 9 1 15916 ± 6121 12 34160 ± 32885 2 8 46666 ± 2740386 2.5 86% 75% 98% Box Model Input Fluxes Sum: (atom s sec’1 (m shore)’1) 7451 ± 2542 21223 + 6155 2 8 96285 ± 2740395 2.8 100% 100% 100% o 0 0 Returning to Huntington Beach, the loss term JD e c a y in Equation 2-14 is equal to the short-lived Ra isotope inventory measured in surface water (2-10B). Coincidentally, given the uncertainty of the individual fluxes, the sum of the three inputs (tidal exchange with estuaries, shoreline, and seafloor) is equivalent to the inventory of Ra observed in surface water for each short-lived Ra isotope. This suggests that the net effect of the longshore advective fluxes (JI n - Jo u t) is negligible. Fortunately, this can be tested. The longshore advection flux depends on the longshore advection rate u (m s'1 ), which averages between 3 and 10 cm s'1 (Hickey, 1993; KOMEX et al., 2003; MCB, 2002; Noble et al., 2003). The residence time of water in San Pedro Bay can be estimated by dividing the length of San Pedro Bay shoreline (19 km) by the advection rate. Using this range of values, the residence time of water is between 2.2 and 7.3 days. If advection is strictly parallel to the shoreline and independent of distance offshore, continuity requires that uin = u0 U t = u, so that: Jm~ Jout = C ,n — C qui) = J TE + JsL + JsF~ J D e c a y (2-17) where C is the average Ra concentration and AL is the cross-sectional area (mixed layer depth x distance offshore) of exchange. The average Ra concentration C can be calculated based on the observed inventory I in the surface mixed layer: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. During the summer of 2001, an average nearshore advection rate of 5 cm s'1 towards the south was measured at Huntington Beach (Noble et al., 2003). Since flow is towards the south, C I n can be computed from the inventory measured at Sunset Beach and C 0 u t from the inventory measured at Huntington Beach. Given these conditions, the net longshore advective flux for each isotope far greater than the near-zero value required to balance the Ra budgets (Table 2-10B). This is caused by the large inventory in water measured upstream at Sunset Beach and the short residence time, relative to the isotope half-lives, of water in San Pedro Bay. Therefore, Sunset Beach cannot be the only source of water, and Ra, to San Pedro Bay. To balance the short-lived Ra isotope budgets requires a source of low-Ra water. The only possible source of low-Ra water at the north end of San Pedro Bay is water moving onshore. Assuming the water moving onshore has no excess Ra, the fraction of offshore water required to balance the budget for a range of advection rates can be estimated. Combining Equation 2-17 and Equation 2-14, and rearranging to solve for the longshore advection rate at Huntington Beach, Q: Q _ J D e c a y T M + ^ S L + / s f ) (2-19) f C In ~ C Bay Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Where f is the longshore fraction of water that enters the box, and 1-f is the offshore fraction. The concentration in the bay and for water entering the bay can be calculated using Equation 2-18. Combining these equations along with Equation 2-15, the longshore fraction can be computed as: Using the computed averages, a greater fraction of water from the north is required to balance the 2 2 4 Ra budget than the 2 2 3 Ra budget. However, the computed JD e c a y and J*D e c a y fluxes can be varied within their uncertainty to produce consistent results by lowering the 2 2 4 Ra:2 2 3 Ra inventory ratio (Table 2- 12). Alternatively, the short-lived Ra activity in water flowing onshore may not be zero, and have a high 2 2 4 Ra/2 2 3 Ra. However, this is counter-intuitive since the isotope with the longer half-life, 2 2 3 Ra, would be expected to have the higher inventory in offshore water. Regardless of which scenario is correct, the fraction of offshore water required to account for the observed profiles is between 65 and 90%. Changes in this fraction of water would lead to changes in the Ra inventory in surface waters, which is further discussed in Chapter 3. The calculated fraction is relatively insensitive to advection rates greater than 2 cm s'1 . / = D e ca y (2-20) All of the variables on the right hand side have been computed for each isotope. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2-12. Computed fraction of water from Sunset Beach that is advected down the coast to Huntington Beach. Advection (cm sec'1 ) 223 Ra (%) 224 Ra (%) l 10 ± 10 53 ± 47 2 15 ± 10 42 ± 28 3 16 ± 9 38 ± 22 4 17 ± 8 36 ± 19 5 18 ± 8 35 ± 17 6 18 ± 7 34 ± 15 7 1 8 + 7 34 ± 14 8 1 9 + 7 33 ± 14 9 19 ± 7 33 ± 13 10 19 ± 7 33 ± 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Figure 2-11: Average summer circulation in San Pedro Bay, CA. Arrows only represent direction of mean current, not the current speed. Open squares are bouy stations: NOAA Outer Los Angeles Harbor Station 9410660 and Scripps San Pedro Buoy 092. P o rt o f L o s A n g e le s / L o n g B e a c h P a lo s V e rd e s P e n n in s u la 118.3 °W 118.1 117.9 Figure 2-12. Surface water temperature measurements from September 2001. Sunset Beach and Huntington Beach surface water measurements compared to NOAA's LA Outer Harbor station and Scripps San Pedro Bay buoy. U o s e i s- 4) a S s 22 LA Outer Harbor (NOAA. Station) San Pedro Bay (Scripps Buoy) 21 A Samples 20 19 U A 18 17 16 C /i 15 Sep/1 Sep/6 Sep/ll Sep/16 Sep/21 Date (Local Time) Sep/26 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. There is independent evidence for an onshore flow at the north end of San Pedro Bay. During the summer, mean flow through San Pedro Channel (Figure 2-11) is equatorward (Hickey, 1993). Assuming geostrophic flow, this requires a temperature increase offshore from Palos Verdes. As the current moves south, the cold, high density water that was at the shoreline along Palos Verdes can spread into San Pedro Bay. This produces an onshore flow of water into San Pedro Bay. The north part of the bay is then isolated by the formation of a cyclonic, cold-core eddy. Evidence for this are water temperatures in the outer Harbor, which are generally 1-2 °C colder than observed offshore at Huntington Beach (Figure 2-12). In addition, comparing samples collected just a few days apart, nearshore water temperatures at Sunset Beach were often 1-2°C colder than at Huntington Beach. Offshore from Huntington Beach, mean flow is southeast and parallel to the shoreline at a mean rate of 5 cm s'1 that increases to 10 cm s'1 at the shelf break (Noble et al., 2003). Therefore, it is reasonable that during the summer, between Sunset Beach and Huntington Beach, a mean onshore flow transports low-Ra water into the nearshore. Conclusions The processes that introduce short-lived Ra isotopes into the coastal zone of San Pedro Bay have been evaluated. About 90% of the total input flux for both isotopes was from the seafloor, where waves and interactions between currents and seafloor topography pump water through the seafloor. The remainder of the inputs were balanced by inputs from estuaries and shoreline sources. 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Along with the Ra input fluxes, the flux of water through sediments was computed. At about 8 m water depth, the water exchange rate was 27 L m'2 d"1 of water pumped through the top 28 cm. With 1.2 km of seafloor in contact with the mixed layer, the total flux of water through seafloor sediments is 32 m3 d"'(m shoreline)"1 . The flux of water through the beach from tides was found to be 3.3 m3 d'’(m shoreline)"1 , and the flux of water from waves was about 10 m3 d_ 1 (m shoreline)'1 . Combining these fluxes, the total flux of mixed-layer water through the beach and seafloor is 46 m3 d " ‘(m shoreline)"1 . The distribution of Ra sources to the mixed layer is not constant along the shoreline, with large differences to the north and south of Huntington Beach. First, large estuaries north and south of Huntington Beach are important sources of Ra in these regions. Second, there is a decrease in the equilibrium pore water Ra concentration of beach sands from north to south, with a significant drop in concentration south of Newport Canyon. Lower concentrations decrease the Ra flux from both the shoreline and the seafloor. In addition, south of Newport Canyon the seafloor is much steeper, limiting the seafloor Ra input to the mixed layer. For 2 2 4 Ra, the inventory calculated in the surface mixed layer of San Pedro Bay is balanced by the computed inputs. For 2 2 3 Ra, however, the inputs exceeded the inventory. To account for the low 2 2 3 Ra inventory, a fraction of the water entering San Pedro Bay must be low-Ra water moving onshore. Based on other studies at Huntington Beach, flow during summer is generally from the north. But to balance 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the Ra budget, 10 to 35% of the water can flow down the coast from Sunset Beach. The remainder of the water must be from offshore. Variations in the fraction of offshore water are likely to create changes in the Ra surface water inventories at Huntington Beach. Finally, assessing the sources of water and Ra into San Pedro Bay has implications for models that use the mixed layer distribution of 2 2 3 Ra and 2 2 4 Ra to estimate cross shelf mixing rates. First, because of the scale of the study, what happens upstream is critical to the Ra distribution at Huntington Beach. These include greater inputs into nearshore water and an onshore flow of low Ra water. Relative changes between these two sources of water lead to changes in the inventory as well as the nearshore distribution. Before more accurate models can be constructed, the source water to San Pedro Bay must be better characterized. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 3. Calculating cross-shelf mixing rates using short-lived Ra isotopes Abstract To predict the distribution of nutrients and contaminants that enter at the shoreline from estuaries, groundwater or other sources, nearshore mixing rates must be constrained. One potential method to measure the nearshore mixing rate is based on the distribution of short-lived Ra isotopes, 2 2 3 Ra and 2 2 4 Ra. Concentrations were measured at the shoreline and in offshore transects along the coast of San Pedro Bay, CA. Shoreline samples collected in knee-deep water, every two hours for a total of 10 hours, showed significant variability in concentration throughout each sampling day. Most of the variability can be explained by tidal variations in the shoreline input flux. Variability at shorter time scales must reflect fluctuations in the nearshore mixing rate. Surface water Ra concentrations measured in transects perpendicular to the shoreline decreased sharply offshore, with concentration reduced by 50% within 1 km offshore. Integrating the mixed layer concentrations gave the Ra inventory, which varied by about 20% on different sampling days. Using the input fluxes presented in Chapter 2, a 2-D model was developed to compute eddy diffusivity and examine the importance of various sources of water to the Ra distribution at Huntington Beach. The best fit to the composite summertime 2 2 3 Ra and 2 2 4 Ra data indicated an eddy diffusivity of 1.3 ± 0.2 m2 s"1 . Since inputs are greater than the observed inventory, 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. water at the upstream model boundary must be lower in Ra than water leaving the modeled area. However, the upstream region at Sunset Beach has inventories 3-4x greater than Huntington Beach and cannot be the primary source of nearshore water. Instead, offshore water with very low Ra concentrations must move onshore and then down the coast. Variations up to 50% of the mean inventory were observed at Huntington Beach and most likely reflect changes in the fraction of offshore water. The model simulation is most sensitive to changes in the mixing rate within 1.5 km offshore, which may explain the difference between the eddy diffusivity observed in this study and those measured further offshore. (228?) Introduction Rivers and groundwater discharge salts, nutrients, and contaminants into the nearshore coastal ocean. The nearshore water column also exchanges solutes with beaches and seafloor sediments, which is enhanced by wave and tidal pumping. In the coastal ocean, solutes are transported away from their source by dispersion and local currents. Predicting the fate of solutes that enter the coastal ocean from these various sources requires models that accurately describe nearshore circulation. The nearshore region can be divided into three zones: the littoral zone, inner shelf, and middle shelf. The littoral zone extends less than 1 km from the shoreline. In this region currents are wave dominated. The inner shelf extends from the littoral zone to about 10-20 m depth. The water column remains well mixed in this region, although 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. horizontal density contrasts may develop due to intense nearshore heating and the relatively small heat capacity of the shallow water column. Thermal expansion should produce a longshore current in the poleward direction on western facing coasts, which is maximum at the surface and zero at the bottom (Pettigrew and Murray, 1986). In the inner shelf, the effect of Coriolis acceleration is reduced and the local wind stress can overwhelm baroclinic forces, which causes surface currents to generally flow parallel to the underwater topographical variations (Csanady, 1977; Swift and Niedoroda, 1985). This produces net longshore flow with little or no net cross-shore flow. The middle shelf extends from the inner shelf to the shelf break. Deeper waters of the middle shelf are often in geostrophic balance, where the Coriolis force is balanced by a cross-shelf pressure gradient, but bottom and wind stress remain important (Lentz et al., 1999). This region is also influenced by open- ocean circulation, including boundary currents (Gawarkiewicz et al., 1996). Eddy diffusion, or dispersion, dominates transport when the mean current velocity is low, such as the cross shelf transport rates found on the inner shelf. Quantifying eddy diffusion is difficult because it requires measuring the truly turbulent part of a mean flow. As a result, current meters, which provide the mean flow, are ineffective. Instead, eddy diffusion can be quantified by averaging out the turbulent fluctuations over a period of time and relating the resultant flux of a dissolved substance to the mean properties of the water (Lewis, 1997). In this chapter, the 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. distributions of short-lived Ra isotopes are used to compute the nearshore eddy diffusivity. Radium is a naturally occurring radioactive element with four isotopes of varying half-lives. Each radium isotope is produced by the decay of a thorium parent, which is primarily found associated with sediments. Radium produced in the sediments is more soluble than its thorium parent and is released into pore waters. Dissolved Ra is then available to be transported into the water column by benthic processes, such as diffusion, advection, irrigation, and bioturbation, and shoreline processes, such as wave pumping and tidal pumping. With such a wide range of half-lives, the four Ra isotopes, 2 2 3 Ra (t1 / 2 =l 1.4 days), 2 2 4 Ra (t1 / 2 =3.66 days), 2 2 6 Ra (t1 / 2 = 1600 years), and 2 2 8 Ra (t1 / 2 =5.8 years) are useful for measuring a wide range of natural processes. The distribution of each isotope in the coastal ocean is a function of its inputs, the rate of offshore transport, and the rate of radioactive decay. The short-lived isotopes, 2 2 3 Ra and 2 2 4 Ra, have half-lives on the order of coastal ocean processes, and therefore should be useful in calculating nearshore mixing rates. In Chapter 2, a budget for the short-lived Ra isotopes in San Pedro Bay was constructed. This budget demonstrated the importance of different Ra sources to the inventory of Ra in surface water. In addition, longshore variations in the Ra inputs were identified. In this chapter, the inputs computed from this budget are used in a 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. two-dimension model for surface waters, extending from Sunset Beach to Newport Beach. This model computes the offshore distribution of Ra, and by adjusting the cross shelf eddy diffusivity, can be fit to the observed nearshore distribution of Ra. Then, by varying the model parameters, sources of variability of the offshore Ra isotopes distribution are assessed. Study Site San Pedro Bay is a relatively broad continental shelf located near the middle of the Southern California Bight, and is bounded to the north by Palos Verdes Peninsula and to the south by Newport Canyon. The climate is Mediterranean, characterized by wet, mild winters and warm, dry summers. Three rivers discharge into the Bay: the Los Angeles, San Gabriel, and Santa Ana Rivers. Most of the river water is diverted upstream to recharge groundwater. Below these diversions, dry-weather runoff, treated domestic sewage, and industrial discharges produce a minor flow. Tides in Southern California are mixed, with two high tides and two low tides of unequal height. Mean tides varies with location and coastal configuration, but is approximately 1.1m; the spring tides are greater than 2 m. Tidal currents have strong oscillatory motions that result in little net movement over a tidal cycle (Noble et al., 2003). Across much of the shelf, the major tidal currents flow parallel to the shoreline. 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In San Pedro Bay, previous studies have found that currents on the shelf follow the direction of local isobaths, with either a poleward or equatorward mean flow (Hickey, 1993; Karl et al., 1980; Noble et al., 2003). Superimposed on the mean longshore flow are tidal currents that propagate to the north (Karl et al., 1980). On the inner shelf, current fluctuations with a period of 5-10 days dominate, while longer period fluctuations (7-20 days) are observed on the outer shelf (Noble et al., 2003). The surface mean current is generally unrelated to local winds over the shelf (Noble et al., 2003). However, there is some indication that large-scale fluctuations in shelf currents may be generated by coastal-trapped waves propagating from Baja California (Hickey, 1992; Noble et al., 2003). Results S h o r e lin e : Surf zone short-lived Ra concentrations are related to the tides, with higher concentrations during low tide (Figure 3-1). Daily isotope concentration fluctuations were usually about 25%, indicating the complex interaction between Ra inputs from water circulation through beach sands and the offshore removal. The daily 2 2 4 Ra/2 2 3 Ra ratio remained relatively constant. This indicates that the two isotopes co-vary and probably have a coherent source. On four sampling days, the water temperature at the shoreline was also measured (Figure 3-2). In general, the temperature increased throughout the day, and began to cool by the time the last sample was collected, after 16:00 local time. There was no 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-1. Shoreline concentration of 223Ra (left column) and 224Ra (right column) at San Pedro Bay stations. Solid line represents the tidal height taken from tide tables. 70 60 1.5 50 40 30 2 0 B elm ont S hore ------------- ,0.5 9/25/01 19:00:00 1 0 ------ 9/25/01 7:00:00 9/25/01 13:00:00 800 700 1.5 600 200 B elm ont Shore ----------- -0.5 9/25/01 19:00:00 100 ---------- 9/25/01 7:00:00 9/25/01 13:00:00 Date Date 70 800 60 1.5 ~ ™ S 600 1 3 B S' £ 5 0 0 0.5 g £ 400 S 300 50 40 30 2 0 200 H u n tin g to n S ta te B each ------------ 1 -0.5 10/16/0119:00:00 H u n tin g to n S ta te B each 100 1 0 I : 10/16/01 7:00:00 10/16/01 13:00:00 1.5 1 H a C f l 0.5 ? 0 -0.5 Date Date 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-1 Continued. 100 1.5 a 0.5 < Sunset Beach 1 -0.5 6/5/02 19:00:00 6/5/02 7:00:00 6/5/02 13:00:00 Date S CL 6/7/02 7:00:00 6/7/02 13:00:00 Date 70 60 50 40 30 2 0 H untington State Beach 1 0 70 1.5 £ 50 0.5 H untington S tate Beach --------- 1 -0.5 6/12/02 19:00:00 6/12/02 7:00:00 6/12/02 13:00:00 800 700 1.5 S 600 I, 500 S s' 400 a 300 0.5 200 Sunset Beach 1 .0.5 6/5/02 19:00: 1 0 0 I ----- 6/5/02 7:00:00 6/5/02 13:00:00 Date Date 800 700 1.5 ■ _ 600 S S 500 a 3 400 ^ 300 200 0.5 Huntington State Beach ----- 1 . 0 . 5 6/7/02 19:00: 100 6/7/02 7:00:00 6/7/02 13:00:00 800 ^ 700 £ 600 s a 500 3 400 Sfi § 300 1.5 0.5 200 H untington S tate Beach --------- 1 -0.5 6/12/02 19:00:00 1 0 0 » ----- 6/12/02 7:00:00 6/12/02 13:00:00 Date Date 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tides (m ) § Tides (m ) ® Tides (m) Figure 3-1 Continued. 90 80 70 60 50 0.5 40 30 2 0 - Sunset Beach 1 0 ---------- 7/26/02 7:00:00 -0.5 7/26/02 19:00:00 7/26/02 13:00:00 Date 70 60 1.5 50 40 0.5 30 2 0 H untington Beach ---------- -0.5 7/26/02 19:00:00 1 0 ----- 7/26/02 7:00:00 7/26/02 13:00:00 Date a 30 N ew port Beach -0.5 7/26/02 19:00:00 7/26/02 7:00:00 7/26/02 13:00:00 a a CL 5 I" 800 - 1 i ... r 2 700 ■ I ■ 1.5 600 1 « H 500 a 3 400 \ f - °*5 I 300 “ 0 2 00 Sunset Beach 100 ..... i.......... -0.5 7/26/02 7:00:00 7/26/02 13:00:00 Date 7/26/02 19:00:00 800 700 1.5 'a a a - o 600 500 n 400 ?: 300 0.5 2 0 0 H untington Beach 1 -0.5 7/26/0219:00:00 1 0 0 |----- 7/26/02 7:00:00 7/26/02 13:00:00 Date 800 _ 700 fO a 6 o o E S 500 (2 400 S « 300 1.5 0.5 2 0 0 N ew port Beach --------- 1 .0.5 7/26/02 19:00:00 1 0 0 7/26/02 7:00:00 7/26/02 13:00:00 Date Date 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-1 Continued. N orth H untington Beach 2 800 1 .5 700 E 600 1 H s a 0 . 500 » 0 .5 f (2 400 a " 300 0 200 ■0.5 100 N orth H untington Beach 8/19/02 7:00:00 8/19/0213:00:00 Date 8/19/0219:00:00 8/19/02 7:00:00 8/19/02 13:00:00 Date 8/19/0219:00:00 70 800 60 — 700 E 600 3 “ ■ 500 400 50 40 a s S ■ 0.5 3 0.5 3 30 300 2 0 200 N orth H untington Beach N orth Huntington Beach -0.5 8/23/0219:00:00 -0.5 8/23/02 19:00:00 1 0 ----- 8/23/02 7:00:00 100 8/23/02 7:00:00 8/23/02 13:00:00 8/23/0213:00:00 Date Date 70 800 60 700 5 600 3 B S' ’ § ‘ 500 50 40 0.5! ft 400 0.5 3 30 300 2 0 200 Huntington State Beach j q Huntington State Beach 9/6/027:00:00 -0.5 9/6/02 19:00:00 -0.5 9/6/02 19:00:00 100 9/6/027:00:00 9/6/02 13:00:00 9/6/02 13:00:00 Date Date 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-1 Continued. 7 0 800 60 ^ 700 E 600 s Q - 500 1.5 50 40 0.5 30 2 0 200 * * 100 N o rth H untington B each , 6/18/03 7:00:00 N o rth H u n tin g to n B each --------- 1 -0.5 6/18/03 19:00:00 --------- 1 .0.5 6/18/0319:00:00 1 0 6/18/03 7:00:00 6/18/03 13:00:00 6/18/03 13:00:00 Date Date 70 800 60 700 1.5 1.5 g 600 1,500 'S 400 u 300 50 40 0.5 0.5 30 2 0 200 - N o rth H untington B each 1 0 0 I --------—------- L 6/25/03 7:00:00 N o rth H un tin g to n B each --------- 1 .0.5 6/25/03 19:00:00 --------- 1 .0.5 6/25/03 19:00:00 1 0 6/25/03 7:00:00 6/25/03 13:00:00 6/25/03 13:00:00 Date Date 800 700 \ 600 | 500 * 0 « 400 ti 3 300 N 200 1.5 £ 50 0.5 0.5 H un tin g to n S tate Beach H un tin g to n S ta te B each --------- 1 -0.5 8/28/03 19:00:00 --------- 1 .0.5 8/28/03 19:00:00 1 0 0 « ------ 8/28/03 7:00:00 8/28/03 7:00:00 8/28/03 13:00:00 8/28/03 13:00:00 Date Date 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Figure 3-2. Shoreline temperature as a function of time. Solid line represents the tidal height. 20 1.5 1.5 19 5 18 18 0.5 2 S 1 7 17 16 North Huntington Beach Huntington State Beach --------------- -0.5 6/18/03 19:00:00 -------------- -0.5 9/6/02 19:00:00 6/18/03 13:00:00 6/18/03 7 :0 0 :0 0 9/6/02 7:00:00 9/6/02 13:00:00 Date Date 1.5 1.5 S_ 19 cs 18 0 .5 3 0.5 3 North Huntington Beach North Newport Beach --------------- -0.5 8/28/03 19:00:00 --------------- -0.5 6/25/03 19:00:00 8/28/03 13:00:00 8/28/03 7:00:00 6/25/03 13:00:00 6/25/03 7:00:00 Date Date Figure 3-3. Nearshore distribution of 223Ra (left column) and 224Ra (right column). Solid line is exponential fit through data. Model fit parameters are presented in Table 1 . 9/11/01 Sunset Beach ' 70 700 r 6oo I 500 o- 400 r 9/13/01 Huntington Beach" 9/13/01 Huntington Beach £40 'es 30 06 a 20 10 70 700 600 | 500 a, 400 T 1 ------- 9/20/01 Sunset Beach T T T T i 9/20/01 Sunset Beach ‘ 300 100 10 70 C 60 700 *T 600 I £ 400 cs 300 06 a 200 100 4/3/02 Huntington Beach‘ 4/3/02 Huntington Beach 10 Distance Offshore (km) Distance Offshore (km) 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-3 Continued. 700 100 6/5/02 Sunset Beach*. 6/5/02 Sunset Beach 600 | 500; £ a 400 f 3 * “ 200 N 100 60 f « 40 700 70 6/7/02 Huntington Beach ■ E E a T 3 6/7/02 Huntington Beach' 600 500 400 i # 300 Z 200 " 100 700 70 — i--------------1 --------------- 6/12/02 Huntington Beach T 1 1 -------- 6/12/02 Huntington Beach T (ST 600 I E 500 400 300 a 4 0 200 100 Distance Offshore (km) Distance Offshore (km) * Note different scale on y-axis 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-3 Continued. 700 600 500 400 7/26/02 Sunset Beach 7/26/02 Sunset Beach rT 60 E E a "d a 40 100 700 70 7/26/02 Huntington Beach' 7/26/02 Huntington Beach 600 500 400 300 200 100 10 - 70 700 600 500 T T “ 1 1 -------- 7/26/02 Newport Beach ’ 7/26/02 Newport Beach 400 300 200 100 Distance Offshore (km) Distance Offshore (km) 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (fc -u i mdp) vnzzz ( e ™ “ dp) ^£ZZ k - ™ «*dp) e ^ z z z (e.u i uidp) Figure 3-3 Continued. 70 700 8/19/02 Huntington Beach' 60 50 40 < * ? 600 \ 500 £ a 400 30 20 10 200 100 0 700 ^ 600 £ 500 | 400 3 300; os 200 ^ 100 70 T T 1 1 -------- 8/23/02 Huntington Beach ’ 4 6 8 10 0 2 70 60 50 700 Sp 600 5 500 j | 400 "O w 300 M 6 200 6/18/03 Huntington Beach " 6/18/03 Huntington Beach 40 30 20 10 100 0 700 —r 6/25/03 Huntington Beach 6/25/03 Huntington Beach * 600 500 400 S 300 M Pi 200 £! to o 30 - 10 Distance Offshore (km) Distance Offshore (km) 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. clear correlation with the tides or Ra concentration. A nearshore increase in temperature has also been observed at buoys offshore from Huntington Beach and interpreted as the result of diurnal winds (Noble et al., 2003). During the day, winds are generally onshore and wind speeds reach a maximum in the afternoon. Winds push warm water onshore during the day, and relax during the night, with a minimum temperature observed in the early morning. However, this diurnal upwelling is not expected to impact the input of Ra to the mixed layer. Except for the first sample on 8/28/03, the range of temperatures observed at Huntington Beach was always within the temperature range of the mixed layer. Therefore, the source of water upwelled seems to be from the base of the mixed layer. O ffs h o r e : Ra concentration decreases with distance offshore (Figure 3-3). An exponential profile produced an excellent fit to the data for each transect: AC = k C 0e~xla (3-1) where A is the decay constant, C is the concentration at distance x offshore, C0 is the average shoreline concentration and a is the scale distance. In general, 2 2 3 Ra should be transported further offshore than 2 2 4 Ra since it has a longer half-life, increasing its scale distance. This is often, but not always, observed (Table 3-1). Equation 3-1 is also the solution to a one-dimensional diffusion-reaction model, where the scale distance is a function of the mixing rate and the half-life of the isotope. This has been used to interpret the distribution of short-lived Ra isotopes 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Table 3-1. Model derived values for average shoreline concentration and scale distance, along with the computed inventory, are presented. 223R a 224R a Co Scale Diet. Inventory Co Scale D lst Inventory Collect Date (dpm m -3) (ton) (xIO 6 <fcm m '1 ) ( d o m m -3) flan) (x l 0 6 dpm m'1 ) 9 /1 3 /0 1 3 8 .6 ± 2.4 0 .8 8 ± 0 .0 3 0.21 ± 0 .0 2 371 ± 12 0 .9 0 ± 0 .0 2 2 .1 6 ± 0 .0 8 4 /3 /0 2 3 9 .8 ± 7.5 3 .0 8 ± 0.71 1.1 8 ± 0 .3 5 5 4 0 ± 67 2 .2 8 ± 0.31 1 1 .1 8 ± 2 .0 5 6 /7 /0 2 4 4 .5 ± 5.3 0.91 ± 0 .6 7 0 .2 6 ± 0 .1 9 4 0 0 ± 4 9 0 .9 7 ± 0.71 2 .5 9 ± 1.93 6 /1 2 /0 2 3 6 .3 ± 1.6 0 .9 4 ± 0 .2 4 0 .2 2 ± 0 .0 6 4 0 2 ± 12 0 .9 8 ± 0 .1 8 2 .6 4 ± 0 .5 0 7 /2 6 /0 2 4 3 .0 ± 6.2 0 .6 3 ± 0 .5 7 0 .1 4 ± 0 .1 3 4 8 5 ± 6 4 0 .4 5 ± 0 .4 4 0 .9 8 ± 0 .9 8 8 /1 9 /0 2 2 6 .2 ± 2.2 0 .9 0 ± 0 .4 6 0 .1 5 ± 0 .0 8 241 ± 16 0 .7 2 ± 0 .2 9 0 .9 8 ± 0 .4 0 8 /2 3 /0 2 2 4 .3 ± 3.1 0 .4 7 ± 0 .4 0 0 .0 5 ± 0 .0 4 2 9 3 ± 25 0 .3 9 ± 0 .2 4 0.41 ± 0 .2 6 6 /1 8 /0 3 2 0 .3 ± 0 .4 1.21 ± 0 .1 0 0 .1 8 ± 0 .0 2 2 1 9 ± 29 0 .7 7 ± 0.41 0 .9 9 ± 0 .5 4 6 /2 5 /0 3 2 7 .0 ± 1.3 1.53 ± 0 .3 7 0 .3 3 ± 0 .0 8 2 9 7 ± 26 0 .8 7 ± 0 .4 6 1.63 ± 0 .8 7 Sunset Beach 6 /5 /0 2 7 1 .4 ± 8.1 2 .8 2 ± 1.33 1.81 ± 0 .8 8 5 2 4 ± 4 8 1.8 8 ± 0.71 7 .8 3 ± 3 .0 5 7 /2 6 /0 2 55 .3 ± 5.2 0 .8 3 ± 0 .4 9 0 .2 5 ± 0 .1 5 7 2 7 ± 39 0 .6 0 ± 0 .2 0 1.86 ± 0 .6 3 further offshore (> 10 km) in the mid-Atlantic Bight (Moore, 2000). However, because of the importance of longshore flow, longshore changes in the Ra input flux, and the importance of seafloor inputs out to depths of about 1 2 m as outlined in Chapter 2, a 1-D diffusion-reaction model is not valid in this environment. The short-lived Ra isotope inventory (I; atom s' 1 (m shoreline)'1 ) in surface waters can be calculated by integrating the concentration offshore assuming the mixed layer concentration is independent of depth and taking into account the increasing depth of water with distance offshore: ° ° Hm 1 = f f h C d z d x (3-2) 0 0 where Hm , the mixed layer thickness, is a function of x and constrained by water depth. An average Hm of 11.8 m was used for the Huntington Beach transects. In Figure 3-4, the inventory on each sampling day is plotted. The only sampling day during wintertime conditions, when the water column was weakly stratified, was on 4/3/02. Because of the importance of other processes during winter, such as vertical mixing, the following discussion will be based solely on the summertime data, collected between June and September. Samples collected within a week of one another had similar inventories. But over longer periods, the inventory changed. Changes in the shoreline concentration and/or the scale distance produce changes in the inventory. In Figure 3-5, the 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Inventory (106 d p m ( m shoreline)'1) Figure 3-4. Short-lived Ra isotope inventory in surface water offshore of Huntington Beach on each sample date. Solid bar is the 223Ra inventory and the striped bar is the 224Ra inventory. 11.2 9/13 2001 6/7 6/12 7/26 8/19 8/23 2002 Date 6/18 6/25 2003 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-5. Correlations between inventory, scale distance, and shoreline concentra tion for 223Ra (left column) and 224Ra (right column): A) Inventory as a function of scale distance, B) inventory as a function of shoreline concentration, and C) scale distance as a function of shoreline concentration. 0.3 I 0 -2 7 3 s O o L. © d 4 > > 0.1 400 600 800 1000 1200 1400 Scale Distance (m) £ © 0.3 a, a 0.2 ■ d s o © rH ~o.i S e U o a 1 5 B) 223Ra 1 ► 'w ' J Ra -----I F - ' . ...1 .... - -----z q j > — ■ ------ - 50 20 25 30 35 40 45 Shoreline Concentration (dpm m'3) / • “ S P 3 u £ Vi 2.5 s , 2 & d ) 1.5 s o o 1 £ 2 0.5 d 4) > £ 0 3 £ 3 . § Vi 2.5 B 2 s a, 1.5 s e o 1 £ © 0.5 d £ 0 a 1 A l 224R a ,............... „ - H F 1 _ - 1 f 300 400 500 600 700 800 Scale Distance (m) 900 1000 B) 224Ra 150 200 250 300 350 400 450 500 Shoreline Concentration (dpm m'3) 2000 8 1500 a ! 5 Q 1000 C 5 & 500 C) 223Ra 1500 © 1 1000 « 500 5 0 C) 224Ra 15 20 25 30 35 40 45 50 Shoreline Concentration (dpm m"3) 0 150 200 250 300 350 400 450 500 Shoreline Concentration (dpm m'3) 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-6. Summer short-lived Ra isotope inventory as a function of various environmental conditions: A) maximum tidal range during previous 3-days, B) tidal range while sampling, C) mixed layer thickness, and D) mixed layer temperature. r 0.35 'v 0.3 2.5 - 0.25 t “ a 0.15 1.5 e 0.1 t 3 0.05 s £ I 0 " 1 0.5 223, 224, 1.7 1.8 1.9 2.1 2.2 1.7 2 1.8 1.9 2.1 2.2 3-Day Tide Maximum (m) 3-Day Tide (m) B) Ra 0.6 0.8 1 1.2 Tidal Range While Sampling (m ) 0.6 0.8 1 1.2 Tidal Range While Sampling (m ) ■ T 0.35 S ' I ° - 3 1 0.25 f 0.2 0.15 a 0.1 b 2 I 2.5 1.5 0.5 0.05 224, 223, a 6 6 8 1 0 1 2 14 16 18 Mixed Layer Thickness (m) Mixed Layer Thickness (m) ~L 0.35 £ 1 0.3 o .a V I e g 0.2 1 8 0.15 2.5 0.25 1.5 t 01 1 0.05 I 0 0.5 223, 224, 17.5 18 18.5 19 19.5 2 0 20.5 17.5 18.5 19.5 20.5 Mixed Layer Temp (°C) Mixed Layer Temp. (°C) 139 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inventory is plotted as a function of these two variables. While there is significant scatter in the plot of inventory as a function of the shoreline concentration, there is a good correlation between the inventory and the scale distance. This suggests that changes in the offshore Ra distribution at Huntington Beach reflect changes in the Ra inventory and do not reflect changes in the mixing rate. Changes in the inventory may represent changes in the Ra input flux, which may be driven by environmental conditions, such as tidal range or wave height. Sampling was done to cover a broad range of environmental conditions. In Figure 3-6, the inventory is compared to several different variables, including the tidal range while sampling, the maximum tidal range during the previous 3 days, the mixed layer thickness, and the mixed layer temperature. No correlation was found with any of these variables. The importance of wave conditions was tested on 9/6/02 at Huntington Beach. On this day, there was a large swell event, with wave heights > 3 m. The shoreline concentration is shown in Figure 3-1. There was no difference between this day and other sampling days. Therefore, changes in the Ra inventory at Huntington Beach are not the result of changes in environmental conditions, and most likely not the result of changes in local input fluxes. Instead, they may be related to variations in the longshore flow of water. 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-D Mixing Model Short-lived Ra enters the coastal ocean at the seafloor, shoreline, and through tidal exchange with marshes and estuaries (Chapter 2). In the coastal ocean, these isotopes are isolated in the mixed layer, due to a thermal gradient at its base that inhibits vertical mixing. Advection and eddy diffusion transport Ra away from their sources, the distance traveled depending on the transport rate and the half-life of each isotope. To better understand the role of each Ra source to the offshore distribution of short-lived Ra, and to use that distribution to estimate the offshore mixing rate, a 2-D finite-difference model was developed to compute the Ra isotope distribution in coastal surface waters along coast of San Pedro Bay. M o d e l G e o m e tr y : The model covers 15 km of shoreline, from south of Sunset Beach to Newport Beach, and extends 20 km offshore (Table 2-3). There are 30x80 elements to the model. Longshore boxes, labeled with the subscript i and beginning at the north end of the model, are all 500 m wide. Cross-shore boxes, labeled with the subscript j and beginning at the shoreline, are 50 m at the shoreline and increase progressively by 5 m. The Huntington Beach transect is 9.5 km south of the upstream boundary (i=19). Between the shoreline and the mixed layer-seafloor intercept, the change in bathymetry was incorporated into the geometry of each box (Figure 3-7). The seafloor bathymetry was based on measurements made while sampling at Huntington 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3-2. Model parameters used in the two-dimensional finite difference model. GEOMETRY Longshore Boxes (i) 30 Offshore Boxes (j) 50 Box longshore width (m) 500 Shoreline box offshore length (m) 50 Increase in box length offshore (m) 5 Seafloor gradient to 890 m offshore 0.01 Seafloor gradient beyond 890 m 0.00417 Mean mixed layer thickness (m) 11.8 Diffusive Time step (seconds) 162 FLUXES 223Ra Shoreline Flux (dpm sec'1 (m shore)'1) 0.0179 223Ra Seafloor Flux (dpm sec"1 (m shore)'1) 0.2971 224Ra Shoreline Flux (dpm sec'1 (m shore)'1) 0.3454 224Ra Seafloor Flux (dpm sec'1 (m shore)'1) 4.7292 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. W ater D epth (m) Figure 3-7. Model bathymetry within the first 2 km offshore. Each point represents the distance offshore to the midpoint of box. Seaward of 1.5 km, the mixed layer has a constant thickness of 1 1 . 8 m and is isolated from the seafloor. 0 2 4 6 8 1 0 1 2 1.5 2 0 0.5 1 Distance Offshore (km) 143 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Beach and NOAA Nautical Chart 18746. Within this region, the bathymetry is fairly constant. The seafloor slope is 0.01 out to 0.9 km offshore. Beyond this, the seafloor slope is 0.004 out to water depths beyond the thickness of the mixed layer. In the longshore direction, box volumes must remain constant in order to incorporate longshore advection into the model. B o u n d a r i e s : The model has three open boundaries. At the north and south boundary, a one-dimensional, steady-state model is solved that incorporates the seafloor geometry and assumes longshore changes in inputs and mixing rate are negligible. These assumptions, including steady-state, are approximations. However, by running the boundary model to steady state, the inventory in surface water can easily be computed as the sum of the input fluxes. Then, by changing the input fluxes and the mixing rate at the boundary, different Ra distributions can be generated. The role of boundary conditions on the distribution of Ra within the model is an important component of this analysis and is further discussed below. At the offshore boundary, a nominal 2 2 3 Ra and 2 2 4 Ra concentration of lx 10' 4 and lxlO ' 3 dpm m'3 , respectively, was defined. C o m p u ta tio n s : Each box within the model has a mass balance equation, which computes the change in atoms in the box (ANy) during each time step (At): ANy = (Jd if +Jin -Jd e c a y ) At = ACyVj (3-3) 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. These include a diffusive exchange with up to four adjacent boxes (Jd if ), an input flux (Jin ) from the shoreline and seafloor, and a loss by radioactive decay (Jd e c a y)- ANy is also equivalent to the product of the box concentration (Cy) and the box volume (Vj). I n p u t F l u x : Source functions(Jin ) for each isotope were defined in Chapter 2. The first row of boxes has an input from the shoreline. Each box located between the shoreline and the mixed layer-seafloor intercept has an input from the seafloor. For this model, advection is assumed to be from the north. Under these conditions, inputs from the Talbert Marsh and Santa Ana River have a negligible effect on the profiles at Huntington Beach and have been left out of the model. E x c h a n g e : Diffusive exchange occurs between adjacent boxes. Diffusive exchange between two boxes, Jd if, is defined as: J dif= - A K h^ (3-4) Where A is the cross sectional area for the boundary between the two boxes (m2 ), Kh is the eddy diffusivity (m2 s'1 ), AC is the difference in concentration of two adjacent boxes (atoms m'3 ), and Ax is the distance between box midpoints. The eddy diffusivity is the focus of this study. The diffusive exchange flux depends not only on the rate of eddy diffusion, but also on the cross-sectional area, which is a function of water depth. Thus, as the water depth increases offshore, the diffusive exchange rate, the product of the eddy diffusivity and the cross-sectional area, increases. 145 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A d v e c t i o n : Longshore advection is accomplished by moving the entire contents of a box in the desired direction of transport after an appropriate time elapses. This advection time step is defined as the box width divided by the advection rate. Since the time step for advection is much greater than the diffusion time step (hours vs. minutes), the advection time step is defined as an integer multiple of the diffusion time step. Superimposed on the mean flow are tidal fluctuations that produce a northward flow during the flood tide and a southward flow during the ebb tide (Noble et al., 2003). For continuous sources, such as the shoreline and seafloor, this process has little impact since adjacent boxes have very similar fluxes and concentrations. Therefore, the impact of tidal circulation on the distribution of Ra can be ignored. However, tidal variations of the shoreline Ra input flux are important and are explored with this model. R a d i o a c t i v e D e c a y : Radioactive decay is ubiquitous for Ra. The loss of Ra by radioactive decay (Jd e c a y ) for each box is: J d e c a y = ^ j ^ R a ^ i j (3~5) Where X R a is the decay rate for a given Ra isotope and Vj is the box volume. S te a d y - S ta te : The model is designed to find the distribution of isotopes at steady- state, given a defined set of boundary conditions. The rate that the model reaches 146 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. steady-state depends on the advection rate. When there is no advection, the 2 2 3 Ra profile takes 50 days to equilibrate. But, when advection is included, the time to reach steady state decreases significantly because the system is rapidly flushed out. For an advection rate greater than 1.5 cm s'1 , equilibrium is reached in less than 12 days. D a ta C o m p a r is o n : To compare the model results to the data, a routine was developed to find the first box midpoint beyond the point where the sample was collected. The model derived concentration at the sample point (Cm ) was computed by assuming a linear concentration gradient between the two boxes: Solutions for 2 2 3 Ra and 2 2 4 Ra were computed simultaneously, which allowed the chi- squared for each isotope to be combined to find the model parameters that best fit both data sets. Two variables were used to minimize chi-squared. First, the inventory at the upstream boundary was defined as a fraction of the inventory observed at Sunset Beach. The second variable was the eddy diffusivity, Kh . \ (3-6) The best fit to the data was found by minimizing the chi squared statistic: 2 (3-7) 147 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. To simplify the calculation for the composite summertime Ra profiles, an exponential model was fit through the data. Then, at five distances that corresponded sampling stations, the concentration of each isotope was computed. The model could then be fit to those data points for each isotope. Model Results M o d e l T e s ts : Several tests were conducted to ensure that the model was computing profiles properly. To begin, the model was simplified by defining the mixed layer thickness at a constant 10 m depth. Then, the model results were directly compared to a 1-D analytical solution. Two general cases were examined. First, a constant flux was defined at the shoreline (Figure 3-8). Second, the input flux was set at zero and the boundary concentration at the upstream boundary was held constant at 1 0 dpm m 3 , and the model was run to equilibrium with an advection rate of 0 (Figure 3- 9) and 1 cm s' 1 (Figure 3-10). Each model closely matched the predicted profile. The greatest deviation was with longshore advection, which predicted slightly greater concentrations in each box. This is the result of the discretization, which requires the entire contents of a box to be translated longshore in order to simulate advection. Between advection time steps, diffusive transport relaxes the concentration towards the predicted analytical solution. Then, the bathymetry, with a sloping seafloor, was introduced to the model, and the tests described above were repeated (Figure 3-11). Since a 1-D analytical solution 148 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 224R a (d p m m'3) 223R a (d p m m‘3) Figure 3-8. Comparison of 2-D model and 1-D boundary model with a constant mixed layer thickness of 10 m 1-D analytical solution (Input flux = 10 dpm sec"1; Kh = 1.3 m2 sec'1 ). 1200 C onstant 10 m Mixed Layer T hickness 1000 223Ra (1-D Analytical Sol'n) 223Ra (Boundary Model) 223Ra (2-D Model) 800 600 400 200 0 1 2 3 4 5 6 600 Constant 10 m Mixed Layer Thickness 500 224Ra (1-D Analytical Sol'n) 224Ra (Boundary Model) 224Ra (2-D Model) 400 200 100 0 1 2 3 4 5 6 Distance Offshore (m) 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-9. Longshore Ra distribution for 1-D analytical solution and 2-D model with constant mixed layer thickness of 10 m and no seafloor topography. Eddy diffusivity is = 1.3 m2 sec"1, and longshore advection rate is u= 0 cm sec'1. 1 2 s a S a P * f) a 3 « e* ■ * * 1 0 8 6 4 2 0 1 2 1 0 8 I \ 3 I 10 m Mixed Layer Thickness Northern Boundary = 10 dpm rrr3 u = 0 cm sec"1 ... 2 2 3 ( J -1 ) Analytical Sol'n) ■ 223Ra (2-D Model) 0 4 6 8 Longshore Distance (km) 1 0 1 2 0 1 1 1 1 1 10 m Mixed Layer Thickness _ Northern Boundary = 10 dpm n r3 u = 0 cm s e c 1 p 1 ......... 224j^a ([_[) Analytical Sol'n) — v p 1 ............ 224Ra (2-D Model) r _» l - I » * — ? * % r \ % w ~ \ - 1 ______L _________ 1 __________L ______ 0 4 6 8 Longshore Distance (km) 1 0 1 2 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-10. Longshore Ra distribution for 1-D analytical solution and 2-D model with constant mixed layer thickness of 10 m and no seafloor topography. Eddy diffusivity is = 1.3 m2 sec'1, and longshore advection rate is u= 1 cm sec'1. tr> S a « 3 9 3 o * 1 2 1 0 8 1------------------ i ------------------ r 10 m Mixed Layer Thickness Northern Boundary = 10 dpm rrr3 u = 1 cm s e c 1 0 0 1 2 10 'r I e B & • 3 93 04 t t 8 223Ra (1-D Analytical Sol'n) 223Ra (2-D Model) 8 1 0 1 ----------------- T ----------------- T 10 m Mixed Layer Thickness Northern Boundary = 10 dpm n r3 u = 1 cm sec-1 224Ra (1-D Analytical Sol'n) 224Ra (2-D Model) ' V . . 0 0 4 6 8 Longshore Distance (km) 1 0 1 2 1 2 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-11. Comparison of offshore distribution of Ra between 2-D model with seafloor topography and 1-D boundary model (input flux = 10 dpm sec'1; Kj, = 1.3 m2 sec'1 ). 3500 3000 'a 2500 B < § * 2 0 0 0 1500 1000 500 0 0 1 2 3 4 5 6 3000 ~ 2500 ro ■ E g 2000 a 3 5 1500 a 1000 500 0 0 1 2 3 4 5 6 Distance Offshore (m) 152 t---------------- 1 ---------------- 1 r i i » - l i Model With Bathymetry . . . — .. 224iga (Boundary) - - - - - 224Ra (Huntington Beach) I 1 T I I J. tit n i r Model with Bathymetry — 223|^a (Boundary) - - 223Ra (Huntington Beach) •--I— * Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cannot incorporate the changes in mixed layer thickness, the profile at Huntington Beach is compared to the 1-D boundary model. When the true bathymetry is included in the model, the shoreline Ra concentration increases and the scale distance decreases. The difference reflects the importance of water depth on the diffusive transport of solutes. The shallow nearshore water column decreases the diffusive flux, inhibiting the offshore transport of Ra out to about 1.2 km. Beyond this point, the mixed layer has a constant thickness, and the effective mixing rate remains constant. The importance of other mixing rate changes are further explored in the Discussion section. Finally, with the true bathymetry included, the input flux was set at zero, and the upstream boundary concentration was held constant at 10 dpm m'3 . The resulting profiles matched the 1-D analytical solution and were identical to Figure 3-9 and Figure 3-10, since there is no change in the water depth in the longshore direction. M o d e l S e n s i t i v i t y a n d R e s u l t s : To begin, the model was fit to the average summertime profile for 2 2 3 Ra and 2 2 4 Ra at Huntington Beach for three advection rates, 2, 5 and 7 cm s" 1 by independently adjusting the mixing rate, Kh , and the inventory at the northern boundary, f, until a best fit combination was found (Figure 3-12). For advection rates greater than 5 cm s'1 , the model predicts an eddy diffusivity of 1.3 m2 s'1 , and a slightly greater rate at slower advection rates. The model also predicts that the inventory at the northern boundary is less than 8 % of the observed inventory at Sunset Beach. This inventory is equivalent to 32% of the 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-12. Composite summertime Ra profiles at Huntington Beach with model best-fit results for three advection rates (u; cm sec'1 ). Model fit paramaters are the fractional inventory of water from Sunset Beach (f) and the mixing rate (Kh; m2 sec"1 ). i e s a 3 0 5 P 4 rl (N 70 • 223Ra (dpm nr-*) 223Ra (u=2; f=3%; Kh=1.56) 223Ra (u=5; f=8%; Kh=1.31) 60 50 40 30 2 0 1 0 0 4 5 6 2 3 0 1 700 • 224Ra (dpm m'-5 ) 224Ra (u=2; f=3%; Kh=1.56) 224Ra (u=5; f=8%; Kh=1.31) • « ■ 224Ra (ii=7; f=IO%; K h=l.26) 600 /■ s O S 500 s 400 s 300 200 100 4 6 0 2 3 5 1 Distance Offshore (km) 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-13. Chi squared contour plots at three different advection rates as a function of mixing rate (Kh) and Sunset Beach inventory fraction (f). Contours are spaced at approximate integer multiples of the the chi squared minimum (approximately 7.5). u = 2 cm sec e o 10.7 - tu G 3 0.6 0.8 1.2 1.4 1.6 1.8 2 u = 5 cm sec B © O R 10.7 - G < L > > G 1 . 2 1.4 1.6 1 . 8 2 0 . 6 0.8 u = 7 cm sec G o ■ a % 10.7- •G o 8 5.3- C Q t G 3 IZ 1 1.6 1.8 0.6 0.8 1.2 1.4 2 Kh (m2 sec'1 ) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inputs at Huntington Beach. At 2 cm s'1 , the best fit eddy diffusivity increases slightly and the predicted inventory drops to 3%. This requirement of low-Ra water upstream from HB is consistent with the budget presented in Chapter 2. The uniqueness of each model solution can be assessed by plotting a contour map of chi squared for various combinations of Kh and f (Figure 3-13). The minimum value of x 2 is located at the center of each plot, and the contours are integer multiples of the minimum. Assuming that 2x2 is equivalent to one standard deviation, then the uncertainty is 1.3 ±0.2 m2 s' 1 for Kh and 8 ±4% (absolute) for f. Beyond 2x2 , the contours are closer together, implying that the quality of the fit degrades quickly. Several other parameters defined in the model had large uncertainties. These values were changed within their uncertainty and the best fit to the average summertime 2 2 3 Ra and 2 2 4 Ra profiles was determined. First, the shoreline and seafloor inputs were reduced by 50%, which reduced the longshore Ra gradient. With this input and an advection rate of 5 cm s"1 , the inventory of water at the upstream boundary increased slightly, from 8 % to 14% and the mixing rate decreased from 1.3 to 0.8 m2 s' 1 (Figure 3-14). Despite the lower input fluxes, high nearshore concentrations are maintained by decreasing the mixing rate, which reduces offshore transport. Regardless of the magnitude of the inputs, the model could be fit with only a narrow range inventories at the upstream boundary. Changing the inventory at the upstream 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e.i u ludp) BX£ Z Z (e- i u radp) Figure 3-14. Best fit to data given various input fluxes at a longshore advection rate of 5 cm sec"1. Model fitting parameters are the inventory at the boundary (f) and the mixing rate (Kh; m2 sec'1 ). 70 • 223Ra (dpm m"3) ; f= 8%; Kh=1.3) ----------- 223Ra(Inputs=80% ; f= 12%; Kh=1.15) — — - 223Ra (Inputs= 50%; f= 14%; Kh=0.78) ------ - - 223Ra (Inputs= 0.1% f=10% Kh=0.1211) 5 ~ r • 224Ra (dpm m ’3) 7 00 i? Kh=1.3) ----------- 224Ra(Inputs=80% ; f=12% ; Kh=1.15) -------- --- 224Ra(Inputs=50% ; f= 14%; Kh=0.78) ------- - 224Ra (Inputs= 0.1% f= .10% K h=0.12ll) .i.Q 2 3 4 Distance Offshore (km) 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e. r a radp) v n iZ Z (e. m uidp) Figure 3-15. Model results for various Ra inventories at the upstream boundary as a fraction of the inventory at Sunset Beach (f). For all model runs, advection rate is 5 1 9 1 cm sec' 1 and Kh is 1.31 mz sec"1. 70 60 • 223Ra (dpm m "-’) — 223Ra (f=8%) 223Ra(f=0% ) 50 40 30 2 0 1 0 0 0 1 2 3 4 5 6 800 700 • 224Ra (dpm m "-i 224Ra (u=8%) 224Ra (f=0%) ■ ■ * 224Ra ( i= l3% ) — - 224Ra (f=26%) 600 500 400 300 200 100 0 1 2 3 4 5 is! Distance Offshore (km) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. boundary to values beyond 2 standard deviations of the average produced noticeable changes in the modeled Ra profile at Huntington Beach (Figure 3-15). Ra inputs probably decrease as a function of distance offshore. However, the rate of decrease is unknown. For the initial model, the seafloor distribution of inputs was divided equally among the boxes in contact with the mixed layer. Since the seafloor area in contact with each box increased with the box length, the resulting seafloor flux per m2 at the last box 1.5 km offshore decreased to 38% of the shoreline value. When most of the input is at the shoreline, with the last box only 8 % of the shoreline value, the model could not fit the data, particularly for 2 2 3 Ra (Figure 3-16). As inputs are concentrated at the shoreline, the upstream inventory decreases and the mixing rate increases. To produce the observed profiles with this model requires seafloor inputs of Ra across seafloor in contact with the mixed layer. Tides impact the Ra input flux from estuaries, tidal marshes, and water draining out of the beach. As water enters these reservoirs during the flood tide, their Ra input flux to the coastal ocean drops to zero. During the ebb tide, the water draining out of these reservoirs carries Ra into the coastal ocean. Discharge can even continue into the next flood tide, as momentum carries water out of estuaries and beach water table outcrops remain above sea level. To simulate this change in flux through time, the input flux at the shoreline was increased by a factor of 3, but only turned on in the model for 1/3 of a tidal cycle (4.2 hours). Once the model reached steady-state, the resulting changes in concentration over 3 days is presented in Figure 3-17. Local 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-16. Best fit to data given various distribution of seafloor inputs. A l f a is the fraction of seafloor input at furthest box, 1 . 2 km offshore relative to the shoreline box. Longshore advection rate is 5 cm sec'1, f is the fraction of Sunset Beach inventory at the upstream boundary, and Kh is eddy diffusion (m2 sec"1 ). E a 3 a & c* 0 70 • 223Ra (dpm in"-’) 223Ra (Alfa= 38% f= 8%; Kh=1.31) 60 — - 223Ra (Alfa= 8%; f= 3%; Kh= 1.76) 50 40 30 20 10 0 700 • 224Ra (dpm in"-’) 224Ra (Alfa= 38% f= 8%; Kh=1.31) 600 — - 224Ra (Alfa= 8%; f= 3%; Kh=1.76) 100 0 1 4 2 3 5 6 Distance Offshore (km) 160 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-17. Model sensitivity to tidally-driven changes in the shoreline flux. A) Predicted shoreline variability at Huntington Beach over three days for 223Ra (dashed line) and 224Ra (solid line). Average concentration for each is shown on the Y-axis. The lower panels show the offshore distribution of B) 223Ra and C) 224Ra at the start of day 13 (dashed line) compared to the offshore Ra distribution predicted using when the shoreline flux is averaged over the entire tidal cycle. 450 400 | 350 ~ & 7 3 3 300 3 I u > 250 Average^ 223Ra*1 a C S e* o * fS 3 a 3 « 06 • T T C 3 70 60 223R a (Low Tide) 223Ra (No Tide) | 50 40 20 10 0 0.6 0.8 1 0 0.2 0.4 Distance Offshore (km) 700 600 224Ra (Low Tide) 224Ra (No Tide) 500 400 300 200 100 0 0.2 0.4 0.6 Distance Offshore (km) 0.8 1 0 161 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e. r a uidp) v n £ zz (c.u i uidp) v n pzz Figure 3-18. Model results for various mixing rates (Kh; m2 sec-1). For all model runs, the advection rate is 5 cm sec'^ and the upstream inventory is equal to 8 % of the Sunset Beach inventory. 70 60 • 223Ra (dpm m‘3) 223Ra (Kh= 1.31) 223Ka (Kh= .1 ) — - 223Ra (Kh= 2) - - 223Ra (Kh=2.5) 50 40 30 20 10 0 0 1 2 3 4 5 6 800 700 • 224Ra (dpm m ° ) 224Ra (Kh=1.31) 224Ra (Kh=l) — - 224Ra (Kh=2) - " 224'Ra (Kh=2.5) 600 500 400 300 200 100 0 1 2 3 4 5 6 Distance Offshore (km) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. changes in the input flux produced a 10-15% fluctuation in the shoreline concentration. The observed fluctuations were about 25%. This test also showed that simplifying the model by using average inputs is a reasonable assumption. When the shoreline concentration was averaged over a tidal cycle, the average concentration is identical to the shoreline concentration computed with the average input flux data.# In addition, these nearshore fluctuations are quickly averaged out, with an identical concentration predicted at less than 0.5 km offshore (Figure 3-17). Next, the model sensitivity to eddy diffusivity was assessed. Results within the first 1.5 km offshore were most sensitive to changes in Kh (Figure 3-18). Reducing the mixing rate resulted in higher concentrations at the shoreline, while higher mixing rates reduced the concentration. Beyond about 2 km, changes in the mixing rate had only a small impact on the concentration, with slightly higher concentrations at higher mixing rates. The region between 1.3 and 2.0 km offshore for 2 2 3 Ra and 1.0 and 1.5 km offshore for 2 2 4 Ra is less sensitive to changes in the mixing rate, with little variability in concentration over the range of mixing rates modeled. Significant concentration changes in these regions require changes in the Ra inventory at the upstream boundary. The mixing rate, Kh , may vary with time. For example, on time scales of days to weeks, weather systems, lunar tidal cycles, and regional changes in circulation may 163 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. influence nearshore mixing rates. On shorter time scales, local winds and tides may also influence Kh . The time required for the model to relax from one steady-state mixing regime to another can be tested with this model. To accomplish this, the box width was doubled to 1 km to minimize the impact of boundary conditions during the first few days after a change in Kh occurred. In Figure 3-19, the shoreline concentration is plotted as a function of time. On short time scales, the shoreline concentration changes rapidly. After just 15 minutes, both isotopes have evolved >20% of the way to the new equilibrium. Therefore, short-period fluctuations in the mixing rate may contribute to the Ra concentration variability observed at the shoreline. After just 24 hours, the nearshore distribution of 2 2 4 Ra is about 80% of the way to the new equilibrium while 2 2 3 Ra is about 65% (Figure 3-19). In addition, there is only a small change in the Ra distribution beyond 1.5 km (Figure 3-20). Finally, the distribution of Ra at the upstream boundary may be different than at Huntington Beach. For example, at Sunset Beach, higher concentrations are observed further offshore. Or, an onshore flow may bring low-Ra water into the nearshore. The offshore gradient at the upstream boundary was varied by choosing different scale factors (i.e. varying the eddy diffusivity at the boundary, Kb ), and the inventory was held constant. A longshore advection rate of 5 cm s" 1 was assumed and new profiles were calculated (Figure 3-21). For 2 2 4Ra, the distribution at the upstream boundary had little impact on the distribution at Huntington Beach. 2 2 3 Ra 164 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e_ i u uidp) V1 &£ZZ Figure 3-19. Shoreline concentration as a function of time after a change in the mixing rate from K^= 1.3 m2 sec" 1 to K^=3.0 m2 sec"1. Longshore advection rate = 5 cm sec'1. 28 26 24 22 20 18 16 14 12 0 1 2 3 4 5 Time (Days) 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (£_ u i uidp) v n pzz 224R a (d p m m "3) 223R a (d p m m'3) Figure 3-20. Model response to a change in the eddy diffusivity (Kh): steady-state profile at Kh=1.3 m2 sec'1, and 3.0 m2 sec"1 , and profile after 1 day. 70 • 223Ra (dpm m-3) 223Ra (Kh=1.3) ........ 223Ra (Kh=3; 1 day) 60 50 40 30 20 10 0 0 1 2 3 4 5 6 800 700 600 500 400 300 200 100 0 0 1 2 3 4 5 6 Distance Offshore (km) 166 • 224Ra (dpm m-3) 224Ra (Kh=1.3) 224Ra (K h=3:1 day) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-21. Model sensitivity to changes in Ra distribution at the upstream boundary. Beginning with the best fit for an advection rate of 5 cm sec"1, the model derived Ra distribution at Huntington Beach is presented for various mixing rates at the upstream boundary (Kb; m2 sec'1). Mixing rate across the model (Kh) is set at 1.31 m2 sec"1. S f t 3 c e 70 • 223Ra (dpm m-3) 223Ra (u=5 cm sec-1; f= 8%; Kb=Kh=1.31) 223Ra (K h=l.31; Kb=.l) - - - 223Ra (Kh=1.31; Kb=.5) — - 223Ra (Kh=1.31; Kb=2.5) 60 50 40 30 20 10 0 4 0 1 2 3 5 6 700 600 500 • 224Ra (dpm m-3) 224Ra (u=5 cm sec-1; f= 8 C ----------- 224Ra (Kh=1.31; Kb=.l) - - - - - 224Ra (Kh=1.3'l; Kb=.5) 224Ra (Kh=1.31; Kb=2.5) Kb=Kh=1.31) & 400 * 300 ^r - i f — 1 0 2 3 4 Distance Offshore (km) 167 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-22. Composite summertime 228Ra at Huntington Beach with model best-fit results for three advection rates (u; cm sec'1 ). The fractional inventory of water from Sunset Beach and the diffusivity were set based on the model results for 223Ra and 224Ra at each advection rate. The seafloor flux, J (atoms sec' 1 (m shoreline)"1 ), was adjusted to find the best fit to the data. 160 • 228Ra (dpm m-3) 228Ra (u= 2; J= 4.0xl06) 228Ra (u= 5; J= 5.3x106) ' ’ ’ 228Ra (o=7; J= 6 .6 x10(' i 140 120 s 100 .r i 4 6 0 2 3 5 Distance Offshore (km) 168 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was more sensitive. The sensitivity of 2 2 3 Ra should decrease at slower advection rates. Application to 2 2 8 Ra Moore (1987) showed that without knowing the residence time of water on the shelf, the 2 2 8 Ra inventory underestimated its input flux because seasonal upwelling brought low-2 2 8 Ra water to the surface. For this study, the residence time of water, the offshore mixing rate, and the inventory of Ra in water entering the San Pedro Bay have been constrained. Therefore, the model developed can also be used to estimate the 2 2 8 Ra input flux. At each advection rate, the eddy diffusivity and inventory as a function of the inventory at Sunset Beach were defined. Water moving onshore was assumed to have a similar activity to that measured far offshore at Santa Monica Basin and San Pedro Basin, 15 dpm m'3 . The seafloor and shoreline input fluxes can then be adjusted to find the best fit to the mean summertime 2 2 8 Ra surface water profile (Figure 3-22). The best fit required an input flux between 4.0 and 6 . 6 xlO6 atoms s' 1 (m shoreline) ' 1 distributed across the seafloor. This is an order of magnitude greater than the lower limit estimate of 0.5 xlO6 atoms s" 1 (m shoreline) ' 1 estimated by the measured input fluxes in Chapter 2. Since the 2 2 8 Ra flux is sensitive to sediment-water exchange processes occurring at rates much slower than those measured, the 2 2 8 Ra flux computed in Chapter 2 is a lower limit. Thus, an additional 2 2 8 Ra input could come from deeper sediments, 169 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. either at the shoreline or across the seafloor. Assuming the source of additional 2 2 8 Ra is from the seafloor, the volume of sediments required to produce the modeled input flux can be estimated using the 2 2 8 Ra emanation rate estimated in Chapter 2. With an emanation rate of 27 dpm kg"1 , between 5 and 8 m of sediment must be flushed to account for the observed enrichment. This is not reasonable for mechanisms such as bio-irrigation or wave or current induced circulation. Instead, terrestrially-derived groundwater may drive an advection of water through sediments. Assuming pore water has an equilibrium activity of 8 dpm L"1 , then a groundwater flux of 13 m3 d" 1 (m shoreline) ' 1 is required, which is equivalent to upward advection of 1 . 0 cm d" 1 across the seafloor in contact with the mixed layer. This groundwater flux is an upper limit estimate, since low estimates were used for the 2 2 8 Ra pore water equilibrium concentration, the input flux from irrigation of the seafloor, and the input flux from the tidal wedge. Discussion Current meter data offshore from Huntington Beach show that during the summer, surface water is flowing toward the southeast, parallel to the shoreline (Noble et al., 2003). But, the Ra observed in surface water at Sunset Beach is not observed at Huntington Beach. Instead, less than 25% of the inventory observed at Sunset Beach can be contributing to the inventory at Huntington Beach. This estimate is similar to the box model results presented in Chapter 2. This is equivalent to less than 25% of 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the flow at Huntington Beach moving down the coast from Sunset Beach. In order to conserve mass, the remainder of the water must originate offshore. Over periods of days to a week, the short-lived isotope Ra inventories in surface water offshore of Huntington Beach remained relatively stable. This is similar to current meter data that showed flow fluctuations on the inner shelf with periods of 5 to 10 days (Noble et al., 2003). Over longer periods, however, significant changes in the Ra inventories were observed. Inventory fluctuations did not correlate with environmental factors, which may generate variations in the input fluxes. Inventory fluctuations may occur without changing the input flux. For example, Ra inventory changes at Huntington Beach may be generated by inventory changes at the upstream boundary (e.g. Figure 3-15), which may reflect changes in the relative fractions of Sunset Beach and offshore water advected down the coast. Based on this model, inventory fluctuations at Huntington Beach could be accounted for by varying the inventory at the upstream boundary between 0 and 25% of the inventory observed at Sunset Beach. Or, the inventory at the boundary may stay constant, but there may be changes in the residence time of water between the upstream boundary and the Huntington Beach transect. The residence time is controlled by two factors: the longshore advection rate and the distance between the onshore advection and the Huntington Beach transect. For most of these model, a longshore advection rate of 5 cm s' 1 and a distance of 10 km between the boundary and the Huntington Beach transect was used. Given a constant inventory at the upstream boundary, increasing 171 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the advection rate or moving the boundary closer to Huntington Beach would decrease the residence time of water in the upstream region. This decreases the contribution of Ra inputs from the shoreline and seafloor to the inventory, reducing the inventory observed at Huntington Beach. Decreasing the advection rate or increasing the distance to the boundary increases the residence time, increases the contribution of Ra inputs, and increases the inventory (Figure 3-23). With the given data set, it is not possible to distinguish between these different mechanisms. Changes in the mixing rate alters the Ra distribution but not the inventory. By changing the Ra distribution, mixing rate changes would produce a negative correlation between the shoreline concentration and the scale distance in Figure 3-5. Instead, there is clearly a positive correlation. To generate a positive correlation requires both inventory fluctuations and only small mixing rate variations. Large changes in the mixing rate would obscure the observed positive correlation. Changes in the inventory produce changes in the average shoreline Ra concentration (Figure 3-5). Changes in the shoreline concentration were also observed at time scales of hours (Figure 3-1). Two mechanisms were presented to explain the Ra concentration variability at the shoreline. First, by varying the shoreline flux with the phase of the tide, the model captures some of the observed shoreline concentration fluctuations. However, the model underestimates the total variability (calculated 15%, observed 25%), and does not account for higher frequency concentration changes. Short period concentration fluctuations may be produced by 172 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 224R a (d p m m'3) 223R a ( d p m m'3) Figure 3-23. Model results for various advection rates. Using the best fit values for the inventory at the boundary and the mixing rate computed for an advection rate of 5 cm sec'l, the advection rate was reduced to 2 cm sec . 70 • 223Ra (dpm in-3) 223Ra (u=5; f=0.32; Kh=Kb=1.31) 223Ra (u=3; f=0.32; Kh=Kb=1.31.) ■ ■ • 223Ra iu-2; f=0.32: kh=KI>=1.3!) 60 50 40 30 20 10 •is#, 0 0 1 2 3 4 5 6 700 • 224Ra (dpm m-3) 224Ra (u=5; f=0.32; Kh=Kb=1.31) 224Ra (u=3; f=0.32; Kh=Kb=1.31) * - ■ 224Ra (u=2: 1=0.32: Kb=Kb=l.3J) 600 500 300 m 200 ■v • i V v 100 0 1 2 3 4 5 6 Distance Offshore (km) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. changes in the nearshore mixing rate. There is some evidence for a change in the nearshore mixing rate at various time scales. On time scales of hours, winds pile-up water at the shoreline (Noble et al., 2003). This may reduce the rate of offshore transport in the afternoon. On shorter time scales, there may be fluctuations in surf zone circulation. The surf zone at Huntington Beach has a bar-trough morphology. Breaking waves carry water over the bar and into the trough, then flow back offshore at breaks in the bar where rip currents form. Changes in the rip current velocity should reflect changes in the residence time of water in the trough. Fluctuations at the period of infra-gravity waves (0.5 - 4 minutes) are well recognized (MacMahan et al., 2004). In addition, there is some evidence of fluctuations at longer periods (MacMahan et al., 2004). However, it is unclear whether these fluctuations are sufficient to produce the large difference in mixing rate required to produce the observed concentration fluctuations. The estimated eddy diffusivities at Huntington Beach are similar to previous estimates of eddy diffusion in the surf zone (Inman et al., 1971) and at the head of rip currents, just outside the surf zone (Johnson and Pattiaratchi, 2004). But these rates are orders of magnitude smaller than estimates using the same short-lived Ra tracers further offshore (Moore, 2000). This is evidence that the mixing rate may increase as a function of distance offshore. Since the nearshore is the primary source for Ra, this would be analogous to scale-dependant diffusion (Okubo, 1971). 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The model was adjusted to account for an increase in mixing rate with distance offshore, using the following relationship: Kh = cxa (3-8) where c has dimension of m2 'a s'1 , x is distance offshore in m, and a is 1.15 as defined by Okubo (1971; 1976). To fit the model to the data, c and the inventory at the upstream boundary f were varied independently for a longshore advection rate of 5 cm s'1 . Using the model to reduce the y 2 statistic fit the shoreline station, which had the lowest relative uncertainty, but greatly underestimated the concentration at the offshore stations. In Figure 3-24, the model variables were adjusted manually to fit the offshore data. A good fit was found with c equal to 2.13 xlO' 3 m0 8 5 s' 1 and a longshore input equivalent to 30% of the inventory at Sunset Beach. Visually, the model fits the offshore data quite well, and may account for some of the elevated concentrations observed at 3 km offshore. But at the shoreline, the model-predicted concentrations are twice the observed average concentration, which results in a poor fit statistically. There are two possible explanations for high model-predicted shoreline concentrations: Ra inputs that are too high or a Kh that is too low. When total model inputs were reduced by 50%, the Ra distribution changed little. Instead, to reduce the shoreline concentration, the mixing rate must be increased. This was simulated in the model by defining the mixing rate at the seaward end of the first box, 50 m 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-24. Model generated profiles for scale-dependant Kh. Longshore advection is 5 cm sec'1. £ a . 0 4 N 0 70 • 223Ra (dpm nT-5 ) 223Ra (f=8%; Kh=1.31 m2 sec’1) 223Ra (f=30%; Kh = 0.00213xL15) • - - 223Ra <f=30%; Kh = 0.00213 x U 5 ; K ji=K i2) 60 50 40 30 20 10 0 N 200 3 r 224Ra (dpm m‘3) 224Ra (f=8%; Kh=1.31 m2 sec'1) 224Ra (f=30%; Kh = 6x1-15) 224Ra (f=30%; Rii = fix1-15; K ^ -K ^ i $ . . . . 2 3 4 Distance Offshore (km) 5 6 176 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-25. Composite summertime 228Ra at Huntington Beach with model best-fit results for constant mixing rate and a scale-dependant mixing rate. The fractional inventory of water from Sunset Beach and the diffusivity were set based on the model results for 223Ra and 224Ra at each advection rate. The seafloor flux, J (106 atoms sec' 1 (m shoreline)'1 ), was adjusted to find the best fit to the data. 160 • 228Ra (dpm m-3) ------------ 228Ra (u=5; f=8%; Kh=1.31; J = 5.2) ----------- 228Ra (u=5; f=30%; Kh=6x1-15; J = 3.4) 140 r 120 < * - 100 10 4 6 8 0 2 Distance Offshore (km) 177 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. offshore, equal to the second box, 105 m offshore. While the shoreline concentration remained high, it was reduced to within 13% of the average concentration (Figure 3- 24). Further offshore, the distribution is indistinguishable from the previous model. This model was also applied to the 2 2 8 Ra data (Figure 3-25). Again, using the model to reduce the y 2 statistic fit the shoreline station but greatly underestimated the concentration at 1 km offshore. The best fit to the data was found by adjusting the seafloor input flux to 3.4 xlO6 atoms s'(m shoreline)1 , 40% lower than the input flux estimated with the constant Kh model. To improve the fit, either the 2 2 8 Ra seafloor input flux 1 km offshore must be increased and/or the nearshore mixing rate must be increased. Mixing in the surf zone has previously been found to be independent of the scale of diffusion, but instead to be related to the height and period of the waves (Inman et al., 1971). The offshore diffusivity in the surf zone can be approximated using the following relationship (Inman et al., 1971): K h s (H ™ ) b X b (3-9) where (H^),, is the root-mean squared breaker height, Xb is the width of the surf zone, and T is the period of the spectral peak of the wave energy spectra. At Huntington Beach, the surf zone extends approximately 80 m offshore. There are generally two peaks in the wave energy spectra, at about 5 sec for locally produced 178 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. wind waves, and 12 sec for swell originating out in the Pacific Ocean (Karl et al., 1980). Assuming a (Hr m s )b of 0.5 m, values for Kh range between 3 and 8 m2 s'1 , depending on the value of T used. Diffusion in the surf zone decreases with depth (Inman et al., 1971), so this surface water estimate is an upper limit. The change in diffusive processes from breaking waves in the surf zone to transport by eddies of various sizes most likely leads to a decrease in the diffusivity outside of the surf zone. But then at some distance beyond the surf zone, scale dependent mixing must begin. The result is that the mixing rate as a function of distance offshore passes through a minimum outside of the surf zone. For a solute with a nearshore source, this minimum mixing rate is a barrier for exchange. Without a smaller sampling scale, it is impossible to determine from the data the distance offshore where the minimum mixing rate occurs. But the distance can be estimated based on this modeling. The diffusivity computed based on the constant Kh model, 1.3 m2 s'1 , should be strongly influenced by the minimum mixing rate. Then, based on the variable Kh model, this mixing rate corresponds to a scale distance of 270 m. This should be the minimum distance offshore, since the eddies at this scale may be pushed further offshore by the presence of the surf zone. The variable mixing rate estimate can be compared to previous published estimates (Figure 3-26). The mixing rate estimate is best constrained over scale distances used in the model, 10’s of meters up to 10 km (103 to 106 cm). Since the scale relationship 179 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3-26. Diffusivity as a function of distance from a source. Comparison of results from this study with other published results. Box labeled "Surf Zone" outlines range of diffusivity values measured in the surf zone. 1 0 - Okubo, 1971 - Okubo, 1976 - Winant, 1983 Moore, 2000 Tins Study IVar 'Kh) This Study (Const Kh ,4 ,5 , 2 |6 Distance from Source (m) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was the same as that presented by Okubo (1971; 1976), slope of the line must be the same. The absolute value for diffusivity is only slightly greater than Okubo’s revised 1976 mixing rate estimate. This suggests that mixing rates derived from open ocean experiments are valid in coastal environments. However, between the shoreline and the end of the surf zone, the diffusivity estimated for the open ocean is underestimated. Diffusion in the surf zone is much faster than would be predicted based on width of the surf zone (~80 m). Conclusions A two dimensional diffusion-reaction-advection model was developed to compute the nearshore steady-state distribution of Ra isotopes between Sunset Beach and Newport Beach, CA. The input fluxes were based on measurements made at the shoreline and seafloor offshore of Huntington Beach. Based on previous work, summertime mean longshore advection rates are from the north at between 2 and 7 cm s'1 . Using a mean longshore advection rate of 5 cm s'1 , the mean 2 2 3 Ra and 2 2 4 Ra surface-water offshore profiles could be explained with a horizontal eddy diffusivity rate of 1.3 ±0.2 m2 s'1 , and an inventory at the northern boundary of the model of 2.3 xlO3 and 1.2 xlO4 atoms s' 1 (m shoreline) ' 1 for 2 2 3 Ra and 2 2 4Ra, respectively. This inventory is 8 % of the observed inventory in surface water at Sunset Beach, just north of the model boundary. 181 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Over time scales of days to weeks, the inventory of Ra in surface water changed little. However, over longer periods, the inventory was observed to change by as much as 50%. The Ra inventory was found to be a function of the scale distance, which in turn depends on the Ra concentration observed at the first station offshore (~ 1 km). At this distance offshore, the Ra concentration is not sensitive to changes in the mixing rate. In order to change the concentration at this station requires a change in the inventory at the upstream boundary. To fit most of the data requires eddy diffusivities between 1 and 2 m2 s' 1 and an inventory at the northern boundary between 0 and 25% of the inventory at Sunset Beach. This range includes uncertainties in the input rates, the offshore distribution of the inputs, and advection rates greater than 2 cm s'1 . Changes in the Ra inventory at Huntington Beach may also be generated by moving onshore advection closer to Huntington Beach. At the shoreline, fluctuations in Ra concentration observed over the course of a day could be explained with two mechanisms. First, the input flux from the shoreline primarily occurs during the ebb tide. When this input is concentrated over just a fraction of the tidal cycle, 10-15% fluctuations in the shoreline concentration are produced. In addition, fluctuations in the offshore mixing rate may also contribute to the variability in concentration observed at the shoreline. The latter fluctuations may produce 2 0 % changes in the shoreline concentration over time periods less than 1 hour. 182 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mixing in the open ocean is known to be scale dependant, increasing as a function of distance from its source. However, the best model fit to the short-lived Ra data was with a constant mixing rate. In addition, the measured mixing rate was similar to previously measured rates from the surf zone and inner shelf. If scale dependent mixing occurs in the coastal ocean, it must begin at some distance offshore. The model-derived mixing rate of 1.3 m2 s" 1 requires a scale distance of about 270 m. Thus, this represents the minimum distance offshore to where scale-dependent mixing begins. Within the first km offshore, where the greatest Ra concentration gradient occurs and the model is the most sensitive, the change in mixing rate is small and may approximate Fickian diffusion. Further offshore, scale dependant mixing may occur, but the Ra distribution presented here is not sensitive enough to constrain any increase in diffusivity. The modeled 2 2 8 Ra results were consistent with the short-lived isotopes. For 2 2 8Ra, the constant mixing rate model provided a satisfactory fit to the data collected within the first 1.5 km offshore. However, this model was unable to account for elevated concentration measured further offshore and overestimated the observed longshore concentration gradient. With the variable mixing rate model, the longshore concentration gradient could be reduced and the higher offshore concentrations fit, but the nearshore concentration was underestimated. Therefore, a combination of these two models, with a constant mixing rate nearshore that change to scale dependent mixing, may provide the best fit to 2 2 8Ra. 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 4. Identifying coastal sources of 2 2 6 Ra in San Pedro Bay, CA Abstract The coastal 2 2 6 Ra budget is sensitive to two sources: release from suspended sediments in estuaries and groundwater discharge. These two sources were assessed to balance the 2 2 6 Ra budget in San Pedro Bay and Santa Monica Bay, CA and two adjoining estuaries: Talbert Marsh and the Santa Ana River. There was little variability in the 2 2 6 Ra distribution beyond 2 km offshore, with an average activity of 98 ± 1 dpm m'3 . The average activity of water draining out of the Talbert Marsh and Santa Ana River was 10% greater. For the Santa Ana River, the additional 2 2 6 Ra may be derived from the suspended sediments. The Talbert Marsh requires a mean groundwater input of 0.03 cm d'1 . Diffusion and bio-irrigation had a negligible impact on the 2 2 6 Ra budget of these estuaries. The highest 2 2 6 Ra activities were measured at the shoreline within 1 km of these estuaries, but the high activities were not measured in water draining out of the marsh. Shoreline 2 2 6 Ra activities may be increased by localized groundwater discharge. Introduction Coastal aquifers are a conduit where fresh, terrestrially-derived groundwater interacts with seawater. From the aquifer outcrop, a wedge of salt water lies beneath fresh water, with the boundary between the two approximated by the Gyben- Herzberg relationship. Between the salt and fresh water is a zone of dispersion, 184 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. generated by diffusion and tidal pumping (Cooper et al., 1964). As a result of mixing, water discharged from coastal aquifers is often brackish. To maintain the salt balance, seawater must be continually drawn into the aquifer (Kohout, 1960). Because of this circulation and the similarity of chemical reaction occurring, coastal aquifers are identified as “subterranean estuaries” (Moore, 1999). Submarine groundwater discharge (SGD) is important to the chemistry and biology of the coastal ocean but has been difficult to quantify (Burnett et al., 2003; Burnett et al., 2002; Moore, 1996; Moore, 1999; Zekster and Loaiciga, 1993). One possible tracer of SGD is 2 2 6 Ra (Burnett et al., 1990; Moore, 1996). 2 2 6 Ra is a naturally occurring radioactive isotope with a 1600 y half-life. 2 2 6 Ra is produced by a long- lived Th parent, which is primarily found associated with sediments. The solubility of 2 2 6 Ra is partly dependent on salinity and increases significantly between 0 and 4 PSU (Burnett et al., 1990). The 2 2 6 Ra activity of brackish water discharged from coastal aquifers can be 1 0 0 times greater than fresh water collected from the same aquifer and at least 10 times greater than coastal ocean activities (Moore, 1996). With such a large activity difference, SGD has the potential to be an important component of the 2 2 6 Ra budget for the coastal ocean. Another important source of 2 2 6 Ra to the coastal ocean is estuaries (Cochran, 1992). Suspended sediment carried by rivers will desorb 2 2 6 Ra as the salinity increases in the estuary. The dissolved activity responds quickly to changes in solubility, with equilibrium between adsorbed and dissolved 2 2 6 Ra occurring on time scales of 185 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. minutes (Krishnaswami et al., 1982). Desorption of 2 2 6 Ra from suspended sediments accounts for 2 2 6 Ra excesses in several estuaries, including the Pee Dee, Hudson, Amazon, and Mississippi Rivers (Elsinger and Moore, 1980; Key et al., 1985; Li and Chan, 1979; Moore and Scott, 1986). In this study, the distribution of 2 2 6 Ra in the coastal ocean off southern California is presented. Elevated 2 2 6 Ra activities were observed both at the shoreline and in water flowing out of the Talbert Marsh. The source of the excess 2 2 6 Ra in Talbert Marsh, either from suspended sediments or groundwater discharge to the marsh, is assessed. Then, the source of nearshore 2 2 6Ra, either groundwater discharge at the shoreline or outflow from the Talbert Marsh or Santa Ana River, is assessed. Study Site Santa Monica Bay and San Pedro Bay are relatively broad continental shelves located near the middle of the Southern California Bight. The climate is Mediterranean, characterized by wet, mild winters and warm, dry summers. One river discharges into Santa Monica Bay, Ballona Creek. Three rivers discharge into San Pedro Bay: the Los Angeles, San Gabriel, and Santa Ana Rivers. Most of the base flow in these three rivers is diverted upstream to recharge groundwater. Below these diversions, dry-weather runoff, treated domestic sewage, and industrial discharges produce a minor flow. For the Santa Ana River, base flow at USGS gauging station 11088500 in Fullerton is 2000 m3 h r1 . 186 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Talbert Marsh is located adjacent to the Santa Ana River in Huntington Beach. Three flood control channels (Fountain Valley, Talbert, and Huntington Beach) discharge into the upper part of the marsh. Because of the flat topography, a pump station, consisting of a forebay and several pumps, transfers water from the flood control channels to the marsh. The pumps operate intermittently during dry weather periods, and continuously during storms. Typical dry weather, fresh water discharge into the Talbert Marsh is generally less than 95 m3 h r 1 (Grant et al., 2001). During high tides, open ocean water floods the lower reaches of the open channels. During low tides, a brackish mixture of ocean water and runoff drains from the system (Grant et al., 2001). The Talbert Marsh exchanges 2.8 xlO5 m3 of water on an average tidal cycle (Grant et al., 2003). In Santa Monica Bay, four aquifers are known to be in contact with the coastal ocean. At the south end of the bay, between El Segundo and Redondo Beach, the Upper Pico, Silverado and the 200 ft Sand overlie one another. The other major aquifer is the west arm of the Gaspur Aquifer, an old river channel that lies below Ballona Creek. At the north end of the bay, small drainages in the Santa Monica Mountains also discharge to the coastal ocean. Groundwater discharge to San Pedro Bay is inhibited by the Newport-Inglewood Fault Zone. Uplift along the fault, which defines the shoreline from Long Beach to Newport Beach, has offset deep aquifers against impermeable materials and/or produced an impermeable fault gouge. As a result, only the most recent aquifers, 187 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. relict river channels formed during the last glaciation, are in contact with the ocean. These aquifers include the south arm of the Gaspur, the Recent Aquifer at Los Alamitos, and the Talbert Aquifer at Huntington Beach. These aquifers intersect deeper aquifers that have been uplifted across the Newport-Inglewood Fault Zone (Poland, 1959). Across the entire groundwater basin is a regional aquitard that prevents any downward migration of contaminants, from underground storage tanks and fertilizer, into drinking water supplies (Herndon, 1992). Above this aquitard is a shallow, unconfined aquifer, which receives recharge by rain in the winter and year round from watering lawns and plant nurseries. Since the 1950’s, seawater intrusion has been a problem in the coastal aquifers of Los Angeles and Orange Counties. To protect the water quality in the basin, injection wells were installed at the most sensitive locations: along the south part of Santa Monica Bay, at Alamitos Gap, and at Talbert Gap. Ideally, the injection wells would maintain a 2-m freshwater head, preventing seawater intrusion. However, due to sustained deep water table depressions, complex geometry of the barriers, sediment heterogeneities, and difficulties in maintaining injection rates, some seawater does flow through the barriers. With a continued landward migration of seawater in the Alamitos Gap and Talbert Gap throughout the time samples were collected, SGD from these aquifers is unlikely (R. Herndon, pers. comm.). 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Methods Samples were collected in 20 L glass carboys. 2 2 6 Ra was measured based on 2 2 2 Rn ingrowth (Mathieu et al., 1988). Samples were first flushed with He to remove any initial 2 2 2Rn. After 2 weeks, more than 90% of the 2 2 2 Rn has grown into secular equilibrium with 2 2 6Ra. The 2 2 2 Rn was then extracted by recirculating He through the sample and passing it over an activated charcoal cold trap. After 90 minutes, 99% of the Rn is removed from the sample. Then, the Rn is released by heating the charcoal trap to 400°C for seven minutes and transferred to a Lucas scintillation cell. Light pulses produced by the alpha decay of 2 2 2 Rn and its daughters are monitored with a photomultiplier-based counter (Applied Techniques). The efficiencies of each extraction system and counting cell has been computed based on JGOFS, EPA and internal standards. The computation of the sample activity accounts for any disequilibrium between 2 2 2 Rn and 2 2 6 Ra at the time of sample processing. Groundwater samples from the Silverado Aquifer were collected from monitoring wells using a bailer and transferred to 20 L glass carboys. At Los Alamitos, samples were collected at the well head, where water is continuously pumped to limit the extent of seawater intrusion. The salinity of these samples was measured using a flow-through conductivity meter with an accuracy of + 0.02 PSU. 189 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table4-1. 2 2 6 Ra samples collected in Santa Monica Bay and San Pedro Bay: A) offshore samples, B) shoreline samples, and C) estuary samples. Sample Distance “ * Re Salinity Temp. Sample Sample Data Tima Location Depth (m) Offshore flan) (ttom m ) (PSU) (*C) A) Offshore Sample* HB1 1 0 /1 9 /0 0 8:45 Huntington Beach 1.5 0.784 103.7 ± 5.2 HB2 1 0 /1 9 /0 0 10:00 Huntington Beach 1.5 0.711 95.3 ± 4.8 HB3 1 0 /1 9 /0 0 10:40 Huntington Beach 1.5 0.701 118.3 ± 5.9 HB11 1 2 /1 /0 0 9:1 5 Huntington Beach 1.5 0.744 94.8 ± 4.7 HB12 1 2 /1 /0 0 10:20 Huntington Beach 1.5 0.649 98.9 ± 4.9 HB13 1 2 /1 /0 0 11:05 Huntington Beach 1.5 2.433 94.4 ± 4.7 3* 8 /2 8 /0 1 9:00 Huntington Beach 0.5 3.5-8 77.7 ± 3.9 4 8 /2 8 /0 1 12:00 Huntington Beach 0.5 3.5-8 78.3 ± 3.9 5 8 /2 8 /0 1 15:00 Huntington Beach 0.5 3.5-8 97.3 ± 4.9 6 8 /2 8 /0 1 18:00 Huntington Beach 0.5 3.5-8 80.7 ± 4.0 7 8 /2 8 /0 1 21:00 Huntington Beach 0.5 3.5-8 88.6 ± 4.4 1.1 9 /1 1 /0 1 11:20 Sunset Beach 1.5 2.27 84.9 ± 4.2 33.5 20.1 1.2 9 /1 1 /0 1 11:20 Sunset Beach 7.6 2.27 98.6 ± 4.9 33.6 18.4 3.3 9 /1 1 /0 1 17:45 Sunset Beach 1.5 9.22 96.9 ± 4.8 33.6 19.5 5.1 9 /1 3 /0 1 12:55 Huntington Beach 18.0 9.86 97.7 ± 4.9 5.3 9 /1 3 /0 1 12:49 Huntington Beach 1.5 9.86 70.9 ± 3.5 33.6 19.5 7.1 9 /1 3 /0 1 17:00 Huntington Beach 11.0 1.35 93.7 ± 4.7 7.3 9 /1 3 /0 1 16:53 Huntington Beach 1.5 1.35 89.4 ± 4.5 33.4 19.9 8.1 9 /20/01 8:44 Sunset Beach 19.5 9.05 105.8 ± 5.3 8.3 9 /20/01 8:44 Sunset Beach 1.5 9.05 89.4 ± 4.5 33.3 18.1 10.1 9 /2 0 /0 1 12:55 Sunset Beach 8.6 1.65 94.6 ± 4.7 33.3 17.9 10.2 9 /2 0 /0 1 12:55 Sunset Beach 1.5 1.65 107.4 ± 5.4 33.3 19.4 11.3 11/1/01 11:35 Hermosa Beach 1.4 6.16 92.3 ± 4.6 1 7.4 13.1 11/1/01 15:40 Hermosa Beach 1.8 0.58 98.2 ± 4.9 17.0 21.1 4 /3 /0 2 9:45 Huntington Beach 9.1 1.67 102.1 ± 5.1 21.3 4 /3 /0 2 9:45 Huntington Beach 1.5 1.67 112.7 ± 5.6 13.5 24.1 4 /3 /0 2 15:45 Huntington Beach 13.7 11.89 94.3 ± 4.7 24.3 4 /3 /0 2 15:45 Huntington Beach 3.9 11.89 97.9 ± 4.9 12.8 25.1 6 /5 /0 2 10:30 Sunset Beach 21.0 7.28 90.9 ± 4.5 14.9* 25.3 6 /5 /0 2 10:30 Sunset Beach 1.5 7.28 100.1 ± 5.0 32.9 17.3 27.1 6 /5 /0 2 15:55 Sunset Beach 6.1 1.90 95.6 ± 4.8 33.2 17.4 27.2 6 /5 /0 2 15:55 Sunset Beach 1.5 1.90 122.1 ± 6.1 33.1 19.3 28.1 6 /7 /0 2 10:05 Huntington Beach 1.5 1.08 114.8 t 5.7 33.2 18.7 28.2 6 /7 /0 2 10:05 Huntington Beach 6.1 1.08 100.0 ± 5.0 33.2 17.3 31.2 6 /7 /0 2 16:50 Huntington Beach 1 5.1 12.48 91.3 ± 4.6 33.2 14.9* 31.4 6 /7 /0 2 16:50 Huntington Beach 1.5 12.48 91.2 ± 4.6 33.2 17.7 32.1 6 /1 2 /0 2 10:00 Huntington Beach 1.5 0.99 96.8 ± 4.8 33.1 18.8 32.2 6 /1 2 /0 2 10:00 Huntington Beach 7.6 0.99 94.6 t 4.7 33.1 18.5 35.1 6 /1 2 /0 2 16:45 Huntington Beach 1.5 10.97 92.2 ± 4.6 33.3 18.0 35.2 6 /1 2 /0 2 16:45 Huntington Beach 1 5.7 10.97 94.5 ± 4.7 33.3 17. 3* 36.1 6 /2 1 /0 2 10:00 Hermosa Beach 1.5 5.42 92.3 ± 4.6 33.3 19.2 36.2 6 /2 1 /0 2 10:00 Hermosa Beach 23.6 5.42 94.5 ± 4.7 33.3 13.5 38.1 6 /2 1 /0 2 13:40 Hermosa Beach 1.5 0.59 92.5 ± 4.6 33.3 19.5 40.1 6 /2 2 /0 2 9:40 Santa Monica 1.5 1.12 96.4 ± 4.8 33.4 19.3 40.2 6 /2 2 /0 2 9:40 Santa Monica 6.1 1.12 104.8 ± 5.2 33.4 19.2 42.1 6 /2 2 /0 2 14:10 Santa Monica 1.5 5.02 106.4 ± 5.3 33.5 19.0 42.2 6 /2 2 /0 2 14:10 Santa Monica 15.7 5.02 101.3 ± 5.1 44.2 14.1 43.2 7 /2 6 /0 2 11:20 Newport Beach 1.5 0.34 84.8 ± 4.2 33.4 18.5 43.3 7 /2 6 /0 2 11:20 Newport Beach 7.6 0.34 85.4 ± 4.3 33.4 17.6 46.2 7 /2 6 /0 2 15:30 Sunset Beach 1.5 0.72 105.4 ± 5.3 33.3 20.0 46.3 7 /2 6 /0 2 15:30 Sunset Beach 4.6 0.72 101.6 ± 5.1 33.3 16.4 47.1 8 /1 9 /0 2 9:35 Huntington Beach 1.5 9.31 111.0 ± 5.6 33.6 20.2 47.2 8 /1 9 /0 2 9:35 Huntington Beach 24.5 9.31 106.2 ± 5.3 50.1 8 /1 9 /0 2 14:30 Huntington Beach 1.5 0.68 101.9 ± 5.1 33.4 20.3 51.1 8 /2 3 /0 2 9:45 Huntington Beach 1.5 8.58 89.8 ± 4.5 33.5 19.6 54.1 8 /2 3 /0 2 14:10 Huntington Beach 1.5 0.67 84.9 ± 4.2 33.2 20.9 190 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-1 Continued. Sample Sample Date Time Location Sample Depth (m) Distance Offshore (km) 2 2 4 to (dam m"2) Salinity (PSU) Temp. fC ) 54.2 8 /2 3 /0 2 14:10 Huntington Beach 9.1 0.67 93.8 ± 4.7 33.3 17.5 55.1 9 /8 /0 2 9:55 Hermosa Beach 1.5 4.10 101.9 ± 5.1 33.4 18.4 55.2 9 /8 /0 2 9:55 Hermosa Beach 30.1 4.10 101.3 ± 5.1 58.1 9 /8 /0 2 14:40 Hermosa Beach 1.5 0.61 107.5 ± 5.4 33.4 19.7 58.2 9 /8 /0 2 14:40 Hermosa Beach 7.6 0.61 102.8 ± 5.1 33.3 1 3.4 B) Huntington Baach Shoreline Sample*** 12N 1 0 /1 9 /9 7 14:04 Huntington Beach 0.2 0 110.0 ± 5.5 3N 1 0 /1 9 /9 7 13:22 Huntington Beach 0.2 0 130.0 ± 6.5 3S 1 0 /1 9 /9 7 12:09 Huntington Beach 0.2 0 100.0 ± 5.0 3S 1 0 /1 9 /9 7 12:30 Huntington Beach 0.2 0 100.0 ± 5.0 9N 1 1 /2 0 /9 7 14:38 Huntington Beach 0.2 0 100.0 ± 5.0 9N 1 1 /2 0 /9 7 13:39 Huntington Beach 0.2 0 100.0 ± 5.0 3N 1 1 /2 0 /9 7 13:10 Huntington Beach 0.2 0 150.0 ± 7.5 3S (Flood Tide) 1 1 /2 0 /9 7 11:28 Huntington Beach 0.2 0 140.0 ± 7.0 HB4(6N) 1 0 /1 9 /0 0 16:00 Huntington Beach 0.2 0 96.6 ± 4.8 HB5 (3N) 1 0 /1 9 /0 0 17:25 Huntington Beach 0.2 0 102.9 ± 5.1 HB6 (3S) 1 0 /1 9 /0 0 18:45 Huntington Beach 0.2 0 114.0 ± 5.7 HB14(6N) 1 2 /1 /0 0 14:25 Huntington Beach 0.2 0 99.4 ± 5.0 HB15 (3N) 1 2 /1 /0 0 15:20 Huntington Beach 0.2 0 90.1 ± 4.5 HB16 (3S) 1 2 /1 /0 0 16:10 Huntington Beach 0.2 0 136.5 ± 6.8 O Estuary SamplesAA HB7 (Santa Ana River) 1 0 /2 3 /0 0 10:1 5 Santa Ana River 0.2 0 103.3 ± 5.2 HB17 (Santa Ana River) 1 2 /3 /0 0 1 5:20 Santa Ana River 0.2 0 111.9 ± 5.6 Santa Ana River 1 0 /1 9 /9 7 13:00 Santa Ana River 0.2 0 110.0 ± 5.5 Santa Ana River 1 1 /2 0 /9 7 11:59 Santa Ana River 0.2 0 110.0 ± 5.5 HB8 (Talbert Marsh) 1 0 /2 3 /0 0 11:20 Talbert Marsh 0.2 0 97.2 ± 4.9 HB18 (Talbert Marsh) 1 2 /3 /0 0 16:05 Talbert Marsh 0.2 0 117.0 ± 5.9 Tafeert Marsh 1 1 /2 0 /9 7 12:41 Talbert Marsh 0.2 0 140.0 ± 7.0 * Samples collected on 8 /2 8 /0 1 were along a series of transects at Huntington Beach. Transects were equally spaced, with Transect 3 offshore from the Huntington Beach Pier and Transect 6 offshore from the Santa Ana River. ** Numbering for OCSD surf zone monitoring stations included. Numbered from Santa Ana River (0) in -1 ,0 0 0 ft (300 m) increments, (eg. 6N is -6 0 0 0 ft (1800 m) north of River) A Reported tem perature is a maximum. Temperature measured after water pumped to the surface. A A Samples collected at Pacific Coast Highway overpass, -2 5 0 m from the mouth. Bold samples collected during the flood tide. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-2. Groundwater 2 2 6 Ra concentrations. Well D ata C ollected A quifer Tapped Salinity (p er mil) 226 Ra (dpm m ~ 3 ) W est Coast Barrier Observation Wells 6AB 3 /2 5 /9 9 2 0 0 Ft.-Silverado 0.03 4 1 .0 ± 3.0 5YA 3 /2 5 /9 9 2 0 0 Ft.-Silverado 0.12 6 3 .0 ± 4.0 6C-10 3 /2 5 /9 9 2 0 0 Ft.-Silverado 0.48 9 6 0 .0 ± 2 0 .0 8P-49 3 /2 5 /9 9 Silverado 0.63 4 7 0 .0 ± 10.0 Allmttos Barrier Observation Wells 34 -1 8 4 /1 6 /9 9 Alimitos Gap B 0.6 4 4 8 0 .0 ± 10.0 34-17 4 /1 6 /9 9 Alimitos Gap C 0.80 5 3 0 .0 ± 10.0 3 4 s-2 2 p 4 /1 6 /9 9 Alimitos Gap 1 1.05 7 5 0 .0 ± 10.0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (£ _U I u id p ) r ^ )zz Figure 4-1. All 226Ra data as a function of distance offshore. Samples plotted at 0 km were collected in knee-deep water. 160 Huntington Beach S u n set B each N ew p ort B each Santa M onica Bay S a n ta A n a R iv e r and T albert M arsh 140 120 100, '-f-J Distance Offshore (km) 193 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D epth (m) Figure 4-2. 226Ra concentration collected offshore from Huntington Beach as a function of depth. i 1 -------------* ------------ 1 ■ i ------- ----#-------------1 i 1 ------------ • -------- — 1 ! ------------• ------- H - i---------0------------- f------------• ---------- ■ '------------ • --------- H i — i— • ----------- 1 ------------ • --------- — 1 i 1 -------------• ----------- 1 _ ....... ......... ...............i................................ i...i ..................... ! 1 -------: -------• - — i — 1 2 5 _______________ J______________ u ____________ i_ i 1 _____ * : 1 i 1 ______________ 70 80 90 100 110 120 226Ra (dpm m'3) 194 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4-3. Groundwater 226Ra concentration as a function of salinity. 100(P £ 400 0 0.2 0.4 0.6 0.8 Salinity (PSU) 195 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results The activity of 2 2 6 Ra was measured in samples from Santa Monica Bay, San Pedro Bay, and at the mouth of the Santa Ana River and Talbert Marsh (Table 4-1). The 2 2 6 Ra activity in water collected in Santa Monica Bay and San Pedro Bay had an average activity of 98 ± 1 dpm m' 3 (Figure 4-1). The greatest variability was in samples collected within the first 2 km, near the mouths of the Santa Ana River, Talbert Marsh and Anaheim Bay. The highest activities were measured at the shoreline within 1 km of the mouth of the Talbert Marsh and Santa Ana River. These elevated activities were not persistent and occurred on both the flood and ebb tides. Elevated activities were not observed in Santa Monica Bay. Vertically, there was no activity gradient (Figure 4-2). The average activity of samples collected below 1 2 m was the same as mean surface water. Groundwater 2 2 6 Ra activities were measured in several aquifers from wells located seaward of the West Coast Barrier and the Alamitos Barrier (Table 4-2). In general, there is an increase in activity with salinity (Figure 4-3). Differences from the trend probably reflect different 2 2 6 Ra emanation rates or adsorption coefficients in the different groundwater aquifers. Discussion The average activity of water draining out of the Santa Ana River and the Talbert Marsh was greater than the mean coastal ocean activity. One potential source of 196 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 2 6 Ra is suspended sediment carried by the rivers and drainage channels that feed the Talbert Marsh and Santa Ana River estuaries. The 2 2 6 Ra activity of water entering the marsh required to produce the observed 2 2 6 Ra activities can be estimated using a mass balance. Talbert Marsh receives a water flux of about 2400 m3 d"1 , which would require the 2 2 6 Ra concentration of the river to be 1200 dpm m'3 . It is unlikely that the 2 2 6 Ra activity of the water is this high. To generate this 2 2 6 Ra activity requires an extraordinary source of 2 2 6 Ra in the drainage. The Santa Ana River has a slightly larger volume, and receives a greater flux of fresh water. When its 2 2 6 Ra budget is computed, water entering the estuary must have an activity of 135 dpm m'3 . This may not be unreasonable for the Santa Ana River. Benthic inputs- diffusion, irrigation, and groundwater advection- are another possible source of 2 2 6 Ra to the Talbert Marsh and Santa Ana River. The importance of benthic inputs can be assessed by constructing a mass balance budget for the estuary: G/n^Qn + A 'l d h = Q 0u} - C 0u, (4-1) Where Q is the water flux and C is the 2 2 6 Ra concentration of water for the subscripts In during the flood tide and Out during the ebb tide. A is the surface area of the marsh and Jb is the benthic 2 2 6 Ra flux. Ignoring any surface or groundwater inputs and evaporation, then the flux of water on the flood and ebb tide should be equal (QI n = Q0 u t = Q). The flux of water through the estuary can be estimated by defining Q as: 197 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. e = — (4-2) T , „ where 1 is the mean water depth and t r e s is the residence time of water in the marsh. Normalizing Equation 4-1 by the surface area of the marsh and rearranging to solve for Jb , } J b = — (AC 0 „,-AC,„) (4-3) r res In the Santa Ana River and Talbert Marsh, mean water depth is about 1 m and the residence time of water is less than 1 day (Grant et al., 2003). The difference in activity can be estimated as the difference between the coastal ocean (98 dpm m 3 ) and the mean activity of water draining out of the estuaries (108 dpm m"3 ). The required 2 2 6 Rabenthic flux is 10 dpm rn 2 d'1 . Since the water that floods the estuary most likely comes from near the shoreline, where 2 2 6 Ra activities are higher than the coastal ocean average, this estimate is an upper limit. First, the importance of diffusion can be estimated. The pore water distribution of 2 2 6 Ra can be estimated by a steady-state diffusion-reaction model: d C d C 0 — ^ = 0 = 0 ( 1 + m c.p w (4-4) a t a z where Cp w is the pore water concentration, t is time, z is depth, and c j ) is the porosity. The diffusivity corrected for tortuosity is Ds = ( j) 2Dm (Dm =7 xlO" 6 cm2 s' 1 at 18°C and 198 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 PSU) (Li and Gregory, 1974; Ullman and Aller, 1982). In Chapter 2, the measured partition coefficient of beach sands, K, was 8.3. Two boundary conditions are applied to solve this equation. First, the 2 2 6 Ra concentration in the overlying water is negligible, relative to the equilibrium pore water concentration. Second, deep in the sediments, there is no concentration gradient. The solution to this equation is: and XCe q is the equilibrium groundwater activity. The 2 2 6 Ra activities of groundwater collected from coastal wells are expected to be in equilibrium. However, the groundwater 2 2 6 Ra activity ranged over an order of magnitude (Figure 4-3). Concentrations are expected to continue to increase as salinity continues to increase up to 4 PSU (Burnett et al., 1990). Since the highest salinity measured was 1 PSU, the measured activities probably underestimate the equilibrium groundwater activity at higher salinities. An estimate of the 2 2 6 Ra equilibrium groundwater activity of beach sands can be made based on the 2 2 3 Ra equilibrium pore water activity measured in Chapter 2. Both 2 2 3 Ra and 2 2 6 Ra have a long lived U isotope at the beginning of their decay series, and the distribution of their parent isotopes in beach (4-5) where, (4-6) 199 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sand minerals should be similar. Assuming that the emanatation power, then emanation rates and equilibrium pore water activities of these isotopes should be proportional to the activity of the parent isotopes. Applying a 2 3 8 U:2 3 5 U ratio of 21 to a 2 2 3 Ra equilibrium pore water activity of 1600 dpm m "3 , the estimated 2 2 6 Ra equilibrium pore water activity is 3.4 xlO4 dpm rn3 . The diffusive flux can then be computed using Fick’s First Law (Fick, 1855): For an equilibrium pore water activity of 3.4 xlO4 dpm m'3 , the diffusive flux is 0.14 dpm rn 2 d'1 . At most, the diffusive flux may account for less than 2% of the benthic flux required to balance the 2 2 6 Ra budget for the estuary. Second, benthic animal feeding or physical pumping of water through the seafloor may enhance the exchange rate of pore water solutes with the overlying water column (Aller, 1980; Aller, 1982; Emerson et al., 1984; Hammond and Fuller, 1979; Hammond et al., 1977; Huettel and Gust, 1992; Huettel and Webster, 2001; Luedtke and Bender, 1979; McCaffrey et al., 1980; Ziebis et al., 1996). The maximum contribution of irrigation to the 2 2 6 Ra benthic flux can be estimated by remove the remaining 2 2 6 Ra inventory from surficial sediments. This is computed by integrating over the irrigation depth, H: (4-7) 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. n J Irr = - A f C d z = - A C eq( z + a e l 'a ) (4-8) 0 0 Where 1/a is the scale distance, approximately 1 m. Assuming an irrigation depth H of 0.5 m and AC^ of 3.4 xlO4 dpm m "3 , the irrigation flux A JI r r is 4.3 xlO' 3 dpm m 2 d 1 . Since this is less than 1% of the diffusive flux, sediment irrigation is not an important source of 2 2 6 Ra to the estuary budget. Third, groundwater discharge is another potential source of 2 2 6 Ra to an estuary (Burnett et al., 1990). Assuming groundwater advection is evenly distributed across the estuary, the mean groundwater advection rate can be estimated: C eq Using the larger ACe q of 3.4 xlO4 dpm m'3 , uG W is 0.03 cm d"1 . This would have a negligible impact on the water budget, which exchanges water at a rate of 1 m d'1 . The highest activities, 140 and 150 dpm m 3 , were measured at the shoreline, within 1 km of the Talbert Marsh and Santa Ana River. Due to the proximity, it is possible that these estuaries are the source of the high activity water. However, these high concentrations were not measured in the water draining out of the estuaries. It is possible that the highest activity water is discharged at the end of the ebb tide, which was not sampled. At the low tide, the hydraulic gradient is maximized, increasing the rate of groundwater discharge. In addition, at low tide, groundwater can 201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. contribute a greater fraction of the total flow, as the volume of water in the channels decreases. As the fraction of groundwater and river water increases, the salinity should decrease. This has been observed at Talbert Marsh, where the conductivity, a proxy for salinity, reaches a minimum at the end of the ebb tide (Grant et al., 2001). If the source of high 2 2 6 Ra activity water at the shoreline is discharged from the Talbert Marsh or Santa Ana River, then it must also be transported up to 1 km with little dilution. A previous estimate of initial dilution for discharge from the Talbert Marsh was 1.6 (Grant et al., 2001). Assuming the diluting water has an activity of 110 dpm m'3 , to produce a surf zone concentration of 150 dpm in 3 requires an estuary discharge of 174 dpm m"3 . Then, this water must be transported 1 km down the coast with little additional dilution. The longshore change in concentration can be estimated using the circulation model presented in Chapter 3. To improve the resolution, the geometry of the boxes was reduced to 20 m by 20 m. Eddy diffusivity was assumed constant at 1.3 m2 s'1 . The longshore advection rate was set at 30 cm s'1 , which is regularly observed in the surf zone at Huntington Beach (Grant et al., 2001; Kim et al., 2004). Discharge from the estuary fills the surf zone and a momentum jet can even carry water beyond the breakers (Grant et al., 2001). To simulate the discharge of the Talbert Marsh and Santa Ana River, a boundary activity of 150 dpm m' 3 was assigned between the shoreline and 80 m offshore, approximately the width of the surf zone. Then, eddy diffusivity was adjusted to limit the loss of 2 2 6 Ra from the surf zone to less than 10% 202 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. within 1 km. At an advection rate of 30 cm s'1 , an eddy diffusivity of 0.4 m2 s' 1 is required. This is only 30% of the model-derived offshore mixing rate. While the marsh is a potential source of 2 2 6 Ra in the surf zone, the conditions required must be just right: high 2 2 6 Ra activities discharged at the end of the ebb tide, limited initial dilution in the surf zone, below average offshore mixing rates, and a reduced offshore 2 2 6 Ra gradient. Alternatively, the high shoreline concentrations may result from groundwater discharge at the shoreline. Since elevated concentrations were not continuous along the shoreline, any groundwater discharge must be a relatively local phenomenon. Again, the mixing model from Chapter 3 and modified above can be used to estimate the groundwater flux required to increase the shoreline concentration by 40 dpm m'3 . The input flux to one shoreline box was increased until its activity increased by 40 dpm m'3 . The required input flux to the box was 32 dpm s'1 . The model predicts that elevated 2 2 6 Ra activities are only a local phenomenon, with activities quickly returning to background levels as dispersion transports 2 2 6 Ra offshore (Figure 4-4). Increasing the longshore advection rate made little difference in the shoreline 2 2 6 Ra distribution. At 200 m down coast, the activity increase was only 6 dpm m' 3 above background. The shoreline concentration is quickly diluted as 2 2 6 Ra is mixed offshore. A 2 2 6 Ra flux of 32 dpm s' 1 requires a groundwater discharge flux of 4.1 m3 d' 1 (m shoreline)'1 . The hydraulic gradient required to generate this groundwater flux can 203 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4-4. Estimate of 226Ra concentration at the shoreline derived from a 2D advection-diffusion model. Point source of 226Ra occurs in a single shoreline box at 270 m . Longshore advection rate is 5 cm sec'1, and the horizontal eddy diffusivity is 1.3 m2 sec'1. 140 135 130 125 S 120 & 115 110 105 100 0 100 200 300 400 500 600 Longshore Distance (m) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. be estimated. Similar beach sands from Santa Monica Bay have a hydraulic conductivity of 0.05 cm s" 1 (Emery, 1960). Assuming that discharge occurs within 10 m of the shoreline, a hydraulic gradient of 0.0094 is required. This is about 6 times greater than the hydraulic gradient of 0.0016 measured at 6 N (Chapter 2). Given the uncertainty in the measurements and possible seasonal variations (e.g. the hydraulic gradient was measured during dry summer months, while shoreline sampling was during wet winter months), localized groundwater discharge may be sufficient to increase the 2 2 6 Ra activity at the shoreline. The source of high levels of indicator bacteria at Huntington Beach has recently been the focus of intense research (Boehm et al., 2002; Grant et al., 2003; Grant et al., 2001; Kim et al., 2004; KOMEX et al., 2003; Noble et al., 2003). During 2000, the presence of specific viruses indicative of human waste was measured along with the 2 2 6 Ra activity. These viruses are generally associated with high concentrations of indicator bacteria at Huntington Beach. Viruses were not detected in any of these samples (J. Griffith, pers. comm.). It is unlikely that the high 2 2 6 Ra concentrations measured correlate with high concentrations of indicator bacteria, because the high 2 2 6 Ra activities were generally measured in the afternoon, when the concentration of indicator bacteria is at its lowest (Boehm et al., 2002). 205 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Conclusions The distribution of 2 2 6 Ra was measured in Santa Monica Bay and San Pedro Bay. Beyond 2 km offshore, the activity was very stable, with a mean of 98 ± 1 dpm m "3 . Within the first 2 km, activities were more variable, with the greatest 2 2 6 Ra activity observed within 1 km of the Talbert Marsh, Santa Ana River, and Anaheim Bay. The mean activity of water draining out of the Talbert Marsh and Santa Ana River was 108 ± 3 dpm m 3 . Two sources of additional 2 2 6 Ra to the Talbert Marsh and Santa Ana River were identified: release from suspended sediments and benthic inputs. Release from suspended sediments may account for the additional 2 2 6 Ra observed in the Santa Ana River discharge, but not for the Talbert Marsh. Three components of benthic inputs to the Talbert Marsh and the Santa Ana River were computed: the diffusive flux, the irrigation flux, and the groundwater flux. The diffusive and irrigation fluxes contribute only a small fraction of the total benthic flux required to balance the 2 2 6 Ra budget. The remainder of the benthic input must come from groundwater inputs, which are estimated to average 0.03 cm d"1 . The highest 2 2 6 Ra activities were measured at the shoreline, within 1 km of the Talbert Marsh and Santa Ana River. For the estuaries to be the source of the high 2 2 6 Ra activity water requires specific conditions. These include estuary discharge water with higher 2 2 6 Ra activities than those measured, below average offshore 206 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mixing rates, and a reduction in the offshore 2 2 6 Ra gradient. Alternatively, the high 2 2 6 Ra concentrations represent local discharge of groundwater. 207 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 5. Conclusions and Implications For this dissertation, the quartet of Ra isotopes were measured at San Pedro Bay, CA, with much of the work focused at Huntington Beach. Ra isotopes were measured in the coastal ocean, estuaries, and in pore water. Concentrations of each isotope ranged over several orders of magnitude, with the greatest concentrations measured in pore water and the lowest concentration measured > 1 0 km offshore. This reflects that nearshore sediments are the primary source of Ra to the coastal ocean. The Ra flux from sediments depends on both adsorption and the emanation rate. Adsorption increased offshore and was inversely related to the grain size. The sediment emanation rate decreased from Long Beach down to Dana Point, with a 50% drop between Huntington Beach and Newport Beach. Similar results were found for the total sediment activity, which suggests that the emanation rate is controlled mineral composition of the sediments and not a secondary coating on the sediments. Three sources of Ra were identified: seafloor sediments, shoreline beach sands, and estuaries. Beginning with estuaries, the Ra flux was computed as the product of the volume of water exchanged and the Ra activity in the water draining out. The water flux from the Santa Ana River and Talbert Marsh has previously been measured 208 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Grant et al., 2003). The Ra activity of the water draining out of these estuaries was measured and the Ra flux computed. Pore water profiles of 2 2 2Rn, 2 2 3Ra, and 2 2 4 Ra were measured at two stations in about 8 m water depth. Seawater cycling through the upper 28 cm of sediments produced pore water concentrations that were less than equilibrium in this zone. Using a one dimensional diffusion-reaction-non-local exchange model, a water exchange rate of 27 L rn 2 d' 1 was required to fit the data. Using this model, the flux of 2 2 3Ra, 2 2 4Ra, and 2 2 8 Ra were computed. At the shoreline, five wells were installed perpendicular to the shoreline to monitor the water table as a function of the tidal height. The volume of the tidal wedge was computed by integrating between the maximum and minimum water table elevation at each well and found to be 3.3 m3 (m shoreline) 1 d"1 . The flux of water through the beach from wave pumping was estimated at 10 m3 (m shoreline) ' 1 d'1 . After measuring the water exchange rate and emanation rate, the flux of 2 2 3 Ra and 2 2 4 Ra from the shoreline was computed. Combining these input fluxes, the total Ra flux to the surface mixed layer could be estimated. For the stretch of shoreline between Sunset Beach and Newport Canyon, the most important source of 2 2 3 Ra and 2 2 4 Ra was the seafloor, which accounted for 90% of the total input flux. Next in importance were the marshes, and the shoreline was the least important source of short-lived Ra. 209 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. To the north and south of Huntington Beach, the distribution of Ra sources to the mixed layer is different. First, large estuaries north and south of Huntington Beach are important sources of Ra in these regions. Second, there is a decrease in the equilibrium pore water Ra concentration of beach sands from north to south, with a significant drop in concentration south of Newport Canyon. Lower concentrations decrease the Ra flux from both the shoreline and the seafloor. In addition, south of Newport Canyon the seafloor is much steeper, limiting the seafloor Ra input to the mixed layer. The average summertime inventory of 2 2 3 Ra and 2 2 4 Ra in the surface mixed layer was computed offshore from Huntington Beach and Sunset Beach. Surface water inventories at Sunset Beach were 3-4 times greater than observed at Huntington Beach. This most likely represents additional sources of Ra to the mixed layer, such as Anaheim Bay and Alamitos Bay. After constraining the Ra inputs and the inventory in the mixed layer, a box model for Ra in the coastal ocean was developed. Because of the relatively short residence time of water in the box, less than 5 days, longshore advection is an important component of the box model. Previous studies in this region have shown that mean summertime surface currents are southward (Noble et al., 2003; Winant and Bratkovich, 1981). If advection is strictly longshore, then the high Ra activities measured at Sunset Beach should be transported to Huntington Beach. This is not observed, with an inventory at Huntington Beach that is at least 50% less than the 210 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sunset Beach inventory. To balance the Ra budget requires a source of low-Ra, which must come from offshore. Then, the longshore advection from Sunset Beach can only account for 10 to 35% of the total flow. In Chapter 3, the offshore distribution of Ra isotopes was examined. The distribution of 2 2 3 Ra and 2 2 4 Ra in the surface mixed layer is a function of the input flux, eddy diffusivity, and radioactive decay. Given the 2 2 3 Ra and 2 2 4 Ra inputs, a two dimensional diffusion-reaction-advection model was developed to compute the eddy diffusivity based on the nearshore distribution of Ra isotopes between Sunset Beach and Newport Beach, CA. The model includes changes in the mixed layer thickness to account for nearshore changes in bathymetry and Ra inputs from the seafloor in contact with the mixed layer. Two independent variables were tuned to simultaneously generate the best fit to the 2 2 3 Ra and 2 2 4 Ra profiles: the eddy diffusivity and the Ra inventory that enters the model at the northern boundary, computed as a fraction of the observed inventory at Sunset Beach. Summertime mean longshore advection rates are from the north at between 3 and 10 cm s' 1 (Hickey, 1993; KOMEX et al., 2003; MCB, 2002; Noble et al., 2003). The mean surface-water offshore profiles could be explained with a constant horizontal eddy diffusivity rate of 1.3 ± 0.2 m2 s'1 , and an inventory at the northern boundary equivalent to just 8 ± 4 % of the observed inventory in surface water at Sunset Beach. The shoreline Ra concentration varied on time scales of hours to days. At the shortest time scales, concentration fluctuations most likely represent fluctuations in 211 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the mixing rate, which may produce 2 0 % changes in the shoreline concentration at periods less than 1 hour. At tidal periods, the shoreline Ra input flux varies. The bulk of the shoreline flux occurs during the ebb tide, when water is draining out of the beach. When this input is concentrated over just a fraction of the tidal cycle, 10- 15% fluctuations in the shoreline concentration are produced. Over time scales of days to weeks, the mean shoreline concentration and the 2 2 3 Ra and 2 2 4 Ra inventory in surface water changed little. However, over longer periods, the shoreline concentration and inventory was observed to change by as much as 50%. The Ra inventory is not only dependent on the shoreline concentration, but is also correlated with the scale distance, which in turn depends on the Ra concentration observed at the first station offshore (~ 1 km). Based on modeling, changes in the mixing rate have only a very minor impact on the concentration at this station. Instead, to change the concentration in this region requires a change in the inventory at the upstream boundary. In order to fit most of the data, 1 <Kh < 2 and the inventory at the northern boundary is between 0 and 25% of the inventory at Sunset Beach. This range includes uncertainties in the input rates, the offshore distribution of the inputs, and advection rates greater than 2 cm s'1 . Several mechanisms to explain changes in the concentration at the upstream boundary include reducing the rate of onshore advection, moving the onshore flow closer to Huntington Beach, or introducing a mean advection towards the north. 212 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In the open ocean, horizontal diffusivity increases with distance from the source (Okubo, 1971). This may explain the difference between these results and a diffusivity of 250 m2 s'1 measured in the Mid-Atlantic Bight based on the distribution of 2 2 3 Ra and 2 2 4 Ra (Moore, 2000). However, when a simple scale dependant mixing is included in the model, the model cannot accurately fit both the shoreline concentration and the offshore stations. To best fit the data and incorporate scale-dependent mixing, the model must be further refined. In the surf zone, mixing is primarily controlled by the height and period of breaking waves and may not be scale-dependant (Inman et al., 1971). Thus, scale dependent mixing must begin at some distance offshore, outside of the surf zone. On length scales greater than the width of the surf zone, the offshore distribution of a solute with a nearshore source is controlled by the minimum diffusivity, which occurs at the start of scale-dependent mixing. Based on the 1-D constant mixing rate model fit and Okubo’s (1976) scale dependant mixing rate, this point occurs at a scale distance of about 270 m. This would represent a minimum distance offshore, since eddies that dominate mixing at this scale may only occur outside of the surf zone. Beyond a 2 km, the precision of the data is insufficient to constrain the mixing rate or to identify an increase in the rate of mixing. The results of this study suggest that changes in the Ra distribution, and presumably other solutes, is not driven by changes in the horizontal diffusivity. The diffusivity varies over a very narrow range. Instead, changes in the distribution are driven by 213 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. changes in input flux. The input flux can change on tidal periods, primarily for shoreline sources. The longshore input flux, controlled by the longshore advection rate and the surface water Ra inventory, may also change. During the summer, there is an onshore flow between Sunset Beach and Huntington Beach. This flow isolates the north part of the bay, with the Los Angeles and Long Beach Harbors, from the south part of the bay. In addition, this circulation suggests that there is a mean flow in the Outer Harbor from east to west in the summer. The last chapter focused on the initial impetus for this project: to constrain the rate of groundwater discharge based on a mass balance for 2 2 6 Ra in the coastal ocean. Huntington Beach was identified as an ideal location to conduct this research. First, there is the well characterized Talbert Aquifer, which is hydraulically connected to the ocean and has injection wells that maintain a constant freshwater head above sea level. Second, the Newport-Inglewood Fault Zone had effectively sealed off any deeper aquifers. Therefore, any groundwater flow must originate from the Talbert Aquifer. Unfortunately, the injection wells have not been operating at full power and the hydraulic barrier has not been maintained. Seawater intrusion through the Talbert Aquifer has been steady since 1997. Any discharge would have to come from perched, unconfined aquifers that are recharged locally with rainwater and irrigation water from lawns and nurseries. 214 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The distribution of 2 2 6 Ra was measured in Santa Monica Bay and San Pedro Bay. Beyond 2 km offshore, the concentration was very stable, with a mean of 98 ± 9 dpm m'3 . Within the first 2 km, concentrations were more variable, with the greatest 2 2 6 Ra concentration observed near the mouths of the Talbert Marsh, Santa Ana River, and Anaheim Bay. Terrestrially-derived groundwater inputs are the most likely source of 2 2 6 Ra to the Talbert Marsh and the Santa Ana River. The maximum rate of groundwater discharge to the marsh is 0.03 cm d"1 . Elevated shoreline 2 2 6 Ra may be explained by direct groundwater inputs. For the Talbert Marsh or the Santa Ana River to be the primary source of shoreline and nearshore 2 2 6 Ra requires specific conditions. These include estuary discharge water with higher 2 2 6 Ra activities than those measured, below average offshore mixing rates, and a reduction in the offshore 2 2 6 Ra gradient. 215 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. References Aller, R.C., 1980. Quantifying solute distributions in the bioturbated zone of marine sediments by defining an average microenvironment. Geochim. Cosmochim. Acta, 44: 1955-1965. Aller, R.C., 1982. The effects of macrobenthos on chemical properties of marine sediments and overlying water. In: PJ. McCall and MJ.S. Teveez (Editors), Animal-Sediment Relationships: Plenum., pp. 3-52. Aller, R.C., 2001. Transport and reactions in the bioirrigated zone. In: B.P. Boudreau and B.B. Jorgensen (Editors), The Benthic Boundary Layer: Transport processes and biogeochemistry. Oxford University Press, New York, pp. 269- 295. Bascom, W.N., 1951. The relationship between sand size and beach-face slope. Trans. Amer. Geophys. Union, 32: 866-874. Bateman, H., 1910. Solution of a system of differential equations occurring in the theory of radio-active transformations. Proc. Cambridge Phil. Soc., 15: 423. Berelson, W.M., Hammond, D.E. and Eaton, A., 1987. A technique for the rapid extraction of Rn-222 from water samples and a case study. In: B. Graves (Editor), Radon in groundwater. National Water Well Association, pp. 271 - 281. Biscaye, P.E., Flagg, C.N. and Falkowski, P.G., 1994. The shelf edge exchange processes experiment, SEEP-II: An introduction to hypotheses, results and conclusions. Deep-Sea Res. II, 41( 2-3): 231-252. Blake, G.H., 1991. Review of neogene biostratigraphy and stratigraphy of the Los Angeles basin and implications for basin evolution. In: K.T. Biddle (Editor), Active Margin Basins. Memoir. American Association of Petrolium Geologists, pp. 135-184. Boehm, A.B. et al., 2002. Decadal and shorter period variability of surf zone water quality at Huntington Beach, California. Environmental Science & Technology, 36(18): 3885-3892. Bollinger, M.S. and Moore, W.S., 1993. Evaluation of salt marsh hydrology using radium as a tracer. Geochim. Cosmochim. Acta., 57: 2203-2212. 216 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Broecker, W.S., 1965. The application of natural radon to problems in ocean circulation. In: T. Ichiye (Editor), Symposium on Diffusion in Oceans and Fresh Waters. Lamont-Doherty Geological Observatory, Palisades, NY, pp. 116-145. Brownlie, W.R. and Taylor, B.D., 1981. Sediment management for southern California mountains, coastal plains, and shoreline: Part C- Coastal sediment delivery by major rivers in southern California, EQL Rep. Cal. Inst. Tech., Pasadena, CA, pp. 314. Burnett, W.C., Bokuniewicz, H., Huettel, M., Moore, W.S. and Taniguchi, M., 2003. Groundwater and pore water inputs to the coastal zone. Biogeochemistry, 66(1-2): 3-33. Burnett, W.C. et al., 2002. Assessing methodologies for measuring groundwater discharge to the ocean. Eos, Transactions, American Geophysical Union, 83(11): 117-123. Burnett, W.C., Cowart, J.B. and Deetae, S., 1990. Radium in the Suwannee River and Estuary: Spring and river input to the Gulf of Mexico. Biogeochem., 10: 237-255. CalDeptF&G, 1976. The natural resources of Anaheim Bay. California Department of Fish and Game and U.S. Fish and Wildlife Service, Sacramento, CA, 103 pp. CalDWR, 1967. Progress report on the ground water geology of the coastal plain of Orange County. California Department of Water Resources, Sacramento, CA. Callison, J., 1992. Alamitos Seawater Barrier. In: E.G. Heath and W.L. Lewis (Editors), The Regressive Pleistocene Shoreline: Coastal Southern California. Field Trip Guide Book. South Coast Geological Society, Los Angeles, pp. 263-271. Casey, K.S. and Adamec, D., 2002. Sea surface temperature and sea surface height variability in the North Pacific Ocean from 1993 to 1999. J. Geophys. Res. (C Oceans), 107(C8): 3099. Chelton, D.B. and Davis, R.E., 1982. Monthly mean sea-level variability along the west coast of North America. J. Phys. Oceanogr., 12: 757-784. Cochran, J.K., 1979. The geochemistry of 2 2 6 Ra and 2 2 8 Ra in marine deposits. Ph.D. dissertation. Yale University. 217 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cochran, J.K., 1992. The oceanic chemistry of the U and Th-series nuclides. In: M. Ivanovich and R.S. Harmon (Editors), Uranium-series Disequilibrium: Applications to Earth, Marine, and Environmental Sciences. Clarendon Press, Oxford, pp. 384-430. Colbert, S.L., 2001. Numerical simulation of whole core squeezer radon pore water profiles: Methodological considerations and evaluation of benthic fluxes and bio-irrigation and advection rates. Master's Thesis. University of Southern California, Los Angeles, CA, 151 pp. Cooper, H.H., Jr., Kohout, F.A., Henry, H.R. and Glover, R.E., 1964. Sea Water in Coastal Aquifers: Relation of salt water to fresh ground water, Geological Survey Water-Supply Paper, Washington, pp. 84. Corps and LAHD, 1980. Final EIS/EIR, Los Angeles Harbor Deepening Project. U.S. Army Corps of Engineers and Los Angeles Harbor Department, Los Angeles. Crowe, F.J. and Schwartzlose, R.A., 1972. Release and recovery records of drift bottles in the California region, 1955 through 1971, California Cooperative Oceanic Fisheries Atlas. Marine Research Committee, State of California, Sacramento, CA. Csanady, G.T., 1977. The coastal jet conceptual model in the dynamics of shallow seas. In: E.D. Goldberg, I.N. McCave, J.J. O'Brien and J.H. Steele (Editors), The Sea. John Wiley and Sons, New York, pp. 117-144. Dean, R.F., 1973. Heuristic models of sand transport in the surf zone, Proceedings of the First Australian Conference on Coastal Engineering, Sydney, pp. 208- 214. Denny, M.W., 1988. Biology and the Mechanics of the Wave-Swept Environment. Princeton University Press, Princeton, NJ, 344 pp. DiGiacomo, P.M. and Holt, B., 2001. Satellite observations of small coastal ocean eddies in the Southern California Bight. Journal of Geophysical Research, 106(10): 22521-22544. Elsinger, R.J., King, P.T. and Moore, W.S., 1982. Radium-224 in natural waters measured by gamma-ray spectrometry. Analytica Chimica Acta, 144: 277- 281. Elsinger, R.J. and Moore, W.S., 1980. Ra-226 behavior in the Pee Dee River- Winyah Bay Estuary. Earth Planet. Res. Lett., 48: 239- 249. 218 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Emerson, S., Jahnke, R. and Heggie, D., 1984. Sediment-water exchange in shallow- water estuarine sediments. J. Mar. Res., 42: 709-730. Emery, K.O., 1959. Wave patterns off Southern California. J. Mar. Res.: 133-140. Emery, K.O., 1960. The sea off Southern California. John Wiley and Sons, New York, 366 pp. Emery, K.O. and Foster, J.F., 1948. Water tables in marine beaches. J. Mar. Res., 7: 644-654. Everts, C.E., 1993. Wave and current data summary: West Newport Beach, 6 Jan 1992- 5 Mar 1993. Public Works Dept., Newport Beach, pp. 15-30. Faure, G., 1986. Principles of isotope geology. John Wiley & Sons, New York, 589 pp. Felix, D.W. and Gorsline, D.S., 1971. Newport Canyon, California: An example of the effects of shifting loci of sand supply on canyon position. Mar. Geol., 10: 177-198. Fick, A., 1855. Uber Diffusion. Ann. Physik Chemie, 94: 59-86. Gargett, A.E., 1984. Vertical eddy diffusivity in the ocean interior. J. Mar. Res., 42: 359-393. Gawarkiewicz, G.G., Linder, C.A., Lynch, J.F., Newhall, A.E. and Bisagni, J.J., 1996. A surface-trapped intrusion of slope water onto the continental shelf in the Mid-Atlantic Bight. Geophys. Res. Letters, 23(25): 3763-3766. Geiger, H. and Nuttall, J.M., 1911. The ranges of the alpha-particles from various radioactive substances and a relation between range and period of transformation. Phil. Mag., 22: 613. Grant, D.J., 1973. Sediments of the San Pedro Shelf. Masters Thesis. University of Southern California, Los Angeles, CA, 93 pp. Grant, S.B. et al., 2003. Coastal runoff impact study phase II: Sources and dynamics of fecal indicators in the lower Santa Ana River watershed, Final report for the National Water Research Institute, County of Orange, and the Santa Ana Regional Water Quality Control Board, pp. 150. 219 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Grant, S.B. et al., 2001. Generation of enterococci bacteria in a coastal saltwater marsh and its impact on surf zone water quality. Env. Sci. Tech., 35(12): 2407-2416. Halliwell, G.R. and Allen, S.S., 1987. Large-scale sea level response to atmospheric forcing along the west coast of North America, Summer 1973. J. Phys. Oceanogr., 14: 864-886. Hammond, D.E. and Fuller, C., 1979. The use of radon-222 to estimate benthic exchange and atmospheric exchange rates in San Francisco Bay. In: T.J. Conomos (Editor), San Francisco Bay: The Urbanized Estuary. Pacific Div. Amer. Assoc. Adv. Sci., pp. 213-230. Hammond, D.E., Marton, R.A., Berelson, W.M. and Ku, T.-L., 1990. Radium 228 distribution and mixing in San Nicolas and San Pedro Basins, Southern California Borderland. J. Geophys. Res., C95(3): 3321-3335. Hammond, D.E., Simpson, HJ. and Mathieu, G., 1977.2 2 2 Rn distribution and transport across the sediment-water interface in the Hudson River Estuary. J. Geophys. Res., 82: 3913-3920. Hammond, D.E., Zukin, J.G. and Ku, T.L., 1988. The kinetics of radioisotope exchange between brine and rock in a geothermal system. J. Geophys. Res., 93(B11): 13,175-13,186. Hancock, G.J. and Murray, A.S., 1996. Source and distribution of dissolved radium in the Bega River estuary, Southeastern Australia. Earth and Planetary Science Letters, 138: 145-155. Hancock, G.J., Webster, I.T., Ford, P.W. and Moore, W.S., 2000. Using Ra isotopes to examine transport processes controlling benthic fluxes into a shallow estuarine lagoon. Geochim. Cosmochim. Acta., 64(21): 3685-3699. Handin, J.W., 1951. The source, transportation and deposition of beach sediment in Southern California. Tech. Memo U.S. Army Corps Engrs Beach Erosion Board 22: 123. Hendricks, T.J., 1980. Currents in the Los Angeles Area, The Southern California Coastal Water Research Project Annual Report 1979-1980. Southern California Coastal Water Research Project, Long Beach, pp. 243-256. Hendricks, T.J., 1994. Near-Bottom Currents off Southern California, Southern California Coastal Water Research Project 1992-1993 Annual Report. Southern California Coastal Water Research Project, Long Beach, pp. 65-80. 220 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hendricks, T.J. and Stubbs, H.H., 1984. Currents in San Gabriel Canyon, Southern California Coastal Water Research Project 1983-1984 Annual Report. Southern California Coastal Water Research Project, Long Beach, CA, pp. 65-76. Herndon, R.L., 1992. Hydrogeology of the Orange County groundwater basin- An overview. In: E.G. Heath and W.L. Lewis (Editors), The Regressive Pleistocene Shoreline: Coastal Southern California. Field Trip Guide Book. South Coast Geological Society, Los Angeles, pp. 237-259. Hickey, B.M., 1979. The California Current system- hypotheses and facts. Prog. Oceanogr., 8(4): 191-279. Hickey, B.M., 1992. Circulation over the Santa Monica-San Pedro basin and shelf. Prog. Oceanogr., 30: 37-115. Hickey, B.M., 1993. Physical Oceanography. In: M.D. Daily, D.J. Reish and J.W. Anderson (Editors), Ecology of the Southern California Bight. University of California, Los Angeles, pp. 19-70. Huettel, M. and Gust, G., 1992. Impact of bioroughness on interfacial solute exchange in permeable sediments. Mar. Ecol. Prog. Ser., 89: 253-267. Huettel, M. and Webster, I.T., 2001. Porewater flow in permeable sediments. In: B.P. Boudreau and B.B. Jorgensen (Editors), The Benthic Boundary Layer: Transport processes and biogeochemistry. Oxford University Press, New York, pp. 144-179. Huettel, M., Ziebis, W. and Forster, S., 1996. Flow-induced uptake of particulate matter in permeable sediments. Limnol. Oceanogr., 41: 309-322. Huh, C.-A. and Ku, T.L., 1998. A 2-D section of 2 2 8 Ra and 2 2 6 Ra in the Northeast Pacific. Oceanol. Acta., 21(4): 533-542. Huh, C.H. and Beasley, T.M., 1987. Profiles of dissolved and particulate thorium isotopes in the water column of coastal Southern California. Earth Planet. Sci. Lett., 85: 1-10. Imboden, D.M., 1981. Tracers and mixing in the aquatic environment: A critical discussion of diffusion models and an introduction into concepts of non- Flckian transport. Thesis. Swiss Federal Institute of Technology, 137 pp. Inman, D.L. and Bush, B.M., 1975. The coastal challenge. Science, 181: 20-29. 221 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Inman, D.L. and Frautschy, 1966. Coastal engineering. Santa Barbara Specialty Conference. Proc. Am. Soc. Civil Engrs: 511-536. Inman, D.L., Tait, R.J., Komar, P.D. and Nordstrom, C.E., 1968. Dispersion of water and sediment in the surf zone. Scripps Institution of Oceanography, La Jolla, Calif., 60 pp. Inman, D.L., Tait, R.J. and Nordstrom, C.E., 1971. Mixing in the Surf Zone. J. Geophys. Res., 76(15): 3493-3514. Johnson, D. and Pattiaratchi, C., 2004. Transient rip currents and nearshore circulation on a swell-dominated beach. J. Geophys. Res. C, 109: C02026. Kadko, D., 1980.2 3 0 Th, 2 2 6Ra, and 2 2 2 Rn in abyssal sediments. Earth. Planet. Sci. Letters, 49: 360-380. Karl, H.A., Cacchione, D.A. and Drake, D.E., 1980. Erosion and Transport of Sediments and Pollutants in the Benthic Boundary Layer on the San Pedro Shelf, Southern California. U.S. Geological Survey Open File Report 80-386, 54 pp. Kaufman, A.T., R. M. ; Broecker, W. S .; Feely, H. W., 1973. Distribution of 2 2 8 Ra in the World Ocean. Journal of Geophysical Research, 78(36): 8827-8848. Kelly, R.P. and Moran, S.B., 2002. Seasonal changes in groundwater input to a well- mixed estuary estimated by using radium isotopes and implications for coastal nutrient budgets. Limnol. Oceanogr., 47: 1796-1807. Key, R.M., Brewer, R.L., Stockwell, J.H., Guinasso, J., N.L. and Schink, D.R., 1979. Some improved techniques for measuring radon and radium in marine sediments and seawater. Mar. Chem, 7: 251-264. Key, R.M., Stallard, R.F., Moore, W.S. and Sarmiento, J.L., 1985. Distribution and flux of Ra-226 and Ra-228 in the Amazon River Estuary. J. Geophys. Res., 90: 6995-7004. Kim, G., 1993. Actinium-227, Radium-228, and Radium-226 in surface water off California: Development of analutical techniques for Actinium-227. Masters Thesis. University of Southern California, Los Angeles, CA, 91 pp. Kim, J.H., Grant, S.B., McGee, C., Sanders, B.F. and Largier, J., 2004. Locating sources of surf zone pollution: A mass budget analysis for fecal indicator bacteria at Huntington Beach, CA. Environ. Sci. Technol., 38: 2626-2636. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Knauss, K.G., Ku, T.L. and Moore, W.S., 1978. Radium and thorium isotopes in the surface waters of the east Pacific and coastal Southern California. Earth Planet. Sci. Lett., 39: 235-249. Kohout, F.A., 1960. Cyclic flow of salt water in the Biscayne Aquifer of southeastern Florida. J. Geophys. Res., 65(7): 2133-2141. KOMEX, Jones, B. and Fuhrman, J., 2003. AES Huntington Beach Generating Station Surf Zone Water Quality Study, Westminster, CA. Krishnaswami, S., Graustein, W.C., Turekian, K.K. and Dowd, J.F., 1982. Radium, thorium and radioactive lead isotopes in groundwaters: Application to the in situ determination of adsorption-desorption rate constants and retardation factors. Water Res. Res., 18(6): 1633-1675. Langmuir, D., 1997. Aqueous environmental geochemistry. Prentice Hall, Upper Saddle River, NJ, 600 pp. Langmuir, D. and Riese, A.C., 1985. The thermodynamic properties of radium. Geochim. Cosmochim. Acta, 49: 1593-1601. Lentz, S., Guza, R.T., Elgar, S., Feddersen, F. and Herbers, T.H.C., 1999. Momentum balances on the North Carolina inner shelf,. J. Geophys. Res. (C Oceans), 104(C8): 18205-18226. Lentz, S.J., Elgar, S. and Guza, R.T., 2003. Observations of the flow field near the nose of a buoyant coastal current. J. Phys. Oceanogr., 33: 933-943. Lewis, R., 1997. Dispersion in estuaries and coastal waters. John Wiley and Sons, New York, 311 pp. Li and Chan, 1979. Desorption of Ba and Ra-226 from river-borne sediments in the Hudson Estuary. Earth Planet. Sci. Lett., 43: 343-350. Li, Y.-H. and Gregory, S., 1974. Diffusion of ions in seawater and in deep-sea sediments. Geochim. Cosmochim. Acta, 38: 703-714. Luedtke, N.A. and Bender, M.L., 1979. Tracer study of sediment-water interactions in estuaries. Estuar. Coast. Mar. Sci., 9: 643-651. Lynn, R.S. and Simpson, J.J., 1987. California Current system- The seasonal variability of its physical characteristics. J. Geophys. Res., 92(C12): 12947- 12966. 223 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MacMahan, J.H., Reniers, Thornton, E.B. and Stanton, T.P., 2004. Infragravity rip current pulsations. J. Geophys. Res. C, 109: C01033. Magnusen, C.E., 1995. The characterization of Huntington Beach and Newport Beach through fourier grain-shape , grain-size, and longshore current analyses. Masters Thesis. University of Southern California, Los Angeles, CA, 178 pp. Maloney, N. and Chan, K., 1974. Hydrography of harbors, lagoons, and sloughs. In: M.D. Dailey, B. Hill and N. Lansing (Editors), A Summary of Knowledge of Southern California Coastal Zone and Offshore Areas. Southern California Ocean Studies Consortium. Mathieu, G.G., Biscaye, P.E., Lupton, R.A. and Hammond, D.E., 1988. System for measurement of radon-222 at low levels in natural waters. Health Physics, 55: 989-992. MCB, 2002. AES Huntington Beach L.L.C. Generating Station In-Plant Bacteria Monitoring Summary Report, May 2002., Costa Mesa, CA. McCaffrey, R.J. et al., 1980. The relation between pore water chemistry and benthic fluxes of nutrients and manganese in Narragansett Bay, Rhode Island. Limnol. Oceanogr., 25(1): 31-44. McCurdy, R., 1964. Water motion and sediments of northeast San Pedro Bay, CA. Master's Thesis. University of Southern California, Los Angeles, 79 pp. McLachlan, A., 1979. Volumes of sea water filtered by East Cape sandy beaches. S. Afr. J. Sci., 75: 75-79. McLachlan, A., 1982. A model for the estimation of water filtration and nutrient regeneration by exposed sandy beaches. Mar. Envir. Res., 6 : 37-47. McLachlan, A., 1989. Water filtration by dissipative beaches. Limnol. Oceanogr., 34: 774-780. Michel, J., Moore, W.S. and King, P.T., 1981. Gamma-ray spectrometry for determination of Ra-228 and Ra-226 in natural waters. Anal. Chem., 53: 1885-1889. Miller, A J. et al., 1999. Observing and modeling the California current system. EOS Transactions, 80(45): 533-539. 224 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Moore, D.G., 1951. The marine geology of San Pedro Shelf. Master's Thesis. University of Southern California, Los Angeles, CA, 87 pp. Moore, W.S., 1976. Sampling radium-228 in the deep ocean. Deep-Sea Res., 23: 647-651. Moore, W.S., 1987. Radium 228 in the South Atlantic Bight. Journal of Geophysical Research, C, Oceans, 92(5): 5177-5190. Moore, W.S., 1996. Large ground water inputs to coastal waters revealed by 2 2 6 Ra enrichments. Nature, 380: 612-614. Moore, W.S., 1999. The subterranean estuary; a reaction zone of ground water and seawater. Mar. Chem., 65(1-2): 111-125. Moore, W.S., 2000. Determining coastal mixing rates using radium isotopes. Cont. Shelf Res., 20: 1993-2007. Moore, W.S. and Arnold, R., 1996. Measurement of 2 2 3 Ra and 2 2 4 Ra in coastal waters using a delayed coincidence counter. J. Geophys. Res., 101: 1321-1329. Moore, W.S. and Scott, M.R., 1986. Behavior of Ra-226 in the Mississippi River mixing zone. J. Geophys. Res., 91(C12): 14317-14329. Moran, D.E. and Wiebe, K.H., 1992. Holocene deposition and organic soils near Huntington Beach, Orange County, California. In: E.G. Heath and W.L. Lewis (Editors), The Regressive Pleistocene Shoreline: Coastal Southern California. Field Trip Guide Book. South Coast Geological Society, Los Angeles, pp. 137-156. Mu, Y.K., Cheng, A.H.D., Badiey, M. and Bennett, R., 1999. Water wave driven seepage in sediment and parameter inversion based on pore pressure data. Int. J. Num. Anal. Methods Geomech., 23: 1655-1674. Nardin, T.R. and Henyey, T.L., 1978. Pliocene-Pleistocene diastrophism of Santa Monica and San Pedro shelves, California continental boarderland. Amer. Assn. Petrol. Geol. Bull., 62(2): 247-272. Noble, M. et al., 2003. Huntington Beach Shoreline Contamination Investigation, Phase III. U.S. Geological Survey Open File Report 03-62. Okubo, A., 1971. Oceanic diffusion diagrams. Deep Sea Res., 18: 789-802. 225 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Okubo, A., 1976. Remarks on the use of 'diffusion diagrams' in modeling scale dependent diffusion. Deep Sea Res., 23: 1213-1214. Orys-Brink, M., 1983. Sediment characteristics across the Santa Monica and San Pedro Shelves : an environmental comparison. Masters Thesis. University of Southern California, Los Angeles, CA, 233 pp. Osmond, J.K. and Ivanovich, M., 1992. Uranium-series mobilization and surface hydrology. In: M. Ivanovich and R.S. Harmon (Editors), Uranium-series Disequilibrium: Applications to Earth, Marine, and Environmental Sciences. Clarendon Press, Oxford, pp. 259-289. Pettigrew, N.R. and Murray, S.P., 1986. The coastal boundary layer and inner shelf. In: C.H.K. Mooers (Editor), Baroclinic Processes on Continental Shelves, pp. 95-108. Pipkin, B.W., 1985. Santa Monica to Dana Point. In: G. Griggs and L. Savoy (Editors), Living with the California Coast. Duke University Press, Durham, North Carolina, pp. 338-340. Poland, J.F., 1959. Hydrology of the Long Beach-Santa Ana area, California, with special reference to the watertightness of the Newport-Inglewood structural zone. USGS Water Supply Paper, W-1471, 257 pp. Precht, E. and Huettel, M., 2003. Advective pore-water exchange driven by surface gravity waves and its ecological implications. Limnol. Oceanogr., 48: 1674- 1684. Rama and Moore, W.S., 1996. Using the radium quartet for evaluating groundwater input and water exchange in salt marshes. Geochim. Cosmochim. Acta., 60(23): 4645-4652. Rice, R.M., Gorsline, D.S. and Osborne, R.H., 1976. Relationship between sand input from rivers and the composition of sands from the beaches of Southern California. Sedimentology, 23: 689-703. Riedl, R.J., 1971. How much seawater passes through sandy beaches? Int. Revue fes. Hydrobiol., 56: 923-946. Riedl, R.J., Huang, N. and Machan, R., 1972. The subtidal pump: a mechanism of interstitial water exchange by wave action. Mar. Biol., 13: 210-221. Rutherford, E., 1906. The mass and velocity of the a particles expelled from radium and actinium. Phil. Mag., 6(12): 348-371. 226 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sarmiento, J.L., Feely, H.W., Moore, W.S., Bainbridge, A.E. and Broecker, W.S., 1976. The relationship between vertical eddy diffusion and buoyancy gradient in the deep sea. Earth Planet. Sci. Lett., 32: 357-370. Sarmiento, J.L., Rooth, C.G.H. and Broecker, W.S., 1982. Radium 228 as a tracer of basin wide processes in the abyssal ocean. Journal of Geophysical Research. C. Oceans and Atmospheres, 87(12): 9694-9698. Schwalbach, F.P. and Gorsline, D.S., 1985. Holoscene sediment budget for the basins of the California continental boarderland. J. Sediment. Petrol., 55: 829-842. Sherman, L., 1931. A history of Newport Beach, 58-61. Times Mirror Press, Los Angeles. Soule, D.F. and Oguri, M., 1980. The marine environment in Los Angeles and Long Beach Harbor during 1978. Part 17. University of Southern California, Los Angeles. Stevenson, R.E., 1954. The marshlands at Newport Bay, CA. Ph.D. Dissertation. University of Southern California, Los Angeles, 199 pp. Stevenson, R.E., Tibby, R.B. and Gorsline, D.S., 1956. The oceanography of Santa Monica Bay. Hancock Foundation, University of Southern California: 267. Stommel, H., 1949. Horizontal diffusion due to oceanic turbulence. J. Mar. Res., 8 : 199-225. Sun, Y. and Torgersen, T., 1998a. The effects of water content and Mn-fiber surface conditions on measurement by emanation. Mar. Chem., 62: 299. Sun, Y. and Torgersen, T., 1998b. Rapid and precise measurement method for adsorbed 2 2 4 Ra on sediments. Mar. Chem., 61(3-4): 163-171. Sun, Y.I. and Torgersen, T., 2001. Adsorption-desorption reactions and bioturbation transport of 2 2 4 Ra in marine sediments: a one-dimensional model with applications. Mar. Chem., 74: 227-243. Sverdrup, H.U. and Fleming, R.H., 1941. The water off the coast of Southern California, March to July 1937. Scripps Institute Oceanography Bulletin, 4(10): 261-387. 227 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sverdrup, H.U., Johnson, M.W. and Fleming, R.H., 1942. The Oceans: Their physics, chemistry, and general biology. Prentice Hall, Englewood Cliffs, NJ, 1060 pp. Swift, D.J.P. and Niedoroda, A.W., 1985. Fluid and sediment dynamics on continental shelves. In: R.W. Tillman, D.J.P. Swift and R.G. Walker (Editors), Shelf sands and sandstone reservoirs. SEPM Short Course 13, pp. 47-133. Tanner, A.B., 1964. Radon migration in the ground: A review. In: J.A.S. Adams and W.M. Lowder (Editors), The natural radiation environment symposium. U. Chicago Press, Chicago, pp. 161-190. Torgersen, T. et al., 1996.2 2 4 Ra distribution in surface and deep water of Long Island Sound: Sources and horizontal transport rates. Cont. Shelf Res., 16(12): 1545-1559. Ullman, W.J. and Aller, R.C., 1982. Diffusion coefficients in nearshore marine sediments. Limnol. Oceanogr., 27(3): 552-556. Vemulakonda, S.R., Chou, L.W. and Hall, R.W., 1991. Los Angeles and Long Beach Harbors additional plan testing, numerical modeling of tidal circulation and water quality. Technical Report. U.S. Army Engineers Waterways Experimental Station, Vicksburg, MS. Winant, C.D. and Bratkovich, A.W., 1981. Temperature and currents on the southern California shelf: A description of the variability. J. Phys. Oceanogr., 11(1): 71-86. Wright, L.D., 1995. Morphodynamics of inner continental shelves. CRC Press, Boca Raton, FL, 241 pp. Wright, L.D. and Short, A.D., 1984. Morphodynamic variability of surf zones and beaches: A synthesis. Mar. Geol., 56: 93-118. Zekster, I.S. and Loaiciga, H.A., 1993. Groundwater fluxes in the global hydrologic cycle: Past, present, and future. J. Hydro., 144: 405-427. Zeng, E.Y., Bay, S.M., Tran, K. and Alexander, C., 2000. Temporal and spatial distributions of contaminants in sediments of Santa Monica Bay, California, Southern California Coastal Water Research Project 1999-2000 Annual Report, pp. 96-115. 228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ziebis, W., Huettel, M. and Forster, S., 1996. Impact of biogenic sediment topography on oxygen fluxes in permeable seabeds. Mar. Ecol. Prog. Ser., 140: 227-237. Zukin, J.G., Hammond, D.E., Ku, T.-L. and Elders, W.A., 1987. Uranium-thorium series radionuclides in brines and reservoir rocks from two deep geothermal boreholes in the Salton Sea Geothermal Field, southeastern California. Geochim. Cosmochim. Acta, 51: 2719-2731. 229 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix A: Methods I. Mn Fiber Preparation Mn- fibers are prepared based on the technique developed by Moore (1976), using Acrilan B-16, 3 denier per filament, 2 inch cut length, made by Solutia, Inc, formerly Monsanto (Moore, 1976). Into a 0.5 M potassium permanganate solution heated to 80°C, 45 g of fluffed acrylic fiber is added. The fiber used is Acrilan B-16, 3 denier per filament, 2 inch cut length, made by Solutia, Inc, formerly Monsanto. The temperature of the solution is carefully brought back to 75-80°C. Under these conditions, the permanganate is reduced in an exothermic reaction and deposited as a manganese oxide on the fibers. After between 5 and 10 minutes, the reaction is quenched by placing the fibers in a deionized water bath. If the reaction becomes self-sustaining, as evident when the solution boils uncontrollably, the fibers should be quenched immediately. The solution from two batches can then be combined to processes a third batch of fiber before the permanganate solution is ineffective. In most cases, about 30 g of prepared Mn-fiber were used for each sample. II. The RaDeCC System I n t r o d u c t i o n . The Radium Delay Coincidence Counter (RaDeCC) was developed by Moore and Arnold (1996). The RaDeCC system is primarily used to measure the activity of short-lived Ra isotopes adsorbed to manganese dioxide impregnated 230 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. acrylic fibers. It may also be used to measure the Ra emanation rate from sediments (Sun and Torgersen, 1998b). The RaDeCC system uses a delayed coincidence circuitry to distinguish between 2 1 9 Rn (t1 / 2 = 3.96 s) and 2 2 0 Rn (t1 / 2 = 55.6 s) by the decay of their daughter products. From this, the activity of the parent isotopes, 2 2 3 Ra and 2 2 4Ra, can be calculated. The activity of the radioactive parents of 2 2 3 Ra or 2 2 4 Ra can also be measured if equilibrium can be assumed. These include 2 2 8Th, 2 2 8Ra, and 2 2 7 Ac (Moore and Arnold, 1996). O p e r a tio n . The RaDeCC system was purchased from Ralph Arnold, through Scientific Computer Instruments, Columbia, SC. The system circulates a carrier gas, helium, through a closed system. The helium is pumped through a flow meter, a sample chamber containing manganese dioxide impregnated acrylic fibers, a counting cell, and then returns to the pump. The flow meter is graduated between 0- 15, and was calibrated for the flow of helium (Figure A-l). The flow rate is controlled by a valve that bypasses the sample chamber and counting cell. When the valve is closed, all of the flow is through the RaDeCC system. As the valve is opened, a greater fraction of the helium stream is circulated through the pump instead of through the RaDeCC system. The counting cell is a 1.1 L Plexiglas tube with flat Plexiglas end caps and is coated internally with a silver activated ZnS. The cell is then wrapped in aluminum foil and black tape to block out any light. The counting cell is attached to a photomultiplier tube (PMT), which is interfaced to a computer via the RaDeCC circuitry. 231 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. H e F lo w ( L m in1) Figure A-l. Flowmeter reading as a function of the helium flow rate. 10 = -0.379;+ 0.5612x R<= 0.99439 8 6 4 2 0 14 12 8 10 4 6 0 2 Flow Meter Value Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The RaDeCC circuitry records data in three bins: Total counts, 2 1 9 Rn counts, and 2 2 0 Rn counts. The Total bin records every count measured. After an initial count is observed, any decay that occurs between 0.01 and 5.6 msec, is counted in the 2 1 9 Rn bin. Then, from 5.61 to 600 msec, any decays that occur are counted in the 2 2 0 Rn bin. These windows are set based on the half-life of each daughter, and should capture 8 8 % of the 2 1 5 Po (t1 / 2 = 1.78 ms) from 2 1 9 Rn and 91% of the 2 1 6 Po (ty 2 = 150 ms) from 2 2 0Rn. Ideally, for every count in the 2 1 9 Rn or 2 2 0 Rn bin, there should be 2 total counts. However, any decay that occurs while the 2 1 9 Rn or 2 2 0 Rn window is open will be counted. As a result, multiple decays may be counted while the window is open, especially for samples with high count rates (i.e. >30 cpm). D a ta P r o c e s s in g : RaDeCC data is processed using the following method. The number of counts in each bin is recorded every 1 0 minutes, which allows the user to go back through the data and check for irregularities. The most common source of irregularities are power surges during the sample run. Power surges may add a large number, or even a negative number, of counts to one or all of the bins during a count interval. If the counts in any bin during an interval are negative or greater than three times the average, then that time interval and the counts for all three bins are subtracted from the total. Random counts, such as background counts or by the decay of 2 2 2Rn, will produce “chance” counts in the 2 l9 Rn and 2 2 0 Rn bins. The chance cpm is different for each isotope. For 2 1 9Rn, it is calculated using the following formula: 233 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (A-l) Where At is the time the count window is open, C is the count rate (cpm) for each bin recorded by the RaDeCC, T is the total counts, and 219 and 220 for each Rn isotope (Moore and Arnold, 1996). For 2 2 0Rn, there is an additional correction for the possibility of two decays occurring from the decay of 2 1 9Rn: Where C' is the counts of 2 1 9 Rn corrected for background and chance counts. Next, the total corrected counts (2N1 ) can be computed: where subscript i is for isotope 2 2 3 Ra or 2 2 4Ra, Cc is the chance counts per minute (cpm), Cb is the background cpm, t^, is the count time in minutes, and Eto t is the total efficiency. The background activity was measured to be 0.0005 cpm for 2 1 9 Rn and .0020 for 2 2 0Rn. The total efficiency is thoroughly explored below. The 2N' is equivalent to the total number of Rn atoms produced during the counting time. Since Rn is in equilibrium with Ra, this is equivalent to the number of Ra atoms that decayed during this time interval. c (CT - C 2ig- C 2 2 0 ) 2&t | 1. 6 * ( c ; i9)2At l-(C r - C 219- C 220)Ar l - 1 . 6 * ( c ; i9)Ar (A-2) (A-3) 234 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C a lc u la tin g th e I n i t i a l C o n c e n t r a t i o n : From the corrected total counts, the initial concentration of each isotope is calculated. For 2 2 4Ra, this requires accounting for the decay of 2 2 4 Ra since the sample was collected, decay while counting, and any support from 2 2 8Th. Assuming that the 2 2 0 Rn is in equilibrium with 2 2 4Ra, then the rate of decay is: where N, is the total number of atoms, \ is the decay constant, and subscript i represents the isotope 2 2 8 Th for 228 and 2 2 4 Ra for 224. Technically, the ingrowth of 2 2 8 Th from 2 2 8 Ra should also be included when calculating the initial 2 2 4 Ra activity. However, the ingrowth of 2 2 8 Th during a 12-hour count is negligible. The general solution to this equation, first solved by Bateman (Bateman, 1910) is: Where XN° is the initial activity of each isotope. With this equation, the 2 2 4 Ra activity after some initial time (t0 ), defined as the time the sample was collected, can be calculated. When the sample is counted during time interval t2 -t,, the number of decays expected, 2N', can be calculated by integrating Equation A-5. d N d t (A-4) (A-5) (A-6 ) 235 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The solution to Equation A- 6 can then be rearranged to calculate the activity of 2 2 4 Ra on the fibers when the sample was first collected, N2 2 4 ° : k 2UN°2U = (A-7) _ g ~ ^ 2 z { h ~ h ) j ^ , " ' ‘-224 l[ ^ ^228 ^2 2 4 “ ^224 { h ~ h ) j The initial 2 2 8 Th concentration is computed by re-counting the fibers on the RaDeCC system at least two weeks after the sample is collected, after the 2 2 4 Ra has grown into equilibrium with the 2 2 8Th. Again, Equation A-7 is used, replacing the 2 2 4 Ra values with those for 2 2 8Th, and the 2 2 8 Th values with those for 2 2 8Ra. The initial activity of 2 2 8 Ra is later measured using gamma spectroscopy (Elsinger et al., 1982). To further assess the importance of ingrowth of 2 2 8 Th and 2 2 4Ra, the theoretical activity as a function of time for each isotope was computed with and without ingrowth. The difference was computed by taking the difference between these two and normalizing to the activity without ingrowth and is presented in Figure A-2 for three 2 2 4Ra:2 2 8 Th ratios: 1, 10, and 100. These isotope ratios correspond to offshore, midshore, and nearshore samples, respectively. At the nearshore stations, the initial 2 2 4 Ra activity is so high that ingrowth from 2 2 8 Th has little influence on the 2 2 4 Ra activity. However, at the offshore stations, where 2 2 4 Ra and 2 2 8 Th are close to 236 ^^224 + ^22&N2: 228 ^ - 2: ^224 - ^ 2: ~ ^2 2 i :28 i :24 “^ ■ 2 2 4 * 1 f1- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure A-2. Daughter ingrowth vs. time for various 2 2 4 Ra:228Th ratios. In general, the ratio decreases with distance offshore. The 224Ra activity is normalized to the 224Ra activity computed assuming no ingrowth. 100 224Ra:228Th = 1 224Ra:228Th = 10 224Ra:228Th = 100 o 2 4 6 8 10 Time (Days) 237 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure A-3. Daughter ingrowth vs. time for various 2 2 8 Ra:228Th ratios. In general, the ratio decreases with distance offshore. The 228Th activity is normalized to the 228Th activity computed assuming no ingrowth. 228Ra:2 2 8 Th = 250 2 2 8 Ra:2 2 8 Th = 100 00 n 10 - !28Ra:2 2 8 Th = 2 0 2 4 6 8 10 Time (Days) 238 ■ the copyright owner. Furtneriep Reproduced with permission equilibrium, not accounting for the decay of 2 2 8 Th can lead to greatly overestimating the 2 2 4 Ra activity. The importance of 2 2 8 Th ingrowth from 2 2 8 Ra is presented in Figure A-3 for three 2 2 8Ra:2 2 8 Th ratios: 25, 100, and 250. These isotope ratios correspond to offshore, midshore, and shoreline samples, respectively. Despite the relatively long half-life for 2 2 8Th, because of the large 2 2 8Ra:2 2 8 Th ratio at the shoreline, ingrowth of 2 2 8 Th can become important. After 4 days, the 2 2 8 Th activity of shoreline samples may increase by 20%. For offshore samples, ingrowth is less important over the time scale of a few days. Combining these result, for nearshore samples, 2 2 8 Th ingrowth from 2 2 8 Ra is important, however, 2 2 4 Ra ingrowth from 2 2 8 Th is not. At offshore stations, 2 2 8 Th ingrowth from 2 2 8 Ra is not important, however, 2 2 4 Ra ingrowth from 2 2 8 Th is important. Therefore, the ingrowth of 2 2 8 Th from 2 2 8 Ra does not need to be taken into consideration when computing the 2 2 4 Ra activity. The ingrowth of 2 2 4 Ra from 2 2 8 Th does need to be computed. In general, the 2 2 8 Th activity was either near or below the detection limit. The 2 2 8 Th was measured on several large volume samples collected near the center of San Pedro Basin, Santa Monica Basin, and Santa Barbara Basin. The average 2 2 8 Th activity was 0.4 dpm m'3 . This value was used for all of the data. 239 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since 2 2 3 Ra has a relatively long half-life and the 2 2 7 Th activity is so small, ingrowth of 2 2 3 Ra from 2 2 7 Th is less important than ingrowth for 2 2 4Ra. By ignoring ingrowth, the decay rate for 2 2 3 Ra is: clN . d t X XX — — X N 223 v 223 (A-8 ) Following the same steps as above, the solution to this equation to convert measured counts to an activity concentration at the time of sampling is: 7 p j° _ 7 223 223 ” / l 223 1C 223 “^ 2 2 3 ^ 1 (A-9) where the variables are as defined above. P a r e n t I s o t o p e s : The RaDeCC system can also be used to measure the activity of the parent isotopes for 2 2 3 Ra and 2 2 4Ra. For example, after 100 days, 2 2 3 Ra is in equilibrium with 2 2 7Ac. The 2 2 4 Ra parents are more complicated. After just 20 days, 2 2 4 Ra is in equilibrium with 2 2 8Th. However, after two years, 2 2 8 Th is only half way to equilibrium with its parent, 2 2 8Ra. The activity of 2 2 8 Ra can be computed, assuming the 2 3 2 Th concentration is negligible: i X - O (X N -X^ N° e ~ k n t' \ /v 7 V v 77i / 'T h i 'l T h c \ ^Th / \ z . Z / (A-10) 240 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. where the subscripts Ra and Th represent 2 2 8 Ra and 2 2 8Th, respectively. Since 2 2 4 Ra is in equilibrium with 2 2 8Th, the measured ^ 2 2 4 N2 2 4 =^rh NX h . M u l t i p le M e a s u r e m e n t s : Ideally, multiple measurements should be made to confirm the initial count rate. These additional counts can be combined to improve the counting statistics. This is done using the following equation: <A - " > Where A X JN ); is the number of counts during interval i. Using this method, the final dpm is weighted to the number of counts during each measurement. E ffic ie n c y : To compute the total corrected counts, UN' (Equation A-3), the total efficiency must be known. The total efficiency can be divided into three parts: the system efficiency, the cell efficiency, and the emanation efficiency (Sun and Torgersen, 1998a). The mechanisms that control each efficiency are further discussed. The emanation efficiency is the ratio between the number of Rn atoms that recoils into the circulating air and the total number produced by the decays of adsorbed Ra. This ratio is a function of the water content of the fibers, and the release for 2 2 0 Rn is maximized when the water to Mn-fiber weight ratio is between 0.3 and 1 (Sun and 241 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Torgersen, 1998a). In this range, the emanation efficiency is 6 8 % for 2 2 4 Ra (Sun and Torgersen, 1998a). A method was developed to quickly reduce the moisture content on fibers. In the laboratory, approximately 0.5 L of deionized water is flushed through the sample chamber to remove any salt. Next, the sample is attached to a stand and compressed air is passed through the column at a moderate flow rate, approximately 95 L min'1 . First, any bulk water which is saturating the fibers is forced from the fibers. For the remainder of the time, the relatively dry air passed through the fibers removes water vapor. The chamber actually cools as the water inside evaporates. After three minutes, the compressed air is turned off, and any residual water is cleared from the inlet and outlet. Then, the sample chamber is reversed, and compressed air is again passed through the chamber from the outlet side for one minute. This is often noisier than the first flush, because the air whistles through the hole in the rubber stopper at the inlet end. In Table A-l, the ratio of water content in the fiber is computed as a function of the flush time. The experiment was repeated 3 times to test the importance of variable flow rates and volumes of fiber. The cell efficiency is the ratio between the number of decays recorded by the counter and the total number of decays in the cell. This will depend on the PMT voltage, the density of the gas in the cell, and the alpha decay energy. The ability of the PMT to receive signals depends on the voltage. At low voltage, the PMT will not record any counts. As the voltage increases, the PMT efficiency will increase to a point where it 242 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-1. Summary of drying fibers experiment. Fibers were first hand squeezed. Then air was passed over the fibers for 3 minutes at a flow rate of ~95 L min1 . The column was then reversed (air in the outlet), and air passed over the fiber for 1 more minute. F IB E R 1 Mass (a) % Water F IB E R 2 Mass (a) 9 6 Water Dry* 30.9 Dry* 19.7 Hand S queeze 98.6 21 9 % Hand S queeze 91.3 36 4 % 3 minute 59.9 94% 3 minute 42.5 116% 1 minute 58.2 88% 1 minute 41 .8 112% Hand S queeze 9 9 .0 22 0 % Hand S queeze 78.3 29 8 % 3 m inute 57.0 85% 3 m inute 39.3 99% 1 m inute 56.2 82% 1 minute 38.6 96% Hand S queeze 91.4 196% Hand S queeze 7 7 .8 295% 3 minute 56.8 84% 3 minute 39.0 98% 1 minute 56.0 81% 1 minute 3 8.0 93% Hand S queeze 95.3 20 8 % 3 minute 57.4 86% 1 minute 56.5 83% Summary F IB E R 1 Mass (a) 9 6 Water F IB E R 2 Mass (g) 9 6 Water Dry* 30.9 Dry* 19.7 Ave. Hand Squeezi 96.1 21 1 % Ave. Hand Squeez< 82.5 319% Ave. 3 min. 57.8 87% Ave. 3 min. 4 0.2 104% Ave. 1 min. 56.7 84% Ave. 1 min. 3 9 .4 100% * Fibers dried in 60°C oven for 3 days 243 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reaches a plateau (Figure A-4). Increasing the voltage beyond this plateau generates spurious counts and increases the system background. Therefore, the voltage of each cell must be set near the midpoint of the plateau, so that minor voltage fluctuations do not result in changes in efficiency. When the RaDeCC system is used in different computers, the voltage for the plateau may change. The counting efficiency in the cell is sensitive to the density of the carrier gas. As the carrier gas density increases, the distance an alpha particle can travel decreases. To allow for the maximum recoil distance, helium is used as the carrier gas. Unfortunately, during a sample run, some air inevitably diffuses into the system, increasing the carrier gas density. This results in an increase in the flow rate recorded on the flowmeter, which is sensitive to the gas density. The importance of air in the system was tested by adding a known volume of air to the system, then returning the system to room pressure with He. In order to maintain a constant flow rate, the flow rate was first set while there was only helium in the system. The system efficiency as a function of the fraction of air is presented in Table A-2. Because of the decrease in efficiency, any sample run that ended with a flow meter reading greater than 15 was not used. The fraction of air needed to significantly decrease the efficiency is >20%. This is less than the 50% fraction of methane required to reduce the efficiency of 2 2 2 Rn in smaller, 0.11 L and 0.25 L, Lucas cells (D.E. Hammond, unpublished data). 244 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 220R n C P M 220R n CPM Figure A-4. Count rate of 220Rn and 219Rn as a function of voltage. Measurements made during March 2004 and corrected for chance counts. 35 30 25 20 15 10 5 0 1 1 1 Channel 1 I I □ □ □ • - • .i __ __ □ B □ 2l9Rn CPM - • 2 2 0 Rn CPM - ...... S...T. . . . . . . . . . , ....I ...............I ............. 650 700 750 800 850 Voltage (V) 900 35 30 25 20 15 10 5 0 3.5 3 2.5 2 1.5 1 0.5 0 950 I l I Channel 3 D i i o □ □ • - □ # • □ “ □ • • □ - • D 2 1 9 Rii CPM • 2 2 0 Rn CPM ............... i .............. i i I . I 3.5 3 2.5 2 1.5 1 0.5 0 650 700 750 800 850 Voltage (V) 900 950 245 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 219R n C P M 219R n CPM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-2. Efficiency of “ Ra as a function of the fraction of air in RaDeCC system. The total system volume is 1.4 L. The "Flow" column presents the flowmeter reading. Vol Air Added (cm3 ) Fraction of Air Flow Reading 2 2 3 Ra C P M 2 2 4 Ra C P M 2 2 3 Ra Norm. 2 Z 4 Ra Norni 0 0 % 10.5 1.61 20.7 1 0 0 % 1 0 0 % 2 0 0 14% 12.5 1.97 23.3 1 2 2 % 113% 300 2 1 % 15 1.50 19.2 93% 93% 500 36% 15+ 1.58 17.4 98% 84% 700 50% 15+ 1.07 12.7 6 6 % 62% 1400 1 0 0 % 15+ 0.54 6.9 33% 34% N > o\ Table A-3. Half-lives and alpha particle energies for Ra, Rn, and Po. Isotope Half-life Alpha Particle Energy (MeV) 223R a 11.435 day 5.72 224R a 3.66 day 5.67 226R a 1600 y 4.78 228R a 5.76 y — 219Rn 3.96 s 6.82 220Rn 55.6 s 6.29 222Rn 3.8235 day 5.49 215Po 1.78 ms 7.39 216Po 1 50 ms 6.78 218Po 3.05 min 6.00 247 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Higher decay energies produce more energetic alpha particles, which are more likely to interact with the cell and generate a signal. The alpha decay energy is inversely correlated with the isotope's half-life and are presented in Table A-3 (Geiger and Nuttall, 1911; Rutherford, 1906). The Po daughters are more likely to be observed than the Rn parent. For reference, 2 2 2 Rn efficiency in Channel 3 is about 57%. The system efficiency (Es y s ) is defined as the ratio between the activity in the counting cell (ACc e ll) and the Rn production rate (P). The system efficiency can be estimated by developing a mass balance equation for each component of the RaDeCC system: Chamber: P + Q C om = Q C ch + X V chC ch (A-12a) Line In: C in = C ch exp(-AV,.„ I Q ) (A-12b) Counting Cell: Q C in = Q C cell + A V cellC cell (A-12c) Line Out: C O M = C celle x p ( - A V oul/ Q ) (A-12d) where Q is the flow rate and A is the Rn isotope decay constant. The volume, V, and concentration, C, are each for the respective subscript, where o u t is the line between the cell and the chamber, including the pump and flow meter, in is the line between the chamber and the cell, c e ll is the counting cell, and c h is the sample chamber. The chamber and counting cell are treated as well-mixed reservoirs, while “plug” flow is 248 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. assumed in the lines connecting the two. Solving these equations for the system efficiency: - - ( k y M + Q X A -V V fg) _ Q {_ k y /Q) ( A ' 1 3 ) •Qexpl-Al',, IQ, This equation shows that the main controls on system efficiency are the flow rate and the volume in each reservoir. Most decays will occur in the counting cell, which accounts for 80% of the total volume. Most of the remaining volume is in the sample chamber. The lines between the cell and chamber account for less than 5% of the total volume. In Table A-4, the system efficiency is calculated at a variety of flow rates. For 2 2 4Ra, the efficiency is predicted to change little over the range of flow rates. 2 2 3 Ra is more sensitive to changes in the flow rate. The system efficiency was also measured using a standard and changing the flow rate (Table A-4). Changes in the system efficiency are reflected in the relative change in the count rate (CPM). The observations show that the system is more sensitive at low flow rates than predicted. The most likely reason is the system plumbing. Since the flow rate is controlled by circulating the gas stream through the pump, as the flow rate decreases, the residence time in the pump increases. This increase in the residence time is sufficient to reduce the flux of Rn returning to the sample chamber and counting cell. The increase in residence time in the line out (xo u t) can be estimated using Equation 2 and assuming xo u t=Vo u t/Q. At a flow rate of 249 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-4. Efficiency of 2 2 4 Ra as a function of the flow rate. A) Theoretical estimate of system efficiency at different flow rates. The relative efficiency is normalized to a flow rate of 8 L min '. B) Observed changes in count rate (CPM) of at different flow rates. The relative efficiency is computed by normalizing the observed CPM to the observed CPM at a flow rate of 8 L min '. A ) Theoretical System Efficiency Flow Rate 2 2 3 Ra Efficiency 2 2 4 Ra Efficiency (L min’1 ) System Relative System Relative 5.2 0.51 0.90 0.70 0.99 8 0.57 1.00 0.71 1.00 10 0.60 1.04 0.71 1.00 13 0.62 1.09 0.71 1.01 B) Observed System Efficiency Flow Rate 2 2 S Ra Efficiency 2 2«R a Efficiency (L m in1 ) Ota. C P M Relative Ota. C P M Relative 5.2 2.1 0.74 21.5 0.85 8 2.8 1.00 25.2 1.00 10 2.7 0.98 23.7 0.94 13 3.5 1.23 24.9 0.98 250 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-5. RaDeCC Channel 1 2 2 4 Ra system efficiency computed for samples processed on both RaDeCC and by gamma spectroscopy. Sample Gamma D P M Recovery Efficiency Total D P M RaDeCC Ch. 1 C P M RaDeCC System Eff. 403.1 20.56 0.990 20.77 6.46 0.311 403.2 20.22 0.990 20.42 7.23 0.354 403.3 14.6 0.828 17.64 6.59 0.374 403.4 15.52 0.672 23.11 8.59 0.372 403.5 18.62 0.990 18.81 7.23 0.384 Averaae: 0.36 ± 0.03 251 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.2 L min'1 , the residence time in the line out must increase by a factor of 7 to account for the observed decrease in efficiency. S y s te m E f f ic i e n c y : Finally, the total efficiency of the RaDeCC system can then be estimated as the product of these three efficiencies. The estimated efficiency is 29% for 2 2 3 Ra and 46% for 2 2 4 Ra. The actual efficiency was measured two different ways. First, the 2 2 4 Ra efficiency was computed by measuring samples on both the RaDeCC system and by gamma spectroscopy. Five samples were collected at Huntington Beach, and measured using the RaDeCC system. A BaS04 precipitate was then made from each sample and counted using gamma spectroscopy (Elsinger et al., 1982). This method is further discussed in the next section. For 2 2 4 Ra, the gamma spectrometer was standardized with an EPA Diluted Monazite standard (SRM- DM2). The BaS04 precipitation efficiency of some of the samples was >100% and was attributed to incomplete drying. The recovery efficiency for these samples was assumed to be 99%. The average observed efficiency was 0.36± 0.03 (Table A-5). For the efficiency on Ch. 3, a series of samples were run on both systems. A linear fit, forced through the origin, gives the correlation between these two channels (Figure A-5). Later, a standard was developed by soaking fibers in a mixture of dissolved pitchblende and monazite. These standards were chosen, because all of the radiogenic isotopes in each of the decay series should be in equilibrium with the parent isotope. And, Ra, along with its Pa, Ac, and Th parents, should readily adsorb 252 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure A-5. Count rate of 220Rn and 219Rn from samples counted on both channels. Count rate corrected for coincident decays. CL oo 0 .5 <M CM a t Slope = 1 .2 0 5 9 0 0.5 1.5 223Ra Channel 1 CPM 20 u CM CM Slope = 1 .2 0 5 9 0 10 5 15 20 224Ra Channel 1 CPM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to the fiber. The dissolved monazite solution, made with EPA Dilute Monazite standard (SRM-DM2) used was made on 11/3/92 by G.K., and contained 27.45 dpm g-i 2 3 2 ancj 1 8 3 dpm g-1 2 3 0 Th. A new standard was made by dissolving 2.187 g of EPA Diluted Pitchblende standard, SPM-DP2. The resulting standard had an activity of 5.55 dpm/g for 2 3 8 U and 0.255 for 2 3 5 U. For the standard fiber, 44.55 g of the pitchblende standard and 2 g of the monazite standard were diluted with 165 g of water. The final standard fiber, STD 8, made on 2/3/04 had 247.1 dpm 2 3 8 U, 250.78 dpm 2 3 0 Th, 11.38 dpm 2 3 5 U, and 54.89 dpm 2 3 2 Th. A second standard was also made from these same standards on 6/11/04. The final activity of Standard 27 had 55.56 dpm 2 3 8 U, 57.54 dpm 2 3 0 Th, 2.56 dpm of 2 3 5 U, and 29.76 dpm 2 3 2 Th. The system efficiency changed dramatically twice (Table A-6). First, the efficiency dropped considerably when the RaDeCC system was connected to a new computer. The reason was because of a change in the voltage plateau, which had shifted to a higher voltage. The second change was after the voltage was adjusted to the center of the new voltage plateau. Unfortunately, a 2 2 3 Ra standard was not run on the RaDeCC system with the old computer. The 2 2 4 Ra efficiency for the old computer suggests that the voltage was set lower than the plateau voltage. The voltage that corresponds to the old 2 2 4 Ra efficiency can be extracted from the voltage plateau, along with the previous 2 2 3 Ra efficiency. R e - P a c k i n g : Adsorbed Ra is most likely not uniformly distributed on fibers in sample cartridges. Instead, it is probably concentrated near the inlet and along rapid 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-6. Isotope efficiency in each channel. Changes in efficiency are explained in the text. 223Ra Z Z 4Ra Date Ch.l Ch.3 Ch.l Ch.3 9 /0 1 -5 /0 3 0.184 0.241 0.359 0.435 5 /0 3 -3 /0 4 0.075 0.160 0.102 0.243 3/0 4 - 0.247 0.274 0.460 0.461 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flow paths. (Note: The inlet has been defined as the removable cap on the cartridge.) Thus, when counting the fibers, the orientation of the cartridge and repacking the fibers might be critical in producing reproducible results. To test the variability in counts as a result of re-packing, 5 hot samples collected at Huntington Beach were used (Table A-7). These samples were kept in the chamber after sampling. Several different cartridge orientations were tried: 1) cartridges were installed with the water inlet closest to the counter, 2) cartridge outlet closest to the counter, 3) the cartridge was repacked and counted with the cartridge inlet closest to the counter, 4) repacked, with the cartridge outlet closest to the counter. Samples were then counted for as long as was required to obtain approximately 5% counting uncertainty (about 400 counts) for 2 2 0 Rn. All samples were corrected for chance counts and decay since the sample was collected. The results were normalized to the initial counting, which is expected to have the highest count rate. For 2 2 4 Ra, changes in the orientation of the column or repacking may have reduced the count rate a little, however, it is within the range of the counting statistics (Table A-7). For 2 2 3 Ra, the counting statistics resulted in greater uncertainty, but it is still clear that column orientation and repacking significantly influence the count rate. To account for the effect of repacking on the 2 2 3 Ra count rate, samples that were repacked were corrected by an efficiency of 69%. This only impacts some data collected before 6/02. 256 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-7. Sensitivity of counting efficiency to changes in cartridge orientation and repacking. Normalized Activity Sampje__________ ^ R a _______________ !£ £ a R E V E R S E D 201..1 ac 1.27 + 0.45 0.92 + 0.09 201 .2c 0.67 ± 0.20 0.81 + 0.07 201 .3c 0.70 ± 0.19 1.18 + 0.08 201 .4c 0.76 + 0.16 0.95 + 0.05 201,,5ac 0.80 + 0.20 0.96 + 0.06 0.84 ± 0.25 0.96 ± 0.13 R E P A C K E D 201.2d 0.72 ± 0.21 0.84 ± 0.07 201.3h 0.65 ± 0.18 0.95 ± 0.07 201.3d 0.61 ± 0.17 1.15 ± 0.09 201.4d 0.50 ± 0.14 0.89 ± 0.05 201.4h 0.79 ± 0.15 0.90 ± 0.04 201.5ad 0.88 ± 0.19 0.94 ± 0.06 0.69 ± 0.14 0.95 ± 0.11 R E P A C K E D , R E W E T T E D 201.2e 0.59 ± 0.17 0.85 ± 0.07 201.3i 0.54 ± 0.16 1.20 ± 0.09 201.5ae 0.51 ± 0.13 0.80 ± 0.05 201.4i 0.99 ± 0.21 0.88 ± 0.05 0.66 ± 0.23 0.93 ± 0.18 R E P A C K E D , REWETTED, R E V E R S E D 201.3e 0.42 ± 0.13 1.19 ± 0.09 201.3j 0.58 ± 0.16 1.18 ± 0.09 201.4e 0.49 ± 0.13 0.89 ± 0.05 201.4j 0.72 ± 0.16 0.92 ± 0.05 0.55 ± 0.13 1.05 ± 0.16 257 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. III. RaDeCC Method • Computer: - Turn computer on. - From START menu, start RaDeCC program • Cartridge Preparation - Loosely pack fibers into cartridge. Wrap threads in Teflon tape and twist on the cap. - Flush ~0.5 L DIW thru fibers to wet fibers. - Blow air thru fibers from top (threaded end) for 3 min.; flip over and blow air thru for 1 min. Flow rate ~ 95 L min-1 . Avoid any large gaps between pieces of fiber. • RaDeCC Preparation - With tubing in place of cartridges, flush system with helium twice. - Draw vacuum on system for at least 5 minutes to remove any moisture. - Fill with helium. • Sample Run - Attach samples with the top-side up (Inlet = removable cap). - Flush samples and system twice with helium. - Turn on pump and adjust flow rate to 10.5. - On Computer, Press Start button for desired channel. This opens the save dialog. - Save file name (MonthCollected_DayCollected_Station#_Sample#RunLetter) - 8 minutes after starting pump, begin counting (Click RESET button) • Samples Run until >400 220Rn counts recorded (5% counting statistics) • Clean-Up - When sufficient counts have been recorded, press STOP button in RaDeCC window. - Turn off pump, remove samples, attach tubing. - Flush system with helium; Draw vacuum on system for at least 10 minutes to remove moisture. 258 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. Radium Extraction from Mn Fibers The following procedure must be performed in the fume hood. There are four initial solutions which must be prepared: 1 N NH2 OH-HCl in IN HC1, IN HC1 Rinse Solution, 20% H2 S04 Solution, and X M BaCl2 solution. BaCl2 is hydroscopic, and therefore it is important to use dried powder. When the stripping efficiency must be computed, based on the difference in Ba through the extraction and precipitation, it is best to use a non-hydroscopic Ba salt, such as Ba(N03 )2 . Procedure 1. In 1 L Beaker, to 400 ml of 1 N NH2 OH-HCl- IN HC1, add 20 ml BaCl2 solution and heat to 50°C. (Use remaining N NH2 OH-HCl- IN HC1 solution to rinse vacuum flask!) 2. Add fibers and increase temperature to 90-95° (avoid boiling solution). Heat until fibers are completely bleached (< 90 minutes). Stir often to expose entire fiber to solution. While waiting, the vacuum filtration unit can be assembled and the H2 S04 solution can be prepared. 3. Remove from heat and allow to cool. 4. Vacuum filter through qualitative filter paper (Whatman #1), wring out fiber, and rinse with IN HC1 (~300 ml). Wring out fibers again, filter, combine washings. 5. While stirring, Add 25 ml H2 S04 solution to sample solution. 259 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. Allow precipitate to settle. Assemble Filter unit and check for leaks. 7. Collect precipitate by filtering (Whatman #42 filter) remaining liquid and precipitate. Wash with 50 mils IN HC1, then with 50 ml DIW. 8. Place filter and precipitate in oven to dry overnight. 9. Transfer precipitate to a weighed polystyrene test tube. Pack tube by tapping on counter top, but only fill to 3 cm high. Place additional precipitate in a second test tube. Re-weigh sample + test tube. Important numbers to record when making precipitate in order to calculate a) precipitation efficiency (assume 100% efficiency making solution, 1:1 ratio of Ba:Ra when precipitate is made, mass difference between Ba and Ra in precipitate is negligible), and b) ingrowth and decay of daughter products on fibers and in precipitate. 1. Sample Name, Date, Time Collected 2. Weight of BaC12 added to solution 3. Precipitation date, time 4. Weight of test tube 5. Weight of test tube and sample 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V. Gamma Method BaS04 precipitates were measured using a high-purity germanium (HP-Ge) well-type detector. When counting BaS04 precipitates, the following peaks were used: 2 2 6 Ra (186.2 keY), the 2 2 4 Ra, 2 1 2 Pb, and 2 1 4 Pb combined peak (239 keV), 2 l4 Pb (295.2 and 609.3 keV), 2 2 8 Ac (338.4, 911.1, and 968.9 keV), and 2 1 4 Bi (609.3 keV). To compute the initial 2 2 4 Ra activity, the method of Elsinger and co-workers was followed (Elsinger et al., 1982); for the 2 2 8 Ra activity, the method of was followed (Michel et al„ 1981). Two standards were used to compute the counting efficiency of the various peaks used: U.S. EPA Standard Diluted Pitchblende ore SRM-DP2 and U.S. EPA Standard Diluted Monazite ore SRM-DM2. The diluted pitchblende ore had an initial 2 3 8 U activity of 253 pCi g'1 , and 2 3 5 U activity of 12 pCi g 1 on 5/18/76 at 4:00. A test tube was filled to 3 cm with 2.041 g of this standard. The diluted monazite standard had an initial 2 3 2 Th activity of 150 pCi g'1 and 2 3 0 Th activity of 10 pCi g 1 on 5/19/76 at 4:00. A test tube was filled to 3 cm with 1.9564 g of this standard. Standards were run every few months and never varied by more than 1%. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VI. 2 2 6 Ra Method Samples were collected in 20 L glass bottles. 2 2 6 Ra was measured based on 2 2 2 Rn ingrowth (Mathieu et al., 1988). Samples were first flushed with He to remove any initial 2 2 2 Rn. After 2 weeks, more than 90% of the 2 2 2 Rn has grown into secular equilibrium with 2 2 6 Ra. The 2 2 2 Rn is then extracted by circulating He through the sample and passing it over an activated charcoal cold trap. After 90 minutes, 99% of the Rn is removed from the sample. The charcoal trap is then heated to 400°C, releasing the Rn which is transferred to a Lucas scintillation cell. The sample-filled cell is connected to a photomultiplier tube which records the number of alpha particles produced. The efficiencies of each extraction system and counting cell has been computed based on GEOSECS, EPA and internal standards. The computation of the sample activity accounts for any disequilibrium between 2 2 2 Rn and 2 2 6 Ra at the time of sample processing. Pore fluid Rn was analyzed using a rapid radon extraction system (Berelson et al., 1987). In brief, the sample is injected into a stripping chamber in an evacuated system. A single pass of He through the solution strips out the Rn, and the gas is carried over anhydrous CaS04 to remove water vapor, followed by a NaOH impregnated clay to remove C 02 . A Lucas scintillation cell, coated with a ZnS phosphor, collects the gas as the pressure is returned to ambient. The cell is then placed in a detector as outlined by Mathieu and others (Mathieu et al., 1988). At this 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sample volume, the stripping efficiency was calculated to be 95% based on running a second extraction for several samples. 263 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VII. Field Methods- The Radiator Ra can be extracted from seawater at high efficiency by filtering water through manganese-impregnated fibers (Moore, 1976). The Radiator was developed in order to filter large volumes (> 100 L) of seawater from either the shoreline or at designated depths from a small vessel. The Radiator was developed using parts that could easily be purchased at local hardware and marine/boating supply stores, which simplified acquiring parts and making repairs. A diagram of the ship-board Radiator is presented in Figure A-6, and a table of parts is presented in Table A-8. Initially, a DC submersible well pump, which could be lowered to the desired sampling depth, was used to collect samples. While these pumps were inexpensive, they were not designed for marine applications and experienced a short life span. Heartier pumps designed specifically for use on boats were found to have a good flow and last longer. Using a car or power tool 12-V battery as a power source and making other adjustments noted in Table A-8, the Radiator could be to collect samples at the shoreline. The Radiator must be assembled out of reach of the waves to protect the battery. A hose could not be deployed to collect the water from the surf zone because of the long distance to the water and bubbles entrained in the water from breaking waves. Instead, a 10 or 20 L bucket was used to collect samples from knee- depth water. Water was decanted off any sand that collected in the bucket and into a 264 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure A-6 . Schematic diagram of the radium extraction system- the Radiator. Letters and numbers represent corresponding parts in Table A-8 . 2A 2A 2A 40 50 30 K > O n U \ Table A-8 . List of parts to construct the Radiator. Letters and numbers correlate with parts show in Figure A-6 . T U B IN G 1 5 /8" Garden Hose 2 3 /8 ID; 9/16 OD Tygon tubing 2a 2-4 inches 2b 1 5-20 feet 3 9" of 1/4 ID tubing C O N N ECTO RS 10 3 /8 " tubing to 5 /8 " garden hose 20 3 /8 " tubing multibarbed T connector (7.5 cm) 30 3 /8 " tubing to 5 /8 " garden hose (Female);5/8" garden to pipe union; 5.5 cm (Ideally- 3 /8 " tubing to 5 /8 " pipe (male)) 40 5 /8 " pipe (female) to 1/2" male pipe; 1/2" to 1/4" bushing; “ U 1/4" male pipe to 1/4" tubing; 7cm (Ideally- 5 /8 " pipe (female) to 1/4" tubing 50 1 /4 " quick connect H A R D W A R E A Battery operated submersible pump (16cm) (Ben Meadows- 22121 6-DC-40 Purging Pumps $95) R Filter Housing w / 10" 5pm filter; 13x32cm (McMaster Carr: D 4422K3 $25.1 5; 4411 k74 $2.08) C Totalizer; 23 cm: (McMaster Carr: 4 1 19K51 $1 56.25) D Mn02 Filter Cartridge M ISC E L L A N E O U S 2 3 /8 " Hose clamps 7 9 /1 6 " hose clamps Many TieWraps (8 or 8.5") Much Electrical Tape Variation for Shoreline Sampling: Missing: 10, 2a, 2b, 20 A Battery operated self-prime pump (WestMarine) 30 5/8"garden to 5/8"tubing; 5/8"tubing; 5 /8 " tubing to 5 /8 " J V garden; 5 /8 " garden to pipe union Garbage can 5-qallon bucket 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 L (35-gallon) plastic can. The water was allowed to settle for about 10 minutes before the water was filtered through the Radiator. S tr ip p in g E ffic ie n c y : To test the how the mass of fiber influences the efficiency of Ra extraction, a second extraction was made on several samples (Table A-9). These results show that as the mass of fiber in the cartridge decreases, so does the stripping efficiency. For the remainder of the sample collections, a mass of at least 30 g of fiber was used. 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-9. Stripping efficiency of 2 2 3 Ra and 2 2 4 Ra as a function of weight of fiber in the cartridge. Fiber weight measured after drying in 60°C oven for 3 days. Fiber weights in italics are estimates. Sample Dry Fiber WL* (ii) Stripping Efficiency 223Ra 224Rh 103.2 22.58 0.966 0.965 103.5 22.89 0.841 0.910 103.3 24.72 0.911 0.923 103.4 30.02 0.967 0.936 101.1 32 0.926 0.901 102.1 32 1.000 1.000 102.5 32 1.000 1.000 Average 0.922 0.927 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VIII. Shoreline Sampling Protocol A list of field equipment needed for shoreline sampling is presented in Table A-10. 1. Arrive at beach early (7:30-8) Few people, easy parking, long day of sampling ahead - Find beach access point where car can be unloaded. Set gear out of the way and as best out of sight as possible. - move car to safe parking spot for the day. 2. Set up on beach. - 1 like to set up near a lifeguard tower so that shade is nearby. Note where the high-water mark is. Set up equipment inland from this point. If water ever does approach the pumping system, the keeping the battery dry is the top priority. 3. Sample Collection: - using pail, COMPLETELY fill up the garbage can with water from knee deep. This will take 10-15 minutes. Note the time when bucket is full. - Keep as much sand and suspended matter out of the garbage can as possible. Before pumping, allow water in garbage can to settle for 5-10 minutes. - Note new Mn02 filter number, flow totalizer value. Attach new Mn02 filter cartridge (threaded end is inlet; Numbered end is the outlet), put hose in water (about half the depth of the garbage can), begin pumping. Note time. - Pump for 60 min. or until there is only 0.5-1 gallon of water remaining. Tilt garbage can as necessary to get last bit of water. DO NOT LET PUMP RUN DRY! Note the end pumping time and totalizer reading. - Rest up until next sample collection. Collect samples every 2 hours for a total of 5 samples. Note the weather conditions, changes in wind, fog breaking up, changes in waves, cute girl's phone numbers. 4. End of the day - Carry everything back across the beach. Get the car, load the car. Go home. Bring samples to school on the next day. Possible Problems: • Pump does not self-prime: fill garden hose with water to prime, turn pump on and put hose in can. • Totalizer stops working: take it off the pumping system and alternate blowing air through and pouring clean water through until dial starts to move again. 269 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A-10. Packing list for shoreline sampling. Work Items______________ Personal Items Data book pen garbage can bucket garden hose pump radio/book sunscreen lunch/snack USC clothing towel chair filter totalizer Mn02 filter cartridges 12 V battery teflon tape black tape screwdriver 270 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix B Fortran code for program HBRE1, a two-dimensional mixed layer model extending from Sunset Beach to Newport Beach. PROGRAM hbrel C hbrel.f 7/14/04 C C 2-D Radium in surface water - C CONSTANT or variable Diffusion, reaction, Longshore Adv. C Model contains 30x80 boxes C Integer (I,J) define the (longshore, offshore) box, with (1,1) at C the north- east corner. C Longshore Dist. between boxes is equal at 600 m C Box length increase by 5 m per box offshore. C Eddy Diffusivity is CONSTANT with distance offshore. C Box thickness increases out to seafloor-mixed layer intercept. C Ra Input Flux is divided between the shoreline and seafloor C Seafloor flux decreases exponentially offshore C Observed 223Ra and 224Ra for a given transect are input to model. C Model is fit to the data (file=HBAVE) by computing the reduced-chi squared, C which is minimized to compute the best fit parameters. C C Two independent variables: C l) Inventory at the northern boundary as a fraction of steady-state inputs @ C Huntington Beach C 2) Eddy diffusivity C C Model is divided into three sections: MAIN, FUNCTION, and SUBROUTINES: C MAIN program allows data to be input and computes box geometry C FUNCTION does all the work. Computes mass balance for each box. C FUNCTION output is the chi-sq value. C SUBROUTINES do computations used in the model, and include C STEVE transfers variables input from MAIN program to FUNCTION. C FLUXUB computes the flux into each box C BOUND computes the steady-state conc. at N and S boundary. C POWELL, BRENT, and associated subroutines can be added to the C model to vary the 223Ra Flux, 224Ra Flux and mixing rate to minimize the C reduce chi-squared. C C There are two OUTPUT files: 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C HBRE10UT.XLS, which spits out the box number, distance C alongshore and offshore, 223Ra and 224Ra concentrations C CHISQ.xls is a record of Northern Boundary inventories, Eddy C diffusivities and the resulting computed ChiSq. PARAMETER(NDIM=3, NLONG=32, NOFFS=80) COMMON /VARIES/ GEOM, CON, FIELD DIMENSION V(26),FIELD(10,6), VOL(NOFFS,5) DIMENSION GEOM(NLONG,NOFFS,6 ) REAL CON(25),p(NDIM),xi(NDIM,NDIM) CHARACTER PARA(26)*60 INTEGER I, J, IDIST, INTI, ITER REAL MIXED,WIDTH, LENGTH, FRET, SFINTR, LENADD REAL FINAL, ANS C Clear Matrix FIELD(I,J) 5 DO 8 J=l,3 DO 7 1=1,10 FIELD(I,J)=0.0 7 CONTINUE 8 CONTINUE C Option to Input field data from Huntington Beach or use average profile. DO 9 1=1,10 PRINT *,'FIELD DATA1 PRINT *, 'Enter Dist. from Shore (km)(Nearshore->Offshore)' PRINT *, ' ( 8 8 For Average Data, 99 When Finished)’ READ *,FLAG IF (FLAG.EQ.8 8 ) THEN FF=FLAG GO TO 10 ENDIF IF (FLAG.EQ.99) GO TO 15 FIELD(I,1)= FLAG* 1000 PRINT *, 'Enter 223Ra Conc. (dpm/m3)' READ *, FLAG FIELD(I,2)=FLAG PRINT *, 'Enter 224Ra Conc. (dpm/m3)' READ *, FLAG FIELD(I,3)=FLAG 9 CONTINUE 10 CONTINUE C Average Huntington Beach Data (If 8 8 is chosen above) 272 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C Stored in a file named HBAVE with a format F9.3 with one value C for Array FIELD on each line C FIELD(I,J): I=Sample Number; J:l=Distance (m); 2=223Ra conc. (dpm/m3) C 3=224Ra conc. (dpm/m3); 4=Corresponding model box (computed later) C 5=223Ra uncertainty; 6=224Ra uncertainty IF (FF.EQ.8 8 ) THEN OPEN (UNIT=1, FILE-HBAVE', FORM='FORMATTED', 1 ACCESS='SEQUENTIAL', STATUS=’ OLD') 2 FORMAT (F9.3) DO 131=1,10 DO 12 J= l, 6 READ (UNIT=1, FMT=2) FIELD(I,J) 12 CONTINUE 13 CONTINUE ENDFILE (UNIT=1) CLOSE (UNIT=1) ENDIF WRITE (*,'(/lx,al0,2x,al0,2x,al0/)') 1 'Dist. (m)','223Ra Conc',’ 224Ra Conc' 15 CONTINUE DO 161=1,10 WRITE (*,'(2x,F8.1,4x,F6.2,8x,F6.2)') 1 FIELD(I,1), FIELD(I,2), FIELD(I,3) 16 CONTINUE PRINT *,'Did you type it in correctly? (0-9=Yes, 10>No)' READ *,NFLAG IF (NFLAG.GE.10) GO TO 10 C Input Model Variables C Model input taken from best fit Ra model 17 PRINT *, 'HBREls Got ya Covered' PRINT *, 'ADJUSTABLE PARAMETERS ARE:' PARA(l) = 'HBRE1: Mixed Layer Thickness (m)' V(l) = 11.8 PARA(2) = 'Southern Transition Box' V(2) = 29 PARA(3) = 'Tolerance test factor for convergence ( not used)' V(3) = 0.001 PARA(4) = ' Initial Kh Value (m2/sec)' V(4) = 1.3 PARA(5) = ' Longshore Ave. Advection Rate (Pos.=DownCoast;cm/sec)' V(5) = 5.0 273 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PARA(6 ) = 'Shore Ra Grad (Frac. dJ at N relative to Jave) 1 V(6) = 1 PARA(7) = ' Seafloor Flux Scale Length' V(7) = 2E-4 PARA(8 ) = 'Initial Fraction of Ra @ Northern boundary' V(8 ) = 0.32 PARA(9) = ' 223Ra Seafloor Flux (dpra/m/sec)' V(9) = 0.2971 PARA(IO) = '224Ra seafloor Flux (dpm/m/sec)' V(10) = 4.7292 PARA(11) = ' 223Ra Shoreline Flux (dpm/m/sec)' V (ll) = 0.0179 PARA(12) = '224Ra Shoreline Flux (dpm/m/sec)' V(12) = 0.3454 PARA(13) = " V(13) = 0. PARA(14) = '' V(14) = 0. PARA(15) = 'To Turn On Tides=1.0; Just Tidal Advection=2.0' V(15) = 0. PARA(16) = ' Use Okubo (1976) Scale Dep. Mixing=l' V(16) = 0. PARA(17) = '' V(17) = 0. PARA(18) = '' V(18) = 0. PARA(19) = '' V(19) = 0. PARA(20) = 'Ra dist. at N. Boundary scale dist. factor (1=Sunset)' V(20) = 1 PARA(21) = ' 1 V(21) = 0. PARA(22) = '' V(22) = 0. PARA(23) = 'Run Model Only ITime if =1.0' V(23) = 0. PARA(24) = 'TIME TO END SIMULATION (DAYS)' V(24)= 12 PARA(25) = ' TIME STEP IN HOURS' V(25) = 0.045 PARA(26) = 'Input New Field Data' V(26) = 69 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C Adjust model parameters 19 DO 20 1=1,26 20 PRINT 1,1, V(I), PARA(I) 1 FORMAT (13,' = E10.3,' A60) PRINT *, 'ENTER # OF PARAMETER TO CHANGE, 0=RUN, 99 =EXIT READ *,NFLAG IF (NFLAG.EQ.99) GO TO 1000 IF (NFLAG.EQ.0) GO TO 100 IF (NFLAG.EQ.26) THEN PRINT *,'The Old Data Was:' DO 30 1=1,10 PRINT *,FIELD(I,1), FIELD(I,2), FIELD(I,3) 30 CONTINUE GO TO 5 ENDIF IF (NFLAG.GT.26) GO TO 19 PRINT *, 'ENTER NEW VALUE FOR ', PARA(NFLAG),'(',V(NFLAG),')' READ *, V(NFLAG) GO TO 19 100 CONTINUE C Fill VOL array with geometry and location of each box at Huntington Beach C WIDTH=Box Width = 500 m; MIXED= Mixed Layer Thickness (m) C LENGTH = Box Length (Shoreline = 50 m) ; INTI=last box on seafloor C Each Box length increases by LENADD WIDTH = 500 MIXED = V(l) LENGTH = 50 LENADD = 5 C Calculate offshore box midpoint, depth, box volume for C Standard/Huntington Beach profile. C VOL: 1= midpoint, 2 = midpoint depth, 3 = volume, C 4=Seaward distance, 5= Seaward Area C SFINTR=dist to mixed layer-seafloor intercept (m); C Seafloor slope is 0.01 to 890 m, then 0.00417. C Boxes increase in length offshore by 5m IF (MIXED.LE.(8.90)) THEN SFINTR = MIXED/0.01 ELSE SFINTR = 890 + (MIXED - 8.9)/0.00417 ENDIF VOL(l,l) = LENGTH/2 275 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V0L(1,2) = LENGTH*0.01/2 V0L(1,3) = LENGTH* WIDTH* VOL( 1,2) V0L(1,4) = LENGTH V0L(1,5) = WIDTH* VOL(1,4)*0.01 DO 110 J = 2,NOFFS LENADD = 5 LENGTH = LENGTH + LENADD VOL(J,l) = VOL(J-l,l) + LENGTH -LENADD/2 VOL(J,4) = VOL(J-l,4) + LENGTH IF (YOL(J, 1).LE.SFINTR) INTI = J IF (J.EQ.INTI) THEN IF (VOL(J, 1 ).LE.890) THEN VOL(J,2) = VOL(J,1)*0.01 ELSE VOL(J,2) = 8.9 + (VOL(J,1)-890)*0.00417 ENDIF ELSE YOL(J,2) = MIXED ENDIF VOL(J,3) = WIDTH*VOL(J,2)*LENGTH IF (J.EQ.INTI) THEN IF (VOL(J,4).LE.890) THEN V OL(J,5)=WIDTH* VOL(J,4)*0.01 ELSE VOL(J,5)=WIDTH*(8.9+(VOL(J,4)-890)*0.00417) ENDIF ELSE VOL(J,5) = MIXED * WIDTH ENDIF 110 CONTINUE C Find Box that Corresponds to Data Points C FIELD(I,4)= X=YOL(J,l) for Distance, X is for calculated Conc. DO 123 1=1,5 DO 120 J=l,NOFFS IDIST=J ANS=FIELD(I,l)-YOL(J,l) IF (ANS.LE.0) GO TO 122 120 CONTINUE 122 FIELD(I,4)=IDIST 123 CONTINUE 276 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C COMBINE Geometry for each box into function GEOM for use in FUNC(X) C Box Geometry: GEOM(I,J,K): C 1= Longshore Coordinate; J= Offshore Coordinate C K= Variables: EVERYTHING IS IN METERS C GEOM(I,J,K): K = Variables: 1= longshore box distance C 2= Dist. Offshore to box midpt.; 3= Dist. to seaward end of box C 4= Southside box area; 5= Seaward box area; 6 =Volume DO 142 I=l,NLONG DO 140 J=l,NOFFS IF (I.EQ.l) THEN GEOM(I,J,l) = WIDTH/2 ELSE GEOM(I,J,l) = GEOM(I-l,J,l) + WIDTH ENDIF GEOM(I,J,2) = VOL(J,l) GEOM(I,J,3) = VOL(J,4) GEOM(I,J,4) = VOL(J,3)/WIDTH GEOM(I,J,5) = VOL(J,5) GEOM(I,J,6 ) = VOL(J,3) 140 CONTINUE 142 CONTINUE C CON(X) = Constants used in model C 1 = Box width; 2=timestep; 3=time stop; 4=adv. rate (m/s) C 5= seafloor intercept box in N. section; 6 =Boundary scale dist. factor C 7 = Seafloor flux scale dist; 8 = Shore Ra Gradient C 9=223 Seafloor Flux; 10=224 Seafloor Flux C 11= 223 Shoreline Flux Frac-N. 12= 224 Shoreline Flux Frac-N. C 13=14=0 C 15= ID to output to hbrelout.xls file on last/single time C 16=17=18=0; 19= Kh Power Law scale power; C 20= Turn On Tides; 21=SL,TM Tide Flux Factor C 22=23=24=25=0 CON(l) = WIDTH CON(2) = V(25)*3600 CON(3) = V(24)*24*3600 CON(4) = V(5)*0.01 CON(5) = INTI CON(6 ) = V(20) CON(7) = V(7) CON(8 ) = V(6 ) CON(9) = V(9)* WIDTH 277 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CON(IO) = V( 10)* WIDTH CON(ll) = V(11)*WIDTH CON(12) = V( 12)* WIDTH CON(13) = 0 CON(14) = 0 CON(15) = V(23) CON(16) = 0 CON(17) = 0 CON(18) = 0 CON(19) = V(16) CON(20) = V(15) CON(21) = 1 CON(22) = 0 CON(23) = 0 CON(24) = 0 CON(25) = 0 C Put needed data into accessible file, chisq.xls OPEN (UNIT=22, FILE='chisq.xls', FORM='FORMATTED', 1 ACCESS='append', STATUS='unknown') WRITE (UNIT=22, FMT=23) 0,0,0,0,0 23 FORMAT (5(E12.6,1X)) ENDFILE (UNIT=22) CLOSE (UNIT=22) C Variables to minimize chi-squared: C Specify Starting Point P(X:3) C l=Kh (m2/sec), 2=Fraction of Huntington Beach input used in Boundary Model C 3= left open so that model can be expanded C Fractional Tolerance FTOL to know when to finish iterations (NOT USED): FTOL = V(3) P d) = V(4) P(2) = V(8) p(3) = 1 C Define Search Vector XI(3,3) DATA xi/0.5,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.5/ np=NDIM C Call Powell to search for lowest value C (Get it- Powell discovered the Grand Canyon!) IF (CON(15).EQ.O.) THEN CALL POWELL(p,xi,NDIM,np,FTOL,iter,fret) ENDIF 278 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C After Minimum is found, Output Data again and C Run model using these values and output results WRITE (*,'(/lx,al0,2x,al0,2x,al0/)') 1 'Dist. (m)7223Ra Conc','224Ra Cone’ DO 170 1=1,10 WRITE (*,'(2x,F8.1,4x,F6.2,8x,F6.2)') 1 FIELD(I,1), FIELD(I,2), FIELD(I,3) 170 CONTINUE CON(15) = 1.0 FINAL = func(p) C Model will spit out the values and return to the top write(*,'(/lx,a,i3)') 'Iterations:',iter write(*,'(/lx,a/lx,3fl2.6)') 'Minimum found at: ’ ,(p(i),I=l,NDIM) write(*,'(/lx,a,fl2.6,E18.6)') 'Min. function value-,fret,FINAL PRINT*,'' GO TO 19 1000 CONTINUE PRINT *, 'HBRE1- The Real Radium Distribution' PRINT * ," STOP END REAL FUNCTION func(x) C But Wait- That's just the beginning. Now the Function is defined. C Output of func(x) is chi-sq. PARAMETER(NLON= 32, NOFF= 80) COMMON /VARIES/ BOX, CON, FIELD DOUBLE PRECISION DC(NLON,NOFF,l) DOUBLE PRECISION D(NLON,NOFF,2),DD(NLON,NOFF,l) DOUBLE PRECISION C(NLON,NOFF,2) DIMENSION FLUX(NLON,NOFF,2), x(3), BOX(NLON,NOFF,6) DIMENSION FLUX 1 (NLON,NOFF,2), FLUX2(NLON,NOFF,2), FLX(4) DIMENSION CON(25), FIELD(10,6), CDN(NOFF,2), DDN(NOFF,2) REAL DECAY3,DECAY4,KH, KB, SUMSQ, DELT, TSTOP REAL WIDE, ADV, BOXEX, ADTIME, TIDE, AREA(NOFF,2) REAL SCALE, KHO, VARK, KB223, KB224 INTEGER IDELT, NN, NT, NTIDE C Define constants C DELT=time step (sec); TSTOP=stop time (sec) 279 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C ADV = longshore advection rate (m/sec); WIDE= box width (m) C Decay Constants (1/sec) (DECAY3=223;DECAY4=224) DELT = CON(2) TSTOP = CON(3) ADY = CON(4) WIDE = CON(l) DECAY3 = 7.037E-7 DECAY4 = 2.192E-6 C Make the Advection Timestep Work Evenly: C ABSADV = absolute value of adv. rate, C BOXEX= box exchange period/Residence Time C IDELT = integer value for adv. time step relative to DELT ABSADV = ABS(ADV) IF (ABSADV.GT.0.0) THEN BOXEX = WIDE/ABSADV IDELT = BOXEX/DELT DELT = BOXEX/IDELT ENDIF C Check if POWELL chose negative value of X DO 30 1=1,2 IF (x(I).LT.O) THEN x(I) = -x(I) PRINT *, '== MODEL CHOSE VALUE LESS THAN ZERO ENDIF 30 CONTINUE C Diffusivity KH (m2/sec); C AREA(NOFF,2) Stores original box areas KH=x(l) IF (CON(19).GT.O) THEN C Define Parameters from Okubo (1976) C KHO = 0.0680 cmA 0.85; scale=1.15 SCALE=1.15 DO 40 J=l,NOFF AREA(J,l) = BOX(l,J,4) AREA(J,2) = BOX(l,J,5) 40 CONTINUE DO 55 I=2,NLON-l DO 50 J=l,NOFF Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VARK=(KH*(100*BOX(I,J,3))**SCALE)/10000. IF (VARK.LE.KH) THEN B0X(I,J,4) = BOX(I,J,4) B0X(I,J,5) = B0X(I,J,5) ELSE BOX(I,J,4)=BOX(I,J,4)*(0.0680* 1 (100*BOX(I,J,3))**SCALE)/(10000.*KH) B0X(I,J,5) = B0X(I,J,5)*VARK/KH ENDIF 50 CONTINUE 55 CONTINUE ENDIF C USE Subroutine BOUND to compute Boundary condition and C Extrapolate across to set initial model concentrations C Transfer Huntington Beach fluxes into FLX array, C Alter those values for use in BOUND, then replace. PRINT *,'Calculating Boundary Concentrations...' DO 60 1=9,12,1 FLX(I-8) = CON(I) CON(I) = CON(I)*X(2) 60 CONTINUE CALL FLUXUB(x,CON,FLUX) DO 651 = 9,12,1 CON(I)= FLX(I-8) 65 CONTINUE CALL BOUND(FLUX,KH,l ,BOX,CDN,DDN) DO 72 I=l,NLON DO 71 J=l,NOFF C(I,J,2) = CDN(J,2) D(I,J,2) = DDN(J,2) C(I,J,1) = C(I,J,2)*BOX(I,J,6) D(I,J,1) = D(I,J,2)*BOX(I,J,6) 71 CONTINUE 72 CONTINUE C Calculate the flux into each box in the model. CALL FLUXUB(x, CON, FLUX) PRINT *,'AVE N, S FIRST BOX 224 FLUX:',FLUX(19,1,1),FLUX(19,1,2) C Option to include tidal flux changes C IF CON(20) = 1, Double Shoreline and Marsh flux on flood tide (FLUX1) 281 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C Shoreline and Marsh flux = 0 on Ebb tide; Define tidal flux factor= CON(21) C Calculate Ebb Tide Flux (Double Shoreline and Tidal Marsh fluxes) IF (CON(20).EQ.l) THEN CON(21)=2 CALL FLUXUB (x,CON,FLUX 1) CON(21) = 0 CALL FLUXUB(x,CON,FLUX2) CON(21) = 1 ENDIF C Compute changes in each box in one time step and augment each reservoir TIME=0 NN=1 NT=0 NTIDE = 1 90 TIME = TIME + DELT C Chose between ebb and flood tide fluxes if CON(20)=1 IF (CON(20).EQ.1.0) THEN IF (TIME/11250.GE.NFLX) THEN NFLX = NFLX+1 TIDE = COS(NFLX*3.14159/2+3.14159/4) IF (TIDE.GT.0.0) THEN DO 93 I=l,NLON DO 92 J=l,NOFF DO 91 K=l,2 FLUX(I,J,K) = FLUX1(I,J,K) 91 CONTINUE 92 CONTINUE 93 CONTINUE ELSE DO 97 I=l,NLON DO 96 J=l,NOFF DO 95 K=l,2 FLUX(I,J,K) = FLUX2(I,J,K) 95 CONTINUE 96 CONTINUE 97 CONTINUE ENDIF ENDIF ENDIF C COMPUTE DIFFUSIVE CHANGE IN RADIUM ONE TIME STEP (DELT) 282 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C C(I,J,1) = 223Ra atoms * 223 Decay in each box; C C(I,J,2) = 223Ra Concentration (dpm/m3) C D(I,J,1) = 224Ra atoms * 224 Decay in each box; C D(I,J,2) = 224Ra Concentration (dpm/m3) C DC, DD = change in Ra atoms in one time step. C Compute Shoreline and Seaward boxes DO 110 I=2,NLON-l DC(I,NOFF,l)=(BOX(I,NOFF,5)*KH*((C(I,NOFF-l,2)-C(I,NOFF,2))+ 1 (0.0001 -C(I,NOFF,2)))/(BOX(I,NOFF,2)-BOX(I,NOFF-1,2))+ 1 KH*(BOX(I,NOFF,4)*(C(I+l ,NOFF,2)-C(I,NOFF,2))+ 1 BOX(I-1 ,NOFF,4)*(C(I-1 ,NOFF,2)-C(I,NOFF,2)))/WIDE- 1 DECAY 3 *C(I,NOFF, 1 ))*DELT DD(I,NOFF, 1 )=(BOX(I,NOFF,5)*KH*((D(I,NOFF-1,2)-D(I,NOFF,2))+ 1 (O.001-D(I,NOFF,2)))/(BOX(I,NOFF,2)-BOX(I,NOFF-1,2))+ 1 KFI*(BOX(I,NOFF,4)*(D(I+l,NOFF,2)-D(I,NOFF,2))+ 1 BOX(I-l,NOFF,4)*(D(I-l,NOFF,2)-D(I,NOFF,2)))/WIDE- 1 DEC AY4*D(I,NOFF, 1)) *DELT DC(I,1,1 )=(BOX(1,1,5)*KH*(C(I,2,2)-C(1,1,2))/(BOX(I,2,2)- 1 BOX(1,1,2))+KH*(BOX(1,1,4)*(C(I+1,1,2)-C(1,1,2))+ 1 BOX(I-1,1,4)*(C(I-1,1,2)-C(I,l,2)))/WIDE+ 1 FLUX(1,1,1 )-DEC A Y3 *C(I,1,1 ))*DELT DD(1,1,1 )=(BOX(1,1,5)*KH*(D(I,2,2)-D(I,1,2))/(BOX(I,2,2)- 1 BOX(1,1,2))+KH*(BOX(I,1,4)*(D(I+1,1,2)-D(I,l ,2))+ 1 BOX(I-1,1,4)*(D(I-1,1,2)-D(1,1,2)))/WIDE+ 1 FLUX(1,1,2)-DECAY4*D(I,l, 1 ))*DELT C Compute interior boxes DO 105 J=2,NOFF-l DC(I,J,l)=(BOX(I,J-l,5)*KH*(C(I,J-l,2)-C(I,J,2))/ 1 (BOX(I,J,2)-BOX(I,J-l,2))+BOX(I,J,5)*KH* 1 (C(I,J+l,2)-C(I,J,2))/(BOX(I,J+l,2)-BOX(I,J,2))+ 1 KH*(BOX(I,J,4)*(C(I+l,J,2)-C(I,J,2))+ 1 BOX(I-l,J,4)*(C(I-l,J,2)-C(I,J,2)))/WIDE+ 1 FLUX(I,J, 1 )-DECAY3*C(I,J, 1 ))*DELT DD(I,J,l)=(BOX(I,J-l,5)*KH*(D(I,J-l,2)-D(I,J,2))/ 1 (BOX(I,J,2)-BOX(I,J-l,2))+BOX(I,J,5)*KH* 1 (D(I,J+1,2)-D(I,J,2))/(BOX(I,J+l ,2)-BOX(I,J,2))+ 1 KH*(BOX(I,J,4)*(D(I+l,J,2)-D(I,J,2))+ 1 BOX(I-1 ,J,4) * (D(I-1, J,2)-D(I,J,2)))/WIDE+ 283 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 FLUX(I,J,2)-DECAY4*D(I,J,1))*DELT 105 CONTINUE 110 CONTINUE C Sum changes in inventory and calculate new concentrations. DO 115 I=2,NLON-l DO 113 J=l,NOFF C(I,J,1) = C(U,1) + DC(U,1) D(I,J,1) = D(I,J,1) + DD(I,J,1) C(I,J,2) = C(I,J, 1 )/BOX(I,J,6) D(I,J,2) = D(I, J, 1 )/BOX(I, J,6) 113 CONTINUE 115 CONTINUE C Reduce the time step if IF (C(l,l,l).LT.O .OR. C(1,1,2).GT.1E7) THEN PRINT *,'Your Time Step is TOO BIG. Let me find you a better one.' TIME = 0 DELT = 0.9*DELT DO 118 I=2,NLON-l DO 117 J=l,NOFF C(I,J,2) = C(1,J,2) D(IJ,2) = D(1,J,2) C(I,J,1) = C(1,J,1) D(I,J,1) = D(1,J,2) 117 CONTINUE 118 CONTINUE IF (ABSADV.GT.0.0) THEN BOXEX = WIDE/ABSADV IDELT = BOXEX/DELT DELT = BOXEX/IDELT ENDIF GO TO 90 ENDIF C Output so internet connection isn't lost and I know it's Working! IF (TIME.GT.(NN*TSTOP/4)) THEN NN=NN+1 PRINT *,'Time, Box 20 Shoreline Conc for 223Ra and 224Ra Are' PRINT *, TIME, C(20,l,2), D(20,l,2) ENDIF C Output to look at impact of the tides 284 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IF (TIME.GT.2600000) THEN IF (NN.GE.19.9) THEN NN=0 PRINT *, TIME, C(19,l,2), D(19,l,2) ENDIF ENDIF C Compute longshore advection in one advection time step. C Positive is downcoast/equatorward; Negative is upcoast/poleward IF (ADV.EQ.0.0) GO TO 130 NT = NT+1.0 ADTIME = NT/IDELT IF (ADTIME.GE.1.0) THEN NT = 0.0 IF (ADV.GT.0.0) THEN DO 123 I=NLON-l,2,-l DO 121 J=l,NOFF C(I,J,1) = C(I-1,J,1) C(I,J,2) = C(I-1,J,2) D(I,J,1) = D(I-1,J,1) D(I,J,2) = D(I-1,J,2) 121 CONTINUE 123 CONTINUE ELSE DO 127 I=2,NLON-l DO 125 J=l,NOFF C(I,J,1) = C(I+1,J,1) C(I,J,2) = C(I+1,J,2) D(I,J,1) = D(I+1,J,1) D(I,J,2) = D(I+1,J,2) 125 CONTINUE 127 CONTINUE ENDIF ENDIF C TIDES: Move boxes up 2 boxes and down 2 boxes every 12.5 hours 130 CONTINUE IF (CON(20).EQ.O) GO TO 140 IF (TIME/11250.GE.NTIDE) THEN NTIDE = NTIDE+1 TIDE = CO S (NTIDE *3.14159/2+3.14159/4) IF (TIDE.GT.0.0) THEN DO 133 I=NLON-l,2,-l Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DO 131 J=l,NOFF C(I,J,1) = C(I-1,J,1) C(I,J,2) = C(I-1 ,J ,2) D(I,J,1) = D(I-1,J,1) D(I,J,2) = D(I-1,J,2) 131 CONTINUE 133 CONTINUE ELSE DO 137 I=2,NLON-l DO 135 J=l,NOFF C(I,J,1) = C(I+1,J,1) C(I,J,2) = C(I+1,J,2) D(I,J,1) = D(I+1,J,1) D(I,J,2) = D(I+1,J,2) 135 CONTINUE 137 CONTINUE ENDIF ENDIF C Loop Or Move Forward 140 CONTINUE IF (TIME.LT.TSTOP) GO TO 90 C Calculate Reduced Chi-Squared for the Fluntington Beach profile, SUMSQ C Data point is between midpoint of box IA and IA-1. C COMPUTE HUNTINGTON BEACH PROFILE SUM-SQUARES C Contribution for 223Ra SUMSQ=0 DO 160 1=1,5 IA = FIELD(I,4) C Extrapolates concentration to the shoreline IF (IA.EQ.l) THEN SUMSQ=SUMSQ+((FIELD(I,2)- 1 (C(19,1,2)+0.5*(C(20,1,2)-C(20,2,2))))/FIELD(I,5))**2 ELSE C Linear extrapolate conc. between two boxes SUMSQ=SUMSQ+((FIELD(I,2)-(C(20,IA-1,2)- 1 (C(20,IA-1,2)-C(20,IA,2))*(FIELD(I,1)-BOX(20,IA-1,2))/ 1 (BOX(20,IA,2)-BOX(20,IA-1,2))))/FIELD(I,5))**2 ENDIF 160 CONTINUE C Contribution for 224Ra 286 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DO 165 1=1,5 IA = FIELD(I,4) C Extrapolates concentration to the shorelines IF (IA.EQ.l) THEN SUMSQ=SUMSQ+((FIELD(I,3)- 1 (D(20,1,2)+0.5*(D(20,1,2)-D(20,2,2))))/FIELD(I,6))**2 ELSE SUMSQ=SUMSQ+((FIELD(I,3)-(D(20,IA-1,2)- 1 (D(20,IA-1,2)-D(20,IA,2))*(FIELD(I,1 )-BOX(20,IA-1,2))/ 1 (BOX(20,IA,2)-BOX(20,IA-1,2))))/FIELD(I,6))**2 ENDIF 165 CONTINUE C Compute the Chi-Sq Value. func=SUMSQ+(x(3)-1)*(x(3)-l) PRINT *,'Kh, 223Flux, 224Flux, and FUNC Are:1 PRINT *,X(l),X(2),X(3),func C Put needed data into accessible file, chisq.xls OPEN (UNIT=22, FILE='chisq.xls', FORM='FORMATTED', 1 ACCESS='append', STATUS='old') WRITE (UNIT=22, FMT=23) X(1),X(2),X(3),FUNC,SUMSQ 23 FORMAT (5(E12.6,1X)) ENDFILE (UNIT=22) CLOSE (UNIT=22) C FOR THE FINAL OR SINGLE RUN, output distances and conc. C of each box to file hbrel out.xls IF (CON(15).EQ.1.0) THEN OPEN (UNIT=8, FILE-hbrelout.xls1 , FORM-'FORMATTED1 , 1 ACCESS-SEQUENTIAL', STATUS = 'unknown') WRITE (UNIT=8, FMT-9) X(1),X(2),X(3),FUNC 9 FORMAT (FI 1.4,2x,Fl 1.4,2x,Fl 1,4,2x,Fl 1.4) DO 185 I=l,NLON DO 180 J=l,NOFF WRITE (UNIT=8, FMT=9) BOX(I,J,l),BOX(I,J,2),C(I,J,2),D(I,J,2) 180 CONTINUE 185 CONTINUE ENDFILE (UNIT=8) CLOSE (UNIT=8) ENDIF 287 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C IF Cross-sectional Areas were changed, return to original values IF (CON(19).GT.O) THEN DO 205 I=2,NLON-l DO 200 J=l,NOFF BOX(I, J,4)=AREA(J, 1) BOX(I,J,5)=AREA(J,2) 200 CONTINUE 205 CONTINUE ENDIF RETURN END SUBROUTINE FLUXUB(FLXT,CON,FLX) C CALCULATE THE FLUX INTO EACH BOX C FLXT = Total fluxes (dpm/sec/Box) 1=223, 2=224 C Flux is divided into two components: C SFLUX= Shoreline Flux; BFLUX= Benthic/ Seafloor Flux C CON= Frac. of each component: C 7 = Seafloor flux scale dist; 8= Shore Ra Gradient; C 9=223 Seafloor Flux; 10=224 Seafloor Flux C 11= 223 Shoreline Flux Frac-N. 12= 224 Shoreline Flux Frac-N. PARAMETER(NL=3 2, NO=80) REAL FLXT(3),CON(25),FLX(NL,NO,2) INTEGER I, J, INI REAL FLXSF,GRADSF,FLXSL,GRADSL,BFLUX,ALFA C Reset Fluxes DO 102 1=1,NL DO 101 J=l,NO FLX(I,J,1) = 0 FLX(I,J,2) = 0 101 CONTINUE 102 CONTINUE C DEFINE CONSTANTS: INI = N. Seafloor Intercept; C INIS= S. Seafloor Intercept; ALFA =Seafloor Scale INI = CON(5) ALFA = CON(7) C Seafloor flux decreases exponentially offshore. C This is NOT DONE IN TERMS OF DISTANCE! C Instead, exponential decrease is in terms of box number C FLXSF=total seafloor flux; GRADSF= seafloor flux gradient 288 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C FLXSL= shoreline flux; GRADSL= shoreline flux gradient C BFLUX = Flux at shoreline (dpm*Box/m) FLXSF = CON(9)*CON(8) GRADSF = CON(9)*(1-CON(8))/(0.5*NL) FLXSL = CON(l l)*CON(8)*CON(21) GRADSL = CON(11)*CON(21)*(1-CON(8))/(0.5*NL) DO 105 1= 1,NL BFLUX = FLXSF* ALFA/(l-exp(-ALFA*INI)) FLX(I,1,1) = BFLUX*(1-EXP(-ALFA))/ALFA + FLXSL DO 104 J=2,INI FLX(I,J,1) = BFLUX*(EXP(-ALFA*(J- 1))-EXP(-ALFA* J))/ ALFA 104 CONTINUE FLXSL = FLXSL+GRADSL FLXSF = FLXSF+GRADSF 105 CONTINUE C REPEAT TO FIND 224Ra fluxes. FLXSF = CON(10)*CON(8) GRADSF = CON(10)*(1-CON(8))/(0.5*NL) FLXSL = CON(12)*CON(8)*CON(21) GRADSL = CON(12)*CON(21)*(1-CON(8))/(0.5*NL) DO 115 1=1, NL BFLUX = FLXSF*ALFA/(l-exp(-ALFA*INI)) FLX(I,1,2) = BFLUX*(1 -EXP(-ALFA))/ALFA + FLXSL DO 114 J=2,INI FLX(I,J,2) = BFLUX* (EXP(-ALFA* (J-1 ))-EXP(-ALFA* J))/ALFA 114 CONTINUE FLXSL = FLXSL+GRADSL FLXSF = FLXSF+GRADSF 115 CONTINUE RETURN END SUBROUTINE BOUND(FLUXL,KH,IEND,VL,CLON,DLON) C Calculate the concentration at the longshore limits of the model. C Modified from the RADRE model. C Box geometry and Flux distribution is calculated in the main program. C Model determines equilibrium after 60 days PARAMETER (NLON=32, NOFF=80) REAL FLUXL(NLON,NOFF,2),CLON(NOFF,2),DLON(NOFF,2) REAL DCLON(NOFF,l), DDLON(NOFF,l), VL(NLON,NOFF,6) 289 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REAL DECAY3, DECAY4, TIME, TSTOP, DT, KH INTEGER IEND, J, K C Decay Constants (1/sec) (DECAY3=223;DECAY4=224) DECAY3 = 7.037E-7 DECAY4 = 2.192E-6 C Input initial model concentration DO 11 I=l,NOFF CLON(I,l) = 0 CLON(I,2) = 0 DLON(I,l) = 0 DLON(I,2) = 0 11 CONTINUE C Compute changes in each box in one time step and augment each reservoir TIME=0 TSTOP = 60*1440*60 DT = 0.045*3600 100 TIME = TIME + DT C COMPUTE DIFFUSIVE CHANGE IN RADIUM ONE TIME STEP (DT) DCLON(NOFF,l)=(VL(IEND,NOFF,5)*(KH*(CLON(NOFF-l,2)- 1 CLON (NOFF,2))+KH* (0.0001 -CLON(NOFF,2)))/ 1 (VL(IEND,NOFF,2)-VL(IEND,NOFF-1,2))- 1 DECAY 3 *CLON (NOFF, 1 ))*DT DDLON(NOFF, 1 )=(VL(IEND,NOFF,5)*(KH*(DLON(NOFF-1,2)- 1 DLON(NOFF,2))+KH*(0.001-DLON(NOFF,2)))/ 1 (VL(IEND,NOFF,2)-VL(IEND,NOFF-1,2))- 1 DEC A Y4* DLON (NOFF, 1)) * DT DCLON( 1,1 )=(VL(IEND, 1,5)*KH*(CLON(2,2)-CLON( 1,2))/ 1 (VL(IEND,2,2)-VL(IEND,1,2))+ 1 FLUXL(IEND, 1,1 )-DECAY3*CLON( 1,1 ))*DT DDLON( 1,1 )=(VL(IEND, 1,5)*KH*(DLON(2,2)-DLON( 1,2))/ 1 (VL(IEND,2,2)-VL(IEND,1,2))+ 1 FLUXL(IEND, 1,2)-DEC A Y4*DLON( 1,1 ))*DT DO 110 J=2,NOFF-l DCLON(J,l)=(VL(IEND,J-1,5)*KH*(CLON(J-1,2)- 1 CLON (J,2))/( VL(IEND, J,2)-VL(IEND, J -1,2))+ 290 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 VL(IEND,J,5)*KH*(CLON(J+l,2)-CLON(J,2))/ 1 (VL(IEND,J+1,2)-VL(IEND,J,2))+ 1 FLUXL(IEND,J, 1 )-DECAY3*CL0N(J, 1 ))*DT DDLON(J,l)=(VL(IEND,J-l,5)*KH*(DLON(J-l,2)- 1 DLON(J,2))/(VL(IEND,J,2)-VL(IEND,J-l,2))+ 1 VL(IEND,J,5)*KH*(DLON(J+l,2)-DLON(J,2))/ 1 (VL(IEND,J+1,2)-VL(IEND,J,2))+ 1 FLUXL(IEND,J,2)-DECAY4*DLON(J,l))*DT 110 CONTINUE C Sum changes in inventory and calculate new concentrations DO 130 K=l, NOFF CLON(K,l) = CLON(K,l) + DCLON(K,l) DLON(K,l) = DLON(K,l) + DDLON(K,l) CLON(K,2) = CLON(K, 1 )/VL(IEND,K,6) DLON(K,2) = DLON(K, 1 )/VL(IEND,K,6) 130 CONTINUE IF (CLON(1,1).LT.O .OR. CL0N(1,2).GT.1E7) THEN PRINT *,'Your Time Step is TOO BIG. Let me find you a better one.' TIME = 0 DT = 0.9*DT DO 140 J=l,NOFF CLON(J,l) = 0 CLON(J,2) = 0 DLON(J,l) = 0 DLON(J,2) = 0 140 CONTINUE ENDIF IF (TIME.LT.TSTOP) GO TO 100 RETURN END SUBROUTINE powell(p,xi,n,np,ftol,iter,fret) INTEGER iter,n,np,NMAX,ITMAX REAL fret,ftol,p(np),xi(np,np),func EXTERNAL func PARAMETER (NMAX=20, ITMAX=200) C USES func, linmin C Minimization of a function func of n variables, (func is not an C argument, it is a fixed function name.) Input consists of an initial C starting point p(l:n); an initial matrix xi(l:n,l:n) with physical C dimensions np by np, and whose columns contain the initial set of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C directions (usually the n unit vectors); and ftol, the fractional C tolerance in the function value such that failure to decrease by more C than this amount on one iteration signals doneness. On output, p is set C to the best point fount, xi is the then-current direction set, fret is the C returned function value at p, and iter is the number of iterations taken. C Parameters: Maximum expected value of n, and maximum allowed iterations INTEGER i,ibig,j REAL del,fp,fptt,t,pt(NMAX),ptt(NMAX),xit(NMAX) fret=func(p) DO 10 j=l,n pt(j)=pG) 10 CONTINUE iter=0 11 iter=iter+l fp=fret ibig=0 del=0. DO 13 i=l,n DO 12 j=l,n xit(j)=xi(j,i) 12 CONTINUE fptt=fret call linmin(p,xit,n,fret) if (abs(fptt-fret).gt.del) then del=abs(fptt-fret) ibig=i endif 13 CONTINUE if (2.*abs(fp-fret).le.ftol*(abs(fp)+abs(fret))) return if (iter.eq.ITMAX) pause 'powell exceeding maximum iterations' do 14j=l,n pttG)=2.*p(j)-ptG) xit(j)=p(j)-pt(j) pt(i)=p(i) 14 CONTINUE fptt=func(ptt) if (fptt.ge.fp) goto 11 t=2. * (fp-2. * fret+fptt) * (fp-fret-del) * * 2-del * (fp-fptt) * * 2 if (t.ge.O.) goto 11 call linmin(p,xit,n,fret) DO 15 j=l,n xi(j,ibig)=xi(j,n) 292 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. xi(j,n)=xit(j) 15 CONTINUE GOTO 11 END SUBROUTINE linmin(p,xi,n,fret) INTEGER n, NMAX REAL fret,p(n),xi(n),TOL PARAMETER (NMAX=50, TOL=l.e-4) C USES brent, fldim,mnbrak C Given an n-dimensional point p(l:n) and an n-dimensional direction C xi(l:n), moves and resets p to where the function func(p) takes on a C minimum along the direction xi from p and replaces xi by the actual C vector displacement that p was moved. Also returns as fret the value C of func at the returned location p. This is actually accomplished by C calling the routines mnbrak and brent INTEGER j, ncom REAL ax,bx,fa,fb,fx,xmin,xx,pcom(NMAX),xicom(NMAX),brent COMMON /flcom/ pcom,xicom,ncom EXTERNAL fldim ncom=n DO 11 j=l,n pcom (j)=p(j) xicom(j)=xi(j) 11 CONTINUE ax=0 x x -1 call mnbrak (ax,xx,bx,fa,fx,fb,fldim) fret=brent(ax,xx,bx,fldim,TOL,xmin) DO 12 j=l,n xi(j)=xmin*xi(j) pG)=p(j)+xiG) 12 CONTINUE RETURN END FUNCTION fldim(x) INTEGER NMAX REAL fldim,func,x PARAMETER (NMAX=50) C USES func C Used by linmin as the function passed to mnbrak and brent. INTEGER j,ncom Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REAL pcom(NMAX),xicora(NMAX),xt(NMAX) COMMON /flcom/ pcom,xicom,ncom DO 11 j=l,ncom xt(j)=pcom(j)+x*xicom(j) 11 CONTINUE fldim=func(xt) RETURN END SUBROUTINE mnbrak(ax,bx,cx,fa,fb,fc,func) REAL ax,bx,cx,fa,fb,fc,func,GOLD,GLIMIT,TINY EXTERNAL func PARAMETER (GOLD=1.618034, GLIMIT=100., TINY=l.e-20) C Given a function FUNC, and given distinct initial points ax and bx, C this routine searches in the downhill direction (defined by the function as C evaluated at the initial points) and returns new points ax, bx, cx that bracket C a minimum of the function. Also returned are the function values at the C three points fa, fb, fc. Parameters: GOLD is the default ratio by which C successive intervals are magnified; C GLIMIT is the maximum magnification allowed for a parabolic-fit step. REAL dum,fu, q,r,u,ulim fa=func(ax) fb=func(bx) if (fb.gt.fa) then dum-ax ax=bx bx=dum dum=fb fb=fa fa=dum endif cx=bx+GOLD*(bx-ax) fc=func(cx) 1 if (fb.ge.fc) then r=(bx-ax)*(fb-fc) q=(bx-cx)*(fb-fa) u=bx-((bx-cx)*q-(bx-ax)*r)/(2.*sign(max(abs(q-r),TINY),q-r)) ulim=bx+GLIMIT*(cx-bx) if ((bx-u)*(u-cx).gt.O.) then fu=func(u) if (fu.lt.fc) then ax-bx fa=fb 294 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bx-u fb=fu return else if (fu.gt.fb) then cx=u fc=fu return endif u=cx+GOLD * (cx-bx) fu=func(u) else if ((cx-u)*(u-ulim).gt.O.) then fu=func(u) if (fu.lt.fc) then bx=cx cx=u u=cx+GOLD* (cx-bx) fb=fc fc=fu fu=func(u) endif else if ((u-ulim)*(ulim-cx).ge.O.) then u=ulim fu=func(u) else u=cx+GOLD*(cx-bx) fu=func(u) endif ax=bx bx=cx cx=u fa=fb fb=fc fc=fu goto 1 endif return END FUNCTION brent(ax,bx,cx,f,tol,xmin) INTEGER ITMAX real brent,ax,bx,cx,tol,xmin,f,CGOLD,ZEPS EXTERNAL f PARAMETER (ITMAX=100, CGOLD=.3819660, ZEPS=1.0e-10) 295 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C Given a function f, and given a bracketing triplet of abscissas ax, bx, cx C(such that bx is between ax and cx, and f(bx) is less than both f(ax) and f(cx)), C this routine isolates the minimum to a fractional precision of about tol using C Brent's method. The abscissa of the minimum is returned as xmin, and the C minimum function is returned as brent, the returned function value C Parameters: Maximum allowed number of iterations; golden ratio; C and a small number that protects against trying to achieve fractional C accuracy for a minimum that happens to be exactly zero INTEGER iter REAL a,b,d,e,etemp,fu,fv,fw,fx,p,q,r,toll,tol2,u,v,w,x,xm a=min(ax,cx) b=max(ax,cx) v=(bx) w=v x=v e=0. fx=f(x) fv=fx fw=fx DO 11 ITER=1,ITMAX xm=0.5*(a+b) tol 1 =tol*abs(x)+ZEPS tol2=2*toll if (abs(x-xm).le.(tol2-.5*(b-a))) goto 13 if (abs(e).gt.toll) then r=(x-w)*(fx-fv) q=(x-v)*(fx-fw) p=(x-v)*q-(x-w)*r q=2.*(q-r) if (q.gt.O.) p=-p q=abs(q) etemp=e e=d if(abs(p).ge.abs(.5*q*etemp) .or. p.le.q*(a-x).or. 1 p.ge.q*(b-x)) goto 1 d=p/q u=x+d if (u-a.lt.tol2 .or. b-u.lt.tol2) d=sign(toll,xm-x) goto 2 endif 1 if (x.ge.xm) then e=a-x else 296 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. e=b-x endif d=CGOLD*e 2 if (abs(d).ge.toll) then u=x+d else u=x+sign(toll,d) endif fu=f(u) if (fu.le.fx) then if(u.ge.x) then a=x else b=x endif v=w fv=fw w=x fw=fx x=u fx=fu else if (u.lt.x) then a=u else b=u endif if (fu.le.fw .or. w.eq.x) then v=w fv=fw w=u fw=fu else if(fu.le.fv .or. v.eq.x .or.v.eq.w) then v=u fv=fu endif endif 11 CONTINUE pause 'brent exceed maximum iterations' 13 xmin=x brent=fx return END 297 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix C. Data Tables Table C-l. Volume of water pumped at each station. tatlon Date Time Sample Depth (m) Water Depth (m) Volume Pumped (L) 1.1 9 /11/01 11:20 1.5 10 134.6 1.2 9 /11/01 11:20 7.6 10 117.3 1.3 9 /11/01 11:20 1.5 10 38.0 2.1 9/11/01 13:55 19.5 22 128.3 2.2 9 /11/01 13:55 9.6 22 116.0 2.3 9/11/01 13:55 1.5 22 118 3.1 9/11/01 17:45 14.0 27 115.3 3.2 9/11/01 17:45 6.9 27 0.0 3.3 9/11/01 17:45 1.5 27 143.1 4.1 9/13/01 9:09 19.7 82 118.4 4.3 9/13/01 9:16 1.5 82 178.0 5.1 9/13/01 12:55 18.0 212 226.4 5.2 9/13/01 12:49 8.8 212 119.4 5.3 9 /13/01 12:49 1.5 212 133.7 6.1 9/13/01 14:50 12.7 24 59.7 6.2 9/13/01 14:50 6.3 24 123.0 6.3 9/13/01 14:50 1.5 24 117.6 7.1 9/13/01 17:00 11.0 13 1 1 0 . 0 7.2 9/13/01 16:53 1.5 13 1 1 0 . 0 7.3 9/13/01 16:53 1.5 13 1 1 0 . 0 8.1 9/20/01 8:44 19.5 27 158.9 8.2 9/20/01 8:44 9.6 27 179.0 8.3 9/20/01 8:44 1.5 27 192.0 9.1 9/20/01 10:51 18.6 20 97.9 9.3 9/20/01 10:51 1.5 20 93.6 10.1 9/20/01 12:55 8.6 9 74.7 10.2 9/20/01 12:55 1.5 9 153.9 11.1 11/1/01 11:35 16.3 186.5 11.2 11/1/01 11:35 9.3 153.7 11.3 11/1/01 11:35 1.4 228.3 12.1 11/1/01 13:40 15.1 158.3 12.2 11/1/01 13:40 8.6 178.7 12.3 11/1/01 13:40 1.3 180.9 13.1 11/1/01 15:40 1.8 124.9 13.2 11/1/01 15:40 9.1 143.2 13.3 11/1/01 15:40 1.8 188.4 298 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). ttatlon Data Tima Sampia Dapth (m) Watar Dapth (m) Voiuma Pumpad < 21.1 4 /3 /0 2 9:45 9.1 10 118.7 21.3 4 /3 /0 2 9:45 1.5 10 99.0 21.4 4 /3 /0 2 9:45 1.5 10 85.9 22.1 4 /3 /0 2 12:15 19.7 19.8 136.1 22.2 4 /3 /0 2 12:15 12.0 19.8 155.6 22.3 4 /3 /0 2 12:15 6.0 19.8 161.6 22.4 4 /3 /0 2 12:15 1.5 19.8 87.4 23.1 4 /3 /0 2 14:15 19.3 31 121.3 23.2 4 /3 /0 2 14:15 11.0 31 163.9 23.4 4 /3 /0 2 14:15 1.5 31 191.2 24.1 4 /3 /0 2 15:45 13.7 95 195.7 24.2 4 /3 /0 2 1 5:45 7.8 95 225.6 24.3 4 /3 /0 2 15:45 3.9 95 153.3 24.4 4 /3 /0 2 15:45 1.5 95 108.6 25.1 6 /5 /0 2 10:30 21.0 22.9 145.2 25.2 6 /5 /0 2 10:30 12.0 22.9 172.2 25.3 6 /5 /0 2 10:30 1.5 22.9 121.5 26.1 6 /5 /0 2 13:45 11.9 12.8 5.6 26.2 6 /5 /0 2 13:45 1.5 12.8 126.1 27.1 6 /5 /0 2 15:55 6.1 8.2 72.7 27.2 6 /5 /0 2 15:55 1.5 8.2 31.5 28.1 6 /7 /0 2 10:05 1.5 9.1 43.0 28.2 6 /7 /0 2 10:05 6.1 9.1 125.1 28.3 6 /7 /0 2 10:05 1.5 9.1 86.9 29.1 6 /7 /0 2 12:30 12.1 19.8 44.3 29.2 6 /7 /0 2 12:30 16.7 19.8 136.3 29.3 6 /7 /0 2 12:30 1.5 19.8 111.1 30.2 6 /7 /0 2 14:20 17.5 33.5 117.7 30.3 6 /7 /0 2 14:20 10.0 33.5 65.4 30.4 6 /7 /0 2 14:20 1.5 33.5 113.6 31.2 6 /7 /0 2 16:50 15.1 185.9 123.7 31.3 6 /7 /0 2 16:50 8.6 185.9 0.0 31.4 6 /7 /0 2 16:50 1.5 185.9 131.5 32.1 6 /1 2 /0 2 10:00 1.5 8.2 92.1 32.2 6 /1 2 /0 2 10:00 7.6 8.2 76.0 32.3 6 /1 2 /0 2 10:00 4.6 8.2 91.6 33.1 6 /1 2 /0 2 12:00 1.5 19.8 170.6 33.2 6 /1 2 /0 2 12:00 12.1 19.8 93.3 33.3 6 /1 2 /0 2 12:00 16.7 19.8 99.0 299 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Station Date Time Sample Depth (m) Water Depth (m) Volume Pumped ( 34.1 6 /1 2 /0 2 14:20 1.5 30.5 219.2 34.2 6 /1 2 /0 2 14:20 23.6 30.5 115.1 34.3 6 /1 2 /0 2 14:20 20.6 30.5 103.2 35.1 6 /1 2 /0 2 16:45 1.5 85.3 104.4 35.2 6 /1 2 /0 2 16:45 15.7 85.3 86.3 35.3 6 /1 2 /0 2 16:45 13.7 85.3 99.1 36.1 6 /2 1 /0 2 10:00 1.5 76.8 146.1 36.2 6 /2 1 /0 2 10:00 23.6 76.8 179.9 36.3 6 /2 1 /0 2 10:00 20.6 76.8 134.4 37.1 6 /2 1 /0 2 12:00 1.4 30.5 140.5 37.2 6 /2 1 /0 2 12:00 22.9 30.5 142.2 37.3 6 /2 1 /0 2 12:00 20.0 30.5 131.7 38.1 6 /2 1 /0 2 13:40 1.5 9.8 188.8 38.2 6 /2 1 /0 2 13:40 4.6 9.8 117.8 38.3 6 /2 1 /0 2 13:40 7.6 9.8 123.8 39.1 6 /2 1 /0 2 16:10 1.5 85.3 158.8 39.2 6 /2 1 /0 2 16:10 18.3 85.3 123.5 39.3 6 /2 1 /0 2 16:10 15.1 85.3 106.7 40.1 6 /2 2 /0 2 9:40 1.5 10.7 158.1 40.2 6 /2 2 /0 2 9:40 6.1 10.7 136.7 40.3 6 /2 2 /0 2 9:40 9.1 10.7 66.1 41.1 6 /2 2 /0 2 11:50 1.5 21.9 133.9 41.2 6 /2 2 /0 2 11:50 10.6 21.9 133.4 41.3 6 /2 2 /0 2 11:50 19.7 21.9 91.7 42.1 6 /2 2 /0 2 14:10 1.5 53.6 126.2 42.2 6 /2 2 /0 2 14:10 15.7 53.6 161.3 42.3 6 /2 2 /0 2 14:10 12.7 53.6 116.3 43.2 7 /2 6 /0 2 11:20 1.5 10.1 116.4 43.3 7 /2 6 /0 2 11:20 7.6 10.1 99.9 44.2 7 /2 6 /0 2 13:20 1.5 10.4 100.4 44.3 7 /2 6 /0 2 13:20 7.6 10.4 68.1 45.2 7 /2 6 /0 2 14:10 1.5 16.8 113.4 46.2 7 /2 6 /0 2 15:30 1.5 16.8 60.2 46.3 7 /2 6 /0 2 1 5:30 4.6 7.9 126.2 300 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Station Date Time Sample Depth (m) Water Depth (m) Volume Pumped ( 47.1 8 /1 9 /0 2 9:35 1.5 115.8 138.9 47.2 8 /1 9 /0 2 9:35 24.5 115.8 86.3 47.3 8 /1 9 /0 2 9:35 15.2 115.8 5 . 0 48.1 8 /1 9 /0 2 11:40 1.5 30.5 122.8 48.2 8 /1 9 /0 2 11:40 30.5 30.5 125.5 48.3 8 /1 9 /0 2 11:40 19.5 30.5 127.3 49.1 8 /1 9 /0 2 13:20 1.5 19.8 201.6 49.2 8 /1 9 /0 2 13:20 12.1 19.8 160.1 49.3 8 /1 9 /0 2 13:20 19.7 19.8 80.7 50.1 8 /1 9 /0 2 14:30 1.5 9.8 92.7 50.2 8 /1 9 /0 2 14:30 7.6 9.8 165.8 51.1 8 /2 3 /0 2 9:45 1.5 71.6 173.7 51.2 8 /2 3 /0 2 9:45 29.0 71.6 146.9 51.3 8 /2 3 /0 2 9:45 18.0 71.6 128.9 52.1 8 /2 3 /0 2 11:20 1.5 30.5 127.6 52.2 8 /2 3 /0 2 11:20 30.5 30.5 197.2 52.3 8 /2 3 /0 2 11:20 19.1 30.5 253.2 53.1 8 /2 3 /0 2 13:00 1.5 20.1 157.1 53.2 8 /2 3 /0 2 13:00 10.5 20.1 99.6 53.3 8 /2 3 /0 2 13:00 18.0 20.1 235.5 54.1 8 /2 3 /0 2 14:10 1.5 9.8 100.0 54.2 8 /2 3 /0 2 14:10 9.1 9.8 89.1 55.1 9 /8 /0 2 9:55 1.5 76.8 173.6 55.2 9 /8 /0 2 9:55 30.1 76.8 121.3 55.3 9 /8 /0 2 9:55 18.6 76.8 143.1 56.1 9 /8 /0 2 11:40 1.5 30.5 92.2 56.2 9 /8 /0 2 11:40 28.5 30.5 109.4 56.3 9 /8 /0 2 11:40 16.5 30.5 149.0 57.1 9 /8 /0 2 12:35 1.5 19.8 131.4 57.2 9 /8 /0 2 12:35 10.5 19.8 151.6 57.3 9 /8 /0 2 12:35 19.5 19.8 150.0 58.1 9 /8 /0 2 14:40 1.5 10.4 76.6 58.2 9 /8 /0 2 14:40 7.6 10.4 76.3 60 6 /1 8 /0 3 8:30 1.5 6.0 30.9 61.1 6 /1 8 /0 3 10:20 1.5 9.0 170.8 61.2 6 /1 8 /0 3 10:20 1.5 9.0 126.9 61.3 6 /1 8 /0 3 10:20 9.0 9.0 86.4 61.4 6 /1 8 /0 3 10:40 1.5 9.0 319.7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Sample Water Volume Station Pate Time Depth (m) Depth (m) Pumped (L) 62.1 6 /1 8 /0 3 12:30 1.5 17.2 145.6 62.2 6 /1 8 /0 3 12:30 12.1 17.2 158.8 62.3 6 /1 8 /0 3 12:30 15.2 17.2 133.1 63.1 6 /1 8 /0 3 14:35 1.5 27.4 245.5 63.2 6 /1 8 /0 3 14:35 12.0 27.4 177.9 63.3 6 /1 8 /0 3 14:35 19.5 27.4 204.4 64.1 6 /2 5 /0 3 9:30 1.5 10.0 100.3 64.2 6 /2 5 /0 3 9:30 9.1 10.0 108.1 64.3 6 /2 5 /0 3 9:30 1.5 10.0 142.7 65.1 6 /2 5 /0 3 11:10 1.5 19.0 73.8 65.2 6 /2 5 /0 3 11:10 7.6 19.0 114.2 65.3 6 /2 5 /0 3 11:10 16.7 19.0 198.7 66.1 6 /2 5 /0 3 13:10 1.5 28.0 151.5 66.2 6 /2 5 /0 3 13:10 7.4 28.0 116.9 66.3 6 /2 5 /0 3 13:10 17.7 28.0 144.8 lorellne Samples 101.1 9 /25/01 8:35 106.9 101.2 9/25/01 10:40 128.2 101.3 9/25/01 12:35 106.7 101.4 9/25/01 14:35 107.7 101.5 9/25/01 16:35 121.5 101.6 9/25/01 18:35 104.9 102.1 10/16/01 9:12 89.1 102.2 10/16/01 11:10 104.7 102.3 10/16/01 13:15 114.2 102.4 10/16/01 15:15 103.9 102.5 10/16/01 17:20 83.7 103.1 1 /9 /0 2 7:20 129.7 103.2 1/9 /0 2 9:15 118.1 103.3 1 /9 /0 2 11:15 122.9 103.4 1/9 /0 2 13:15 124.9 103.5 1 /9 /0 2 15:15 122.3 201.1 6 /5 /0 2 8:40 124.5 201.2 6 /5 /0 2 10:45 129.5 201.3 6 /5 /0 2 12:40 127.4 201.4 6 /5 /0 2 14:35 116.8 201.5 6 /5 /0 2 16:45 62.7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Sample Water Volume Station Date Time Depth (m) Depth (m) Pumped (L) 202.1 6 /7 /0 2 9:00 128.2 202.2 6 /7 /0 2 10:45 126.8 202.3 6 /7 /0 2 12:40 130.9 202.4 6 /7 /0 2 14:35 120.2 202.5 6 /7 /0 2 16:35 130.2 203.1 6 /1 2 /0 2 9:25 127.6 203.2 6 /1 2 /0 2 11:05 118.3 203.3 6 /1 2 /0 2 13:05 120.1 203.4 6 /1 2 /0 2 15:05 118.2 203.5 6 /1 2 /0 2 17:00 113.3 204.1 6 /2 1 /0 2 7:45 109.7 204.2 6 /2 1 /0 2 9:50 128.6 204.3 6 /2 1 /0 2 11:50 125.1 204.4 6 /2 1 /0 2 13:50 118.8 204.5 6 /2 1 /0 2 15:45 115.0 205.1 6 /2 2 /0 2 8:25 117.9 205.2 6 /2 2 /0 2 10:25 124.4 205.3 6 /2 2 /0 2 12:10 125.7 205.4 6 /2 2 /0 2 14:10 123.1 205.5 6 /2 2 /0 2 15:55 1 1 7 . 0 301.1 7 /2 6 /0 2 8:00 93.5 301.2 7 /2 6 /0 2 9:50 67.4 301.3 7 /2 6 /0 2 11:50 132.1 301.4 7 /2 6 /0 2 13:50 56.0 301.5 7 /2 6 /0 2 15:50 74.6 302.1 7 /2 6 /0 2 7:50 75.5 302.2 7 /2 6 /0 2 9:50 78.6 302.3 7 /2 6 /0 2 11:55 75.8 302.4 7 /2 6 /0 2 14:05 37.7 302.5 7 /2 6 /0 2 15:55 91.8 303.1 7 /2 6 /0 2 8:30 95.8 303.2 7 /2 6 /0 2 10:20 102.6 303.3 7 /2 6 /0 2 12:15 90.1 303.4 7 /2 6 /0 2 14:10 71.5 303.5 7 /2 6 /0 2 15:55 74.2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Sample Water Volume Station Date Time Depth (m) Depth (m) Pumped (L) 401.1 8 /1 9 /0 2 9:30 117.3 401.2 8 /1 9 /0 2 11:20 108.3 401.3 8 /1 9 /0 2 13:10 97.7 401.4 8 /1 9 /0 2 15:00 100.3 401.5 8 /1 9 /0 2 17:30 106.7 402.1 8 /2 3 /0 2 8:00 1 1 0 . 0 402.2 8 /2 3 /0 2 11:50 9 0 . 5 402.3 8 /2 3 /0 2 14:00 112.8 402.4 8 /2 3 /0 2 15:55 113.6 402.5 8 /2 3 /0 2 17:50 110.5 403.1 9 /6 /0 2 7:35 75.6 403.2 9 /6 /0 2 9:30 76.4 403.3 9 /6 /0 2 11:30 75.3 403.4 9 /6 /0 2 13:30 74.3 403.5 9 /6 /0 2 15:30 73.5 404.1 9 /8 /0 2 7:30 100.3 404.2 9 /8 /0 2 9:30 107.1 404.3 9 /8 /0 2 11:20 104.1 404.4 9 /8 /0 2 13:05 92.0 404.5 9 /8 /0 2 15:05 90.8 501.1 6 /1 8 /0 3 10:15 0.0 501.2 6 /1 8 /0 3 12:00 19.7 501.3 6 /1 8 /0 3 14:05 31.4 501.4 6 /1 8 /0 3 16:00 19.3 502.1 6 /2 5 /0 3 10:30 69.3 502.2 6 /2 5 /0 3 12:30 71.2 502.3 6 /2 5 /0 3 14:15 64.7 502.4 6 /2 5 /0 3 16:15 53.8 502.5 6 /2 5 /0 3 17:35 38.6 504.1 8 /2 8 /0 3 11:10 65.9 8 /2 8 /0 3 13:15 504.2 8 /2 8 /0 3 15:45 103.3 504.3 8 /2 8 /0 3 18:15 93.5 304 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C -l (Continued). Station Date Time Sample Depth (cm) Water Depth (m) Volume Pumped (L Pore Water Samples Kg 505.11 9 /1 1 /0 3 10:00 7.5 0.34825 505.12 9 /1 1 /0 3 10:00 27.5 0.32525 505.13 9 /1 1 /0 3 10:00 47.5 0.41441 505.14 9 /1 1 /0 3 10:00 77.5 0.34766 505.2A 9 /1 1 /0 3 10:55 7.5 0.32969 505.2B 9 /1 1 /0 3 10:55 27.5 0.42603 505.2D 9 /1 1 /0 3 10:55 77.5 0.39689 505.BEACH> 9 /1 1 /0 3 13:25 7.5 0.45696 505.BEACH> 9 /1 1 /0 3 13:30 27.5 0.49415 505.BEACH2 9 /1 1 /0 3 13:45 47.5 0.45111 05.BEACHC/ 9 /1 1 /0 3 13:55 77.5 0.50838 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2. Temperature and salinity data from each station. BOLD numbers were measured after water pumped to the surface and collected in a bucket. Station Depth Salinity Temp. Depth Salinity Temp. 1 Cm ) CPSU) CC) Station (m) (PSU) C O 0.0 33.3 21.2 5 0.0 33.6 20.0 1.5 33.5 20.1 1.5 33.6 19.5 3.0 33.6 19.8 3.0 33.6 19.4 4.6 33.6 19.1 4.6 33.6 19.3 6.1 33.6 18.6 6.1 33.6 19.3 7.6 33.6 18.4 7.6 33.6 19.3 9.1 33.6 18.3 9.1 33.6 19.3 10.6 33.6 17.9 10.6 33.6 19.3 12.1 33.6 19.3 0.0 33.5 20.8 0.0 33.6 19.3 1.5 33.5 20.0 3.0 33.6 18.9 6 0.0 33.2 20.4 4.6 33.6 18.8 1.5 33.3 20.1 6.1 33.6 18.3 2.9 33.3 19.4 7.6 33.6 17.7 4.1 33.4 19.0 9.1 33.5 17.1 5.3 33.4 18.9 10.6 33.6 16.4 6.6 33.5 18.8 12.1 33.5 16.0 7.5 33.5 18.7 0.0 33.6 20.7 8.7 33.5 18.6 10.0 33.5 18.6 0.0 33.6 19.7 0.0 33.6 19.7 1.5 33.6 19.5 1.5 33.5 19.6 2.9 33.6 19.0 2.9 33.5 19.4 4.3 33.6 18.8 5.3 33.6 18.5 7 0.0 33.4 20.0 6.2 33.6 18.1 1.5 33.4 19.9 6.5 33.6 17.6 3.0 33.4 19.8 6.9 33.6 17.5 4.4 33.5 19.7 7.8 33.5 17.1 5.7 33.5 19.4 0.0 33.7 19.3 6.9 33.5 18.9 7.5 33.5 18.7 0.0 33.5 19.3 8.2 33.5 18.2 1.5 33.5 19.2 8.6 33.4 17.6 3.0 33.5 19.2 0.0 33.5 19.8 4.6 33.5 19.2 6.1 33.5 19.2 7.6 33.5 19.2 9.1 33.5 19.2 10.6 33.5 19.2 12.1 33.5 19.2 0.0 33.5 19.2 306 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Depth Salinity Temp. Station (nO fPSU) f°C) 8 0.0 33.3 18.1 1.1 33.3 18.1 2.2 33.3 18.1 3.2 33.3 18.1 4.3 33.4 18.1 5.4 33.4 18.1 6.5 33.4 18.1 7.5 33.4 18.1 8.6 33.4 18.1 0.0 33.4 18.1 9 0.0 33.4 18.2 1.5 33.4 18.2 2.9 33.4 18.2 4.4 33.4 18.1 5.7 33.5 17.8 7.2 33.5 17.7 8.3 33.5 17.7 9.7 33.5 17.6 10.6 33.4 17.4 0.0 33.4 18.2 10 0.0 33.2 19.9 1.5 33.3 19.4 3.0 33.3 18.8 4.6 33.3 18.3 6.1 33.3 18.3 7.5 33.3 18.1 8.8 33.3 17.9 9.7 33.3 17.6 10.6 33.4 17.3 0.0 33.3 19.9 11 0.0 17.0 2.0 17.4 4.0 17.4 6.0 17.4 7.9 17.4 9.8 16.8 11.8 16.3 16.3 15.0 Depth Salinity Temp. Station (iri) (PSin C O 12 0.0 17.3 2.0 17.3 4.0 17.3 6.0 17.3 7.9 17.0 9.8 17.0 11.8 16.8 13.8 16.8 15.1 17.0 13 0.0 17.0 2.0 17.0 3.9 17.0 5.9 16.9 7.7 16.8 9.7 15.1 9.1 16.8 21 0.0 14.3 1.5 13.5 3.0 13.2 4.6 12.6 6.1 12.4 7.6 12.2 22 0.0 14.2 1.5 13.8 3.0 13.2 4.4 13.1 5.9 13.1 7.4 12.9 8.6 12.8 10.0 12.7 23 0.0 13.0 1.5 12.9 2.9 12.9 4.3 12.7 5.7 12.7 7.2 12.4 8.3 12.3 9.7 12.2 307 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Station 24 25 26 27 Depth Cm ) Salinity CPSU) Temp. C *C ) Station Depth (m) Salinity CPSU) Temp. C ’C) 0.0 12.9 28 0.0 33.2 18.7 1.4 12.9 1.5 33.2 18.7 2.9 12.9 3.0 33.2 18.7 4.3 12.8 4.6 33.2 18.6 5.7 12.8 6.1 33.2 17.3 6.9 12.7 7.6 33.2 16.7 8.3 12.6 9.1 33.2 16.5 9.7 12.6 11.0 12.6 29 0.0 33.3 17.8 1.5 33.2 17.8 0.0 32.9 17.3 3.0 33.2 17.7 1.5 32.9 17.3 4.6 33.2 17.7 3.0 32.9 17.3 6.1 33.3 17.7 4.6 32.9 17.2 7.6 33.3 17.6 6.1 32.9 17.1 9.1 33.3 17.6 7.6 32.9 17.1 10.6 33.3 17.5 9.1 32.9 17.1 12.1 33.3 17.3 10.6 32.9 16.8 16.8 33.2 17.4 12.1 32.9 16.4 12.0 33.2 17.1 30 0.0 33.3 18.2 21.0 33.0 14.9 1.5 33.3 18.1 3.0 33.3 18.1 0.0 33.2 17.6 4.5 33.3 18.1 1.5 33.1 17.4 6.0 33.3 18.0 3.0 33.1 17.2 7.5 33.3 17.8 4.5 33.1 17.1 9.0 33.3 17.5 5.9 33.1 16.9 10.5 33.3 17.4 7.4 33.1 16.8 12.0 33.3 17.3 8.8 33.2 16.8 12.2 33.2 17.9 10.3 33.2 16.7 21.3 33.2 16.9 11.8 33.2 16.7 11.9 33.2 17.3 31 0.0 33.2 17.7 1.5 33.2 17.7 0.0 33.1 19.3 3.0 33.2 17.7 1.5 33.1 19.3 4.4 33.2 17.6 2.9 33.1 19.2 5.7 33.2 17.6 4.4 33.2 19.1 6.6 33.2 17.5 5.9 33.2 17.4 7.9 33.2 17.5 9.0 33.2 16.7 8.7 33.2 17.5 11.8 33.2 16.3 10.0 33.2 16.5 10.5 33.2 15.5 8.6 33.2 16.2 15.1 33.2 14.9 308 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Station 32 33 34 Depth (m) Salinity CPSU) Temp. (*C) 0.0 33.1 18.8 1.5 33.1 18.8 3.0 33.1 18.7 4.6 33.1 18.6 6.1 33.1 18.5 7.6 33.1 18.5 9.1 33.1 18.5 0.0 33.3 18.3 1.5 33.2 18.2 2.9 33.2 18.1 4.4 33.2 18.0 5.9 33.2 17.9 7.4 33.2 17.9 8.8 33.2 17.8 10.3 33.2 17.8 11.8 33.2 17.8 13.2 33.2 17.7 12.2 33.3 18.4 16.8 33.2 18.0 0.0 33.2 18.3 1.4 33.2 18.1 2.8 33.2 17.7 4.1 33.2 17.7 5.5 33.2 17.6 6.6 33.2 17.5 7.9 33.2 17.5 9.2 33.2 17.4 10.6 33.2 17.3 11.9 33.2 17.2 20.6 33.2 17.3 23.6 33.2 16.5 Station Dopth Salinity Temp. 35 36 Cm ) CPSU) C*c) 0.0 33.3 18.0 1.5 33.3 18.0 2.9 33.3 18.0 4.1 33.3 18.0 5.0 33.3 17.9 5.4 33.3 17.9 6.5 33.3 17.9 7.5 33.3 17.9 8.6 33.3 17.9 9.7 33.3 17.9 13.7 33.2 17.7 15.7 33.3 17.3 0.0 33.3 19.2 1.5 33.3 19.2 3.0 33.3 19.2 4.4 33.3 19.2 5.7 33.3 19.1 7.2 33.3 19.1 8.6 33.3 19.1 10.0 33.3 18.7 11.5 33.3 18.2 12.9 33.3 18.0 13.8 33.3 17.2 15.2 33.3 16.5 16.6 33.3 16.3 18.0 33.3 15.2 19.3 33.3 14.7 20.7 33.3 13.9 22.1 33.3 13.7 23.5 33.3 13.5 24.9 33.3 13.1 309 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Station Depth Salinity Temp. Cm) CPSU) C*c) 0.0 33.4 19.4 1.5 33.4 19.4 3.0 33.4 19.2 4.6 33.4 19.2 6.1 33.4 19.2 7.6 33.4 19.1 9.1 33.3 18.0 10.6 33.3 17.2 12.1 33.3 16.7 13.5 33.3 16.3 15.0 33.3 15.4 16.5 33.3 14.7 18.0 33.2 14.0 19.5 33.2 13.8 21.0 33.2 13.5 22.5 33.2 13.0 24.0 33.2 12.9 25.5 33.2 12.9 27.0 33.2 12.8 0.0 33.3 19.5 1.5 33.3 19.5 3.0 33.3 19.5 4.6 33.3 19.4 6.1 33.3 19.3 7.6 33.3 19.2 9.1 33.3 19.0 10.6 33.2 16.6 37 38 Station Depth Cm ) Salinity CPSU) Temp. CC) 0.0 33.3 19.5 1.4 33.3 19.5 2.3 33.3 19.4 3.5 33.3 19.3 4.7 33.3 19.3 4.4 33.4 19.2 5.2 33.4 19.2 6.1 33.4 19.1 7.0 33.4 19.1 7.9 33.4 19.0 8.7 33.4 18.9 9.6 33.4 18.9 10.5 33.4 18.9 11.4 33.4 18.8 12.2 33.4 18.8 13.1 33.4 18.5 14.0 33.4 18.4 22.5 33.4 14.7 0.0 33.3 19.3 1.5 33.4 19.3 3.0 33.4 19.3 4.6 33.4 19.3 6.1 33.4 19.2 7.6 33.4 18.7 9.1 33.4 17.2 10.6 33.4 15.9 39 40 310 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Depth Salinity Temp. Depth Salinity Temp. Station 41 42 43 Cm ) CPSU) C *C ) Station Cm ) CPSU) (*C) 0.0 33.3 19.5 44 0.0 33.3 19.6 1.5 33.3 19.1 1.5 33.3 19.6 3.0 33.3 19.1 3.0 33.3 18.5 4.6 33.3 19.0 4.6 33.3 18.1 6.1 33.3 18.9 6.1 33.3 18.0 7.6 33.3 18.9 7.6 33.3 17.7 9.0 33.3 18.9 9.1 33.3 15.4 10.5 33.3 18.9 12.0 33.3 18.8 45 0.0 33.3 20.0 13.5 33.3 18.6 1.5 33.3 19.9 15.0 33.3 18.2 3.0 33.3 19.5 16.5 33.3 16.9 4.6 33.3 17.6 18.0 33.3 15.8 6.1 33.3 16.5 19.5 33.3 14.8 7.6 33.3 16.0 21.0 33.3 13.6 9.0 33.3 15.4 22.5 33.3 13.5 10.5 33.3 15.1 24.0 33.3 13.4 12.0 33.3 14.5 13.5 33.3 14.3 0.0 33.5 19.1 15.0 33.3 13.7 1.5 33.5 19.0 16.5 33.3 13.4 2.9 33.4 19.0 4.4 33.4 18.8 46 0.0 33.3 20.0 5.0 33.3 16.3 1.5 33.3 20.0 5.8 33.3 16.2 3.0 33.3 19.5 6.5 33.3 16.1 4.6 33.3 16.4 7.5 33.3 16.0 6.1 33.3 16.0 8.6 33.3 15.5 7.6 33.3 15.8 9.8 33.3 15.3 11.8 33.3 15.1 12.2 33.3 14.3 14.0 33.2 14.2 15.7 44.2 14.1 0.0 33.4 18.6 1.5 33.4 18.5 3.0 33.4 18.5 4.6 33.4 18.2 6.1 33.4 18.0 7.6 33.4 17.6 9.1 33.4 16.5 311 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Depth Salinity Station 47 Temp. Depth Salinity Temp. 48 (m) (psm CC) Station (m) (PSU) CC) 0.0 33.6 20.2 49 0.0 33.5 20.0 1.5 33.6 20.2 1.5 33.5 20.0 2.9 33.6 20.2 3.0 33.5 19.9 4.1 33.6 20.2 4.6 33.5 19.9 5.3 33.6 20.1 6.1 33.5 19.9 6.2 33.6 19.0 7.6 33.5 19.8 7.0 33.5 18.2 9.1 33.5 19.7 8.2 33.5 17.7 10.6 33.5 19.5 8.6 33.5 17.4 12.1 33.5 19.3 9.7 33.5 17.1 13.5 33.5 19.0 9.8 33.5 17.0 15.0 33.5 18.3 10.8 33.5 16.8 16.5 33.5 17.7 11.8 33.5 15.7 18.0 33.5 17.3 12.7 33.5 15.4 19.5 33.5 15.8 13.7 33.5 15.2 14.7 33.5 14.8 50 0.0 33.4 20.4 15.7 33.5 14.5 1.5 33.4 20.3 16.7 33.4 14.4 3.0 33.4 20.0 17.6 33.4 14.3 4.6 33.4 19.7 18.6 33.4 14.3 6.1 33.4 19.5 7.5 33.5 19.4 0.0 33.5 20.2 9.0 33.5 18.4 1.5 33.5 20.1 10.5 33.5 17.5 3.0 33.5 20.0 4.6 33.5 19.9 51 0.0 33.5 19.6 6.1 33.5 19.9 1.5 33.5 19.6 7.6 33.5 19.8 3.0 33.5 19.6 9.1 33.5 19.6 4.6 33.5 19.6 10.6 33.5 19.5 6.1 33.5 19.6 12.1 33.5 18.9 7.6 33.5 19.3 13.5 33.5 18.3 9.1 33.5 18.6 15.0 33.5 16.7 10.6 33.5 17.5 16.5 33.5 16.0 12.0 33.4 16.3 18.0 33.5 1 5.5 13.5 33.4 16.1 19.5 33.4 15.0 15.0 33.4 15.2 21.0 33.4 14.7 16.5 33.4 14.9 22.5 33.4 13.8 18.0 33.4 14.6 24.0 33.4 13.3 19.5 33.4 14.2 25.5 33.4 13.1 20.6 33.4 13.9 27.0 33.4 12.9 22.1 33.4 13.7 28.5 33.4 12.8 23.6 33.4 13.6 24.3 33.4 13.4 25.8 33.4 13.0 27.2 33.4 12.7 312 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Station Depth Salinity Temp. Depth Salinity Temp. 52 53 (m) CPSU) CC) Station Cm ) CPSU) (*c) 0.0 33.4 20.0 54 0.0 33.2 21.0 1.5 33.4 19.9 1.5 33.2 20.9 3.0 33.4 19.8 3.0 33.2 20.6 4.6 33.4 19.6 4.6 33.3 20.0 6.1 33.4 19.3 6.1 33.3 19.8 7.5 33.5 19.1 7.6 33.3 19.4 9.0 33.5 18.9 9.1 33.3 17.5 10.5 33.5 18.8 10.6 33.2 16.2 12.0 33.5 17.9 13.2 33.5 16.9 55 0.0 33.4 18.5 14.7 33.4 15.2 1.5 33.4 18.4 16.2 33.4 15.0 3.0 33.4 18.4 17.7 33.4 14.8 4.6 33.4 18.2 19.1 33.4 14.2 6.1 33.4 17.4 20.6 33.4 13.9 7.6 33.3 14.8 22.1 33.4 13.6 9.1 33.3 14.1 23.6 33.4 13.3 10.6 33.3 13.4 25.0 33.4 13.0 12.1 33.3 13.0 26.5 33.4 12.8 13.7 33.3 12.8 28.0 33.4 12.4 15.0 33.3 12.7 16.5 33.3 12.1 0.0 33.3 20.2 17.7 33.3 11.9 1.5 33.3 20.2 19.1 33.3 11.9 3.0 33.3 19.8 20.0 33.3 11.8 4.6 33.3 19.4 21.5 33.3 11.7 6.1 33.4 19.0 22.9 33.3 11.6 7.6 33.4 18.8 24.3 33.3 11.6 9.1 33.4 18.6 25.8 33.3 11.6 10.6 33.4 18.4 27.2 33.3 11.5 12.1 33.4 16.5 13.5 33.4 15.8 15.0 33.4 15.3 16.5 33.4 14.9 18.0 33.4 13.7 19.5 33.3 12.9 313 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Depth Salinity Station 56 Temp. Depth Salinity Temp. 57 58 (m) (psm c c ) Station (m) (PSU) C*c) 0.0 33.4 19.0 60 0.0 32.8 19.3 1.5 33.4 19.0 1.5 32.2 19.1 3.0 33.4 18.2 3.0 32.3 17.6 4.6 33.4 15.4 4.6 32.5 16.4 6.1 33.4 14.3 7.6 33.3 13.7 61 0.0 33.1 17.9 9.1 33.3 13.5 1.5 33.1 17.9 10.6 33.3 13.3 3.0 33.1 17.9 12.1 33.3 13.2 4.6 33.1 17.9 13.7 33.3 13.0 6.1 33.1 17.5 15.0 33.3 12.5 7.6 33.1 17.4 16.5 33.3 12.3 9.1 33.1 17.4 18.0 33.3 12.3 19.5 33.3 12.1 62 0.0 33.1 18.0 21.0 33.3 12.0 1.5 33.1 18.0 22.5 33.3 12.0 3.0 33.1 18.0 23.6 33.3 11.9 4.6 33.1 18.0 25.0 33.3 11.9 6.1 33.1 17.9 26.5 33.3 11.8 7.6 33.1 17.8 28.0 33.3 11.6 9.1 33.1 17.7 10.6 33.1 17.5 0.0 33.4 19.7 12.1 33.1 17.3 1.5 33.4 19.6 13.7 33.1 17.1 3.0 33.4 18.2 15.2 33.1 15.5 4.6 33.3 15.0 16.7 33.1 14.4 6.1 33.3 14.1 15.5 33.1 16.9 7.6 33.3 13.3 9.1 33.3 13.0 10.6 33.3 13.0 12.1 33.3 12.8 13.7 33.3 12.8 15.0 33.3 12.8 16.5 33.3 12.5 18.0 33.3 12.3 19.5 33.3 12.3 0.0 33.4 19.9 1.5 33.4 19.7 3.0 33.4 18.8 4.6 33.4 15.5 6.1 33.3 14.4 7.6 33.3 13.4 9.1 33.3 13.3 10.6 33.3 13.3 314 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-2 Continued. Station 63 64 Depth Salinity Temp. Depth Salinity Temp. (m) CPSU) CC) Station (m) CPSU) (*C) 0.0 18.3 33.0 65 (Con't] 15.0 33.1 16.1 1.5 18.3 33.0 16.2 33.1 13.8 3.0 18.3 33.0 17.7 33.1 13.5 4.6 17.8 33.0 7.6 33.1 19.5 6.1 16.7 33.0 16.7 32.8 14.9 7.6 15.9 33.0 9.1 15.6 33.0 66 0.0 33.1 19.6 10.5 15.3 33.0 1.5 33.1 19.7 12.0 15.1 33.0 3.0 33.1 19.6 13.5 14.9 33.0 4.6 33.1 19.3 15.0 14.6 33.0 6.0 33.1 19.1 16.5 14.4 33.0 7.5 33.1 19.0 18.0 13.2 33.0 9.0 33.1 18.9 19.1 12.8 33.0 10.5 33.1 18.7 20.6 12.8 33.0 11.8 33.1 18.0 22.1 12.0 33.0 13.2 33.1 17.5 23.6 11.6 33.0 14.7 33.1 17.2 24.3 11.6 33.0 16.2 33.1 16.4 25.8 11.5 33.0 17.7 33.1 15.6 27.2 11.5 33.0 19.1 33.1 15.5 12.0 33.0 15.8 20.6 33.1 13.9 19.0 33.0 14.1 22.1 33.1 13.5 23.6 33.1 12.0 0.0 33.0 19.8 24.3 33.1 12.0 1.5 32.9 19.7 25.8 33.1 11.8 3.0 32.9 19.7 27.2 33.1 11.8 4.6 33.0 19.4 7.4 33.2 19.2 6.1 33.0 19.1 17.7 32.7 16.5 7.6 33.0 18.7 9.0 33.0 17.6 10.5 33.0 17.3 0.0 33.0 20.1 1.5 33.0 19.9 3.0 33.0 19.6 4.6 33.0 19.6 6.1 33.1 19.4 7.6 33.1 19.2 9.1 33.1 18.8 10.6 33.1 17.7 12.1 33.1 17.5 13.5 33.1 17.3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-3. Temperature and salinity at beach stations. Station Date Time Tamp (*C) Sal CPSU) 403.1 9 /6 /0 2 7:35 17.6 33.3 403.2 9 /6 /0 2 9:30 17.4 33.3 403.3 9 /6 /0 2 11:30 18.2 33.3 403.4 9 /6 /0 2 13:30 18.5 33.3 403.5 9 /6 /0 2 15:30 17.9 33.3 404.1 9 /8 /0 2 7:30 18.2 404.2 9 /8 /0 2 9:30 18.6 404.3 9 /8 /0 2 11:20 18.7 404.4 9 /8 /0 2 13:05 20.2 404.5 9 /8 /0 2 15:05 20.9 501.1 6 /1 8 /0 3 10:15 17.2 501.2 6 /1 8 /0 3 12:00 18.4 501.3 6 /1 8 /0 3 14:05 17.8 501.4 6 /1 8 /0 3 16:00 18.2 502.1 6 /2 5 /0 3 10:30 19.6 502.2 6 /2 5 /0 3 12:30 19.7 502.3 6 /2 5 /0 3 14:15 19.8 502.4 6 /2 5 /0 3 16:15 20.0 502.5 6 /2 5 /0 3 17:35 19.9 504.1 8 /2 8 /0 3 11:10 15.9 32.6 8 /2 8 /0 3 13:15 17.0 32.5 504.2 8 /2 8 /0 3 15:45 17.5 32.8 504.3 8 /2 8 /0 3 18:15 17.2 32.8 316 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-4: Average station location, distance offshore, and water depth. Drift was computed by the change in location between the beginning and end o f sample collection. Average Distance Depth Drift Station Location_________ Latitude Longitude Offhora (km) ftri) flan) 1 Anaheim Bay 33.7146 118.1056 2.27 10 0.0 2 Anaheim Bay 33.6973 118.1397 5.39 22 0.0 3 Anaheim Bay 33.6643 118.1523 9.22 27 5.1 4 Shelf 33.5807 1 18.2238 19.98 82 0.0 5 Huntington Beach 33.5599 118.0265 9.86 212 2.9 6 Huntington Beach 33.6024 117.9672 2.97 24 2.8 7 Huntinqton Beach 33.6230 1 17.9735 1.35 13 1.3 8 Anaheim Bay 33.6634 118.1460 9.05 27 1.7 9 Anaheim Bay 33.7010 118.1345 4.82 20 1.1 10 Anaheim Bay 33.7156 118.0955 1.65 9 2.4 11 Hermosa Beach 33.8613 118.4756 6.16 80 1.1 12 Hermosa Beach 33.8640 118.4329 2.56 50 2.1 13 Hermosa Beach 33.8654 118.4105 0.58 10 0.0 21 Huntington Beach 33.6577 118.0311 1.67 14 0.0 22 Huntington Beach 33.6392 118.0509 4.41 19 0.0 23 Huntington Beach 33.6207 118.0758 7.48 38 2.1 24 Huntinqton Beach 33.5901 118.1058 11.89 84 3.1 25 Sunset Beach 33.6926 118.1724 7.28 22.9 1.7 26 Sunset Beach 33.7083 118.1068 2.89 12.8 1.8 27 Sunset Beach 33.7148 118.0991 1.90 8.2 0.0 28 Huntington Beach 33.6613 118.0244 1.08 9.1 0.0 29 Huntington Beach 33.6433 118.0461 3.77 19.8 0.0 30 Huntington Beach 33.6215 118.0705 7.09 33.5 2.0 31 Huntinqton Beach 33.5851 118.1088 12.48 185.9 0.8 32 Huntington Beach 33.6622 118.0245 0.99 8.2 0.0 33 Huntington Beach 33.6433 118.0460 3.77 19.8 0.0 34 Huntington Beach 33.6278 118.0632 6.12 30.5 0.0 35 Huntinqton Beach 33.5930 118.0948 10.97 85.3 2.6 36 Hermosa Beach 33.8571 118.4641 5.42 76.8 2.5 37 Hermosa Beach 33.8628 118.4303 2.40 30.5 0.0 38 Hermosa Beach 33.8637 118.4101 0.59 9.8 0.0 39 Hermosa Beach 33.8525 118.5173 10.12 85.3 2.6 40 Santa Monica 34.0158 118.5217 1.12 10.7 0.0 41 Santa Monica 34.0046 118.5343 2.82 21.9 0.0 42 Santa Monica 33.9903 118.5510 5.02 53.6 4.4 43 Newport Beach 33.6029 1 17.9279 0.34 10.1 0.0 44 Huntington Beach 33.6616 118.0243 1.05 10.4 0.0 45 Huntington Beach 33.6477 118.0403 3.07 16.8 0.0 46 Sunset Beach 33.7097 118.0743 0.72 7.9 0.0 47 Huntington SB 33.5891 118.0634 9.31 115.8 2.8 48 Huntington SB 33.6080 1 18.0358 6.09 30.5 0.0 49 Huntington SB 33.6239 118.0050 2.96 19.8 0.0 50 Huntington SB 33.6359 1 17.9830 0.68 9.8 0.0 51 Huntington SB 33.5937 118.0576 8.58 71.6 2.2 52 Huntington SB 33.6078 118.0356 6.09 30.5 0.0 53 Huntington SB 33.6239 118.0050 2.96 20.1 0.0 54 Huntinqton SB 33.6359 117.9830 0.67 9.8 0.0 55 Herm osa Beach 33.8429 1 1 8 .4 4 5 4 4 .1 0 76.8 0.8 56 Hermosa Beach 33.8559 118.4244 2.10 30.5 0.0 57 Hermosa Beach 33.8590 118.4175 1.40 19.8 0.0 58 Hermosa Beach 33.8615 118.4092 0.61 10.4 0.0 317 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-4 (Continued). Station Location Average Latitude Longitude Distance Offhore flan) Depth C m ) Drift (ton) 60 Fish Harbor 33.736 118.269 0.000 6.0 0.0 61 Huntington Beach 33.6624 118.0229 0.94 9.0 0.0 62 Huntington Beach 33.6456 118.0405 3.28 17.2 0.0 63 Huntinqton Beach 33.6326 118.0535 5.15 27.4 0.0 64 Huntington Beach 33.661 1 118.0244 1.10 10.0 0.0 65 Huntington Beach 33.6455 118.0405 3.29 19.0 0.0 66 Huntinqton Beach 33.6323 118.0534 5.16 28.0 0.0 Shoreline Samples Station L O C A T IO N Latitude Longitude 101 Huntington Beach 33.63559 117.97062 102 Belmont Shore 33.75666 118.1471 103 Hermosa Beach 33.86488 11 8.40484 201 Sunset Beach 33.72023 118.07796 202 Huntington Beach 33.64736 117.99245 203 Huntington Beach 33.64736 117.99245 204 Hermosa Beach 33.86488 118.40484 205 Santa Monica 34.03351 1 18.53669 301 Sunset Beach 33.72023 118.07796 302 >rth Huntington Bee 33.64736 117.99245 303 Newport Beach 33.60703 117.9287 401 ngton Beach-Power 33.6392 117.97745 402 ngton Beach-Power 33.6392 117.97745 403 Huntington Beach 33.6392 117.97745 404 Hermosa Beach 33.86488 11 8.40484 501 Huntington Beach 33.64736 117.99245 502 Huntington Beach 33.64736 117.99245 503 Hermosa Beach 33.86488 118.40484 504 Newport Beach 33.62666 1 17.95338 505 Huntinqton Beach 33.63559 117.97062 318 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-5. Mixed layer thickness and surface temperature. BOLD numbers are minimum values Mixed Layer Temp. (*C) Station Location Date Thickness (m) 9 1.5m 1 Anaheim Bay 9 /11/01 7 20.1 2 Anaheim Bay 9 /11/01 7 20 3 Anaheim Bay 9 /11/01 6 19.5 4 Shelf 9 /13/01 12.1 19.2 5 Huntington Beach 9/13/01 12.1 19.5 6# Huntington Beach 9/1 3 /0 1 10 20.1 7# Huntington Beach 9/1 3 /0 1 8.5 19.9 8 Anaheim Bay 9/2 0 /0 1 8.6 18.1 9 Anaheim Bay 9/20/01 11 18.2 10 Anaheim Bay 9/20/01 9 19.4 11 Hermosa Beach 11/1/01 12 Hermosa Beach 11/1/01 13 Hermosa Beach 11/1/01 21 Huntington Beach 4 /3 /0 2 22 Huntington Beach 4 /3 /0 2 23 Huntington Beach 4 /3 /0 2 24 Huntington Beach 4 /3 /0 2 25 Sunset Beach 6 /5 /0 2 12 17.3 26 Sunset Beach 6 /5 /0 2 12 17.4 27 Sunset Beach 6 /5 /0 2 12 19.3 28 Huntington Beach 6 /7 /0 2 7 18.7 29 Huntington Beach 6 /7 /0 2 15 17.8 30 Huntington Beach 6 /7 /0 2 20 18.1 31 Huntington Beach 6 /7 /0 2 10 17.7 32 Huntington Beach 6 /1 2 /0 2 9.11 18.8 33 Huntington Beach 6 /1 2 /0 2 17 18.2 34 Huntington Beach 6 /1 2 /0 2 22 18.1 35 Huntington Beach 6 /1 2 /0 2 15 18 36 Hermosa Beach 6 /2 1 /0 2 14 19.2 37 Hermosa Beach 6 /2 1 /0 2 10 19.4 38 Hermosa Beach 6 /2 1 /0 2 10 19.5 39 Hermosa Beach 6 /2 1 /0 2 14 19.5 40 Santa Monica 6 /2 2 /0 2 10 19.3 41 Santa Monica 6 /2 2 /0 2 16 19.1 42 Santa Monica 6 /2 2 /0 2 5 19 43 Newport Beach 7 /2 6 /0 2 8 18.5 44 Huntington Beach 7 /2 6 /0 2 8 19.6 45 Huntington Beach 7 /2 6 /0 2 5 19.9 46 Sunset Beach 7 /2 6 /0 2 4 20 319 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T able C-5 Station (Continued). Location Data Mixed Layer Thickness (m) Temp. CC) • 1.5m 47 Huntington Beach 8/19/02 10 20.2 48 Huntington Beach 8/19/02 14 20.1 49 Huntington Beach 8/19/02 18 20 50 Huntinqton Beach 8/19/02 10 20.3 51 Huntington Beach 8/23/02 11 19.6 52 Huntington Beach 8/23/02 13 19.9 53 Huntington Beach 8/23/02 11 20.2 54 Huntinqton Beach 8/23/02 10 20.9 55 Hermosa Beach 9/8/02 7 18.4 56 Hermosa Beach 9/8/02 4 19 57 Hermosa Beach 9/8/02 4 19.6 58 Hermosa Beach 9/8/02 4 19.7 60 Fish Harbor 6/18/03 2.5 19.1 61 Huntington Beach 6/18/03 10 17.9 62 Huntington Beach 6/18/03 14 18 63 Huntinqton Beach 6/18/03 6 18.3 64 Huntington Beach 6/25/03 8 19.7 65 Huntington Beach 6/25/03 9 19.9 66 Huntinqton Beach 6/25/03 10.5 19.7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6. Tides for 3 days around each sampling date. Statlons/Locatlon/Date Date Time Height Cm) Tide 1 0 /1 8 /9 7 4:48 0.42 Low 1 0 /1 8 /9 7 11:00 1.95 High 1 0 /1 8 /9 7 17:55 -0.13 Low 1st Shoreline 1 0 /1 9 /9 7 0:18 1.34 High Huntington Beach 1 0 /1 9 /9 7 5:29 0.58 Low 10/19/97 1 0 /1 9 /9 7 11:43 1.83 High 1 0 /1 9 /9 7 18:52 -0.03 Low 1 0 /2 0 /9 7 1:26 1.22 High 1 0 /2 0 /9 7 6:18 0.74 Low 1 0 /2 0 /9 7 12:32 1.68 High 1 0 /2 0 /9 7 19:57 0.08 Low 1 1 /1 9 /9 7 1:24 1.18 High 1 1 /1 9 /9 7 5:52 0.89 Low 1 1 /1 9 /9 7 11:47 1.51 High 1 1 /1 9 /9 7 19:22 0.1 Low 2nd Shoreline 1 1 /2 0 /9 7 2:43 1.2 High Huntington Beach 1 1 /2 0 /9 7 7:22 0.94 Low 11/20/97 1 1 /2 0 /9 7 12:51 1.34 High 1 1 /2 0 /9 7 20:25 0.2 Low 1 1 /2 1 /9 7 3:52 1.26 High 1 1 /2 1 /9 7 9:13 0.9 Low 1 1 /2 1 /9 7 14:17 1.21 High 1 1 /2 1 /9 7 21:27 0.27 Low 1 0 /1 8 /0 0 2:17 1.08 High 1 0 /1 8 /0 0 6:24 0.85 Low 1 0 /1 8 /0 0 12:55 1.65 High 1 0 /1 8 /0 0 20:48 0.1 Low Sustainable CHy HB1-6 1 0 /1 9 /0 0 4:12 1.09 High Huntington Beach 1 0 /1 9 /0 0 7:51 0.97 Low 1 0 /1 9 /0 0 1 0 /1 9 /0 0 14:14 1.56 High 1 0 /1 9 /0 0 22:13 0.08 Low 1 0 /2 0 /0 0 5:39 1.2 High 1 0 /2 0 /0 0 10:02 0.97 Low 1 0 /2 0 /0 0 15:51 1.52 High 1 0 /2 0 /0 0 23:25 0.03 Low 321 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 1 0 /2 2 /0 0 0:21 -0.02 Low 1 0 /2 2 /0 0 7:06 1.48 High 1 0 /2 2 /0 0 12:41 0.62 Low 1 0 /2 2 /0 0 18:27 1.6 High H B 7, H B 8 1 0 /2 3 /0 0 1:07 -0.03 Low 10/23/00 1 0 /2 3 /0 0 7:39 1.61 High 1 0 /2 3 /0 0 13:31 0.41 Low 1 0 /2 3 /0 0 19:24 1.64 High 1 0 /2 4 /0 0 1:47 0 Low 1 0 /2 4 /0 0 8:10 1.72 High 1 0 /2 4 /0 0 14:15 0.22 Low 1 0 /2 4 /0 0 20:14 1.64 High 1 0 /2 5 /0 0 2:22 0.07 Low 1 1 /3 0 /0 0 0:47 1.06 High 1 1 /3 0 /0 0 4:26 0.91 Low 1 1 /3 0 /0 0 10:40 1.56 High 1 1 /3 0 /0 0 18:30 0.1 Low Sustainable City H B 11-16 1 2 /1 /0 0 2:04 1.07 High Huntington Beach 1 2 /1 /0 0 5:18 0.99 Low 12/1/00 1 2 /1 /0 0 11:22 1.42 High 1 2 /1 /0 0 19:25 0.18 Low 1 2 /2 /0 0 3:21 1.12 High 1 2 /2 /0 0 6:57 1.03 Low 1 2 /2 /0 0 12:21 1.29 High 1 2 /2 /0 0 20:23 0.24 Low HB17, HB18 1 2 /3 /0 0 4:12 1.2 High 12/3/00 1 2 /3 /0 0 9:09 0.98 Low 1 2 /3 /0 0 13:48 1.17 High 1 2 /3 /0 0 21:19 0.29 Low 1 2 /4 /0 0 4:45 1.3 High 1 2 /4 /0 0 10:36 0.83 Low 1 2 /4 /0 0 15:26 1.11 High 1 2 /4 /0 0 22:08 0.32 Low 322 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date_______Date Time Ht. On) Tide 1,2,3 Anaheim Bay 9/11/01 4,5,6,7 Huntington Beach 9/13/01 8,9,10 Anaheim Bay 9/20/01 101 Belmont Shore 9/25/01 9/10/01 4:30 0.94 High 9/10/01 7:40 0.88 Low 9/10/01 15:01 1.49 High 9/10/01 23:13 0.32 Low 9/11/01 6:36 1.03 High 9/11/01 9:45 0.96 Low 9/11/01 16:29 1.56 High 9/12/01 0:21 0.15 Low 9/12/01 7:21 1.15 High 9/12/01 11:31 0.91 Low 9/12/01 17:43 1.69 High 9/13/01 1:11 -0.02 Low 9/13/01 7:54 1.27 High 9/13/01 12:39 0.78 Low 9/13/01 18:45 1.84 High 9/14/01 1:54 -0.16 Low 9/14/01 8:25 1.40 High 9/14/01 13:32 0.61 Low 9/14/01 19:38 1.96 High 9/19/01 5:00 0.09 Low 9/19/01 11:13 1.82 High 9/19/01 17:33 0.08 Low 9/19/01 23:42 1.58 High 9/20/01 5:36 0.30 Low 9/20/01 11:52 1.79 High 9/20/01 18:29 0.12 Low 9/21/01 0:40 1.36 High 9/21/01 6:13 0.51 Low 9/21/01 12:34 1.72 High 9/21/01 19:34 0.20 Low 9/22/01 1:55 1.16 High 9/24/01 6:00 1.10 High 9/24/01 9:43 0.99 Low 9/24/01 15:56 1.46 High 9/24/01 23:49 0.20 Low 9/25/01 7:08 1.20 High 9/25/01 11:35 0.96 Low 9/25/01 17:20 1.47 High 9/26/01 0:47 0.14 Low 9/26/01 7:46 1.29 High 9/26/01 12:41 0.87 Low 9/26/01 18:23 1.53 High 9/27/01 1:30 0.09 Low 323 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 10/15/01 2:37 -0.06 Low 10/15/01 8:54 1.82 High 10/15/01 15:02 0.05 Low 10/15/01 21:08 1.77 High 10/16/01 3:13 0.04 Low 10/16/01 9:26 1.91 High 10/16/01 15:47 -0.08 Low 10/16/01 21:56 1.68 High 10/17/01 3:47 0.17 Low 10/17/01 9:59 1.96 High 10/17/01 16:32 -0.14 Low 10/17/01 22:45 1.55 High 10/18/01 4:21 0.34 Low 10/31/01 1:57 0.35 Low 10/31/01 8:10 1.76 High 10/31/01 14:45 0.07 Low 10/31/01 20:52 1.37 High 11/1/01 2:20 0.43 Low 11/1/01 8:33 1.80 High 11/1/01 15:19 0.00 Low 11/1/01 21:31 1.30 High 11/2/01 2:45 0.52 Low 11/2/01 8:59 1.83 High 11/2/01 15:57 -0.03 Low 11/2/01 22:15 1.22 High 11/3/01 3:11 0.62 Low 1 /8 /0 2 5:10 1.75 High 1 /8 /0 2 12:24 0.04 Low 1 /8 /0 2 18:37 1.05 High 1 /8 /0 2 23:21 0.58 Low 1 /9 /0 2 5:55 1.84 High 1 /9 /0 2 13:14 -0.13 Low 1 /9 /0 2 19:39 1.09 High 1 /1 0 /02 0:12 0.63 Low 1 /1 0 /02 6:36 1.90 High 1 /1 0 /02 13:57 -0.24 Low 1 /1 0 /02 20:28 1.13 High 1 /1 1 /02 0:57 0.66 Low 102 Huntington Beach 10/16/01 11, 12,13 Hermosa Beach 11/ 1/01 103 Hermosa Beach 1/9/02 324 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 4 /2 /0 2 7:15 -0.01 Low 4 /2 /0 2 14:04 0.94 High 4 /2 /0 2 18:08 0.74 Low 21,22,23,24 4 /3 /0 2 0:43 1.49 High Huntington Beach 4 /3 /0 2 8:43 0.05 Low 4/3 /0 2 4 /3 /0 2 16:25 0.95 High 4 /3 /0 2 19:36 0.88 Low 4 /4 /0 2 2:04 1.37 High 4 /4 /0 2 10:14 0.04 Low 4 /4 /0 2 17:52 1.06 High 4 /4 /0 2 21:58 0.89 Low 4 /5 /0 2 3:41 1.33 High 6 /4 /0 2 0:20 0.64 Low 6 /4 /0 2 5:33 1.08 High 6 /4 /0 2 11:49 0.29 Low 6 /4 /0 2 18:39 1.42 High 201,25,26,27 6 /5 /0 2 1:06 0.47 Low Sunset Beach 6 /5 /0 2 6:38 1.07 High 6/5/02 6 /5 /0 2 12:25 0.36 Low 6 /5 /0 2 19:03 1.51 High 6 /6 /0 2 1:45 0.3 Low 6 /6 /0 2 7:33 1.07 High 6 /6 /0 2 12:57 0.43 Low 6 /6 /0 2 19:27 1.61 High 202,28,29,30,31 6 /7 /0 2 2:20 0.14 Low Huntington Beach 6 /7 /0 2 8:22 1.08 High 6/7/02 6 /7 /0 2 13:27 0.5 Low 6 /7 /0 2 19:51 1.69 High 6 /8 /0 2 2:54 0 Low 6 /8 /0 2 9:06 1.09 High 6 /8 /0 2 13:57 0.55 Low 6 /8 /0 2 20:18 1.77 High 6 /9 /0 2 3:28 -0.11 Low 325 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height (m) Tide 6 /1 1 /0 2 4:42 -0.25 Low 6 /1 1 /0 2 11:18 1.09 High 6 /1 1 /0 2 15:33 0.71 Low 6 /1 1 /0 2 21:54 1.88 High 203,32,33,34,35 6 /1 2 /0 2 5:23 -0.28 Low Huntington Beach 6 /1 2 /0 2 12:07 1.08 High 6/12/02 6 /1 2 /0 2 16:12 0.76 Low 6 /1 2 /0 2 22:32 1.86 High 6 /1 3 /0 2 6:07 -0.27 Low 6 /1 3 /0 2 13:01 1.08 High 6 /1 3 /0 2 16:58 0.81 Low 6 /1 3 /0 2 23:16 1.79 High 6 /2 0 /0 2 0:53 0.2 Low 6 /2 0 /0 2 6:45 1.13 High 6 /2 0 /0 2 12:10 0.36 Low 6 /2 0 /0 2 18:46 1.84 High 204,36,37,38,39 6 /2 1 /0 2 1:47 -0.03 Low Hermosa Beach 6 /2 1 /0 2 7:54 1.14 High 6/21/02 6 /2 1 /0 2 12:56 0.46 Low 6 /2 1 /0 2 19:26 1.95 High 205,40,41,42 6 /2 2 /0 2 2:35 -0.21 Low Santa Monica 6 /2 2 /0 2 8:54 1.16 High 6/22/02 6 /2 2 /0 2 13:41 0.54 Low 6 /2 2 /0 2 20:05 2.02 High 6 /2 3 /0 2 3:21 -0.33 Low 6 /2 3 /0 2 9:48 1.17 High 6 /2 3 /0 2 14:24 0.61 Low 6 /2 3 /0 2 20:45 2.05 High 7 /2 5 /0 2 5:06 -0.22 Low 7 /2 5 /0 2 11:40 1.26 High 7 /2 5 /0 2 16:21 0.71 Low 7 /2 5 /0 2 22:28 1.87 High 301,302,303,43,44,45,46 7 /2 6 /0 2 5:40 -0.14 Low Nearshore SPBay 7 /2 6 /0 2 12:13 1.26 High 7/26/02 7 /2 6 /0 2 17:01 0.72 Low 7 /2 6 /0 2 23:04 1.75 High 7 /2 7 /0 2 6:11 -0.03 Low 7 /2 7 /0 2 12:47 1.27 High 7 /2 7 /0 2 17:44 0.74 Low 7 /2 7 /0 2 23:40 1.61 High 326 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 8 /1 8 /0 2 1:38 -0.05 Low 8 /1 8 /0 2 8:23 1.19 High 8 /1 8 /0 2 12:47 0.82 Low 8 /1 8 /0 2 19:01 1.85 High 8 /1 9 /0 2 2:23 -0.13 Low 8 /1 9 /0 2 9:03 1.26 High 8 /1 9 /0 2 13:40 0.77 Low 8 /1 9 /0 2 19:48 1.9 High 8 /2 0 /0 2 3:02 -0.17 Low 8 /2 0 /0 2 9:35 1.31 High 8 /2 0 /0 2 14:24 0.71 Low 8 /2 0 /0 2 20:30 1.92 High 8 /2 2 /0 2 4:06 -0.14 Low 8 /2 2 /0 2 10:30 1.37 High 8 /2 2 /0 2 15:36 0.6 Low 8 /2 2 /0 2 21:40 1.87 High 8 /2 3 /0 2 4:34 -0.07 Low 8 /2 3 /0 2 10:54 1.39 High 8 /2 3 /0 2 16:09 0.57 Low 8 /2 3 /0 2 22:13 1.8 High 8 /2 4 /0 2 4:59 0.01 Low 8 /2 4 /0 2 11:18 1.4 High 8 /2 4 /0 2 16:43 0.55 Low 8 /2 4 /0 2 22:45 1.69 High 8 /2 5 /0 2 5:24 0.13 Low 9 /5 /0 2 2:59 -0.22 Low 9 /5 /0 2 9:22 1.43 High 9 /5 /0 2 14:34 0.52 Low 9 /5 /0 2 20:42 2.04 High 9 /6 /0 2 3:33 -0.25 Low 9 /6 /0 2 9:52 1.54 High 9 /6 /0 2 15:19 0.38 Low 9 /6 /0 2 21:27 2.04 High 9 /7 /0 2 4:08 -0.21 Low 9 /7 /0 2 10:23 1.64 High 9 /7 /0 2 16:05 0.26 Low 9 /7 /0 2 22:13 1.97 High 401,47,48,49,50 Huntington Beach 8/19/02 402,51,52,53,54 Huntington Beach 8/23/02 403 Huntington Beach 9/6/02 327 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 9 /7 /0 2 4:08 -0.21 Low 9 /7 /0 2 10:23 1.64 High 9 /7 /0 2 16:05 0.26 Low 9 /7 /0 2 22:13 1.97 High 9 /8 /0 2 4:43 -0.11 Low 9 /8 /0 2 10:58 1.71 High 9 /8 /0 2 16:55 0.19 Low 9 /8 /0 2 23:01 1.81 High 9 /9 /0 2 5:19 0.06 Low 9 /9 /0 2 11:34 1.76 High 9 /9 /0 2 17:48 0.16 Low 9 /9 /0 2 23:55 1.59 High 9 /1 0 /0 2 5:55 0.26 Low 9 /1 0 /0 2 12:15 1.77 High 9 /1 0 /0 2 18:50 0.17 Low 6 /1 7 /0 3 6:44 -0.35 Low 6 /1 7 /0 3 13:40 1.17 High 6 /1 7 /0 3 17:57 0.81 Low 6 /1 8 /0 3 0:03 1.76 High 6 /1 8 /0 3 7:35 -0.22 Low 6 /1 8 /0 3 14:40 1.2 High 6 /1 8 /0 3 19:08 0.86 Low 6 /1 9 /0 3 0:58 1.56 High 6 /1 9 /0 3 8:27 -0.07 Low 6 /1 9 /0 3 15:39 1.25 High 6 /1 9 /0 3 20:35 0.86 Low 6 /2 4 /0 3 1:25 0.31 Low 6 /2 4 /0 3 7:19 1 High 6 /2 4 /0 3 12:16 0.57 Low 6 /2 4 /0 3 18:54 1.62 High 6 /2 5 /0 3 2:05 0.16 Low 6 /2 5 /0 3 8:16 1.01 High 6 /2 5 /0 3 12:52 0.64 Low 6 /2 5 /0 3 19:23 1.69 High 6 /2 6 /0 3 2:41 0.03 Low 6 /2 6 /0 3 9:05 1.04 High 6 /2 6 /0 3 13:26 0.7 Low 6 /2 6 /0 3 19:53 1.76 High 404,55,56,57,58 Hermosa Beach 9 /8/02 501,60,61,62,63 Huntington Beach 6/18/03 502, 64,65,66 Huntington Beach 6/25/03 328 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-6 (Continued). Statlons/Locatlon/Date Date Time Height Cm) Tide 8 /2 1 /0 3 0:33 0.38 Low 8 /2 1 /0 3 8:00 1.01 High 8 /2 1 /0 3 10:40 0.97 Low 8 /2 1 /0 3 17:28 1.5 High 8/2 2 /0 3 1:23 0.23 Low 8 /2 2 /0 3 8:29 1.09 High 8 /2 2 /0 3 12:07 0.96 Low 8 /2 2 /0 3 18:24 1.6 High 8 /2 3 /0 3 2:01 0.1 Low 8 /2 3 /0 3 8:51 1.16 High 8 /2 3 /0 3 13:01 0.89 Low 8 /2 3 /0 3 19:10 1.72 High 8 /2 4 /0 3 2:33 -0.02 Low 8 /2 6 /0 3 21:05 1.96 High 8 /2 7 /0 3 4:02 -0.18 Low 8 /2 7 /0 3 10:22 1.43 High 8 /2 7 /0 3 15:37 0.51 Low 8 /2 7 /0 3 21:44 1.95 High 8 /2 8 /0 3 4:32 -0.14 Low 8 /2 8 /0 3 10:50 1.5 High 8 /2 8 /0 3 16:19 0.43 Low 8 /2 8 /0 3 22:24 1.88 High 8 /2 9 /0 3 5:02 -0.05 Low 8 /2 9 /0 3 11:20 1.57 High 8 /2 9 /0 3 17:05 0.37 Low 8 /2 9 /0 3 23:08 1.73 High 9 /2 /0 3 2:21 1.09 High 9 /2 /0 3 7:22 0.67 Low 9 /2 /0 3 14:10 1.67 High 9 /2 /0 3 21:50 0.26 Low 9 /3 /0 3 4:33 0.99 High 9 /3 /0 3 8:22 0.84 Low 9 /3 /0 3 15:23 1.67 High 9 /3 /0 3 23:25 0.15 Low 9 /4 /0 3 6:40 1.06 High 9 /4 /0 3 10:08 0.95 Low 9 /4 /0 3 16:47 1.7 High 9 /5 /0 3 0:38 0 Low 503 Huntington Beach 8/20/03 504 Newport Beach 8/28/03 505 Huntington Beach 9/3/03 329 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table C-7. Tide summary. Tidal range while sampling and the maximum tidal range over between the day before and day after sampling. Tidal Range Whk 3-day Max Sampling Day________Stations_________________ Location__________ Sampling (m)_________ Tidal Range Cm) 9/11/01 Anaheim Bay 1,2,3 0.60 1.55 9/13/01 Huntington Beach 4,5,6,7 0.49 2.12 9/20/01 Anaheim Bay 8,9,10 1.49 1.74 9/25/01 Belmont Shore 101 0.51 1.44 10/16/01 Huntington Beach 102 1.99 2.10 11/1/01 Hermosa Beach 11,12,13 1.80 1.85 1 /9 /0 2 Hermosa Beach 103 1.22 2.15 4 /3 /0 2 Huntington Beach 21,22,23,24 0.90 1.50 6 /5 /0 2 Sunset Beach 201,25,26,27 0.71 1.47 6 /7 /0 2 Huntington Beach 202,28,29,30,31 0.58 1.88 6 /1 2 /0 2 Huntington Beach 203,32,33,34,35 0.32 2.16 6 /2 1 /0 2 Hermosa Beach 204,36,37,38,39 0.68 2.23 6 /2 2 /0 2 Santa Monica 205,40,41,42 0.62 2.38 7 /2 6 /0 2 Nearshore SPBay 301,302,303,43,44,45,46 0.54 2.09 8 /1 9 /0 2 Huntington Beach 401,47,48,49,50 0.49 2.09 8 /2 3 /0 2 Huntington Beach 402,51,52,53,54 0.82 2.01 9 /6 /0 2 Hermosa Beach 403 1.16 2.29 9 /8 /0 2 Hermosa Beach 404,55,56,57,58 1.52 2.18 6 /1 8 /0 3 Huntington Beach 501,60,61,62,63 1.42 2.11 6 /2 5 /0 3 Huntington Beach 502, 64,65,66 0.37 1.73 8 /2 0 /0 3 Huntington Beach 503 0.13 1.74 8 /2 8 /0 3 Newport Beach 504 1.07 2.14 9 /3 /0 3 Huntington Beach 505 0.83 1.70 Los Anqeles Average 1.15

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Numerical simulation of whole core squeezer radon pore water profiles: Methodological considerations and evaluation of benthic fluxes and rates of bio-irrigation and advection

PDF

Integrated geochemical and hydrodynamic modeling of San Diego Bay, California

PDF

Fractionation of nitrogen isotopes during early diagenesis

PDF

Natural decay series isotopes in surface waters, bottom waters, and plankton from the East Pacific

PDF

Geochemical studies of beryllium isotopes in marine and continental natural water systems

PDF

Paleoenvironments and paleoecology of the disaster taxon Lingula in the aftermath of the end-Permian mass extinction: Evidence from the Dinwoody Formation (Griesbachian) of southwestern Montana...

PDF

Numerical modeling of long period events at the Piton de la Fournaise by the use of object-oriented scientific computing

PDF

Oxygen isotope constraints on the petrogenesis of the Sybille intrusion of the Proterozoic Laramie Anorthosite Complex

PDF

Oxygen isotopic evidence for fluid infiltration in the Mount Stuart batholith, Washington

PDF

Paleoecology and depositional paleoenvironments of Pleistocene nearshore deposits, Las Animas, Baja California Sur, Mexico

PDF

Estuarine sediment transport and Holocene depositional history, Upper Chesapeake Bay, Maryland

PDF

Stable isotopes in live benthic foraminifera from the Southern California borderland

PDF

Multi-proxy studies of climate variability in central China: Subdecadal to centennial records in stalagmite from Budda Cave

PDF

The origin of enigmatic sedimentary structures in the Neoproterozoic Noonday dolomite, Death Valley, California: A paleoenvironmental, petrographic, and geochemical investigation

PDF

Using small events to measure the evolution of the strain and stress fields before the 1992 Landers, California earthquake

PDF

A proxy for reconstructing histories of carbon oxidation in the Northeast Pacific using the carbon isotopic composition of benthic foraminifera

PDF

Silicic and germanic acids: Laboratory determination of their molecular diffusivities and field study of their benthic fluxes along the California margin

PDF

Carbon isotopic heterogeneity and limited oxygen isotopic homogeneity in the cretaceous Cucamonga terrane, southeastern San Gabriel Mountains, California

PDF

Geochemical and sulfur isotope study of Red Sea geothermal systems

PDF

Observation on the long-period variability of the Gulf Stream, downstream of Cape Hatteras

Asset Metadata

Creator Colbert, Steven Laurence (author)

Core Title Radium isotopes in San Pedro Bay, California: Constraint on inputs and use of nearshore distribution to compute horizontal eddy diffusion rates

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag geochemistry,OAI-PMH Harvest,physical oceanography

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-569455

Unique identifier UC11335435

Identifier 3155396.pdf (filename),usctheses-c16-569455 (legacy record id)

Legacy Identifier 3155396.pdf

Dmrecord 569455

Document Type Dissertation

Rights Colbert, Steven Laurence

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA