Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Fourier grain-shape analysis of quartz sand from the eastern and central Santa Barbara littoral cell, Southern California
(USC Thesis Other)
Fourier grain-shape analysis of quartz sand from the eastern and central Santa Barbara littoral cell, Southern California
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
FOURIER GRAIN-SHAPE ANALYSIS OF QUARTZ SAND FROM THE EASTERN AND CENTRAL SANTA BARBARA LITTORAL CELL, SOUTHERN CALIFORNIA, by Rory Anthony Robinson A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (Geological Sciences) May 1993 Copyright 1993 Rory Anthony Robinson UMI Number: EP58831 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMI Dissertation PublisNng UMI EP58831 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90007 This thesis, written by R ory A n th o n y R o b in s o n under the direction of h.i s Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillm ent of the requirements fo r the degree of M a ste r “ “ i c a l S c i e n c e s ) Dean Date . M . a.r. c. h J . !J 9 9 3 THESIS COMMITTEE Chairman CONTENTS Page FIGURES v TABLES vii ACKNOWLEDGMENTS viii ABSTRACT ix INTRODUCTION 1 Purpose 1 Location 1 Previous Work 3 REGIONAL SETTING 6 Regional Geography 6 Regional Geology 9 Stratigraphy and Geologic History 9 Areal Geology of the Drainage Basins 13 Regional Oceanography 16 Current Sediment Budget Analysis 20 METHODOLOGY 28 Introduction 28 Sample Collection 31 Sample Preparation 38 ii Page Fourier Methodology 39 Data Processing 43 STATISTICAL PROCEDURES 45 Introduction 45 One-way Analysis of Variance 50 Duncan’s New Multiple Range Test 51 Factor Analysis 53 Hotelling’s V Test 54 Discriminant Function Analysis 55 RESULTS 56 Sample Comparison Using ANOVA 58 Sample Relationships from Factor Analysis 61 Source Sample Comparison Using Hotelling’s V Test 70 Chi-Square Analysis of Shape Frequency Distribution 71 DISCUSSION 74 Sand Sources for the Study Area 77 Boundaries of Potential Littoral Subcells 84 Character of Sediment Transport Pathways 85 Temporal Changes in Grain-Shape Composition 86 Implications for Computation of Sand Budgets 91 iii Page CONCLUSIONS REFERENCES APPENDICES Appendix A: Appendix B Appendix C Appendix D Appendix E : : Listing of names and code numbers for the samples used in this study. Sample location and time code numbers are defined in Table 2. : Listing of the VAX/VM S C computer program BYWAVEIT written by Rory Robinson. : Listing of the VAXA/MS C computer program CONVRTIT written by Rory Robinson. : Listing of the VAXA/MS C computer program STATSIT written by Rory Robinson. : Listing of the VAXA/MS C computer program RECASTIT written by Rory Robinson. 96 99 105 106 110 121 135 141 iv FIGURES Figure 1. Location of principal study area and total extent of sampling with reference to nearby cities and landforms. Circled numbers represent major highways. 2. Geomorphologic features of the central and eastern Santa Barbara Littoral Cell. Shelf outline is at -200 meters (MLLW). 3. Surficial sediment transport pathways in the nearshore region of the mainland shelf, of the Santa Barbara Littoral Cell (Kolpack, 1986). 4. Surface water circulation in the Santa Barbara Channel (Kolpack, 1986). 5. Schematic diagram showing the conceptual sediment sources and sinks as well as idealized transport pathways for beach sand in a defined coastal reach (Osborne and Yeh, 1991). 6. Subcells of the Santa Barbara Littoral Cell as defined by Noble Consultants (1989). The zero shoreline mile is located at the Santa Barbara-San Luis Obispo County line. 7. Regional sample map for the Santa Barbara Littoral Cell. Bathymetry is in meters. Flatchered area is shown in Figure 8. 8. Sample map for the Ventura area in the southeastern portion of the Santa Barbara Littoral Cell. Bathymetry is in meters. 9. Generalized flow chart for statistical processing of Fourier amplitude values. 10. Solution from factor analysis of the 61 samples used in this study. Sample numbers refer to the computer codes outlined in Appendix A. Circled samples are end-members for factor one. Doubly- circled samples are end-members for factor two. Page 2 7 17 18 21 23 33 35 46 66 v Figure 11. Scanning electron photomicrograph of selected quartz grains from three samples. The elongate and rough quartz grains in A and B are from granodioritic source rock exposed along Sand Canyon Creek. Photomicrographs C and D are less elongate quartz grains reworked from sedimentary strata exposed along Santa Paula Creek and Ventura River, respectively. The grain in photomicrograph C is more rough than the grain illustrated in D. Scale bar equals 200 microns. 12. Relationships between and among various sedimentary sources as determined from the factor scores. 13. Relationships between and among various beach samples as determined from the factor scores. Plot includes all 0 m MLLW samples collected from June, 1985 to February, 1992. 14. Relationships between and among upper and lower shoreface samples as determined from the factor scores. Samples were collected in June, 1985. 15. Relationships between and among samples collected in October, 1985 and July, 1990 from the Ventura Harbor and samples collected from June, 1985 to March, 1992 as determined from the factor scores. 16. Relationships between and among five beaches collected in June, 1991, October, 1991 and February, 1992 as determined from the factor scores. These beaches are Emma Woods, Surfer’s Point, Marina Park, Hollywood and Ormond. TABLES Table 1. Generalized stratigraphic column for the Santa Ynez Mountains. 2. Sample sets used in the present study. 3. Beach nomenclature used in this study. 4. Process algorithm for recording individual maximum projection outlines of quartz grains from dry grain mounts. 5. Sample Input files for BMDP. A) Sample file for ANOVA and Duncan’s New Multiple Range test. B) Sample file for the factor analysis. C) Sample file for the Hotelling’s T-squared test. D) Sample file for discriminant function analysis. 6. The design for the one-way Analysis of Variance as used in this study (Davis, 1986). 7. Results of significance testing for spatial variation in sets of beach samples collected at different times. Probabilities marked with double asterisks are very significant, whereas single asterisks are significant. 8. Results of significance testing for temporal consistency of shape composition in sets of beach samples at four non-source beaches. Probabilities marked by a single asterisk are significant. 9. Factor scores solution for factors one and two as determined from analysis of the 61 used samples in this study. 10. Samples tested using the Hotelling’s T2 test. 11. Results of chi square testing of individual amplitude values. All shaded squares indicate a significant difference at the five percent level. Page 10 32 37 41 49 52 60 62 64 70 73 vii ACKNOWLEDGMENTS I would like to acknowledge all the professors of the Department of Geological Sciences who furthered my knowledge in this field. I would like to especially thank Dr. Robert Osborne, who as my primary advisor, sponsored me on this project. I would also like to thank the other members of my Committee, namely Dr. Donn Gorsline and Dr. Bernard Pipkin, for their time and effort in improving this manuscript. I also would like to thank Jinyou Liu and Tim Fogarty for their assistance. Lastly, I would like to thank my younger bother Jim Robinson for his help in sample processing. This project was supported in part by a grant from the Graduate Student Research Fund provided by the Department of Geological Sciences. viii ABSTRACT The two-dimensional shape composition of detrital quartz grains from the medium sand fraction of 61 samples collected from the eastern and central Santa Barbara Littoral Cell was determined using Fourier analysis. Nine samples sets were obtained by various organizations from the shoreface (0 to -9 m MLLW) of beaches, rivers, coastal cliffs and dunes as well as the inner continental shelf (-12 to -17 m MLLW) between June, 1969 and March, 1992. The Ventura and Santa Clara River samples are statistically homogeneous with respect to quartz grain shape. Littoral and inner shelf samples from the Ventura area are composed principally of resedimented sand from the Ventura and Santa Clara Rivers, but also contain sand from an unidentified source exposed on the adjacent inner shelf. No statistically significant change in the grain-shape composition of upper shoreface (0 m MLLW) samples in the Ventura area occurs within the observed sample sets, which represent a period of 23 years. Seasonal variation within the Ventura area is marked by an increase in the abundance of more elongate grains on the upper shoreface during the summer months, and a decrease in the abundance of such grains during the oceanographic winter. The increase in more elongate grains during the summer most likely reflects the onshore transport of shelf-derived grains through ridge and runnel systems, whereas the decrease in the abundance of such grains reflects mixing with pre-existing backshore sand containing more equant grains during winter storms. The selective transport of more elongate grains is evidenced by the enrichment of such grains within the Ventura Harbor (-7 m MLLW). x INTRODUCTION Purpose This study examines the southeastern portion of the Santa Barbara Littoral Cell in south central California for the purposes of identifying local sand sources, identifying the boundaries of potential littoral subcells, delineating the character of sediment transport pathways between adjacent subcells, and describing temporal changes of these patterns. These issues were investigated using Fourier grain-shape analysis. Although certain aspects of the Santa Barbara Littoral Cell have been studied by numerous workers (Drake and others, 1972; Bloom, 1979; Kolpack, 1986; and Noble Consultants, 1989), research concerning the short- and long-term variation of the shape composition for medium quartz sand grains has not yet been performed. Such study should lead to a more complete understanding of the natural processes that affect the stability of the beaches within this littoral cell. Location The principal study area extends from approximately 4 km northwest of the mouth of the Ventura River to a position 5 km south to the mouth of the Santa Clara River (Fig. 1). Sampling also was performed throughout the central and eastern parts of the Santa Barbara Littoral Cell, which extends from Point Conception to Point Mugu and includes approximately 150 km of 1 ARROYO BURRO CREEK ( j o T ) - SANTA BARBARA OJAI SANTA CLARA R, VENTURA R. I 26 COAL OIL PT. KM SANTA BARBARA CHANNEL PT. HUENEME ANACAPA IS. PT. MUGU SAN TA CRUZ ISLAND Figure 1 . Location of principal study area and total extent of sampling with reference to nearby cities and landforms. Circled numbers represent major highways. coast line in both Santa Barbara and Ventura Counties (Emery, 1960). This region is easily accessible via well maintained highways and roads. The most traveled of these roads is Interstate Highway 101, commonly known as the Ventura Freeway, which traverses the entire study area. Previous Work Santa Barbara and Ventura Counties are rich in both onshore and offshore petroleum reserves; therefore, the geology of the area has been well studied. The first geological reconnaissance of the area was performed for the War Department by the Parke expedition (1857). Arnold (1907) mapped the geology and oil resources of the Summerland District around Santa Barbara. Kew (1924) produced a regional geologic map of part of Los Angeles and Ventura Counties. Putnam (1942) described the geomorphology of the Ventura region. In 1954, the California Division of Mines and Geology described the regional geology of the Ventura area. Winterer and Durham (1962) assembled a geologic map of the Ventura Basin, and Dibblee (1966) prepared a regional geologic map of the Santa Ynez Mountains. In 1973, a geologic map (1:48,000) of Ventura County was compiled by the California Division of Mines an Geology. Yeats (1976) described Neogene tectonics in the Ventura Basin. Jackson and Yeats (1982) summarize the structural evolution of the Carpinteria basin. Yerkes and others (1987) described the Quaternary deformation of the Ventura area. Recent papers by Nilsen (1984 3 and 1987) have attempted to decipher the tectonic significance of the Transverse Ranges. From 1987 to 1992, the Dibblee Geological Foundation has published the geology for most of the 7.5 minute quadrangles within the study area. The marine geology and oceanography of the region has been equally well studied. The following works are pertinent to this study. Drake and others (1972) described the sediment transport patterns within the study region following a major flood. Fischer and others (1972) described the Quaternary shelf deposits within the study area and throughout southern California. In 1975, the California Division of Mines and Geology published a geologic map of the offshore geology of California. Rice and others (1976) described the relationship between river and beach sand from Point Conception to the United States-Mexico border. Smith (1978) described the grain-surface microtexture of sediments from the Santa Clara River and Santa Barbara Littoral Cell. Taylor (1981) delineated the major sedimentologic processes of the southeastern portion of the study region. Kolpack (1985) described surface flow patterns within the Santa Barbara Channel. An unpublished U.S. Army Corps report dated January, 1985, describes beach sand characteristics in and around the Ventura Harbor. A U.S. Army Corps of Engineers report dated June, 1989, entitled, "Feasibility Report - Ventura Harbor," summarizes U.S. Army Corps of Engineers data concerning the 4 Santa Barbara Littoral Cell. Dahlen and others (1990) describe the late Quaternary geologic history of the Ventura Shelf. Pipkin and Proctor (1992) described the nature and rates of coastal erosion in southern California. Bloom (1979) used Fourier grain-shape analysis to study the sediment patterns within and south of the Santa Clara River to Hueneme Canyon. Noble Consultants Inc. (1989) prepared a report for the Beach Erosion Authority for Control Operations and Nourishment (BEACON). This report, named the BEACON study, presents a comprehensive sand management plan for the central and eastern parts of the Santa Barbara Littoral Cell. 5 REGIONAL SETTING Regional Geography Regional geography can be separated into three subjects of interest. These are the nearshore subaerial geomorphology, the nearshore subaqueous geomorphology, and the hydrologic character of major river systems in the study area. The nearshore subaerial geomorphology is dominated by the western extension of California’s east-trending Transverse Ranges, known locally as the Santa Ynez Mountains. From Point Conception east to the Ventura River, these mountains parallel the coast and create a rugged landscape with little or no coastal plain. The Santa Ynez Mountains attain elevations exceeding 2,400 m above sea level (Dibblee, 1966). The area south of the Ventura River to Point Mugu occurs along the Pacific border of the Oxnard Plain. The Oxnard Plain is a broad, gently-sloping alluvial deposit located between the Santa Ynez Mountains on the north and the Santa Monica Mountains to the southeast (Fig. 2) (Yeats, 1976). The Santa Monica Mountains trend east like the Santa Ynez Mountains, but have only moderate relief achieving elevations up to 1,000 m above sea level (Kew, 1924). The nearshore subaqueous geomorphology of the study area is dominated by the submarine expression of the Oxnard Plain, known as the 6 SANTA YNEZ MOUNTAINS KM SANTA BARBARA MARINA VENTURA, R VENTURA SHELF 34 ° 15 VENTURA HARBOR CHANNEL ISLAND HARBOR PORT HUENEME MUGU SHELF |J r f 11 f n ; r ANACAPA IS. • 9\\- Figure 2. Geomorphologic features of the centra! and eastern Santa Barbara Littoral Cell. Shelf outline is at -200 meters (MLLW). Ventura shelf. The Ventura Shelf extends from approximately 6 to 15 km southwest of the shoreline to the Santa Barbara Channel and is a gently- sloping feature with depths to 100 m (Kolpack, 1986; Dahlen and others, 1990). The Santa Barbara Channel trends east and attains depths which exceed 200 m. The Ventura shelf narrows toward the east and west and is broken on the southeast by Hueneme Canyon, which trends due south of Point Hueneme. The small shelf between Hueneme Canyon and Mugu Canyon is named the Mugu shelf as is the eastern extension of the Ventura Shelf, which is now isolated by the two submarine canyons (Kolpack, 1986). River drainage basins within the study area can be separated into three separate watersheds (Brownlie and Taylor, 1981). These are the Santa Clara River system, the Ventura River system, and the Santa Ynez Mountain Front. The Ventura and Santa Clara Rivers occur along the northern part of the Oxnard Plain. These are the only major rivers in the study area known to contribute significant volumes of sand to local beaches (Kolpack, 1986; Noble Consultants, 1989). The Ventura River drainage area is 585 km2 with a total of 243 km2 (41.5 %) controlled by the Matilija and Lake Casitas reservoirs (Brownlie and Taylor, 1981). The Santa Clara River drainage area is 4,219 km2 with a total of 1,539 km2 (36.5 %) controlled by the Bouquet, Lake Piru, Lake Pyramid, and Lake Castaic reservoirs (Brownlie and Taylor, 1981). These two rivers are distinctive in that they have the highest sediment 8 discharge in southern California, together averaging approximately 1.4 x 106 m3 /yr (Brownlie and Taylor, 1981). Further, much of the drainage areas for these rivers are not controlled and flood during times of heavy precipitation (Brownlie and Taylor, 1981; U.S. Army Corps of Engineers, 1989). The Santa Ynez Mountain Front river system consists of several short and steep mountain streams and drains a total area of approximately 1,080 km2 . These streams extend from the coastal divide to the shoreline and range from approximately 10 to 20 km in length. Less than 10 percent of these mountain streams are controlled by debris basins (Brownlie and Taylor, 1981). Regional Geology This section is divided into two parts. The first is an overview of the stratigraphy and geologic history of the region and the second is a description of the areal geology within the Santa Clara, Ventura, and Santa Ynez Mountain Front River drainage basins. Stratigraphy and Geologic History Table 1 is a generalized stratigraphic column for the study region. The oldest rocks within the western Transverse Ranges are late Cretaceous and belong to the Franciscan Formation (Norris and Webb, 1990). The Franciscan Formation consists of a mixed assemblage of deep-water marine strata which represent a former accretionary wedge. This accretionary wedge was part of 9 Table 1. Generalized stratigraphic column for the Santa Ynez Mountains. Epoch Formations Dominant Lithology Holocene Un-named Alluvium and Terrace (marine and non-marine) Clay, Silt, Sand and Gravel Saugus (non-marine) Sandstone Pleistocene Los Posas (marine) Sandstone Santa Barbara (marine) Sandstone Pliocene Pico (marine) Shale Repetto (marine) Shale Sisquoc (marine) Shale Monterey (marine) Shale and Sandstone Miocene Rincon (marine) Shale Blanca (marine) Shale Vaqueros (marine) Sandstone Gaviota (marine) Sandstone Oligocene Sespe (non-marine) Sandstone and Conglomerate Cold water (marine) Sandstone Sacate (marine) Shale Eocene Cozy Dell Matilija (marine) Shale (marine) Sandstone Anita (marine) Shale Juncal (marine) Shale Paleocene Pozo (marine) Shale Jalama (marine) Sandstone and Shale Cretaceous Franciscan (marine) Accretionary Complex 10 a relatively slow and highly-oblique subduction zone at the western edge of a north-trending volcanic arc (Nilsen, 1987). Latest Cretaceous marked the beginning of the Laramide orogeny which changed the character of subduction in the region. The angle of the subducting plate is thought to have become more shallow causing volcanism to move eastward, away from the existing volcanic arc (Yeats, 1987). The Jalama and Pozo Formations were deposited during this time (Dibblee, 1982). These rocks represent deep-water marine strata deposited along the edge of the allochthonous terranes of the Transverse, Salinian, and Peninsular blocks that existed southwestward of the volcanic arc (Nilsen, 1987). By late Paleocene, these allochthonous terranes and their associated peripheral sediments were sutured to the edge of the continent. The Juncal, Anita, Matilija, Cozy Dell, Sacate, and Coldwater Formations were deposited after this accretionary event in the newly created Santa Ynez basin (Norris and Webb, 1990). This basin in thought to have been a pronounced north- trending depression on the west side of the sutured terranes. The source for the Eocene sediments is believed to have been from the Mojave and Peninsular blocks, which due to thrusting from the newly accreted terranes were topographically high (Nilsen, 1987). The Santa Ynez basin was uplifted by the start of the Oligocene, due to the approach of the Pacific-Farallon spreading ridge (Nilsen, 1984). Fluvial 11 sedimentation, predominantly of strata assigned to the Sespe Formation was dominant within the Santa Ynez basin during early and middle Oligocene time. The Oligocene Santa Ynez basin received sediment from northern highlands underlain by the Franciscan Formation and eastern highlands underlain by granitoids, possibly belonging to the Mojave block (Nilsen, 1984). With the subduction of the Pacific-Farallon spreading ridge, Miocene time marked the initiation of the San Andreas fault system (Crowell, 1975). This transform fault system in believed to have caused the clockwise rotation of the Transverse Ranges, from an original northerly trend to the present westerly trend (Luyendyk and Flornafius, 1987). This faulting and the subsequent rotation also caused the formation of several relatively small but deep basins, most notably the Ventura basin, where marine strata assigned to the Vaqueros, Blanca, Rincon, Monterey and Sisquoc Formations were deposited (Norris and Webb, 1990). This setting of transform tectonics and small basin sedimentation persisted well into the Pleistocene Epoch. Much of the Ventura basin has received almost continual sedimentation since the late Miocene; however, uplift and rotation along the northern, eastern and southern sides of the Ventura basin has produced local unconformites (Winterer and Durham, 1962). During Pliocene and the early Pleistocene time, the Ventura basin experienced deep-marine deposition of the Repetto and Pico Formations followed by shallow-marine deposition of the Santa Barbara 12 Sand (Yeats and Rockwell, 1991). By the middle to late Pleistocene, areas of the Ventura basin became emergent and fluvial sand of the Saugus Formation was deposited (Dibblee, 1966). Sediment deposited within the basin during Pliocene and Pleistocene time came from granitoid sources to the east and sedimentary sources to the northeast. These sediments were transported in a westward direction into the basin along an ancestral predecessor to the Santa Clara River (Yeats and Rockwell, 1991). During late Pleistocene time, sediment transport into the Ventura basin was diverted by Hueneme and Mugu submarine canyons into the Santa Monica basin (Dahlen and others, 1990). Until less than 1,000 years ago, the Santa Clara River debouched near Port Hueneme, supplying sediment to Hueneme Submarine Canyon and to Santa Monica basin (Yeats and Rockwell, 1991). Presently, the Santa Clara River enters the Pacific Ocean in the Santa Buena Ventura area, and thus is currently supplying the Ventura/Santa Barbara basin. Areal Geology of the Drainage Basins The relative areal percent of dominantly sand bearing rock formations that underlie surface soil within each river and stream drainage basin is presented below. Unless noted otherwise, all of the following areal percentages were determined from 7.5-minute quadrangle geologic map sheets published by the Dibblee Geological Foundation (1987 to 1992). All 13 percentages were determined by point counts along an arbitrary grid system and are rounded to the nearest whole number in increments of five. The Santa Clara River system consists of the main river, which trends broadly eastward, and six major north-trending tributaries (Santa Paula, Sespe, Hopper, Piru, Castaic and Bouquet Creeks). Of the six main tributaries, Piru and Castaic Creeks are dammed near the point where they merge with the Santa Clara River, and thus can be discounted as major sediment sources. Santa Paula Creek drains only Tertiary sedimentary rocks, of which the most extensive sand-contributing rocks are the Matilija (25%), Los Posas (10%), Coldwater (10%), and Saugus (5%) Formations. Sespe Creek, like Santa Paula Creek, drains only Tertiary sedimentary rocks. Of the rocks in Sespe Creek, the most areally extensive sand-producing formations are the Sespe (25%), Coldwater (10%), Matilija (10%), Juncal (<5%), Jalama (<5%), and Vaqueros (<5%) Formations. Sand in Hopper Canyon Creek is derived mostly (45%) from the sandstone member of the Monterey Formation with only minor areas (5%) underlain by sand derived from the Pico and Saugus Formations. The short segments of Piru and Castaic Creeks that drain into the Santa Clara River may receive sand from either the Saugus (30%), Monterey (10%), or the Pico (5%) Formations. Bouquet Creek as well as the Santa Clara River east of the junction with Bouquet Creek comprise the eastern watershed of the Santa Clara River and drain roughly equal areas of 14 crystalline and sedimentary bedrock. All of the following areal percentages for the eastern watershed are from Smith (1978). Within the eastern watershed, 12 percent of the total area is underlain by Tertiary marine rocks and 38 percent of the total area is underlain by Tertiary non-marine rocks, predominantly that of the Sespe, Saugus and Mint Canyon Formations. Crystalline terranes include 18 percent composed of granitoids and 23 percent underlain by schistose to gneissic metamorphic rocks. Approximately 8 percent of the total area in the eastern watershed in underlain by anorthosite and less than 1 percent is underlain by volcanics. The Ventura River system consists of the main north-trending river and several smaller branching creeks known as Coyote, Santa Ana, Matilija and San Antonio/Reeves Creeks. The whole Ventura watershed drains Tertiary sedimentary rocks. However, Coyote, Santa Ana and Matilija Creeks are controlled by reservoirs, and thus can not contribute sand to the main river. The remaining creeks predominantly drain the eastern side of the basin, adjacent to the Santa Clara watershed. Thus, in the available watershed, the most areally extensive sand-producing formations are the Sespe (20%), Coldwater (10%), Matilija (5%), Jalama (5%), Vaqueros (<5%), and Saugus (<5%). The Santa Ynez Mountain Front, from Coal Oil Point to the Ventura River, consists of numerous small streams that exclusively drain Tertiary sedimentary 15 rocks. The areal percentages for this system represent regional averages, and are not the composition of any one stream. The most areally extensive sand-producing formations are the Sespe (35%), Coldwater (20%), Matilija (10%), Santa Barbara (5%), Vaqueros (<5%), and Jalama (<5%). Regional Oceanography The southern California coast is characterized by an overall southerly or easterly longshore flow (Fig. 3) (Emery, 1960). Typically longshore flow is limited to the shoreface which in the study area typically ranges from 0 to 9 m below Mean Lower Low Water (MLLW) with a transition to the offshore environment located between 9 and 18 m (Howard and Reineck, 1981). Surface currents within the Santa Barbara Channel are characterized by a southeasterly flow along the north side of the Channel Islands (California Current) with a strong northwesterly flow along the inner to middle part of the Ventura Shelf (Anacapa Current) (Fig. 4). This reversal of flow is caused by an eddy generated from the topography of the Channel Islands (Drake and others, 1972). These flow patterns are influenced by the four wind, wave, and swell patterns that affect the central and eastern portions of the Santa Barbara Littoral Cell. These patterns include the Eastern Pacific High, Eastern Pacific Low, Tropical Cyclones, and Southern Hemisphere Low (Noble Consultants, 1989). 16 34 40 ■SANTA YNEZ RIVER EL PT. CONCEPTION SANTA BARBARA 34 2 0 S. CLAR. SEDIMENT TRANSPORT ■ C I H U I K m 0 10 20 L M ' : 119° 30 120 00 ' 120 30' F ig u r e 3. S u r fic ia l s e d i m e n t t r a n s p o r t p a t h w a y s in t h e n e a r s h o r e r e g i o n o f t h e m a i n l a n d s h e l f , o f t h e S a n t a B a r b a r a Littorial C e ll ( K o l p a c k , 1 9 8 6 ) . SURFACE CURRENTS km 20 CAP ITA N SANTA • B A R B A R A VENTURA 34 ANACAPA 0 0' SANTA CRUZ SANTA ROSA F ig u r e 4 . S u r f a c e w a t e r c ir c u la t io n in t h e S a n t a B a r b a r a C h a n n e l ( K o l p a c k , 1 9 8 6 ) . The Eastern Pacific High (EPH) is responsible for the prevailing easterly- directed waves throughout much of the year, especially in the late spring, summer and early fall. During the late spring and early summer, the center of the this weather system is located several hundred kilometers west of the study area, and when combined with low pressure systems over Nevada the resulting winds can at rare instances exceed 80 km per hour (U.S. Army Corps of Engineers, 1989). During late spring and early fall, the focus of this system shifts northward, and due to its more northern position, affects the study area less frequently. At times, very high pressure builds over Nevada which causes a reverse of this normal flow resulting in strong northeasterly winds, known locally as the Santa Ana Winds (U.S. Army Corps of Engineers, 1989). The Eastern Pacific Low (EPL) is the storm pattern which creates the largest waves and typically occurs from November to April. These storms are directed west toward the study area by cyclogenesis (the counterclockwise rotation of winds around a low pressure system) centered in the mid-Pacific. The area is affected by strong east and southeasterly winds prior to storm fronts. The storms themselves are accompanied by easterly-directed waves resulting from strong west to northwesterly winds. The waves created by the EPL may be intensified by high river discharge during these storms. This 19 weather condition is responsible for the transport of large amounts of sediment within a short time (U.S. Army Corps of Engineers, 1989). The Tropical Cyclones (TC) storm pattern develops off the coast of Mexico during the summer months. If these systems approach southern California they rapidly dissipate as they travel over the cold water off the California coast. Some of the waves generated by these storms do enter the study area, but the sedimentologic impact is thought to be minor (U.S. Army Corps of Engineers, 1989). The Southern Hemisphere Low (SHL) storm pattern is caused by large low pressure systems moving eastward across the South Pacific during the southern hemisphere winter from May to October. The very large waves created by these storms can travel across the Pacific and break against beaches in the study area. Usually these waves are 1 to 1.5 m high; however, these waves can exceed 3 m (U.S. Army Corps of Engineers, 1989). Current Sediment Budget Analysis Various sediment budgets have been computed for the study area (U.S Army Corps of Engineers, 1989). Sedimentary budgets attempt to delineate areas from which beach sands were derived (sources), the direction and rates of sand transport (transport pathways) and areas where beach sands are ultimately deposited (sinks) (Fig. 5). Sources are typically considered to include river sediment discharge, cliff or beach erosion, offshore wind 20 River BlufT& Wind & Harbor Beach Dredging & Sediment Dune Overwash Sediment Nour- Sand Mining Discharge Erosion Transport Sink ishment Losses o o x: a < / ! V i O f i C c at I t I t I COASTAL REACH: ♦ * Oflshore/Onshore Sediment Transport Submarine Canyon Transport £ t .c S i F ig u r e 5. S c h e m a t i c d i a g r a m s h o w i n g t h e c o n c e p t u a l s e d i m e n t s o u r c e s a n d s i n k s a s w e ll a s i d e a l i z e d t r a n s p o r t p a t h w a y s for b e a c h s a n d in a d e f i n e d c o a s t a l r e a c h ( O s b o r n e a n d Y e h , 1 9 9 1 ) . transport, beach nourishment activities, onshore transport from the inner shelf, and longshore transport into an area. Sand sinks include overwash transport, harbor sedimentation, dredging and mining operations, longshore transport out of an area, offshore transport to the inner shelf, and transport down submarine canyons (U.S. Army Corps of Engineers, 1989). If the above factors can be accurately estimated for a given coastal reach, predictions can be made concerning the future state of the shoreline. The most recent and comprehensive sediment budget analysis for the present study area was performed for Beach Erosion Authority for Control Operations and Nourishment (BEACON) by Noble Consultants, Incorporated (1989). This study separated the Santa Barbara Littoral Cell into seven subcells (Fig. 6). These subcells are not smaller isolated coastal reaches within the larger littoral cell; but rather, Noble Consultants (1989) uses these subcells to perform coastal calculations and devise maintenance strategies for the region based on manageable segments. In short, these coastal divisions are based on necessity or convenience rather than physical reality. The location, basis and character of each subcell is discussed below. The first subcell extends from Point Conception to Santa Barbara Harbor. The relatively large region that this cell encompasses is rural and undeveloped with a very poor sedimentologic data base. The downcoast end of this cell was chosen, because it coincides with a man-made harbor that traps littoral 22 o o U I LU C O C D U I U I O c o C O c o e g h -U J U I O C J O SUB CELL I (TO POINT C O N CEPTIO N ) SUBCELL 2 SUBCELL 3 u i o 0 0 LU O C Q = 3 C O UI o CD 60' 100 — I — 1 2 0 I 140 SHORELINE MILES Figure 6. Subcells of the Santa Barbara Littorial Cell as defined by Noble Consultants (1989). The zero shoreline mile is located at the Santa Barbara-San Luis Obispo County line. 23 sand, and thus provides documented sand volumes based on dredging activity. The second and third subcells extend from Santa Barbara Harbor to the mouth of the Ventura River, and may be considered as a single coastal reach. However, for descriptive purposes, Noble Consultants (1989) choose to separate the two cells at Rincon Point. Subcell four is relatively small and occurs between the mouth of the Ventura River and Ventura Harbor. The definition of this subcell was based on work performed by the U.S. Army Corps of Engineers, which suggested that this coastal reach was dominated by fluvial sediment derived from the Ventura River. The downcoast end of the cell (Ventura Harbor) also was chosen for volume control, based on dredging records. The fifth subcell extends from the Ventura Harbor to Channel Island Harbor. This cell has good volume control and includes the Santa Clara River, which is thought to be the dominant sediment source for this coastal segment (U.S. Army Corps of Engineers, 1989). The sixth and smallest subcell occurs between Channel Island and Port Hueneme Harbors and is artificially replenished. The seventh subcell lies between Port Hueneme Harbor and Mugu Submarine Canyon. This section of beach is unique in that there are no major sand sources known other than longshore transport from the west and possible shelf input. The eastern boundary of the Santa Barbara Littoral cell is Mugu Submarine Canyon. 24 Based on volumetric estimations Noble Consultants (1989) calculated that the total volume of sand in the first subcell is composed of approximately 31 percent which "leak" around Point Conception, 9 percent of which is added from cliff erosion, and 60 percent of which is fluvial discharge from small mountain streams. It is assumed that the net sand transport out of the first subcell equals the sand bypassing rate at Santa Barbara Harbor, which is 230.000 m3 /year and is placed on East Beach. After passing the Santa Barbara Harbor, sand travels east through the second subcell, with a total of 64.000 m3 of sand per year added to the littoral flow by either cliff or beach erosion and by fluvial discharge from mountain streams. Of the 294,000 m3 of sand per year that enters the third subcell, 46,000 m3 /year is lost to the very near shelf. The fourth subcell gains approximately 214,000 m3 /year of sand from both the Ventura River sediment discharge and beach erosion around Pierpoint Bay. An average of 490,000 m3 /year of sand is dredged from Ventura Harbor and is thus bypassed to Spinnaker and Santa Clara River Beach. Of this bypassed sand, as much as 76,000 m3 /year may come into the harbor from the east or downcoast. The fifth subcell transports an average of 842,000 m3 /year of sand into Channel Islands Harbor. Additions to this subcell are from the Santa Clara River and the river’s nearshore flood delta. Only minor amounts of sand are lost to the dunes at McGrath State Beach. The sixth subcell looses sand to Port Hueneme Harbor and is 25 artificially replenished at a rate of 50,000 m3 /year. Sand that enters both Channel Island and Port Hueneme Harbors is ultimately bypassed to the northwestern end of subcell seven (Hueneme and Ormond Beach). Sand lost to local beach growth in the northern part of subcell seven is approximately offset by the sand gained from beach erosion around Mugu Lagoon, in the south. Thus, sand exits the seventh and last subcell at a rate of approximately 842,000 m3 /year, of which at least 90 percent is intercepted by Mugu Submarine Canyon. It should be noted that although the study performed by Noble Consultants (1989) is internally consistent, several of the assumptions concerning sand volumes and transport rates have little observational support. These critical assumptions are based on work performed by the U.S. Army Corps of Engineers (1989) and include: 1) net sand transport within the littoral zone is predominantly southeast or downcoast, 2) onshore-offshore transport is related only to storm and seasonal activity, 3) there is no significant movement of sand to or from the inner to middle shelf, 4) there is no net effect from the seasonal movement of sand to and from the shoreface, and 5) there is a conservation of volume. Transport rates presented by Noble Consultants (1989) were calculated using Beach Profile Analysis and checked against dredging records. Transport directions were computed by short-term wave hindcasting. The down canyon sand transport rate for Mugu Submarine 26 Canyon was obtained from Bailard (1985). Cliff and beach erosion were determined by review of historic photographs and maps. Lastly, river discharge estimates are based on a combination of calculated and recorded amounts. 27 METHODOLOGY Introduction Fourier grain shape analysis (FGSA) has been used for numerous sediment source and transport problems since first introduced by Schwarcz and Shane (1969). The following is a description of the FGSA technique as described by Osborne and Yeh (1991). Ehrlich and Weinberg (1970) describe a closed-form Fourier method to analyze the observed variation of two-dimensional, maximum-projection, grain-shape area. Grain shape may be estimated by an expansion of the periphery radius as a function of angle about the grain’s center of gravity by a Fourier series. In Fourier analysis, a series of sine and cosine curves with periods equal to fundamental harmonics is fit to the observed data by a least-square technique. Fundamental harmonics are the prime fractions (1/2, 1/3, 1/4, . . . 1/N), where N equals half the number of digitized points used to define the periphery of a grain. As the number of fundamental harmonics is increased, the computed curve converges with observed data. The highest frequency that can be estimated is the Nyquist frequency, which is equal to twice the distance between the successive observations. If the Nyquist 28 frequency is exceeded, error may be introduced by the incorporation of irresolvable high frequencies into lower frequencies (aliasing). The radius is given by: oo R( 0 ) = Ro + 2 Rn cos( 0 n - PAn) n a I Where theta is the polar angle measured from an arbitrary reference line. The first term in the series Ro is equivalent to the average radius of the grain in the maximum projection orientation. For the reminder of the terms, n is the harmonic order, Rn is the harmonic amplitude, and PAn is the phase angle. The phase angle appears to provide little additional grain-shape information, and therefore is not considered further. It is important to note that the n’th harmonic contributes to the explanation of the observed shape variation as a figure with n "bumps." For example, the "zeroth" harmonic is a centered circle with an area equal to that of the maximum projection; the first harmonic is an off-centered circle; the second is a figure eight; the third is a trefoil; etc. The center of gravity of the maximum-projection shape is used as the origin of the radius expansion to simplify interpretation of the Fourier series. Coordinates of points along the periphery of the maximum- 29 projection outline are required for the Fourier expansion. At least twice the number of such points must be known as the number of the highest desired harmonic. The initial origin of the periphery points may be arbitrary, because a later transformation places the origin at the center of gravity of the maximum grain-projection area. If a harmonic or periodic function exists within the data, the amplitude of the sine and cosine curves with periods close to the natural harmonic will be considerably larger than the amplitudes of other harmonics in the sequence. Although conceptually similar to the closed-form methodology described by Ehrlich and Weinberg (1970), the methodology employed in this study makes use of the newer and more widely used Fast Fourier Transform (FFT). This procedure involves the calculation of many values of the line spectrum using the FFT computer algorithm to produce a smoothed estimate of the continuous spectrum. The FFT algorithm, as its name implies, is extremely rapid and requires only nlog2 n arithmetic operations rather than the n2 operations as do alternative methods. The reader is referred to Brigham (1974), Bloomfield (1976), and Bendat and Piersol (1971) for extensive treatments of the mathematically complex FFT. 30 Sample Collection This study is based on the analysis of 61 sand samples collected from beaches, rivers, dunes, cliffs and inner shelf areas of the central and eastern parts of the Santa Barbara Littoral Cell. Sample sets used in this study were collected from June, 1969, to March, 1992 by several different organizations (Table 2). Table 3 contains a listing of beach nomenclature as used in this study. The locations for each of the samples is shown in Figures 7 and 8. Appendix A lists all names and code numbers for the samples used in this project. A total of 41 foreshore-beach samples, approximately 500 cm3 in volume, were collected at the 0 m MLLW elevation. Three samples from the Ventura River and four from the Santa Clara River were collected from active channels upstream of the areas affected by tides. The Ventura River samples were collected approximately 2 km from the river mouth at an elevation of +9 m (MLLW). Two of the Santa Clara River samples were obtained approximately 6 km from the river mouth at an elevation of +12 m (MLLW). The other two Santa Clara River samples were collected in Santa Paula Creek and south Sand Canyon Creek tributaries. The sample from Santa Paula Creek was collected 1 km upstream of the creek mouth at an elevation of +150 m. The sample from south Sand Canyon Creek was taken at the intersection of Sand Canyon Creek and Ravenhill Road, approximately 6 km from the creek mouth 31 Table 2. Sample sets used in the present study. o 2 c: Sample Sample Sets * * - t — » C O o 3 Names 1 2 3 4 5 6 7 8 9 1 Arroyo Burro Beach X 2 Lead better Beach X 3 East Beach X X X 4 Emma Woods Beach X X X X X 5 Surfer’s Point Beach (0 m) X X X X X 6 Surfer’s Point Beach (-9 m) X 7 Santa Buena Ventura Beach X 8 9 Marina Park Beach (0 m) Marina Park Beach (-9 m) ------- X X X ------ ------ X X X ------- 10 Harbor Channel A X X 11 Harbor Channel B X X 12 Spinnaker Beach (0 m) X X X X 13 Spinnaker Beach (-9 m) X 14 Santa Clara River Beach (0 m) X X X X 15 Santa Clara River Beach (-9 m) X 16 Mandalay Beach X 17 Hollywood Beach X X X 18 19 Hueneme Beach Ormond Beach — ------ —■ X X X X X — 20 Point Mugu Beach X X X 21 Arroyo Burro Cliff X 22 Ventura River X X X 23 Santa Clara River X X 24 Santa Paula Creek X 25 South Sand Canyon Creek X 26 McGrath Dune Field A X 27 McGrath Dune Field B X 28 Ventura Shelf A X 29 Ventura Shelf B X * Sampling Dates and Responsible Organization or Individual 1) June, 1969, Sedimentology Laboratory, University of Southern California. 2) June 4 to 6, 1985, U. S. Army Corps of Engineers. 3) October 22 to 25, 1985, U. S. Army Corps of Engineers. 4) August 30 to September 1, 1988, Noble Consultants. 5) July 31, 1990, U. S. Army Corps of Engineers. 6) April 26 to June 10, 1991, R. Robinson. 7) October 10, 1991, R. Robinson. 8) February 21, 1992, R. Robinson. 9) March 31, 1992, R. Robinson. 32 Figure 7. Regional sample map for the Santa Barbara Littoral Cell. Bathymetry is in meters. Hatchered area is shown in Figure 8. 33 ARROYO ; I BURR 0 / « CR. ^ 7 o SANTA BARBARA CARPINTE Rl A o _ RINCON PT. SANTA BARBARA MARINA SANTA PAULA ■ " 24 PITAS PT -70 = jS<ZZ) ^VENTURA VENTURA HARBOR' 23 EXPLANATION /X O X N A R D CHANNEL ISLAND HARBOR PORT HUENEME 20 Location Number PT. MUGU Km 0 0 Figure 8. Sample map for the Ventura area in the southeastern portion of the Santa Barbara Littoral Cell. Bathymetry is in meters. JO ctr VEN TU R A PIER VENTURA VENTURA HARBOR SANTA CLARA RIVER 2 6 2 8 2 7 2 9 CHANNEL ISLAND HARBOR EXPLANATION PORT HUENEME HUENEME / PIER 27 Location Num ber km 36 at an elevation of +610 m. The collection site for south Sand Canyon Creek is not shown on either Figures 7 to 8 due to its distant location. Two dune grab samples were collected near the crest of the McGrath State Beach dune field at an elevation of approximately +4 m (MLLW). The only cliff sample was obtained from a sandstone lens in a weathered outcrop of Monterey Shale exposed in the bluffs immediately north of Arroyo Burro Beach at an elevation of approximately +5 m (MLLW). The four -7 m (MLLW) harbor samples, the two shelf samples (-12 and -17 m MLLW), and the four -9 m Table 3. Beach nomenclature used in this study. Beach Names Known Used in This Studv Pseudonyms Arroyo Burro Beach — Leadbetter Beach — East Beach Santa Barbara Beach Emma Woods Beach The Under-Pass Surfer’s Point Beach The Fairgrounds, East Ventura River Beach Santa Buena Ventura Beach Ventura Pier Beach Marina Park Beach North Ventura Harbor Beach Spinnaker Beach South Ventura Harbor Beach Santa Clara River Beach Santa Clara River Mouth Bar McGrath Beach The Dunes Mandalay Beach — Hollywood Beach — Hueneme Beach Port Hueneme Beach, Hueneme Pier Beach Ormond Beach The Field Point Mugu Beach The Rock, Mugu Beach 37 (MLLW) samples were obtained by vibracoring, preformed by the Los Angeles District of the U.S. Army Corps of Engineers. The February 21, 1992 sample set was collected ten days after a large oceanic storm which affected the study area. This storm was the fifth such oceanic storm to affect the study area between November 14, 1991 and February 11, 1992. Each of these five oceanic storms had waves that ranged from 2.0 to as great as 3.3 m (Osborne and others, 1992). Sample Preparation The sand for this project was prepared by the method described by Osborne and Yeh (1990): Quartz grains from the 0.25 to 0.50 mm (1.0-2.0 phi) or medium-sand fraction from each sample were used for the FGSA. Samples were first wet-sieved to obtain the desired size fraction to minimize attrition of the quartz grains during sieving. Sieving was followed by drying in convection ovens at 40° C. After drying, all strongly magnetic minerals were removed from the samples using a hand magnet. The samples were then treated in a solution of 10 percent hydrochloric acid with stannous chloride crystals for 20 minutes to remove iron-oxide coatings and carbonate cement (Carver, 1971). The hydrochloric-acid residue was removed by rinsing the samples in de-ionized water. The quartz grains were 38 etched clean by washing in a hydrofluoric-acid bath for 1 minute. Such etching did not significantly alter the shape of the grains (Schultz, 1980). All samples were then rinsed in de-ionized water and dried thoroughly. Samples were examined under a petrographic microscope using a reflecting light source. Quartz grains were picked by using a very small brush and were dry mounted on a glass slide, which enhanced the contrast between the grain and its background (Ehrlich and Weinberg, 1970). Adequate contrast is essential in the operation of the digitizing program. Fourier Methodology Grain boundaries are recorded using the system available in the Sedimentary Petrology Laboratory at the University of Southern California. Hardware specifications consist of a Video-Microscope connected to a Digital Corporation VAX-Station 3200. The Video-Microscope consists of a standard binocular petrographic microscope and a black-and-white video camera connected via a ordinary camera attachment. The VAX-Station 3200 is a stand-alone 32-bit work-station. The unit is based on the KA650 Central Processing Unit typically found on the larger VAX systems. The VAX-Station 3200 comes with a mouse, keyboard and Ethernet Communication Module, which enables digital communication with other systems. The Sedimentary 39 Petrology Laboratory’s VAX is further equipped with 16 million bytes of Random Access Memory, a 650 megabyte internal hard-drive, an external magneto-optic diskette, and a video subsystem module which is connected to a second video screen and the video camera. The VAX-Station presently runs the VMS control-language operating-system. Grain recording and Fourier calculation are performed by the Grain- Shape Analysis (GSA) program which was written by Tim Fogarty, and consists of several complex Fortran routines. The process algorithm for recording individual grain shapes from the dry grain mounts is show in Table 4. A minimum of 200 quartz grain outlines were obtained for each sample. The GSA program determines grain outlines from a standard multi-pixel single frame image "grabbed" from the video camera. With the proper contrast settings the grain will appear dark in relation to the "lighted" background. The boundary algorithm determines the grain outline, by assuming that the maximum contrast in the image occurs along the grain boundary, or the transition from the dark grain (high tone pixels) to the lighter background (low tone pixels). These areas of maximum contrast, high tone pixels which adjoin a low tone pixel, are recorded in an arbitrary Cartesian coordinate (X-Y) system as the boundary points. Further, the center of gravity of the grain outline also is computed. These points are then written to a storage device as a binary data file with a \BPT file type identifier extension. 40 Table 4. Process algorithm for recording individual maximum projection outlines of quartz grains from dry grain mounts. Procedure to Digitize Grains: 1) Check to ensure that the microscope lens is at the correct magnification and the magnification factor is appropriate. 2) Find, center, and focus the microscope on the new grain to digitize. 3) Highlight and then select "Grab Image from Camera." 4) Visually inspect the new image on the computer screen. If the image is fuzzy you must re-grab the image. If there are dust spots around the outside of the grain, highlight and then select the "Remove Bad Pixels" option. Then, set these spots to zero. If there are glare holes within the grain, highlight and then select the "Remove Bad Pixels" option. Then, set these holes to one. There are some cases when no amount of pixel editing can clean up the grain image, when this occurs skip and discard this grain and return to step 1. 5) Highlight and then select "Find Boundary Points." 6) Visually inspect the chosen boundary point image on the computer screen. If there are numerous bad points you must re-Grad the image. 7) If all is well, then highlight and then select "Save X-Y Coordinates to File." 8) Return to step 1. 41 Grains of a single sample are written in series, one after another, into the same BPT data file. At some later time, the boundary point data is recalled and prepared for the Fast Fourier Transform (FFT). This preparation includes conversion of the X-Y Cartesian coordinate boundary points into polar coordinates of radius versus angle. The radius is the distance from any one boundary point to the centroid of the grain outline with the angle being measured from a horizontal line which bisects the centroid. Next, because the mathematics behind the FFT require data to be equally spaced and centered about the zero, the irregular polar boundary points are first linearly interpolated to 128 evenly- spaced data points with the mean radius then being subtracted from each of the 128 points. It should be noted that this radius is later placed in the final amplitude spectrum set. The Fast Fourier Transform may now be calculated using the 128 evenly- spaced data points. The amplitudes for each wave number are then considered to equal the square root of the summation of the square of the real and imaginary FFT values. However, because the input data is real, half of the wave numbers represent reflections, and are therefore omitted. This entire process results in amplitude values for each of the 64 different wave numbers for each grain. The amplitude values and their respective wave numbers embody a mathematical representation of the grain, which is later 42 used for statistical analysis. However, after the amplitudes are determined they are written to a storage device as a binary data file with an ’.AMP’ file type identifier extension. Values for a single grain are written into the same AMP file in series, from the lowest to highest wave number, with data for successive grains of the same sample following. At present, the GSA program can calculate the Fourier amplitude values using one of two methods. The first method, termed the standard method, is precisely what is described is the preceding paragraph. For the second method, termed the normalized method, the GSA program initially calculates the Fourier amplitude values as in the standard method, but then divides all values by the amplitude value of the zeroth harmonic. This procedure to "normalize" the amplitude values, reduces any effect due to grain size variation. This method also decreases the total variance of the sample set. Data Processing After creating the AMP files for a given project, the Fourier data contained in those files must be prepared for statistical analysis. The binary AMP files are converted to ASCII files by using either the BYWAVEIT, CONVRTIT, or STATSIT programs, which were created by the author for this project (Appendices B, C and D). These programs create ASCII data files with varying formats from one or more of the binary AMP files. The ASCII files are then electronically transferred through a serial connection cable to an 43 IBM PC via a program called KERMIT. Once these files are on the IBM PC, they can be accessed by any number of statistical packages. The commercially available statistical package called "Bio-Medical Data Processing" (BMDP), developed by BMDP Statistical Software, Inc. was used to perform all statistical calculations. 44 STATISTICAL PROCEDURES Introduction The statistical tests used in this study are One-Way Analysis of Variance, Duncan’s New Multiple Range Test, Factor Analysis, Hotelling’s T2 Test, and Discriminant Function Analysis. A flow chart for statistical processing is shown in Figure 9. Data processing can be divided into three broad phases, each of which is described below. In the first phase, termed significance testing, the equivalency in shape composition of two or more samples is tested. For this project, significance testing consists of a one-way ANOVA design which is employed to determine whether there are important differences in samples collected at the same time from different locations, samples collected at different times from the same location, and the amplitude values for a single wave number for several samples. For the second phase, termed composition determination, several different tests are used to describe the shape composition of the beach samples in terms of percentages of associated source samples. Source samples are defined by a combination of geologic observation and data analysis which include significance testing using the Hotelling V statistic and factor analysis. Specifically, factor analysis is used to identify end-member 45 Figure 9. Generalized flow chart for statistical processing of Fourier amplitude values. 46 "si Statistical Process - General Logic Flow Chart PHASE 2 PHASE 1 PHASE 3 ^T e st S o u rc e s ^ w By use of Hotelling T Squared test. Not Distinct Different Interpret Results Analysis Complete Interpret Results Analysis Complete From Discriminant Function Analysis and RECASTIT. Composition Determination ^Single Harmonic Testing^ Harmonic by Harmonic testing of Geologic Hypothesis by ANOVA or Pearson Chi Square. Significance Testing ANOVA testing of Amplitude data to determine sample Inter-relations. 1) One Location-Varying Times 2) Many Locations-Same Time 3) Combination of the Above Go to Phase 2 when finished. ^ Source Selection ^ Sources are chosen by a combination of: 1) Geologic Observation. 2) Results from Significance Testing. 3) Factor Analysis. Note: All Phase 2 tests are based on Amplitude values for harmonics 2-24. f Significance Testing ^ and Data Description of Single Harmonic Amplitude values ANOVA testing and Factor Analysis of Single Harmonic Amplitude data. Chosen Harmonic is based on the single harmonic results. 1) Return to Source Selection, choose different Sources, and retest. 2) Retest with same Sources but use only Select Harmonics found to cause the most variance. 3) Composition Determination Not Possible. Go to Phase 3 (Single Harmonic Testing). Base conclusions on results of Significance, Factor and Single Harmonic testing. Options: samples, which are the most diverse with respect to each factor. Once the operator has selected the probable source samples, the Hotelling V Test may be used to formally test whether the shape composition of the source samples differ by a statistically significant amount. In some cases, it is not possible to obtain useful shape composition results and the third phase, termed single harmonic testing, may be preformed. This phase consists of a harmonic by harmonic test between amplitude values from selected samples. These samples represent operator- defined geologic hypothesis concerning the nature of the sample set. The testing is performed so that a single harmonic may be isolated that may differentiate between or among critical sets of samples. If this harmonic can be isolated, ANOVA and factor analyses may be performed to further describe the structure of the data set. Each of the statistical tests were calculated by the BMDP Statistical Software Program. This program is command based and operates by reading directions from an input file (Dixon, 1990). Typical input files for each of the above tests can be found in Table 5. Input files consist of several paragraph command words, which follow a slash (/). Each paragraph has several sub-command options which also must be given. The /INPUT, /VARIABLE, /GROUP, and /TRANSFORM paragraphs are common to all input files. The /INPUT sub-command defines the data file name, type, and format 48 T a b l e 5 . S a m p l e I n p u t f i l e s fo r B M D P . A ) S a m p l e file for A N O V A a n d D u n c a n ’ s N e w M u ltip le R a n g e t e s t . B ) S a m p l e file fo r t h e f a c t o r a n a l y s i s . C ) S a m p l e file for t h e H o t e l l i n g ’ s T - s q u a r e d t e s t . D ) S a m p l e file for d i s c r i m i n a n t f u n c t i o n a n a l y s i s . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ / INPUT / VARIABLE / GROUP / HISTOGRAM / COMPARISON / END FILE IS ’ SAMPLE.DAT’ . FORMAT IS FREE. VARIABLES ARE 4. NAMES ARE SAMPLE. GRAIN. WAVE. AMP, CODES(SAMPLE) = 1 TO 5. CODES( GRAIN) = 1 TO 200. C0DES(WAVE) = 1 TO 24. GROUPING = SAMPLE. VARIABLE = AMP. DUNCAN. / INPUT / VARIABLE / FACTOR / PLOT / END FILE IS 'SAMPLE.DAT’ . FORMAT IS FREE. VARIABLES ARE 23. RECLENGTH = 500. NAMES ARE W2, W3. W4, W5. W6, W7. W8, W9, W10. W l l , W12, W13. W14, W15. W16, W17. W18, W19, W20 W21, W22, W23, W24. METHOD = PCA. CONSTANT = 0 .2 5 . IN IT IA L = 2. FINAL = 2. FSC0RE = 2. B / INPUT / VARIABLE / GROUP / TW0GR0UP / END FILE IS ’ SAMPLE.DAT’ . FORMAT IS FREE. VARIABLES ARE 4. NAMES ARE SAMPLE. GRAIN. WAVE. AMP, CODES( SAMPLE) = 1 TO 5. CODES( GRAIN ) = 1 TO 200. CODES( WAVE ) = 1 TO 24. GROUPING = SAMPLE. VARIABLE = AMP. HOTELLING. c / INPUT / VARIABLE / GROUP / DISCRIMINANT / PLOT / END FILE IS ’ SAMPLE.DAT’ . FORMAT IS FREE. VARIABLES ARE 12. RECLENGTH = 500. NAMES ARE SAMPLE, Wl, W4, W7, W 9 W12. W13, W16, W18. W19, W20. W22 GROUPING = SAMPLE. CODES!SAMPLE) = 1 TO 6. NAMES(SAMPLE) = A. B, C. D. E ,F. USE = 1, 3, 4, 5. D CO to be used in the calculation. The /VARIABLE and /GROUP sub-commands define the format, names, and ranges of the variables to be used. The /TRANSFORM sub-commands mathematically converts the input data by a specified function. However, to understand the results of these tests, one must first appreciate how the results are obtained. Thus, the following is a brief synopsis of each of the above statistical procedures. One-Way Analysis of Variance One-way analysis of variance (ANOVA) is a method used to test the equivalency of samples by partitioning the sum of squares. ANOVA compares the variance of multiple sample populations as well as subsets of such populations. The total variance of the tested sample set is separated into among sample variance, within sample variance and an error term (Alder and Roessler, 1977). The rationale being that if their variances are similar, then these sample populations are likely to be from the same parent population. ANOVA requires that data consist of random samplings from normally distributed parent populations that have equal variance and no interaction effects (Davis, 1986). If the assumptions are met then the null hypotheses (H0 ) states that the samples are equal and are likely drawn from the same parent population, with the alternative (HJ that at least one sample is different. 50 A one-way ANOVA without data replication design was employed for this study. Test data consists of normalized amplitude values for the first through the 24th harmonic for each grain in every sample tested. Further, the significance level was chosen to be five percent (alpha = 0.05) for all tests. An example ANOVA table for the above design and the appropriate equations can be found in Table 6. The /HISTOGRAM paragraph directs BMDP to preform the ANOVA (Table 5a). Duncan’s New Multiple Range Test Duncan’s New Multiple Range Test was preformed in conjunction with ANOVA. Although ANOVA usually can discern when there is significant difference between or among multiple samples or sets thereof, it can not identify which sample or samples are responsible for the observed significant difference. The Duncan test provides a series of shortest significant ranges with which to compare difference between means (Alder and Roessler, 1977). Samples which are not significantly different are grouped together, whereas dissimilar samples are grouped separately. The basic assumptions for the Duncan Test are identical to ANOVA (Alder and Roessler, 1977). In the Duncan’s New Multiple Range Test, the mean for each of the samples to be tested is determined and then placed in a table listed from highest to lowest. Based on the number of sample means, the significance level, the number of degrees of freedom, and the standard error of the mean, 51 Table 6. The design for the one-way Analysis of Variance as used in this study (Davis, 1986). Source Sum of Squares Degrees of Freedom Mean Square F Ratio Main Effect (between) v . Tf G x i j ) 2 ^ 1 n ■ , N C - 1 SSme DFme MS me m s Sd Sample Deviations (within) 2 T ^ Z . i X u - 2 ^ N - C SSsd dfsd Total ^ v 2 (SXi j) 2 ^ ' i A u im N - 1 Where: j = treatments; i = variates; T = treatment totals; n = observations; N = total number of observations; C = total number of treatments; x = data; SS = Sum of Squares; DF = Degrees of Freedom; MS = Mean Square. 52 the shortest significant ranges are computed. The standard error of the mean is the square root of the within sample mean square from the analysis of variance table, divided by the square root of the number of observations on which each of the means is based (Alder and Roessler, 1977). The difference between any two sample means is then compared to shortest significant ranges, and if lower is considered not significantly different. In the present study, the Duncan’s New Multiple Range Test was used to identify the samples which were responsible for the rejection null hypotheses associated with ANOVA. A five percent significance level (alpha = 0.05) was chosen for all tests. A /COMPARISON Paragraph following a /HISTOGRAM paragraph directs BMDP to preform a Duncan Test if the results of ANOVA show at least one significant difference (Table 5a). Factor Analysis Factor analysis was developed by experimental psychologists in the 1930’s, to identify patterns in multivariate data matrices (Davis, 1986). The term "factor" represents the intent of the analysis, which is to arrange the data along hypothetical axes which account for much of the variance of the system. Thus, in factor analysis, there is no significance testing of the data set and no probability values are generated. The information is plotted from a new theoretical perspective for the purposes of visually identifying trends within the data set (Davis, 1986). The axes or "factors" on which the data 53 are plotted are determined from eigenvalues and eigenvectors obtained from the covariance matrix, which is, in turn, obtained from a square "R" matrix consisting of the original data matrix, multiplied by its transpose (Davis, 1986). In this project, factor analysis is employed to identify possible source samples which occur as end members along one or more factor axes. This was determined from the averaged amplitude values for the second through the 24th harmonic for each of the samples considered. The amplitude value for the first harmonic was not considered in the factor analysis since its average is typically one order of magnitude greater than the other amplitude values and thus which would bias the resulting factors. Numerous types of factor analyses exist and the specific type used for this study is termed a "Q-mode, maximum variance, principle component analysis, determined from the covariance matrix" (Dixon, 1990). The /FACTOR paragraph directs BMDP to preform the analysis (Table 5b). Hotelling’s T2 Test The Hotelling’s T2 Test is a technique to compare the multivariate means between two sample populations. This particular test is preformed to verify that the chosen source samples, obtained from either geologic reasoning or factor analysis, are statistically distinct with respect to mean values. This test consists of determining the multidimensional mean vector for each sample, as well as the associated variance-covariance matrices (Alder and Roessler, 54 1977). Then, using matrix algebra, the variance-covariance matrix, the inverse of the variance-covariance matrix, and the difference from each of the mean vectors is multiplied. Finally, the resultant matrix is multiplied by the number of observations to obtain the T2 value. Once the T2 value is determined, the associated F value is the T2 value multiplied by the resultant of the difference of the number observations and the number of measurements divided by the degrees of freedom (Alder and Roessler, 1977). The Hotelling’s V Test requires that data consist of random samples from normally-distributed parent populations that have no interaction effects (Alder and Roessler, 1977). As in ANOVA, the test data consists of normalized amplitude values for the first through the 24th harmonic for each grain in every sample tested. The /TWOGROUP paragraph directs BMDP to preform the T test (Table 5c). A significance level of five percent (alpha = 0.05) was chosen for all tests. Discriminant Function Analysis Discriminant function analysis (DFA) is one of the most powerful and versatile multivariate statistical techniques available (Davis, 1986). As used in this study, DFA consists of four stages. In the first stage, the source sample sets must be defined using the methods described above. The second stage consists of computing a mathematical function, termed the discriminant function, from the amplitude values of the first through the 24th harmonic for each grain in the operator-defined source sample sets. 55 This mathematical function consists of finding a transform which gives the minimum ratio of the difference between the group multivariate means to the multivariate variance within the defined groups (Davis, 1986). Conceptually, the DFA takes the data matrix and rotates an imaginary orthagonal axis through multidimensional space until the sample vector clusters are both minimized in area and maximized in separation. The interested reader should refer to Davis (1986) for explanation of the matrix mathematics for the DFA. In the third stage each grain of each non-source samples is assigned to the most likely source field based on a maximum likelihood solution (Davis, 1986). The shape composition of each non-source sample is then determined from the assignments of each of the 200 grains. As in the first stage of DFA, all results are calculated from the amplitude values of the first through the 24th harmonic for each grain. A fourth stage is needed to correct for any "overlap" of the source fields. DFA assumes that the sources are completely unique; however, this is rarely the case with sources typically overlapping to some degree (Davis, 1986). Thus, percent composition as determined by DFA must be proportionally "recast" to correct for this condition. Recasting is accomplished via the method described by Ahlschwede (1988) and is preformed by the RECASTIT program, which was encoded by the author (Appendix E). Once the 56 percentages have been recast they can then be used for geologic interpretation. DFA assumes that the source samples are statistically distinct and consist of random samples from normally-distributed parent populations that have no interaction effects. The /DISCRIMINANT paragraph directs BMDP to preform both the second and third stages of the analysis (Table 5d). 57 RESULTS Presentation of the results is separated into four sections based on type of statistical test. These sections are sample comparison using ANOVA, sample relationships from factor analysis, source sample comparison using Hotelling’s T2 test and Chi-square analysis of shape frequency distributions. Sample Comparison Using ANOVA Amplitude values for the first through the 24th harmonic of each grain for groups of two or more samples were compared using ANOVA. These tests were preformed to investigate both the temporal and spacial consistency of sample grain-shape composition. The temporal consistency of fluvial grain-shape composition was tested by comparing samples taken from the same river at various times. The Ventura River was sampled in June, 1969, June, 1985 and March, 1992. The Santa Clara River was sampled in June, 1985 and March, 1992. Using one way ANOVA, the Ventura River samples showed no significant difference (probability = 0.69), as did the Santa Clara River samples (probability = 0.18). It should be understood that these probabilities measure the likelihood that samples of the same river belong to the same or equivalent parent population. The spatial consistency with respect to grain-shape composition of the dunes and the inner shelf sand was tested by comparing samples taken at 58 the same time but from different locations within each sand body. Both the dune and the inner continental shelf were sampled twice. Sets of dune and shelf samples were each collected at the same time, namely June, 1991 and August, 1988, respectively, but at different locations. Using one-way ANOVA, these sets of sand bodies showed no significant differences with respect to location at a given time (Dune probability = 0.23; Shelf probability = 0.76). The only cliff sample, Arroyo Burro cliff, was compared to Arroyo Burro Beach. The grain-shape compositions of the two samples were found to be statistically homogeneous using one-way ANOVA (probability = 0.28). The spatial variation of the shape composition of beaches throughout the study area in June 1985, October 1985, June 1991, October 1991 and February 1992 was tested (Table 7). Using one-way ANOVA, it was found that the amplitude values for the first through the 24th harmonic for each grain in the tested samples displayed significant variation only in the October 1985 and October 1991 sample sets. Within the October 1985 sample set, the Santa Clara River Beach sample displayed the most variance. Within the October 1991 sample set, the samples from East Beach, Spinnaker Beach, Hueneme Beach, and Ormond Beach collectively accounted for the most variance. The temporal variation in grain-shape composition of Emma Woods Beach, Surfer’s Point Beach, Marina Park Beach, Spinnaker Beach, and Santa 59 T a b l e 7 . R e s u l t s o f s i g n i f i c a n c e t e s t i n g fo r s p a t i a l v a r ia t io n in s e t s o f b e a c h s a m p l e s c o l l e c t e d a t d if f e r e n t t i m e s . P r o b a b i li t i e s m a r k e d w ith d o u b l e a s t e r i s k s a r e v e r y s ig n if ic a n t , w h e r e a s s i n g l e a s t e r i s k s a r e s i g n i f i c a n t . 2 / 2 1 / 9 2 Emma Surfer’s Marina Holly Orm P r o b .= S a m p l e S e t . Woods Point Park wood ond 0 . 0 9 10/10/91 Arroyo Lead- East Emma Surfer’s Marina Spinn Santa Holly Huen Orm Point P r o b .= S a m p l e S e t . Burro better Beach Woods Point Park aker Clara wood eme ond Mugu 0.00 ** 4/26/91 - 6/10/91 East Emma Surfer’s Marina Spinn Santa Holly Huen Orm Point P r o b .= S a m p l e S e t . Beach Woods Point Park aker Clara wood eme ond Mugu 0 . 3 3 10/22/85 - 10/25/85 Surfer’s Marina Harbor Harbor Spinn-! Santa P r o b .= S a m p l e S e t . Point Park A B aker Clara 0.01 * 6/4/85 - 6/6/85 i Surfer’s1 Marina; Spinn-, Santa P r o b .= S a m p l e S e t . Point Park aker Clara 0.06 a o Clara River Beach also was examined (Table 8). Using one-way ANOVA, it was found that significant variation occurred within the Spinnaker Beach and Santa Clara River Beach sample sets. Within the Spinnaker Beach sample set, the October, 1991 sample displayed the most variance, whereas the October, 1985 sample accounted for the most variance within the Santa Clara River Beach sample set. In both the Spinnaker and Santa Clara River Beach sample sets, there was less variance between and among the two seasonal samples from the same year (1985 or 1991) than between and among samples from the two different years. Four beaches have samples available for both 0 m MLLW and -9 m MLLW elevations. Thus, samples of the same beach were compared to one another to measure the shape variation from the upper to lower shoreface. No significant variation occurs within the Spinnaker Beach sample set (probability = 0.77) or Santa Clara River Beach sample set (probability = 0.15). However, significant variation does occur within the Surfer Point Beach sample set (probability = 0.03) and Marina Park Beach sample set (probability = 0.03). Sample Relationships from Factor Analysis A Q-mode, maximum variance, principle component analysis was performed on the averaged amplitude values for the second through the 24th harmonic for all the 61 samples used in this study. The resulting two factor 61 Table 8. Results of significance testing for temporal consistency of shape composition in sets of beach samples at four non-source beaches. Probabilities marked by a single asterisk are significant. Emma Woods 6/4 8/30 4/26 10/10 2/21 Prob.= Beach 1985 1988 1991 1991 1992 0.38 Surfer’s Point 6/4 10/22 4/26 10/10 2/21 Prob.= Beach 1985 1985 1991 1991 1992 0.56 Marina Park 6/4 10/22 4/26 10/10 2/21 Prob.= Beach 1985 1985 1991 1991 1992 0.5 4 Spinnaker Beach 6/4 1985 10/22 1985 4/26 1991 10/10 1991 I I * n t - o o C L o Santa Clara River Beach 6/4 1985 10/22 1985 4/26 1991 10/10 1991 Prob.= 0.01 * 62 solution, as calculated from the factor scores, describes 77 percent of the total variance in the system (factor one = 54 percent; factor two = 23 percent) (Table 9). The sample identification numbers used for the factor analysis correspond to the computer code numbers defined in Appendix A. The resulting factor score plot (Fig. 10) has a sample spread with four end- member samples, two for each factor. Samples 56 and 40 are the end- members for factor one, whereas samples 51 and 8 are the end-members for factor two. Based on the average amplitude values of these end-members, factor one is interpreted to represent asperity or grain roughness, with asperity increasing in the positive direction. Factor two is interpreted to represent grain elongation, with elongation increasing in the positive direction. Thus sample 56 would be the be elongate with a high asperity, whereas sample 40 would be moderately elongate with a very low asperity. These interpretations of asperity and elongation for factors one and two respectively were evaluated further by use of scanning electron microscopy (SEM). Sample 56 on the factor scores plot is from Sand Canyon Creek. The photomicrographs of this sample and its position on the factor scores plot indicate that the grain is very elongate with high asperity (Fig. 11a and b). A SEM photomicrograph of a grain from Santa Paula Creek (Fig. 11c) indicates that it is less elongate than Sand Canyon Creek, but still has a high asperity, a fact that is in agreement with the factor scores plot (sample 53 on 63 Table 9. Factor scores solution for factors one and two as determined from analysis of the 61 samples used in this study. 64 Sample Number Factor One Factor Two 1 0.365 -0.686 2 0.860 -0.870 3 -0.158 0.940 4 -0.513 -1.704 5 0.149 0.114 6 -0.921 -0.149 7 0.304 -0.656 8 0.728 -2.309 9 0.204 -1.002 10 0.160 0.154 11 1.286 -0.739 12 -1.040 1.072 13 0.274 0.800 14 0.717 -1.152 15 -0.555 0.565 16 -0.937 0.810 17 0.224 -1.722 18 0.639 1 .204 19 0.901 1.275 20 0.840 -0.390 21 0.010 1 .238 22 -0.566 -0.192 23 -0.946 0.388 24 -0.730 -0.565 25 0.661 -0.771 26 -0.202 0.925 27 0.654 0.493 28 1.162 0.514 29 1 .677 -1.188 30 -0.091 0.635 31 0.021 -0.645 32 1 .003 1.054 33 -0.859 -0.750 34 0.568 -0.510 35 -1.062 -0.651 36 -0.074 1 .785 37 2.336 -0.411 38 -0.324 -1.517 39 -2.109 0.012 40 -2.416 0.262 41 -0.745 -0.028 42 -0.295 0.581 43 -1.258 -0.082 44 -0.203 -0.431 45 -1.561 0.626 46 -0.802 0.217 47 0.547 -0.292 48 0.284 0.038 49 -0.542 0.634 50 -0.341 -1.038 51 -0.438 2.858 52 -1.137 -1.693 53 2.305 -0.688 54 0.457 0.696 55 -0.348 1.343 56 3.301 2.177 57 -0.339 -0.245 58 -0.533 -1.162 59 -0.416 0.493 60 0.206 0.517 61 -0.383 -0.181 Figure 10. Solution from factor analysis of the 61 samples used in this study. Sample numbers refer to the computer codes outlined in Appendix A. Circled samples are end-members for factor one. Doubly- circled samples are end-members for factor two. 66 Factor Two + 3 . 6 0 + 1 . 8 0 0 - 1 . 8 0 - 3 . 6 0 a Ventura Harbor Channel A •36 South Sand Canyon Creek © Arroyo Burro Beach c o V-* cd O ) c o LU O ) c ‘g o C O C D o c • 39 •55 • 21 • 12 3 18 • 16 49 26 * 13 • 45 1 • • 30 ' ^ 15* , *42 3U *60 •19 •32 23 • 46 *41 59 •43 .61 22 *57 .35 • 24 44 •33 • 50 • 58 <27 *28 -10 5 • 48 •31 •• • 47 7 «34 • 20 38 •52 • 4 •9 *25 • 2 • 14 • 17 • 11 • 29 • 37 • 53 Point Mugu Beach T“ - 3 .7 5 Increasing Asperity 'I I "T I 0 + . 9 3 8 + 1 . 8 8 + 2 . 8 1 + 3 . 7 5 - 2 .8 1 - 1.88 - .9 3 8 C D •v j Factor One Figure 11. Scanning electron photomicrograph of selected quartz grains from three samples. The elongate and rough quartz grains in A and B are from granodioritic source rock exposed along Sand Canyon Creek. Photomicrographs C and D are less elongate quartz grains reworked from sedimentary strata exposed along Santa Paula Creek and Ventura River, respectively. The grain in photomicrograph C is more rough than the grain illustrated in D. Scale bar equals 200 microns. 68 69 plot). Sample 55 on the factor scores plot indicates that this sample should have an intermediate elongation and a lower asperity than the previous two samples. The SEM photomicrograph of a grain from sample 55 (Fig. 1 1 d) shows these characteristics. Source Sample Comparison Using Hotelling’s T2 Test Hotelling’s T2 test was used to measure the similarity within source samples and within end-member samples from the factor scores plot. This step, as noted earlier, is a necessary prerequisite for discriminant function analysis. The source and end-member samples that were analyzed are listed in Table 10 and are possible sand sources for only the eastern third of the Santa Barbara Littoral Cell, namely between Emma Woods Beach and Point Mugu Beach. Table 10. Samples tested using the Hotelling V test. Source Samples End-Member Samples Dune A-June, 1991 Arroyo Burro Beach-October, 1991 Dune B-June, 1991 Sand Canyon Creek-March, 1992 Ventura River-March, 1992 Harbor A-July, 1988 Santa Clara River-March, 1992 Point Mugu Beach-June 1991 Santa Paula Creek-March, 1992 Sand Canyon Creek-March, 1992 Emma Woods Beach-June, 1991 Emma Woods Beach-October, 1991 Point Mugu Beach-June, 1991 Point Mugu Beach-October, 1991 70 The results from the source sample testing showed that the grain-shape composition of the sample obtained from Sand Canyon Creek in March, 1992, is significantly different (alpha = 0.05) from all other samples. Further, the sample obtained from Emma Woods Beach in October, 1991 is significantly different from the March, 1992 samples obtained from the Santa Clara River and Santa Paula Creek. However, the differences between both the Santa Clara River and Santa Paula Creek samples and the June, 1991 Emma Woods Beach sample are not significant. The March, 1992 Sand Canyon Creek sample is significantly different (alpha = 0.05) from all other end-member samples. However, Harbor A, Arroyo Burro, and Point Mugu Beach samples are statistically homogeneous with respect to grain-shape composition. Since the March, 1992 Sand Canyon Creek sample was the only sample that is statistically distinct from all other source or end-member samples, there is an insufficient number of distinct sample sets to produce a meaningful solution using discriminant function analysis. Chi-Square Analysis of Shape Frequency Distribution Any given set of grain-shape distributions may in some cases have similar mean amplitude values but distinctly different shape frequency distributions. To evaluate this possibility, various geologic hypothesis (i.e. groups of samples) were tested by comparing all the amplitude values of a 71 single harmonic, from the first through the 24th harmonic (Table 11). All testing was performed using the Pearson chi-squared test with a significance level of 0.05. Hypothesis 1 compares samples from Emma Woods Beach with Point Mugu Beach. This hypothesis was used to compare the shape composition of sand which enters and/or exits the eastern third of the Santa Barbara Littoral Cell. There was found to be no significant difference between any of the amplitude distributions for the first 24 harmonics. Hypotheses 2 through 5 compare samples from the various rivers and tributary streams to each other. The hypotheses tested are: #2) the Santa Clara River versus the Ventura River samples, #3) the Santa Clara River versus Sand Canyon Creek samples, #4) the Santa Clara River versus Santa Paula Creek samples and #5) the Ventura River versus Sand Canyon Creek samples. Collectively these hypothesis were used to examine the grain-shape composition of the rivers. The only significant differences are the Sand Canyon Creek sample versus the Santa Clara and Ventura River samples. More than 80 percent of the harmonics for these samples displayed statistically significant differences. Hypotheses 6 to 9 were proposed to investigate the relationship between the two rivers and either Emma Woods Beach and Point Mugu Beach. The hypotheses examined are: #6) Emma Woods Beach versus the Ventura River, 72 Table 11. Results of chi square testing of individual amplitude values. All shaded squares indicate a significant difference at the five percent level. 73 Hypotheses 1) Emma Woods vs. Point Mugu 2) Santa Clara River vs. Ventura River 3) Santa Clara River vs. South Sand Canyon Creek 4) Santa Clara River vs. Santa Paula Creek 5) Ventura River vs. South Sand Canyon Creek 6) Emma Woods vs. Ventura River 7) Emma Woods vs. Santa Clara River 8) Point Mugu vs. Ventura River 9) Point Mugu vs. Santa Clara River 10) Shelf vs. Ventura River 11) Shelf vs. Santa Clara River 12) Dune vs. nearby beach 13) Northern Beaches vs. Southern Beaches 14) Winter Beaches vs. Storm Beaches 15) Northern Beaches 16) Southern Beaches 17) Storm Beaches Harmonics 23 24 "si #7) Emma Woods Beach versus the Santa Clara River, #8) Point Mugu Beach versus the Ventura River and #9) Point Mugu Beach versus the Santa Clara River. Only hypothesis 8 and 9 were found to have any dissimilarities. Harmonics 2, 3 and 6 tested significant for hypothesis 8, whereas harmonic 3 tested significant for hypothesis 9. The degree of similarity between the two rivers and the shelf samples was investigated with hypotheses 10 and 11. Harmonics 4 and 7 were found to be significantly different between the shelf and the Santa Clara River samples, whereas harmonic 4 displayed a statistically significant difference between the shelf and the Ventura River samples. Hypothesis 12 compares the shape composition of the dune samples of McGrath Beach to the adjacent Santa Clara River Beach sample. This hypothesis was used to determine the relationship of the dunes to the actively transported sand of the upper shoreface. Significant differences were found to occur between harmonics 1 and 19. The northern beaches of the eastern third of the Santa Barbara Littoral Cell (Emma Woods, Surfers Point, and Marina Park) were compared with southern beaches (Hollywood, Ormond, and Point Mugu). These comparisons, hypothesis 13, were used to compare the changes in shape composition across the study area. No significant difference occurred between any of the amplitude distributions for the first 24 harmonics. 75 The influence of large storms was examined by hypothesis 14. Several end-of-winter samples collected from five different beaches were compared to samples collected from the same beaches after the large storms of 1992. Harmonics number 5, 7, 12, 18 and 23 displayed statistically significant differences. Hypotheses 15, 16, and 17 examine the variation within northern beaches, within southern beaches and within post-storm beaches as analyzed in hypotheses 13 and 14. The only dissimilarities were found within the post storm beaches, with harmonics 5, 7, 12 and 18 displaying statistically significant differences. 76 DISCUSSION This discussion is divided into the following five parts: sand sources for the study area, boundaries of potential littoral subcells, character of sediment transport pathways, temporal changes in grain-shape composition, and implications for computation of sand budgets. Sand Sources for the Study Area Figure 12 depicts relationships between and among various sedimentary sources as determined from the factor scores. The sample obtained from Sand Canyon Creek represents first-cycle sand from an acid-plutonic source rock, and plots as very elongate with the highest asperity present in the sample set. The sample obtained from Santa Paula Creek, contains resedimented marine sand from the Matilija, Los Posas, and Coldwater Formations. This sample is more equant and has a high asperity. Sand from the mouth of the Santa Clara River plots between the samples from Santa Paula and Sand Canyon Creeks and could be considered as a mixture of the two. However, according to the factor analysis, Hotelling’s T2 , and chi- square testing, the grain-shape composition of the Ventura River, a river with no plutonic rocks in its watershed, is indistinguishable from that of the Santa Clara River. These analyses also indicate that the Sand Canyon Creek sample is markedly different from all other samples in the data set. Therefore, 77 Figure 12. Relationships between and among various sedimentary sources as determined from the factor scores. 78 Factor Two +3.60 4-1.80 0 -1.80 -3.60 a c o " -4— 1 03 C D C o L U C D C 'co c n C D o c River/Shelf sample field. Sample labels as follows: V = Ventura River, C = Santa Clara River and S = inner shelf Dune sample (#12) —^ collected near Santa ^ Clara River Mouth Sample from the Monterey Formation (#52) First-cycle sand collected from Sand Canyon Creek (#56) Sample from Santa Paula Creek (#53) □D ■ Dune sample (#4) collected in the approximate center of the McGrath Dune Field 1— -3.75 Increasing Asperity -2.81 - 1.88 -.938 I - f.938 4-1.88 4-2.81 4-3.75 ■vj CO Factor One first-cycle sand similar to that of Sand Canyon Creek is not a major source of sand for the Santa Clara River. As much as 35 percent of the total watershed of the Ventura River is underlain by the Sespe, Coldwater, and Matilija Formations, of which the Sespe Formation accounts for almost half of the exposed rocks capable of yielding sand. Thus, the similarity of the Santa Clara River with the Ventura River, suggests that the Santa Clara River also must receive a significant part of its sand from the Sespe Formation. As indicated by the factor analysis (Fig. 12), the inner shelf receives much of its sand from the Ventura and Santa Clara Rivers. However, hypotheses 10 and 11 demonstrate that the inner shelf contains a population of quadrate grains as signified by a dissimilarity in the fourth harmonic. This cannot be due to abrasion in the surf zone, since harmonics 4 represents a fundamental grain shape which would not significantly change with abrasion. Further, other samples that have gone through the surf zone do not have high values for the fourth harmonic (hypotheses 6 through 9). Therefore, this difference in the fourth harmonic, must be due to sand derived from another part of the shelf. The sample labeled as "Dune A" was obtained from the McGrath dune field, just southeast of Santa Clara River Beach. The factor plot shows sample Dune A to have a slightly less asperity than the Santa Clara River 80 Beach sample. This decrease in asperity may be due to the abrasion during eolian transport. The second dune sample (Dune B) is notably less elongate than both of the river samples as well as the Dune A sample. Dune B plots near the June, 1969 Emma Woods Beach sample, which may indicate that the dunes have a shape composition related to past activity. Whether or not the dunes are compositionally zoned, Hotelling’s T2 values indicate that the dune sand samples are similar to each other. The dunes also are broadly similar to the sand samples collected on the upper shoreface. The only significant differences were identified by chi-square testing (hypothesis 12) and occurred in the first and 19th harmonics. These differences may be due to both selective transport and abrasion from the beach to the dune field. Figure 13 shows the relationships between and among beach samples and suggests that there is an increase in the asperity of beach sand samples from the western to eastern parts of the study area. Further, the Ventura River, Santa Clara River and the inner shelf sand samples are much more elongate than most of the beach samples. Therefore, there must be some unsampled sand sources to account for the more equant grain-shape compositions present in these beaches. This additional source (termed: western source) for the Santa Barbara area must have both low asperity and a moderately low elongation; whereas in the area west of the Santa Clara River, the source (termed: eastern source) must have a high asperity and low 81 Figure 13. Relationships between and among various beach samples as determined from the factor scores. Plot includes all 0 m MLLW samples collected from June, 1985 to February, 1992. 82 Factor Two o o o o + 3 . 6 0 + 1 . 8 0 0 - 1 . 8 0 A C o "5 O) C o L U O) c *00 03 ( D o c - 3 . 6 0 - - 3 . 7 5 j. EXPLANATION 45 Sample field for the western beaches (Arroyo Burro, Leadbetter, East and Emma Woods Beach). Sample field for the middle beaches (Surfer's Point, Marina Park, Spinnaker, Santa Clara Bar Beach). Sample field for eastern beaches (Man<............... Hueneme, Orrr Mugu Beach). beaches (Mandalay, Hollywood, Hueneme, Ormona and Point River/Shelf sample field (from Figure 12). Increasing Asperity -2.81 -1.88 .938 0 +.938 + 1.88 +2.81 + 3 . 7 5 Factor One elongation. The western source may reflect either discharge from creeks that drain the Santa Ynez Mountains or an inner shelf source. The eastern source must be shelf-derived since there are no other rivers or cliffs that supply sand to the area. This eastern source also may account for the differences in the shape composition identified by chi-square testing in hypotheses 8 and 9 (Point Mugu Beach versus the Ventura River and Point Mugu Beach versus the Santa Clara River, respectively). Boundaries of Potential Littoral Subcells ANOVA testing of the spatial variation along the coast indicates that changes in grain-shape composition from the west to east are not as great as local changes. Specifically, the variance is largest in samples collected from those beaches that receive dredged sand, and include East Beach, Spinnaker Beach, Santa Clara River Beach, Hueneme Beach, and Ormond Beach. Significant compositional change occurs between the samples obtained in October, 1985 and October, 1991. This shape compositional difference may be caused by the bypassing effort most of which takes place during the summer months. Chi-square testing indicates that the sand that composes the beaches between Emma Woods and Point Mugu has identical shape-frequency distributions (hypotheses 1, 13, 15 and 16). Thus, there are no identifiable 84 sedimentary boundaries between Emma Woods Beach and Point Mugu Beach provided bypassing of sand from the local harbors continues. As determined from Figure 13, there is a difference in the grain-shape composition of sand between the western and eastern parts of the study area. A consequence of which is that there may be little net longshore transport from the Santa Barbara area. This fact suggests the possibility that there may be a separation in the coastal reach between the Santa Barbara and Ventura areas. Alternatively, this difference in the grain-shape composition of sand between the two areas may be due only to mixing, since the net longshore sand transport increases by 260 percent in the Ventura area and any Santa Barbara sand would be significantly diluted (Noble Consultants, 1989). Character of Sediment Transport Pathways The sand from Emma Woods to Point Mugu Beach has a similar shape composition (determined from chi-square hypotheses 1, 13, 15 and 16), which suggests that the sand in this area is well mixed due to bi-directional seasonal transport. The fact that sand with this shape composition extends from the Ventura and Santa Clara River mouths, the principal sources, to Point Mugu supports the concept that there is a net southeasterly longshore flow within at least the eastern third of the Santa Barbara littoral cell, (Emery, 1960; Kolpack, 1986; Noble Consultants, 1989). However, as previously discussed, 85 there also is a net onshore transport of sand from the shelf which causes a broad shift towards higher asperity and lower elongation values in the beaches east of the Ventura River. There is a shift in the grain-shape composition of June, 1985 littoral sand samples toward higher asperity from the upper to lower shoreface (Fig. 14). The shift in asperity may be caused either by a combination of abrasion of sand entrained in the upper shoreface or selective transport of grains with a higher asperity from the upper to the lower shoreface or, more likely, the adjacent inner continental shelf. Sand samples collected from the entrance channel to the Ventura Harbor are noticeably more elongate than beach sand within the region, probably due as well to selective transportation of more elongate grains into the harbor (Fig. 15). Temporal Changes in Grain-Shape Composition The grain-shape composition of sediment from Ventura and Santa Clara Rivers was determined from ANOVA testing to have not significantly changed from 1969 to 1992. Temporal variation of four beaches within the Ventura area also was evaluated using one-way ANOVA. Significant variation was found to occur among sample sets obtained at four different times from both Spinnaker and Santa Clara River Beaches (Tables 7 and 8). These beaches receive varying amounts of sand dredged from Ventura Harbor, which may be the principal cause of the observed changes. On average, however, there is 86 Figure 14. Relationships between and among upper and lower shoreface samples as determined from the factor scores. Samples were collected in June, 1985. 87 00 00 + 3 .6 0 + 1 .8 0 o 0 o CO L L ■1.80 -3.60 i c o o d O ) c o UJ O ) c ‘ g o 0 3 C D o c -3.75 E X P L A N A T I O N Sample field for the lower shoreface (-9 m MLLW). Sample field for the upper shoreface (0 m MLLW). (C) Santa Clara Bar Beach @) Marina Park Beach (?) Sufer’s Point Beach (S) Spinnaker Beach Increasing Asperity -2.81 -1 .88 ■.938 0 Factor + .93 8 +1.88 +2.81 One + 3 .7 5 Figure 15. Relationships between and among samples collected in October, 1985 and July, 1990 from the Ventura Harbor and samples collected from June, 1985 to March, 1992 as determined from the factor scores. 89 A c o "■ + — I 0 3 O) c o L U O) C 'co 0 3 03 b c "+ + i : B + ii'J .+ • • - m m + + + + +/ ■ ■ E X P L A N A T IO N + Sample field for samples from the Ventura Harbor Channel. 'A' harbor samples are from the inner channel area. 'B‘ harbor samples are from the outer channel area. Sample field for the beaches near the Ventura Harbor (Marina Park, Spinnaker, Santa Clara Bar Beach). Increasing Asperity - 3 . 7 5 T T ^ 8 8 ~ 9 3 8 0 + . 9 3 8 + 1 8 8 + 2 . 8 1 + 3 . 7 5 Factor One no significant long-term temporal variation in the grain-shape composition of littoral beach sand samples collected between 1969 and 1992 within the study area. Seasonal variation within the study area as determined from the factor scores is shown in Figure 16. This plot indicates that there is a general accumulation of more elongate grains on the upper shoreface during the summer months. This is probably due to the onshore transport of more elongate shelf-derived grains through ridge and runnel systems. This accumulation is mixed with pre-existing backshore sand during or after winter storms to produce sand with an intermediate composition with respect to elongation. By the end of the oceanographic winter, storm-generated mixing produces the less elongate grain-shape field. There is significant variation in the shape composition of storm beaches in the 5th, 7th, 12th, and 18th harmonics as determined from chi-square hypotheses 14 and 17. This variation may be to sediment influx from the shelf during storms, or from the different wave conditions that exist during storms. Implications for Computation of Sand Budgets Several critical assumptions used in computation of the sand budget analysis for the Ventura area (Noble Consultants, 1989) are listed below. 91 Figure 16. Relationships between and among five beaches collected in June, 1991, October, 1991 and February, 1992 as determined from the factor scores. These beaches are Emma Woods, Surfer’s Point, Marina Park, Hollywood, and Ormond. 92 X X A c o H 3 D > C _o L U O ) c 'co d S (D o c + E X P L A N A T IO N Sample field for the selected June, 1991 beaches (end-of-wlnter). Sample field for the selected October, 1991 beaches (end-of-summer). Sample field for the selected February, 1992 beaches (post-storm). 59 Increasing Asperity I .. -2 .8 1 i + . 9 3 8 i + 1.88 i + 2 . 8 1 — i p + 3 . 7 5 - 3 . 7 5 - 1.88 - . 9 3 8 Factor One 1) Net sand transport within the littoral zone is predominantly southeast or downcoast. 2) The Ventura and Santa Clara Rivers are the dominate sources of sand for the beaches in the Ventura area. 3) There is no significant movement of sand to or from the inner and middle continental shelf. 4) There is no net effect from the seasonal movement of sand to and from the shoreface. 5) Onshore-offshore transport is related only to storms. This study supports the assumption that the net longshore transport within the shoreface in the Ventura area is to the southeast. There is a noticeable seasonal change in the shape composition of beach sand within the Ventura area. However, there is no statistically significant change in the shape composition of beach sand between 1969 and 1992, which supports the assumption that there is no net effect from the seasonal movement of sand to and from the shoreface. This study supports the assumption that the Ventura and Santa Clara Rivers are the dominate source of littoral sand between Emma Woods and Point Mugu Beach. However, it was determined that the inner shelf within the Ventura area also may be an important source of littoral sand. This finding contradicts assumption three, which has been used in the calculation of past sand budgets. Due to the similarity in the 94 grain-shape composition of upper shoreface and most inner shelf sand, it was not possible to evaluate whether or not onshore-offshore sand transport is related only to storms. 95 CONCLUSIONS Fourier grain-shape analysis, analysis of variance, Hotelling’s V test, factor analysis and chi-square testing of grain-shape composition was used to identify local sand sources, to identify the boundaries of potential littoral subcells, to describe the character of sediment transportation and to describe temporal changes in these patterns for the central and eastern parts of the Santa Barbara Littoral Cell. Analysis of variance and Hotelling’s T2 test were used to identify the differences in grain-shape composition between various samples. Factor analysis and chi-square grain-shape distribution testing was used to describe the differences that were found between the samples in this study. Based on these analyses, the following conclusions can be made. 1) The Ventura and Santa Clara Rivers contain similar populations of quartz grain shapes, most of which probably are derived from the Sespe Formation. The grain-shape composition of these rivers was statistically consistent between 1985 and 1992. 2) Little, if any, first-cycle sand from granitic sources similar to that obtained from Sand Canyon Creek occurs in the rivers, shoreface or inner continental shelf within the study area. 96 3) Sand samples from the inner shelf in the Ventura area are composed of sand from both the Ventura and Santa Clara Rivers as well as at least one other source on the inner continental shelf. 4) Samples from the Ventura River, Santa Clara River and inner shelf contain a larger number of more elongate grains than most of the beach samples. 5) There is an overall increase in the asperity of sand grain shapes from the central to eastern parts of the Santa Barbara Littoral Cell. 6) In the Ventura area, littoral sand is composed of both river- and shelf-derived sand, which results in a grain-shape composition that has high asperity and moderately low elongation. 7) Littoral sand in the Santa Barbara area is derived from a combination of sediment discharged from the streams of the Santa Ynez Mountains and the onshore movement of sand from the shelf, which results in a moderately elongate grain-shape composition with a very low asperity. 8) There are no identifiable, natural sedimentary boundaries between Emma Woods Beach and Point Mugu Beach as long as dredging operations continue at the Santa Barbara, Ventura, Port Hueneme, and Channel Island Harbors. 9) Sand in the Ventura area is areally well distributed along the upper shoreface due to a net southeasterly longshore transport. 97 10) Littoral sand in the upper shoreface of the study area has a lower asperity than grains in the lower shoreface possibly due to abrasion of grains entrained in the upper shoreface and selective transport of rougher grains from the upper to the lower shoreface or, more likely, from the adjacent inner continental shelf. Selective transport also may be the cause of the observed increases in the proportion of elongate grains deposited in Ventura Harbor. 11) There is a general accumulation of more elongate grains with an moderate asperity on beaches (MLLW) during the summer months. This is probably due to the onshore transport of more elongate shelf-derived grains through ridge and runnel systems. This accumulation is mixed with pre existing backshore sand during or after storms to produce sand with an intermediate composition with respect to elongation. By the end of the oceanographic winter, storm-generated mixing produces a sand with a higher percent of more equant grains. 12) There are no long-term changes in the grain-shape composition of littoral sand samples collected at MLLW within the study area from 1969 to 1991. 98 REFERENCES Ahlschwede, K.S., 1988, Sources and net littoral transport of sand in San Diego and southern Orange Counties, southern California: Fourier grain- shape analysis: Unpublished M.S. Thesis, University of Southern California, Los Angeles, California, 135 p. Alder, H.L., and Roessler, E.B., 1977, Introduction to probability and statistics: W.H. Freeman and Company, New York, 315 p. Arnold, R., 1907, Geology and oil resources of the Summerland District: United States Geological Survey Bulletin 322, 93 p. Bailard, J.A., 1985, Beach Erosion and Seawall Assessment at Mugu Beach, California: Technical Memorandum number M-42-86-02, Naval Civil Engineering Laboratory, Port Hueneme, California, 42 p. Bendat, J.S., and Piersol, A.G., 1971, Random data: Analysis and measurement procedures: Wiley-lnterscience Incorporated, New York, 407 p. Bloom, L., 1979, The Relationships among river, beach and submarine canyon sands in the Southern Santa Barbara Littoral Cell, Ventura County, California, Fourier grain-shape analysis: Unpublished M.S. Thesis, University of Southern California, Los Angeles, California, 115 p. Bloomfield, P., 1976, Fourier analysis of time series: An introduction: Wiley- lnterscience Inc., New York, 258 p. Brigham, E.O., 1974, The fast Fourier transform: Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 252 p. Brownlie, W.R., and Taylor, B.D., 1981, Sediment management for southern California mountains, coastal plains and shoreline: California Institute of Technology, Pasadena, California, EQL Report Number 17-C, p. C26- C110. California Division of Mines and Geology, 1954, Geology of southern California: Bulletin 170, 312 p. 99 California Division of Mines and Geology, 1973, Geologic map of southern Ventura County, California; showing mineral deposits: compiled from many sources: Special Report 14, Plate 3, scale 1:48,000. California Division of Mines and Geology, 1975, Offshore surficial geology of California: Map Sheet 26, scale 1:48,000. Crowell, J.C., ed., 1975, San Andreas fault in southern California, California Division Mines and Geology Special Report number 118, 118 p. Dahlen, M.Z., Osborne, R.H., and Gorsline, D.S., 1990, Late Quaternary history of the Ventura mainland shelf, California: Marine Geology, volume 94, p. 317-340. Davis, J.C., 1973, Statistics and data analysis in geology: John Wiley & Sons Incorporated, New York, 646 p. Dibblee, T.W., 1966, Geologic of the central Santa Ynez Mountains, Santa Barbara County, California: California Division of Mines and Geology, Bulletin number 186, 340 p. Dibblee, T.W., 1982, Geology of the Santa Ynez-Topatopa mountains, southern California: in Fife, D.L., and Minch, J.A., eds., Geology and mineral wealth of the California Transverse Ranges: South Coast Geological Society, p. 7-26. Dibblee, T.W., 1992, Geologic Maps Within or Adjacent to Los Padres National Forest: Dibblee Geological Foundation, Santa Barbara, maps numbers DF-1 to DF-19, DF-21, DF-26 to DF-29, DF-34, and DF-37 to DF-42, scale 1:24000. Dixon, W.J., 1990, BMDP Statistical Software, Manual, Volumes 1-2: BMDP Statistical Software, Inc., Los Angeles, 629 p. Drake, D.E., 1972, Sediment transport on the Santa Barbara-Oxnard Shelf, Santa Barbara Channel, California: in Swift, D.J.P., Duane, D.B., and Pilkey, O.H., eds., Shelf sediment transport: Process and pattern: Stroudsburg, Dowden, Hutchinson, and Ross Incorporated, p. 307-331. Ehrlich, R., and Weinberg, B., 1970, An exact method for characterization of grain shape: Journal of Sedimentary Petrology, volume 40, p. 205-212. 100 Emery, K.O., 1960, The sea off southern California: John Wiley & Sons Incorporated, New York, 366 p. Fischer, P.J., Kreutzer, P.A., Morrison, L.R., Rudat, J.H., Ticken, E.J., Webb, J.F., Woods, M.M., Berry, R.W., Henry, M.J., Hoyt, D.H., and Young, M., 1983, Study of Quaternary shelf deposits (sand and gravel) of southern California: State of California, Department of Boating and Waterways, Beach Erosion Control Project, Sacramento, number FR 82-11, 74 p. Howard, J.D., and Reineck, H.E., 1981, Depositional facies of high-energy beach-to-offshore sequence: Comparison with low-energy sequence: American Association of Petroleum Geologists, volume 65, p. 807-830. Jackson, P.A., and Yeasts, R.S., 1982, Structural evolution of the Carpinteria Basin, western Transverse Ranges, California: American Association of Petroleum Geologists Bulletin, volume 66, number 7, p. 805-829. Kew, W.S.W., 1924, Geology and resources of a part of the Los Angeles and Ventura Counties, California: United States Geological Survey Bulletin, number 753, 202 p. Kolpack, R.L., 1986, Sedimentology of the mainland nearshore region of Santa Barbara Channel, California: in Knight, R.J., and McLean, J.R., eds., Shelf sands and sandstones: Canadian Society of Petroleum Geologists, Alberta, p. 57-72. Luyendyk, B.P., and Hornafius, J.S., 1987, Neogene crustal rotations, fault slip, and basin development in southern California: in Ingersoll, R.V., and Ernst, W.G., eds., Cenozoic basin development of coastal California: Prentice-Hall, Incorporated, Englewood Cliffs, New Jersey, p. 259-283. Nilsen, T.H., 1984, Oligocene tectonics and sedimentation of California: Sedimentary Geology, volume 38, p. 305-336. Nilsen, T.H., 1987, Paleogene tectonics and sedimentation of coastal California: jn Ingersoll, R.V., and Ernst, W.G., eds., Cenozoic basin development of coastal California: Prentice-Hall, Incorporated, Englewood Cliffs, New Jersey, p. 81-123. Noble Consultants, 1989, Coastal sand management plan, Santa B arb ara/V entura County Coastline - Appendices: Noble Consultants, Incorporated, Irvine, p. A-1 - 1-21. 101 Noble Consultants, 1989, Coastal sand management plan, Santa B arb ara/V entura County Coastline - Main Report: Noble Consultants, Incorporated, Irvine, 186 p. Norris, R.M., and Webb, R.W., 1990, Geology of California: John Wiley & Sons Incorporated, New York, 541 p. Osborne, R.H., Liu, J., and Lee, A., 1992, Evaluation of the littoral sand composition of November, 1991, and February, 1992, foreshore sample sets: Fourier grain-shape analysis: Technical Report for Oceanside Monitoring Program, Oceanside Harbor Experimental Sand Bypass System, Los Angeles District, U.S. Army Corps of Engineers, 31 p. Osborne, R.H., and Yeh, C.C., 1991, Fourier grain-shape analysis of coastal and inner continental-shelf sand samples: Oceanside Littoral Cell, Southern Orange and San Diego Counties, Southern California: Society for Sedimentary Geology, Special Publication number. 46, Tulsa, p. 51- 66. Parke, J.G., 1857, Explorations for a railroad route from the Mississippi River to the Pacific (1854 to 1856): A.O.P. Nicholson, Washington D.C., volumes 1 to 7. Pipkin, B.W., Robertson, H.S., and Mills, R.S., 1992, Coastal erosion in southern California - an overview: jn Pipkin, B.W., and Proctor, R.J., Engineering geology practice in southern California: Association of Engineering Geologists, Southern California Section, Special Publication number 4, p. 461-483. Putnam, W.C., 1942, Geomorphology of the Ventura region, California: Geology Society of America Bulletin, volume 53, p. 691-754. Schultz, D.J., 1980, The effects of hydrofluoric acid on quartz grain shape- Fourier grain shape analysis: Journal of Sedimentary Petrology, volume 50, p. 644-645. Schwarcz, H.P., and Shane, R.C., 1969, Measurement of particle shape by Fourier analysis: Sedimentology, volume 13, p. 213-231. 102 Smith, D.R., 1978, Quartz grain-surface microtexture from sediments of the Santa Clara River and Santa Barbara Littoral Cell, California: Unpublished M.S. Thesis, University of Southern California, Los Angeles, California, 98 P- Taylor, B.D., 1981, Inland sediment movements by natural processes: iri Sediment management for southern California mountains, Coastal plains and shoreline: EQL Report number. 17-B, Environmental Quality Laboratories., California Institute of Technology, Pasadena, California, p. 77-89. U.S. Army Corps of Engineers, 1985, Geotechnical Studies for Ventura Marina Sedimentation Study: Los Angeles, California, 40 p. U.S. Army Corps of Engineers, 1989, Feasibility report - Ventura Harbor, Ventura County, California - Main report and environmental assessment: Los Angeles, California, 119 p. U.S. Army Corps of Engineers, 1989, Feasibility report - Ventura Harbor, Ventura County, California - Technical report: Los Angeles, California, p. D-1 - G-47. Winterer, E.L., and Durham, D.L., 1962, Geology of southeastern Ventura Basin, Los Angeles County, California: United States Geological Survey Bulletin, number 334-H, 366 p. Yeats, R.S., 1976, Neogene tectonics of the central Ventura Basin: in Fritsche, A.E., eds., The Neogene symposium (San Francisco), Society for Sedimentary Geology, Pacific Section, p. 19-32. Yeats, R.S., 1987, Changing tectonic styles in Cenozoic basins of southern California: in Ingersoll, R.V., and Ernst, W.G., eds., Cenozoic basin development of coastal California: Prentice-Hall, Incorporated, Englewood Cliffs, New Jersey, p. 284-298. Yeats, R.S., and Rockwell, T.K., 1991, Quaternary geology of the Ventura and Los Angeles Basins, California: in Morrison, R.B., eds., Quaternary nonglacial geology: conterminous United States: Geological Society of America; Decade of North American Geology, volume K-2, p. 185-189. 103 Yerkes RF Sarna-Wojcicke, A.M., and Lajoie, K.R., 1987, Geology and Quaternary deformation of the Ventura area: United States Geologic Survey Professional Paper, number 1339, p. 169-178. 104 APPENDICES 105 APPENDIX A Listing of names and code numbers for the samples used in this study. Sample location and time code numbers are defined in Table 2. 106 Computer Sample Sample Other Known Names Codes for Location Time and Code Numbers this Studv Code # Code # for the Same Samples RORVS 01 14 6 908+05-B, used by the writer. RORVS 02 12 6 872+89-B, used by the writer. RORVS 03 3 6 — RORVS_04 27 6 Sample Bags are reversed! Bulk bag is marked as Dune A. RORVS 05 8 6 839+71-B, used by the writer. RORVS 06 4 6 — RORVS 07 5 6 712+36-B, used by the writer. RORVS 08 20 6 — RORVS 09 19 6 — RORVS 10 17 6 — RORVS 11 18 6 — RORVSJ2 26 6 Sample Bags are reversed! Bulk bag is marked as Dune B. RORVS_13 22 1 VR 1969, used by USC Sedimentology Lab. RORVSJ4 22 2 VMS-42A/R-3, used by U.S. Army Corps of Engineers. RORVSJ5 23 2 VMS-36/SCR-1, used by U.S. Army Corps of Engineers. RORVS_16 20 1 Point Mugu 1969, used by USC Sedimentology Lab. RORVS_17 4 1 Emma Woods 1969, used by USC Sedimentology Lab. RORVSJ8 12 2 VMS-17, used by U.S. Army Corps of Engineers. RORVSJ9 14 3 VMS-29, used by U.S. Army Corps of Engineers. RORVS_20 5 3 VMS-7, used by U.S. Army Corps of Engineers. RORVS_21 12 3 VMS-22, used by U.S. Army Corps of Engineers. RORVS_22 8 2 VMS-9, used by U.S. Army Corps of Engineers. RORVS_23 8 3 VMS-14, used by U.S. Army Corps of Engineers. RORVS_24 5 2 VMS-2, used by U.S. Army Corps of Engineers. 107 RORVS_25 14 2 RORVS_26 28 4 RORVS_27 29 4 RORVS_28 6 2 RORVS 29 18 7 RORVS_30 13 2 RORVS 31 20 7 RORVS 32 12 7 RORVS 33 4 7 RORVS 34 3 7 RORVS_35 4 4 RORVS 36 19 7 RORVS_37 15 2 RORVS 38 8 7 RORVS 39 2 7 RORVS 40 1 7 RORVS 41 17 7 RORVS_42 16 4 RORVS 43 5 7 RORVS_44 7 4 RORVS_45 3 4 RORVS 46 14 7 RORVS_47 9 2 RORVS_48 10 3 RORVS_49 1 1 5 RORVS_50 11 3 RORVS 51 10 5 VMS-24, used by U.S. Army Corps of Engineers. SC-16, used by Noble Consultants. SC-15, used by Noble Consultants. VMS-45, used by U.S. Army Corps of Engineers. VMS-60, used by U.S. Army Corps of Engineers. ST-7/Emma Woods 1988, used by Noble Consultants. VMS-68, used by U.S. Army Corps of Engineers. ST-10/Mandalay 1988, used by Noble Consultants. ST-8/Buena Ventura 1988, used by Noble Consultants. ST-1/Santa Barbara 1988, used by Noble Consultants. VMS-52, used by U.S. Army Corps of Engineers. VMS-78/Harbor E, used by U.S. Army Corps of Engineers. VH-90-20, used by U.S. Army Corps of Engineers. VMS-74/Harbor A, used by U.S. Army Corps of Engineers. VH-90-9, used by U.S. Army Corps of Engineers. 108 RORVS 52 21 7 RORVS_53 24 9 RORVS 54 23 9 RORVS 55 22 9 RORVS 56 25 9 RORVS 57 8 8 RORVS 58 5 8 RORVS 59 4 8 RORVS 60 17 8 RORVS 61 19 8 109 APPENDIX B Listing of the VAX/VMS C computer program BYWAVEIT written by Rory Robinson. 110 / * BYWAVEIT * / / * VAX-VMS V ersion: 6-16-92 * / / * W ritte n By Rory Robinson. * / / * For USC SED PET LAB - Dr Osborne * / / * Creates Wave Histograms from B inary .AMP * / / * F ile s , Then W rites Results to ASCII F ile s . * / / * L ib ra rie s * / #in clu d e s td io #in clu d e s t d li b # in clud e ctype / * Global V ariab les * / l'nt form; / * 1 = FACTOR; 2 = ANOVA * / in t o r ie n t; / * 1 = SAMPLES IS COLUMNS; 2 = SAMPLES IS RO W S * / in t f i l e s , g ra in s , waves; in t num_cells; f lo a t high, low; f lo a t c e ll_ fa c ; const in t m a x_ file = 100; const in t max_grains = 400; f lo a t amp_val[1 0 0 ][4 0 0 ][6 4 ]; in t h is t_ c o u n t[1 0 0 ][5 0 ]; char o l d _ f i l e s [1 0 0 ][8 0 ], n e w _ file [80 ]; char *old_p, *new_p; char prompt; / * Main Program * / m ain() C s t a r t : p r i n t_ i n f o ( ); workarea: f i nd_goaI( ) ; read_bi n_amp<); c a lc _ h ilo ( ); ch e c k _ c e lI( ) ; ca lc _ h is to g ra m s (); wri te_asc i i_ in f o ( ); done: p r i n t f ( "\nBy-Wave C a lc u la tio n s C om plete.\nM); ex i t ( 0 ) ; > * * * * * * * * * * * * * * * * * * * * * * * y / * Subroutines * / y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y help_opt i o n () C printf("\nBYW AVEIT HELP\n\n"); p r i n t f C This program w i l l create histogram s o f a s in g le F ourier ha rm on ic\n"); p r i n t f ( "from the Two Byte B inary .AMP f i l e s generated by the GSA F ourier P rog ram .\n") printf("BYWAVEIT can consider m u ltip le .AMP f i l e s at once and w i l l w r ite the d a ta \n ") p r i n t f ( " i n t o a s in g le ASCII f i l e using one of two BMDP f i l e form ats. T hese\n"); p r in tf( " fo r m a ts are fo r the Factor A n a lysis and the ANOVA or T -Test. The o u tp u t\n ") ; p r i n t f ( " f i I e s generated by t h is program can be used d ir e c t ly by BMDP. BYWAVE I T \ n " ); p rin tf("a s s u m e s th a t the .AMP f i l e s co n ta in 64 F ourier harmonics (Zero th ro u g h \n "); p r i n t f ( " t h e 63) lis te d one g ra in a f t e r a n othe r. \ n \ n " ); pauseO; p r i n t f (" The program w i ll ask fo r the number and names of a l l th e \n " ) ; p r i n t f ( " in p u t .AMP f i l e s . Do not fo rg e t to type the .AMP extension when\n"); p r in t f( " e n te r in g the f i l e names. Next, you w i l l need to type the o u tp u t\n ") ; p r i n t f ( " f i l e name w ith extension. You w i l l now be prompted fo r th e \n " ) ; p rin tf("n u m b e r of gra in s in each input f i l e . Note: Each f i l e may h a v e \n "); p rin tfC 'm o re than t h is number of g ra in s ; but NO LESS. From th is p o in t o n \n "); p r in t f( " y o u need on ly fo llo w the program menus to s e le c t the o u tp u t\n "); p r in t f( " fo r m a t you w is h .\n \n " ) ; pauseO ; p r i n t f C Once BYWAVEIT has determined the d e fa u lt design fo r the h is to g ra m ,\n "); p r in t f( " y o u w i l l have the o p tio n to a lte r the Upper and Lower boundary of th e \n " ) ; p r in tf( " h is to g r a m , as w ell as the number of c e lls in i t . Note: the maximum\n"); p r in t f( " p o s s ib le number of c e lls fo r any given histogram is 5 0 .\n \n " ) ; p r i n t f C Be sure to RECORD the histogram design (Upper and Lower B ou nd arie s\n"); p r in tf( " a n d the number of c e lls ) , because t h is in fo rm a tio n w i l l NOT be w r it t e n \ n " ); p r i n t f ( " i n t o the output f i l e , and thus, w i l l be LOST! \ n \ n " ); pa use( ) ; p r i n t f ( " The Factor A na lysis format can be created in two ways. F ir s t , w ith \n " ) ; p rin tf("s a m p le s as the columns and the c e lls from lowest to highest as the ro w s .\n "); p r i n t f ( "Second, w ith the c e lls from lowest to highest as the columns and th e \n M); p r i n t f ( "samples as the rows. \ n \ n " ); p r i n t f C The ANOVA of T-Test ouput f i l e format is b e lo w :\n "); p r i n t f C Sample # C ell # C ell C o un t\n"); p r i n t f C (Note: C ell number 1 is always the lowest cel I . ) \ n \ n " ); pause( ) ; p r in t f C T h is program can c u r r e n tly read up t o : \ n " ) ; p rin tf("% 3 d .AMP input f i l e s c o n ta in in g up to \n " , m a x f i l e ) ; p rin tf("% 3 d gra in s at a tim e \n \n ", max_grains); pauseO; ex i t ( 0 ); J 'k - k - k - k 'k 'k 'k 'k - k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k - k 'k - k 'k - k - k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k -k 'k 'k -k -k 'k 'k 'k 'k * ★ f pauseO / * Pauses Screen T i l l Key Pressed * / C fflu s h ( s td in ); p r i n t f ( "<Press RETURN to continue.;-" ); prompt = g e tc h a rO ; p r i n t f ( " \ n \ n " ) ; j* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *j p r i n t_ i n f o ( ) C pr i n t f ( 1 1 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ypn^- p r i n t f C * BYWAVEIT - W ritte n by Rory Robinson * \ n " ) ; p r i n t f C * (c) Copyright 1992 * \ n " ) ; p r i n t f C * VAX-VMS Version: 6-16-92 * \ n M); p r i n t f C * Creates histrogram s from b in a ry .AMP * \ n " ) ; p r i n t f C * f i l e s fo r s t a t is t ic a l a n a ly s is . * \ n " ) ; pr i nt f ( * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * \ n M)■ p r i n t f C * Type: 'O' at the Number of F ile s * \ n " ) ; p r i n t f C * prompt to see program d ir e c tio n s . * \ n " ) ; pr i nt f ( n* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ) • > 112 y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y f i nd_goaI() { in t z, loop, match; j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y / * ENTER IN NUMBER OF FILES - HELP OPTION ALSO * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y f i nd_a: f flu s h ( s td in ); p r i n t f ("\n E n te r Number of F ile s to Convert (ty p e 'O' fo r he lp )\n ? " ) ; z = s c a n f("X d ", S f ile s ); p ri n t f ( H\ n " ); i f (z == 0) f p r i n t f C * ERROR - Bad Format - Enter an in te g e r nu m ber!\n"); goto fin d _ a ; > i f ( f i l e s == 0) h e lp _ o p tio n (); else i f ( f i l e s < 0) C p r i n t f C * ERROR - Minimum Number of F ile s = 1 \nM); goto fin d _ a ; > else i f ( f i l e s > m a x _ file ) C p r i n t f C * ERROR - Maximum Number of F ile s = %3d\n", m a x _ file ); goto fin d _ a ; > y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * LOOP TO ENTER INPUT FILE NAMES * / y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y fo r ( lo o p = 0; loop < f i l e s ; loop++ ) C fflu s h ( s td in ); p r in t fC \n E n t e r F ile Name fo r b in a ry input f i l e number %2d\n? ", (lo o p + 1 )); old_p = gets( & o ld _ file s [ lo o p ] [0] ); p r i n t f C \ n " ) ; fo r (z=0; z < s trle n (o ld _ p ); z++) { old_ p[z] = to u p p e r(o ld _ p [z ]); } > ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j /* ENTER IN OUTPUT FILE NAME * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j f i nd_b: fflu s h ( s td in ); p r in t fC \n E n t e r ASCII Output F ile Name\n? " ) ; new_p = gets( & ne w _ file[0] ); p r i n t f C \ n " ) ; fo r (z=0; z < strlen(n ew _p); z++) ( new_p[z] = toupper(new _p[z]); > match = 0; fo r (z=0; z < file s ; z++) ( i f ( ! s trcm p (& n e w _ file [0 ], & o ld _ f ile s [ z ] )) match = 1; 113 > i f (match = = 1 ) { p r i n t f C * ERROR - Output F ile Name Must Be D iffe r e n t From Input F ile Nam es!\n"); goto fin d _ b ; > I *****************************I /* ENTER IN NUMBER OF GRAINS * / f i nd_c: fflu s h ( s td in ); p r i n t f ("\n E n te r Number of Grains in Each Input F ile \n ? 1 1 ); z = s c a n f("/id ", &grains ); p r i n t f ( " \ n " ) ; i f (z == 0) f p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ) ; goto fin d _ c ; > i f (g ra in s < 2) { p r i n t f C * ERROR - Minimum Number of Grains = 2 \ n " ); goto fin d _ c ; > else i f (g ra in s > max__grains) f p r i n t f C * ERROR - Maximum Number of Grains = %3d\n", max_grains); goto fin d _ c ; > j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j /* CHOICE FOR BMDP FILE FORMAT * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * J f i nd_d: f flu s h ( s td in ); p r i n t f ("\n O p tio n s fo r Output F ile F o rm a t:\n \n "); p r in tf( " < 1 > Columnar Format (BMDP Factor A n a ly s is ). \ n " ) ; p r in tf( " < 2 > BMDP ANOVA or T-Test F o rm a t.\n \n M); p r in tf( " T y p e 1 or 2 and press R eturn.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == ' 1 ' ) form = 1; else i f (prompt == ' 2 ' ) form = 2; else goto fin d _ d ; j j /* O rie n ta tio n of Columnar Format */ ^ -k -k ifk -k -k " k ic k -k ic -k -k -ie -k -k -k -k -k -k -k -k -k -k -k -k ic k -k -k -k -k -k -k -k -k -k j i f (form == 1) < f i nd_e: fflu s h ( s td in ); p r i n t f ("\n O p tio n s fo r Columnar O r ie n ta tio n :\n \n " ) ; p r in tf( " < 1 > Samples are Columns, C e lls are R ow s.\n"); p r in tf( " < 2 > C e lls are Columns, Samples are R o w s.\n \n"); p r in tf( " T y p e 1 or 2 and press R eturn.\n? " ) ; 114 prompt = g e tc h a rO ; p r in t f ( " X n " ) ; i f (prompt == '1 ') o rie n t = 1; else i f (prompt == ' 2 ' ) o rie n t = 2; else goto fin d _ e ; > J * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * J / * Determine Which Wave to Transfer * / z* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * I f i n d _ f: fflu s h ( s td in ); p r i n t f ("\n E n te r Harmonic Number to be used to Create Histograms (2 - 63)\n? z = scan fC % d ", Swaves ); p r i n t f C \ n " ) ; i f (z == 0) < p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ); goto f in d _ f; > i f (waves < 2) C p r i n t f C * ERROR - Lowest A llow able Wave = 2 \n M); goto fin d _ f; > else i f (waves > 6 3 ) { p r i n t f C * ERROR - Maximum A llow able Wave = 6 3 \n "); goto fin d _ f; > read_bin_amp() C FILE * fp ; in t z, fn , gn, hn, e r r ; prin tf("X n R e a d in g From B inary Input F ile s .X n "); ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * LOOP TO READ BINARY DATA FROM FILES * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j f o r ( f n = 0; fn < f i l e s ; fn++ ) { i f ( ( f p = fopen ( & o l d _ f i le s [ f n ] [0 ], "rb " )) == 1\ 0 1) { p r i n t f C * ERROR - Cannot Open Input Fi le! \nO peration Aborted! \ n " ) ; ex i t ( 0 ) ; > I *********************************i /* LOOP TO READ NUMBER OF GRAINS * / I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j fo r(g n = 0; gn < g ra in s ; gn++ ) C 115 / * LOOP TO READ NUMBER OF HARMONICS * / I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j fo r(h n = 0; hn < 64; hn++ ) { e rr = fre a d ( &_val [fn ] [gn] [h n ], s iz e o f(a m p _ v a l[fn ][g n ][h n ]), 1, fp ); i f ( ( e r r == 0) && ( f e o f ( f p ) != 0 )) { p r i n t f C * ERROR - F ile Read E rro r! \nOperat ion Aborted! \ n " ) ; e x it ( 0 ); > > > fc lo s e ( fp ); > > I *********************************************************j w rite _ a s c i i_ in f o ( ) L FILE * fp ; in t z, fn , gn, c l, e r r ; p r i n t f ("\nS a vin g Output F ile . \ n " ) ; I j /* OPENING OUTPUT FILE * / j***********************j i f ( ( f p = fopen (& n e w _ file [0 ], "w" )) == 1\ 0 1) < p r i n t f C * ERROR - Cannot Create Output Fi le! \nO peration Aborted! \ n " ) ; ex i t ( 0 ) ; > ^ * * * * * * * * * * * * * * * * * * * * * * f /* LOOP TO WRITE DATA * / I ********■*■*■*■*■*■*■*■*■*■*■*■*■*■*■ j i f (form = = 1 ) / * FACTOR ANALYSIS FORMAT * / t i f (o r ie n t = = 1 ) { f o r ( c l = 0; c l < num_cells; cl++) { fo r ( fn = 0; fn < f i l e s ; fn++) { f p r i n t f ( f p , "%d " , h is t_ c o u n t[fn ] [ c l ] ); > f p r i n t f ( f p , "X n "); > > else / * ORIENT IS 2 * / { f o r ( fn = 0; fn < f i l e s ; fn++) t fo r ( c l = 0; c l < num_cells; cl++) { f p r i n t f ( f p , "%d ", h is t_ c o u n t[fn ] [ c l ] ); > f p r i n t f ( f p , " \ n " ) ; 116 > > e lse i f (form == 2) / * ANOVA - T-TEST FORMAT * / C fo r ( fn = 0; fn < f i l e s ; fn++) C fo r ( c l = 0; c l < num_cells ; cl++) C f p r i n t f ( f p , "%3d %2d %3d\n", (fn + 1 ), (c l+ 1 ), h i s t _ c o u n t [ f n ] [ c l] ); > > > fc lo s e ( fp ); / * CLOSE OUPUT FILE * / > ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * J c a lc _ h ilo ( ) C in t fn , gn; p rin tf(" \n D e te rm in in g Histogram C ell B o u n d rie s .\n "); high = anp_val [0] [0] [waves]; low = amp_val[0] [0] [waves]; fo r ( fn = 0 ; fn < f i l e s ; fn++) C fo r (gn = 0; gn < g ra in s ; gn++) C i f (high < a m p _ v a l[fn ][g n ][w a v e s ]) high = am p_vaI[fn][gn] [waves]; i f (low > am p _val[fn][gn] [waves]) low = am p_vaI[fn][gn] [waves]; > > / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j ch e c k _ c e lI( ) L in t z; num_cells = 20; check_a: p p j n \ n****************************************\n\n. . j - p r i n t f ( "H istogram D e s ig n :\n \n "); p rin tf(" L o w e r histro gra m boundary = % f\n", low); p rin tf(" U p p e r histro gra m boundary = % f\n", h ig h ); prin tf("N u m b e r of c e lls = %2d\n", num _cells); p p j ^ II ^ p ^* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ) • check_b: fflu s h ( s td in ); pr i n t f ("\nO pt i ons: \ n \ n " ) ; p r in tf( " < 1 > Accept C u rre n tly Defined H is to g ra m .\n "); p r in tf( " < 2 > Change Defined P a ra m e te rs .\n \n "); p r in tf( " T y p e 1 or 2 and press R eturn.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; 117 i f (prompt == ' 2 ') f ^ * * * * * * * * * * * * * * * * * * y / * Input New Data * / ^ * * * * * * * * * * * * * * * * * * y p r i n t f ("\n C u rre n t Lower Histrogram Boundary = % f\n", low); check_c: fflu s h ( s td in ); p r i n t f ( “ \nE nter New Lower Histogram Boundary\n? " ) ; z = s ca n f("% f", Slow ); p r i n t f ( " \ n " ) ; i f (z == 0) f p r i n t f C * ERROR - Bad Format - Enter an in te g e r nu m be r!\n"); goto check_c; > i f ( Iow < 0.0 ) f p r i n t f C * ERROR - No Negative Values A llo w e d !\n "); goto check_c; > else i f (low > 99.0) t p r i n t f C * ERROR - Maximum A llow able Value = 9 9 "); goto check_c; > p r in tfC \n C u r r e n t Upper Histrogram Boundary = % f\n "f h ig h ); check_d: fflu s h ( s td in ); p r in t fC \n E n t e r New Upper Histogram Boundary\n? " ) ; z = sc a n f("% f", Shigh ); p r i n t f C \ n " ) ; i f (z == 0) C p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ) ; goto check_d; > i f ( high <= Iow) t p r i n t f C * ERROR - Upper Value Must be G reater than Lower Value! \ n M); goto checked; > else i f (high > 100.0) f p r i n t f C * ERROR - Maximum A llow able Value = 100"); goto check_d; > p r in tfC \n C u r r e n t Number of C e lls = %2d\n", num _cells); check_e: fflu s h ( s td in ); p r i n t f ("\n E n te r New Number o f C e lls \n ? " ) ; z = scanfC% dMf &num_cells ); p r i n t f C \ n " ) ; i f (z == 0) f p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ); goto check_e; > i f (num_cells < 10) f 118 p r in t fC 1 1 * ERROR - Minimum Number of C e lls = 1 0 \n "); goto check_e; > else i f (num _cells > 50) C p r in t fC " * ERROR - Maximum Number of C e lls = 5 0 "); goto check_e; > goto check_a; > / * * * * * * * * * * * * * * * * * * J / * E x it From Loop * / y * * * * * * * * * * * * * * * * * * y else i f Cprompt != 111) goto check_b; > y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y calc_histogram sC) f in t z, fn , gn, c l; p r in t fC "\nD eterm ining H istogram s. \ n " ); / * * * * * * * * * * * * * * * * * * * * * j / * Set C e lls to Zero * / fo r Cfn = 0; fn < f i l e s ; fn++) f fo r Cel = 0; c l < num_cells; cl++) C h is t_ c o u n t [f n ][ c l] = 0; > > y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y / * Determ ining C e ll Factor * / c e ll_ fa c = Chigh - low)/num _celIs; y********************************j /* Determ ining Histogram Makeup * / y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y fo r Cfn = 0; fn < f i l e s ; fn++) f fo r Cgn = 0; gn < g ra in s ; gn++) f fo r Cel = 0; c l < num_cells; cl++) C i f Cel < Cnum_cells - 1)) f i f C C am p_val[fn][gn][w aves] >= C low + C ce ll_ fa c*cl)) ) && C am p_vaI[fn][gn][w aves] < Clow+Ccell_fac*Cc1+1))) ) ) h is t_ c o u n t[fn ][c l]+ + ; > else C i f C C am p_val[fn][gn][w aves] >= C low + C ce ll_ fa c*cI)) ) && 119 > } ( am p_valtfn][gn][w aves] <= ( lo w + (c e ll_ fa c *(c L + 1 ) ) ) ) ) h is t_ c o u n t[fn ][c l]+ + ; / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j /* END SOURCE CODE - BYWAVE IT * / ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j 120 APPENDIX C Listing of the VAX/VMS C computer program CONVRTIT written by Rory Robinson. 121 / * CONVRTIT * / / * VAX-VMS V ersion: 2-2-92 * / / * W r itte n By Rory Robinson. * / / * For USC SED PET LAB - Dr Osborne * / / * C onverts B in a ry .AMP F ile s to ASCII F ile s . * / / * L ib ra rie s * / #inclu de s td io #include s t d li b #include ctype / * Global V ariab les * / in t average; / * 0 = NO; 1 = YES; MUST BE SET SEPARATELY * / in t form; / * 1 = FACTOR; 2 = ANOVA; 3 = DISCRIMINANT * / in t o r ie n t; / * 1 = SAMPLES IS COLUMNS; 2 = SAMPLES IS ROW S * / in t dtype; / * RANGE OF 1 TO 6 - DEPENDS ON FORMAT VAR FOR MEANING * / in t s e lc [2 4 ]; i n t f i l e s , g ra in s , waves; co n st i n t m a x _ file = 100; co n st i n t m ax_grains = 400; f lo a t amp_val [100] [400] [6 4 ]; f lo a t ave_amp[100] [64] ; char t e x t [8 0 ], o l d _ f i l e s [1 0 0 ][8 0 ], n e w _ file [100]; char *te s t_ p , *o ld _p, *new_p; char prompt; / * Main Program * / ma i n() C s t a r t : p r i n t_ i n f o ( ) ; w o rk a re a : f i n d _ go a I( ) ; read_bi n_amp() ; i f (average = = 1 ) calc_average(); wri te_asci i_amp(); done: p r i n t f ( "\nC onversion C a lc u la tio n s C o m p le te .\n "); e x it( 0 ); } I *********************************************************I / * Subroutines * / I *********************************************************I h e lp _ o p t i o n ( ) C pr i n t f ("\nC0NVRTIT HELP\n\n"); p r i n t f C This program w i l l convert the Two Byte B inary .AMP f i l e s \ n " ) ; p rin tf(" g e n e ra te d by the GSA F ourier program in to ASCII f i l e s fo r use in \n " ) p r in t f ( " o t h e r s t a t is t ic s packages. F urthe r, the program can average th e \n ") p r in t f( " F o u r ie r harmonics fo r each .AMP f i l e . \ n \ n " ) ; p r i n t f C CONVRTIT can transform m u ltip le .AMP f i l e s at once and w i l l \ n " ) p r i n t f ( " w r it e the data in to a s in g le ASCII f i l e using one of three BMDP\n">; p r i n t f ( " f i l e form ats. These formats are fo r the Factor A n a lysis, th e \n "); printf("ANOVA or T -Test, and the D iscrim in ant Function A n a ly s is . The\n">; p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f pause( p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f pause( p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f pause( p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f p r i n t f pause( p r i n t f p r i n t f p r i n t f p r i n t f pause( p r i n t f p r i n t f p r i n t f pause( ex i t ( 0 ouput f i l e s generated by th is program can be used d i r e c t ly b y \n "); BMDP.\n\nM); CONVRTIT assumes th a t the .AMP f i l e s con tain 64 F o u r ie r \n " ); harmonics (Zero through the 63) lis te d one g ra in a fte r a n o th e r. \ n " ); Note: This program a u to m a tic a lly excludes the Zero harmonic when\n"); c re a tin g the output f i l e . \ n \ n " ) ; The program w i ll ask fo r the number and names of a l l th e \n " ) ; input .AMP f i l e s . Do not fo rg e t to type the .AMP extension when\n"); e n te rin g the f i l e names. Next, you w i ll need to type the o u tp u t\n "); f i l e name w ith extension. You w i l l now be prompted fo r th e \n " ) ; number of gra in s in each input f i l e . Note: Each f i l e may h a ve \n "); more than th is number of g ra in s ; but NO LESS. From th is p o in t o n \n "); you need on ly fo llo w the program menus to s e le c t the o u tp u t\n "); format you w is h . \n \ n " ) ; The Factor A nalysis format can be created in two w a y s .\n "); F ir s t , w ith samples as the columns and the harmonics as the ro w s .\n "); Second, w ith harmonics as the columns and the samples as th e \n " ) ; rows. The f i r s t format supports output of Raw data ( a l l \ n " ) ; harmonics fo r a l l g ra in s ), Averaged data (th e average of e a c h \n "); harmonic fo r th a t .AMP f i l e ) . Selected data (your choice of th e \n " ) ; f i r s t 24 harmonics fo r a l l g ra in s ), and Averaged Selected d a ta \n "); (th e average of each of your choice of the f i r s t 24 harmonics) . \ n " ); The second format supports only Averaged data and Averaged\n"); Selected d a ta .\n \n " ) ; The ANOVA or T-Test format supports three output m e th o d s,\n "); which in clud e: Raw data, Averaged data, and S ingle Harmonic d a ta \n "); ( in which on ly a s in g le harmonic number is o u tp u t). The o u tp u t\n ") ; f i l e format is b e lo w :\n \n "); Raw data - 4 columns as fo llo w s : \ n " ) ; Sample # Grain # Harmonic # Fourier Harm onic\n"); Averaged data - 3 columns as f o llo w s : \ n " ) ; Sample # Harmonic # Fourier Harm onic\n"); S ingle data - 3 columns as fo llo w s : \ n " ); Sample # Grain # F ourier H arm onic\n\n"); The D iscrim in ant Function A na lysis format supports tw o \n "); output methods, which in clud e: Raw data (Maximum of only 2 4 \n "); harmonics) and Selected data. The output f i l e format is be Iow: \ n \ n " ); Sample # [harmonics, from 1 ( l e f t ) to 24 ( r i g h t ) ] \ n \ n " ); This program can c u r r e n tly read up t o : \ n " ) ; %3d .AMP input f i l e s con tain ing up to \n " , m a x _ file ); %3d g ra in s at a tim e \n \n ", max_grains); ); p a u se() / * Pauses Screen T ill Key Pressed * / C f f l u s h ( s t d in ) ; p r i n t f ( "<P ress RETURN to c o n tin u e .>" ); prompt = g e tc h a r O ; p r i n t f ( " \ n \ n M) ; > p r i n t _ i n f o ( ) 123 c pr i n t f ( " \ n * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * \ n " ) ; p r in tfC " * CONVRTIT - W ritte n by Rory Robinson * \ n " ) ; p r i n t f C * (c) Copyright 1992 * \ n M); p r i n t f C * VAX-VMS V ersion: 2-2-92 * \ n M); p r i n t f C * Converts b in a ry .AMP f i l e s to ASCII * \ n " ) ; p r i n t f C * f i l e s fo r s t a t is t ic a l a n a ly s is . * \ n M); p r in t fC 1 1 \n M) 1 p r i n t f C * Type: '0 ' at the Number of F ile s * \ n " ) ; p r i n t f C * prompt to see program d ir e c tio n s . * \ n M); > f i nd_goalC) t in t z, loop, match; average = 0; / * ENTER IN NUMBER OF FILES - HELP OPTION ALSO * / f i nd_a: fflushC s td in ); p r in t fC \n E n t e r Number of F ile s to Convert (type '0 ' fo r he lp )\n ? " ) ; z = scanfC% d", S file s ); p r i n t f C \ n M); i f (z == 0) p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ) ; goto fin d _ a ; > i f ( f i l e s == 0) h e lp _ o p tio n (); else i f ( f i l e s < 0) t p r i n t f C * ERROR - Minimum Number of F ile s = 1\nN); goto find _a ; > else i f ( f i l e s > m a x_ file ) t p r i n t f C * ERROR - Maximum Number of F ile s = %3d\n", max f i l e ) ; goto fin d _ a ; > / * ENTER IN INPUT FILE NAMES * / fo r ( lo o p = 0; loop < f i l e s ; loop++ ) t f flu s h ( s td in ); p r i n t f ("\n E n te r F ile Name fo r b in a ry input f i l e number %2d\n? ", (lo o p + 1 )); old_p = gets( & o ld _ file s [ lo o p ] [0] ); p r i n t f C \ n M); fo r (z=0; z < s trle n (o ld _ p ); z++) T old_ p[z] = to u p p e r(o ld _ p [z ]); } > / * ENTER IN OUTPUT FILE NAME * / f ind_b: fflu s h ( s td in ); p rin tf(" X n E n te r ASCII Output F ile Name\n? " ) ; new_p = gets( & new _file[0] ); p r i n t f C \ n M); fo r (z=0; z<strlen(new _p); z++) T new_p[z] = toupper(new _p[z]); > 124 match = 0; fo r (z=0; z < file s ; z++) i f ( ! strcm p(& ne w _ file[0 ], & o ld _ f ile s [ z ] )) match = 1; > i f (match == 1) p r i n t f C * ERROR - Output F ile Name Must Be D iffe r e n t From Input F ile Nam es!\n"); goto fin d _ b ; > / * ENTER IN NUMBER OF GRAINS * / f i nd_c: f flu s h ( s td in ); p r i n t f ("\n E n te r Number of Grains in Each Input F ile \n ? " ) ; z = sca n fC X d ", &grains ); p r i n t f C \ n n); i f (z == 0) p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \n n); goto fin d _ c ; > i f (g ra in s < 2) p r i n t f C * ERROR - Minimum Number of Grains = 2 \n " ) ; goto fin d _ c ; > else i f (g ra in s > max_grains) t p r i n t f C * ERROR - Maximum Number of Grains = %3d\n", max_grains); goto fin d _ c ; > j*******************************j / * CHOICE FOR BMDP FILE FORMAT * / !*******************************j f i nd_d: f flu s h ( s td in ); p r i n t f ("\n O p tions fo r Output F ile F o rm a t:\n \n "); p r in tfC < 1 > Columnar Format (BMDP Factor A n a ly s is ) .\n " ) ; p r i n t f ( n<2> BMDP ANOVA or T-Test Format. \ n " ); p r i n t f ( M <3> BMDP D iscrim in a n t Function F o rm a t.\n \n "); p r i n t f ("Type 1, 2, or 3 and press Return. \n? 1 1 ); prompt = g e tc h a rO ; p r in t f C X n " ) ; i f (prompt == ' 1 ' ) form = 1; else i f (prompt == '2 ') form = 2; else i f (prompt == '3 ') form = 3; else goto fin d _ d ; y*************************************j / * LOGIC SEPARATION FOR FORMAT INPUT * / f - k - k - k ic * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j i f (form == 1) / * OUTPUT ORIENTATION INPUT * / f i nd_e: f flu s h ( s td in ); 125 p r i n t f ("\n O p tio n s fo r Columnar O r ie n ta tio n :\n \n " ) ; p r in tf( " < 1 > Samples are Columns, Harmonics are R ow s.\n"); p r in tf( " < 2 > Harmonics are Columns, Samples are R o w s.\n \n"); p r in tf( " T y p e 1 or 2 and press R eturn.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == ' 1 ' ) o rie n t = 1; e lse i f (prompt == ' 2 ' ) o rie n t = 2; e lse goto find _e ; / * LOGIC SEPARATION FOR COLUMN ORIENTATION INPUT * / i f ( o r ie n t == 1) C / * DATA TYPE INPUT * / f i n d _ f: fflu s h ( s td in ); p r i n t f ("\n O p tio n s fo r (Columns = Samples) Data O utput: \ n \ n " ); p r in tf( " < 1 > Raw D a ta .\n M); p r in tf( " < 2 > Averaged D a ta .\n "); p r in tf( " < 3 > Selected D a ta .\n "); p r in tf( " < 4 > Averaged Selected D a ta .\n \n "); p rin tf( " T y p e 1, 2, 3 or 4 and press Return.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == ' 1 ' ) dtype = 1; e lse i f (prompt == ' 2 ' ) dtype = 2; e lse i f (prompt == l 3 ‘ ) dtype = 3; e lse i f (prompt == ' 4 ' ) dtype = 4; e lse goto fin d _ f; / * LOGIC SEPARATION FOR COLUMN ORIENTATION INPUT * / > else C / * DATA TYPE INPUT * / f ind_g: f flu s h ( s td in ); p r i n t f ("\n O p tio n s fo r (Columns = Harmonics) Data O utput: \ n \ n " ); p r in tf( " < 1 > Averaged D a ta .\n "); p r in tf( " < 2 > Averaged Selected Data. \ n \ n " ); p r i n t f ( "Type 1 or 2 and press Return.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == ' 1 ' ) dtype = 2; else i f (prompt == ' 2 ' ) dtype = 4; else goto fin d _ g ; / * LOGIC END FOR COLUMN ORIENTATION INPUT * / > ^ ic 'k ic 'k 'k ic ic ic ic ic ic ic ic ic ic ic ic ic ic ic ic ic 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k ^ / * LOGIC SEPARATION FOR FORMAT INPUT * / / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j > else i f (form == 2) 126 c / * DATA TYPE INPUT * / f i nd_h: f flu s h ( s td in ); p r i n t f ("\n O p tio n s fo r (ANOVA and T-Test) Data O utput: \ n \ n " ); p r in tf( " < 1 > Raw D a ta .\n "); p r in tf( " < 2 > Averaged D a ta .\n M); p r in tf( " < 3 > S in g le Harmonic D a ta .\n \n "); p r in tf( " T y p e 1, 2, or 3 and press R eturn.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == ' 1 ' ) dtype = 1; e lse i f (prompt == ' 2 ' ) dtype = 2; e lse i f (prompt == ' 3 ' ) dtype = 5; e lse goto fin d _ h ; I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * LOGIC SEPARATION FOR FORMAT INPUT * / > else C / * DATA TYPE INPUT * / f i nd_ i : f flu s h ( s td in ); p r i n t f ("\n O p tions fo r (D is c rim in a n t Function) Data O utput: \ n \ n " ); p r in tf( " < 1 > Raw D a ta .\n "); p r in tf( " < 2 > Selected D a ta .\n \n "); p r in tf( " T y p e 1 or 2 and press Return.\n? " ) ; prompt = g e tc h a rO ; p r i n t f ( " \ n " ) ; i f (prompt == '1') dtype = 6; / * SPECIAL CASE * / e ls e i f (prompt == ' 2 ' ) dtype = 3; else goto f in d _ i; I* LOGIC END FOR FORMAT INPUT * / > j •k ic 'k 'k 'k -k -k -k -k -k -k -k -k -k -k -k 'k iC 'k 'k 'k 'k 'k iC 'k ic iC 'k 'k 'k 'k 'k 'k 'k -k -k -k j / * TURN O N AVERAGE VAR SWITCH * / ^ 'k ic - k - k 'k 'k 'k id c iC 'k iC 'k iC 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k iC 'k iC 'k 'k 'k 'k 'k J i f (dtype == 2) average = 1; i f (dtype = = 4 ) f average = 1; waves = 24; > ^ ic ic 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k i c iC 'k iC 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k i c 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k ic 'k 'k 'k ic ic ^ /* INPUT OF HARMONIC BOUNDRIES O N BASIS OF DTYPE VAR * / j - k • k 'k 'k 'k 'k -k -k -k - k - k - k if'k -k 'k 'k 'k 'k 'k 'k 'k 'k 'k * * * * * * * * * * * * * * * * ★ ★ ★ * * * * * * * * * * J i f ( ( d ty p e ==1) J| (d ty p e == 2 )) f 127 f i n d _ j: fflu s h ( s td in ); p r i n t f ( M \nE nter Number of Harmonics to Transfer (2 - 63)\n? " ) ; z = scan f("X d", Swaves ); p r i n t f ( " \ n " ) ; i f (z == 0) p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \n " ) ; goto f in d _ j; > i f (waves < 2) p r i n t f C * ERROR - Minimum Number of Harmonics = 2 \n " ) ; goto f in d _ j; > else i f (waves > 6 3 ) p r i n t f C * ERROR - Maximum Number of Harmonics = 6 3 \n "); goto f in d _ j; > / * INPUT OF HARMONIC BOUNDRIES ON BASIS OF DTYPE VAR * / ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j > else i f (dtype = = 5 ) f i nd_k: fflu s h ( s td in ); p rin tf(" X n E n te r Harmonic Number to Create ANOVA/T-Test F ile With (1 - 63)\n? " ) ; z = scanf("%dM, &waves ); p r i n t f ( " \ n " ) ; i f (z == 0) C p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! Xn1 1 ); goto fin d _ k ; > i f (waves < 1) p r i n t f C * ERROR - Minimum Harmonics Number = 1 \n "); goto fin d _ k ; > else i f (waves > 6 3 ) p r i n t f C * ERROR - Maximum Harmonics Number = 6 3 \n M); goto fin d _ k ; > j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * INPUT OF HARMONIC BOUNDRIES O N BASIS OF DTYPE VAR * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j > else i f (dtype == 6) f i n d _ l : f flu s h ( s td in ); p rin tf(" X n E n te r Number of Harmonics to Transfer (2 - 24)\n? 1 1 ); z = scanf ("/id ", &waves ); p r i n t f C \ n M); i f (z == 0) C 128 p r i n t f C * ERROR - Bad Format - Enter an in te g e r num ber!\n"); goto f in d _ l; > i f (waves < 2) £ p r i n t f C * ERROR - Minimum Number of Harmonics = 2 \n M); goto fin d _ l; > e lse i f (waves > 2 4 ) £ p r i n t f C * ERROR - Maximum Number of Harmonics = 24\nM); goto fin d _ l; > I *****************************************************j / * INPUT OF HARMONIC BOUNDRIES O N BASIS OF DTYPE VAR * / !*****************************************************j > else i f ((d ty p e == 3) jJ (dtype == 4 )) £ find_m: p r i n t f ("\n S e le c t Which Harmonics to T r a n s fe r.\n \n " ) ; fo r ( lo o p = 1 ; loop <= 24; loop++ ) £ f i nd_n: fflu s h ( s td in ); p r in tf( " T r a n s fe r Harmonic Number %2d (Y/N)? " , loop); prompt = g e tc h a rO ; p r i n t f C \ n " ) ; i f ((prom pt == 1Y1) JJ (prompt == l y ' ) ) s e lc llo o p ] = 1; else i f ((prompt == 1N1) [[ (prompt == ' n ' ) ) s e lc llo o p ] = 0; e lse goto find _n ; > p r in tf( " \n T h e fo llo w in g Harmonic Numbers w i ll be T ra n s fe re d :\n \n "); fo r ( lo o p = 1 ; loop <= 24; loop++ ) i f (s e lc llo o p ] == 1) p rin tf("% 2 d , ", loop); > p r i n t f ( " \ n \ n " ) ; f i nd_o: f flu s h ( s td in ); p r i n t f C I s the above l i s t Correct (Y/N)? " ) ; prompt = g e tc h a rO ; p r i n t f C \ n " ) ; i f ((prom pt == 1N1) jj (prompt == ' n ' ) ) goto find_m; else i f ((prompt != ’ Y1) && (prompt != ' y 1)) goto f i n d o ; j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j /* END OF INPUT OF HARMONIC BOUNDRIES O N BASIS OF DTYPE VAR * / ^ / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y > > 129 I *********************************************************I read_bi n_amp() C FILE * fp ; in t z, fn , gn, hn, e r r ; p r i n t f ("\nReading From B inary Input F ile s A n " ) ; / * LOOP TO READ BINARY DATA FROM FILES * / f o r ( f n = 0 ; fn < f i l e s ; fn++ ) C i f ( ( f p = fopen ( & o ld _ f ile s [ f n ] [0 ], "r b " )) == 1\ 0 1) C p r i n t f C * ERROR - Cannot Open Input Fi le! \nO peration A b o rte d !\n "); e x it( 0 ); > / * LOOP TO READ NUMBER OF GRAINS * / fo r(g n = 0; gn < g ra in s ; gn++ ) C / * LOOP TO READ NUMBER OF HARMONICS * / fo r(h n = 0; hn < 64; hn++ ) C e r r = fread( &_val [fn ] [gn] [hn] , si zeof (amp_va I [ f n] [gn] [hn] ), 1, fp ); i f ( ( e r r == 0) && ( f e o f ( f p ) != 0)) C p r i n t f C * ERROR - F ile Read E r r o r ! \nOperat ion Aborted! \ n " ); ex i t ( 0 ) ; } } > fc lo s e ( fp ); > > j*********************************************************j w rite _ a s c i i_amp() C FILE * f p ; in t z, fn , gn, hn, e r r ; p r in tfC \n S a v in g Output F ile . \ n n); / * OPENING OUTPUT FILE * / i f ( ( f p = fopen (& n e w _ file [0 ], "w" )) == 1\ 0 1) < p r i n t f C * ERROR - Cannot Create Output Fi le! \nOperat ion Aborted! \ n " ) ; ex i t ( 0 ); > / * LOOP TO WRITE DATA * / i f (form = = 1 ) / * FACTOR ANALYSIS FORMAT * / C i f ( o r ie n t =- 1) C i f (dtype = = 1 ) / * RAW DATA * / C fo r (gn = 0; gn < g ra in s ; gn++) C f o r (hn = 1; hn <= waves; hn++) C 130 f o r ( f n = 0; fn < f i l e s ; fn++) { f p r i n t f ( f p , "%f ", am p_vaI[fn][gn] [hn] ); > f p r i n t f ( f p , " \ n " ) ; > > > else i f (dtype == 2) / * AVERAGED DATA * / fo r (hn = 1; hn <= waves; hn++) f o r ( fn = 0; fn < f i l e s ; fn++) f p r i n t f ( f p , "%f " , ave_amp[fn] [h n ]); > f p r i n t f ( f p , " \ n " ) ; > > else i f (dtype == 3) / * SELECTED DATA * / fo r (gn = 0; gn < g ra in s ; gn++) fo r (hn = 1; hn <= 24; hn++) C i f (s e lc [h n ] == 1) C f o r ( fn = 0; fn < f i l e s ; fn++) f p r i n t f ( f p , "%f ", amp_val[fn] [gn] [h n ]); > f p r i n t f ( f p , " \ n " ) ; > > > > else / * DTYPE IS 4 AND IS AVERAGED SELECTED DATA * / C f o r (hn = 1; hn <= 24; hn++) i f (s e lc lh n ] == 1) fo r ( f n = 0; fn < f i l e s ; fn++) f p r i n t f ( f p , "%f ", ave _am p[fn][hn]); > f p r i n t f ( f p , " \ n " ) ; > > > > else / * ORIENT IS 2 * / C i f (dtype == 2) / * AVERAGED DATA FOR FACTOR ANALYSIS * / fo r ( fn = 0; fn < f i l e s ; fn++) fo r (hn = 1; hn <= waves; hn++) f p r i n t f ( f p , "%f ", ave _am p[fn][hn]); > f p r i n t f ( f p , " \ n " ) ; > 131 > else / * DTYPE IS 4 AND IS AVERAGED SELECTED DATA FOR FACTOR * / L fo r ( fn = 0 ; fn < f i l e s ; fn++) L fo r (hn = 1; hn <= 24; hn++) L i f (s e lc th n ] == 1) f p r i n t f ( f p , "%f ", ave_amp [fn ] [hn] ); > f p r i n t f ( f p , " \ n " ) ; } > > > else i f (form == 2) / * ANOVA - T-TEST FORMAT * / L i f (dtype == 1) / * RAW DATA FOR ANOVA O R T-TEST * / L fo r ( fn = 0 ; fn < f i l e s ; fn++) L fo r (gn = 0; gn < g ra in s ; gn++) L fo r (hn = 1; hn <= waves ; hn ++) L f p r i n t f ( f p , "%3d %3d %2d % f\n", (fn + 1 ), (gn+1), hn, am p_val[fn][gn] [h n ]); > > > > else i f (dtype == 2) / * AVERAGE - ANOVA O R T-TEST * / L fo r ( fn = 0; fn < f i l e s ; fn++) L fo r (hn = 1; hn <= waves ; hn ++) L f p r i n t f ( f p , "%3d %2d % f\n", (fn + 1 ), hn, ave_am p[fn][hn]); > > > else / * DTYPE IS 5 AND IS FOR SINGLE W AVE OUPUT * / L f o r ( f n = 0; fn < f i l e s ; fn++) L fo r (gn = 0; gn < g ra in s ; gn++) L f p r i n t f ( f p , "%3d %3d % f\n", (fn + 1 ), (gn+1), am p_val[fn][gn] [waves] ); > > > > else / * DISCRIMINANT FUNCTION FORMAT * / L i f (dtype == 6) / * RAW DATA * / L fo r ( fn = 0; fn < f i l e s ; fn++) L fo r (gn = 0; gn < g ra in s ; gn++) L f p r i n t f ( f p , "%3d ", (fn + 1 )); fo r (hn = 1; hn <= waves; hn++) 132 { f p r i n t f ( f p , "%f ", a m p _ v a l[fn ][g n ][h n ]); > f p r i n t f ( f p , "X n "); > > > else / * DTYPE IS 3 AND IS FOR SELECTED DATA * / ( fo r ( fn = 0; fn < f i l e s ; fn++) f fo r (gn = 0; gn < g ra in s ; gn++) f f p r i n t f ( f p , "%3d ", (fn + 1 )); f o r (hn = 1; hn <= 24; hn++) f i f (s e lc th n ] == 1) f p r i n t f ( f p , "%f ", amp_val [fn ] [gn] [h n ]); > f p r i n t f ( f p , "X n "); > > > > fc lo s e ( fp ) ; / * CLOSE OUPUT FILE * / } I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *************************I calc_average() in t z, fn , gn, hn; p rin tf("X n A v e ra g in g Harmonic Data.Xn1 1 ); / * ZERO AVERAGE AMP VARIABLES * / fo r ( fn = 0; fn < f i l e s ; fn++) ( fo r(h n = 0; hn < waves; hn++) < _ ave_amp[fn] [hn] = 0.0; > > / * CALCULATING SUMS * / fo r ( f n = 0; fn < f i l e s ; fn++) fo r (hn = 0; hn <= waves; hn++) f o r (gn = 0; gn < g ra in s ; gn++) t ave_amp[fn] [hn] = ave_amp[fn] [hn] + amp_val [fn ] [gn] [hn] ; > > > / * DIVIDES FOR AMP AVERAGES * / fo r ( fn = 0; fn < f i l e s ; fn++) { fo r (hn = 0; hn <= waves; hn++) f ave_am p[fn][hn] = ave_amp[fn][hn] / g ra in s ; > > 133 I *********************************************************I / * END SOURCE CODE - CONVRTIT * / j ■ k -k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k 'k y 134 APPENDIX D Listing of the V A X /V M S C computer program STATSIT written by Rory Robinson. 135 / * STATSIT * / / * VAX-VMS V ersion: 6-1-92 * / / * W ritte n By Rory Robinson. * / / * For USC SED PET LAB - Dr Osborne * / / * C a lculates s t a t is t ic s from .AMP f i l e . * / / * L ib ra rie s * / #in clu d e s td io # include s t d lib # include ctype # inclu de math / * Global V ariab les * / in t g ra in s ; const in t waves = 63; const in t max_grains = 400; f lo a t amp_val[4 0 0 ][6 4 ]; f Io a t ave_amp[6 4 ]; f lo a t vrc_am p[64]; f lo a t stndev_amp[64]; char o ld _ f i le [ 8 0 ] , n e w _ file [80 ]; char *o ld _ p , *new_p; char prompt; / * Main Program * / m ain() { s t a r t : p r in t _ in f o ( ) ; workarea: f i nd_goaI( ) ; read_bi n_amp(); c a lc _ s ta ts ( ); wri te_asci i_answ er(); done: p rin tfC " \ n S t a t is t ic a l C a lc u la tio n s C om ple te.\n "); ex i t ( 0 ) ; > / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * / / * Subroutines * / / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * / help_opt i o n () L p r i n t f ("\nSTATSIT HELP\n\n"); p r in tfC " This program w i l l c a lc u la te the average, the variance and th e \n ") p rin tfC 's ta n d a rd d e v ia tio n o f each harmonic from a s in g le B inary .AMP f i l e \ n " ) p r in tfC " w ith up to 400 g ra in s . The re s u lts are then w r itte n in to an A S C II\n") p r in tfC 'o u tp u t f i l e . \ n \ n “ ); pauseC); ex i t C 0 ); 136 pause(> / * Pauses Screen T i l l Key Pressed * / C fflushC s td in ); p rin tfC "<Press RETURN to c o n tin u e s " ); prompt = g e tc h a rO ; p rin tfC "\n \n " > ; ^*********************************************************j p rin t_ in fo C > C p r in t fC 1 1 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * \ n 1 1 ) ■ p r in t fC " * STATSIT - W ritte n by Rory Robinson * \ n " ) ; p r in tfC " * Cc) Copyright 1992 * \ n " ) ; p r in t fC " * VAX-VMS V ersion: 6-1-92 * \ n " ) ; p r in t fC " * C alculates s t a t is t ic s from the * \ n " ) ; p r in tfC " * B inary .AMP f i l e s . * \ n " ); p r in t fC 1 1 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * \ n 1 1 )■ p r in tfC " * Type: 'HELP1 at the F ile Name * \ n " ) ; p r in tfC " * prompt to see program d ire c tio n s . * \ n " ) ; pr i ntfC )■ > j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j f ind_goaIC) t in t z, match; / * ENTER IN NUMBER OF FILES - HELP OPTION ALSO * / f ind_a: fflu s h C s td in ); p rin tfC "\n E n te r F ile Name fo r B inary Input F ile \n ? " ) ; old_p = getsC & o ld _ file [0 ] ); p r i n t f ( " \ n " ) ; fo r Cz=0; z<strlenC old_p>; z++) { old_ p[z] = tou pperC old_p[z]); > i f C! strcm pC & o ld _file [0], "HELP")) help_optionC); / * ENTER IN OUTPUT FILE NAME * / f i nd_b: fflu s h C s td in ); p rin tfC "\n E n te r ASCII Output F ile Name\n? " ) ; new_p = getsC & new _file[0] ); p r in t f C " \ n " ) ; fo r Cz=0; z<strlenCnew_p); z++) { _ new_p[z] = toupperCnew_p[z] ) ; > match = 0; i f C! strcm pC & new _file[0], & o ld _ f ile [0 ]) ) match = 1; else match = 0; i f Cmatch == 1) t p r in tfC " * ERROR - Output F ile Name Must Be D iffe r e n t From Input F ile Names!\n"); goto fin d _ b ; > 137 / * ENTER IN NUMBER OF G R A IN S * / f i nd_c: f flu s h ( s td in ); p r i n t f ( M \nE nter Number o f Grains in Input F ile \n ? " ) ; z = scanf("% d"f &grains ); p r i n t f C \ n " ) ; i f (z == 0) C p r i n t f C * ERROR - Bad Format - Enter an in te g e r nu m b e r!\n "); goto fin d _ c ; > i f (g ra in s < 2) p r i n t f C * ERROR - Minimum Number of Grains = 2 \n M); goto fin d _ c ; > else i f (g ra in s > max_grains) t p r i n t f C * ERROR - Maximum Number of Grains = %3d\n", max g ra in s ); goto fin d _ c ; } > read_bi n_amp() C FILE * fp ; in t z, gn, hn, e r r ; p rin tfC \n R e a d in g From B inary Input F ile . \ n " ) ; / * LOOP TO READ BINARY DATA FROM FILES * / i f ( ( f p = fopen ( & o ld _ f ile [0 ], "rb " )) == ' \ 0 ' ) t p r i n t f C * ERROR - Cannot Open Input F i le! \nOperat i on Aborted! \n " ); ex i t ( 0 ); } / * LOOP TO READ NUMBER OF GRAINS * / fo r(g n = 0; gn < g ra in s ; gn++ ) t / * LOOP TO READ NUMBER OF HARMONICS * / fo r(h n = 0; hn <= 63; hn++ ) £ e rr = fread ( &_val[gn] [h n ], s iz e o f(a m p _ v a l[g n ][h n ]), 1, fp ); i f ( ( e r r == 0) && ( f e o f ( f p ) != 0 )) £ p r i n t f C * ERROR - F ile Read E rro r! \nO peration Aborted! \ n " ) ; ex i t ( 0 ); > > > fc lo s e ( fp ); / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * I wri te_asci i_answ er() C 138 FILE * fp ; in t z, gn, hn, e r r ; p r i n t f ("\nS aving Output F ile . \ n " ) ; / * OPENING OUTPUT FILE * / i f ( ( f p = fopen (&new_fiLe[0 ], "w" )) == 1\ 0 '> p r i n t f ( " * ERROR - Cannot Create Output F ile ! \nO peration A b o rte d !\n "); exitC 0 ); > fo r (hn = 0; hn <= 63; hn ++) f p r i n t f ( f p , "%2d %f %f % f\n", hn, ave_amp[hn], vrc_amp[hn], stndev_amp[hn]); > fcloseC fp ); / * CLOSE OUPUT FILE * / > I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j c a lc _ s ta ts ( ) T in t z, gn, hn; p r i n t f ("\n P ro c e s s ing Harmonic D a ta .\n "); j - k * * * * * * * * * * * * * * * * * j / * ZERO VARIABLES * / yaHriHriHriHriHr * * * * * * * * * * * * * j fo r(h n = 0; hn <= 63; hn++) i ave_amp[hn] = 0 .0 ; vrc_amp[hn] = 0 .0 ; stndev_amp[hn] = 0.0 ; > y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * y / * CALCULATING SUMS FOR AVERAGE * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j fo r (hn = 0; hn <= 63; hn++) i fo r (gn = 0; gn < g ra in s ; gn++) C ave_amp[hn] = ave_amp[hn] + amp_val [gn] [hn] ; > > / * DIVIDES FOR AMP AVERAGES * / fo r (hn = 0; hn <= 63; hn++) C ave_amp[hn] = ave_amp[hn] / g ra in s ; > y * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * CALCULATING SUMS FOR VARIANCE * / j * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j fo r (hn = 0; hn <= 63; hn++) 139 c fo r (gn = 0; gn < g ra in s ; gn++) t vrc_amp[hn] = vrc_amp[hn] + pow((amp_val[gn][hn] - ave am p[hn]), 2); > > / * DIVIDES FOR VARIANCE * / fo r (hn = 0; hn <= 63; hn++) vrc_amp[hn] = vrc_amp[hn] / g ra in s ; > / * SQUARE ROOTS FOR STANDARD DEVIATION * / fo r (hn = 0; hn <= 63; hn++) C stndev_amp[hn] = s q rt( vrc_amp[hn] ); > /* END SOURCE CODE - STATSIT * / I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j 140 APPENDIX E Listing of the VAX/VMS C computer program RECASTIT written by Rory Robinson. 141 / * RECASTIT * / / * VAX/VMS V ersion: 1-12-92 * / / * W ritte n By Rory Robinson * / / * For USC SED PET LAB - Dr Osborne * / / * Recasting Program fo r Beach Data. * / / * L ib ra rie s * / #include s td io # include s t d li b # include ctype / * Global V ariab les * / in t beaches, sources; f lo a t old_beaches[1 0 0 1 ][1 1 ]; f lo a t new_beaches[1 0 0 1 ][1 1 ]; f lo a t source_num[11 ][1 1 ]; char beach_names[1 0 0 1 ][8 0 ]; char header [11] [8 0 ]; char te x t [80], o ld _ f ile [80], n e w _ file [8 0 ]; char *beach_p, *header_p, *old_p, *new_p; char prompt; / * Main Program * / mai n ( ) C s t a r t : p r in t_ in fo ( ); workarea: read_data( ); c a lc _ d a ta (); wri te _ d a ta (); done: p r i n t f ( "\nR ecasting C a lc u la tio n s C om plete.\n\nM); ex i t ( 0 ); > ! •k 'k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k -k 'k 'k 'k -k -k 'k 'k -k -k -k -k -k -k 'k -k -k 'k -k 'k -k J / * Subroutines * / h e lp _ o p tio n () C p ri nt p r in t p ri nt p ri nt p r in t p r in t p ri nt p ri nt p r in t pause p r in t p ri nt p r in t p r in t p ri nt p r i n t f ( 1 1 j M \n\n\nRECASTIT HELP\n\n"); 1 1 This program w i ll RECAST beach percentage data using th e \n "); "source percentage m a trix . A ll percentages should be from th e \n ") ; "d is c rim in a n t fu n c tio n a n a ly s is . This program can handle up to 1 0 \n "); " d if fe r e n t sources and recast up to 1000 beaches at a t im e .\n \n " ) ; " The program reads in data from a standard ASCII te x t f i l e , \ n " ) ; "co nve rts the data and then outputs to another ASCII te x t f i l e . IfN n"), "you do not know how to crea te an input or source f i l e , the program \n"), " w i l l help you to make o n e .\n \n "); " Input or source f i l e s should have the fo llo w in g fo rm a t: \ n \ n " ); sourcel source2 source3 ■ |\n " ) sourcel source2 source3 ! \ n " ) 75.0 20.0 5.0 | \n " ) 10.0 80.0 10.0 | \ n " ) 15.0 20.0 65.0 ! \ n " ) 142 p ri n p ri n p rin p r i n p ri n p ri n p ri n p ri n p rin p rin p r i n pause( p r i n p ri n p ri n p r i n p ri n p rin p ri n p ri n pause( ex i t ( j beach 1 | e t c . . . 55.5 24.3 20.2 | \ n " ) ; \n » ); \ n " ) ; | \ n \ n " ); "The f i r s t lin e must o n ly contain source names. The source m a tr ix \n " ) ; "must be o rie n te d as shown (That is the same from le f t to r ig h t a s \n "); " i t is top to bottom ). The beach numbers must correspond to th e \n " ) ; "source headings at the top of the columns. Beach and source names\n"); "must have NO spaces in them and can be no longer than 10 c h a ra c te rs . \ n " ); " A ll h o riz o n ta l rows should add up to 100 percent; however, the program \n"), "does not check t h is . \ n \ n " ) ; " The reca sting equation used in th is ro u tin e is as fo llo w s : \ n \ n " ) ; 'NEW_B(1 ,A) = 0LD_B(1 ,A) x S0URCE(1,A) "NEW_B(1 ,B) TOTAL OF SOURCE COLUMN A 0LD_B(1 ,A) X S0URCE(2,A) TOTAL OF SOURCE COLUMN A ■A' IS CHANGED TO 'B '\ n " ) ; AND SUM CONTINUES UNTIL\n"), ALL COLUMNS ARE USED.\n\n"); ■A1 IS CHANGED TO 1B' \ n " ); AND SUM CONTINUES UNTIL\n"), ALL COLUMNS ARE USED.\n\n"), "Formula continues u n t il a l l of NEW_B(1,X) has been c a Ic u la te d .\n \n " ) ; ); pause() / * Pauses Screen T i l l Key Pressed * / ■ C fflu s h ( s td in ); p r i n t f ( "<Press RETURN to c o n tin u e .>" ); prompt = g e tc h a rO ; p r i n t f ( " \ n \ n " ) ; > p ri n t_ i n f o ( ) ■ C p r i n t f ( M \ ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * \ nii j - p r i n t f C * RECASTIT - W ritte n by Rory Robinson * \ n " ) ; p r i n t f C * (c) Copyright 1992 * \ n " ) ; p r i n t f C * VAX-VMS Version: 1-12-92 * \ n M); p r i n t f C * Recasts Beach % % by using Source %% * \ n " ) ; p r i ntf( ) ■ p r i n t f C * Type: 'HELP' at the Source F ile Name * \ n " ) ; p r i n t f C * prompt to see program d ir e c tio n s . * \ n " ) ; p ri n tfC 1 1 * * * * * * * * * * * * * * * * * * * * * * * '* * * * * * * * * * * * ■ * * * * * \ n \n " ) * read_data() t FILE * fp ; in t z, row, c o l; / * Clear the input b u ffe r and get w ith gets. * / f flu s h ( s td in ); p r in t f( " E n t e r Source F ile Name (type 'none' to en ter data from keyboard)\n? " ) ; old_p = gets( o ld _ f ile ); fo r (z=0; z < s trle n (o ld _ p ); z++) ( old_ p[z] = to u p p e r(o ld _ p [z ]); > 143 i f (■strcm p(old_p, "NONE")) keyin _ d a ta (); i f < !strcm p(old_p, "HELP")) h e lp _ o p tio n (); Ioop_a: f flu s h ( s td in ); p r i n t f ("\n E n te r Output F ile Name\n? " ) ; new_p = gets< n e w _ file ); fo r (z=0; z<strlen(new _p); z++) { new_p[z] = toupper(new _p[z]); } i f ( ! strcmp(new_p, o ld_ p)) f p r i n t f ( " \ n * ERROR - F ile Names Must Be D if f e r e n t ! " ) ; goto toop_a; > p r in t f < " \ n " ) ; Ioop_b: p r in t f( " E n t e r Number of Sources (type 0 to e x it) \n ? " ) ; z = scanf("% d", &sources ); i f (z == 0) t fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ) ; goto loop_b; > i f (sources == 0) e x it( 0 ); else i f (sources < 2) C p r i n t f C * ERROR - Minimum Number of Sources = 2 \n "); goto loop_b; > else i f (sources > 10) p r i n t f C * ERROR - Maximum Number of Sources = 1 0 \n "); goto loop_b; > Ioop_c: p r in t f( " E n t e r Number of Beach Samples to Recast (type 0 to e x it) \n ? z = scanf("% d"f Sbeaches ); i f (z == 0) t fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter an in te g e r number! \ n " ) ; goto loop_c; > i f (beaches == 0) e x it( 0 ); else i f (beaches < 0) C p r i n t f C * ERROR - Minimum Number of Beaches = 1 \n "); goto loop_c; > else i f (beaches > 1000) t p r i n t f C * ERROR - Maximum Number of Beaches = 1000\n"); goto loop_c; > / * Read Data From F ile * / p rin tf("\n R e a d in g From Source F ile . \ n " ) ; i f ( ( f p = fopen (old_p, " r " )) == 1\ 0 1) c p r i n t f ( " * ERROR - Cannot Open Source F i l e ! \nO peration A b o rte d !\n N); ex i t ( 0 ); > / * READS SOURCE HEADER * / fo r (co l= 1 ; col<=sources; col++) C fs c a n f( fp , "%s", & header[col] ); > / * READS SOURCE MATRIX - DISCARDS SOURCE NAMES * / fo r (row=1; row<=sources; row++) C fscanfC fp , "Xs", Stext ); fo r (col=1 ; col<=sources; col++) C fscanfC fp , "%f1 1 , &source_numlrow] [c o l] ); > > / * READS BEACH NAMES AND MATRIX * / fo r (row=1; row<=beaches; row++) C fs c a n f( fp , "Xs", &beach_names[row] ); fo r (col=1 ; col<=sources; col++) fscanfC fp , "%f", & old_beaches[row ][col] ); > > fc lo s e ( fp ); J*********************************************************j keyi n_ da ta () FILE * fp ; in t z, row, c o l; / * ENTER IN FILE NAME * / k e y _ s ta rt: p r i n t f ( " \n \n * * Keyboard E ntry * * \ n \ n M); fflushC s td in ); p r i n t f ( "E nter Source F ile Name For The New Data\n? " ) ; old_p = getsC o ld _ f ile ); fo r (z=0; z < s trle n (o ld _ p ); z++) C old_p[z] = to u p p e r(o ld _ p [z ]); > p r i n t f ( " \ n " ) ; / * ENTER IN NUMBER OF SOURCES * / keyin_a: p r in tfC 'E n te r Number of Sources (type 0 to e x it) \n ? " ) ; z = scanf("JCd", &sources ); i f (z == 0) C fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter an in te ge r number! \ n “ ); goto keyin_a; > i f (sources == 0) e x it( 0 ); else i f (sources < 2) 145 c p r i n t f C * ERROR - Minimum Number of Sources = 2 \n "); goto keyin_a; } else i f (sources > 10) f p r i n t f C * ERROR - Maximum Number of Sources = 1 0 \n "); goto keyin_a; > / * LOOP TO ENTER IN SOURCE DATA * / col = 1; row = 1; keyi n_b: fflu s h ( s td in ); p r in t fC \n \ n E n t e r Name For Source Number %2d\n? ", row ); header_p = gets( header[row] ); fo r (z=0; z<strlen(header_p); z++) { header_p[z] = tou ppe r(he ade r_p [z]); > p r i n t f C \ n " ) ; keyi n_c: p r in t fC E n t e r Percent of Source %2d in Source %2d\n? ", c o l, row ); z = sc a n fC % f", &source_num[row] [c o l] ); i f (z == 0) C fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter a num ber!\n"); goto keyin_c; } else i f (source_num [row ][col] < 0.0) f p r i n t f C * ERROR - P o s itiv e Numbers O n ly !\n ") ; goto keyin_c; > else i f (source_num [row ][col] > 100.0) f p r i n t f C * ERROR - Maximum P ossible Percent = 100\nM); goto keyin_c; i f (co l < sources) f col = col + 1; goto keyin_c; > else f i f (row < sources) f row = row + 1 ; col = 1; goto keyin_b; > > p r i n t f ( " \ n \ n " ) ; / * ENTER IN BEACH NAMES AND DATA * / beach_a: p r in t fC E n t e r Number of Beach Samples (type 0 to e x it) \n ? " ) ; z = scanf("% d", Sbeaches ); 146 i f (z == 0) C fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter an in te ge r num ber!\n“ ); goto beach_a; > i f (beaches == 0) e x it( 0 ); else i f (beaches < 0) C p r i n t f C * ERROR - Minimum Number of Beaches = 1 \n "); goto beach_a; > else i f (beaches > 1000) { p r i n t f C * ERROR - Maximum Number of Beaches = 1000\n"); goto beach_a; > / * LOOP TO ENTER IN BEACH DATA * / col = 1; row = 1; beach_b: f flu s h ( s td in ); p r i n t f ("\n \n E n te r Name For Beach Number %2d\n? ", row ); beach_p = gets( beach_names[row] ); fo r (z=0; z<strlen(beach_p); z++) { beach_p[z] = toupper(beach_p[z] ); > p r i n t f ( " \ n " ) ; beach_c: p r in t fC E n t e r Percent of Source %2d in Beach %2d\n? ", c o l, row ); z = scan f("% f", & old_beaches[row ][col] ); i f (z == 0) C fflu s h ( s td in ); p r i n t f C * ERROR - Bad Format - Enter a number! \ n " ) ; goto beach_c; > else i f (old_ bea che s[ro w ][co l] < 0.0) { p r i n t f C * ERROR - P o s itiv e Numbers O n ly !\n M); goto beach_c; > else i f (old_ bea che s[ro w ][co l] > 100.0) C p r i n t f C * ERROR - Maximum P ossible Percent = 100\nM); goto beach_c; f (col < sources) col = col + 1; goto beach_c; Ise i f (row < beaches) C row = row + 1; col = 1; goto beach_b; 147 > } o p ti on_loop: f flu s h ( s td in ); p r i n t f ("\n\nChoose an O ption: <S>ave data, <R>e-enter data, <E>xit : " ) ; prompt = g e tc h a rO ; i f ((prompt == 'R') jj (prompt == 'r') ) goto ke y_ sta rt; else i f ((prom pt == 'E') Jj (prompt == 'e')) e x it( 0 ); else i f ((prompt != 'S') && (prompt != 's')) goto option_loop; / * SAVE NEW DATA TO A SOURCE FILE FORMAT * / p r i n t f ("\n \n S a vin g New Source F ile .\n " > ; i f ((fp = fopen (old_p, "w" )) == 1\ 0 1 > L p r i n t f C * ERROR - Cannot Create Source Fi le! \nO peration Aborted! \ n " ) ; ex i t ( 0 ) ; / * SAVES SOURCE HEADER * / i f (sources <= 6 ) L f p r i n t f ( f p , " " ) ; / * 15 SPACES TO PAD HEADER * / fo r (col=1; col<=sources; col++) C f p r i n t f ( fp , "%-10s", &header[col] ); > f p r i n t f ( f p , " \ n " ) ; > else C f p r i n t f ( f p , " " ) ; / * 10 SPACES TO PAD HEADER * / fo r (col=1 ; col<=sources; col++) C f p r i n t f ( fp , "%-7s", S headerlcol] ); > f p r i n t f ( f p , M \n n); / * SAVES SOURCE MATRIX - WRITES SOURCE NAMES * / i f (sources <= 6 ) C fo r (row=1; row<=sources; row++) C f p r i n t f ( fp , "%-15s ", &header[row] ); fo r (col=1; col<=sources; col++) C f p r i n t f ( fp , "% -6.2f ", source_num [row][col] >; > f p r i n t f ( f p , " \ n " ) ; > > else { fo r (row=1; row<=sources; row++) C f p r i n t f ( fp , "X -IO s", &header[row] ); fo r (col=1; col<=sources; col++) C f p r i n t f ( fp, "% -6.2f ", source_num [row][col] ); > f p r i n t f ( f p , " \ n " ) ; 148 > > f p r i n t f ( f p , M \n M); / * SAVES BEACH NAMES AND MATRIX * / i f (sources <= 6 ) C fo r (row=1; row<=beaches; row++) t f p r i n t f ( fp , "%-15s " , &beach_names[row] ); fo r (col=1 ; col<=sources; col++) C f p r i n t f ( fp , H%-6.2f " , old_beaches[row ][col] ); > f p r i n t f ( f p , "X n "); > > el se C fo r (row=1; row<=beaches; row++) f p r i n t f ( fp , "/i-IO s ", &beach_names[row] ); fo r (col=1; col<=sources; col++) t f p r i n t f ( fp , "% -6.2f " , old_beaches[row ][col] ); > f p r i n t f ( f p , "X n "); > > fc lo s e ( fp ); / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * / wri te _ d a ta () C FILE * fp ; in t z, row, c o l; p rin tf("X n S a v in g Output F ile .X n "); i f ((fp = fopen (new_p, "w" )) == 1\ 0 1) C p r i n t f C * ERROR - Cannot Create Output Fi le! \nO peration Aborted! \ n " ) ; e x it( 0 ); > / * SAVES SOURCE HEADER * / i f (sources <= 6 ) ( f p r i n t f ( f p , » " ) ; / * 15 SPACES TO PAD HEADER * / fo r (col=1; col<=sources; col++) C f p r i n t f ( fp , n%-10sM, &header[col] ); > f p r i n t f ( f p , " \ n " ) ; > else C f p r i n t f ( fp , " " ) ; / * 10 SPACES TO PAD HEADER * / fo r (col=1 ; col<=sources; col++) C f p r i n t f ( fp , "%-7s", &header[col] ); 149 > f p r i n t f ( f p , " \ n " ) ; > / * SAVES BEACH NAMES AND MATRIX * / i f (sources <= 6 ) ( fo r (row=1; row<=beaches; row++) C f p r i n t f ( fp , "%-15s ", &beach_names[row] ); fo r (col=1 ; col<=sources; col++) { f p r i n t f ( fp , "% -6.2f " , new_beaches[row][col] ) > f p r i n t f ( f p , " \ n " ) ; > > el se < fo r (row=1; row<=beaches; row++) < f p r i n t f ( fp , "%-10s", &beach_names[row] ); fo r (col=1 ; col<=sources; col++) C f p r i n t f ( fp , "% -6.2f ", new_beaches[row][col] ); > f p r i n t f ( f p , " \ n " ) ; > > fc lo s e ( fp ); > ^ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * J c a lc _ d a ta () t in t z, y, row, c o l, s_col; flo a t s _ to ta l[1 1 ]; p rin tf("X n R e c a s tin g Beach D ata.Xn"); / * ZERO TOTALS * / fo r (z=0; z<11; z++) { s _ to ta l[z ] = 0.0; > / * ZERO NEW_BEACHES VARIABLES * / fo r (z=0; z<101; z++) C fo r(y = 0 ; y< 11; y++) < ! new_beaches [z] [y] = 0.0; > > / * GETS SOURCE COL TOTALS * / fo r (col= 1 ; col<=sources; col++) C fo r (row=1; row<=sources; row++) C s _ to ta l[c o l] = s _ to ta l[c o l] + source_num [row ][col]; > } / * RECASTS BEACH DATA * / / * format fo r a rra y v a ria b le s : source_num [row ][col]; old _ b e a ch e s[ro w ][co l]; new _beaches[row ][col]; * / fo r (row=1; row<=beaches; row++) C fo r {co l=1 ; col<=sources; col++) C fo r (s_col=1; s_col<=sources; s_col++) C new_beaches[row][col] = new_beaches[row][col] + (( old_beaches[row ][s_col] * source_num [col][s_co / s _ to ta l[s _ c o l] ); > > > > / * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j / * END SOURCE CODE - RECASTIT * / I * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * j
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Grain-shape sorting of medium-sized quartz sand by eolian action
PDF
Fourier grain-shape analysis of quartz sand from the Santa Monica Bay Littoral Cell, Southern California
PDF
The relationships among river, beach and submarine canyon sands in the Southern Santa Barbara littoral cell, Ventura County, California, Fourier grain shape analysis.
PDF
Quartz grain-shape modification in high-gradient fluvial environments: Fourier grain-shape analysis
PDF
Quartz grain-surface microtextures from sediments of the Santa Clara River and Santa Barbara littoral cell, California
PDF
Depositional systems of the mid-Tertiary Gene Canyon and Copper Basin Formations, eastern Whipple Mountains, California
PDF
Depositional environments of the Neogene Hungry Valley Formation: Sedimentary response to the initiation of the San Andreas Fault, Ridge Basin, Southern California
PDF
Mineralogic profiles for sand transport within southern California littoral cells: Cliffs, rivers and beaches
PDF
Sedimentary structures in vibra-cores from the Oxnard Shelf, California
PDF
Facies and depositional environments of Guadalupian strata (Brushy Canyon and Cherry Canyon Formations), northwestern Delaware Basin, west Texas
PDF
Carbonate microfacies of the Upper Monte Cristo Limestone and the Lower Bird Spring Group at Mountain Springs, Clark County, Nevada
PDF
The effects of biological activity on transport of dissolved species across the sediment-water interface of San Francisco Bay
PDF
A scanning electron microscopy study of Pleistocene and Holocene sand samples from Santa Monica Bay: Southern California.
PDF
Petrology and economic value of beach sand from southern Monterey Bay, California
PDF
Engineering geology of Santa Ynez Canyon, Santa Monica Mountains, Los Angeles, California
PDF
Engineering properties of Southern California borderland sediments
PDF
Quartz Grain-Shape Variation Within An Individual Pluton: Granite Mountain, San Diego County, California
PDF
Holocene sedimentation in Santa Monica Basin, California
PDF
Distribution of dinoflagellate cysts in sediments of Santa Monica Basin, California
PDF
Geology and engineering properties of the Portugese tuff, Palos Verdes Hills, California
Asset Metadata
Creator
Robinson, Rory Anthony
(author)
Core Title
Fourier grain-shape analysis of quartz sand from the eastern and central Santa Barbara littoral cell, Southern California
Degree
Master of Science
Degree Program
Geological Sciences
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Sedimentary Geology
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c30-130096
Unique identifier
UC11225698
Identifier
usctheses-c30-130096 (legacy record id)
Legacy Identifier
EP58831.pdf
Dmrecord
130096
Document Type
Thesis
Rights
Robinson, Rory Anthony
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
Tags
Sedimentary Geology