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
/
Distribution and impact of algal blooms leading to domoic acid events in southern California
(USC Thesis Other)
Distribution and impact of algal blooms leading to domoic acid events in southern California
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
DISTRIBUTION AND IMPACT OF ALGAL BLOOMS LEADING TO DOMOIC ACID EVENTS IN
SOUTHERN CALIFORNIA
by
Erica Lee Seubert
_______________________________________________
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOLOGICAL SCIENCES)
August 2013
Copyright 2013 Erica Lee Seubert
ii
Dedication
I dedicate this dissertation to my grandfather and childhood best friend, Arthur Malek,
and to my adulthood best friend, Alec Alders. Arthur Malek was my first ‘pupil’, dutifully
completeing all of his homework and acing all the exams given by Miss General Lee. Alec Alders
has tolerated being tasked as an additional lab assistant and manuscript proofreader, with a
brilliant smile and patience I had only previously seen in my first student.
iii
Acknowledgements
The journey towards a successful completion of a doctoral thesis is not one that is taken
alone, and it begins well before being accepted into a doctoral program. My success is a direct
reflection of the love and support I have received from my family and friends throughout my
life. During my time at USC, I have had the pleasure to call David Caron my advisor and several
post-docs, graduate students, lab technicians and undergraduate students as my lab mates.
The work present in this dissertation could not have been accomplished without their
assistance.
Chapter 1 (Algal toxins and reverse osmosis desalination operations: Laboratory bench
testing and field monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid) was
completed with assistance from my co-authors Shane Trussell, John Eagleton, Astrid Schnetzer,
Ivona Cetinić, Phil Lauri, Burton Jones and David Caron, and supported with funding from a
NOAA MERHAB grant (NA05NOS4781228), USC Sea Grant (NA07OAR41700008) and a
collaborative grant between the West Basin Municipal Water District, Department of Water
Resources and USC. Additional assistance with sample collection and processing was provided
by Sarah Barber, Lindsay Darjany, Mark Donovan and Elizabeth Fitzpatrick. The work was
published Water Research, volume 46 pages 6563-6573 in 2012.
Chapter 2 (Seasonal and annual dynamics of harmful algae and algal toxins revealed
through weekly monitoring at two coastal ocean sites off southern California, USA) was
completed with assistance from my co-authors Alyssa Gellene, Meredith Howard, Paige
Connell, Matthew Ragan, Burton Jones, Jennifer Runyan and David Caron, and supported with
funding from NOAA grants (NA05NOS4781228, NA11NOS4780052) and a subaward of
iv
NA08OAR4320894. Additional sample collection and processing assistance was provided by
Sarah Barber, Lindsay Darjany, Cody Nelson, Astrid Schnetzer, Christine Tung and Warren
Yamashita. The work was published in Environmental Science and Pollution Research in 2013,
DOI 10.1007/s11356-012-1420-0.
Chapter 3 (Development, comparison and validation using ELISAs for the analysis of
domoic acid in California sea lion body fluids) was completed with assistance from my co-
authors Meredith Howard, Rapheal Kudela, Thomas Stewart, Wayne Litaker, Richard Evans and
David Caron, and supported with funding from NOAA grants (NA05NOS4781228,
NA11NOS4780052) and a subaward of NA08OAR4320894, a USC Sea Grant award
(NA06OAR4170012), the California Ocean Protection Council (R/OPCCONT-12 A 10) and the
Central and Northen California Ocean Observing System. The work has been accepted for
publication in the Journal of the Association of Official Analytical Chemists.
Chapter 4 (Phytoplankton community response to the addition of treated sewage
effluent) was completed with assistance from my co-authors Alyssa Gellene, Victoria Campbell,
Jayme Smith, George Robertson and David Caron, with funding from a NOAA ECOHAB grant
(NA11NOS4780052), a USC Sea Grant award (NA06OAR4170012) and a collaborative grant
between the Orange County Sanitation District and USC. Additional sample collection and
assistance was provided by Alec Alders, Connor Fox, Troy Gunderson, Oliver Hayward and Alex
Yuen.
v
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables vii
List of Figures viii
Abstract 1
Chapter 1: Algal toxins and reverse osmosis desalination operations:
Laboratory bench testing and field monitoring of domoic acid, saxitoxin,
brevetoxin and okadaic acid
5
Abstract 5
Introduction 6
Materials and Methods 11
Results 18
Discussion 27
Conclusions 31
References 32
Chapter 2: Seasonal and annual dynamics of harmful algae and algal toxins
revealed through weekly monitoring at two coastal ocean sites off southern
California, USA
37
Abstract 37
Introduction 38
Materials and Methods 43
Results 48
Discussion 65
Conclusions 71
References 72
Chapter 3: Development, comparison and validation using ELISAs for the
analysis of domoic acid in California sea lion body fluids
81
Abstract 81
Introduction 81
Materials and Methods 86
Results and Discussion 93
Conclusions 110
References 113
vi
Chapter 4: Phytoplankton community response to the addition of treated
sewage effluent
117
Abstract 117
Introduction 118
Materials and Methods 121
Results 126
Discussion 142
Conclusions 146
References 148
Bibliography 153
Appendix 1: Chasing elusive harmful algal blooms
Article for the Catalina Marine Society Magazine, Ocean Bights
169
vii
List of Tables
Table 2-1. (A) Information on major and minor bloom events identified at Newport pier using
chlorophyll a concentrations to define blooms. (B) Information on major and minor blooms
events identified at Newport Pier defined by abundances of cells in the P. seriata size class from
Pseudo-nitzschia cell counts………………………………………………………..………………………………………….53
Table 2-2. (A) Information on major and minor bloom events identified at Redondo Beach pier
using chlorophyll a concentrations to define blooms. (B) Information on major and minor
bloom events identified at Redondo Beach pier based upon the abundance of P. seriata size
class cells from Pseudo-nitzschia cell counts……………………………………………………………………………56
Table 2-3. Results of multiple regression analysis of the Newport pier dataset. Prior to analysis,
samples were identified as blooms based upon the P. seriata size class abundance definition
and all non-bloom values were removed. Negative correlations are identified with a minus sign
(-)…………………………………………………………………………………………………………………………………………….63
Table 2-4. Results of multiple regression analysis of the Redondo Beach pier dataset. Prior to
analysis, samples were identified as blooms based upon the P. seriata size class abundance
definition and all non-bloom values were removed. Negative correlations are identified with a
minus sign (-). Asterisks denote failed regressions without statistical significance, low F values
(<2) and low power values………………………………………………………………………………………………………65
Table 3-1. Summary of DA concentrations and analysis methods reported in the literature for
several pinniped species………………………………………………………………………………………………………….85
Table 3-2. R
2
and slope values obtained from linear regression lines plotted for each method
platform and body fluid versus the known spike concentration of DA……………………………..……..95
Table 3-3. Measurements of DA concentrations naturally present in California sea lion body
fluids across the three platforms. The original DA concentration determined by MS ELISA with
the modified DA analysis protocol within one month of receipt of samples at USC. Values
reported for MS NonSPE, BS NonSPE and MS SPE analysis during the validation study in fall
2009 are averages of triplicate replicates………………………………………………………………….…………109
Table 4-1. Nutrient concentrations measured in the initial whole seawater sample and the
nutrient concentrations of the effluent sample used in the Pre-Diversion (9/6/2012) and Mid-
Diversion (9/20/2012) experiments. The N:P ratio is reported by atoms and samples below the
method detection limit are reported as “bd”……………………………………………………………………..126
viii
List of Figures
Figure 1-1. Diagram of the bench-scale RO experimental setup. The dashed line in the RO unit
depicts the RO membrane. Sampling locations for the reservoir, permeate and retentate are
noted by the large black arrows. …………………………………………………………………………………………….13
Figure 1-2. Location of the pilot desalination plant operated by the West Basin Municipal Water
District in El Segundo, CA, USA, co-located at the El Segundo Generating Station……………………15
Figure 1-3. (A) The change in concentration of extracellular DA over the duration of the bench-
scale RO experiment (96h), as measured in the reservoir (light gray triangles), permeate (open
triangles) and retentate (dark gray triangles) streams. The open triangles denote
concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO
membrane over the duration of the experiment, represented by changes in the specific (SPF)
and constant flux (GFD)…………………………………………………………………………………………………………..19
Figure 1-4. (A) The change in concentration of extracellular STX over the duration of the bench-
scale RO experiment (48h), as measured in the reservoir (light gray triangles), permeate (open
triangles) and retentate (dark gray triangles) streams. The open triangles denote
concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO
membrane over the duration of the experiment, represented by changes in the specific (SPF)
and constant flux (GFD)…………………………………………………………………………………………………………21
Figure 1-5. (A) The change in concentration of extracellular PbTx over the duration of the
bench-scale experiment (48h), as measured in the reservoir (light gray triangles), permeate
(open triangles) and retentate (dark gray triangles) streams. The open triangles denote
concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO
membrane over the duration of the experiment, represented by changes in the specific (SPF)
and constant flux (GFD)…………………………………………………………………………………………………………23
Figure 1-6. The average monthly intracellular (top panel) and extracellular (bottom panel) DA
concentration detected in the pre-microfiltrated intake water during the 2005-2009 sampling
period; bars are color coded based on sample year. The maximum DA concentration observed
for each month, regardless of year, is marked on the graph by a black circle. The open circle
denotes concentrations under the limit of detection of the ELISA………………………………………..25
Figure 1-7. The average monthly extracellular STX concentration detected in the pre-
microfiltrated intake water during the 2008-2009 sampling period; bars are color coded based
on sample year. The maximum extracellular STX concentration observed for each month,
regardless of year, is marked on the graph by a black circle. The open circle denotes
concentrations under the limit of detection of the ELISA. The star signifies the months samples
were not taken due to plant inoperation………………………………………………………………………………26
ix
Figure 2-1. Location of the SCCOOS weekly HAB monitoring piers in southern California,
maintained through a collaboration of five university laboratories, beginning in June 2008. USC
maintains the Newport pier sampling location in Orange County, CA. Redondo Beach pier
(noted on the map with a star), located in Los Angeles County, CA, was added to the USC HAB
monitoring effort in 2010……………………………………………………………………………………………………..44
Figure 2-2. Temperature and salinity measured at Newport pier with the SCCOOS automated
sensor is plotted from June 30, 2008, to December 19, 2011. Nitrate, phosphate and silicate
measured in the discrete samples from the same time period are also plotted……………………50
Figure 2-3. Temperature measured during the discrete sampling at Redondo Beach Pier from
January 26, 2010, to December 19, 2011, is plotted in conjunction with the salinity
measurements taken with the SCCOOS automated sensor at Santa Monica Pier during the same
period……………………………………………………………………………………………………………………………………52
Figure 2-4. (A) Chlorophyll a concentrations measured in the weekly samples collected at
Newport pier from June 30, 2008, to December 19, 2011. The major bloom threshold
chlorophyll concentration of 12.9 µg/L and minor bloom threshold of 9.74 µg/L are shown by
the dashed and dotted lines, respectively. The four major blooms are marked by the grey stars
and the twelve minor blooms are marked by the white stars. (B) All samples that were not
identified as major or minor blooms by the chlorophyll concentration definition were removed,
and only bloom sample chlorophyll concentrations are plotted with the black circles. The major
and minor bloom thresholds are shown by the dashed and dotted lines, respectively. pDA
concentrations measured in any of the chlorophyll defined major and minor are plotted with
the light grey triangles, pDA samples below the detection limit were not included……………..54
Figure 2-5. (A) Chlorophyll a concentrations measured in the weekly samples collected at
Redondo Beach pier from January 26, 2010, to December 19, 2011. The major bloom threshold
chlorophyll concentration of 9.13 µg/L and minor bloom threshold of 5.66 µg/L are shown by
the dashed and dotted lines, respectively. The four major blooms are marked by the grey stars
and the four minor blooms are marked by the white stars. (B) All samples that were not
identified as major or minor blooms by the chlorophyll concentrations definition were removed
and only bloom sample chlorophyll concentrations are plotted. The major and minor bloom
thresholds are shown by the dashed and dotted line, respectively. Domoic acid was not
measured in any of the chlorophyll bloom samples from Redondo Beach Pier……………………55
Figure 2-6. (A) The concentration of Pseudo-nitzschia seriata size class cells measured in the
Newport pier weekly samples from June 30, 2008 to December 19, 2011 are plotted in the top
panel figure. The major bloom threshold concentration of 88,000 cells/L and minor bloom
threshold of 40,000 cells/L are plotted as the dashed and dotted lines, respectively. The eight
major blooms are marked by the grey stars and the twelve minor blooms are marked by the
white stars. (B) Only the identified blooms by the P. seriata size class definition, major and
minor, are plotted in the bottom panel with black circles and the associated pDA concentrations
measured in the same sample are plotted as the grey triangles, pDA samples below the
x
detection limit were not included. The major and minor bloom thresholds are plotted as the
dashed and dotted lines, respectively………………………………………………………………………..…………..59
Figure 2-7. (A) The concentration of Pseudo-nitzschia seriata size class cells measured in the
Redondo Beach pier weekly samples from January 26, 1010, to December 19, 2011, are plotted.
The major bloom threshold concentration of 110,000 cells/L and minor bloom threshold of
56,000 cells/L are plotted as the dashed and dotted line, respectively. The four major blooms
are marked by the grey stars and the minor blooms are marked by the white stars. (B) Only the
identified blooms by the P. seriata size class definition, major and minor, are plotted as the
black circles with the associated domoic acid concentrations measured in the same sample
plotted as the grey triangles, pDA samples below the detection limit were not included. The
major and minor bloom thresholds are plotted as the dashed and dotted lines, respectively….60
Figure 3-1. Flow diagram outlining the validation design for California sea lion body fluid
samples……………………………………………………………………………………………………………………………………89
Figure 3-2. DA concentrations measured using different analytical platforms and protocols in
samples of AF spiked with DA standard. The dotted lines in all graphs show the expected DA
concentration based on the amount of DA added to each sample. Error bars represent
standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned
results represented by black circles and the solid black line, the MS SPE-cleaned results
represented by white circles and the long dashed and dotted line, the BS ELISA non-SPE-cleaned
results represented by gray circles and the gray line, and the LC-MS results represented by black
triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-
SPE-cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned);
(C) DA concentrations obtained using the MS ELISA (SPE-cleaned) plotted with the LC-MS
results; (D) DA concentration obtained using the MS ELISA (non-SPE-Cleaned) represented by
the black circles and black line and the DA concentration results obtained using the BS ELISA
(non-SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS
(SPE-cleaned) results. ……………………………………..……………………………………………………………………94
Figure 3-3. DA concentrations measured using different analytical platforms and protocols in
samples of CSF spiked with DA standard. The dotted lines in all graphs show the expected DA
concentration based on the amount of DA added to each sample. Error bars represent
standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned
results represented by black circles and the solid black line, the MS SPE-cleaned results
represented by white circles and the long dashed and dotted line, the BS ELISA non-SPE-cleaned
results represented by gray circles and the gray line, and the LC-MS results represented by black
triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-
SPE-cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned);
(C) DA concentrations obtained using the MS ELISA (SPE-cleaned) plotted with the LC-MS
results; (D) DA concentration obtained using the MS ELISA (non-SPE-Cleaned) represented by
the black circles and black line and the DA concentration results obtained using the BS ELISA
xi
(non-SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS
(SPE-cleaned) results………………………………………………………………………………………………………….98
Figure 3-4. DA concentrations measured using different analytical platforms and protocols in
samples of serum spiked with DA standard. The dotted lines in all graphs show the expected DA
concentration based on the amount of DA added to each sample. Error bars represent
standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned
results represented by black circles and the solid black line, the MS SPE-cleaned results
represented by white circles and the long dashed and dotted line, the BS ELISA non-SPE-cleaned
results represented by gray circles and the gray line, and the LC-MS results represented by black
triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-
SPE-cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned);
(C) DA concentrations obtained using the MS ELISA (SPE-cleaned) plotted with the LC-MS
results; (D) DA concentration obtained using the MS ELISA (non-SPE-Cleaned) represented by
the black circles and black line and the DA concentration results obtained using the BS ELISA
(non-SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS
(SPE-cleaned) results………………………………………………………………………………………………………….102
Figure 3-5. DA concentrations measured using different analytical platforms and protocols in
samples of urine spiked with DA standard. The dotted lines in all graphs show the expected DA
concentration based on the amount of DA added to each sample. Error bars represent
standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned
results represented by black circles and the solid black line, the MS SPE-cleaned results
represented by white circles and the long dashed and dotted line, the BS ELISA non-SPE-cleaned
results represented by gray circles and the gray line, and the LC-MS results represented by black
triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-
SPE-cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned);
(C) DA concentrations obtained using the MS ELISA (SPE-cleaned) plotted with the LC-MS
results; (D) DA concentration obtained using the MS ELISA (non-SPE-Cleaned) represented by
the black circles and black line and the DA concentration results obtained using the BS ELISA
(non-SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS
(SPE-cleaned) results…………………………………………………………………………………………………………….106
Figure 4-1. Location of the study area in Orange County, California, with the locations water
was collected for the Pre-Diversion and Mid-Diversion effluent incubation experiments are
marked in white…………………………………………………………………………………………………………………….122
Figure 4-2. Chlorophyll α concentrations (A) and the abundances of diatoms (B),
picoeukaryotes (C), Prochlorococcus spp. (D), Synechoccus spp. (E) and heterotrophic bacteria
(F) measured during the Pre-Diversion effluent addition experiment with surface water are
plotted separately for each treatment. Results from the true controls are plotted with open
circles and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent
additions with black circles and solid black lines, 1:100 effluent additions with dark grey circles
xii
and solid dark grey lines and 1:1000 effluent additions with light grey circles and solid light grey
lines……………………………………………………………………………………………………………………………………….129
Figure 4-3. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured
during the Pre-Diversion effluent addition experiment with surface water are plotted separately
for each treatment. Results from the true controls are plotted with open circles and dashed
lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with black
circles and solid black lines, 1:100 effluent additions with dark grey circles and solid dark grey
lines and 1:1000 effluent additions with light grey circles and solid light grey lines. The method
detection limit for each nutrient analyte is marked on the graphs with a dotted light grey
line……………………………………………………………………………………………………………………………………….131
Figure 4-4. Chlorophyll α concentrations (A) and the abundances of diatoms (B),
picoeukaryotes (C), Prochlorococcus spp. (D), Synechoccus spp. (E) and heterotrophic bacteria
(F) measured during the Pre-Diversion effluent addition experiment with water from the DCM
are plotted separately for each treatment. Results from the true controls are plotted with open
circles and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent
additions with black circles and solid black lines, 1:100 effluent additions with dark grey circles
and solid dark grey lines and 1:1000 effluent additions with light grey circles and solid light grey
lines……………………………………………………………………………………………………………………………………..133
Figure 4-5. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured
during the Pre-Diversion effluent addition experiment with water from the DCM are plotted
separately for each treatment. Results from the true controls are plotted with open circles and
dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with
black circles and solid black lines, 1:100 effluent additions with dark grey circles and solid dark
grey lines and 1:1000 effluent additions with light grey circles and solid light grey lines. The
method detection limit for each nutrient analyte is marked on the graphs with a dotted light
grey line…………………………………………………………………………………………………………………………….134
Figure 4-6. Chlorophyll α concentrations (A) and the abundances of diatoms (B),
picoeukaryotes (C), Prochlorococcus spp. (D), Synechoccus spp. (E) and heterotrophic bacteria
(F) measured during the Mid-Diversion effluent addition experiment are plotted separately for
each treatment. Results from the true controls are plotted with open circles and dashed lines,
Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with black circles
and solid black lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and
1:1000 effluent additions with light grey circles and solid light grey lines………………………..136
Figure 4-7. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured
during the Mid-Diversion effluent addition experiment are plotted separately for each
xiii
treatment. Results from the true controls are plotted with open circles and dashed lines, Milli-
Q® controls with open circles and dotted lines, 1:10 effluent additions with black circles and
solid black lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and
1:1000 effluent additions with light grey circles and solid light grey lines. The method detection
limit for each nutrient analyte is marked on the graphs with a dotted light grey line…………138
Figure 4-8. MDS plots using Bray-Curtis Similarities for the phytoplankton community
composition of each treatment throughout the Pre-Diversion experiments with surface water
(A) and DCM water (B), the Mid-Diversion experiment (C) and the first three days of both the
Pre-Diversion and Mid-Diversion experiments (D). Pre-Diversion surface samples are plotted
with upward pointing triangles, DCM samples with downward pointing triangles and the Mid-
Diversion samples are plotted with circles. All T0 samples are colored a light grey, true controls
are in black, Milli-Q® controls are black outlined symbols, 1:10 effluent treatments are in red,
1:100 effluent treatments are in yellow and the 1:1000 effluent treatments are in blue.
Samples with Bray-Curtis Similarities of 40, 60 and 80 are circled on each plot in green, dark
blue and light blue, respectively………………………………………………………………………………………….140
Supplementary Figure 4-1. Chlorophyll a concentrations (A) and the abundances of diatoms (B),
picoeukaryotes (C), Prochlorococcus spp. (D), Synechoccus spp. (E) and heterotrophic bacteria
(F) measured during the Mid-Diversion effluent addition experiment are plotted separately for
each treatment. Results from the vitamin and trace metal addition is plotted with black circles
and solid black lines, 1:10 effluent with vitamins and trace metals is plotted with dark grey
circles and solid dark grey lines and 1:10 effluent mimic with vitamins and trace metals is
plotted with light grey circles and solid light grey lines………………………………………………………..147
1
ABSTRACT
The term harmful algal bloom (HAB) is used to describe any bloom of microalgae that
has a detrimental impact to the local ecosystem and/or economy. The impacts of a HAB to an
ecosystem can include death or injury to local wildlife through the production of toxins,
decreased oxygen concentrations, physical damage, decreased light availability or food web
disturbance. The economic impacts can be reduction in tourism, human illness, reduced fishing
effort or interruption of desalination plant operations. The occurrence and intensity of HABs
have been increasing globally during the past few decades, whether this increase can be
attributed to enhanced awareness and monitoring, or to a dramatic upswing in the
development of HAB events remains unresolved.
A variety of HAB-forming species of microalgae occur in southern California, and several
of these species are known to produce potent neurotoxins. The impact of algal toxin presence
on both the intake and reverse osmosis (RO) desalination process and whether or not the
naturally occurring algal toxins can pass through the RO membrane and into the desalination
product was addressed through bench-scale RO experiments and monitoring for algal toxins at
a pilot RO desalination plant. Concentrations exceeding maximal values previously reported
during natural blooms were used in the laboratory experiments, with treatments comprised of
50 µg/L of domoic acid (DA), 2 µg/L of saxitoxin (STX) and 20 µg/L of brevetoxin (PbTx). None
of the algal toxins used in the bench-scale experiments were detectable in the desalinated
product water. Monitoring for intracellular and extracellular toxin concentrations of DA, STX,
PbTx and okadaic acid (OA) within the intake and desalinated water from a pilot RO
desalination plant in El Segundo, CA, was conducted from 2005 to 2009. During the five-year
2
monitoring period, DA and STX were detected sporadically in the intake waters but never in the
desalinated water. PbTx and OA were not detected in either the intake or desalinated water.
The results of this study demonstrate the potential for HAB toxins to be inducted into coastal
RO intake facilities, and the ability of typical RO operations to effectively remove these toxins.
The ability to accurately and rapidly identify an emerging HAB event is of high
importance. Monitoring of HAB species and other pertinent chemical/physical parameters at
two piers in southern California, Newport and Redondo Beach, was used to investigate the
development of a site-specific bloom definition for identifying emerging DA events. The
neurotoxin DA is produced by the chain forming diatom Pseudo-nitzschia, and it is the most
common HAB organism in southern California. Emphasis was given to abundances of the P.
seriata size category of Pseudo-nitzschia due to the prevalence of this size class in the region.
P. seriata bloom thresholds were established for each location based on deviations from their
respective long-term mean abundances, allowing the identification of major and minor blooms.
Sixty five percent of blooms identified at Newport Beach coincided with measurable DA
concentrations, while 36% of blooms at Redondo Beach coincided with measurable DA. Bloom
definitions allowed for increased specificity in multiple regression analysis of environmental
forcing factors significant to the presence of DA and P. seriata. The strongest relationship
identified was between P.seriata abundances two weeks following upwelling events at Newport
Beach.
Blooms of Pseudo-nitzschia can develop at depth in offshore waters not encompassed
by coastal HAB monitoring programs. California sea lions are predominately associated with DA
mortality events on the US west coast undoubtedly due to their large population sizes and
3
overlapping distribution with Pseud-nitzschia. Quantifying the amount of DA in these animals
and correlating this information with the presence of DA in phytoplankton and the local food
web has become a research focus for many scientists. However differences in materials,
equipment, technical capability, budgets and objectives of the various groups and/or agencies
involved in this work have influenced the DA quantification platforms employed. The
performance of two commercially available enzyme-linked immunosorbent assays for the
analysis of DA in a spectrum of California sea lion body fluids was compared to the results
obtained with liquid chromatography-mass spectrometry of the same samples. The results
indicated differences among these approaches, presumably owing to matrix effects (particularly
urine) and antibody reactivities. This information implies that care should be taken in
attempting to compare datasets generated using different analytical platforms and interpreting
the results of published studies.
The Orange County Sanitation District diverted flow of secondarily treated effluent from
a discharge pipe located 8.0 km offshore at 60 m depth to a pipe located 1.6 km from shore at
17 m depth for three weeks in September of 2012. Two incubation experiments were
performed to examine the influence of treated effluent at various dilutions on natural, coastal
phytoplankton communities, the first initiated a week prior to the diversion (‘Pre-Diversion’)
and the second initiated a week after the start of the diversion (‘Mid-Diversion’). The overall
community response observed in both experiments following effluent addition was an increase
in diatom and picoeukaryote abundances, a decrease in picophotocyanobacteria and a
dramatic increase in heterotrophic bacteria abundance. The 1:10 effluent additions yielded
significant increases in chlorophyll a concentrations, although the Pre-Diversion 1:10
4
experiments exhibited a lag in response to effluent addition. The DA producing diatom
Pseudo-nitzschia was present throughout both experiments, however DA production was only
detected in the Mid-Diversion experiment. The highest concentration of DA measured, 0.42 ±
0.057 µg/L coincided with phosphate and silicate concentrations below the detection limit of
the method, suggesting limitation by these macronutrients.
5
Chapter 1: Algal toxins and reverse osmosis desalination operations: laboratory bench
testing and field monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid.
ABSTRACT
The occurrence and intensity of harmful algal blooms (HABs) have been increasing
globally during the past few decades. The impact of these events on seawater desalination
facilities has become an important topic in recent years due to enhanced societal interest and
reliance on this technology for augmenting world water supplies. A variety of harmful bloom-
forming species of microalgae occur in southern California, as well as many other locations
throughout the world, and several of these species are known to produce potent neurotoxins.
These algal toxins can cause a myriad of human health issues, including death, when ingested
via contaminated seafood. This study was designed to investigate the impact algal toxin
presence has on both the intake and reverse osmosis (RO) desalination process; most
importantly, whether or not the naturally occurring algal toxins can pass through the RO
membrane and into the desalination product. Bench-scale RO experiments were conducted to
explore the potential of extracellular algal toxins contaminating the RO product.
Concentrations exceeding maximal values previously reported during natural blooms were used
in the laboratory experiments, with treatments comprised of 50 µg/L of domoic acid (DA), 2
µg/L of saxitoxin (STX) and 20 µg/L of brevetoxin (PbTx). None of the algal toxins used in the
bench-scale experiments were detectable in the desalinated product water. Monitoring for
intracellular and extracellular toxin concentrations of DA, STX, PbTx and okadaic acid (OA)
within the intake and desalinated water from a pilot RO desalination plant in El Segundo, CA,
6
was conducted from 2005 to 2009. During the five-year monitoring period, DA and STX were
detected sporadically in the intake waters but never in the desalinated water. PbTx and OA
were not detected in either the intake or desalinated water. The results of this study
demonstrate the potential for HAB toxins to be inducted into coastal RO intake facilities, and
the ability of typical RO operations to effectively remove these toxins.
INTRODUCTION
Burgeoning worldwide population has amplified societal interest in developing new and
reliable sources of potable water, particularly in naturally arid, highly populated regions. One
example of this trend is the semi-arid region of southern California, USA, which has been
confronted with shortages of freshwater availability since the early 1900s. Several
municipalities and public works agencies have begun to examine seawater desalination as a
local source of potable water that can augment existing supplies and address the
environmental concerns associated with the current imported water supply.
Seawater desalination faces a number of challenges in order to contribute significantly
as an environmentally safe and cost-effective source of potable water in California. One factor
complicating the use of desalinated seawater for human consumption is the increasing
frequency and impact of nuisance and harmful algal blooms (HABs) in coastal southern
California waters (Lewitus et al. 2012). Increases in HAB events have also been documented
worldwide, with anthropogenic influences being identified as significant contributors to these
increases (Anderson et al. 2002, Glibert et al. 2005, Heisler et al. 2008, Kudela et al. 2008, Paerl
& Paul 2012). HABs can negatively impact coastal seawater desalination facilities through
7
accumulations of high microalgal biomass near intake pipes that increase the solids load to
prefiltration processes, increase membrane fouling or biofouling in reverse osmosis (RO)
membranes, and increased chemical consumption (Abdul Azis et al. 2000, Ladner et al. 2010,
Zhang et al. 2011, Franks et al. url). These concentrated plumes of algal biomass may result in
the presence of significant levels of extracellular algal toxins in the intake water and a reduction
in dissolved oxygen concentrations as the bloom undergoes decomposition that can make
treatment more challenging(Caron et al. 2010). Seawater desalination facilities in the Persian
Gulf and the Gulf of Oman have experienced temporary shutdown of their plants during periods
of high algal biomass blooms near their intake pipes, until algal biomass sufficiently decreased
(Pankrantz 2008, Nazzal 2009, Richlen et al. 2010).
While investigations have been conducted to address pretreatment strategies for the
removal of high microalgal biomass load prior to introducing the feedwater to the RO process
(Pearce et al. 2004, Kim & Yoon 2005, Kwon et al. 2005, Castaing et al. 2010, Desormeaux et al.
2011, Vardon et al. 2011), the fate of microalgal toxins within the desalination process has not
been adequately addressed. The wide variety of toxins currently known to be produced in
naturally occurring microalgal blooms and the possibility for more than one type of toxin to
occur at any given time (Van Dolah 2000) complicate our understanding of the impact that
toxin presence may have on human health via the desalinated product water. In southern
California, microalgal species known to produce domoic acid (DA), saxitoxin (STX), brevetoxin
(PbTx) and okadaic acid (OA) are frequent, albeit highly variable contributors to the local
microalgal community (Caron et al. 2010, Lewitus et al. 2012). The interaction between the
potential presence of these toxins and the RO desalination process are an important
8
consideration for existing and future coastal desalination operations in the area. It has been
demonstrated in previous work on predicting molecule rejection that molecular weight alone
may not determine the rejection capability of a membrane (Verliefde et al. 2006). Rejection
capabilities can be impacted by the hydrophobicity of the molecule, surface charge, operating
conditions such as the specific flux, and feed water composition. Therefore, it is useful to
empirically examine the retention or passage of a molecule of interest.
DA is a neurotoxin produced by species of the diatom genus Pseudo-nitzschia in
southern California (Caron et al. 2010). It is a hydrosoluble molecule with a molecular mass of
311.14 g/mole. It is generally known that RO membranes are designed to efficiently reject the
vast majority of molecules greater than approximately 250g/mole, making DA an interesting
toxin to challenge an RO membrane. DA is seasonally present in the southern California region
and therefore has raised concern for successful operation of a desalination plant in the area.
Human exposure to DA is typically a consequence of consuming contaminated seafood
(generally filter-feeding fish and shellfish). The ensuing condition is referred to as Amnesic
Shellfish Poisoning (ASP) and results in symptoms ranging from gastroenteritis (vomiting,
diarrhea, abdominal cramps) to confusion, memory loss, disorientation, seizures, coma and/or
cranial nerve palsies or death (Perl et al. 1990a, Wright et al. 1990).
STX is the parent compound of the group of neurotoxins classified as the Paralytic
Shellfish Toxins (PSTs). The known producer of PSTs in southern California is the dinoflagellate
Alexandrium catenella (Caron et al. 2010, Garneau et al. 2011, Lewitus et al. 2012). STX is the
most potent of the more than 30 identified PSTs, and is classified as a chemical weapon in
Schedule 1 of the Chemical Weapons Convention (Llewellyn 2006). The hydrosoluble STX
9
molecule has a molecular mass of 299.3 g/mol, making it closer than DA to the theoretical
molecular weight cutoff of an RO membrane. STX is not as pervasive as DA in the southern
California area, but recent research has continued to document STX along the California coast
(Jester et al. 2009b, Garneau et al. 2011), highlighting it as a concern for successful operation of
desalination facilities in the area. Due to the potency of STX, A. catenella does not need to be a
dominant member of the microalgal community in order to constitute a significant risk to
human health (Burkholder et al. 2006). Consumption of a lethal dose can result in death within
hours due to muscular paralysis and respiratory difficulty. Worldwide, over 2000 illnesses each
year can be attributed to consumption of seafood contaminated with PSTs, with a 5-10%
mortality rate (Hallegraeff 2003).
PbTxs are a suite of neurotoxins that can be produced by raphidophyte algae found in
southern California, specifically Chattonella marina, Heterosigma akashiwo, and Fibrocapsa
japonica (Caron et al. 2010, Lewitus et al. 2012). There are 13 derivatives of PbTx that have
been identified, the most common to the marine environment are PbTx-2, PbTx-3 and PbTx-9
and of these three, PbTx-2 and PbTx-3 are the most potent (Baden et al. 2005). The molecular
masses of the liposoluble PbTxs are large at approximately 899 g/mol. While their size would
predict successful rejection by RO membranes, their hydrophobic nature may cause unique
solute-membranes interactions, impacting rejection capabilities. Humans that consume
seafood containing PbTxs may experience Neurotoxic Shellfish Poisoning (NSP), which causes
symptoms of nausea, vomiting, abdominal cramps, paresthesia and respiratory illness and/or
failure (Kirkpatrick et al. 2004).
10
OA is a member of a suite of toxins identified as Diarrhetic Shellfish Toxins (DSTs) which
includes dinophysistoxins and pectenotoxins (Caron et al. 2010, Lewitus et al. 2012). DSTs can
be produced by members of the dinoflagellate genus Prorocentrum, although they are more
commonly produced by the genus Dinophysis. The two Dinophysis species on the west coast of
the US known to have the capability of producing OA are D. acuminata and D. fortii (Yasumoto
et al. 1980, Murata et al. 1982, Yasumoto et al. 1985). OA is a liposoluble molecule with a
molecular mass of 805 g/mol, much larger than the theoretical molecular weight rejection
capabilities of RO membranes but solute-membrane interactions may be negatively impacted
by it’s hydrophobicity. Ingestion of seafood containing OA and other DSTs leads to Diarrhetic
Shellfish Poisoning (DSP) in humans. Symptoms of DSP include abdominal cramps,
inflammation of the intestinal tract, and diarrhea (Hallegraeff 2003).
Microalgae that are capable of producing DA, STX, PbTx and OA are known to be present
in southern California waters, although to date only DA and STX have been routinely observed
(Anderson et al. 2006, Busse et al. 2006, Mengelt 2006, Schnetzer et al. 2007, Sekula-Wood et
al. 2009, Garneau et al. 2011, Lewitus et al. 2012). The seasonality of these latter toxins
throughout the fifteen coastal California counties has been established from an analysis of
shellfish tissues by the Marine Biotoxin Monitoring Program (MBMP) of the California
Department of Public Health (CDPH) during the period 2002-2007 (Langlois 2007, Caron et al.
2010). On average, DA concentrations exhibit a strong maximum in spring and minor maximum
in fall, while STX concentrations exhibit a maximum in late summer to early fall.
Experimental studies were conducted in the laboratory using a bench-scale RO
apparatus to examine if extracellular DA, STX or PbTx at concentrations at or above those
11
observed during substantial blooms of toxin-producing microalgae in the region pose a threat
for passage through RO membrane during standard operations at the RO desalination plant
operated by the West Basin Municipal Water District. Operating conditions (i.e. pressure,
specific flux and recovery) and feed water composition (i.e. temperature and pH) reflected the
conditions experienced at their facility allowing accurate prediction of algal toxin rejection. OA
was not included in the bench-scale experiments due to the lack of a commercially available
ELISA platform capable of analyzing extracellular OA. In addition, the intake and resulting
desalinated water from the pilot desalination plant in El Segundo, CA, was monitored for
various microalgal toxins over a five-year period in order to assess the potential for naturally
occurring algal toxins to be inducted into seawater desalination operations in southern
California. The monitoring program offered insight into the magnitude and duration of toxic
HABs in a coastal location during this period.
MATERIALS AND METHODS
Experiments using bench-top reverse osmosis. Experiments designed to directly test the
efficacy of RO to remove microalgal toxins common to southern California waters were
conducted using a bench-scale RO setup. Monitoring of microalgal toxins in the extracellular
phase is not customary and the extracellular concentrations chosen for the bench-scale RO
experiments were based upon previously published values for intracellular concentrations, with
the assumption that the pretreatment process (micro or ultrafiltration) may efficiently disrupt
cells, and in the worst-case scenario release all toxin previously contained within the cell into
the extracellular phase (Ladner et al. 2010, Desormeaux et al. 2011). In general, the microalgal
12
toxin concentrations used for the bench-scale RO challenge were derived to exceed typical
concentrations of extracellular microalgal toxins observed in US west coast seawater (e.g. ~10
times greater). Concentrations of 50 µg/L extracellular DA (Sigma Aldrich®; St. Louis, MO), 2
µg/L extracellular STX (National Research Council, Institute for Marine Biosciences; Halifax,
Canada) and 20 µg/L extracellular PbTx-2 (World Ocean Solutions, LLC; Durham, NC) were used
in the laboratory experiments.
The bench-scale RO experiments were undertaken using a SEPA® CF II Membrane
Element Cell (Osmonics Inc., Minnetonka, MN) and SCW4+ RO membranes (Hydranautics,
Oceanside, CA). This instrument was selected because it could provide a cross-flow
environment, not simple dead-end filtration, that more closely represents full-scale RO facilities
and was capable of operating under the pressure required for seawater applications. The
experimental setup was constructed as a closed system, with water pumped from a reservoir
(10 L) containing the algal toxin onto the RO membrane apparatus at a specified pressure, and
both permeate and retentate were returned to the reservoir (Fig. 1-1). The reservoir in the
bench-scale RO experiments consisted of 10L of seawater previously passed through a
microfiltration process (20 µm), and 10L of concentrated seawater collected from the retentate
side of the pilot desalination plant in a 20L polycarbonate carboy. The resulting mixture
targeted a conductivity of approximately 85mS/cm to represent the midpoint of salinity in the
feed/concentrate stream for a typical desalination system with 50% recovery.
13
Figure 1-1. Diagram of the bench-scale RO experimental setup. The dashed line in the RO unit depicts the RO membrane.
Sampling locations for the reservoir, permeate and retentate are noted by the large black arrows.
The closed system approach was used in order to maintain the overall toxin
concentration during each experiment. Retentate water was passed through a cooling water
bath prior to its return to the reservoir to minimize increases in water temperature, which
occurred due to a high pressure pump used in the RO system. Membranes were conditioned at
the start of experiments using dechlorinated tap water (tap water treated with 16 µM sodium
metabisulfate), followed by rinses with a 0.20 M MgSO
4
solution, dechlorinated tap water, a 0.5
M NaCl solution, dechlorinated tap water, a 50/50 mixture of natural, 0.2 µm filtered seawater
and the concentrated seawater retentate noted above. The extensive conditioning procedure
was designed to establish a steady state operating condition prior to the addition of toxin. The
procedure allowed for the confident removal of preservative present on the new membrane
and for the membrane to stabilize following the initial compaction experienced after first
exposure to high pressure. The conditioned membrane was then challenged with toxin, which
was added to the reservoir of the experimental setup. Pressure of the water pumped from the
14
reservoir into the RO unit was operated at a target pressure of 900 psi and adjusted throughout
the duration of the experiments in order to maintain a permeate flux of 9 gallons per square
foot of membrane per day (GFD) and a cross flow velocity at an average of 0.5m/sec.
Monitoring for conductivity, flux and specific flux (gallons per square foot of membrane per day
per unit pressure; SPF) were conducted at regular intervals throughout the duration of each
experiment.
The use of a high pressure pump to pass water through the RO membrane resulted in a
rise in water temperature, which was addressed by immersing the tubing of the SEPA cell
(retentate side) in a 16°C water bath. A consistent water temperature of 23°C was maintained
in the reservoir, approaching the water temperature typically observed at the pilot desalination
plant at El Segundo, CA. A mass balance calculation of the amount of toxin in the reservoir and
RO unit system during each experiment was conducted in order to ensure that degradation of
microalgal toxins during the bench-scale experiments due to light and temperature fluctuations
was negligible.
Field sampling. The intake and resulting desalinated water (permeate) from the pilot
desalination plant located in El Segundo, CA, was examined for the presence and
concentrations of the algal toxins DA, STX, PbTx and OA to test the efficacy of toxin removal by
the RO process (Figure 1-2). These microalgal toxins were chosen due to their known
occurrence in southern California and/or the occurrence of the potential toxin-producing
microalgae. Sampling was conducted on a weekly basis during months of full plant operation
between 2005 and 2009. The pilot plant operated by the West Basin Municipal Water District
15
(Carson, CA) was co-located at the El Segundo Generating Station (33°55'49.88"N,
118°26'8.31"W). The intake pipe was located at a depth of approximately 10 m in Santa
Monica Bay, CA. The intake water was sampled following passage through a micro-prefiltration
system, and the permeate was sampled post-RO-desalination. Samples to be processed for DA,
STX and OA were collected in one liter polycarbonate bottles and stored at 4°C for no longer
than 24 hours until processing. Samples to be processed for PbTx were collected in one liter
glass bottles, due to the tendency of PbTx to adsorb to plastic, and stored at 4°C for no longer
than 24 hours until processing. Samples for the analysis of intracellular toxins were collected by
filtering 200-400 mL of the intake water onto GF/F Whatman filters, which concentrated the
cells present in the water and improved the limit of detection for the different algal toxins
analyzed. Extracellular algal toxin samples were analyzed on aliquots of seawater collected
from the filtrate of the intake water, and from the permeate water of the RO system. All
intracellular and extracellular samples were stored at -20°C until analysis via ELISAs.
Figure 1-2. Location of the pilot desalination plant operated by the West Basin Municipal Water District in El Segundo, CA, USA,
co-located at the El Segundo Generating Station.
16
Analysis of algal toxins. Concentrations of DA were analyzed using the Amnesic Shellfish
Poisoning ELISA from Biosense™ Laboratories (Bergen, Norway) for field samples collected
between 2005 and 2008, employing adaptations for intracellular samples and extracellular
seawater analyses (Schnetzer et al. 2007). The limit of detection for this method was 0.1 µg/L
for extracellular samples and 0.003 µg/L for intracellular samples (the latter value was
dependent on the volume of sample filtered). All other DA samples were analyzed using the
Mercury Science DA ELISA (Durham, NC). The samples from the permeate of the DA bench-
scale experiment were analyzed on both DA ELISA platforms. The limit of detection for the
latter method was 0.2 µg/L for extracellular samples and 0.007 µg/L for intracellular samples.
Samples for the determination of STX, PbTx and OA were analyzed using ELISA kits available
from Abraxis (Warminster, PA). The limit of detection for extracellular STX samples was 0.02
µg/L (intracellular STX was not measured as part of this study). The STX ELISA is capable of
detecting STX with 100% efficiency and other PSTs at ≤29%, as reported by the manufacturer,
reflecting the ability to identify STX but the low efficiency to detect the other analogues. Data
reported here on STX is described in µg/L of STX and not STX equivalents. A detection limit of
0.008 µg/L was achieved for intracellular PbTx and 0.1 µg/L for extracellular PbTx. The PbTx
ELISA is reported by the manufacturer to detect PbTx 5 at 127%, deoxy PbTx 2 at 133%, PbTx 2
at 102% and PbTx 3 at 100% efficiency. The limit of detection for intracellular OA samples was
0.008 ng OA/mL. The OA ELISA was capable of detecting OA 100%, as well as DTX 1 and DTX 2
with 50% efficiency, as reported by the manufacturer. Detection of intracellular toxins were
more sensitive than the analyses for extracellular toxins because significant volumes of water
were filtered for the measurement of intracellular toxins. For intracellular toxin analysis (DA,
17
PbTx and OA) samples were extracted in 3mL of 10% methanol (Fisher) and sonicated for 60
seconds. Following sonication, the samples were centrifuged for 10 minutes at 4000rpm.
Dilutions of the supernatant were prepared with the sample dilution buffer specified by the
individual ELISA assays. Extracellular toxin samples (DA, STX, PbTx and OA) were vortexed
briefly and diluted with the sample dilution buffer. Intracellular and extracellular samples were
handled per manual instructions for each ELISA platform but were optimized prior to sample
analysis in order to minimize matrix effects and false positives. Tests for false positives were
conducted by analyzing the extraction fluid (10% methanol) and toxin-free seawater at various
dilutions on each ELISA platform. The dilution employed to analyze the laboratory experiments
was selected based on these results. Efficacy of the optimization was tested using filtered
seawater spiked with known concentrations of toxins (positive controls). Based on the results
of these procedures, intracellular DA concentrations were measured at a minimum dilution of
1:25 on the Biosense ELISA and a minimum 1:10 dilution on the Mercury Science ELISA.
Extracellular DA concentrations were measured at a minimum 1:10 dilution on the Biosense
ELISA and a minimum 1:2 dilution on the Mercury Science ELISA. Intracellular and extracellular
STX and PbTx as well as extracellular OA samples were measured at a minimum 1:10 dilution.
Completed ELISA plates were read at 425nm on a ThermoMax Microplate reader (Molecular
Devices; Sunnyvale, CA) and data compiled by SoftMax Pro software (Molecular Devices;
Sunnyvale CA).
18
RESULTS
Bench-top RO experiments with DA, STX and PbTx. The RO process employed in this study
resulted in the reduction of DA, STX and PbTx in the permeate of the benchtop setup in all
experiments, as evidenced by the lack of detectable concentrations of any of the toxins tested.
No DA was detected in the permeate at any time during the experiment, regardless of the DA
ELISA used for analysis (Figure 1-3A). The DA concentration in the reservoir averaged 53.8 ± 6.9
µg/L throughout the 96-hour experiment, with no perceptible or consistent change over the
course of the trial (Figure 1-3A). DA concentration in the retentate of the RO setup averaged
52.9 ± 6.4 µg/L over the 96-hour experiment, and no significant trend in DA concentration in
the retentate was observed during the experiment (Figure 1-3A). The maintenance of a mass
balance for DA in the reservoir and retentate signified that the toxin was not significantly
degraded during the course of the bench-scale experiment due to light, temperature or
microbial processes, and was not appreciably adsorbed or absorbed onto the membrane or
other surfaces of the experimental setup. The flux (GFD) and specific flux (SPF) for the DA trial
averaged 8.41 ± 0.65 gfd and 0.026 ± 0.001 gfd/psi (average ±1 standard deviation; Figure 1-
3B), respectively. The applied pressure ranged from 780 to 880 psi in the DA trial in order to
maintain a permeate flux near the 9 gfd target. The relative constancy of these values
throughout the duration of the 96-hour experiment indicated that membrane performance
remained stable during the experiment, with no noticeable fouling that would result in changes
in these parameters.
19
Figure 1-3. (A) The change in concentration of extracellular DA over the duration of the bench-scale RO experiment (96h), as
measured in the reservoir (light gray triangles), permeate (open triangles) and retentate (dark gray triangles) streams. The
open triangles denote concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO membrane
over the duration of the experiment, represented by changes in the specific (SPF) and constant flux (GFD).
(A)
(B)
20
No STX was detected in any of the permeate samples collected for the duration of the
48-hour experiment. The concentration of STX in the reservoir averaged 2.5 ± 0.4 µg/L and the
concentration in the retentate averaged 2.7 ± 0.3 µg/L (Figure 1-4A). The reservoir water
contained 0.4 µg/L extracellular STX prior to the addition of the 2.0 µg/L STX standard, resulting
in a higher concentration of STX in the reservoir for the duration of the experiment from the
expected value of 2.0 µg/L . Toxin levels in the reservoir indicated STX was not significantly
degraded due to light, temperature or microbial processes, and was not appreciably adsorbed
or absorbed onto the membrane or other surfaces of the experimental setup.
Membrane performance during the STX experiment yielded flux and specific flux values
that averaged 9.10 ± 0.20 gfd and 0.033 ± 0.001 gfd/psi, respectively, and indicated no
consistent change during the 48-hour experiment (Figure 1-4B). The applied pressure in the
STX trial ranged from 492 to 787 psi in order to maintain a permeate flux near the 9 gfd target.
21
Figure 1- 4. (A) The change in concentration of extracellular STX over the duration of the bench-scale RO experiment (48h), as
measured in the reservoir (light gray triangles), permeate (open triangles) and retentate (dark gray triangles) streams. The
open triangles denote concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO membrane
over the duration of the experiment, represented by changes in the specific (SPF) and constant flux (GFD).
(A)
(B)
22
No PbTx was detected in any of the permeate samples collected throughout the
duration of the experiment, while the reservoir maintained an average concentration of PbTx-2
of 23.0 ±2.0 µg/L and the retentate maintained an average concentration of 20.5 ± 2.2 µg/L
(Figure 1-5A). No change in the concentration of PbTx-2 was observed in the reservoir during
the experiment, while slight changes appeared to occur in the retentate (approximately 20%).
Maintenance of mass balance for PbTx-2 in the combined volume of reservoir and retentate
throughout the experiment, indicated PbTx-2 was not significantly degraded due to light,
temperature or microbial processes, and was not appreciably adsorbed or absorbed onto the
membrane or other surfaces of the experimental setup. Flux and specific flux for the bench-
scale RO setup during the PbTx experiment averaged 9.38 ± 0.34 gfd and 0.031 ± 0.002 gfd/psi,
respectively, with no marked changes in these parameters over the course of the 48-hour
experiment (Figure 1-5B). The applied pressure in the PbTx trial ranged from 780 to 880 psi to
maintain a permeate flux near the target of 9 gfd.
23
Figure 1- 5. (A) The change in concentration of extracellular PbTx over the duration of the bench-scale experiment (48h), as
measured in the reservoir (light gray triangles), permeate (open triangles) and retentate (dark gray triangles) streams. The
open triangles denote concentrations under the limit of detection of the ELISA method used. (B) Fouling of the RO membrane
over the duration of the experiment, represented by changes in the specific (SPF) and constant flux (GFD).
(A)
(B)
24
Algal toxins in the intake and desalinated water of a pilot desalination plant. The intake and
desalinated water at the El Segundo pilot desalination plant was monitored for intracellular and
extracellular DA during the 2005 – 2009 study period, for a total 128 samples. Intracellular DA
was measured in the intake water most commonly in the spring and early summer months
(Figure 1-6). The highest intracellular DA concentration of 4.0 µg/L was measured in the intake
water during April 2006. Extracellular DA was measured in the intake water most commonly in
the spring and early summer months, concomitant with high intracellular DA concentrations,
but was also detected infrequently in winter months (Figure 1-6). The highest extracellular DA
concentration of 10.1 µg/L was detected in the intake water in April 2008. No DA was measured
in the desalinated water throughout the extent of the field experiment.
25
Figure 1-6. The average monthly intracellular (top panel) and extracellular (bottom panel) DA concentration detected in the
pre-microfiltrated intake water during the 2005-2009 sampling period; bars are color coded based on sample year. The
maximum DA concentration observed for each month, regardless of year, is marked on the graph by a black circle. The open
circle denotes concentrations under the limit of detection of the ELISA.
26
A total of 25 samples of the intake and desalinated waters were monitored for
extracellular STX during 2008 and 2009. STX was detected in nearly every month sampled
(Figure 1-7), with the highest extracellular STX concentration of 0.3 µg/L measured during April
2009. No STX was detected in the desalinated water throughout the experiment.
Figure 1-7. The average monthly extracellular STX concentration detected in the pre-microfiltrated intake water during the
2008-2009 sampling period; bars are color coded based on sample year. The maximum extracellular STX concentration
observed for each month, regardless of year, is marked on the graph by a black circle. The open circle denotes concentrations
under the limit of detection of the ELISA. The star signifies the months samples were not taken due to plant inoperation.
Monitoring for intracellular and extracellular PbTx in the intake and desalinated water
was conducted only in March and April of 2009. PbTx has not been previously demonstrated in
the region, although raphidophyte species capable of producing PbTx have been documented in
these coastal waters (Caron et al. 2010, Lewitus et al. 2012). Samples were collected multiple
27
times a week, for a total of 12 samples. The 12 samples of the intake and desalinated water
yielded no detectable levels of intracellular or extracellular PbTx.
Monitoring of the intake water for intracellular OA occurred during 2008 and 2009. OA
has not been previously detected in microalgal samples in southern California, but it has been
detected in shellfish samples in central California (Sutherland 2008). None of the 22 samples
analyzed had detectable levels of intracellular OA.
DISCUSSION
The bench-scale experiments performed in this study were designed to investigate
whether algal toxins commonly encountered in coastal ecosystems would be effectively
removed from the permeate water produced by commercial RO desalination processes. The
membranes, pressures and fluxes employed in our bench-scale apparatus, the toxins and
concentrations employed in our laboratory experiments, modeled the full-scale RO desalination
plant operated by the West Basin Municipal Water District for the Redondo Beach area in
southern California, making the results relevant for other desalination plants planning to
operate in the area. The toxin concentrations used in the study were chosen to represent peak
challenge concentrations, but were still ecologically relevant concentrations of toxins in coastal
waters. The experiments were performed for durations of time that might mimic peak
concentrations of these toxins in natural blooms occurring in the region.
Past work in the southern California area (Anderson et al. 2006, Busse et al. 2006,
Mengelt 2006, Schnetzer et al. 2007, Sekula-Wood et al. 2009), analyses of DA concentrations
in shellfish from the MBMP data collected by the CDPH, (Langlois 2007, Caron et al. 2010) and
28
monitoring results from the present study have demonstrated that DA is the most prevalent
algal toxin in southern California waters. This toxin constitutes a potential threat to human
health in many other regions globally (Lelong et al. 2012, Trainer et al. 2012), therefore the
results of this study are widely relevant to desalination operations. Our results demonstrated
that RO desalination was effective in removing all detectable levels of DA from the RO
permeate, even at concentrations typical of an extreme, natural, toxic DA event (up to 50 µg/L
of extracellular DA).
The concentration of extracellular DA used to challenge the RO membrane in the bench-
scale experiment of this study is comparable to DA concentrations that have been reported in
the literature for naturally occurring outbreaks of DA in the southern California region (see
Caron et al. 2010 for a review). It is important to note that reports of DA in the literature have
frequently focused on concentrations of intracellular toxin rather than extracellular
concentrations. The former values are societally relevant because human exposure to DA
typically is the result of the consumption of seafood that has become contaminated with DA
through the ingestion of toxin-containing cells of Pseudo-nitzschia. That trophic connection has
led to a strong correlation between DA in planktonic algal organisms and ASP events.
However, it is probable that the prefiltration process used to remove the algal biomass load in
seawater prior to RO desalination causes morphological damage to microalgal cells, resulting in
the release of toxic compounds contained within the cells and may be responsible for a portion
of the extracellular DA observed in the intake water from the El Segundo pilot plant over the
course of this study.
29
Detectable quantities of extracellular DA were present in the intake water of the pilot
desalination plant during most months of the year during the present study, while intracellular
DA was detected the first six months of the year (Figure 1-6). These differences between the
detection of extracellular and intracellular DA cannot easily be attributed to methodological
differences because our intracellular DA analyses had a lower limit of detection than analyses
for extracellular toxin. Another potential driver of the observed discrepancies could be due to
prefiltration of the intake water, noted above. Prefiltration of the intake water may cause
either the removal or breakage of cells, consequently decreasing the presence of intracellular
and increasing the presence of extracellular DA in our samples. Alternatively, active release of
DA by Pseudo-nitzschia spp. could explain why detectable concentrations of extracellular DA
were more prevalent in the samples than intracellular DA. Numerous hypotheses have been
explored in hope of explaining the process of production and release of DA by Pseudo-nitzschia
spp. Laboratory experiments have shown that DA may function as a metal chelator to facilitate
the acquisition of iron and copper from the environment (Bates et al. 2000, Rue & Bruland
2001, Maldonado et al. 2002, Wells et al. 2005), as a deterrent to the grazing activities of
zooplankton (Maneiro et al. 2005, Bargu et al. 2006, Olson et al. 2006, Olson & Lessard 2010)
and/or as an allelopathic compound to retard the growth of other algal species competing for
nutrients (Subba Rao et al. 1995, Lundholm et al. 2005).
Monitoring the concentrations of intracellular and extracellular DA in the intake water
of the pilot desalination plant in El Segundo, CA confirmed the strong spring seasonality of
these events in southern California waters previously indicated by analyses of the MBMP
shellfish DA data (Langlois 2007, Caron et al. 2010). Nevertheless, DA was not detected in the
30
permeate water samples from the pilot desalination plant at any time during the five-year
monitoring period despite the presence of high concentrations of intracellular DA
concentrations (4.0 µg/L in April 2006) and extracellular DA (10.1 µg/L in April 2008). The lack
of detectable DA concentrations in the permeate suggest the RO process is effective in
removing this dangerous neurotoxin from the desalination product.
Even the smallest concentrations of STX in oceanic waters are reason for human health
concern, due to its high toxicity and associated human mortality rates (Hallegraeff 2003). Our
bench-scale RO experiment with STX indicated that RO desalination was effective in reducing
the concentration of STX from seawater below detectable levels when the concentration in the
reservoir was as high as 2 µg/L. An analysis of PST concentrations in shellfish along the
California coast by the MBMP indicates that in recent years, STX has been more prevalent in
northern California than in southern California (Langlois 2007, Caron et al. 2010). Accordingly,
most previous studies of this class of toxins on the US west coast have focused primarily on
occurrence in central and northern California as well as Oregon and Washington (Jester 2008,
Lefebvre et al. 2008, Jester et al. 2009a). However, our results demonstrate that the
extracellular form of STX was present in the El Segundo pilot desalination plant intake water
samples throughout much of the year (Figure 1-7). A recent study conducted in the
neighboring King Harbor, City of Redondo Beach, CA (Garneau et al. 2011) indicated the
sporadic occurrence of intracellular STX and the dinoflagellate, Alexandrium catenella, a
putative producer of this toxin. These findings of the pervasiveness of extracellular STX in
southern California highlight the need for a clearer understanding of the distribution and
31
concentration of STX in the region, as well as information on the potential impact of this toxin
with respect to desalination operations.
PbTx and OA are not yet recognized as significant contributors to the mixture of
microalgal toxins commonly observed in southern Californian coastal waters. OA has been
detected in shellfish from Monterey Bay, CA, (Sutherland 2008) but has not been reported in
microalgal samples. PbTx has not been reported in shellfish or microalgal samples in California
to date. The concentration of PbTx-2 used in the bench-scale RO experiment was based on the
potential production for PbTx-2 demonstrated in laboratory studies of the three raphidophytes
that have been observed in southern Californian waters, Heterosigma akashiwo, Fibrocapsa
japonica and Chattonella marina (C. Tomas, personal comm.). PbTxs are large molecules, and
would be expected to be removed effectively by RO. Indeed, the bench-scale RO system
employed in this study reduced concentrations of extracellular PbTx-2 below the limit of
detection when the concentration in the reservoir was as high as 20 µg/L . PbTxs and OA were
not detected in southern California during the monitoring conducted in this study, but known
producers of PbTxs from the Raphidophyte class of microalgae and OA producers of the
dinoflagellate genus Dinophysis have been observed in local waters (Caron et al. 2010).
CONCLUSIONS
This study examined the potential impact that the presence of extracellular algal toxins
common to coastal waters of southern California may have on the quality of RO permeate by
challenging a bench-scale RO unit with high concentrations of DA, STX and PbTx. None of the
three toxins tested were detectable in the permeate over the duration of the experiments.
32
Field monitoring of intracellular and extracellular algal toxins in the intake waters of a pilot
desalination plant located in El Segundo, CA, were conducted over a five year period (2005 –
2009). DA and STX were present in the water throughout much of the year, with DA present
most commonly in the spring and early summer months and STX present in all four seasons.
Field monitoring of the intake waters was coupled with analysis of algal toxins in the
desalination product. No detectable levels of the algal toxins were observed during the five
years of monitoring.
REFERENCES
Abdul Azis, P., I. Al-Tisan, M. Al-Daili, T. Green, A. Dalvi, and M. Javeed. 2000. Effects of
environment on source water for desalination plants on the eastern coast of Saudi
Arabia. Desalination 132:29-40.
Anderson, C., M. Brzezinski, L. Washburn, and R. Kudela. 2006. Circulation and environmental
conditions during a toxigenic Pseudo-nitzschia australis bloom in the Santa Barbara
Channel, California. Mar Ecol Prog Ser 327:119-133.
Anderson, D., P. Glibert, and J. Burkholder. 2002. Harmful algal blooms and eutrophication:
Nutrient sources, composition and consequences. Estuaries 25:704-726.
Baden, D., A. Bourdelais, H. Jacocks, S. Michelliza, and J. Naar. 2005. Natural and derivative
brevetoxins: historical background, multiplicity and effects. Environ Health Persp
113:621-625.
Bargu, S., K. Lefebvre, and M. Silver. 2006. Effect of dissolved domoic acid on the grazing rate of
Euphasia pacifica Mar Ecol Prog Ser 312:169-175.
Bates, S., C. Leger, M. Satchwell, and G. Boyer. 2000. The effects of iron on domoic acid
production by Pseudo-nitzschia multiseries. Pages 320-323 in G. Hallegraeff, S.
Blackburn, C. Bolch, and R. Lewis, editors. 9th International Conference on Harmful
Algal Blooms. Intergovernmental Oceanographic Commision, Hobart, Tasmania.
Burkholder, J., R. Azanza, and Y. Sako. 2006. The ecology of harmful dinoflagellates. Pages 53-66
in E. Graneli and J. Turner, editors. Ecology of Harmful Algae. Springer Heidelberg,
Berlin.
33
Busse, L., E. Venrick, R. Antrobus, P. Miller, V. Vigilant, M. Silver, C. Mengelt, L. Mydlarz, and B.
Prezelin. 2006. Domoic acid in phytoplankton and fish in San Diego, CA, USA. Harmful
Algae 5:91-101.
Caron, D., M.-E. Garneau, E. Seubert, M. Howard, L. Darjany, A. Schnetzer, I. Cetinic, G. Filteau,
P. Lauri, B. Jones, and S. Trussell. 2010. Harmful algae and their potential impacts on
desalination operations off southern California. Water Res 44:385-416.
Castaing, J., A. Masse, M. Pontie, V. Sechet, J. Haure, and P. Jaouen. 2010. Investigating
submerged ultrafiltration (UF) and microfiltration (MF) membranes for seawater pre-
treatment dedicated to removal of undesirable micro-algae. Desalination 253:71-77.
Desormeaux, E., P. Meyerhofer, H. Luckenbach, and R. Kudela. 2011. Pilot-testing multiple
pretreatment systems for seawater desalination. Inter Desalination Assoc J First
Quarter:42-52.
Franks, R., M. Wilf, N. Voutchkov, P. Murkute, and J. Kizer. A pilot study using seawater reverse
osmosis membranes in combination with various pretreatments to meet the challenges
of Pacific seawater desalination. Accessed January 24, 2012: Retrieved at
http://www.membranes.com/docs/papers/New%20Folder/A%20PILOT%20STUDY%20
USING%20SEAWATER%20REVERSE%20OSMOSIS%20MEMBRANES.pdf.
Garneau, M.-E., A. Schnetzer, P. Countway, A. Jones, E. Seubert, and D. Caron. 2011.
Examination of the seasonal dynamics of the toxic dinoflagellate Alexandrium catenella
at Redondo Beach, California, by quantitative PCR. Appl Environ Micro 77:7669-7680.
Glibert, P., S. Seitzinger, C. Heil, J. Burkholder, M. Parrow, L. Codispoti, and V. Kelly. 2005. The
role of eutrophication in the global proliferation of hamrful algal blooms. Oceanogr
18:198-209.
Hallegraeff, G. 2003. Harmful algal blooms: a global overview. Pages 25-49 in G. Hallegraeff, D.
Anderson, and A. Cembella, editors. Manual on Harmful Marine Microalgae.
Monographs on Oceanographic Methodology. UNESCO Publishing, Paris.
Heisler, J., P. Glibert, J. Burkholder, D. Anderson, W. Cochlan, W. Dennison, Q. Dortch, C.
Gobler, C. Heil, E. Humphries, A. Lewitus, R. Magniun, H. Marshall, K. Sellner, D.
Stockwell, D. Stoecker, and M. Suddleson. 2008. Eutrophication and harmful algal
blooms: a scientific consensus. Harmful Algae 8:3-13.
Jester, R. 2008. An investigation into the prevalence of Alexandrium derived toxins in marine
food webs. University of California, Santa Cruz, Santa Cruz.
Jester, R., K. Baugh, and K. Lefebvre. 2009a. Presence of Alexandrium catenella and paralytic
shellfish toxins in finfish, shellfish and rock crabs in Monterey Bay, California, USA. Mar
Biol 156:493-504.
34
Jester, R., K. Lefebvre, G. Langlois, V. Vigilant, K. Baugh, and M. Silver. 2009b. A shift in the
dominant toxin-producing algal species in central California alters phycotoxins in food
webs. Harmful Algae 8:291-298.
Kim, S.-H., and J.-S. Yoon. 2005. Optimization of microfiltration for seawater suffering from red
tide contamination. Desalination 182:315-321.
Kirkpatrick, L., D. Squicciarini, L. Backer, W. Clark, J. Benson, Y. Cheng, D. Johnson, R. Pierce, J.
Zaias, G. Bossart, and D. Baden. 2004. Literature review of Florida red tide: implications
for human health effects. Harmful Algae 3:99-115.
Kudela, R., J. Lane, and W. Cochlan. 2008. The potential role of anthropogenically derived
nitrogen in the growth of harmful algae in California, USA. Harmful Algae 8:103-110.
Kwon, B., N. Park, and J. Cho. 2005. Effect of algae on fouling and efficiency of UF membranes.
Desalination 179:203-214.
Ladner, D., D. Vardon, and M. Clark. 2010. Effects of shear on microfiltration and ultrafiltration
fouling by marine bloom-forming algae. J Membrane Sci 356:33-43.
Langlois, G. 2007. Marine Biotoxin Monitoring Program Annual Report. in. California
Department of Public Health for California Department of Fish and Game, Sacramento.
Lefebvre, K., B. Bill, A. Erickson, K. Baugh, L. O'Rourke, P. Costa, S. Nance, and V. Trainer. 2008.
Characterization of intracellular and extracellular saxitoxin levels in both field and
cultured Alexandrium spp. samples from Sequim Bay, Washington. Mar Drugs 6:103-
116.
Lelong, A., H. Hegaret, P. Soudant, and S. Bates. 2012. Pseudo-nitzschia (Bacillariophyceae)
specoes, domoic acid and amnesic shellfish poisoning: revisting previous paradigms.
Phycologia 51:168-216.
Lewitus, A., R. Horner, D. Caron, E. Garcia-Mendoza, B. Hickey, M. Hunter, D. Huppert, R.
Kudela, G. Langlois, J. Largier, E. Lessard, R. RaLonde, J. Rensel, P. Strutton, V. Trainer,
and J. Tweddle. 2012. Harmful algal blooms along the North American west coast
region: History, trends, causes and impacts. Harmful Algae 19:133-159.
Llewellyn, L. 2006. Saxitoxin, a toxic marine natural product that targets a multitude of
receptors. Nat Prod Rep 23:200-222.
Lundholm, N., P. Hansen, and Y. Kotaki. 2005. Lack of allelopathic effects of the domoic acid-
producing marine diatom Pseudo-nitzschia multiseries. Mar Ecol Prog Ser 288:21-33.
Maldonado, M., M. Hughes, E. Rue, and M. Wells. 2002. The effect of Fe and Cu on growth and
domoic acid production in Pseudo-nitzschia australis. Limnol Oceanogr 47:515-526.
Maneiro, I., P. Iglesias, C. Guisande, I. Riveiro, A. Barreiro, S. Zervoudaki, and E. Graneli. 2005.
Fate of domoic acid ingested by the copepod Arcatia clausi. Mar Biol 148:123-130.
35
Mengelt, C. 2006. Ultravioloet photoecology, dark survival and seasonal abundance of Pseudo-
nitzschia australis and P. multiseries in coastal waters of central California. Dissertation.
University of California, Santa Barbara.
Murata, M., M. Shimatani, Y. Sugitani, Y. Oshima, and T. Yasumoto. 1982. Isolation and
structural elucidation of the causative toxin of the diarrhetic shellfish poisoning. B Jpn
Soc Sci Fish. 48:549-552.
Nazzal, N. 2009. 'Red tide' shuts desalination plant. in Gulf News, Dubai, UAE.
Olson, M., and E. Lessard. 2010. The influence of the Pseudo-nitzschia toxin, domoic acid, on
microzooplankton grazing and growth: A laboratory and field experiment. Harmful
Algae 9:240-247.
Olson, M., E. Lessard, C. Wong, and M. Bernhardt. 2006. Copepod feeding selectivity on
microplankton including the toxigenic diatoms Pseudo-nitzschia spp., in the coastal
Pacific Northwest. Mar Ecol Prog Ser 326:207-220.
Paerl, H., and V. Paul. 2011. Climate change: Links to global expansion of harmful
cyanobacteria. Water Res 46:1349-1363.
Pankrantz, T. 2008. Red tides close desal plants. Water Desalination Rep 44.
Pearce, G., S. Talo, K. Chida, A. Basha, and A. Gulamhusein. 2004. Pretreatment options for
large scale SWRO plants: case studies of UF trials at Kindasa, Saudi Arabia, and
conventional pretreatment in Spain. Desalination 167:175-189.
Perl, T., L. Bedard, T. Kosatsky, J. Hockin, E. Todd, L. McNutt, and R. Remis. 1990. Amnesic
Shellfish Poisoning: a new clincal syndrome due to domoic acid. Can Diseases Weekly
Rep 16:7-8.
Richlen, M., S. Morton, E. Jamali, A. Rajan, and D. Anderson. 2010. The catastrophic 2008-2009
red tide in the Arabian gulf region, with observations on the identification and
phylogeny of the fish-killing dinoflagellate Cochlodinium polykrikoides. Harmful Algae
9:163-172.
Rue, E., and K. Bruland. 2001. Domoic acid binds iron and copper: a possible role for the toxin
production by the marine diatom Pseudo-nitzschia Mar Chem 76:127-134.
Schnetzer, A., P. Miller, R. Schaffner, B. Stauffer, B. Jones, S. Weisberg, P. DiGiacomo, W.
Berelson, and D. Caron. 2007. Blooms of Pseudo-nitzschia and domoic acid in the San
Pedro Channel and Los Angeles harbor areas of the Southern California Bight, 2003-
2004. Harmful Algae 6:372-387.
Sekula-Wood, E., A. Schnetzer, C. Benitez-Nelson, C. Anderson, W. Berelson, M. Brzezinski, J.
Burns, D. Caron, I. Cetinic, J. Ferry, E. Fitzpatrick, B. Jones, P. Miller, S. Morton, R.
Schaffner, D. Siegel, and R. Thunell. 2009. Rapid donward transport of the neurotoxin
domoic acid in coastal waters. Nat Geosci 2:272-275.
36
Subba Rao, D., Y. Pan, and S. Smith. 1995. Allelopathy between Rhizosolenia alata (Brightwell)
and the toxigenic Pseudo-nitzschia pungens f. multiseris (Hasle). in P. Lassus, G. Arzul, E.
Erad, P. Gentien, and Marcaillou, editors. Harmful Marine Algal Blooms. Lavoisier,
Intercept Ltd, Paris.
Sutherland, C. 2008. Diarrhetic shellfish toxins linked to local Dinophysis populations in the
California coastal waters of Monterey Bay. University of California, Santa Cruz, Santa
Cruz.
Trainer, V., S. Bates, N. Lundholm, A. Thessen, W. Cochlan, N. Adams, and C. Trick. 2012.
Pseudo-nitzschia physiological ecology, phylogeny, toxicity, monitoring and impacts on
ecosystem health. Harmful Algae 14:271-300.
Van Dolah, F. 2000. Marine algal toxins: Origins, health effects, and their increased occurrence.
Environ Health Persp 108:133-141.
Vardon, D., M. Clark, and D. Ladner. 2011. The potential of laser scanning cytometry for early
warning of algal blooms in desalination plant feedwater. Desalination 277:193-200.
Verliefde, A., N. Van Vliet, G. Amy, B. Van der Bruggen, and J. van Dijk. 2006. A semi-
quantitative method for prediction of the rejection of uncharged organic
micropollutants with nanofiltration. Water Practice and Technology 1: doi:
10.2166/WPT.2006084.
Wells, M., C. Trick, W. Cochlan, M. Hughes, and V. Trainer. 2005. Domoic acid: the synergy of
iron, copper and the toxicity of diatoms. Limnol Oceanogr 50:1908-1917.
Wright, J., C. Bird, A. de Freitas, D. Hampson, J. McDonald, and M. Quilliam. 1990. Chemistry,
biology and toxicology of domoic acid and its isomers. Can Diseases Weekly Rep 16:21-
26.
Yasumoto, T., M. Murata, Y. Oshima, M. Sano, G. Matsumoto, and J. Clardy. 1985. Diarrhetic
shellfish toxins. Tetrahedron 41:1019-1025.
Yasumoto, T., Y. Oshima, W. Sugawara, Y. Fukuyo, H. Oguri, T. Igarashi, and N. Fujita. 1980.
Identification of Dinophysis fortii as the causative organism of diarrhetic shellfish
poisoning. B Jpn Soc Sci Fish 46:1405-1411.
Zhang, M., S. Jiang, D. Tanuwidjaja, N. Voutchkov, E. Hoek, and B. Cai. 2011. Composition and
variability of biofouling organisms in seawater reverse osmosis desalination plants. Appl
Environ Micro 77:4390-4398.
Reprinted from Water Research, Volume 46 Issue 19, EL Seubert, S Trussell, J Eagleton, A
Schnetzer, I Cetinić, P Lauri, BH Jones, DA Caron, Algal toxins and reverse osmosis desalination
operations: laboratory bench testing and field monitoring of domoic acid, saxitoxin, brevetoxin
and okadaic acid, pages 6563-6573, Copyright 2012, with permission from Elsevier.
37
Chapter 2: Seasonal and annual dynamics of harmful algae and algal toxins revealed through
weekly monitoring at two coastal ocean sites off southern California, USA
ABSTRACT
Reports of toxic harmful algal blooms (HABs) attributed to the diatom Pseudo-nitzschia
spp. have been increasing in California during the last several decades. Whether this increase
can be attributed to enhanced awareness and monitoring, or to a dramatic upswing in the
development of HAB events remains unresolved. Given these uncertainties, the ability to
accurately and rapidly identify an emerging HAB event is of high importance. Monitoring of
HAB species and other pertinent chemical/physical parameters at two piers in southern
California, Newport and Redondo Beach, was used to investigate the development of a site-
specific bloom definition for identifying emerging domoic acid (DA) events. Emphasis was given
to abundances of the P. seriata size category of Pseudo-nitzschia due to the prevalence of this
size class in the region. P. seriata bloom thresholds were established for each location based
on deviations from their respective long-term mean abundances, allowing the identification of
major and minor blooms. Sixty five percent of blooms identified at Newport Beach coincided
with measurable DA concentrations, while 36% of blooms at Redondo Beach coincided with
measurable DA. Bloom definitions allowed for increased specificity in multiple regression
analysis of environmental forcing factors significant to the presence of DA and P. seriata. The
strongest relationship identified was between P.seriata abundances two weeks following
upwelling events at Newport Beach.
38
INTRODUCTION
Substantial increases in microalgal biomass in planktonic ecosystems, generally
observed as increases in chlorophyll a concentrations or cell abundances, serve as the
foundation of highly productive oceanic food webs, spawning productive fisheries and foraging
areas for marine mammals, birds and other large predators (Legendre 1990). However, toxic or
harmful algal blooms (HABs) produced by a few species of microalgae can have negative
impacts on local food webs as well as threaten human health. Nearly 300 of the >4000
currently described species of marine microalgae are considered capable of forming HABs and
approximately 80 of those species are known to be capable of producing compounds that are
toxic to co-occurring marine species and/or humans (Sournia 1995). HAB events are highly
diverse in their taxonomic composition, spatiotemporal distributions and detrimental effects,
complicating the understanding of their ecology, reducing the accuracy of predicting outbreaks
and impeding the development of successful management strategies (Smayda 1997, Zingone &
Enevoldsen 2000, Anderson et al. 2012). Anthropogenically influenced changes in climate and
nutrient loading is in part responsible for the global increase in the incidence, magnitude and
duration of HAB events (Paerl 1997, Van Dolah 2000, Anderson et al. 2002, Glibert et al. 2005,
Heisler et al. 2008, Kudela et al. 2008, Paerl & Paul 2012). The impact anthropogenic driven
change will have on a given region will be determined by the HAB organisms present and the
magnitude of the change experienced in environmental conditions responsible for local HAB
initiation, maintenance and demise.
HABs along the coastline of California have been common for many years, although
poorly characterized until relatively recently. The first dependable records most likely began
39
with the documentation of massive red tides attributed to the dinoflagellate Lingulodinium
polyedrum that have occurred sporadically along the coast since the beginning of the 20
th
century (Torrey 1902, Allen 1938, 1946, Gregorio & Pieper 2000, Shipe et al. 2008, Omand et
al. 2011). Yessotoxin production by L. polyedrum in California strains has been confirmed
(Howard et al. 2008), but this toxin has failed to be accredited as the cause of any instance of
marine animal or human illness in the area. Numerous other potentially toxic species of
microalgae have also been documented in California waters, including several raphidophytes
and dinoflagellate species within the genera Akashiwo, Alexandrium, Cochlodinium, and
Dinophysis (Howard et al. 2007, Jessup et al. 2009, Jester et al. 2009b, Caron et al. 2010,
Garneau et al. 2011, Lewitus et al. 2012).
Members of the diatom genus Pseudo-nitzschia are one of the few non-flagellated
microalgae currently known to be capable of toxin production. The toxin they produce, domoic
acid (DA), is also produced by a closely related diatom, Nitzschia navis-varingica (Lundholm &
Moestrup 2000, Kotaki et al. 2004, Lundholm et al. 2004, Lelong et al. 2012), and potentially by
the diatom Amphora coffeaeformis (Maranda et al. 1990, Sala et al. 1998, Bates 2000). The
capability of Pseudo-nitzschia to produce the powerful neurotoxin was originally discovered on
Prince Edward Island, Canada, in 1987 when more than 150 people were sickened and three
perished after consuming blue mussels (Mytilus edulis) contaminated with DA during a bloom
of P. multiseries (Bates et al. 1989, Wright et al. 1989). Human consumption of shellfish
contaminated with the neurotoxin DA causes amnesic shellfish poisoning (ASP), the main
symptoms include disorientation, gastrointestinal illness, memory loss and even death. Pseudo-
nitzschia spp. has been reported as a frequent contributor to the microalgal community in
40
coastal waters of California since the early 1900s (Allen 1934, 1936, Fryxell et al. 1997),
however toxin production by these species was not documented until 1991 following the
poisoning and deaths of brown pelicans (Pelecanus occidentalis) and Brandt’s cormorants
(Phalacrocorax penicillatus) in Santa Cruz, CA , corresponding with a bloom of P. australis (Fritz
et al. 1992, Work et al. 1993). DA continues to be the cause of mortality events of marine
mammals and birds along the U.S. west coast and elsewhere (Lefebvre et al. 1999, Scholin et al.
2000, de la Riva et al. 2009, Fire et al. 2009, Fire et al. 2010, Hall & Frame 2010, Lefebvre et al.
2010) although no human deaths have been reported since the initial Canadian ASP outbreak in
1987. The threat of DA poisoning in humans from seafood other than shellfish has been
highlighted in studies of exposure via recreational fishing activities (Busse et al. 2006, Mazzillo
et al. 2010), as well as the potential for DA to be transferred to trophic levels not directly
consuming toxic Pseudo-nitzschia cells (Busse et al. 2006, Vigilant & Silver 2007, Kvitek et al.
2008, Mazzillo et al. 2011).
DA has been detected in strains of Pseudo-nitzschia from California, other US Pacific
coast states and in multiple other locations around the globe (Anderson et al. 2006, Schnetzer
et al. 2007, Lelong et al. 2012, Stauffer et al. 2012, Trainer et al. 2012, Schnetzer et al.
Accepted). The locations in which DA has been detected are highly variable with respect to
chemical and physical oceanography, meteorology and nutrient inputs. The ability of multiple
species of Pseudo-nitzschia to produce DA and the inconsistent nature of toxin production
imply that more than one ‘trigger’ may be involved in stimulating DA production, a possibility
that is supported by several laboratory studies. Numerous hypotheses exist regarding the
conditions that can give rise to DA production in Pseudo-nitzschia including biological, chemical
41
and physical aspects of the ecosystem (Lelong et al. 2012, Lewitus et al. 2012). A definitive
cause for blooms of Pseudo-nitzschia spp. and DA production has not been determined,
hindering attempts to predict these events, in spite of the wealth of research on the subject.
Anthropogenic eutrophication and climate change does not currently appear to be the reason
behind Pseudo-nitzschia bloom events and DA production (Lewitus et al. 2012) although those
factors have been found to be responsible for the increased incidence of other HAB events
globally. A general lack of predictability of DA outbreaks remains the primary motivation for
implementing HAB monitoring programs that include Pseudo-nitzschia and DA. Models are
now emerging that provide some basic predictive power, but they are still relatively early in
their development, generally requiring direct measurements of microalgal abundances and
toxin concentrations for parameterization and validation (Anderson et al. 2009, Lane et al.
2009, Anderson et al. 2010). Long-term datasets of microalgal abundance and environmental
parameters are essential for documenting blooms and to extend the understanding of potential
environmental conditions leading to them (Kim et al. 2009, Glibert et al. 2010, Frolov et al.
2012).
The Southern California Coastal Ocean Observing System (SCCOOS) HAB Monitoring
program was initiated in 2008. The program entails weekly samples collected for monitoring
the abundances of potentially harmful microalgal species, particulate DA and nutrient
concentrations at pier locations along the coast of southern California. Sampling is carried out
by investigators at California Polytechnic University, University of California Santa Barbara,
University of California Los Angeles, University of Southern California and the Scripps Institution
of Oceanography, University of California San Diego. Automated sensor packages maintained
42
by SCCOOS provide basic chemical/physical parameters on a continuous basis and daily discrete
samples for temperature and salinity collected as a part of the Scripps Institution of
Oceanography Shore Station Program occurs at several of the SCCOOS HAB monitoring
locations. In this report, information collected for 3.5 years at the pier in the City of Newport
Beach, Orange County (33°36'N, 117°55'W) and for 2 years at the pier in the City of Redondo
Beach, Los Angeles County (33°50'N, 118°23’W) were analyzed in order to establish a baseline
for identifying algal blooms in the region based on easily-collected parameters and to
determine whether this information could be used to specifically identify blooms of Pseudo-
nitzschia and DA events. A statistical analysis of these datasets was conducted with the
objective of improving our understanding of the conditions that give rise to toxic Pseudo-
nitzschia blooms in southern California and to investigate algal bloom definitions. Historically,
algal blooms have been arbitrarily defined, with large variations in definitions between
locations, laboratories and researchers. Most investigations into identifying bloom events have
centered around unusually high chlorophyll a concentrations, focusing on deviations from a
mean chlorophyll concentration specific to the location being studied (Carstensen et al. 2007,
Henson & Thomas 2007, Allen et al. 2008, Kim et al. 2009). These studies have differed in the
time span over which the mean chlorophyll concentration has been determined and the
method of collection of the chlorophyll concentrations (i.e. satellite imagery, in vivo
fluorescence, fluorometric or spectrophotometric analysis of chlorophyll concentration from
discrete water samples filtered onto and extracted from membrane filters).
43
MATERIALS AND METHODS
Sample Collection and Processing. Weekly sampling at Newport Pier in Orange County,
California, began on June 30, 2008, as a part of the SCCOOS HAB Monitoring Program (Fig 2-1).
Surface seawater (2 liters) was collected in acid-washed (5% HCl) polycarbonate bottles every
Monday, kept cool and out of direct sunlight until processing in the laboratory approximately 1-
2 hours after collection. Net tow samples were collected in conjunction with the whole
seawater samples with a 20 µM net (Sea-gear, Melbourne, FL) for relative abundance
determinations of microalgae present. In January 2010, the HAB monitoring effort was
expanded to include the Redondo Beach pier in Los Angeles County, using similar methods (Fig
2-1). Samples for dissolved inorganic nutrients were collected at Newport pier by filtering
approximately 30 mL of water through a 0.2 µm syringe filter using syringes that were pre-
washed with 5% HCl and rinsed three times with sample water. Samples were dispensed into
50 mL polypropylene conical tubes and frozen at -20°C upon arrival at the laboratory until
analysis. Nutrient samples were not collected at the Redondo Beach pier location during the
two year monitoring period. Concentrations of nitrate plus nitrite (0.2 µM limit of detection),
nitrite (0.1 µM limit of detection), ammonium (0.1 µM limit of detection), phosphate (0.1 µM
limit of detection) and silicate (1.0 µM limit of detection) were measured with ± 5% precision
on a QuikChem 8000 flow injection analyzer (Lachat Instruments; Loveland, CO, USA) by the
Analytical Lab at the Marine Sciences Institute at University of California Santa Barbara.
Samples for chlorophyll and particulate DA (pDA) concentrations were collected in duplicate by
vacuum filtration of 100mL and 200mL samples, respectively, onto GF/F Whatman filters.
Samples for determining the concentration of chlorophyll a were extracted in 100% acetone for
44
24 hours at -20°C and analyzed on a calibrated laboratory fluorometer (TD-700; Turner Designs
Inc, Sunnyvale, CA, USA) using the 5% HCl acidification method for the correction for
phaeopigment (Parsons et al. 1984). The use of a 100% acetone extraction with 24 hour
storage at -20°C has been proven to be equally as robust as the traditional 90% acetone
extraction with 24 hour storage in a refrigerator (Caron 2001). Prior to analysis, pDA sample
filters were extracted in 3 mL of 10% methanol, sonicated for 30 seconds and centrifuged for 10
minutes at 4000 rpm. The resulting supernatant was analyzed using the Mercury Science Inc.
DA Enzyme-Linked ImmunoSorbent Assay (ELISA; Durham, NC) following the methods described
in Seubert et al. (2012). The detection limit for the ELISA assay used was 0.02 µg/L.
Figure 2-1. Location of the SCCOOS weekly HAB monitoring piers in southern California, maintained through a collaboration of five university
laboratories, beginning in June 2008. USC maintains the Newport pier sampling location in Orange County, CA. Redondo Beach pier (noted on
the map with a star), located in Los Angeles County, CA, was added to the USC HAB monitoring effort in 2010.
45
Subsamples for characterizing the microalgal community composition were preserved
with 4% formaldehyde and examined by inverted light microscopy at 400x after settling 25 mL
in Utermöhl chambers for 24 hours (Utermohl 1958). Forty fields of view were counted, giving
a limit of detection of 3,000 cells/L. Samples were stored at room temperature in glass bottles,
out of direct sunlight, until enumeration within 1 day to 1 week of collection. As a part of the
SCCOOS HAB Monitoring Program, microalgal community analysis is focused on the
identification of potential HAB formers known to be present in southern California waters.
Organisms specifically identified are the dinoflagellates Akashiwo sanguinea, Alexandrium spp.,
Dinophysis spp., Lingulodinium polyedrum and Prorocentrum spp. and the diatom genus
Pseudo-nitzschia, all other cells in these classes counted were grouped into the categories
‘other dinoflagellate’ or ‘other diatom’. Other HAB organisms known to not preserve well that
can be present, such as Cochlodinium, Heterosigma and Phaeocystis, were recorded in the
relative abundance determinations of live net tows from each location. Conclusive
identification of Pseudo-nitzschia to species is not possible without using molecular methods
(Scholin et al. 1996, Miller & Scholin 1998, Lundholm et al. 2002, Hubbard et al. 2008) or
electron microscopy (Hasle et al. 1996, Hasle & Syvertsen 1997); consequently Pseudo-nitzschia
cells were divided into two size classes, the P. seriata size class with frustule widths greater
than 3 µm, and the P. delicatissimia size class with frustule widths smaller than 3 µm (Hasle &
Syvertsen 1997), which is easily accomplished by light microscopy.
Ancillary Physical Data. Rainfall data was obtained from the University of California, Davis,
Integrated Pest Management Program weather station at the Santa Ana Fire Station (NCDC #
46
7888, 33°45’N 117°52’W) in Orange County, CA, for use with the Newport pier dataset and the
Santa Monica weather station (CIMIS # 99, 34°3’N 118°29’W) in Los Angeles County, CA, for use
with the Redondo Beach pier dataset (www.ipm.ucdavis.edu). Rainfall data was collected as
daily totals measured by an eight inch diameter gauge and the information was binned into
weekly totals. Daily river discharge data for the Santa Ana river was obtained from the US
Geological Survey station #11078000 (33°39’N 117°54’W) and binned into weekly totals for
inclusion in the Newport pier dataset (http://waterdata.usgs.gov/nwis/). The weekly binning
was performed by totaling information from the last sample date until the next sample
collected, the majority of which were seven day totals, with eight day totals needing to be
made when samples were collected on days other than Monday and six day totals when
sampling returned to Monday collection. There is no major river that discharges in close
proximity to the Redondo Beach pier location and therefore river discharge was not included in
the Redondo dataset. Information on water temperature, salinity and chlorophyll fluorescence
was collected at Newport pier by SCCOOS using automated sensors. The Newport pier
automated station contains a Sea-Bird Electronics (Bellevue, WA, USA) 16plus SeaCAT
conductivity and a temperature and pressure meter with measurements collected every four
minutes. Temperature data collected by the Newport pier SCCOOS sensor package was
obtained from the SCCOOS website (www.sccoos.org) for each day of sampling at the Newport
pier. The temperature information from the automated sensor was verified by comparing
sensor data to discrete temperature information manually collected in conjunction with
discrete sample collection. Salinity data from the SCCOOS sensor package was discovered to be
impacted by bio-fouling on the sensor when compared to salinity data collected by the SCCOOS
47
manual shore station program at Newport pier. The manual shore station data was used in
replacement of the flawed salinity sensor data in the regression analyses. The manual shore
station program collects salinity data with the Guildline, model 8410, inductive salinometer and
is available on the University of California San Diego Shore Station website
(ftp://ftp.iod.ucsd.edu/shore/). Sensor data was not available for the Redondo Beach pier but
temperature data was collected weekly simultaneous to discrete sample collection. Salinity
information was obtained from a SCCOOS automated station located within Santa Monica Bay
at the Santa Monica Pier, approximately 20 km north of the Redondo Beach pier.
Statistical Analyses. Simple linear and multiple linear regressions were performed in SigmaPlot
(v. 11.0.0, Systat Software, Inc.) on both the Newport and Redondo Beach pier datasets.
Multiple linear regression on log
10
-tranformed pDA + 1 and P. seriata + 1 concentrations was
carried out using a backward step-wise approach in which independent variables were
iteratively removed from the analysis based upon multi-collinearity (VIF > 4), non-significant
contributions (p > 0.05) and low F-values (< 2). Multiple regressions of the Newport pier
dataset were computed using the independent variables temperature, salinity, Santa Ana river
discharge and rainfall, chlorophyll, nitrate, ammonium, nitrite, phosphate and silicate
concentrations, and ratios of nutrient concentrations, silicate to nitrate, silicate to phosphate
and phosphate to nitrate. All independent variables failed tests for normality; temperature,
salinity, chlorophyll, ammonium, and phosphate were log
10
-transformed while the remaining
variables Santa Ana river discharge, rainfall, nitrate, nitrite, silicate, silicate to nitrate ratio,
silicate to phosphate ratio and the phosphate to nitrate ratio were log
10
-transformed after an
48
addition transform of 1. Multiple regressions of the Redondo Beach pier dataset were
computed using the independent variables temperature, salinity, rainfall and chlorophyll.
Temperature, salinity and chlorophyll were log
10
-transformed and rainfall was log
10
-
transformed following an addition transform of 1. Regressions were computed with no time lag
of environmental variables as well as with one and two week time lagging of environmental
variables, to investigate a possible delay in the biological response to environmental forcing
factors. Chlorophyll concentrations were included within the regression analysis to investigate
a potential relationship between overall microalgal biomass and DA events. The adjusted R
2
values for each regression were used to compare the non-lagged, as well as one and two week
lagged multiple regressions. The adjusted R
2
takes into account samples size and variable
number, weakening the likelihood of R
2
being artificially maximized by the inclusion of non-
significant factors.
RESULTS
Environmental Conditions at Newport and Redondo Beach Piers. Temperatures recorded by
the SCCOOS automated sensor at Newport pier ranged from 12.3 to 23.7°C, with an average
temperature of 16.9 ± 2.33 °C , during the 3.5 year monitoring period (Fig 2-2). A smooth
seasonal trend of warming during the summer months and cooling during the winter was
interrupted by sharp decreases in temperature due most likely to storm events and upwelling in
the region. Salinity ranged from 32.5 to 34.6 psu, with an average salinity of 33.5 ± 0.23 psu
(Fig 2-2). Concentrations of nitrate, phosphate and silicate demonstrated numerous, sporadic
increases, with the highest concentrations often coinciding with decreases in temperature due
49
to the presence of upwelled water (Fig 2-2). Detectable nitrate concentrations ranged from
0.21 to 13 µM, with an overall average concentration of 1.2 ± 2.0 µM. Detectable nitrite
concentrations ranged from 0.10 to 0.96 µM, with an overall average concentration of 0.13 ±
0.16 µM. Ammonium concentrations were always above the detection limit of the method (see
Materials and Methods) and ranged from 0.21 to 31.6 µM, with an overall average
concentration of 3.19 ± 4.31 µM. Phosphate concentrations were always above the detection
limit of the method used (see Materials and Methods) and ranged from 0.11 to 1.8 µM, with an
overall average concentration of 0.32 ± 0.20 µM. Detectable silicate concentrations ranged
from 1.0 to 14 µM, with an average concentration of 3.1 ± 2.4 µM. Temperatures recorded in
the weekly discrete sampling at Redondo Beach pier ranged from 13.2 to 22.0°C, with an
average temperature of 17.2 ± 2.33°C, during the two year monitoring period (Fig 2-3). Salinity
measurements taken from the SCCOOS automated sensor on Santa Monica pier, approximately
20 km north of Redondo Beach pier, ranged from 31.3 to 33.4 psu, with an average
concentration of 32.8 ± 0.47 psu (Fig 2-3).
50
Figure 2-2. Temperature and salinity measured at Newport pier with the SCCOOS automated sensor is plotted from June 30,
2008, to December 19, 2011. Nitrate, phosphate and silicate measured in the discrete samples from the same time period are
also plotted.
51
Minor bloom thresholds for each dataset were established by first removing all
chlorophyll values above the major bloom threshold (to remove the influence of those extreme
values on the mean) and then the new average chlorophyll a concentrations were determined.
The average chlorophyll concentration for Newport pier after the removal of the major bloom
values was 3.88 ± 2.93 µg/L, and a minor bloom threshold of 9.74 µg/L was defined as two
standard deviations above that average (Table 2-1A). The average chlorophyll a concentration
for the Redondo Beach pier dataset after removal of the major bloom values was 2.32 ± 1.67
µg/L, yielding a minor bloom threshold of 5.66 µg/L (Table 2-2A). These treatments of the
datasets enabled identification of four major and twelve minor blooms at the Newport pier site
for a total of 16 bloom events during the 3.5 year dataset (Table 2-1A; Fig 2-4A). Four major
and four minor blooms events were identified at Redondo Beach pier for a total of 8 bloom
samples in the 2 year dataset (Table 2-2A; Fig 2-5A).
52
Figure 2-3. Temperature measured during the discrete sampling at Redondo Beach Pier from January 26, 2010, to December
19, 2011, is plotted in conjunction with the salinity measurements taken with the SCCOOS automated sensor at Santa Monica
Pier during the same period.
Major and minor bloom events defined by anomalously high concentrations of
chlorophyll (Figs 2-4A and 2-5A) were compared to detectable concentrations of domoic acid in
the same samples to examine the potential for identifying toxic Pseudo-nitzschia spp. blooms
based on total microalgal biomass (i.e. chlorophyll concentration). Detectable concentrations
of pDA in the Newport pier dataset occurred during only one major bloom and three minor
bloom events, corresponding to 25% of the samples identified as blooms (Table 2-1A; Fig 2-4B).
Particulate DA was not detected in any of the eight samples identified as blooms in the
Redondo Beach pier dataset based on chlorophyll concentrations (Table 2-2A; Fig 2-5B).
53
Table 2-1. (A) Information on major and minor bloom events identified at Newport pier using chlorophyll a
concentrations to define blooms. (B) Information on major and minor blooms events identified at Newport Pier
defined by abundances of cells in the P. seriata size class from Pseudo-nitzschia cell counts.
(A) Major Blooms Minor Blooms
Threshold 12.9 µg/L 9.74 µg/L
No. Identified 4 12
% Occurrence 2% 7%
No. Co-Occurring with [pDA] 1 3
% with [pDA] 25% 25%
Total Bloom Samples 16
Total % with [pDA] 25%
(B) Major Blooms Minor Blooms
Threshold 88,000 cells/L 40,000 cells/L
No. Identified 8 12
% Occurrence 4% 7%
No. Co-Occurring with [pDA] 5 8
% with [pDA] 63% 67%
Total Bloom Samples 20
Total % with [pDA] 65%
54
Figure 2-4. (A) Chlorophyll a concentrations measured in the weekly samples collected at Newport pier from June 30, 2008, to
December 19, 2011. The major bloom threshold chlorophyll concentration of 12.9 µg/L and minor bloom threshold of 9.74
µg/L are shown by the dashed and dotted lines, respectively. The four major blooms are marked by the grey stars and the
twelve minor blooms are marked by the white stars. (B) All samples that were not identified as major or minor blooms by the
chlorophyll concentration definition were removed, and only bloom sample chlorophyll concentrations are plotted with the
black circles. The major and minor bloom thresholds are shown by the dashed and dotted lines, respectively. pDA
concentrations measured in any of the chlorophyll defined major and minor are plotted with the light grey triangles, pDA
samples below the detection limit were not included.
55
Figure 2-5. (A) Chlorophyll a concentrations measured in the weekly samples collected at Redondo Beach pier from January 26,
2010, to December 19, 2011. The major bloom threshold chlorophyll concentration of 9.13 µg/L and minor bloom threshold of
5.66 µg/L are shown by the dashed and dotted lines, respectively. The four major blooms are marked by the grey stars and the
four minor blooms are marked by the white stars. (B) All samples that were not identified as major or minor blooms by the
chlorophyll concentrations definition were removed and only bloom sample chlorophyll concentrations are plotted. The major
and minor bloom thresholds are shown by the dashed and dotted line, respectively. Domoic acid was not measured in any of
the chlorophyll bloom samples from Redondo Beach Pier.
56
Table 2-2. (A) Information on major and minor bloom events identified at Redondo Beach pier using chlorophyll a
concentrations to define blooms. (B) Information on major and minor bloom events identified at Redondo Beach pier based
upon the abundance of P. seriata size class cells from Pseudo-nitzschia cell counts.
(A) Major Blooms Minor Blooms
Threshold 9.13 µg/L 5.66 µg/L
No. Identified 4 4
% Occurrence 4% 4%
No. Co-Occurring with [pDA] 0 0
% with [pDA] 0% 0%
Total Bloom Samples 8
Total % with [pDA] 0%
(B) Major Blooms Minor Blooms
Threshold 110,000 cells/L 56,000 cells/L
No. Identified 4 7
% Occurrence 4% 7%
No. Co-Occurring with [pDA] 2 2
% with [pDA] 50% 29%
Total Bloom Samples 11
Total % with [pDA] 36%
Establishing Bloom Conditions Based on Pseudo-nitzschia Cell Abundances. Many HAB species
do not need to attain exceedingly high cell abundances or dominate total microalgal biomass in
order to cause ecosystem damage and present a human health risk. Thus, it was not surprising
to find that a chlorophyll-based definition for blooms at our coastal sites poorly identified
situations in which pDA production by Pseudo-nitzschia spp. was detectable. Correlations with
cell abundances of Pseudo-nitzschia were investigated as an alternative means of identifying
potential emerging DA events.
57
Pseudo-nitzschia spp. abundances were divided into two size classes in the cell count
data collected for the SCCOOS HAB Monitoring Program; a P. seriata size class and a P.
delicatissima size class (see Methods and Materials). The P. seriata size class contains species
of Pseudo-nitzschia that are known to be capable of DA production. Documented toxic events
in southern California have been most often attributed to P. australis and P. multiseries, both
members of the P. seriata size class (Anderson et al. 2006, Busse et al. 2006, Schnetzer et al.
2007, Schnetzer et al. Accepted). The P. delicatissima size class contains some species capable
of toxin production, however blooms of this size class are rarely associated with toxic events
(Smith et al. 1991, Adams et al. 2000, Stehr et al. 2002, Orsini et al. 2004).
Cellular abundance data collected for the P. seriata size class at the Newport and
Redondo Beach piers were used to establish major and minor Pseudo-nitzschia bloom events in
a manner analogous to that used with the chlorophyll a concentrations, described above. Long-
term means of P. seriata size class cell abundance data were calculated for each location and
major bloom thresholds were established based on two standard deviations from the overall
mean values for each sampling site. A long-term mean for the Newport pier samples was
calculated as 14,000 ± 37,000 cells/L, yielding a major bloom threshold of 88,000 cells/L of the
P. seriata group (Table 2-1B). After removal of the major blooms values, a new long term mean
was determined to be 7,100 ± 16,000 cells/L, creating a minor bloom threshold of 40,000 cells/L
of the P. seriata group. The long-term mean for the Redondo Beach pier samples was
determined to be 19,000 ± 46,000 cells/L, yielding a major bloom threshold of 110,000 cells/L
of the P. seriata group (Table 2-2B). The long term mean for these samples following removal
58
of the major bloom samples was 11,000 ± 22,500 cells/L, yielding a minor bloom threshold of
56,000 cells/L.
These treatments of the cell abundance data identified eight major and twelve minor
blooms of the P. seriata size class at the Newport pier sampling site for a total of twenty bloom
samples (Table 2-1B; Fig 2-6A), and four major and seven minor blooms at the Redondo Beach
pier sampling site for a total of eleven bloom samples (Table 2-2B; Figure 2-7A). The bloom
events identified using the abundances of the P. seriata size class were compared to pDA
concentrations at each location (Figs 2-6B and 2-7B). Thirteen of the Newport pier samples
identified as blooms corresponded to samples with detectable pDA concentrations (Table 2-1B),
while four of the eleven Redondo Beach pier samples identified as blooms had detectable pDA
concentrations (Table 2-2B). Bloom events at the Newport pier site, defined using the
abundances of the P. seriata size class, corresponded to detectable pDA concentrations in 65%
of the samples (Table 2-1B) compared to 25% correspondence for a bloom definition based on
anomalously high chlorophyll concentrations (Table 2-1A). Measureable pDA concentrations
occurred in 29 of the 178 total samples (16%) collected at Newport pier and 13 of the 29 (45%)
were detected during blooms of the P. seriatia size class. The samples with detectable pDA that
were not identified as blooms ranged from 0.060 to 0.34 µg/L, with an average concentration of
0.075 ± 0.073 µg/L, while the samples corresponding to identified blooms had pDA
concentrations that ranged from 0.046 to 3.0 µg/L, with an average concentration of 0.35 ±
0.82 µg/L.
59
Figure 2-6. (A) The concentration of Pseudo-nitzschia seriata size class cells measured in the Newport pier weekly samples from
June 30, 2008 to December 19, 2011 are plotted in the top panel figure. The major bloom threshold concentration of 88,000
cells/L and minor bloom threshold of 40,000 cells/L are plotted as the dashed and dotted lines, respectively. The eight major
blooms are marked by the grey stars and the twelve minor blooms are marked by the white stars. (B) Only the identified
blooms by the P. seriata size class definition, major and minor, are plotted in the bottom panel with black circles and the
associated pDA concentrations measured in the same sample are plotted as the grey triangles, pDA samples below the
detection limit were not included. The major and minor bloom thresholds are plotted as the dashed and dotted lines,
respectively.
60
Figure 2-7. (A) The concentration of Pseudo-nitzschia seriata size class cells measured in the Redondo Beach pier weekly
samples from January 26, 1010, to December 19, 2011, are plotted. The major bloom threshold concentration of 110,000
cells/L and minor bloom threshold of 56,000 cells/L are plotted as the dashed and dotted line, respectively. The four major
blooms are marked by the grey stars and the minor blooms are marked by the white stars. (B) Only the identified blooms by
the P. seriata size class definition, major and minor, are plotted as the black circles with the associated domoic acid
concentrations measured in the same sample plotted as the grey triangles, pDA samples below the detection limit were not
included. The major and minor bloom thresholds are plotted as the dashed and dotted lines, respectively.
Bloom events at the Redondo Beach pier site, defined using the abundances of the P.
seriata size class, corresponded to detectable pDA concentrations in 36% of the bloom samples
(Table 2-2B) compared to 0% correspondence for a bloom definition based upon anomalously
61
high chlorophyll concentrations (Table 2A). Measurable pDA concentrations occurred in 23 of
the total 98 samples (23%) collected at the Redondo Beach sampling site, and 4 of the 23 (17%)
were detected during blooms of the P. seriata size class. The samples with detectable pDA that
were not identified as blooms ranged in concentration from 0.048 to 0.30 µg/L, with an average
concentration of 0.082 ± 0.069 µg/L, and the bloom associated pDA concentrations range from
0.041 to 0.57 µg/L, with an average concentration of 0.21 ± 0.25 µg/L.
Correlating Pseudo-nitzschia Blooms and DA Events to Environmental Variables. Multiple and
simple linear regression analyses were carried out on each of the complete datasets from
Newport and Redondo Beach piers and then compared to regression analysis on a subset of
samples defined as blooms of the P. seriata size class at each location. The comparison of the
regression results was performed to discern if the identification of bloom events using the
abundances of Pseudo-nitzschia in the P. seriata size class improved the specificity of statistical
analysis by eliminating samples that were unrelated to Pseudo-nitzschia or toxic events.
The first set of regression analyses on the entire 3.5 year Newport pier dataset were
undertaken to identify significant relationships between ammonium, chlorophyll a, nitrate,
nitrite, phosphate and silicate concentrations, ratios of nutrient concentrations for silicate to
nitrate, silicate to phosphate and phosphate to nitrate, Santa Ana river discharge, rainfall,
temperature and salinity to pDA and P. seriata size class concentrations (Table 2-3). The most
significant relationships identified on the dataset without time lagging were obtained with
multiple linear regressions. Salinity, chlorophyll a concentrations and the ratio of silicate to
phosphate concentrations were significantly correlated, albeit weakly, with pDA concentrations
62
(R
2
adj = 0.0844, p < 0.001). The salinity and chlorophyll values were positively correlated with
pDA and silicate to phosphate ratios were negatively correlated with pDA. A significant and
slightly stronger relationship was identified between temperature, chlorophyll and nitrite
concentrations, and ratios of silicate to nitrate with P. seriata size class concentrations (R
2
adj =
0.193, p < 0.001). Temperature, nitrite concentration and the ratio of silicate to nitrate
concentrations were negatively correlated with P. seriata whereas chlorophyll was positively
correlated. A one week time lag of environmental data revealed a significant relationship
between temperature, salinity, chlorophyll concentration and the ratio of silicate to nitrate with
pDA concentrations (R
2
adj = 0.0676, p = 0.007). Temperature was negatively correlated to pDA
while salinity, chlorophyll and the ratio of silicate to nitrate were positively correlated with
pDA. The same parameters were significantly correlated with P. seriata (R
2
adj = 0.171, p <
0.001), temperature was negatively correlated and salinity, chlorophyll and the ratio of silicate
to nitrate were positively correlated with P. seriata. A two week time lag of environmental data
revealed a significant positive correlation between chlorophyll and the silicate to nitrate ratio
with pDA concentrations (R
2
adj = 0.0559, p = 0.005). Temperature was negatively correlated
with P. seriata and the silicate to nitrate ratio was positively correlated with P. seriata with a
two week time lag (R
2
adj = 0.615, p = 0.001).
63
Table 2-3. Results of multiple regression analysis of the Newport pier dataset. Prior to analysis, samples were identified as
blooms based upon the P. seriata size class abundance definition and all non-bloom values were removed. Negative
correlations are identified with a minus sign (-).
Variable Time Lag R
2
adj F P Significant Variables
pDA
Concentration
None 0.125 3.707 0.07 - Silicate : Phosphate
1 week 0.537 6.228 0.004
Chlorophyll, Salinity, Silicate : Nitrate, -
Temperature
2 week 0.113 2.078 0.16 Salinity, Chlorophyll
P. seriata
Abundance
None 0.613 8.904 .002
- Ammonium, - Phosphate : Nitrate,
-Temperature
1 week 0.242 2.913 0.069
- Ammonium, - Phosphate : Nitrate,
- Salinity
2 week 0.653 8.997 0.001 Rainfall, - Salinity, Silicate, Temperature
Bloom events identified by anomalously high abundances of the P. seriata size class
were then used to select a subset of data for analysis. Samples not identified as blooms were
removed in an effort to improve specificity of the statistical analysis. All identified major (8)
and minor (12) blooms of the P. seriata size class were used in regression analyses investigating
relationships between ammonium, chlorophyll, nitrate, phosphate and silicate concentrations,
ratios of nutrient concentrations silicate to nitrate, silicate to phosphate and phosphate to
nitrate, Santa Ana river discharge, rainfall, temperature and salinity to pDA concentrations and
P. seriata size class abundances. The most significant relationship identified for pDA
concentrations was with one week lagged temperature, salinity, chlorophyll and silicate to
nitrate ratio (R
2
adj = 0.537, p = 0.004). Temperature was negatively correlated while salinity,
chlorophyll and silicate to nitrate ratio were positively correlated with pDA. Salinity and
chlorophyll concentrations were also positively correlated with pDA concentrations with a two
week lag (R
2
adj = 0.113, p = 0.16), although the relationship was less significant than with a one
64
week time lag. Without time lagging of the data, the most significant relationship identified
was with a simple linear regression and a negative correlation between the silicate to
phosphate ratio and pDA concentrations (R
2
adj = 0.125, p = 0.07). The most significant
relationship identified in linear regressions with P. seriata size class abundances was with two
week time lagged temperature, salinity, rainfall and silicate concentration information (R
2
adj =
0.653, p = 0.001). Salinity was negatively correlated while temperature, rainfall and silicate
concentrations were positively correlated with P. seriata abundances.
The initial regression analyses performed with the entire two year Redondo Beach pier
dataset investigated relationships between chlorophyll concentrations, rainfall, temperature
and salinity to pDA concentrations. Regressions failed to produce statistical significance
between any variable (without time lagging) and pDA concentrations, as demonstrated by high
p values (>0.05), low F values (<2) and low power values (Table 2-4). F values improved using a
one week and two week time lag of variables, but p values remained insignificant (>0.05).
Regressions with the same variables to abundance of the P. seriata size class revealed a
significant positive relationship with chlorophyll concentration (R
2
adj = 0.0386, p = 0.032),
although the relationship had very weak predictive power.
65
Table 2-4. Results of multiple regression analysis of the Redondo Beach pier dataset. Prior to analysis, samples were identified
as blooms based upon the P. seriata size class abundance definition and all non-bloom values were removed. Negative
correlations are identified with a minus sign (-). Asterisks denote failed regressions without statistical significance, low F values
(<2) and low power values.
Variable Time Lag R
2
adj F P Significant Variables
pDA
Concentration
None ** ** ** **
1 week 0.124 2.416 0.155 - Chlorophyll
2 week 0.293 5.144 0.050 - Chlorophyll
P. seriata
Abundance
None ** ** ** **
1 week 0.358 6.581 0.03 Salinity
2 week ** ** ** **
Samples from the Redondo Beach pier dataset meeting the criteria of a bloom event
defined by the abundances of Pseudo-nitzschia in the P. seriata size class were also examined
for relationships to environmental variables. Multiple regressions were performed using the
subset including major (4) and minor (7) blooms to investigate relationships between pDA
concentrations and chlorophyll concentrations, rainfall, temperature and salinity (Table 2-4).
Negative correlations between chlorophyll concentration, lagged one week and two weeks
prior to pDA concentrations, were identified but lacked significance (p = 0.155 and 0.050,
respectively). Regressions were repeated with the same environmental variables to investigate
relationships with abundances of Pseudo-nitzschia in the P. seriata size class. The only
significant relationship was identified from a simple linear regression between the Pseudo-
nitzschia abundances and salinity values lagged one week (R
2
adj = 0.358, p = 0.03).
DISCUSSION
The goals of the present study were: (1) to employ a long-term, weekly dataset of
microalgal biomass, community composition and pDA concentrations to establish site-specific,
66
quantitative metrics to identify emerging blooms of potentially toxic species of the diatom
genus Pseudo-nitzschia and DA events resulting from these species; (2) to establish the
relationship between blooms and DA events with the intent of identifying easily-acquired
measurements that might indicate an emerging HAB event; and (3) to provide an investigative
approach for relating Pseudo-nitzschia blooms and DA events to chemical and physical
environmental parameters co-occurring or preceding these events.
Defining Blooms and Their Relation to DA Events. The ability to rapidly identify an emerging
HAB event is a prerequisite for preventing human exposure and for taking steps that might
minimize or mitigate potential ecological impacts. Unfortunately, microalgal blooms have been
rather arbitrarily defined, or have been defined based on the particular harmful algal species
involved. Microalgal blooms have generally been defined based on total algal biomass, often
using chlorophyll concentration (either extracted chlorophyll analyses or in vivo chlorophyll
fluorescence) as a proxy for algal biomass. Only a few studies have taken a rigorous,
quantitative approach to defining blooms using this parameter (Carstensen et al. 2007, Henson
& Thomas 2007, Allen et al. 2008, Kim et al. 2009). Moreover, only some HAB species dominate
the microalgal assemblage so strongly that total microalgal biomass is a reasonable proxy for
their abundance. Species in this latter category include massive blooms of the nuisance brown
tide algae, Aureococcus anophagefferens and Aureoumbra lagunensis, the dinoflagellates
Lingulodinium polyedrum, or the toxic flagellate Prymnesium parvum that can constitute most
of the algal pigment present in water samples collected during blooms of these species (Allen
1938, Omand et al. 2011, Roelke et al. 2011, Gobler & Sunda 2012). Blooms of A.
67
anophagefferans, for example, have been defined based on cell abundances, indicating
abundances at which this species dominates the pelagic food web and poses the potential for
ecological damage (Gastrich & Wazniak 2002). In contrast, species of the dinoflagellates genus
Alexandrium that produce paralytic shellfish toxins can constitute a significant human health
risk even at relatively low abundances (i.e. a small percentage of the total biomass of the
microalgal community in a sample).
Defining phytoplankton blooms based on anomalously high chlorophyll concentrations
(two standard deviations above a long-term mean) indicated the presence of numerous major
or minor bloom events over the 3.5 or 2 year period of observation at the Newport pier or
Redondo Beach pier, respectively (Tables 2-1A and 2-2A; Figs 2-4A and 2-5A). While these
events indicate significant increases in microalgal biomass at these sites and times, blooms
defined in this manner were very poor predictors of detectable concentrations of pDA (Tables
2-1B and 2-2B; Figs 2-4B and 2-5B). Blooms defined on anomalously high abundances of the P.
seriata size class of Pseudo-nitzschia spp. also allowed for the identification of numerous bloom
events, with blooms defined in this manner yielding greater correspondence to samples that
also exhibited detectable concentrations of pDA (Tables 2-1B and 2-2B; Figs 2-6B and 2-7B).
Documented DA events during blooms of Pseudo-nitzschia spp. in the southern California
region have been attributed to members of the P. seriata size class, most commonly to P.
multiseries and P. australis (Anderson et al. 2006, Schnetzer et al. 2007, Lewitus et al. 2012,
Schnetzer et al. Accepted). The history of the P. seriata size class involvement in DA events in
our region was the motivation for focusing on this size class for the basis of an emerging event
definition.
68
Developing a simple P. seriata abundance-based metric to establish the presence of a
potentially emerging Pseudo-nitzschia bloom and DA event would reduce the time and cost that
is inherent in more sophisticated, albeit more informative, approaches. Technology is only now
becoming available for in situ measurements of DA. The Environmental Sample Processor (ESP)
is capable of quantifying pDA concentrations in situ, and is also capable of identifying a suite of
Pseudo-nitzschia species that might be the source of the toxin (Greenfield et al. 2006,
Greenfield et al. 2008, Doucette et al. 2009). However, the cost and maintenance logistics of
the instrument are presently prohibitive for most scientific studies, routine monitoring and
regulatory programs. Alternatively, direct toxin measurements performed on microalgal
samples returned to a laboratory can also provide direct confirmation of the presence and
quantities of pDA in samples, but toxin analysis by ELISAs, high performance liquid
chromatography or liquid chromatography/mass spectrometry require fairly extensive sample
preparation and thus significant time between sample collection and the availability of the
data. Emerging molecular approaches, such as microarray-based technologies, for quantifying
HAB species will provide speed and power in time and our approach will provide some
bounding on where and when to sample intensively for their application (Ahn et al. 2006,
Gescher et al. 2008, Kegel et al. 2012).
Until these more-insightful methods and instruments become easily accessible and less
costly, the ability to infer an emerging DA event based upon unusual increases in the
abundance of the P. seriata size class of Pseudo-nitzschia cells provides a crude, yet useful and
relatively rapid assessment of developing events that can applied to large sample sets.
Moreover, continued observations and documentation of blooms of Pseudo-nitzschia at
69
Newport and Redondo Beach piers and other coastal sites along the California coast will
augment the datasets that are presently available in this region, allowing for more accurate
differentiation of toxic and non-toxic events. Defining a site-specific Pseudo-nitzschia bloom
metric was also motivated by a desire to improve our understanding of the environmental
forcing factors that give rise to these events in our region. Past studies of Pseudo-nitzschia
blooms in California have typically used a definition of 10,000 cells/L of either size class as a
bloom threshold for species in this genus (Fehling et al. 2006, Howard et al. 2007, Jester et al.
2009b, Lane et al. 2009). Based on this definition, Pseudo-nitzschia attained bloom abundances
in 65 of the 178 total samples collected at Newport pier. Pseudo-nitzschia cells were observed
in a total of 87 of the 178 samples, and therefore 75% percent of the samples with Pseudo-
nitzschia collected at Newport pier during the 3.5 year monitoring period would be considered
bloom samples using the non-site specific bloom metric. This number was far greater than the
number of samples exhibiting detectable pDA concentrations. Applying the same 10,000 cells/L
bloom definition to P. seriata size class abundances only instead of the abundance of all
Pseudo-nitzschia cells would still identify 38 of the 53 samples in which P. seriata was observed
as bloom samples (72%). Basing a bloom definition on a long-term mean of the abundances of
the P. seriata size class of Pseudo-nitzschia reduced the number of samples that potentially
contained detectable pDA concentrations, and thereby improved our ability to investigate
relationships between emerging toxic events and environmental forcing factors.
Correlating Bloom or Toxic Events to Environmental Forcing Factors. The most significant
relationship observed in the regression analysis of the datasets from either pier and pDA
70
concentrations from identified bloom samples occurred at Newport pier with a one week time
lag of chlorophyll, salinity, silicate to nitrate and temperature (R
2
adj = 0.537, p = 0.004, Table 3).
Chlorophyll, salinity and the silicate to nitrate ratio were positively correlated while
temperature was negatively correlated with pDA concentrations. The positive correlation with
chlorophyll concentrations observed in the preceding week may indicate that the detection of
pDA during bloom events follows periods that are conducive to general microalgal growth. This
result is consistent with experimental studies that have observed a link between DA production
by Pseudo-nitzschia species and stationary growth (Bates 1998). A positive correlation with
salinity and negative correlation with temperature suggests a correlation between pDA
production and upwelling events as water brought to the surface in an upwelling event will be
colder, slightly more saline and rich in nutrients. Recent research in the nearshore
environment of southern California has shown that upwelling can stimulate Pseudo-nitzschia
population growth specifically, eventually stimulating pDA production (Schnetzer et al.
Accepted). Silicate to nitrogen ratios for marine diatoms have been shown to be 1 Si: 1 N but
may fluctuate according to species due to differences in the silica content of their frustule. The
positive correlation to high silicate to nitrate ratios the week prior to detectable pDA
concentrations may signify an excess of silicate present for the stimulation of diatom growth,
although Pseudo-nitzschia has been shown to be lightly silicified and capable of prospering in
low silicate conditions (Fehling et al. 2006).
In the regression analysis with P. seriata abundances as the dependent variable there
was a positive correlation between silicate concentrations two weeks prior to identified P.
seriata blooms in conjunction with positive correlations to rainfall and temperature and a
71
negative correlation to salinity at Newport pier (R
2
adj = 0.653, p = 0.001, Table 2-3). The switch
between a negative correlation with temperature for pDA concentrations to a positive
correlation with temperature for P. seriata abundances may demonstrate the difference in
environmental parameters that stimulate Pseudo-nitzschia growth versus parameters that
stimulate pDA production directly. Application of the P. seriata size class bloom identification
metric to larger datasets collected in the region will allow researchers to separate events that
have occurred in which Pseudo-nitzschia growth has been stimulated without inciting pDA
production for comparison to events in which both growth and pDA production has been
promoted. The information gleaned from the statistical relationships identified here can direct
future laboratory experimentation and field sampling efforts as we try to further elucidate the
conditions that stimulate pDA production in southern California.
CONCLUSION
The identification, and eventually prediction, of emerging toxic bloom events of Pseudo-
nitzschia along the coast of southern California is imperative given the recent documentation of
increases in frequency and severity of the occurrence of DA in these waters and poisoning
events resulting from this neurotoxin (Lewitus et al. 2012). An empirically defined, site-specific
bloom definition using Pseudo-nitzschia cell abundance information was investigated in this
study to provide researchers with a relatively rapid and inexpensive tool for identifying the
onset of potentially toxic events. This information constitutes an initial screening tool with
which to guide more intensive or expensive observations and measurements in the future as
well as be applied to larger datasets collected in the region allowing for deeper analysis into the
72
causation of historical Pseudo-nitzschia events. The differences in results obtained in the
regression analysis for Newport and Redondo Beach piers demonstrated that while bloom
events could still be identified with the P. seriata abundance based metric, the specificity of
statistical relationships identified were weak at the Redondo Beach location at which only basic
microalgal community information was collected. We suggest for monitoring programs that
include objectives for the understanding of Pseudo-nitzschia events nutrient and reliable
salinity information should be collected.
REFERENCES
Adams N, Lesoing M, Trainer V (2000) Environmental conditions associated with domoic acid in
razor clams on the Washington coast. J Shellfish Res 19: 1007-1015
Ahn S, Kulis D, Erdner D, Anderson D, Walt D (2006) Fiber-optic microarray for simultaneous
detection of multiple harmful algal bloom species. Appl Environ Microbiol 72: 5742-
5749
Allen J, Smyth T, Siddorn J, Holt M (2008) How well can we forecast high biomass algal bloom
events in a eutrophic coastal sea? Harmful Algae 8:70-76
Allen W (1934) Marine plankton diatoms of lower California in 1931. Bot Gaz 95:485-492
Allen W (1936) Occurrence of marine plankton diatoms in a ten-year series of daily catches in
southern California. Am J Bot 23:60-63
Allen W (1938) "Red Water" along the west coast of the United States in 1938. Science 88:55-56
Allen W (1946) "Red water" in La Jolla Bay in 1945. T Am Microsc Soc 65:149-153
Anderson C, Brzezinski M, Washburn L, Kudela R (2006) Circulation and environmental
conditions during a toxigenic Pseudo-nitzscia australis bloom in the Santa Barbara
Channel, California. Mar Ecol Prog Ser 327:119-133
Anderson C, Sapiano M, Prasad M, Long W, Tango P, Brown C, Murtugudde R (2010) Predicting
potentially toxigenic Pseudo-nitzschia blooms in the Chesapeake Bay. J Mar Sys 83:127-
140
73
Anderson C, Siegel D, Kudela R, Brzezinski M (2009) Empirical models of toxigenic Pseudo-
nitzschia blooms: Potential use as a remote detection tool in the Santa Barbara
Channel. Harmful Algae 8:478-492
Anderson D, Cembella A, Hallegraeff G (2012) Progress in understanding harmful algal blooms:
Paradigm shifts and new technologies for research, monitoring and management. Annu
Rev Mar Sci 4:143-176
Anderson D, Glibert P, Burkholder J (2002) Harmful algal blooms and eutrophication: Nutrient
sources, composition and consequences. Estuaries 25:704-726
Bates S (1998) Ecophysiology and metabolism of ASP toxin production. In: Anderson D,
Cembella A, Hallegraeff G (eds) Physiological ecology of hamrful algal blooms. Springer-
Verlag, Heidelberg
Bates S (2000) Domoic-acid-producing diatoms: Another genus added! J Phycol 36:978-985
Bates S, Bird C, de Freitas A, Foxall R, Gilgan M, Hanic L, Johnson G, McCulloch A, Odense P,
Pocklington R, Quilliam M, Sim P, Smith J, Subba Rao D, Todd E, Walter J, Wright J
(1989) Pennate diatom Nitzschia pungens as the primary source of domoic acid, a toxin
in shellfish from eastern Prince Edward Island, Canada. Can J Fish Aquat Sci 46:1203-
1215
Busse L, Venrick E, Antrobus R, Miller P, Vigilant V, Silver M, Mengelt C, Mydlarz L, Prezelin B
(2006) Domoic acid in phytoplankton and fish in San Diego, CA, USA. Harmful Algae
5:91-101
Caron D (2001) Protistan Herbivory and Bacterivory. In: Paul J (ed) Methods in Microbiology,
Book 30. Academic Press, Ltd, St. Petersburg, FL
Caron D, Garneau M-E, Seubert E, Howard M, Darjany L, Schnetzer A, Cetinic I, Filteau G, Lauri
P, Jones B, Trussell S (2010) Harmful algae and their potential impacs on desalination
operations off southern California. Water Res 44:385-416
Carstensen J, Henriksen P, Heiskanen A (2007) Summer algal blooms in shallow estuaries:
definition, mechanisms and link to eutrophication. Limnol Oceanogr 52:370-384
de la Riva G, Johnson C, Gulland F, Langlois G, Hayning J, Rowles T, Mazet J (2009) Association
of an unusual marine mammal mortality event with Pseudo-nitzschia spp. blooms
along the southern California coastline. J Wildl Dis 45:109-121
74
Doucette G, Mikulski C, Jones K, King K, Greenfield D, Marin III R, Jensen S, Roman B, Elliot C,
Scholin C (2009) Remote, subsurface detection of the algal toxin domoic acid onboard
the Environmental Sample Processor: Assay development and field trials. Harmful
Algae 8:880-888
Fehling J, Davidson K, Bolch C, Tett P (2006) Seasonality of Pseudo-nitzschia spp.
(Bacillariophyceae) in western Scottish waters. Mar Ecol Prog Ser 323:91-105
Fire S, Wang Z, Berman M, Langlois G, Morton S, Sekula-Wood E, Benitez-Nelson C (2010)
Trophic transfer of the harmful algal toxin domoic acid as a cause of death in a Minke
Whale (Balaenoptera acutorostrata) stranding in southern California. Aquatic Mammals
36:342-350
Fire S, Wang Z, Leighfield T, Morton S, McFee W, McLellan W, Litaker R, Tester P, Hohn A,
Lovewell G, Harms C, Rotstein D, Barco S, Costidis A, Sheppard B, Bossart G, Stolen M,
Durden W, Van Dolah F (2009) Domoic acid exposure in pygmy and dwarf sperm whales
(Kogia spp.) from southeastern and mid-Atlantic US waters. Harmful Algae 8:658-664
Fritz L, Quilliam M, Wright J, Beale A, Work T (1992) An outbreak of domoic acid poisoning
attributed to the pennate diatom Pseudo-nitzschia australis. J Phycol 28:439-442
Frolov S, Kudela R, Bellingham J (2012) Monitoring of harmful algal blooms in the era of
diminishing resources: A case study of the U.S. west coast. Harmful Algae
http://dx.doi.org/10.1016/j.hal.2012.11.001
Fryxell G, Villace M, Shapiro L (1997) The occurence of the toxic diatom genus Pseudo-nitzschia
(Bacillariophyceae) on the West Coast of the USA, 1920-1996: a review. Phycologia
36:419-437
Garneau M-E, Schnetzer A, Countway P, Jones A, Seubert E, Caron D (2011) Examination of the
seasonal dynamics of the toxic dinoflagellate Alexandrium catenella at Redondo Beach,
California, by quantitative PCR. Appl Environ Microbiol 77:7669-7680
Gastrich M, Wazniak C (2002) A brown tide bloom index based on the potential harmful effects
of the brown tide alga, Aureococcus anophagefferens. Aquat Ecosyst Health 5:435-441
Gescher C, Metfies K, Medlin L (2008) The ALEX CHIP - Development of a DNA chip for
identification and monitoring of Alexandrium. Harmful Algae 7
Glibert P, Allen J, Bouwman A, Brown C, Flynn K, Lewitus A, Madden C (2010) Modeling of HABs
and eutrophication: Status, advances, challenges. J Marine Syst 83:262-275
75
Glibert P, Seitzinger S, Heil C, Burkholder J, Parrow M, Codispoti L, Kelly V (2005) The role of
eutrophication in the global proliferation of hamrful algal blooms. Oceanography
18:198-209
Gobler C, Sunda W (2012) Ecosystem disruptive algal blooms of the brown tide species,
Aureococcus anophagefferens and Aureoumbra lagunensis. Harmful Algae 14:36-45
Greenfield D, Marin III R, Doucette G, Mikulski C, Jones K, Jensen S, Roman B, Alvarado N,
Feldman J, Scholin C (2008) Field applications of the second-generation Environmental
Sample Processor (ESP) for remote detection of harmful algae: 2006-1007. Limnol
Oceanogr: Methods 6:667-679
Greenfield D, Marin III R, Jensen S, Massion E, Roman B, Feldman J, Scholin C (2006) Application
of environmental sample processor (ESP) methodology for quantifying Pseudo-nitzschia
australis using ribosomal RNA-targeted probes in sandwich and fluorescent in situ
hybridization formats. Limnol Oceanogr: Methods 4:426-435
Gregorio D, Pieper R (2000) Investigations of red tides along the southern California coast. Bull
South Calif Acad Sci 99:147-160
Hall A, Frame E (2010) Evidence of domoic acid exposure in harbour seals from Scotland: A
potential factor in the decline in abundance. Harmful Algae 9:489-493
Hasle G, Lange C, Syvertsen E (1996) A review of Pseudo-nitzschia, with special reference to the
Skagerrak, North Atlantic, and adjacent waters. Helgolander Meeresun 50:131-175
Hasle G, Syvertsen E (1997) Marine Diatoms. In: Carmelo T (ed) Identifying Marine
Phytoplankton. Academic Press, San Diego
Heisler J, Glibert P, Burkholder J, Anderson D, Cochlan W, Dennison W, Dortch Q, Gobler C, Heil
C, Humphries E, Lewitus A, Magniun R, Marshall H, Sellner K, Stockwell D, Stoecker D,
Suddleson M (2008) Eutrophication and harmful algal blooms: a scientific consensus.
Harmful Algae 8:3-13
Henson S, Thomas A (2007) Interannual variability in timing of bloom initiation in the California
Current System. J Geophys Res 112:C080007
Howard M, Cochlan W, Ladizinsky N, Kudela R (2007) Nitrogenous preference of toxigenic
Pseudo-nitzschia australis (Bacillariophyceae) from field and laboratory experiments.
Harmful Algae 6:206-217
Howard M, Jones A, Schnetzer A, Countway P, Tomas C, Kudela R, Hayashi K, Chia P, Caron D
(2012) Quantitative real-time polymerase chain reaction for Cochlodinium fulvescens
76
(Dinophyceae), a harmful dinoflagellate from California coastal waters. J Phycol 48:384-
393
Howard M, Silver M, Kudela R (2008) Yessotoxin detected in mussel (Mytilus californicus) and
phyotplankton samples from the US west coast. Harmful Algae 7:646-652
Hubbard K, Rocap G, Armbrust E (2008) Inter- and intraspecific community structure within the
diatom genus Pseudo-nitzschia (Bacillariophyceae). J Phycol 44:637-649
Jessup D, Miller M, Ryan J, Nevins H, Kerkering H, Mekebri A, Crane D, Johnson T, Kudela R
(2009) Mass stranding of marine birds caused by a surfactant-producing red tide. PLoS
One 4:e4550. doi:4510.1371/journal.pone.0004550
Jester R, Lefebvre K, Langlois G, Vigilant V, Baugh K, Silver M (2009) A shift in the dominant
toxin-producing algal species in central California alters phycotoxins in food webs.
Harmful Algae 8:291-298
Kegel J, Del Amo Y, Medlin L (2012) Introduction to project MIDTAL: its methods and samples
from Arcachon Bay, France. Environ Sci Pollut Res DOI 10.1007/s11356-012-1299-9
Kim H, Miller A, McGowan J, Carter M (2009) Coastal phytoplankton blooms in the southern
California bight. Prog Oceanogr 82:137-147
Kotaki Y, Lundholm N, Onodera H, Kobayashi K, Bajarias F, Furio E, Iwataki M, Fukuyo Y, Kodama
M (2004) Wide distribution of Nitzschia navis-varingica, a new domoic acid-producing
benthic diatom found in Vietnam. Fisheries Sci 70:28-32
Kudela R, Lane J, Cochlan W (2008) The potential role of anthropogenically derived nitrogen in
the growth of harmful algae in California, USA. Harmful Algae 8:103-110
Kvitek R, Goldberg J, Smith G, Doucette G, Silver M (2008) Domoic acid contamination within
eight representative species from the bethic food web of Monterey Bay, California,
USA. Mar Ecol Prog Ser 367:35-47
Lane J, Raimondi P, Kudela R (2009) Development of a logistic regression model for the
prediction of toxigenic Pseudo-nitzschia blooms in Monterey Bay, California. Mar Ecol
Prog Ser 383:37-51
Lefebvre K, Powell C, Busman M, Doucette G, Moeller P, Silver J, Miller P, Hughes M, Singaram
S, Silver M, Tjeerdema R (1999) Detection of domoic acid in northern anchovies and
California sea lions assocaited with an unusual mortality event. Nat Toxins 7:85-92
Lefebvre K, Robertson A, Frame E, Colegrove K, Nance S, Baugh K, Wiedenhoft H, Gulland F
(2010) Clinical signs and histopathology associated with domoic acid poisoning in
77
northern fur seals (Callorhinus ursinus) and comparison of toxin detection methods.
Harmful Algae 9:374-383
Legendre L (1990) The significance of microalgal blooms for fisheries and for the export of
particulate organic carbon in oceans. J Plankton Res 12:681-699
Lelong A, Hegaret H, Soudant P, Bates S (2012) Pseudo-nitzschia (Bacillariophyceae) species,
domoic acid and amnesic shellfish poisoning: revisting previous paradigms. Phycologia
51:168-216
Lewitus A, Horner R, Caron D, Garcia-Mendoza E, Hickey B, Hunter M, Huppert D, Kudela R,
Langlois G, Largier J, Lessard E, RaLonde R, Rensel J, Strutton P, Trainer V, Tweddle J
(2012) Harmful algal blooms along the North American west coast region: History,
trends, causes and impacts. Harmful Algae 19:133-159
Lundholm N, Daugbjerg N, Moestrup O (2002) Phylogeny of the Bacillariophyceae with
emphasis on the genus Pseudo-nitzschia (Bacillariphyceae) based on partial LSU rDNA.
Eur J Phycol 37:115-134
Lundholm N, Hansen P, Kotaki Y (2004) Effect of pH on growth and domoic acid production by
potentially toxic diatoms of the genera Pseudo-nitzschia and Nitzschia. Mar Ecol Prog
Ser 273:1-15
Lundholm N, Moestrup O (2000) Morphology of the marine diatom Nitzschia navis-varingica,
sp. nov. (Bacillariophyceae), another producer of the neurotoxin domoic acid. J Phycol
36:1162-1174
Maranda L, Wang R, Musauda K, Shimizu Y (1990) Investigations of the source of domoic acid in
mussels. In: Graneli E, Sundstrom B, Edler L, Anderson D (eds) Toxic Marine
Phytoplankton. Elsevier Science Publishing Co., New York
Mazzillo F, Field J, Staaf D, Carter M, Ohman M (2011) A note on the detection of the
neurotoxin domoic acid in beach-stranded Dosidicus gigas in the southern California
bight. CalCOFI Rep 52:109-115
Mazzillo F, Pomeroy C, Kuo J, Ramondi P, Prado R, Silver M (2010) Angler exposure to domoic
acid via consumption of contaminate fishes. Aquatic Biology 9:1-12
Miller P, Scholin C (1998) Identification and enumeration of cultured and wild Pseudo-nitzschia
(Bacillariophyceae) using species-specific LSU rRNA-targeted fluorescent probes and
filter-based whole cell hybridization. J Phycol 34:371-382
78
Omand M, Leichter J, Franks P, Guza R, Lucas A, Feddersen F (2011) Physical and biological
processes underlying the sudden surface appearance of a red tide in the nearshore.
Limnol Oceanogr 56:787-801
Orsini L, Procaccini G, Sarno D, Montresor M (2004) Multiple rDNA ITS-types within the diatom
Pseudo-nitzschia delicatissima (Bacillariophyceae) and their relative abundances across
a spring bloom in the Gulf of Naples. Mar Ecol Prog Ser 271:87-98
Paerl H (1997) Coastal eutrophication and harmful algal blooms: importance of atmospheric
deposition and groundwater as "new" nitrogen and other nutrient sources. Limnol
Oceanogr 42:1154-1165
Paerl H, Paul V (2012) Climate change: Links to global expansion of harmful cyanobacteria.
Water Res 46:1349-1363
Parsons T, Maita Y, Lalli C (1984) A Manual of Chemical and Biological Methods for Seawater
Analysis. Pergamon Press, Oxford
Roelke D, Grover J, Brooks B, Glass J, Buzan D, Southard G, Fries L, Gable G, Schwierzke-Wade L,
Byrd M, Nelson J (2011) A decade of fish-killing Prymnesium parvum blooms in Texas:
roles of inflow and salinity. J Plankton Res 33:243-253
Sala S, Sar E, Ferrario M (1998) Review of the materials reported as containing Amphora
coffeaeformia (Agardh) Kutzing in Argentina. Diatom Res 13:323-336
Schnetzer A, Jones B, Schaffner R, Cetinic I, Fitzpatrick E, Miller P, Seubert E, Caron D (accepted)
Coastal upwelling linked to toxic Pseudo-nitzschia australis blooms in Los Angeles
coastal waters, 2005-2007. J Plank Res
Schnetzer A, Miller P, Schaffner R, Stauffer B, Jones B, Weisberg S, DiGiacomo P, Berelson W,
Caron D (2007) Blooms of Pseudo-nitzschia and domoic acid in the San Pedro Channel
and Los Angeles harbor areas of the Southern California Bight, 2003-2004. Harmful
Algae 6:372-387
Scholin C, Buck K, Britschgi T, Cangelosi G, Chavez F (1996) Identification of Pseudo-nitzschia
australis (Bacillariophyceae) using rRNA-targeted probes in whole cell and sandwich
hybridization formats. Phycologia 35:190-197
Scholin C, Gulland F, Doucette G, Benson S, Busman M, Chavez F, Cordaro J, DeLong R, De
Vogelaere A, Harvey J, Haulena M, Lefebvre K, Lipscomb T, Loscutoff S, Lowenstine L,
Marin III R, Miller P, McLellan W, Moeller P, Powell C, Rowles T, Silvagni P, Silver M,
Spraker T, Trainer V, Van Dolah F (2000) Mortality of sea lions along the central
California coast linked to a toxic diatom bloom. Nat 403:80-84
79
Seubert E, Trussell S, Eagleton J, Schnetzer A, Cetinic I, Lauri P, Jones B, Caron D (2012) Algal
toxins and reverse osmosis desalination operations: Laboratory bench testing and field
monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid. Water Res 46:6563-
6573
Shipe R, Leinweber A, Gruber N (2008) Abiotic controls of potentially harmful algal blooms in
Santa Monica Bay, California. Continental Shelf Res 28:2584-2593
Smayda T (1997) What is a bloom? A comentary. Limnol Oceanogr 42:1132-1136
Smith J, Pauley K, Cormeir P, Angus R, Odense P, O'Neil D, Quilliam M, Worms J (1991)
Population dynamics and toxicity of various species of Dinophysis and Nitzschia from
the southern Gulf of St. Lawerence. Can Tech Rep Fish Aquat Sci 1799:25
Sournia A (1995) Red tide and toxic marine phytoplankton of the world ocean: an inquiry into
biodiversity. In: Lassus P, Arzul G, Erard E, Gentien P, Marcaillou C (eds) Harmful marine
algal blooms. Lavoisier, Intercept Ltd, Paris
Stauffer B, Gellene A, Schnetzer A, Seubert E, Oberg C, Sukhatme G, Caron D (2012) An
oceanographic, meteorological, and biological 'perfect storm' yields a massive fish kill.
Mar Ecol Prog Ser 468:231-243
Stehr C, Connell L, Baugh K, Bill B, Adams N, Trainer V (2002) Morphological, toxicological and
genetic differences among Pseudo-nitzschia (Bacillariophyceae) species in inland
embayments and outer coastal waters of Washington state, USA. J Phycol 38:55-65
Torrey H (1902) An unusual occurrence of dinoflagellata on the California coast. Amer Nat
36:187-192
Trainer V, Bates S, Lundholm N, Thessen A, Cochlan W, Adams N, Trick C (2012) Pseudo-
nitzschia physiological ecology, phylogeny, toxicity, monitoring and impacts on
ecosystem health. Harmful Algae 14:271-300
Utermohl H (1958) Zur gewassertypenfrage tropischer seen. Verhandlungen Internationalen
Vereinigung Limnologie 13:236-251
Van Dolah F (2000) Marine algal toxins: Origins, health effects and their increased occurrence.
Enviro Health Perspect 108:133-141
Vigilant V, Silver M (2007) Domoic acid in bethic flatfish on the continental shelf of Monterey
Bay, California, USA. Mar Biol 151:2053-2062
80
Work T, Barr B, Beale A, Fritz L, Quilliam M, Wright J (1993) Epidemiology of domoic acid
poisoning in brown pelicans (Pelecanus occidentalis) and Brandt's cormorants
(Phalacrocorax penicillatus) in California. J Zoo Wildl Med 24:54-62
Wright J, Boyd R, deFreitas A, Falk M, Foxall R, Jamieson W, Laycock M, McCulloch A, McInnes
A, Odense P, Pathak V, Quilliam M, Ragan M, Sim P, Thibault P, Walter J, Gilgan M,
Richard D, Dewar D (1989) Identification of domoic acid, a neuroexcitatory amino acid,
in toxic mussles from eastern Prince Edward Island. Can J Chem 67:481-490
Zingone A, Enevoldsen H (2000) The diversity of harmful algal blooms: a challenge for science
and management. Ocean Coast Manage 43:725-748
Reprinted from Environmental Science and Pollution Research,DOI: 10.1007/s11356-012-1420-
0, EL Seubert, AG Gellene, MDA Howard, P Connell, M Ragan, BH Jones, J Runyan, DA Caron,
Seasonal and annual dynamics of harmful algae and algal toxins revealed through weekly
monitoring at two coastal ocean sites off southern California, USA, Copyright 2013, with kind
permission of Springer Science and Business Media.
81
Chapter 3: Development, comparison and validation using ELISAs for the analysis of domoic
acid in California sea lion body fluids
ABSTRACT
Mortalities of California sea lions (Zalophus californianus) attributed to the neurotoxin
domoic acid (DA) produced by the diatom Pseudo-nitzschia have occurred repeatedly along the
US west coast since the late 1990s. Quantifying the amount of DA in these animals and
correlating this information with the presence of DA in phytoplankton and the local food web
has become a research focus for many scientists. However differences in materials, equipment,
technical capability, budgets and objectives of the various groups and/or agencies involved in
this work have influenced the DA quantification platforms employed. The goal of the present
study was to compare the performance of two commercially available enzyme-linked
immunosorbent assays for the analysis of DA in a spectrum of California sea lion body fluids and
to compare the results with liquid chromatography-mass spectrometry of the same samples.
The results indicated differences among these approaches, presumably owing to matrix effects
(particularly urine) and antibody reactivities. This information implies that care should be taken
in attempting to compare datasets generated using different analytical platforms and
interpreting the results of published studies.
INTRODUCTION
The diatom Pseudo-nitzschia has been known as a common member of the
phytoplankton community in California since the early 1900s (Allen 1934, 1936, Fryxell et al.
1997), however the capability of Pseudo-nitzschia to produce the neurotoxin domoic acid (DA)
82
and the threat that toxin can pose for human and wildlife health was not identified until the
end of the century. DA can be bio-accumulated in organisms feeding directly on toxic Pseudo-
nitzschia cells such as zooplankton (Bargu et al. 2002, Leandro et al. 2010), shellfish (Bates et al.
1989, Wright et al. 1989, Blanco et al. 2006) and planktivorous fish (Work et al. 1993, Lefebvre
et al. 1999, Lefebvre et al. 2001, Busse et al. 2006, Del Rio et al. 2010), which in turn serve as
vectors transporting the toxin to higher trophic levels of the food web. The first outbreak of
human illness occurred in 1987 on Prince Edward Island, Canada, when over 100 people were
sickened and three died after consuming DA contaminated blue mussels (Bates et al. 1989,
Wright et al. 1989, Perl et al. 1990b). DA has not caused any widespread human illnesses since
the initial Canadian outbreak, a situation that is most likely a consequence of the extensive
coastal monitoring programs for DA implemented by health departments worldwide since
1987. Conversely, DA continues to be the cause of marine bird (Fritz et al. 1992, Work et al.
1993, Sierra Beltran et al. 1997) and marine mammal (Scholin et al. 2000, Torres de la Riva et al.
2009, Fire et al. 2010, Leandro et al. 2010, Wang et al. 2012) mortality events in areas where
Pseudo-nitzschia occurs and these mortalities are often the first sign of an emerging DA event
in a given area.
The marine mammal predominately associated with DA mortality events on the US west
coast is the California sea lion (CSL; Zalophus californianus), undoubtedly due to their large
population sizes and overlapping distribution with Pseudo-nitzschia in coastal waters (Scholin et
al. 2000, Bejarano et al. 2008, Torres de la Riva et al. 2009, Bargu et al. 2010). There are two
types of DA exposure currently identified in CSLs, acute DA toxicosis that occurs when a CSL is
exposed to a single high dose of DA and chronic DA toxicosis that occurs when a CSL is
83
repeatedly exposed to sub-lethal concentrations of DA (Gulland et al. 2002, Goldstein et al.
2008). Presumably these different types of DA exposure influence the magnitude of DA
concentrations present in the body fluids of stranded animals at the time of rescue. Females
comprise the majority of CSL strandings and DA exposure not only threatens their health but
the health of a fetus they may be carrying (Brodie et al. 2006, Ramsdell & Zabka 2008, Goldstein
et al. 2009). The impact of DA on CSLs following rehabilitation can be seen in alterations of CSL
behavior, movement, dive pattern and survival (Gulland et al. 2002, Thomas et al. 2010).
Identifying strong positive correlations between the presence of DA producing Pseudo-
nitzschia and CSL strandings is impeded by several factors: the type of exposure (acute or
chronic), the amount of time between exposure and the time of stranding and rescue, the
health of the animal admitted to the rehabilitation center, the vector responsible for the
exposure, the amount of DA produced by Pseudo-nitzschia and the abundance of the Pseudo-
nitzschia cells producing DA. Unlike dinoflagellates that can form conspicuous blooms capable
of altering the color of the water, Pseudo-nitzschia does not typically reach cell abundances
large enough to be visibly noticeable. Moreover, coastal monitoring programs for Pseudo-
nitzschia and DA are routinely established at surface water stations accessible from the shore,
where the presence of DA in the phytoplankton and/or shellfish is most likely to overlap with
human activities. Yet, marine animals can be in contact with Pseudo-nitzschia blooms present
offshore and/or in thin layers (Rines et al. , McManus et al. 2008, Rines et al. 2010) and the
strandings of these animals often are the first sign of an emerging DA event prior to the Pseudo-
nitzschia cells being physically transported to shore and subsequently seen by a coastal
monitoring program.
84
The range of DA concentrations reported for fluids and solids collected from stranded
pinnipeds vary in range, fluid type, collection and DA quantification protocols (Table 3-1).
Multiple platforms exist for the measurement of DA including mouse bioassay, high
performance liquid chromatography (HPLC), liquid chromatography-mass spectrometry (LC-
MS), receptor binding assay (RBA) and enzyme linked immunosorbent assay (ELISA). Each
method has its own array of advantages and potential shortcomings, and laboratories generally
choose the methodology that will best suit their needs in terms of cost, technical sophistication
and research goals. For example, commercially available ELISAs generally offer lower cost
alternatives to analytical chemical approaches (i.e. – HPLC, LC-MS) that require a signification
investment of equipment and technical expertise. Several studies have implemented the rapid
and lower cost methodologies (i.e. – RBA and ELISA) as a pre-screening tool prior to analysis by
chemical methods (Brodie et al. 2006, Goldstein et al. 2008, Goldstein et al. 2009).
85
Table 3-1. Summary of DA concentrations and analysis methods reported in the literature for several pinniped species.
Species
Amniotic
Fluid
(ng/mL)
Cerebral
Spinal Fluid
(ng/mL)
Feces
(µg/g)
Gastric
Fluid
(ng/mL)
Serum
(ng/mL)
Urine
(ng/mL)
Method Reference
California sea lions
(Zalophus
californianus)
2.5 – 152.0 HP-LC
(Lefebvre et
al. 1999)
1.3 – 182.0 RBA
(Lefebvre et
al. 1999)
1.31 –
182.01µg/mL
170.0 – 200.0 30.0 – 3720.0 RBA
(Scholin et
al. 2000)
4.0 – 34.0
0.5 – 15.0 7.0 – 261.0 LC-MS
(Brodie et al.
2006)
1.0 – 82.02 HP-LC
(Goldstein
et al. 2008)
3.0 – 200.0 2.0 – 3720.0 LC-MS/MS
(Goldstein
et al. 2008)
3.0 – 9.3
0.3 – 44.0 2.0 – 17.6 LC-MS/MS
(Goldstein
et al. 2009)
1.4 – 96.8 HP-LC
(Bargu et al.
2010)
0.2 – 96.8 HP-LC
(Bargu et al.
2012)
Harbor seals
(Phoca vitulina)
10.0
0.002 – 0.063 8.0 – 10.0 2.0 – 16.0
Biosense
ELISA
(Hall &
Frame 2010)
Northern fur seals
(Callorhinus ursinus)
20.0 ng/g
0.002 – 18.6 2 – 286 ng/g
1.0 – 2784.0
ng/g
Biosense
ELISA
(Lefebvre et
al. 2010)
0.53 – 2.80
190.0 –
13661.0 ng/g
HP-LC
(Lefebvre et
al. 2010)
0.44 – 54.73 811.0 – 828.0
371.0 – 5630.0
ng/g
LC-MS/MS
(Lefebvre et
al. 2010)
46.47
512 – 12693
ng/g
RBA
(Lefebvre et
al. 2010)
86
The primary objective of the present study was to compare the performance of several
of the methodologies available for quantification of DA in CSL body fluids, thereby enabling
some degree of extrapolation across existing datasets and to provide context for comparing
new results to previously published studies. This objective was met through (1) validating a
protocol adapted for the measurement of DA by ELISA in sea lion body fluids that minimized the
sample volume required and reduced sample handling procedures; (2) comparing the
performance of two commercially available ELISAs, the monoclonal antibody ELISA
manufactured by Mercury Science, Inc. (MS) and the polyclonal antibody ELISA manufactured
by Biosense (BS) using the modified protocol; (3) comparing the results from ELISA with a well-
established analytical method utilizing LC-MS. The CSL body fluids used in the validation study
include amniotic fluid (AF), cerebral spinal fluid (CSF), serum and urine.
METHODS
DA quantification methods. Three method platforms for the analysis of DA (two commercially
available ELISAs and LC-MS) were compared for their ability to accurately measure DA in CSL
body fluids and to provide information on how to compare data collected using these different
platforms. Each fluid type selected (AF, CSF, serum and urine) has unique properties that may
cause interferences for ELISA or LC-MS methodologies, warranting examination of each fluid
individually. The MS ELISA is a monoclonal antibody assay developed by the National
Oceanographic and Atmospheric Association Centers for Coastal Ocean Science, National Ocean
Service, the Northwest Fisheries Science Center with Mercury Science, Inc., (Durham, NC). It
has been validated for the analysis of DA in shellfish tissues and in dissolved and particulate
87
phytoplankton samples (Litaker et al. 2008, Seubert et al. 2012). The BS ELISA is a polyclonal
antibody based assay developed by Biosense Laboratories (Bergen, Norway), and has been
validated by both single and inter-laboratory studies for the analysis of DA in shellfish tissues
(Kleivdal et al. 2007, Olson & Lessard 2010) and for the analysis of DA concentrations present in
rat serum and brain samples (Hesp et al. 2005). An Agilent 6130 LC-MS system operated in
positive electrospray ionization mode with an Agilent Zorbax Rapid Resolution column and
Selected Ion Monitoring of DA (312 amu) was used for LC-MS analysis generally following the
method of Wang et al. (2007). Quantification was based on peak area and an external standard
curve using National Research Council Canada CRM-DA-f standards. Peaks were confirmed
based on the presence of daughter fragments at 266 and 248 amu. Since the objective of the
study was to compare methods and matrices, the unknown samples were run blind and not
corrected for matrix effects using standard addition or an internal standard.
Sample Collection and Selection for Validation Study. Samples of AF, CSF, serum and urine
were obtained from stranded CSLs treated by the Pacific Marine Mammal Center (PMMC;
Laguna, CA) during 2007 and 2009. Following collection, samples were stored at -20°C at
PMMC, transported frozen to the University of Southern California (USC; Los Angeles, CA) and
once again stored at -20°C until analysis via ELISA. Samples collected in 2007 were initially
analyzed by the BS ELISA and samples collected in 2009 were initially analyzed by the MS ELISA,
typically within one month following receipt at USC. Fluid samples that were determined to be
below the detection limit of the respective ELISA platforms and contained sufficient volume
after the initial analysis were employed in a ‘spike and recovery’ study conducted during the fall
88
of 2009 that utilized simultaneous analysis by all three platforms. Samples collected during
2009 that yielded measurable DA concentrations using the MS ELISA and contained sufficient
remaining volume to be analyzed simultaneously on all three platforms, were also stored and
included in the fall 2009 study to allow for comparison of naturally DA positive fluid samples
across all three platforms. These latter samples also allowed determination of any DA
degradation that may have occurred in the samples during storage.
Modified ELISA protocol. A modified ELISA protocol was developed in 2009 with the primary
goal of minimizing the required sample volume and reducing sample preparation. We reasoned
that a methanol extraction step typically used in the extraction of DA from solid matrices (i.e. –
phytoplankton cells, shellfish tissues) might not be necessary for fluid samples as the DA would
already be in the dissolved form. Fluid samples were vortexed for 1 minute, diluted 1:25 with
the sample buffer provided by the respective ELISA manufacturer and the diluted sample briefly
vortexed immediately prior to pipetting onto the ELISA plate. The expected limit of detection
for each ELISA platform was calculated from the plate sensitivity reported by the manufacturer
and adjusted for the 1:25 minimum dilution. The BS ELISA is reported to have a 0.01 ng/mL
plate sensitivity and therefore the limit of detection for a sample diluted 1:25 is expected to be
0.25 ng/mL. The MS ELISA is reported to have a 0.1 ng/mL plate sensitivity and the limit of
detection expected is 2.5 ng/mL.
Method comparison study. Analysis of CSL body fluids using the modified protocol for both
ELISAs and LC-MS, were performed within one week in October 2009 at the University of
89
California Santa Cruz, in order to minimize degradation of DA that may occur in samples over
long storage periods. Samples selected for the spike and recovery portion of the comparison
study were previously determined to be below detection of the respective ELISA platform
during their initial receipt and analysis in 2007 or 2009. A portion of the samples measured in
the spring of 2009 that had quantifiable concentrations of DA using the MS ELISA and the
modified protocol were re-analyzed simultaneously on all three platforms. The study was
designed to allow the following comparisons (1) performance of the modified ELISA protocol on
the analysis of DA concentrations in AF, CSF, serum and urine samples spiked with known
concentrations of DA; (2) comparison of the results obtained with the modified protocol by the
MS and BS ELISAs; (3) comparison of results obtained from solid-phase extraction (SPE) cleaned
spiked samples analyzed by LC-MS and the MS ELISA; (4) comparison of LC-MS results on SPE
cleaned samples to MS and BS ELISA results without SPE cleaning (Figure 3-1).
Figure 3-1. Flow diagram outlining the validation design for California sea lion body fluid samples.
CSL Body Fluid
Sample
Aliquot for
ELISAs and LCMS
BS and MS ELISA
non-SPE-cleaned
analysis
SPE-cleanup
MS ELISA SPE-
cleaned analysis
LC-MS SPE-
cleaned analysis
Aliquot spiked
with known DA
concentration
BS and MS ELISA
non-SPE-cleaned
analysis
SPE-cleanup
MS ELISA SPE-
cleaned analysis
LC-MS SPE-
cleaned analysis
90
Preparation of standards. The DA standard used to spike CSL fluid samples and to prepare
standard curves of LC-MS analysis was obtained from the National Research Council, Canada
(CRM-DA-f; Ottawa, Ontario). Two standard curves were prepared for the LC-MS analysis, one
in Milli-Q water and the second in LC-MS grade 50% methanol. A subset of the Milli-Q
standards were SPE-cleaned (see section below) prior to analysis by LC-MS in order to quantify
the amount of DA lost during the cleanup procedure. The LC-MS standard curves were made
through serial dilution with final concentrations of: 1, 2, 5, 20, 50, 100, 250, 500 ng/mL. A
standard of 1000 ng/mL was prepared in Milli-Q to use in the spiking of AF, CSF, serum and
urine samples to the following concentrations: 12.5, 15, 18, 21.5, 26, 31, 37, 44.5, 53.5, 64, 77,
92.5, 110 ng/mL. A matrix free 1 ng/mL Milli-Q standard was used for assessing ELISA platform
performance during the study.
LC-MS sample handling procedure. The procedure used for LC-MS analysis was modified from
a procedure previously employed for seawater and phytoplankton samples, described in Wang
et al. (2007). The same samples analyzed via ELISA methods were also analyzed on LC-MS but
were cleaned using Bond Elut SPE columns with LRC-C18 resin (Varian, Inc., now Aglient
Technologies, Santa Clara, CA). The columns were conditioned prior to the addition of sample
by vacuum filtering 10 mL of 100 % methanol followed by 10 mL of LC-MS grade water (Fisher
Scientific, Pittsburgh, PA). Samples were acidified with 0.5 mL 2:5:93 formic
acid:methanol:water and 4 mL 5 % formic acid was added prior to introduction to the SPE
column. Samples were pipetted into the column and 4 mL of 1.5 % formic acid was added to
the sample followed by vacuum filtration and the extract discarded. The final extraction step
91
used vacuum filtration of 3 mL of 50 % methanol onto the column and the resulting extract
collected for analysis. The method detection limit of 0.48 ng/mL was determined using seven
spiked Milli-Q samples with SPE cleanup.
Analysis of spike and recovery of DA results. The quality of the fluid samples collected by
rehabilitation centers can be impacted by multiple factors, as discussed in the introduction, and
fluid samples from a number of individual animals were employed in the spike and recovery
portion of the study to account for this variability. The samples were divided into two portions,
one portion of each sample remained unspiked and the other portion was spiked with DA
standard at known concentrations (Figure 3-1). The unspiked portions were run on each
platform, with non-SPE cleaned samples analyzed using the MS and BS ELISAs and SPE-cleaned
samples analyzed by MS ELISA and LC-MS. The remaining, spiked portions of each sample were
vortexed for 1 minute and then analyzed as noted in the previous sections detailing the
modified ELISA protocol and LC-MS sample handling procedure.
ELISA samples were run in triplicate and spread across two plates to account for any
inter- or intra-plate variability that may have arisen during commercial fabrication. DA
concentrations for each ELISA sample were determined from the average results of duplicate
wells and calculated using Excel spreadsheets provided by the respective manufacturers. BS
ELISA results were quantified using an Excel Macro that employs a 4-parameter logistics curve-
fitting model to produce a standard curve for each plate and sample concentrations were
determined by extrapolating from the standard curve. MS ELISA results were calculcated using
an Excel spreadsheet that used a ratio between the maximal absorbance signal (a control
92
containing no DA) and the absorbance signal of the sample in conjunction with constants for
the midpoint and slope of a standard curve determined by the manufacturer when the method
was developed. Both quantification spreadsheets automatically identified samples outside of
the working range of the ELISA (i.e. – too dilute or too concentrated) and calculated coefficients
of variation (CV) for the duplicate sample wells. Samples with CVs greater than 15 % were
eliminated as recommended by the manufacturers because the high amount of variation makes
it difficult to identify which of the duplicate wells can be used to calculate an accurate DA
concentration.
The measured DA concentrations determined by the respective platforms were plotted
against the known DA concentration spiked into each sample. The ELISA and LC-MS samples
were treated blindly, with no correction of ELISA or LC-MS results for the known DA
concentrations spiked into the samples. This was done in order to more appropriately mimic
the manner in which CSL body fluid samples of an unknown DA concentration might be handled
in different laboratories. The slopes and R
2
values of the linear regressions plotted for each
platform were recorded and the slopes statistically compared. The linear regressions were
carried out with SigmaPlot (v 11.0.0, Systat Software, Inc.). A Student’s t test was computed as
the difference between the two slopes being compared and divided by the standard error of
the difference between the slopes. The comparison of slopes using the Student’s t test has
been identified as reliable for determining agreement between two methods and agreement to
the ideal equality line with a slope of 1 (Westgard & Hunt 1973, Bland & Altman 1986). Deming
regressions were performed to test for linear relationships between platform results using the
XLSTAT Macro for Microsoft Excel (v 2012, Addinsoft SARL).
93
RESULTS AND DISCUSSION
Spike and recovery of DA in CSL AF samples. The DA concentrations measured by each
method for the spiked CSL AF samples were plotted versus the known DA concentrations added
to each sample and linear regressions performed to determine the R
2
and slope (Figure 3-2A;
Table 3-2). The highest R
2
value of 0.94 was obtained for the MS ELISA results using non-SPE-
cleaned samples, indicating good linearity across the range of DA concentrations examined.
The slope obtained for the MS non-SPE-cleaned regression line was 0.67, lower than and
statistically different from the ideal slope of 1 (t
0.05,2,8
> 2.31). The BS ELISA results using non-
SPE-cleaned samples also showed good linearity across the range of DA concentrations as
indicated by an R
2
of 0.90. The slope of the regression line was 1.4, higher than and statistically
different from the ideal slope of 1 (t
0.05,2,10
> 2.23). The results from the SPE-cleaned samples
analyzed on the MS ELISA had a high R
2
(0.88), but the slope of the linear regression line was
0.29, much lower than and statistically different from the ideal slope of 1 (t
0.05,2,10
> 2.23). The
slope of 0.87 obtained for the LC-MS measured SPE-cleaned samples not statistically
distinguishable from the ideal slope of 1 (t
0.05,2,7
< 2.37) and the R
2
value of 0.75 indicated a
relatively good linearity across the range of DA concentrations tested.
94
Figure 3-2. DA concentrations measured using different analytical platforms and protocols in samples of CSF spiked with DA
standard. The dotted lines in all graphs show the expected DA concentration based on the amount of DA added to each sample.
Error bars represent standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned results
represented by black circles and the solid black line, the MS SPE-cleaned results represented by white circles and the long
dashed and dotted line, the BS ELISA non-SPE-cleaned results represented by gray circles and the gray line, and the LC-MS
results represented by black triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-SPE-
cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned); (C) DA concentrations obtained using
the MS ELISA (SPE-cleaned) plotted with the LC-MS results; (D) DA concentration obtained using the MS ELISA (non-SPE-
Cleaned) represented by the black circles and black line and the DA concentration results obtained using the BS ELISA (non-
SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS (SPE-cleaned) results.
95
Table 3-2. R
2
and slope values obtained from linear regression lines plotted for each method platform and body fluid versus the
known spike concentration of DA.
Mercury Science ELISA Biosense ELISA LC-MS
Non-SPE Cleaned SPE Cleaned Non-SPE Cleaned SPE Cleaned
R
2
Slope R
2
Slope R
2
Slope R
2
Slope
Amniotic
Fluid
0.94 0.67 0.88 0.29 0.89 1.42 0.75 0.87
Cerebral
Spinal
Fluid
0.82 0.78 0.55 0.11 0.92 1.52 0.48 0.70
Serum 0.85 0.38 0.86 0.29 0.94 0.73 0.99 1.07
Urine 0.55 0.38 0.84 0.24 0.36 0.71 0.19 0.38
Comparison of the slopes calculated using each analytical approach with the AF spiked
samples revealed that the slope obtained using the MS ELISA on non-SPE-cleaned samples
(0.67) was not statistically different from the slope of the LC-MS regression line (0.87; t
0.05,2,15
<
2.13), while the slopes obtained for the remaining platforms were found to be statistically
different from one another (Figure 3-2A). The latter result was not unexpected given the basic
dissimilarities in sample preparation and analytical methods. Results obtained for the non-SPE-
cleaned samples measured by the MS ELISA were plotted versus the BS ELISA results and an R
2
and slope were determined (Figure 3-2B). The R
2
of 0.89 indicated good linearity between the
results of the two ELISA methods and a Deming regression analysis was performed to confirm a
linear relationship. The computed p-value of 0.95 for the Deming regression is greater than α =
0.05, and supported the null hypothesis that the relationship between the MS and BS ELISA
non-SPE-cleaned AF results was linear. The slope of the Deming regression was 2.4, most likely
influenced by the MS ELISA slightly underestimating DA concentrations at the higher
concentrations investigated and the BS ELISA overestimating DA concentrations at the higher
concentrations.
96
The results for the SPE-cleaned AF samples measured by LC-MS and the MS ELISA were
plotted versus one another to compare samples that were handled in the same manner but
analyzed by these two different methods (Figure 3-2C). The R
2
of 0.73 demonstrated relatively
good linearity between the two methods, further confirmed through a Deming regression with
computed p-value of 0.34, greater than α = 0.05, supporting the null hypothesis that the
relationship was linear. The low slope of 0.27 for the regression line indicated significant
underestimation of DA concentrations on SPE-cleaned AF samples by the MS ELISA.
Underestimation of DA concentrations in the SPE-cleaned AF samples was also evidenced by
the slope of 0.20 obtained when these values were plotted versus the known spike
concentrations (Figure 3-2A). SPE cleaning produced an extract comprised of 50 % methanol,
which may have an inhibitory effect on ELISA antibodies and therefore underestimate DA
concentrations. The low slope values obtained when plotting the MS results for SPE-cleaned
samples of AF versus the spiked DA concentration (Figure 3-2A) and the low slope obtained in a
Deming regression performed versus LC-MS results (Figure 3-2C) on the same samples indicate
that SPE-cleaning may be contraindicated for analysis of AF samples on an ELISA platform.
The results obtained by both ELISAs with non-SPE-cleaned AF samples were plotted
versus the LC-MS results on the SPE-cleaned AF samples (Figure 3-2D). The R
2
values for the
regression lines of the MS and BS ELISA results were 0.60 and 0.56 respectively, indicating a
moderate linearity present between the ELISA and LC-MS measured values. Deming
regressions of the MS and BS ELISA results versus the LC-MS results produced p-values of 0.34
in each regression and supported the null hypothesis that the MS and BS ELISA results on non-
SPE-cleaned AF samples have a linear relationship with the LCMS SPE-cleaned results.
97
Spike and recovery of DA in CSL CSF samples. DA concentrations measured by each method
for the spiked CSL CSF samples were plotted versus the known DA concentrations and linear
regression performed to determine the R
2
and slope values (Figure 3-3A; Table 3-2). The
highest R
2
value of 0.92 was tabulated for the BS ELISA results on non-SPE-cleaned CSF samples,
indicating good linearity across the range of DA concentrations measured. The slope of the BS
ELISA regression line was 1.5, greater and statistically different than the ideal slope of 1 (t
0.05,2,10
> 2.23). The BS ELISA overestimated DA concentrations in CSF samples, similar to the result
obtained for the AF samples analyzed by the BS ELISA discussed above. The slope of the
regression line of MS ELISA results on non-SPE-cleaned CSF samples was 0.78 and not
statistically different than the ideal slope of 1 or the slope of the LC-MS regression line (0.70;
t
0.05,2,18
< 2.10, t
0.05,2,8
< 2.31, respectively). The R
2
of 0.82 for the non-SPE-cleaned CSF samples
analyzed using the MS ELISA indicated a relatively good linearity across the range of DA
concentrations tested. The LC-MS regression line had a relatively poor R
2
value (0.48), but the
slope of the regression line was not statistically different than the ideal slope of 1 (0.70; t
0.05,2,6
< 2.45). The low R
2
value may have been a consequence of matrix influence on the ability of
the LC-MS to quantify DA concentrations, or DA lost in the SPE-cleanup procedure. This could
be addressed in the future by including standard additions for the unknown samples, or by
using the same matrix for the standard curve. The SPE-cleaned CSF samples that were analyzed
on the MS ELISA platform exhibited relatively good linearity with a R
2
value of 0.55, but the
slope of the regression line was relatively flat (0.11) and was statistically different from the
ideal slope of 1 (t
0.05,2,5
> 2.57). Whether this results was a consequence of the impact of 50 %
98
methanol extract on the ELISA, or the SPE- cleanup procedure itself is unknown. However,
these results verified previous findings that the SPE-cleanup step was contraindicated for ELISA.
Figure 3-3. DA concentrations measured using different analytical platforms and protocols in samples of serum spiked with DA
standard. The dotted lines in all graphs show the expected DA concentration based on the amount of DA added to each sample.
Error bars represent standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned results
represented by black circles and the solid black line, the MS SPE-cleaned results represented by white circles and the long
dashed and dotted line, the BS ELISA non-SPE-cleaned results represented by gray circles and the gray line, and the LC-MS
results represented by black triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-SPE-
cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned); (C) DA concentrations obtained using
the MS ELISA (SPE-cleaned) plotted with the LC-MS results; (D) DA concentration obtained using the MS ELISA (non-SPE-
Cleaned) represented by the black circles and black line and the DA concentration results obtained using the BS ELISA (non-
SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS (SPE-cleaned) results.
99
The CVs calculated for non-SPE-cleaned CSF samples were higher using the BS ELISA
than the MS ELISA, similar to the results obtained with the analysis of the AF samples discussed
above. Non-SPE-cleaned and SPE-cleaned samples measured by the MS ELISA both had average
CVs of 4 % and no results were discarded for CVs > 15 %. In contrast, the non-SPE-cleaned
samples analyzed by the BS ELISA had an average CV of 14 %, with 20 results having CVs > 15 %
(and were therefore discarded).
The slopes of the regression lines obtained for the non-SPE-cleaned samples analyzed by
the MS and BS ELISAs when plotted versus the known DA concentration were found to be
statistically different (t
0.05,2,18
> 2.13; Figure 3-3A). The absolute values obtained for the two
methods slightly overestimated (BS) or underestimated (MS) the DA concentrations relative to
the known amounts of DA spiked into these samples. When the results for each ELISA were
plotted against each other, the R
2
of 0.92 indicated good linearity across the range of DA
concentrations measured by each platform, but the BS ELISA yielded values that were
consistently higher than values obtained using the MS ELISA (Figure 3-3B). A Deming regression
confirmed the null hypothesis of a linear relationship between the MS and BS ELISA results with
the computed p-value of 0.082 (> α = 0.05).
Results for SPE-cleaned CSF samples analyzed by the MS ELISA and LC-MS were plotted
versus one another and an R
2
and slope value calculated (Figure 3-3C). Although samples were
handled in an identical manner prior to analysis, the SPE-cleaning only appeared to impact the
results of the MS ELISA analysis, presumably due to the effect of the 50 % methanol extract
produced by SPE-cleaning. The impact on toxin detection capabilities of SPE-cleaned CSF
samples analyzed by the MS ELISA was reflected in the slope value of 0 and confirmed the
100
finding that SPE was contraindicated for the analysis of DA samples by ELISA. A Deming
regression computed a p-value of 0.51 (> α = 0.05) and supported the null hypothesis that the
relationship between the LC-MS and MS ELISA results was linear, in spite of the flat slope and a
low R
2
value of 0.15.
The results of non-SPE-cleaned CSF samples using both ELISAs were plotted versus the
LC-MS results on the SPE-cleaned CSF samples and Deming regressions performed to test for
linear relationships (Figure 3-3D). R
2
values of 0.59 and 0.54 were obtained for the BS and MS
ELISA results of non-SPE-cleaned CSF samples, respectively. The moderate linearity inferred
from the R
2
value for the BS ELISA results was confirmed with a Deming regression with a
computed p-value of 0.70 (> α = 0.05) supporting the null hypothesis of a linear relationship.
The Deming regression computed a p-value of 0.99 (> α = 0.05) to support the null hypothesis
that the non-SPE-cleaned MS ELISA results had a linear relationship with the SPE-cleaned LCMS
results.
Spike and recovery of DA in CSL serum samples. The DA concentrations measured by each
method for spiked CSL serum samples were plotted versus the known DA concentrations and
linear regression performed to determine the R
2
and slope (Figure 3-4A; Table 3-2). The
regression line slope of DA concentrations in SPE-cleaned serum samples analyzed using LC-MS
was 1.1 and not statistically different from the ideal slope of 1 (t
0.05,2,6
< 2.45; R
2
value was
0.99). The absolute values obtained by LC-MS were all slightly less than the expected (spiked)
values, indicating a slight loss of DA, presumably during the SPE cleaning, or possibly
suppression of ionization due to matrix effects; again this could be corrected in the future with
101
the use of matrix-specific standard curve or with internal spikes of known concentrations. The
BS ELISA results on non-SPE cleaned serum samples returned the second highest R
2
values with
a 0.94 and a slope of 0.73 that was not statistically different than the ideal slope of 1 (t
0.05,2,3
<
3.18). The absolute toxin concentrations for the BS ELISA method were all very similar to the
expected (spiked) concentrations of DA, but the results may be misleading because the number
of samples plotted was reduced considerably by the elimination of sample results yielding poor
CV values. The CVs calculated for the BS ELISA averaged 18 %, with 23 results discarded (CVs >
15 %). The frequency of high CV values obtained on the BS ELISA may have been influenced by
the color of the serum samples analyzed. Serum is blood plasma with the fibrinogens removed
and should not contain any red or white blood cells. The removal of red and white blood cells is
affected by the efficiency of the centrifugation and the quality of the serum sample collected
from the animal. We speculate that the presence of red blood cells (or material from lysed
cells) in a serum sample may interfere with colorimetric determination of DA concentration
determination by ELISA methods. However, this effect was not observed for the MS ELISA
analysis on either the non-SPE or SPE-cleaned serum samples. The non-SPE and SPE-cleaned
sample CVs for the MS ELISA averaged 4 % and no results were discarded for high CVs.
102
Figure 3-4. DA concentrations measured using different analytical platforms and protocols in samples of serum spiked with DA
standard. The dotted lines in all graphs show the expected DA concentration based on the amount of DA added to each sample.
Error bars represent standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned results
represented by black circles and the solid black line, the MS SPE-cleaned results represented by white circles and the long
dashed and dotted line, the BS ELISA non-SPE-cleaned results represented by gray circles and the gray line, and the LC-MS
results represented by black triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-SPE-
cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned); (C) DA concentrations obtained using
the MS ELISA (SPE-cleaned) plotted with the LC-MS results; (D) DA concentration obtained using the MS ELISA (non-SPE-
Cleaned) represented by the black circles and black line and the DA concentration results obtained using the BS ELISA (non-
SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS (SPE-cleaned) results.
103
The R
2
value determined for the MS ELISA for non-SPE-cleaned serum samples was 0.85
and the SPE-cleaned samples returned a R
2
value of 0.86. The R
2
values indicate good linearity
across the range of DA concentrations measured, however all the DA concentrations analyzed
by using the MS ELISA platform were less than the expected values, and the slopes of 0.38 and
0.29, respectively, were statistically different than the ideal slope of 1 (t
0.05,2,6
> 2.45, t
0.05,2,4
>
2.78; Figure 3-4A).
The R
2
of 0.08 and slope of 0.35 obtained when plotting the results of the non-SPE-
cleaned serum samples measured using both ELISAs indicated a lack of agreement between the
two platforms (Figure 3-4B). However, the number of samples included in the comparison plot
was small (4) because many of the samples measured using the BS ELISA were discarded due to
CVs > 15 %. Deming regression analysis computed a p-value of 0.76 (> α = 0.05) supporting the
null hypothesis that there is a linear relationship between the non-SPE-cleaned serum samples
analyzed by the MS and BS ELISAs.
The regression for the SPE-cleaned serum samples measured by LC-MS and the MS
ELISA yielded an R
2
of 0.89, indicating a good linearity between the measured DA
concentrations (Figure 3-4C). However, most of the MS ELISA values were less than the values
obtained using LC-MS, particularly for samples with high concentrations of DA added, as
indicated by the slope of the regression line (0.27). Deming regression analysis computed a p-
value of 0.98 (> α = 0.05) supporting the null hypothesis of a linear relationship between SPE-
cleaned serum samples analyzed by the MS ELISA and LC-MS.
The results obtained from non-SPE-cleaned serum samples using the two ELISA
platforms were plotted versus the LC-MS results (Figure 3-4D). The R
2
for the BS ELISA analysis
104
was 0.92, indicating good linearity between the measured LC-MS and BS ELISA measured DA
concentrations. However, relatively few samples were compared because many of the BS ELISA
results were discarded due to high CVs, and those samples that were compared were all higher
than values obtained by LC-MS. The R
2
value for the MS ELISA results was 0.29, indicating poor
linearity between DA concentrations in spiked serum samples measured by LC-MS and the MS
ELISA. Deming regressions computed p-values of 0.76 (> α = 0.05) for the MS and BS ELISA non-
SPE-cleaned serum results in comparison to the SPE-cleaned LC-MS results. The null hypothesis
was supported in both instances, with the non-SPE-cleaned MS ELISA results had a linear
relationship with the LC-MS SPE-cleaned results and the non-SPE-cleaned BS ELISA results had a
linear relationship with the LC-MS SPE-cleaned results.
Spike and recovery of DA in CSL urine samples. The DA concentrations measured by each
method for the spiked CSL urine samples were plotted versus the known DA concentrations and
linear regressions performed to determine the R
2
and slope values (Figure 3-5A; Table 3-2). The
highest R
2
(0.84) was obtained for the values determined using the MS ELISA on SPE-cleaned
urine samples. However, the slope of 0.24 was lower than and statistically different from the
ideal slope of 1, and all absolute values were considerably less than values anticipated from the
DA concentrations added to the samples. The non-SPE-cleaned samples analyzed by the MS
and BS ELISAs were 0.55 and 0.36, respectively, but the absolute values obtained were
generally more similar to the expected (spiked) concentrations of DA. Therefore, the SPE-
cleaning improved the precision of replicate samples analyzed by the MS ELISA, but the results
underestimated the known DA concentrations, especially at the higher concentrations of DA
105
used in the spiked samples. The R
2
value of 0.19 obtained for the LC-MS measured results was
the lowest R
2
value not only the analysis of the urine samples but in all CSL body fluids tested by
LC-MS. Similarly, the R
2
values for the MS and BS ELISA results on non-SPE-cleaned urine
samples were the lowest R
2
values of all body fluids tested by these platforms.
The poor performance of the urine analysis by the three platforms may be attributable
to the high salt content of CSL urine, which is 2.5x more concentrated than seawater and 7-8x
more concentrated than their blood. The concentrated urine produced by these animals is a
mechanism for riding their bodies of excess salt and reduce freshwater loss. It is possible that
the salt inhibits antibody performance in the ELISA platforms, affects the loss of DA from
samples during the SPE cleaning procedure and/or impacts the efficiency of ion formation
during LC-MS analysis. SPE cleaning is recommended for the analysis of DA concentrations in
seawater by LC-MS as the salt presence impacts MS signal stability (Wang et al. 2007).
106
Figure 3-5. DA concentrations measured using different analytical platforms and protocols in samples of urine spiked with DA
standard. The dotted lines in all graphs show the expected DA concentration based on the amount of DA added to each sample.
Error bars represent standard deviations of triplicate replicates. (A) Results from all platforms, MS non-SPE-cleaned results
represented by black circles and the solid black line, the MS SPE-cleaned results represented by white circles and the long
dashed and dotted line, the BS ELISA non-SPE-cleaned results represented by gray circles and the gray line, and the LC-MS
results represented by black triangles and the long dashed line; (B) DA concentrations measured using the BS ELISA (non-SPE-
cleaned) plotted with the concentrations obtained using the MS ELISA (non-SPE-cleaned); (C) DA concentrations obtained using
the MS ELISA (SPE-cleaned) plotted with the LC-MS results; (D) DA concentration obtained using the MS ELISA (non-SPE-
Cleaned) represented by the black circles and black line and the DA concentration results obtained using the BS ELISA (non-
SPE-cleaned) represented by gray circles and the gray line are plotted versus the LC-MS (SPE-cleaned) results.
107
Direct comparisons of the non-SPE-cleaned urine samples analyzed by MS and BS ELISAs
(R
2
= 0.08; Figure 3-5B), the SPE-cleaned samples analyzed by MS ELISA and LC-MS (R
2
= 0.14;
Figure 3-5C), and the non-SPE samples analyzed by the MS and BS ELISAs to results of LC-MS (R
2
= 0.05, 0.06, respectively; Figure 3-5D) all produced low R
2
values. These results reflected the
poor relationships identified between each of the methods and the expected DA concentrations
spiked into urine samples (Figure 3-5A; Table 3-2). Further, a large number of samples were
discarded from the ELISA results due to high CVs. The non-SPE-cleaned samples analyzed using
the MS ELISA had an average CV of 6 %, but 5 sample results were discarded for exceeding 15
%. The CVs calculated for the SPE-cleaned urine samples measured using the MS ELISA
averaged 3 % and no results were discarded. The BS ELISA analysis of non-SPE-cleaned samples
averaged 14 % with 24 sample results returning CVs of greater than 15 %.
Analysis of DA in CSL body fluids from natural samples. The modified DA protocol described in
this study for the analysis of DA in CSL body fluids was adopted at USC in 2009. Samples
collected from stranded CSLs in Orange County, California, by PMMC were analyzed within one
month of sample receipt using the modified protocol on the MS ELISA. Samples with DA
concentrations thought to be significant enough to survive storage at -20°C until the planned
validation study were set aside to be reanalyzed by the three platforms simultaneously in the
fall of 2009. DA degradation in particulate phytoplankton samples has been shown to be highly
variable (Quay et al. 2011) and can potentially influence successful quantification of DA
concentrations of CSL body fluids after long term storage. AF samples collected in the spring of
108
2009 were below the detection limit of the MS ELISA and therefore no naturally positive AF
samples were available for analysis.
Available fluid samples from eleven different animals were analyzed simultaneously
using the four methods (Table 3-3). The majority of the samples (8 animals) were stored at -
20°C for six months, samples from two animals were stored for eight months and samples from
one animal were stored for nine months. The amount of DA degradation observed in individual
samples analyzed by the MS ELISA using the modified protocol was highly variable. There was a
49 % decrease in the DA concentration measured in the urine sample stored for 9 months, an
average decrease 40% in DA concentrations for urine samples stored for eight months and an
average 43 % decrease in DA concentrations for urine samples stored for six months for an
overall average loss of 44 %. The individual percent decreases in DA concentrations measured
by the MS ELISA in urine samples ranged from 4 % to 72 %, and in two instances the DA
concentrations measured were higher than the original DA concentrations measured from fresh
samples. Four serum samples were stored for six months and losses of DA ranged from 3 % to
64 % when measured by MS ELISA. Two of the four samples decreased in DA concentration
during storage to levels below the detection of the MS ELISA. One of the CSF samples stored
for six months decreased 54 % in DA concentration by MS ELISA analysis and the other
decreased to below the detection limit of the MS ELISA. It is recommended that body fluid
samples be analyzed for DA content promptly following receipt as DA degradation in samples
during storage was found to be highly variable.
109
Table 3-3. Measurements of DA concentrations naturally present in California sea lion body fluids across the three platforms. The original DA concentration determined by MS
ELISA with the modified DA analysis protocol within one month of receipt of samples at USC. Values reported for MS NonSPE, BS NonSPE and MS SPE analysis during the
validation study in fall 2009 are averages of triplicate replicates.
Cerebral Spinal Fluid (ng/mL) Serum (ng/mL) Urine (ng/mL)
Animal
MS
NonSPE
(Orig)
MS
NonSPE
BS
NonSPE
MS
SPE
LC-MS
MS
NonSPE
(Orig)
MS
NonSPE
BS
NonSPE
MS
SPE
LC-MS
MS
NonSPE
(Orig)
MS
NonSPE
BS
NonSPE
MS SPE LC-MS
Z-09-01-23-014 547.7
281.9 ±
57.6
335.3 ±
56.2
117.6 ±
25.4
410.87
Z-09-02-26-032 603.5
293.6 ±
97.9
362.5 ±
40.3
158.4 ±
32.9
500.46
Z-09-02-26-033 366.7 260.8 ± 0 265.5 ± 0
191.5 ±
4.3
358.43
Z-09-04-19-058 202.4
194.3 ±
20.0
108.2 ±
32.6
239.67
Z-09-04-24-065 13.9 bd
8.6 ±
1.1
bd 8.79 4904.9
1366.7 ±
71.7
1838.5 ±
371.7
906.1 ±
0
1828.9
Z-09-04-25-066 74.6 34.1 ± 5.0
55.6 ±
1.3
bd 4.92 12519.5
4483.0 ±
429.6
2800.0
± 325.5
6396.4
Z-09-04-26-067 15.6
16.1 ±
4.5
18.1 ±
3.4
12.6
± 1.9
16.56 1119.8
323.9 ±
17.0
1059.8
Z-09-04-26-068 49.8
54.5 ±
9.5
23.8 ±
2.0
106.84
Z-09-04-27-072 894.5
2884.0 ±
806.1
228.4 ±
109.2
2273.8
Z-09-04-28-076 10.3 bd
3.4 ±
.09
bd bd 20.0
7.3 ±
0.9
12.8 ±
1.2
10.0
± 0
7.75 7456.5
4345.0 ±
527.5
4903.4 ±
83.0
1817.3
± 227.0
4401.8
Z-09-04-30-078 9.4 bd bd bd 29.1 692.9
462.4 ±
45.9
108.8 ±
7.7
2275.5
110
The impact of SPE-cleaning of naturally positive samples destined for ELISA analysis
could be directly quantified in the simultaneous analysis of the stored samples (Table 3-3). The
SPE-cleaned urine samples averaged of 51 % lower DA concentrations than the non-SPE-
cleaned samples analyzed by the MS ELISA (ranged 19 % to 92 % lower). The SPE-cleaned urine
samples analyzed by the MS ELISA were consistently lower than the LC-MS measured DA
concentrations for the same sample. The MS ELISA SPE-cleaned samples averaged 67 % of the
LC-MS measured concentrations (range of 47 % to 95 % lower). These results were consistent
with the results obtained from the spike and recovery portion of this validation study in that
SPE-cleaning did not improve DA analysis of CSL body fluids using ELISA methods. The 50 %
methanol extract obtained from the SPE clean-up appeared to inhibit antibody performance
leading to underestimation of the DA concentrations.
Conclusions
The purpose of this study was to provide a comparison of DA concentrations reported in
CSL body fluids across methods, laboratories and datasets for a spectrum of matrices. The
amount, quality and type of body fluid available for DA analysis from an individual animal are
variable and highly dependent on the health status of the animal when brought into
rehabilitation facilities. In addition, animals suffering from chronic DA toxicosis could be
sickened at lower concentrations of DA in their body than animals afflicted with acute DA
toxicosis. Accurate determination of toxin concentration is useful for assessing these different
conditions and relating animal body burdens to the occurrence of toxin in local planktonic
environments. The platform and methodology used by a given laboratory will be determined
111
by the materials, equipment, cost and technical ability as well as the specific research goals of
the project. The availability of multiple platforms to a single laboratory is a rarity, and
therefore the ability to extrapolate results across platforms is essential.
The modified protocol described in the present study for the analysis of CSL body fluids
by an ELISA platform was validated as a protocol for analysis of DA concentrations. SPE-
cleaning was shown to be contraindicated for the removal of matrix effects stemming from the
different body fluid matrices because the 50% methanol extract was more inhibitive to
antibody performance than the individual fluid types. False positives were not observed using
either ELISA platform, regardless of sample type, using a 1:25 minimal dilution without a
methanol extraction step. Overall, the BS ELISA yielded accurate DA concentrations or slightly
overestimated DA concentrations, possibly due to the multiple epitopes targeted by the
polyclonal antibodies. Whether a slightly higher dilution of body fluid samples would have
overcome this inhibition was not tested. Overestimation of DA concentrations using the BS
ELISA to analyze body fluids was also noted in previous studies comparing BS ELISA measured
results of DA in rat serum and brain samples, relative to LC-MS measurements (Hesp et al.
2005). The MS ELISA slightly underestimated DA concentrations in general, possibly due to
greater sensitivity of the monoclonal antibody to the body fluid matrices, relative to the
polyclonal antibodies of the BS ELISA. However, the BS ELISA results were less reproducible
than MS ELISA results, with the BS ELISA values yielding more high standard deviations and CVs.
A total of 77 samples analyzed by the BS ELISA were discarded due to high CVs while only 6
samples analyzed by the MS ELISA were discarded. High CVs can be a consequence of operator
error during analysis (i.e. – pipette error, use of an un-homogenized sample, procedural error),
112
matrix impacts on antibody performance, and/or a manufacturing error. It is unknown if matrix
effects were solely responsible for the higher CVs observed in the BS ELISA analysis.
The linear relationship between the spiked DA concentrations and that measured by the
BS and MS ELISAs for the AF, CSF and serum samples, as well as the fact that both assays
generally correlated well with each other, may allow for matrix corrected assays to be
developed. This would involve applying a correction factor for each assay that adjusts the
average ELISA values to match those obtained by LC-MS. Testing of additional samples could
then be used to confirm whether the relationship holds across samples from different animals.
If so, both assays may be used to rapidly obtain reasonable estimates of DA concentrations in
AF, CSF and serum samples.
The accurate correlation of DA present in phytoplankton samples with unusual marine
mammal mortality events in a given area relies upon the positive identification of DA presence
in body fluids and the magnitude of the DA present. The ELISA and LC-MS methods described
in the present study have been shown to be capable of meeting these goals, without
introducing a significant risk of false positives. The platforms demonstrated relatively good
agreement (high R
2
values) with known DA concentrations added to CSL body fluid samples,
and the linearity observed when platform results were directly compared verified that the
magnitude of DA concentrations measured by each platform were comparable. Urine was
identified as a complicated matrix, most likely due to the high salt content, and further
investigation is needed to determine if additional sample manipulation and/or clean-up
procedures can reduce the interference observed in the urine results of the present study. It is
113
therefore recommended that DA concentrations for urine samples be an average of multiple
replicates, regardless of platform being used.
REFERENCES
Allen W (1934) Marine plankton diatoms of lower California in 1931. Bot Gaz 95:485-492
Allen W (1936) Occurrence of marine plankton diatoms in a ten-year series of daily catches in
southern California. Am J Bot 23:60-63
Bargu S, Goldstein T, Roberts K, Li C, Gulland F (2012) Pseudo-nitzschia blooms, domoic acid and
related California sea lion strandings in Monterey Bay, California. Mar Mammal Sci
28:237-253
Bargu S, Powell C, Coale S, Busman M, Doucette G, Silver M (2002) Krill: a potential vector for
domoic acid in marine food webs. Mar Ecol Prog Ser 237:209-216
Bargu S, Silver M, Goldstein T, Roberts K, Gulland F (2010) Complexity of domoic acid-related
sea lion strandings in Monterey Bay, California: Foraging patterns, climate events and
toxic blooms. Mar Ecol Prog Ser 418:213-222
Bates S, Bird C, de Freitas A, Foxall R, Gilgan M, Hanic L, Johnson G, McCulloch A, Odense P,
Pocklington R, Quilliam M, Sim P, Smith J, Subba Rao D, Todd E, Walter J, Wright J (1989)
Pennate diatom Nitzschia pungens as the primary source of domoic acid, a toxin in
shellfish from eastern Prince Edward Island, Canada. Can J Fish Aquat Sci 46:1203-1215
Bejarano A, Gulland F, Goldstein T, St Leger J, Hunter M, Schwacke L, Van Dolah F, Rowles T
(2008) Demographics and spatio-temporal signature of the biotoxin domoic acid in
California sea lions (Zalophus californianus) stranding records. Mar Mammal Sci 24:899-
912
Blanco J, Acosta C, Marino C, Muniz S, Martin H, Morono A, Correa J, Arevalo F, Salgado C
(2006) Depuration of domoic acid from different body compartments of the king scallop
Pecten maximus grown in raft culture and natural bed. Aquat Living Resour 19:257-265
Bland J, Altman D (1986) Statistical methods for assessing agreement between two methods of
clinical measurement. Lancet i:307-310
Brodie E, Gulland F, Greig D, Hunter M, Jaakola J, St Leger J, Leighfield T, Van Dolah F (2006)
Domoic acid reproductive failure of California sea lions (Zalophus californianus). Mar
Mammal Sci 22:700-707
Busse L, Venrick E, Antrobus R, Miller P, Vigilant V, Silver M, Mengelt C, Mydlarz L, Prezelin B
(2006) Domoic acid in phytoplankton and fish in San Diego, CA, USA. Harmful Algae
5:91-101
114
Del Rio R, Bargu S, Baltz D, Fire S, Peterson G, Wang Z (2010) Gulf menhaden (Brevoortia
patronus): A potential vector of domoic acid in coastal Louisiana food webs. Harmful
Algae 10:19-29
Fire S, Wang Z, Berman M, Langlois G, Morton S, Sekula-Wood E, Benitez-Nelson C (2010)
Trophic transfer of the harmful algal toxin domoic acid as a cause of death in a Minke
whale (Balaenoptera acutorostrata) stranding in southern California. Aquat Mammals
36:342-350
Fritz L, Quilliam M, Wright J, Beale A, Work T (1992) An outbreak of domoic acid poisoning
attributed to the pennate diatom Pseudo-nitzschia australis. J Phycol 28:439-442
Fryxell G, Villace M, Shapiro L (1997) The occurrence of the toxic diatom genus Pseudo-nitzschia
(Bacillariophyceae) on the West Coast of the USA, 1920-1996: A review. Phycologia
36:419-437
Goldstein T, Mazet J, Zabka T, Langlois G, Colegrove K, Silver M, Bargu S, Van Dolah F, Leighfield
T, Conrad P, Barakos J, Williams D, Dennison S, Haulena M, Gulland F (2008) Novel
symptomatology and changing epidemiology of domoic acid toxicosis in California sea
lions (Zalophus californianus): An increasing risk to marine mammal health. Proc R Soc B
275:267-276
Goldstein T, Zabka T, DeLong R, Wheeler E, Ylitalo G, Bargu S, Silver M, Leighfield T, Van Dolah
F, Langlois G, Sidor I, Dunn J, Gulland F (2009) The role of domoic acid in abortion and
premature parturition of California sea lions (Zalophus californianus) on San Miguel
Island, California. J Wildlife Dis 45:91-108
Gulland F, Haulena M, Fauquier D, Langlois G, Lander M, Zabka T, Duerr R (2002) Domoic acid
toxicity in California sea lions (Zalophus caliornianus): clinical signs, treatment and
survival. Vet Rec 150:475-480
Hall A, Frame E (2010) Evidence of domoic acid exposure in harbour seals from Scotland: A
potnetial factor in the decline in abundance? Harmful Algae 9:489-493
Hesp B, Harrison J, Selwood A, Holland P, Kerr D (2005) Detecton of domoic acid in rat serum
and brain by direct competitive enzyme-linked immunosorbent assay (cELISA). Anal
Bioanal Chem 383:783-786
Kleivdal H, Kristiansen S, Nilsen M (2007) Single-Laboratory validation of Biosense direct
competitive enzyme-linked immunosorbent assay (ELISA) for determination of domoic
acid toxins in shellfish. J AOAC Int 90:1000-1010
Leandro L, Rolland R, Roth P, Lundholm N, Wang Z, Doucette G (2010) Exposure of the North
Atlantic right whale Eubalaena glacialis to the marine algal biotoxin, domoic acid. Mar
Ecol Prog Ser 398:287-303
115
Lefebvre K, Dovel S, Silver M (2001) Tissue distribution and neurotoxic effects of domoic acid in
a prominent vector species, the northen anchovy Engraulis mordax. Mar Biol 138:693-
700
Lefebvre K, Powell C, Busman M, Doucette G, Moeller P, Silver J, Miller P, Hughes M, Singaram
S, Silver M, Tjeerdema R (1999) Detection of domoic acid in northern anchovies and
California sea lions associated with an unusual mortality event. Nat Toxins 7:85-92
Lefebvre K, Robertson A, Frame E, Colegrove K, Nance S, Baugh K, Wiedenhoft H, Gulland F
(2010) Clinical signs and histopathology associated with domoic acid poisoning in
northern fur seals (Callorhinus ursinus) and comparison of toxic detection methods.
Harmful Algae 9:374-383
Litaker R, Stewart T, Eberhart B, Wekell J, Trainer V, Kudela R, Miller P, Roberts A, Hertz C,
Johnson T, Frankfurter G, Smith G, Schnetzer A, Schumacker J, Bastian J, Odell A,
Gentien P, Le Gal D, Hardison D, Tester P (2008) Rapid enzyme-linked immunosorbent
assay for detection of the algal toxin domoic acid J Shellfish Res 27:1301-1310
McManus M, Kudela R, SIlver M, Steward G, Donaghay P, Sullivan J (2008) Cryptic blooms: Are
thin layers the missing connection? Estuar Coast 31:396-401
Olson M, Lessard E (2010) The influence of the Pseudo-nitzschia toxin, domoic acid, on
microzooplankton grazing and growth: A laboratory and field experiment. Harmful
Algae 9:240-247
Perl T, Bedard L, Kosatsky T, Hockin J, Todd E, Remis R (1990) An outbreak of toxic
encephalopathy caused by eating mussels contaminated with domoic acid. New Engl J
Med 322:1775-1780
Quay J, Hayashi K, Kudela R Watch wha you watch for: Degradation of particulate domoic acid
(pDA) field samples. Proc Sixth Symposium of Harmful Algae in the US
Ramsdell J, Zabka T (2008) In utero domoic acid toxicity: a fetal basis to adult disease in the
California sea lion (Zalophus californianus). Mar Drugs 6:262-290
Rines J, Donaghay P, Dekshenieks M, Sullivan J, Twardowski M Thin layers and camouflage:
Hidden Pseudo-nitzschia spp. (Bacillariophyceae) populations in a fjord in the San Juan
Islands, Washington, USA. Mar Ecol Prog Ser 225:123-137
Rines J, McFarland M, Donaghay P, Sullivan J (2010) Thin layers and species-specific
characterization of the phytoplankton community in Monterey Bay, California, USA.
Cont Shelf Res 30:66-80
Scholin C, Gulland F, Doucette G, Benson S, Busman M, Chavez F, Cordaro J, DeLong R, De
Vogelaere A, Harvey J, Haulena M, Lefebvre K, Lipscomb T, Loscutoff S, Lowenstine L,
Marin III R, Miller P, McLellan W, Moeller P, Powell C, Rowles T, Silvagni P, Silver M,
116
Spraker T, Trainer V, Van Dolah F (2000) Mortality of sea lions along the central
California coast linked to a toxic diatom bloom. Nature 403:80-84
Seubert E, Trussell S, Eagleton J, Schnetzer A, Cetinic I, Lauri P, Jones B, Caron D (2012) Algal
toxins and reverse osmosis desalination operations: Laboratory bench testing and field
monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid. Water Res 46
Sierra Beltran A, Palafox-Uribe M, Grajales-Montiel J, Cruz-Villacorta A, Ochoa J (1997) Sea bird
mortality at Cabo San Lucas, Mexico: Evidence that toxic diatom blooms are spreading.
Toxicon 35:447-453
Thomas K, Harvey J, Goldstein T, Barakos J, Gulland F (2010) Movement, dive behavior and
survival of California lions (Zalophus californianus) posttreatment for domoic acid
toxicosis. Mar Mammal Sci 26:36-52
Torres de la Riva G, Johnson C, Gulland F, Langlois G, Heyning J, Rowles T, Mazet J (2009)
Association of an unusual marine mammal mortality event with Pseudo-nitzschia spp.
blooms along the southern California coastline. J Wildlife Dis 45:109-121
Wang Z, King K, Ramsdell J, Doucette G (2007) Determination of domoic acid in seawater and
phytoplankton by liquid chromatography-tandem mass spectrometry. J Chromatogr A
1163:169-176
Wang Z, Maucher-Fuquay J, Fire S, Mikulski C, Haynes B, Doucette G, Ramsdell J (2012)
Optimization of solid-phase extraction and liquid chromatography-tandem mass
spectrometry for the determination of domoic acid in seawater, phytoplankton and
mammalian fluids and tissues. Anal Chim Acta 715:71-79
Westgard J, Hunt M (1973) Use and interpretation of common statistical tests in method-
comparison studies. Clin Chem 19:49-57
Work T, Barr B, Beale A, Fritz L, Quilliam M, Wright J (1993) Epidemiology of domoic acid
poisoning in brown pelicans (Pelecanus occidentalis) and Brandt's cormorants
(Phalacrocorax pencillatus) in California. J Zoo Wildlife Med 24:54-62
Wright J, Boyd R, de Freitas A, Falk M, Foxall R, Jamieson W, Laycock M, McCulloch A, McInnes
A, Odense P, Pathak V, Quilliam M, Ragan M, Sim P, Thibault P, Walter J (1989)
Identification of domoic acid, a neuroexcitatory amino acid, in toxic mussels from
eastern Prince Edward Island. Can J Chem 67:481-490
117
Chapter 4: Phytoplankton Community Response to the Addition of Treated Sewage Effluent
ABSTRACT
The Orange County Sanitation District diverted flow of secondarily treated effluent from
a discharge pipe located 8.0 km offshore at 60 m depth to a pipe located 1.6 km from shore at
17 m depth for three weeks in September of 2012. Two incubation experiments were
performed to examine the influence of treated effluent at various dilutions on natural, coastal
phytoplankton communities, the first initiated a week prior to the diversion (‘Pre-Diversion’)
and the second initiated a week after the start of the diversion (‘Mid-Diversion’). The overall
community response observed in both experiments following effluent addition was an increase
in diatom and picoeukaryote abundances, a decrease in picophotocyanobacteria and a
dramatic increase in heterotrophic bacteria abundance. The 1:10 effluent additions yielded
significant increases in chlorophyll a concentrations, although the Pre-Diversion 1:10
experiments exhibited a lag in response to effluent addition. Bray Curtis similarity comparisons
of the Pre- and Mid-Diversion experiment results revealed the communities present at the start
of each experiment were only 25 % similar. These differences in community composition
suggest the Pre-Diversion communities may have been inhibited by the 170 µM addition of
ammonium while the Mid-Diversion community that did not exhibit a lag in response to 1:10
effluent addition was not inhibited by the 219 µM addition of ammonium. The domoic acid
producing diatom Pseudo-nitzschia was present throughout both experiments, however domoic
acid production was only detected in the Mid-Diversion experiment. The highest concentration
of domoic acid measured, 0.42 ± 0.057 µg/L, coincided with phosphate and silicate
118
concentrations below the detection limit of the method, suggesting limitation by these
macronutrients.
INTRODUCTION
The impact of anthropogenic influence on the marine environment is readily apparent in
the enhanced nutrient loading of coastal waters (Smith et al. 1999, Paerl et al. 2002, Smith et al.
2006) and destruction of coastal habitats by sea level rise (Sahagian et al. 1994, Nicholls et al.
1999), ocean acidification (Fabry et al. 2008, Hoegh-Guldberg et al. 2010) and the expansion of
hypoxic and anoxic zones (Chan et al. 2008, Stramma et al. 2010, Deutsch et al. 2011). The
recent observed global increase in severity and incidence of harmful algal blooms (HABs) has
been correlated with the human driven processes of global climate change (Moore et al. 2008,
Fu et al. 2012) and eutrophication (Paerl 1997, Anderson et al. 2002, Glibert et al. 2006). Not
all HAB events can be directly linked to anthropogenic influences (Kudela et al. 2008, Davidson
et al. 2012) and their occurrence in southern California is not a modern phenomenon (Torrey
1902, Allen 1946).
Diatoms of the genus Pseudo-nitzschia are common to southern California waters and
HAB events attributed to domoic acid (DA) producing members of this genus occur most
commonly in the spring (Caron et al. 2010, Seubert et al. 2012, Seubert et al. 2013). DA is a
potent neurotoxin that disrupts proper functioning of the hippocampus in higher mammals (Xi
& Ramsdell 1996, Hampson & Manalo 1998) and DA blooms in southern California are
accompanied by the sickening and death of scores of marine mammals and birds through DA
toxicosis (Schnetzer et al. 2007, Torres de la Riva et al. 2009, Fire et al. 2010). These blooms
119
have been linked to upwelling events in the area (Seubert et al. 2013, Schnetzer et al. Accepted)
and anthropogenic nutrient loading is not currently considered to be a factor in bloom
development (Lewitus et al. 2012). The growth rate and DA production of Pseudo-nitzschia has
been shown to vary with nitrogenous sources (Bates et al. 1993, Hillebrand & Sommer 1996,
Howard et al. 2007, Cochlan et al. 2008, Thessen et al. 2009) and a previous study stimulated
Pseudo-nitzschia growth in laboratory experiments with varying dilutions of secondarily treated
effluent obtained from a Canadian treatment plant (Pan & Subba Rao 1997). Therefore any
alteration to nutrient loading in the southern California area must consider any impact to
Pseudo-nitzschia growth and/or DA production.
The over 22 million people living in southern California comprise roughly 60 % of the
state’s total community. Nineteen municipal wastewater treatment facilities are in operation
to support this large and growing community, and three of the four largest facilities operate in
the coastal ocean off of Los Angeles and Orange counties, each averaging treated effluent
discharges of over 100 million gallons per day. These effluent discharges are relatively constant
while the natural nutrient loading processes of upwelling and storm-water runoff events are
seasonal and stochastic. Upwelling most commonly occurs in the spring, when upwelling
favorable winds dominate, while sporadic rain events occur during November to March.
Previous studies have focused on upwelling events as the principal process stimulating primary
production in the southern California coastal region (Eppley et al. 1978, Eppley et al. 1979), but
recent research has demonstrated effluent discharges may drive productivity, particularly
during periods of upwelling relaxation (Howard et al. in revision). The magnitude of nutrient
concentrations introduced to the coastal ocean from each of these processes is dissimilar and
120
the dominant nitrogenous source in each is also disparate, with nitrate the primary nitrogenous
form in upwelled water and ammonium the primary nitrogenous form in effluent. Since
nitrogenous source preference differs among many phytoplankton species (Goldman & Glibert
1982, Howard et al. 2007), there is a need to understand how phytoplankton community
structure can be influenced by the presence of a nutrient source overwhelmingly dominated by
ammonium, such as treated effluent.
The Orange County Sanitation District (OCSD) in Fountain Valley, CA, USA, averages a
discharge of over 200 million gallons per day of secondarily treated effluent. During secondary
treatment, organic matter is biologically oxidized and greater than 85 % of the total suspended
solids and biochemical oxygen demand is removed. Inorganic nutrients that are a product of
the biological oxidation can stimulate autotrophic growth of phytoplankton when discharged
into the coastal ocean. Ocean outfall pipes for effluent discharge are designed to minimize
their impact by discharging far from shore and below the euphotic zone, and initial dilution of
effluent is increased through the use of multiple diffuser ports. The main discharge pipe used
by OCSD since 1972 is located 8.0 km from shore at a depth of 60 m, has a diameter of 3.1 m,
and it is estimated to obtain an initial dilution of 1:350. In September 2012, OCSD diverted
their primary effluent discharge from the 8.0 km pipe to an older pipe located 1.6 km from
shore in order to perform necessary maintenance and repairs to the 8.0 km pipe. The older
pipe discharges at 17 m and it is estimated to obtain an initial dilution of 1:36. The purpose of
the present study was to understand and document potential impact of the diversion of treated
effluent to a shallow, near-shore environment on phytoplankton standing stock and community
composition. Towards this end, two incubation experiments were performed to examine the
121
influence of various dilutions of treated effluent on a natural, coastal, phytoplankton
community collected one a week prior to the planned OCSD diversion, and another performed
a week following the start of the diversion event.
MATERIALS AND METHODS
Study area and experimental design. Incubation experiments were performed in September
2012, utilizing natural phytoplankton communities collected offshore of Orange County, CA,
USA (Figure 4-1). The first experiment occurred one week prior to the planned diversion of the
OCSD effluent discharge and the second occurred one week following the start of the diversion
event. The ‘Pre-Diversion’ experiment was initiated on September 6, 2012, and evaluated
differences between a surface and deep chlorophyll maximum (DCM) phytoplankton
community response to the addition of treated effluent. Water was collected in Niskin bottles
as a component of conductivity-temperature-depth (CTD) casts from 5 m to 100 m. The depth
of the DCM was determined utilizing real-time in situ fluorometry data (33°30'40" N, 118° 3'32"
W; Figure 4-1). The surface community was collected at 5 m and the DCM community was
collected at 40 m. The ‘Mid-Diversion’ experiment was initiated on September 20, 2012, and
examined the influence of treated effluent addition on a DCM community and also the
potential for effluent inhibition of growth and for secondary limitation of growth due to
micronutrient limitation. Water was collected in the same manner as the Pre-Diversion
experiment, at a station slightly inshore of the first experiment in order to collect a DCM
community containing high biomass, at 17 m (33°30'40" N, 118° 3'32" W; Figure 4-1). Sample
water for the incubation experiments was collected in acid-washed (5 % HCl) 20 L carboys
122
directly from the Niskin bottles, taking care to minimize bubbling and agitation during the
transfer. The carboys were protected from light and kept cool during transport to shore for
experimental setup.
Figure 4-1. Location of the study area in Orange County, California, with the locations where water was collected for the Pre-
Diversion and Mid-Diversion effluent incubation experiments are marked in white.
The incubation experiments were performed in acid-washed, 4-L polycarbonate bottles
that were incubated at ambient water temperature with neutral density screening
approximating 50 % ambient light levels, in Los Angeles Harbor at the Southern California
Marine Institute in San Pedro, CA. The core treatments tested in triplicate for both the Pre- and
Mid-Diversion experiments were (1) a true control with no additions, (2) a Milli-Q® control, and
OCSD effluent additions representing dilutions of (3) 1:10, (4) 1:100 and (5) 1:1000. The
123
volumes of all treatments were kept consistent and the amount of freshwater (effluent or Milli-
Q®) added reduced the salinity in the treatment bottles from approximately 35 ppt to 32 ppt.
The Milli-Q® control was employed in order to determine any community changes that
occurred solely due to the reduction in salinity. The Mid-Diversion experiment included
additional treatments to examine the role of potentially inhibitory substances present in
effluent and the impact of micronutrient limitation. The presence of inhibitory substances was
tested by adding an ‘effluent mimic’ at a 1:10 dilution for comparison to the 1:10 addition of
OCSD effluent. The effluent mimic prepared in the laboratory with Milli-Q® water targeted a
final concentration of 2870 µM ammonium (NH
4
), 63.5 µM phosphate (PO
4
), 10 µM silicate
(SiO
3
), 500 pM of vitamins B
1
and B
7
, 5000pM of vitamin B
12
and a 1:100 dilution of the trace
metal solution used in L1 enriched seawater medium (Guillard & Hargraves 1993). The
macronutrient concentrations in the mimic were based on concentrations measured in 2010 by
Howard et al. (in revision). The impact of micronutrient limitation was investigated in two
ways. First, a dilution of 1:10 OCSD effluent was supplemented with vitamins and trace metals
for comparison to the 1:10 addition of OCSD effluent only. Second, vitamins and trace metals
were added to another treatment for comparison to the controls. The vitamins and trace
metals that supplemented the 1:10 OCSD effluent addition and in the micronutrient only
addition treatment were in the same concentrations as the effluent mimic solution.
Sample collection and processing. Samples were taken for analysis at the initiation of each
experiment, and the treatment bottles were subsampled approximately every 24 hours during
the first three days of the experiments for chlorophyll a, particulate DA (pDA), dissolved DA
124
(dDA), flow cytometry, preserved cell counts and dissolved inorganic nutrient analysis.
Subsamples were taken again at the close of the experiment. The Pre-Diversion experiment
was ended after 7 days and the Mid-Diversion experiment was ended after 6 days of
incubation. Samples for the analysis of dissolved inorganic nutrients were collected from the
initial seawater sample, the effluent used in the dilution experiment, and from each treatment
bottle during the first three days of incubation. Approximately 20 mL of sample was syringe
filtered through a 0.2 µm syringe filter and dispensed into acid-washed plastic scintillation vials
and frozen at -20° C until analysis. Nutrient analysis for concentrations of nitrate plus nitrite
(NO
3
+ NO
2
; 0.2 µM limit of detection), NO
2
(0.1 µM limit of detection), NH
4
(0.1 µM limit of
detection), PO
4
(0.1 µM limit of detection) and SiO
3
(1.0 µM limit of detection) were performed
at the Analytical Lab at the Marine Sciences Institute at the University of California, Santa
Barbara on a QuikChem 8000 flow injection analyzer (Lachat Instruments; Loveland, CO) with ±
5 % precision. Samples for chlorophyll a and pDA concentrations were collected in duplicate by
vacuum filtration of 10 to 100 mL and 200 mL samples, respectively, onto grade F glass fiber
filters (Sterlitech, Kent, WA) and frozen at -20° C until analysis. The volume filtered for
chlorophyll a analysis was reduced as needed from 100 mL when concentrations exceeded the
calibrated range of the fluorometer. Filters were extracted in 4 mL of 100 % acetone for 24
hours at -20° C and analyzed on a calibrated Trilogy® laboratory fluorometer with the
chlorophyll a non-acidification module (7200-046; Turner Designs Inc., Sunnyvale, CA). Filters
for pDA were extracted in 3 mL of 10 % methanol and analyzed using the Mercury Science Inc.
DA Enzyme-Linked ImmunoSorbent Assay (ELISA; Durham, NC) following the methods described
in Seubert et al. (2012) with a 0.02 µg /L limit of detection. Flow cytometry samples for the
125
abundances of heterotrophic bacteria, picoeukaryotes, Prochlorococcus spp. and
Synechococcus spp. were pre-screened through an 80 µm Nitex fabric and preserved with a
final concentration of 1 % formalin, flash frozen in liquid nitrogen and stored at -80° C until
analysis on a four-color, dual-laser FACSCalibur flow cytometer (BD Biosciences, San Jose, CA).
Whole seawater samples for phytoplankton community composition were preserved with a
final concentration of 1 % formalin, stored in the dark at 4° C and examined by inverted light
microscopy at 400x after settling in Utermöhl chambers for 24 hours (Utermohl 1958).
Organisms greater than 10 µm in size were identified first by group and then to genus and
species, when possible. Cells identified as Pseudo-nitzschia were divided into size classes based
upon frustule width, as conclusive species identification is not possible without electron
microscopy (Hasle et al. 1996, Hasle & Syvertsen 1997) or use of molecular methods (Scholin et
al. 1996, Hubbard et al. 2008). The P. seriata size class is identified as having frustule widths
greater than 3 µm and the P. delicatissima size class has frustule widths smaller than 3 µm
(Hasle & Syvertsen 1997).
Statistical analysis and comparisons of the community composition data obtained from
the settled cell counts from each experiment were performed using PRIMER v6 (Clarke & Gorley
2006). Species composition data was log (x+1) transformed prior to Bray-Curtis similarity
calculations. The similarity data was plotted using the Cluster and Multi-Dimensional Scaling
(MDS) functions available in the PRIMER v6 software (Clarke 1993). The different treatments
from within one experiment were compared to each other and the entire results of each
experiment were compared directly. The direct comparison of the Pre-Diversion and Mid-
126
Diversion experiments were over the first three days of the experiment as the end day of each
experiment was different (7 days in the Pre-Diversion and 6 days in the Mid-Diversion).
RESULTS
Pre-Diversion experiment. The Pre-Diversion experiment investigated the response of both a
surface and DCM community to the addition of OCSD effluent. The reasoning for examining
these depths is that the 8.0 km pipe discharges at 60 m depth and could potentially influence a
DCM phytoplankton community while the 1.6 km pipe discharges at 17 m depth and can
potentially influence a surface community. In addition, an upwelling event can transport an
offshore DCM community to the near coast euphotic zone, exposing it to higher irradiances and
effluent discharges located in shallow water. Both treatments were incubated at surface
irradiance and the DCM community investigated experienced higher nutrient concentrations
associated with the effluent additions as well as increased irradiance (Table 4-1).
Table 4-1. Nutrient concentrations measured in the initial whole seawater sample and the nutrient concentrations of the
effluent sample used in the Pre-Diversion (9/6/2012) and Mid-Diversion (9/20/2012) experiments. The N:P ratio is reported by
atoms and samples below the method detection limit are reported as “bd”.
Date Source Type NH
4
(µM) NO
3
(µM) NO
2
(µm) PO
4
(µM) SiO
3
(µM)
N:P
9/6/2012
Surface Seawater 1.39 0.23 bd 0.20 4.19 8.10
DCM Seawater 3.08 2.18 0.10 0.22 4.78 24.4
OCSD Effluent 1,690 336 104 22.1 585 96.4
9/20/2012
DCM Seawater 0.25 bd bd 0.10 bd 2.50
OCSD Effluent 2,190 421 59.3 22.3 594 119
Effluent Mimic 2,870 - - 63.5 10 45.2
127
Pre-Diversion surface. Chlorophyll a concentrations in the 1:10 effluent treatment were lower
than the 1:100 and 1:1000 treatments during the first three days of incubation, but then
increased dramatically to 210 ± 35.3 µg/L by day seven, the highest of all treatments and all
time points (Figure 4-2A). The increase in chlorophyll a concentration in the 1:10 treatment
was a >1,000-fold increase over the initial concentration, 0.20 ± 0.006 µg/L. Chlorophyll a
concentrations in the 1:100 and 1:1000 treatments did not show a lag increase following the
addition of effluent as seen in the 1:10 treatment. The 1:100 treatment increased 110-fold
over initial concentrations, reaching 22.0 ± 2.40 µg/L at experiment close. The 1:1000
treatment increased 16.3-fold in chlorophyll a concentrations over the initial value during the
first three days of incubation. The chlorophyll a concentration decreased in that treatment
between day three and day seven, with the final chlorophyll a concentration of 0.92 ± 0.18 µg/L
only a 4.6-fold increase from initial concentrations.
The phytoplankton community in the surface at the time of collection was dominated by
diatoms, and it remained dominated by diatoms (85 % of community or higher) in all
treatments throughout the length of the incubation experiment. Diatom abundance in the 1:10
treatment of 33,000 ± 12,000 cells/mL on day seven was the highest measured of all treatments
and time points (Figure 4-2B). The 1:100 and 1:1000 effluent treatments yielded the highest
concentrations of diatoms during the first three days of the experiment compared to the 1:10
treatment and the controls, mirroring results observed for chlorophyll a concentrations.
Picoeukaryote abundance reached 1.0 x 10
6
± 5.7 x 10
5
cells/mL at the end of the experiment in
the 1:10 treatment (Figure 4-2C). The 1:100 and 1:1000 treatments had concentrations of
picoeukaryotes higher than the 1:10 treatment during the first three days of the incubation,
128
however the two treatments decreased in picoeukaryote concentration between day three and
day seven.
129
Figure 4-2. Chlorophyll a concentrations (A) and the abundances of diatoms (B), picoeukaryotes (C), Prochlorococcus spp. (D),
Synechococcus spp. (E) and heterotrophic bacteria (F) measured during the Pre-Diversion effluent addition experiment with
surface water are plotted separately for each treatment. Results from the true controls are plotted with open circles and
dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with black circles and solid black lines,
1:100 effluent additions with dark grey circles and solid dark grey lines and 1:1000 effluent additions with light grey circles and
solid light grey lines.
130
Abundances of picophotocyanobacteria, Prochlorococcus and Synechococcus, did not
show dramatic, positive responses as the microplanktonic and picoplanktonic eukaryotic
phytoplankton (Figures 4-2D and 4-2E). Heterotrophic bacteria concentrations are highest in
the 1:10 effluent treatment throughout the experiment, contrary to the initial delay in the 1:10
effluent treatment concentrations of chlorophyll α, diatoms, picoeukaryotes, Prochlorococcus
and Synechococcus observed from day one to day three (Figure 4-2F).
Nutrients in the effluent were highly elevated and the N:P ratios markedly different than
the coastal waters from which the experiment water was collected (Table 4-1). The surface
water used in the incubation experiments had a N:P ratio of 8.10 and NH
4
was the dominant
nitrogen source at 1.39 µM. The effluent sample had a N:P ratio of 96.4 with NH
4
the dominant
nitrogen source at 1,690 µM, NO
3
and NO
2
are also present in high concentrations at 336 and
104 µM, respectively. The initial nutrient concentrations in the effluent dilution treatments
were calculated according to the effluent addition to which the nutrient concentrations
measured in the initial water sample were added. Concentrations of all macronutrients in the
1:10 treatment remained relatively constant throughout the experiment (Figure 4-3A-E).
Ammonium concentrations in the 1:100 treatment did not noticeably decrease from the initial
18.3 µM until day three when it was measured at 10.7 ± 1.12 µM. The PO
4
concentration was
the only other nutrient to decrease noticeably, reaching a concentration below the detection
limit of the method (0.1 µM) by day three. NO
3
, NO
2
and SiO
3
did not show obvious decreases
in concentrations over the first three days of the incubation in the 1:100 treatment. The SiO
3
concentration in the Milli-Q® control and the effluent addition treatments that were diluted in
131
Milli-Q® prior to dosing the experiment bottles (1:100 and 1:1000 treatments) were higher than
expected, suggesting the Milli-Q® water may have contained SiO
3
.
Figure 4-3. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured during the Pre-Diversion
effluent addition experiment with surface water are plotted separately for each treatment. Results from the true controls are
plotted with open circles and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with
black circles and solid black lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and 1:1000 effluent
additions with light grey circles and solid light grey lines. The method detection limit for each nutrient analyte is marked on the
graphs with a dotted light grey line.
Pre-Diversion DCM. The chlorophyll a concentrations in the 1:10 treatment decreased
significantly during the first three days of the Pre-Diversion experiment conducted with the
DCM phytoplankton community (Figure 4-4A). The highest concentration was measured in this
treatment on day seven, 175 ± 21.9 µg/L, only a 273-fold increase over the initial chlorophyll a
concentration of 0.64 µg/L compared to >1,000-fold increase seen in the 1:10 treatment with
surface water. The 1:100 treatment had the second highest chlorophyll a concentration on day
seven with 54.0 ± 4.8 µg/L, an 84.4-fold increase over the initial chlorophyll a concentration.
132
The 1:1000 treatment had a 11.3-fold increase over the initial value, with a chlorophyll a
concentration on day seven of 7.21 ± 2.18 µg/L.
The phytoplankton community present in the DCM at the start of the experiment was
dominated by diatoms, approximately 70 %, and increased in abundance in all treatments to
reach virtually 100 % dominance (Figure 4-4B). The 1:10 treatment had the lowest diatom
concentration of any treatment during the first three days of the experiment until increasing by
day seven to 19,000 ± 11,000 cells/mL. The 1:100 treatment had the highest abundance of
diatoms at day seven at 26,000 ± 3,500 cells/mL. The picoeukaryotes had a similar trend in
abundance between the treatments as seen in the diatom concentrations, with the 1:10
treatment having the lowest concentration during the first three days of incubation (Figure 4C).
The highest concentration of picoeukaryotes was achieved in the 1:10 treatment on day seven
with a value of 1.1 x 10
6
± 6.2 x 10
5
cells/mL, which is also the highest picoeukaryote
concentration measured in the Pre-Diversion community experiments.
Abundances of Prochorococcus and Synechococcus did not exhibit positive responses to
the addition of effluent, contrary to the stimulatory results observed in the planktonic
eukaryotes (Figures 4-4D and 4-4E). The highest concentration of heterotrophic bacteria was
measured in the 1:10 treatment, reaching a final concentration of 1.6 x 10
7
± 3.3 x 10
6
cells/mL
(Figure 4-4F). Bacterial abundances reached in the experiment with a community from the
DCM are lower than those measured in the surface water treatments, in spite of the starting
concentrations in the initial communities being similar, 3.9 x 10
6
present in the DCM versus 4.0
x 10
6
present in the surface.
133
Figure 4-4. Chlorophyll a concentrations (A) and the abundances of diatoms (B), picoeukaryotes (C), Prochlorococcus spp. (D),
Synechococcus spp. (E) and heterotrophic bacteria (F) measured during the Pre-Diversion effluent addition experiment with
water from the DCM are plotted separately for each treatment. Results from the true controls are plotted with open circles
and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with black circles and solid black
lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and 1:1000 effluent additions with light grey circles
and solid light grey lines.
134
Nutrient concentrations in the DCM water collected for the experiment were slightly
above concentrations in the surface water collected, but the concentrations and N:P ratio of
the effluent was still markedly different (Table 4-1). The dominant nitrogen source in the DCM
was NH
4
at a concentration of 3.08 µM, followed by 2.18 µM of NO
3
and 0.10 µM of NO2. The
PO
4
and SiO
3
concentrations of 0.22 and 4.78 µM, respectively, were similar to the
concentrations measured in the surface water. Concentrations measured during the first three
days of the incubation experiment remained relatively consistent without a noticeable
decrease, and the Milli-Q® water used in the Milli-Q® controls and effluent dilutions may have
contained SiO
3
(Figure 4-5).
Figure 4-5. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured during the Pre-Diversion
effluent addition experiment with water from the DCM are plotted separately for each treatment. Results from the true
controls are plotted with open circles and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent
additions with black circles and solid black lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and
1:1000 effluent additions with light grey circles and solid light grey lines. The method detection limit for each nutrient analyte
is marked on the graphs with a dotted light grey line.
135
Mid-Diversion experiment. The chlorophyll a concentration of 4.10 ± 0.61 µg/L in the water
collected for the Mid-Diversion experiment was greater than the surface (0.20 ± 0.006 µg/L)
and DCM (0.64 µg/L) communities sampled for the Pre-Diversion experiments. Both the true
and Milli-Q® controls decreased in chlorophyll a concentration throughout the experiment with
final concentrations of 0.67 ± 0.09 and 0.76 ± 0.08 µg/L in the true and Milli-Q® controls,
respectively (Figure 4-6A). The highest chlorophyll a concentration was measured on the sixth
day in the 1:10 treatment, 61.2 ± 1.69 µg/L, a 14.9-fold increase over the initial concentration.
This concentration is lower than the highest concentration seen in the Pre-Diversion
experiment, 210 ± 35.3 µg/L, both in absolute concentration and in the amount fold increase, a
14.9-fold increase versus a >1,000-fold increase observed in the Pre-Diversion experiment with
surface water.
Diatoms dominated the DCM community sampled for the experiment, comprising nearly
100 % of the community throughout the experiment (Figure 4-6B). Following an initial increase
in all treatments, diatoms decreased between day three and the close of the experiment at day
six. Picoeukaryote abundances were the highest in the 1:10 treatment, reaching 8.6 x 10
4
± 9.1
x 10
3
cells/mL on day six (Figure 4-6C). The 1:100 and 1:1000 treatments decreased in
picoeukaryote abundance between day three and experiment close on day six.
Abundances of Prochlorococcus and Synechococcus decreased in all treatments (Figure
4-6D and 4-6E) and heterotrophic bacteria were the highest in the 1:10 treatment throughout
the incubation period (Figure 4-6F). The lack of response by the picocyanobateria compared to
the positive responses observed in the microplanktonic and picoplanktonic eukaryotes, is
similar to the results obtained in the Pre-Diversion experiments.
136
Figure 4-6. Chlorophyll a concentrations (A) and the abundances of diatoms (B), picoeukaryotes (C), Prochlorococcus spp. (D),
Synechococcus spp. (E) and heterotrophic bacteria (F) measured during the Mid-Diversion effluent addition experiment are
plotted separately for each treatment. Results from the true controls are plotted with open circles and dashed lines, Milli-Q®
controls with open circles and dotted lines, 1:10 effluent additions with black circles and solid black lines, 1:100 effluent
additions with dark grey circles and solid dark grey lines and 1:1000 effluent additions with light grey circles and solid light grey
lines.
137
The nutrient concentrations in the DCM water collected for the Mid-Diversion
experiment were lower in concentration than either Pre-Diversion water sample (Table 4-1).
The dominant form of nitrogen was NH
4
with a concentration of 0.25 µM while NO
3
and NO
2
were below the method detection limit (0.2 and 0.1 µM, respectively) and the resulting N:P
ratio was 2.50. The OCSD effluent collected for the Mid-Diversion experiment had similar PO
4
and SiO
3
concentrations as the sample collected for the Pre-Diversion, but differed most
noticeably in the NH
4
and NO
2
concentrations. The NH
4
concentration of 2,190 µM was 30 %
higher than the Pre-Diversion effluent and the NO
2
concentration of 59.3 µM was 43 % lower
than the Pre-Diversion effluent. The effluent mimic that was prepared in the laboratory based
upon measurements made in 2010 was higher in concentration in NH
4
and PO
4
than the Mid-
Diversion effluent, but the N:P ratio was significantly lower, 45.2 versus 119 in the effluent
sample.
Macronutrient concentrations decreased in all treatments, in contrast to the relatively
stable concentrations measured in the effluent treatments of the Pre-Diversion experiments
(Figure 4-7). The 1:100 treatment had the most remarkable changes in macronutrient
concentrations; PO
4
was below detection by the second day of incubation and SiO
3
was below
detection by the third day. The decreases in macronutrient concentration in the 1:100
treatment by day three ranged from 40 to 85 % from initial values whereas the decreases seen
in the surface and DCM 1:100 treatments of the Pre-Diversion experiment ranged from 0 to 42
%.
138
Figure 4-7. Nutrient concentrations of NH
4
(A), NO
3
(B), NO
2
(C), PO
4
(D), and SiO
3
(E) measured during the Mid-Diversion
effluent addition experiment are plotted separately for each treatment. Results from the true controls are plotted with open
circles and dashed lines, Milli-Q® controls with open circles and dotted lines, 1:10 effluent additions with black circles and solid
black lines, 1:100 effluent additions with dark grey circles and solid dark grey lines and 1:1000 effluent additions with light grey
circles and solid light grey lines. The method detection limit for each nutrient analyte is marked on the graphs with a dotted
light grey line.
Pre-Diversion versus Mid-Diversion community composition. Results from the preserved
whole seawater cell counts were used in Bray Curtis Similarity calculations and MDS plots
comparing the communities in the Pre-Diversion surface, Pre-Diversion DCM and the Mid-
Diversion experiments. Comparisons were made within results from each experiment and
subsequently between all three experiments.
The Pre-Diversion surface experiment was dominated by the Pseudo-nitzschia
delicatissima size class in both the initial sample and throughout the experiment in all
treatments except the 1:10 effluent addition. The P. delicatissima size class comprised only 16
% of the 1:10 treatment community at its highest abundance, and the diatom Cylindrotheca
spp. was instead dominant throughout. The P. seriata size class was also present throughout
139
the experiment in all treatments, although typically below 10 % of the community. Samples
collected for pDA were analyzed for all treatments and time points but failed to be above the
detection limit of the method (0.02 µg/L). The Bray Curtis Similarity comparison of
phytoplankton communities in the Pre-Diversion surface treatments revealed the 1:10
treatment community sampled after one day of incubation was less than 40 % similar to any
other sample collected during the experiment (Figure 4-8A). The communities present in the
1:100 and 1:1000 treatments ranged from 60 to 80 % similar throughout the experiment.
The DCM community present at the start of the Pre-Diversion experiment was
dominated by the diatom Navicula spp. and the P. delicatissima size class was below the
detection ability of the settled counts (3 cells/mL). The 1:10 treatment had low abundances of
the P. delicatissima size class throughout the experiment, ranging from 0 to 13 % of the
community. By the end of the experiment the P. delicatissima size class was the dominant
member of the community in both the true and Milli-Q® controls and effluent treatments of
1:100 and 1:1000, comprising 70 % or more of the community. The P. seriata size class was a
minor contributor to the experiment community, akin to the results described above from the
surface community. Samples were analyzed for pDA and failed to be above the detection limit.
The community observed in the 1:10 treatment in samples collected after one and two days of
incubation were 60 % similar to the community observed in the initial DCM community,
according to the Bray Curtis Similarity analysis (Figure 4-8B). Both controls and the 1:100 and
1:1000 treatment communities present throughout the experiment were 60 % similar to one
another in the samples collected after one and two days of incubation. The communities were
80 % similar in the samples collected on day three and at the close of the experiment.
140
Figure 4-8. MDS plots using Bray-Curtis similarities for the phytoplankton community composition of each treatment throughout the Pre-Diversion experiments with surface
water (A) and DCM water (B), the Mid-Diversion experiment (C) and the first three days of both the Pre-Diversion and Mid-Diversion experiments (D). Pre-Diversion surface
samples are plotted with upward pointing triangles, DCM samples with downward pointing triangles and the Mid-Diversion samples are plotted with circles. All T0 samples are
colored a light grey, true controls are in black, Milli-Q® controls are black outlined symbols, 1:10 effluent treatments are in red, 1:100 effluent treatments are in yellow and the
1:1000 effluent treatments are in blue. Samples with Bray-Curtis similarities of 40, 60 and 80 are circled on each plot in green, dark blue and light blue, respectively.
141
The community composition of the initial Mid-Diversion community was dominated by
diatoms, however none of the individual genera identified comprised more than 25 % of the
community. The common diatom genera observed (between 10 to 25 % abundance) were
Asterionellopsis, Chaetoceros, Leptocylindrus and the P. delicatissima and P. seriata size classes.
The P. delicatissima and P. seriata size class was present in all treatments and time points,
ranging 12 to 48 % and 3 to 28 % of the community, respectively. Measurable pDA
concentrations were present in all treatments and time points, although the initial water
sample was below detection. The two highest pDA concentrations occurred in the 1:100
effluent treatment in samples collected on the second and third day, 0.23 ± 0.025 and 0.42 ±
0.057 µg/L, respectively. Bray Curtis Similarity analysis revealed the communities observed
throughout the experiment in the various treatments were 60 % similar (Figure 4-8C). The 1:10
and 1:100 treatment communities present in samples from the second and third day of the
experiment were 80 % similar.
Direct comparison of the Pre-Diversion surface, Pre-Diversion DCM and the Mid-
Diversion experiment community composition results revealed that the Mid-Diversion
experiment was 40 % similar to only a small number of samples from the Pre-Diversion
experiments (Figure 4-8D). All Mid-Diversion experiment samples are 60 % similar to one
another, similar to the result received when the Mid-Diversion treatments were compared
solely to one another. Only the true control sample from the second day of the Mid-Diversion
experiment was 60 % similar to any Pre-Diversion sample; specifically, it was 60 % similar to the
second day samples of the Pre-Diversion surface treatments of 1:100 and 1:1000, the third day
142
samples of the Pre-Diversion DCM treatments of the controls, 1:100 and 1:1000, and the Pre-
Diversion surface samples from the 1:100 and 1:1000 treatments on day three.
DISCUSSION
The 1:10 effluent addition investigated in the present study represents a ‘worst case
scenario’ that may occur when agencies divert flow to older discharge systems in times of
maintenance and emergency. Designs of modern sewage outfalls rely on increasing the dilution
of effluent upon discharge in order to diminish the impact on the marine environment. The
2012 OCSD diversion event diverted effluent flow from a pipe 8.0 km from shore with an
estimated initial dilution of 1:200 to an older pipe located 1.6 km from shore with an estimated
initial dilution of 1:20. A previous outfall diversion to an older system operated by the Hyperion
Water Treatment Plant, occurred in Santa Monica Bay, Los Angeles, CA, for several days in 2006
and was followed by a phytoplankton bloom of HAB dinoflagellates correlated with
environmental parameters indicative of effluent plume water (Reifel et al. accepted). The
dilution of the Hyperion effluent plume at their older pipe located 1.6 km from shore was
estimated to be 1:13, but observations made during the short diversion determined the initial
dilution was actually 1:11 (Reifel et al. accepted). This study aimed to understand if a HAB
event would accompany the OCSD diversion and whether or not increased productivity near
the area of diversion could be attributed directly to the stimulation of autotrophic growth by
the presence of effluent.
143
Effluent impact on phytoplankton community composition. Nutrient uptake by phytoplankton
is directly influenced by genetics, biochemistry and physiology (Dortch et al. 1984, Dortch 1990,
Cochlan et al. 1991, Kudela et al. 1997, Lomas & Glibert 2000), and these constraints determine
the effect nitrogenous source will have on phytoplankton community structure. The
nitrogenous source dominate in secondarily treated effluent is NH
4
, a reduced form of N that is
often assumed to be preferred over NO
3
by many phytoplankton species, but is also known to
inhibit NO
3
uptake in some species. In San Francisco Bay for example, high concentrations of
NH
4
introduced to the system primarily through effluent discharge, has been implicated in the
suppression of NO
3
uptake and extensive phytoplankton blooms develop only after NH
4
concentrations are drawn down and NO
3
uptake is allowed to occur (Wilkerson et al. 2006,
Dugdale et al. 2007). Studies off southern California have reported areas of enhanced
phytoplankton growth associated with the location of effluent discharges (Allan Hancock
Foundation 1964, Eppley et al. 1972), suggesting phytoplankton communities in the area may
be capable of responding to NH
4
input.
The two effluent samples employed in the incubation experiments had relatively similar
concentrations of macronutrients and comparable N:P ratios, but absolute values and N:P
ratios differed markedly from natural sources of N and P (Table 4-1). Diatom and picoeukaryote
abundances increased in response to effluent addition in both the Pre- and Mid-Diversion
experiments, while the picophotocyanobacteria decreased in abundance (Figures 4-2 and 4-4).
Bray Curtis Similarity values revealed that the phytoplankton community present at the start of
the Pre- and Mid-Diversion experiments were 25 % similar to one another and only a small
144
subset of Pre-Diversion experiment samples were 40 % similar to communities observed
throughout the Mid-Diversion experiment (Figure 4-8D).
The 1:10 effluent treatments in the Pre-Diversion experiments exhibited a lag in
response to effluent addition, with increases in chlorophyll a concentrations and abundances of
diatoms and picoeukaryotes not occurring until after the first two days of incubation. The 1:10
treatment added 170 µM of NH
4
, which is well over two orders of magnitude above NH
4
concentrations present naturally (1.39 µM in the surface and 1.34 µM in the DCM). The
phytoplankton communities collected from the surface and DCM for the Pre-Diversion
experiment may have been unable to respond to the high NH
4
addition due either to NH
4
inhibition or nitrogenous source preference (i.e. – NO
3
). This is represented well in the MDS
plot of Bray-Curtis similarities comparing the Pre-Diversion DCM treatments, revealing the 1:10
treatment community present in the day one and day two samples were 60 % similar to the
community present at the start of the experiment (Figure 8B). Increases in chlorophyll a,
diatom and picoeukaryote concentrations do not occur until the third day of incubation in the
1:10 treatment and the community present in the sample was only 40 % similar to the original
community.
The 1:10 treatment in the Mid-Diversion experiment did not exhibit a lag in response to
effluent addition, suggesting the community present at the start of the experiment was capable
of responding to NH
4
input. The samples collected from the 1:10 and 1:100 treatments had
community compositions 80 % similar to one another during the first three days of incubation
(Figure 8C), as well as corresponding concentrations of chlorophyll a, diatoms, picoeukaryotes
and picophotocyanobacteria (Figure 4-6). Results from the 1:10 effluent mimic addition
145
support the hypothesis that the Mid-Diversion experiment community was capable of
responding to NH
4
addition (Figure 4-S1). The concentrations of NH
4
added in the 1:10 effluent
and effluent mimic treatments were similar in magnitude (219 and 287 µM, respectively)
although the N:P ratios were markedly different (Table 4-1). Chlorophyll a concentrations and
abundances of diatoms and picoeukaryotes increased in response to the addition of effluent
mimic and the picophotocyanobacteria decreased (Figure 4-S1). The only source of N added in
the effluent mimic was NH
4
, and if the community had been unable to utilize NH
4
increased
growth would not have been observed immediately.
Stimulation of HAB events. Pseudo-nitzschia from both size classes were observed in both the
Pre- and Mid-Diversion experiments. The Pre-Diversion experiments did not yield detectable
pDA concentrations while the Mid-Diversion experiments had detectable pDA concentrations in
all treatments and time points except for the original sample. The highest pDA concentrations
occurred in the 1:100 treatment, with a pDA concentration of 0.23 ± 0.025 µg/L measured on
day two and 0.42 ± 0.057 µg/L measured on day three. The increase in pDA concentration from
day one (0.064 ± 0.017 µg/L) to the second day coincided with a decrease in PO
4
concentration
from 0.19 ± 0.015 µM in day one to below detection by day two (0.1 µM detection limit). The
PO
4
concentration remained below detection on day three and the SiO
3
concentration
decreased from 2.52 ± 0.48 µM on day two to below detection by the third day. The
correspondence between the highest pDA concentration and undetectable PO
4
and SiO
3
concentrations is in agreement with laboratory and field studies that have revealed a
relationship between enhanced pDA production and nutrient limitation by PO
4
and/or SiO
3
(Pan
146
et al. 1996a, Pan et al. 1996b, Fehling et al. 2004, Schnetzer et al. 2007, Sun et al. 2011, Tatters
et al. 2012, Seubert et al. 2013). The higher abundances of the P. seriata size class observed in
the Mid-Diversion treatments presumably had an impact on the detection of DA, as blooms
associated with high concentrations of pDA have been attributed mainly to members of this
size class, especially in southern California (Busse et al. 2006, Schnetzer et al. 2007, Seubert et
al. 2013, Schnetzer et al. Accepted).
CONCLUSIONS
The present study investigated the response of natural phytoplankton communities to several
dilutions of OCSD secondarily treated effluent in two separate incubation experiments. The
1:10, 1:100 and 1:1000 effluent additions increased chlorophyll a concentrations and the
abundances of diatoms, picoeukaryotes and heterotrophic bacteria. The 1:10 treatment
represented a worst case scenario and had the most dramatic effect on chlorophyll a
concentration increase. The 1:1000 effluent addition introduced nutrients at concentrations
lower than the magnitude that would be expected during an upwelling event, but higher than
the ambient concentrations observed at the start of either experiment. The results indicate
that while effluent additions to natural phytoplankton communities at a 1:1000 dilution can
stimulate the community, the degree of stimulation is markedly less than what would occur if
high dilution rates were not achieved.
147
Supplementary Figure 4-1. Chlorophyll a concentrations (A) and the abundances of diatoms (B), picoeukaryotes (C),
Prochlorococcus spp. (D), Synechoccus spp. (E) and heterotrophic bacteria (F) measured during the Mid-Diversion effluent
addition experiment are plotted separately for each treatment. Results from the vitamin and trace metal addition is plotted
with black circles and solid black lines, 1:10 effluent with vitamins and trace metals is plotted with dark grey circles and solid
dark grey lines and 1:10 effluent mimic with vitamins and trace metals is plotted with light grey circles and solid light grey lines.
148
REFERENCES
Allan Hancock Foundation (1964) An investigation on the fate of organic and inorganic wastes
discharged into the marine environment and their effects on biological productivity.
California State Water Quality Control Board, Standard Agreement No 12-17
Allen W (1946) "Red Water" in La Jolla Bay in 1945. T Am Microsc Soc 65:149-153
Anderson D, Glibert P, Burkholder J (2002) Harmful algal blooms and eutrophication: Nutrient
sources, composition and consequences. Estuaries 25:704-726
Bates S, Worms J, Smith J (1993) Effects of ammonium and nitrate on growth and domoic acid
production by Nitzschia pungens in batch culture. Can J Fish Aquat Sci 50:1248-1254
Busse L, Venrick E, Antrobus R, Miller P, Vigilant V, Silver M, Mengelt C, Mydlarz L, Prezelin B
(2006) Domoic acid in phytoplankton and fish in San Diego, CA, USA. Harmful Algae
5:91-101
Caron D, Garneau M-E, Seubert E, Howard M, Darjany L, Schnetzer A, Cetinić I, Filteau G, Lauri P
(2010) Harmful algae and their potential impacts on desalination operations off
southern California. Water Res 44:385-416
Chan F, Barth J, Lubchenco J, Kirincich A, Weeks H, Peterson W, Menge B (2008) Emergence of
anoxia in the California current large marine ecosystem. Science 319:920
Clarke K (1993) Non-parametric multivariate analyses of changes in community structure. Aust J
Ecol 18:117-143
Clarke K, Gorley R (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth
Cochlan W, Harrison P, Denman K (1991) Diel periodicity of nitrogen uptake by marine
phytoplankton in nitrate-rich environments. Limnol Oceanogr 36:1689-1700
Cochlan W, Herndon J, Kudela R (2008) Inorganic and organic nitrogen uptake by the toxigenic
diatom Pseudo-nitzschia australis (Bacillariophyceae). Harmful Algae 8:111-118
Davidson K, Gowen R, Tett P, Bresnan E, Harrison P, McKinney A, Milligan S, Mills D, Silke J,
Crooks A-M (2012) Harmful algal blooms: How strong is the evidence that nutrient
ratios and forms influence their occurrence? Estuar Coast Shelf S 115:399-413
Deutsch C, Brix H, Ito T, Frenzel H, Thompson L (2011) Climate-forcced variability of ocean
hypoxia. Science 333:336-339
Dortch Q, Clayton Jr. J, Thoresen S, Ahmed S (1984) Species differences in accumulation of
nitrogen pools in phytoplankton. Mar Biol 81:237-250
Dortch Q (1990) The interaction between ammonium and nitrate uptake in phytoplankton. Mar
Ecol Prog Ser 61:183-201
149
Dugdale R, Wilkerson F, Hogue V, Marchi A (2007) The role of ammonium and nitrate in spring
bloom development in San Francisco Bay. Estuar Coast Shelf S 73:17-29
Eppley R, Carlucci A, Holm-Hansen O, Kiefer D, McCarthy J, Williams P (1972) Evidence for
eutrophication in the sea near southern California coastal sewage outfalls - July; 1970.
CalCOFI Rep 16:74-83
Eppley R, Renger E, Harrison W (1979) Nitrate and phytoplankton production in southern
California coastal waters. Limnol Oceanogr 24:483-494
Eppley R, Sapienza C, Renger E (1978) Gradients in phytoplankton stocks and nutrients off
southern California in 1974-76. Estuar Coast Mar S 7:291-301
Fabry V, Seibel B, Feely R, Orr J (2008) Impacts of ocean acidifcation on marine fauna and
ecosystem processes. ICES J Mar Sci 65:414-432
Fehling J, Davidson K, Bolch C, Bates S (2004) Growth and domoic acid production by Pseudo-
nitzschia seriata (Baccilariophyceae) under phosphate and silicate limitation. J Phycol
40:674-683
Fire S, Wang Z, Berman M, Langlois G, Morton S, Sekula-Wood E, Benitez-Nelson C (2010)
Trophic transfer of the harmful algal toxin domoic acid as a cause of death in a Minke
whale (Balaenoptera acutorostrata) stranding in southern California. Aquat Mammals
36:342-350
Fu F, Tatters A, Hutchins D (2012) Global change and the future of harmful algal blooms in the
ocean. Mar Ecol Prog Ser 470:207-233
Glibert P, Harrison J, Heil C, Seitzinger S (2006) Escalating worldwide use of urea - a global
change contributing to coastal eutrophication. Biogeochemistry 77:441-463
Goldman J, Glibert P (1982) Comparative rapid ammonium uptake by four species of marine
phytoplankton. Limnol Oceanogr 27:814-827
Guillard R, Hargraves P (1993) Stichochrysis immobilis is a diatom, not a chrysophyte.
Phycologia 32:234-236
Hampson D, Manalo J (1998) The activation of glutamate receptors by kainic acid and domoic
acid. Nat Toxins 6:153-158
Hasle G, Lange C, Syvertsen E (1996) A review of Pseudo-nitzschia, with special reference to the
Skagerrak, North Atlantic, and adjacent waters. Helgolander Meeresun 50:131-175
Hasle G, Syvertsen E (1997) Marine Diatoms. In: Carmelo T (ed) Identifying Marine
Phytoplankton. Academic Press, San Diego
Hillebrand H, Sommer U (1996) Nitrogenous nutrition of the potentially toxic diatom Pseudo-
nitzschia pungens f. multiseries Hasle. J Plank Res 18:295-301
150
Hoegh-Guldberg O, Mumby P, Hooten A, Steneck R, Greenfield P, Gomez E, Harvell C, Sale P,
Edwards A, Caldeira K, Knowlton N, Eakin C, Iglesias-Prieto R, Muthiga N, Bradbury R,
Dubi A, Hatziolos M (2010) Coral reefs under rapid climate change and ocean
acidification. Science 318:1737-1742
Howard M, Cochlan W, Ladizinsky N, Kudela R (2007) Nitrogenous preference of toxigenic
Pseuod-nitzschia australis (Bacillariophyceae) from field and laboratory experiments.
Harmful Algae 6:206-217
Howard M, Sutula M, Caron D, Chao Y, Farrara J, Frenzel H, Jones B, Robertson G, McLaughlin K,
Sengupta A (in revision) Anthropogenic nutrient sources rival natural sources on small
scales in the coastal waters of the southern California bight. Limnol Oceanogr
Hubbard K, Rocap G, Armbrust E (2008) Inter- and intraspecific community structure within the
diatom genus Pseudo-nitzschia (Bacillariophyceae). J Phycol 44:637-649
Kudela R, Cochlan W, Dugdale R (1997) Carbon and nitrogen uptake response to light by
phytoplankton during an upwelling event. J Plank Res 19:609-630
Kudela R, Lane J, Cochlan W (2008) The potential role of anthropogenically derived nitrogen in
the growth of harmful algae in California, USA. Harmful Algae 8:113-110
Lewitus A, Horner R, Caron D, Garcia-Mendoza E, Hickey B, Hunter M, Huppert D, Kudela R,
Langlois G, Largier J, Lessard E, RaLonde R, Rensel J, Strutton P, Trainer V, Tweddle J
(2012) Harmful algal blooms along the North American west coast region: History,
trends, causes and impacts. Harmful Algae 19:133-159
Lomas M, Glibert P (2000) Comparisons of nitrate uptake, storage and reduction in marine
diatoms and flagellates. J Phycol 36:903-913
Moore S, Trainer V, Mantua N, Parker M, Laws E, Backer L, Fleming L (2008) Impacts of climate
variability and futre climate change on harmful algal blooms and human health. Environ
Health 7:S4
Nicholls R, Hoozemans F, Marchand M (1999) Increasing flood risk and wetland losses due to
global sea-level rise: regional and global analyses. Global Environ Chang 9:S69-S87
Paerl H (1997) Coastal eutrophication and harmful algal blooms: Importance of atmospheric
deposition and groundwater as "new" nitrogen and other nutrient sources. Limnol
Oceanogr 42:1154-1165
Paerl H, Dennis R, Whithall D (2002) Atmospheric deposition of nitrogen: Implications for
nutrient over-enrichment of coastal waters. Estuaries 25:677-693
Pan Y, Subba Rao D (1997) Impacts of domestic sewage effluent on phytoplankton from
Bedford Basin, Eastern Canada. Mar Poll Bull 34:1001-1005
151
Pan Y, Subba Rao D, Mann K, Brown R, Pocklington R (1996a) Effects of silicate limitation on
production of domoic acid, a neurotoxin, by the ditaom Pseudo-nitzschia multiseries. I.
Batch culture studies. Mar Ecol Prog Ser 131:225-233
Pan Y, Subba Rao D, Mann K, Li W, Harrison W (1996b) Effects of silicate limitation on
production of domoic acid, a neurotoxin, by the diatom Pseudo-nitzschia multiseries. II.
Continuous culture studies. Mar Ecol Prog Ser 131:235-243
Reifel K, Corcoran A, Cash C, Shipe R, Jones B (accepted) Effects of a surfacing effluent plume on
a coastal phytoplankton community. Cont Shelf Res:
http://dx.doi.org/10.1016/j.csr.2013.1004.1012
Sahagian D, Schwartz F, Jacobs D (1994) Direct anthropogenic contributions to sea level rise in
the twentieth century. Nature 367:54-57
Schnetzer A, Jones B, Schaffner R, Cetinic I, Fitzpatrick E, Miller P, Seubert E, Caron D
(submitted) Coastal upwelling linked to toxic Pseudo-nitzschia australis blooms in Los
Angeles coastal waters, 2005-2007. J Plank Res
Schnetzer A, Miller P, Schaffner R, Stauffer B, Jones B, Weisberg S, DiGiacomo P, Berelson W,
Caron D (2007) Blooms of Pseudo-nitzschia and domoic acid in the San Pedro Channel
and Los Angeles harbor areas of the Southern California Bight, 2003-2004. Harmful
Algae 6:372-387
Scholin C, Buck K, Britschbi T, Cangelosi G, Chavez F (1996) Identification of Pseudo-nitzschia
australis (Bacillariophyceae) using rRNA-targeted probes in whole cell and sandwich
hybridization formats. Phycologia 35:190-197
Seubert E, Gellene A, Howard M, Connell P, Ragan M, Jones B, Runyan J, Caron D (2013)
Seasonal and annual dynamics of harmful algae and algal toxins revealed through
weekly monitoring at two coastal ocean sites off southern California, USA. Environ Sci
Poll DOI 10.1007/s11356-012-1420-0
Seubert E, Trussell S, Eagleton J, Schnetzer A, Cetinic I, Lauri P, Jones B, Caron D (2012) Algal
toxins and reverse osmosis desalination operations: Laboratory bench testing and field
monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid. Water Res 46:6563-
6573
Smith V, Joye S, Howarth R (2006) Eutrophication of freshwater and marine ecosystems. Limnol
Oceanogr 51:351-355
Smith V, Tilman G, Nekola J (1999) Eutrophication: impacts of excess nutrient inputs on
freshwater, marine and terrestrial ecosystems. Environ Poll 100:179-196
Stramma L, Schmidtko S, Levin L, Johnson G (2010) Ocean oxygen minima expansions and their
biological impacts. Deep Sea Research Part I: Oceanogr Res Papers 57:587-595
152
Sun J, Hutchins D, Feng Y, Seubert E, Caron D, Fu F (2011) Effects of changing pCO2 and
phosphate availability on domoic acid production and physiology of the marine harmful
bloom diatom Pseudo-nitzschia multiseries. Limnol Oceanogr 56:829-840
Tatters A, Fu F, Hutchins D (2012) High CO2 and silicate limitation synergistically increase the
toxicity of Pseudo-nitzschia fraudulenta. PLoS ONE 7:e32116
Thessen A, Bowers H, Stoecker D (2009) Intra- and interspecies differences in growth and
toxicity of Pseudo-nitzschia while using different nitrogen sources. Harmful Algae 8:792-
810
Torres de la Riva G, Johnson C, Gulland F, Langlois G, Heyning J, Rowles T, Mazet J (2009)
Association of an unusual marine mammal mortality even with Pseudo-nitzschia spp.
blooms along the southern California coastline. J Wild Diseas 45:109-121
Torrey H (1902) An unusual occurrence of Dinoflagellata on the California coast. Am Nat
36:187-192
Utermohl H (1958) Zur gewassertypenfrage tropischer seen. Verhandlungen Internationlen
Vereinigung Limnologie 13:236-251
Wilkerson F, Dugdale R, Hogue V, Marchi A (2006) Phytoplankton blooms and nitrogen
productivity in San Francisco Bay. Estuar Coast 29:401-416
Xi D, Ramsdell J (1996) Glutamate receptors and calcium entry mechanisms for domoic acid in
hippocampal neurons. NeuroReport 7:1115-1120
153
Bibliography
Abdul Azis P, Al-Tisan I, Al-Daili M, Green T, Dalvi A, Javeed M (2000) Effects of environment on
source water for desalination plants on the eastern coast of Saudi Arabia. Desalination
132:29-40
Adams N, Lesoing M, Trainer V (2000) Environmental conditions associated with domoic acid in
razor clams on the Washington coast. J Shellfish Res 19:1007-1015
Ahn S, Kulis D, Erdner D, Anderson D, Walt D (2006) Fiber-optic microarray for simultaneous
detection of multiple harmful algal bloom species. Appl Environ Microbiol 72:5742-5749
Allan Hancock Foundation (1964) An investigation on the fate of organic and inorganic wastes
discharged into the marine environment and their effects on biological productivity.
California State Water Quality Control Board, Standard Agreement No 12-17
Allen J, Smyth T, Siddorn J, Holt M (2008) How well can we forecast high biomass algal bloom
events in a eutrophic coastal sea? Harmful Algae 8:70-76
Allen W (1934) Marine plankton diatoms of lower California in 1931. Bot Gaz 95:485-492
Allen W (1936) Occurrence of marine plankton diatoms in a ten-year series of daily catches in
southern California. Am J Bot 23:60-63
Allen W (1938) "Red Water" along the west coast of the United States in 1938. Science 88:55-56
Allen W (1946) "Red Water" in La Jolla Bay in 1945. T Am Microsc Soc 65:149-153
Anderson C, Brzezinski M, Washburn L, Kudela R (2006) Circulation and environmental
conditions during a toxigenic Pseudo-nitzschia australis bloom in the Santa Barbara
Channel, California. Mar Ecol Prog Ser 327:119-133
Anderson C, Sapiano M, Prasad M, Long W, Tango P, Brown C, Murtugudde R (2010) Predicting
potentially toxigenic Pseudo-nitzschia blooms in the Chesapeake Bay. J Mar Sys 83:127-
140
Anderson C, Siegel D, Kudela R, Brzezinski M (2009) Empirical models of toxigenic Pseudo-
nitzschia blooms: Potential use as a remote detection tool in the Santa Barbara
Channel. Harmful Algae 8:478-492
Anderson D, Cembella A, Hallegraeff G (2012) Progress in understanding harmful algal blooms:
Paradigm shifts and new technologies for research, monitoring and management. Annu
Rev Mar Sci 4:143-176
154
Anderson D, Glibert P, Burkholder J (2002) Harmful algal blooms and eutrophication: Nutrient
sources, composition and consequences. Estuaries 25:704-726
Baden D, Bourdelais A, Jacocks H, Michelliza S, Naar J (2005) Natural and derivative brevetoxins:
historical background, multiplicity and effects. Environ Health Persp 113:621-625
Bargu S, Lefebvre K, Silver M (2006) Effect of dissolved domoic acid on the grazing rate of
Euphasia pacifica Mar Ecol Prog Ser 312:169-175
Bargu S, Powell C, Coale S, Busman M, Doucette G, Silver M (2002) Krill: a potential vector for
domoic acid in marine food webs. Mar Ecol Prog Ser 237:209-216
Bargu S, Silver M, Goldstein T, Roberts K, Gulland F (2010) Complexity of domoic acid-related
sea lion strandings in Monterey Bay, California: Foraging patterns, climate events and
toxic blooms. Mar Ecol Prog Ser 418:213-222
Bates S (1998) Ecophysiology and metabolism of ASP toxin production. In: Anderson D,
Cembella A, Hallegraeff G (eds) Physiological ecology of harmful algal blooms. Springer-
Verlag, Heidelberg
Bates S (2000) Domoic-acid-producing diatoms: Another genus added! J Phycol 36:978-985
Bates S, Bird C, de Freitas A, Foxall R, Gilgan M, Hanic L, Johnson G, McCulloch A, Odense P,
Pocklington R, Quilliam M, Sim P, Smith J, Subba Rao D, Todd E, Walter J, Wright J (1989)
Pennate diatom Nitzschia pungens as the primary source of domoic acid, a toxin in
shellfish from eastern Prince Edward Island, Canada. Can J Fish Aquat Sci 46:1203-1215
Bates S, Leger C, Satchwell M, Boyer G The effects of iron on domoic acid production by Pseudo-
nitzschia multiseries. In: Hallegraeff G, Blackburn S, Bolch C, Lewis R (eds). Proc 9th
International Conference on Harmful Algal Blooms. Intergovernmental Oceanographic
Commision
Bates S, Worms J, Smith J (1993) Effects of ammonium and nitrate on growth and domoic acid
production by Nitzschia pungens in batch culture. Can J Fish Aquat Sci 50:1248-1254
Bejarano A, Gulland F, Goldstein T, St Leger J, Hunter M, Schwacke L, Van Dolah F, Rowles T
(2008) Demographics and spatio-temporal signature of the biotoxin domoic acid in
California sea lions (Zalophus californianus) stranding records. Mar Mammal Sci 24:899-
912
Blanco J, Acosta C, Marino C, Muniz S, Martin H, Morono A, Correa J, Arevalo F, Salgado C
(2006) Depuration of domoic acid from different body compartments of the king scallop
Pecten maximus grown in raft culture and natural bed. Aquat Living Resour 19:257-265
Bland J, Altman D (1986) Statistical methods for assessing agreement between two methods of
clinical measurement. Lancet i:307-310
155
Brodie E, Gulland F, Greig D, Hunter M, Jaakola J, St Leger J, Leighfield T, Van Dolah F (2006)
Domoic acid reproductive failure of California sea lions (Zalophus californianus). Mar
Mammal Sci 22:700-707
Burkholder J, Azanza R, Sako Y (2006) The ecology of hamrul dinoflagellates. In: Graneli E,
Turner J (eds) Ecology of Harmful Algae, Book 189. Springer Heidelberg, Berlin
Busse L, Venrick E, Antrobus R, Miller P, Vigilant V, Silver M, Mengelt C, Mydlarz L, Prezelin B
(2006) Domoic acid in phytoplankton and fish in San Diego, CA, USA. Harmful Algae
5:91-101
Caron D (2001) Protistan Herbivory and Bacterivory. In: Paul J (ed) Methods in Microbiology,
Book 30. Academic Press, Ltd, St. Petersburg, FL
Caron D, Garneau M-E, Seubert E, Howard M, Darjany L, Schnetzer A, Cetinic I, Filteau G, Lauri
P, Jones B, Trussell S (2010) Harmful algae and their potential impacts on desalination
operations off southern California. Water Res 44:385-416
Carstensen J, Henriksen P, Heiskanen A (2007) Summer algal blooms in shallow estuaries:
definition, mechanisms and link to eutrophication. Limnol Oceanogr 52:370-384
Castaing J, Masse A, Pontie M, Sechet V, Haure J, Jaouen P (2010) Investigating submerged
ultrafiltration (UF) and microfiltration (MF) membranes for seawater pre-treatment
dedicated to removal of undesirable micro-algae. Desalination 253:71-77
Chan F, Barth J, Lubchenco J, Kirincich A, Weeks H, Peterson W, Menge B (2008) Emergence of
anoxia in the California current large marine ecosystem. Science 319:920
Clarke K (1993) Non-parametric multivariate analyses of changes in community structure. Aust J
Ecol 18:117-143
Clarke K, Gorley R (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth
Cochlan W, Harrison P, Denman K (1991) Diel periodicity of nitrogen uptake by marine
phytoplankton in nitrate-rich environments. Limnol Oceanogr 36:1689-1700
Cochlan W, Herndon J, Kudela R (2008) Inorganic and organic nitrogen uptake by the toxigenic
diatom Pseudo-nitzschia australis (Bacillariophyceae). Harmful Algae 8:111-118
Davidson K, Gowen R, Tett P, Bresnan E, Harrison P, McKinney A, Milligan S, Mills D, Silke J,
Crooks A-M (2012) Harmful algal blooms: How strong is the evidence that nutrient
ratios and forms influence their occurrence? Estuar Coast Shelf S 115:399-413
de la Riva G, Johnson C, Gulland F, Langlois G, Hayning J, Rowles T, Mazet J (2009) Association
of an unusual marine mammal mortality event with Pseudo-nitzschia spp. blooms along
the southern California coastline. J Wildl Dis 45:109-121
156
Del Rio R, Bargu S, Baltz D, Fire S, Peterson G, Wang Z (2010) Gulf menhaden (Brevoortia
patronus): A potential vector of domoic acid in coastal Louisiana food webs. Harmful
Algae 10:19-29
Desormeaux E, Meyerhofer P, Luckenbach H, Kudela R (2011) Pilot-testing multiple
pretreatment systems for seawater desalination. Inter Desalination Assoc J First
Quarter:42-52
Deutsch C, Brix H, Ito T, Frenzel H, Thompson L (2011) Climate-forcced variability of ocean
hypoxia. Science 333:336-339
Dortch Q (1990) The interaction between ammonium and nitrate uptake in phytoplankton. Mar
Ecol Prog Ser 61:183-201
Dortch Q, Clayton Jr. J, Thoresen S, Ahmed S (1984) Species differences in accumulation of
nitrogen pools in phytoplankton. Mar Biol 81:237-250
Doucette G, Mikulski C, Jones K, King K, Greenfield D, Marin III R, Jensen S, Roman B, Elliot C,
Scholin C (2009) Remote, subsurface detection of the algal toxin domoic acid onboard
the Environmental Sample Processor: Assay development and field trials. Harmful Algae
8:880-888
Dugdale R, Wilkerson F, Hogue V, Marchi A (2007) The role of ammonium and nitrate in spring
bloom development in San Francisco Bay. Estuar Coast Shelf S 73:17-29
Eppley R, Carlucci A, Holm-Hansen O, Kiefer D, McCarthy J, Williams P (1972) Evidence for
eutrophication in the sea near southern California coastal sewage outfalls - July; 1970.
CalCOFI Report 16:74-83
Eppley R, Renger E, Harrison W (1979) Nitrate and phytoplankton production in southern
California coastal waters. Limnol Oceanogr 24:483-494
Eppley R, Sapienza C, Renger E (1978) Gradients in phytoplankton stocks and nutrients off
southern California in 1974-76. Estuar Coast Mar Sci 7:291-301
Fabry V, Seibel B, Feely R, Orr J (2008) Impacts of ocean acidifcation on marine fauna and
ecosystem processes. ICES Journal of Marine Science 65:414-432
Fehling J, Davidson K, Bolch C, Bates S (2004) Growth and domoic acid production by Pseudo-
nitzschia seriata (Baccilariophyceae) under phosphate and silicate limitation. J Phycol
40:674-683
Fehling J, Davidson K, Bolch C, Tett P (2006) Seasonality of Pseudo-nitzschia spp.
(Bacillariophyceae) in western Scottish waters. Mar Ecol Prog Ser 323:91-105
157
Fire S, Wang Z, Berman M, Langlois G, Morton S, Sekula-Wood E, Benitez-Nelson C (2010)
Trophic transfer of the harmful algal toxin domoic acid as a cause of death in a Minke
Whale (Balaenoptera acutorostrata) stranding in southern California. Aquat Mammals
36:342-350
Fire S, Wang Z, Leighfield T, Morton S, McFee W, McLellan W, Litaker R, Tester P, Hohn A,
Lovewell G, Harms C, Rotstein D, Barco S, Costidis A, Sheppard B, Bossart G, Stolen M,
Durden W, Van Dolah F (2009) Domoic acid exposure in pygmy and dwarf sperm whales
(Kogia spp.) from southeastern and mid-Atlantic US waters. Harmful Algae 8:658-664
Franks R, Wilf M, Voutchkov N, Murkute P, Kizer J (url) A pilot study using seawater reverse
osmosis membranes in combination with various pretreatments to meet the challenges
of Pacific seawater desalination. Accessed January 24, 2012:Retrieved at
http://www.membranes.com/docs/papers/New%20Folder/A%20PILOT%20STUDY%20U
SING%20SEAWATER%20REVERSE%20OSMOSIS%20MEMBRANES.pdf
Fritz L, Quilliam M, Wright J, Beale A, Work T (1992) An outbreak of domoic acid poisoning
attributed to the pennate diatom Pseudo-nitzschia australis. J Phycol 28:439-442
Frolov S, Kudela R, Bellingham J (2012) Monitoring of harmful algal blooms in the era of
diminishing resources: A case study of the U.S. west coast. Harmful Algae
http://dx.doi.org/10.1016/j.hal.2012.11.001
Fryxell G, Villace M, Shapiro L (1997) The occurence of the toxic diatom genus Pseudo-nitzschia
(Bacillariophyceae) on the West Coast of the USA, 1920-1996: a review. Phycologia
36:419-437
Fu F, Tatters A, Hutchins D (2012) Global change and the future of harmful algal blooms in the
ocean. Mar Ecol Prog Ser 470:207-233
Garneau M-E, Schnetzer A, Countway P, Jones A, Seubert E, Caron D (2011) Examination of the
seasonal dynamics of the toxic dinoflagellate Alexandrium catenella at Redondo Beach,
California, by quantitative PCR. Appl Environ Microbiol 77:7669-7680
Gastrich M, Wazniak C (2002) A brown tide bloom index based on the potential harmful effects
of the brown tide alga, Aureococcus anophagefferens. Aquat Ecosys Health 5:435-441
Gescher C, Metfies K, Medlin L (2008) The ALEX CHIP - Development of a DNA chip for
identification and monitoring of Alexandrium. Harmful Algae 7
Glibert P, Allen J, Bouwman A, Brown C, Flynn K, Lewitus A, Madden C (2010) Modeling of HABs
and eutrophication: Status, advances, challenges. J Marine Syst 83:262-275
Glibert P, Harrison J, Heil C, Seitzinger S (2006) Escalating worldwide use of urea - a global
change contributing to coastal eutrophication. Biogeochemistry 77:441-463
158
Glibert P, Seitzinger S, Heil C, Burkholder J, Parrow M, Codispoti L, Kelly V (2005) The role of
eutrophication in the global proliferation of hamrful algal blooms. Oceanogr 18:198-209
Gobler C, Sunda W (2012) Ecosystem disruptive algal blooms of the brown tide species,
Aureococcus anophagefferens and Aureoumbra lagunensis. Harmful Algae 14:36-45
Goldman J, Glibert P (1982) Comparative rapid ammonium uptake by four species of marine
phytoplankton. Limnol Oceanogr 27:814-827
Goldstein T, Mazet J, Zabka T, Langlois G, Colegrove K, Silver M, Bargu S, Van Dolah F, Leighfield
T, Conrad P, Barakos J, Williams D, Dennison S, Haulena M, Gulland F (2008) Novel
symptomatology and changing epidemiology of domoic acid toxicosis in California sea
lions (Zalophus californianus): an increasing risk to marine mammal health. P R Soc B
275:267-276
Goldstein T, Zabka T, DeLong R, Wheeler E, Ylitalo G, Bargu S, Silver M, Leighfield T, Van Dolah
F, Langlois G, Sidor I, Dunn J, Gulland F (2009) The role of domoic acid in abortion and
premature parturition of California sea lions (Zalophus californianus) on San Miguel
Island, California. J Wildlife Dis 45:91-108
Greenfield D, Marin III R, Doucette G, Mikulski C, Jones K, Jensen S, Roman B, Alvarado N,
Feldman J, Scholin C (2008) Field applications of the second-generation Environmental
Sample Processor (ESP) for remote detection of harmful algae: 2006-1007. Limnol
Oceanogr: Methods 6:667-679
Greenfield D, Marin III R, Jensen S, Massion E, Roman B, Feldman J, Scholin C (2006) Application
of environmental sample processor (ESP) methodology for quantifying Pseudo-nitzschia
australis using ribosomal RNA-targeted probes in sandwich and fluorescent in situ
hybridization formats. Limnol Oceanogr: Methods 4:426-435
Gregorio D, Pieper R (2000) Investigations of red tides along the southern California coast. Bull
South Calif Acad Sci 99:147-160
Guillard R, Hargraves P (1993) Stichochrysis immobilis is a diatom, not a chrysophyte.
Phycologia 32:234-236
Gulland F, Haulena M, Fauquier D, Langlois G, Lander M, Zabka T, Duerr R (2002) Domoic acid
toxicity in California sea lions (Zalophus caliornianus): clinical signs, treatment and
survival. Vet Rec 150:475-480
Hall A, Frame E (2010) Evidence of domoic acid exposure in harbour seals from Scotland: A
potnetial factor in the decline in abundance? Harmful Algae 9:489-493
Hallegraeff G (2003) Harmful algal blooms: a global overview. In: Hallegraeff G, Anderson D,
Cembella A (eds) Manual on Harmful Marine Microalgae Monographs on
Oceanographic Methodology. UNESCO Publishing, Paris
159
Hampson D, Manalo J (1998) The activation of glutamate receptors by kainic acid and domoic
acid. Nat Toxins 6:153-158
Hasle G, Lange C, Syvertsen E (1996) A review of Pseudo-nitzschia, with special reference to the
Skagerrak, North Atlantic, and adjacent waters. Helgolander Meeresun 50:131-175
Hasle G, Syvertsen E (1997) Marine Diatoms. In: Carmelo T (ed) Identifying Marine
Phytoplankton. Academic Press, San Diego
Heisler J, Glibert P, Burkholder J, Anderson D, Cochlan W, Dennison W, Dortch Q, Gobler C, Heil
C, Humphries E, Lewitus A, Magniun R, Marshall H, Sellner K, Stockwell D, Stoecker D,
Suddleson M (2008) Eutrophication and harmful algal blooms: a scientific consensus.
Harmful Algae 8:3-13
Henson S, Thomas A (2007) Interannual variability in timing of bloom initiation in the California
Current System. J Geophys Res 112:C080007
Hesp B, Harrison J, Selwood A, Holland P, Kerr D (2005) Detecton of domoic acid in rat serum
and brain by direct competitive enzyme-linked immunosorbent assay (cELISA). Anal
Bioanal Chem 383:783-786
Hillebrand H, Sommer U (1996) Nitrogenous nutrition of the potentially toxic diatom Pseudo-
nitzschia pungens f. multiseries Hasle. J Plank Res 18:295-301
Hoegh-Guldberg O, Mumby P, Hooten A, Steneck R, Greenfield P, Gomez E, Harvell C, Sale P,
Edwards A, Caldeira K, Knowlton N, Eakin C, Iglesias-Prieto R, Muthiga N, Bradbury R,
Dubi A, Hatziolos M (2010) Coral reefs under rapid climate change and ocean
acidification. Science 318:1737-1742
Howard M, Cochlan W, Ladizinsky N, Kudela R (2007) Nitrogenous preference of toxigenic
Pseudo-nitzschia australis (Bacillariophyceae) from field and laboratory experiments.
Harmful Algae 6:206-217
Howard M, Silver M, Kudela R (2008) Yessotoxin detected in mussel (Mytilus californicus) and
phyotplankton samples from the US west coast. Harmful Algae 7:646-652
Howard M, Sutula M, Caron D, Chao Y, Farrara J, Frenzel H, Jones B, Robertson G, McLaughlin K,
Sengupta A (in revision) Anthropogenic nutrient sources rival natural sources on small
scales in the coastal waters of the southern California bight. Limnol Oceanogr
Hubbard K, Rocap G, Armbrust E (2008) Inter- and intraspecific community structure within the
diatom genus Pseudo-nitzschia (Bacillariophyceae). J Phycol 44:637-649
Jessup D, Miller M, Ryan J, Nevins H, Kerkering H, Mekebri A, Crane D, Johnson T, Kudela R
(2009) Mass stranding of marine birds caused by a surfactant-producing red tide. PLoS
One 4:e4550. doi:4510.1371/journal.pone.0004550
160
Jester R (2008) An investigation into the prevalence of Alexandrium derived toxins in marine
food webs. PhD, University of California, Santa Cruz, Santa Cruz
Jester R, Baugh K, Lefebvre K (2009a) Presence of Alexandrium catenella and paralytic shellfish
toxins in finfish, shellfish and rock crabs in Monterey Bay, California, USA. Mar Biol
156:493-504
Jester R, Lefebvre K, Langlois G, Vigilant V, Baugh K, Silver M (2009b) A shift in the dominant
toxin-producing algal species in central California alters phycotoxins in food webs.
Harmful Algae 8:291-298
Kegel J, Del Amo Y, Medlin L (2012) Introduction to project MIDTAL: its methods and samples
from Arcachon Bay, France. Environ Sci Pollut Res DOI 10.1007/s11356-012-1299-9
Kim H, Miller A, McGowan J, Carter M (2009) Coastal phytoplankton blooms in the southern
California bight. Prog Oceanogr 82:137-147
Kim S-H, Yoon J-S (2005) Optimization of microfiltration for seawater suffering from red tide
contamination. Desalination 182:315-321
Kirkpatrick L, Squicciarini D, Backer L, Clark W, Benson J, Cheng Y, Johnson D, Pierce R, Zaias J,
Bossart G, Baden D (2004) Literature review of Florida red tide: implications for human
health effects. Harmful Algae 3:99-115
Kleivdal H, Kristiansen S, Nilsen M (2007) Single-Laboratory validation of Biosense direct
competitive enzyme-linked immunosorbent assay (ELISA) for determination of domoic
acid toxins in shellfish. J AOAC Int 90:1000-1010
Kotaki Y, Lundholm N, Onodera H, Kobayashi K, Bajarias F, Furio E, Iwataki M, Fukuyo Y, Kodama
M (2004) Wide distribution of Nitzschia navis-varingica, a new domoic acid-producing
benthic diatom found in Vietnam. Fisheries Sci 70:28-32
Kudela R, Cochlan W, Dugdale R (1997) Carbon and nitrogen uptake response to light by
phytoplankton during an upwelling event. J Plank Res 19:609-630
Kudela R, Lane J, Cochlan W (2008) The potential role of anthropogenically derived nitrogen in
the growth of harmful algae in California, USA. Harmful Algae 8:103-110
Kvitek R, Goldberg J, Smith G, Doucette G, Silver M (2008) Domoic acid contamination within
eight representative species from the bethic food web of Monterey Bay, California, USA.
Mar Ecol Prog Ser 367:35-47
Kwon B, Park N, Cho J (2005) Effect of algae on fouling and efficiency of UF membranes.
Desalination 179:203-214
161
Ladner D, Vardon D, Clark M (2010) Effects of shear on microfiltration and ultrafiltration fouling
by marine bloom-forming algae. J Membrane Sci 356:33-43
Lane J, Raimondi P, Kudela R (2009) Development of a logistic regression model for the
prediction of toxigenic Pseudo-nitzschia blooms in Monterey Bay, California. Mar Ecol
Prog Ser 383:37-51
Langlois G (2007) Marine Biotoxin Monitoring Program Annual Report. California Department of
Public Health for California Department of Fish and Game, Sacramento
Leandro L, Rolland R, Roth P, Lundholm N, Wang Z, Doucette G (2010) Exposure of the North
Atlantic right whale Eubalaena glacialis to the marine algal biotoxin, domoic acid. Mar
Ecol Prog Ser 398:287-303
Lefebvre K, Bill B, Erickson A, Baugh K, O'Rourke L, Costa P, Nance S, Trainer V (2008)
Characterization of intracellular and extracellular saxitoxin levels in both field and
cultured Alexandrium spp. samples from Sequim Bay, Washington. Mar Drugs 6:103-116
Lefebvre K, Dovel S, Silver M (2001) Tissue distribution and neurotoxic effects of domoic acid in
a prominent vector species, the northen anchovy Engraulis mordax. Mar Biol 138:693-
700
Lefebvre K, Powell C, Busman M, Doucette G, Moeller P, Silver J, Miller P, Hughes M, Singaram
S, Silver M, Tjeerdema R (1999) Detection of domoic acid in northern anchovies and
California sea lions associated with an unusual mortality event. Nat Toxins 7:85-92
Lefebvre K, Robertson A, Frame E, Colegrove K, Nance S, Baugh K, Wiedenhoft H, Gulland F
(2010) Clinical signs and histopathology associated with domoic acid poisoning in
northern fur seals (Callorhinus ursinus) and comparison of toxin detection methods.
Harmful Algae 9:374-383
Legendre L (1990) The significance of microalgal blooms for fisheries and for the export of
particulate organic carbon in oceans. J Plankton Res 12:681-699
Lelong A, Hegaret H, Soudant P, Bates S (2012) Pseudo-nitzschia (Bacillariophyceae) species,
domoic acid and amnesic shellfish poisoning: Revisiting previous paradigms. Phycologia
51:168-216
Lewitus A, Horner R, Caron D, Garcia-Mendoza E, Hickey B, Hunter M, Huppert D, Kudela R,
Langlois G, Largier J, Lessard E, RaLonde R, Rensel J, Strutton P, Trainer V, Tweddle J
(2012) Harmful algal blooms along the North American west coast region: History,
trends, causes and impacts. Harmful Algae 19:133-159
Litaker R, Stewart T, Eberhart B, Wekell J, Trainer V, Kudela R, Miller P, Roberts A, Hertz C,
Johnson T, Frankfurter G, Smith G, Schnetzer A, Schumacker J, Bastian J, Odell A,
162
Gentien P, Le Gal D, Hardison D, Tester P (2008) Rapid enzyme-linked immunosorbent
assay for detection of the algal toxin domoic acid J Shellfish Res 27:1301-1310
Llewellyn L (2006) Saxitoxin, a toxic marine natural product that targets a multitude of
receptors. Nat Prod Rep 23:200-222
Lomas M, Glibert P (2000) Comparisons of nitrate uptake, storage and reduction in marine
diatoms and flagellates. J Phycol 36:903-913
Lundholm N, Daugbjerg N, Moestrup O (2002) Phylogeny of the Bacillariophyceae with
emphasis on the genus Pseudo-nitzschia (Bacillariphyceae) based on partial LSU rDNA.
Eur J Phycol 37:115-134
Lundholm N, Hansen P, Kotaki Y (2004) Effect of pH on growth and domoic acid production by
potentially toxic diatoms of the genera Pseudo-nitzschia and Nitzschia. Mar Ecol Prog
Ser 273:1-15
Lundholm N, Hansen P, Kotaki Y (2005) Lack of allelopathic effects of the domoic acid-producing
marine diatom Pseudo-nitzschia multiseries Mar Ecol Prog Ser 288:21-33
Lundholm N, Moestrup O (2000) Morphology of the marine diatom Nitzschia navis-varingica,
sp. nov. (Bacillariophyceae), another producer of the neurotoxin domoic acid. J Phycol
36:1162-1174
Maldonado M, Hughes M, Rue E, Wells M (2002) The effect of Fe and Cu on growth and domoic
acid production in Pseudo-nitzschia australis. Limnol Oceanogr 47:515-526
Maneiro I, Iglesias P, Guisande C, Riveiro I, Barreiro A, Zervoudaki S, Graneli E (2005) Fate of
domoic acid ingested by the copepod Arcatia clausi. Mar Biol 148:123-130
Maranda L, Wang R, Musauda K, Shimizu Y (1990) Investigations of the source of domoic acid in
mussels. In: Graneli E, Sundstrom B, Edler L, Anderson D (eds) Toxic Marine
Phytoplankton. Elsevier Science Publishing Co., New York
Mazzillo F, Field J, Staaf D, Carter M, Ohman M (2011) A note on the detection of the
neurotoxin domoic acid in beach-stranded Dosidicus gigas in the southern California
bight. CalCOFI Rep 52:109-115
Mazzillo F, Pomeroy C, Kuo J, Ramondi P, Prado R, Silver M (2010) Angler exposure to domoic
acid via consumption of contaminate fishes. Aquat Biol 9:1-12
McManus M, Kudela R, SIlver M, Steward G, Donaghay P, Sullivan J (2008) Cryptic blooms: Are
thin layers the missing connection? Estuar Coast 31:396-401
163
Mengelt C (2006) Ultravioloet photoecology, dark survival and seasonal abundance of Pseudo-
nitzschia australis and P. multiseries in coastal waters of central California. Dissertation,
University of California, Santa Barbara
Miller P, Scholin C (1998) Identification and enumeration of cultured and wild Pseudo-nitzschia
(Bacillariophyceae) using species-specific LSU rRNA-targeted fluorescent probes and
filter-based whole cell hybridization. J Phycol 34:371-382
Moore S, Trainer V, Mantua N, Parker M, Laws E, Backer L, Fleming L (2008) Impacts of climate
variability and futre climate change on harmful algal blooms and human health. Environ
Health 7:S4
Murata M, Shimatani M, Sugitani Y, Oshima Y, Yasumoto T (1982) Isolation and structural
elucidation of the causative toxin of the diarrhetic shellfish poisoning. B Jpn Soc Sci Fish
48:549-552
Nazzal N (2009) 'Red tide' shuts desalination plant. Gulf News, Dubai, UAE
Nicholls R, Hoozemans F, Marchand M (1999) Increasing flood risk and wetland losses due to
global sea-level rise: regional and global analyses. Global Environ Chang 9:S69-S87
Olson M, Lessard E (2010) The influence of the Pseudo-nitzschia toxin, domoic acid, on
microzooplankton grazing and growth: A laboratory and field experiment. Harmful
Algae 9:240-247
Olson M, Lessard E, Wong C, Bernhardt M (2006) Copepod feeding selectivity on microplankton
including the toxigenic diatoms Pseudo-nitzschia spp., in the coastal Pacific Northwest.
Mar Ecol Prog Ser 326:207-220
Omand M, Leichter J, Franks P, Guza R, Lucas A, Feddersen F (2011) Physical and biological
processes underlying the sudden surface appearance of a red tide in the nearshore.
Limnol Oceanogr 56:787-801
Orsini L, Procaccini G, Sarno D, Montresor M (2004) Multiple rDNA ITS-types within the diatom
Pseudo-nitzschia delicatissima (Bacillariophyceae) and their relative abundances across
a spring bloom in the Gulf of Naples. Mar Ecol Prog Ser 271:87-98
Paerl H (1997) Coastal eutrophication and harmful algal blooms: importance of atmospheric
deposition and groundwater as "new" nitrogen and other nutrient sources. Limnol
Oceanogr 42:1154-1165
Paerl H, Dennis R, Whithall D (2002) Atmospheric deposition of nitrogen: Implications for
nutrient over-enrichment of coastal waters. Estuaries 25:677-693
Paerl H, Paul V (2012) Climate change: Links to global expansion of harmful cyanobacteria.
Water Res 46:1349-1363
164
Pan Y, Subba Rao D (1997) Impacts of domestic sewage effluent on phytoplankton from
Bedford Basin, Eastern Canada. Mar Pollut Bull 34:1001-1005
Pan Y, Subba Rao D, Mann K, Brown R, Pocklington R (1996a) Effects of silicate limitation on
production of domoic acid, a neurotoxin, by the ditaom Pseudo-nitzschia multiseries. I.
Batch culture studies. Mar Ecol Prog Ser 131:225-233
Pan Y, Subba Rao D, Mann K, Li W, Harrison W (1996b) Effects of silicate limitation on
production of domoic acid, a neurotoxin, by the diatom Pseudo-nitzschia multiseries. II.
Continuous culture studies. Mar Ecol Prog Ser 131:235-243
Pankrantz T (2008) Red tides close desal plants. Water Desalination Report 44
Parsons T, Maita Y, Lalli C (1984) A Manual of Chemical and Biological Methods for Seawater
Analysis. Pergamon Press, Oxford
Pearce G, Talo S, Chida K, Basha A, Gulamhusein A (2004) Pretreatment options for large scale
SWRO plants: case studies of UF trials at Kindasa, Saudi Arabia, and conventional
pretreatment in Spain. Desalination 167:175-189
Perl T, Bedard L, Kosatsky T, Hockin J, Todd E, McNutt L, Remis R (1990a) Amnesic Shellfish
Poisoning: a new clincal syndrome due to domoic acid. Can Diseas Weekly Rep 16:7-8
Perl T, Bedard L, Kosatsky T, Hockin J, Todd E, Remis R (1990b) An outbreak of toxic
encephalopathy caused by eating mussels contaminated with domoic acid. New Engl J
Med 322:1775-1780
Quay J, Hayashi K, Kudela R Watch wha you watch for: Degradation of particulate domoic acid
(pDA) field samples. Proc Sixth Symposium of Harmful Algae in the US
Ramsdell J, Zabka T (2008) In utero domoic acid toxicity: a fetal basis to adult disease in the
California sea lion (Zalophus californianus). Mar Drugs 6:262-290
Reifel K, Corcoran A, Cash C, Shipe R, Jones B (accepted) Effects of a surfacing effluent plume on
a coastal phytoplankton community. Cont Shelf
Res:http://dx.doi.org/10.1016/j.csr.2013.1004.1012
Richlen M, Morton S, Jamali E, Rajan A, Anderson D (2010) The catastrophic 2008-2009 red tide
in the Arabian gulf region, with observations on the identification and phylogeny of the
fish-killing dinoflagellate Cochlodinium polykrikoides. Harmful Algae 9:163-172
Rines J, Donaghay P, Dekshenieks M, Sullivan J, Twardowski M Thin layers and camouflage:
Hidden Pseudo-nitzschia spp. (Bacillariophyceae) populations in a fjord in the San Juan
Islands, Washington, USA. Mar Ecol Prog Ser 225:123-137
165
Rines J, McFarland M, Donaghay P, Sullivan J (2010) Thin layers and species-specific
characterization of the phytoplankton community in Monterey Bay, California, USA.
Cont Shelf Res 30:66-80
Roelke D, Grover J, Brooks B, Glass J, Buzan D, Southard G, Fries L, Gable G, Schwierzke-Wade L,
Byrd M, Nelson J (2011) A decade of fish-killing Prymnesium parvum blooms in Texas:
roles of inflow and salinity. J Plank Res 33:243-253
Rue E, Bruland K (2001) Domoic acid binds iron and copper: a possible role for the toxin
production by the marine diatom Pseudo-nitzschia Mar Chem 76:127-134
Sahagian D, Schwartz F, Jacobs D (1994) Direct anthropogenic contributions to sea level rise in
the twentieth century. Nature 367:54-57
Sala S, Sar E, Ferrario M (1998) Review of the materials reported as containing Amphora
coffeaeformia (Agardh) Kutzing in Argentina. Diatom Res 13:323-336
Schnetzer A, Jones B, Schaffner R, Cetinic I, Fitzpatrick E, Miller P, Seubert E, Caron D (Accepted)
Coastal upwelling linked to toxic Pseudo-nitzschia australis blooms in Los Angeles
coastal waters, 2005-2007. J Plank Res
Schnetzer A, Miller P, Schaffner R, Stauffer B, Jones B, Weisberg S, DiGiacomo P, Berelson W,
Caron D (2007) Blooms of Pseudo-nitzschia and domoic acid in the San Pedro Channel
and Los Angeles harbor areas of the Southern California Bight, 2003-2004. Harmful
Algae 6:372-387
Scholin C, Buck K, Britschbi T, Cangelosi G, Chavez F (1996) Identification of Pseudo-nitzschia
australis (Bacillariophyceae) using rRNA-targeted probes in whole cell and sandwich
hybridization formats. Phycologia 35:190-197
Scholin C, Gulland F, Doucette G, Benson S, Busman M, Chavez F, Cordaro J, DeLong R, De
Vogelaere A, Harvey J, Haulena M, Lefebvre K, Lipscomb T, Loscutoff S, Lowenstine L,
Marin III R, Miller P, McLellan W, Moeller P, Powell C, Rowles T, Silvagni P, Silver M,
Spraker T, Trainer V, Van Dolah F (2000) Mortality of sea lions along the central
California coast linked to a toxic diatom bloom. Nat 403:80-84
Sekula-Wood E, Schnetzer A, Benitez-Nelson C, Anderson C, Berelson W, Brzezinski M, Burns J,
Caron D, Cetinic I, Ferry J, Fitzpatrick E, Jones B, Miller P, Morton S, Schaffner R, Siegel D,
Thunell R (2009) Rapid donward transport of the neurotoxin domoic acid in coastal
waters. Nat Geosci 2:272-275
Seubert E, Gellene A, Howard M, Connell P, Ragan M, Jones B, Runyan J, Caron D (2013)
Seasonal and annual dynamics of harmful algae and algal toxins revealed through
weekly monitoring at two coastal ocean sites off southern California, USA. Environ Sci
Pollut R:DOI 10.1007/s11356-11012-11420-11350
166
Seubert E, Trussell S, Eagleton J, Schnetzer A, Cetinic I, Lauri P, Jones B, Caron D (2012) Algal
toxins and reverse osmosis desalination operations: Laboratory bench testing and field
monitoring of domoic acid, saxitoxin, brevetoxin and okadaic acid. Water Res 46:6563-
6573
Shipe R, Leinweber A, Gruber N (2008) Abiotic controls of potentially harmful algal blooms in
Santa Monica Bay, California. Cont Shelf Res 28:2584-2593
Sierra Beltran A, Palafox-Uribe M, Grajales-Montiel J, Cruz-Villacorta A, Ochoa J (1997) Sea bird
mortality at Cabo San Lucas, Mexico: Evidence that toxic diatom blooms are spreading.
Toxicon 35:447-453
Smayda T (1997) What is a bloom? A comentary. Limnol Oceanogr 42:1132-1136
Smith J, Pauley K, Cormeir P, Angus R, Odense P, O'Neil D, Quilliam M, Worms J (1991)
Population dynamics and toxicity of various species of Dinophysis and Nitzschia from the
southern Gulf of St. Lawerence. Can Tech Rep Fish Aquat Sci 1799:25
Smith V, Joye S, Howarth R (2006) Eutrophication of freshwater and marine ecosystems.
Limnology and Oceanography 51:351-355
Smith V, Tilman G, Nekola J (1999) Eutrophication: impacts of excess nutrient inputs on
freshwater, marine and terrestrial ecosystems. Environmental Pollution 100:179-196
Sournia A (1995) Red tide and toxic marine phytoplankton of the world ocean: an inquiry into
biodiversity. In: Lassus P, Arzul G, Erard E, Gentien P, Marcaillou C (eds) Harmful marine
algal blooms. Lavoisier, Intercept Ltd, Paris
Stauffer B, Gellene A, Schnetzer A, Seubert E, Oberg C, Sukhatme G, Caron D (2012) An
oceanographic, meteorological, and biological 'perfect storm' yields a massive fish kill.
Mar Ecol Prog Ser 468:231-243
Stehr C, Connell L, Baugh K, Bill B, Adams N, Trainer V (2002) Morphological, toxicological and
genetic differences among Pseudo-nitzschia (Bacillariophyceae) species in inland
embayments and outer coastal waters of Washington state, USA. J Phycol 38:55-65
Stramma L, Schmidtko S, Levin L, Johnson G (2010) Ocean oxygen minima expansions and their
biological impacts. Deep Sea Res Part I: Oceanogr Res Papers 57:587-595
Subba Rao D, Pan Y, Smith S (1995) Allelopathy between Rhizosolenia alata (Brightwell) and the
toxigenic Pseudo-nitzschia pungens f. multiseris (Hasle). In: Lassus P, Arzul G, Erad E,
Gentien P, Marcaillou (eds) Harmful Marine Algal Blooms. Lavoisier, Intercept Ltd, Paris
Sun J, Hutchins D, Feng Y, Seubert E, Caron D, Fu F (2011) Effects of changing pCO2 and
phosphate availability on domoic acid production and physiology of the marine harmful
bloom diatom Pseudo-nitzschia multiseries. Limnol Oceanogr 56:829-840
167
Sutherland C (2008) Diarrhetic shellfish toxins linked to local Dinophysis populations in the
California coastal waters of Monterey Bay. MS MS Thesis, University of California, Santa
Cruz, Santa Cruz
Tatters A, Fu F, Hutchins D (2012) High CO2 and silicate limitation synergistically increase the
toxicity of Pseudo-nitzschia fraudulenta. PLoS One 7:e32116
Thessen A, Bowers H, Stoecker D (2009) Intra- and interspecies differences in growth and
toxicity of Pseudo-nitzschia while using different nitrogen sources. Harmful Algae 8:792-
810
Thomas K, Harvey J, Goldstein T, Barakos J, Gulland F (2010) Movement, dive behavior and
survival of California lions (Zalophus californianus) posttreatment for domoic acid
toxicosis. Mar Mammal Sci 26:36-52
Torres de la Riva G, Johnson C, Gulland F, Langlois G, Heyning J, Rowles T, Mazet J (2009)
Association of an unusual marine mammal mortality event with Pseudo-nitzschia spp.
blooms along the southern California coastline. J Wild Diseas 45:109-121
Torrey H (1902) An unusual occurrence of dinoflagellata on the California coast. Amer Nat
36:187-192
Trainer V, Bates S, Lundholm N, Thessen A, Cochlan W, Adams N, Trick C (2012) Pseudo-
nitzschia physiological ecology, phylogeny, toxicity, monitoring and impacts on
ecosystem health. Harmful Algae 14:271-300
Utermohl H (1958) Zur gewassertypenfrage tropischer seen. Verhandlungen Internationalen
Vereinigung Limnologie 13:236-251
Van Dolah F (2000) Marine algal toxins: Origins, health effects, and their increased occurrence.
Environ Health Persp 108:133-141
Vardon D, Clark M, Ladner D (2011) The potential of laser scanning cytometry for early warning
of algal blooms in desalination plant feedwater. Desalination 277:193-200
Verliefde A, Van Vliet N, Amy G, Van der Bruggen B, van Dijk J (2006) A semi-quantitative
method for prediction of the rejection of uncharged organic micropollutants with
nanofiltration. Water Practice Tech 1:doi: 10.2166/WPT.2006084
Vigilant V, Silver M (2007) Domoic acid in bethic flatfish on the continental shelf of Monterey
Bay, California, USA. Mar Biol 151:2053-2062
Wang Z, King K, Ramsdell J, Doucette G (2007) Determination of domoic acid in seawater and
phytoplankton by liquid chromatography-tandem mass spectrometry. J Chromatogr A
1163:169-176
168
Wang Z, Maucher-Fuquay J, Fire S, Mikulski C, Haynes B, Doucette G, Ramsdell J (2012)
Optimization of solid-phase extraction and liquid chromatography-tandem mass
spectrometry for the determination of domoic acid in seawater, phytoplankton and
mammalian fluids and tissues. Anal Chim Acta 715:71-79
Wells M, Trick C, Cochlan W, Hughes M, Trainer V (2005) Domoic acid: the synergy of iron,
copper and the toxicity of diatoms. Limnol Oceanogr 50:1908-1917
Westgard J, Hunt M (1973) Use and interpretation of common statistical tests in method-
comparison studies. Clin Chem 19:49-57
Wilkerson F, Dugdale R, Hogue V, Marchi A (2006) Phytoplankton blooms and nitrogen
productivity in San Francisco Bay. Estuar Coasts 29:401-416
Work T, Barr B, Beale A, Fritz L, Quilliam M, Wright J (1993) Epidemiology of domoic acid
poisoning in brown pelicans (Pelecanus occidentalis) and Brandt's cormorants
(Phalacrocorax penicillatus) in California. J Zoo Wildl Med 24:54-62
Wright J, Bird C, de Freitas A, Hampson D, McDonald J, Quilliam M (1990) Chemistry, biology
and toxicology of domoic acid and its isomers. Can Diseas Weekly Rep 16:21-26
Wright J, Boyd R, deFreitas A, Falk M, Foxall R, Jamieson W, Laycock M, McCulloch A, McInnes
A, Odense P, Pathak V, Quilliam M, Ragan M, Sim P, Thibault P, Walter J, Gilgan M,
Richard D, Dewar D (1989) Identification of domoic acid, a neuroexcitatory amino acid,
in toxic mussles from eastern Prince Edward Island. Can J Chem 67:481-490
Xi D, Ramsdell J (1996) Glutamate receptors and calcium entry mechanisms for domoic acid in
hippocampal neurons. NeuroReport 7:1115-1120
Yasumoto T, Murata M, Oshima Y, Sano M, Matsumoto G, Clardy J (1985) Diarrhetic shellfish
toxins. Tetrahedron 41:1019-1025
Yasumoto T, Oshima Y, Sugawara W, Fukuyo Y, Oguri H, Igarashi T, Fujita N (1980) Identification
of Dinophysis fortii as the causative organism of diarrhetic shellfish poisoning. B Jpn Soc
Sci Fish 46:1405-1411
Zhang M, Jiang S, Tanuwidjaja D, Voutchkov N, Hoek E, Cai B (2011) Composition and variability
of biofouling organisms in seawater reverse osmosis desalination plants. Appl Environ
Microbiol 77:4390-4398
Zingone A, Enevoldsen H (2000) The diversity of harmful algal blooms: a challenge for science
and management. Ocean Coast Manage 43:725-748
169
Appendix 1: Chasing elusive harmful algal blooms
Article in the Catalina Marine Society Publication, Ocean Bights, Winter 2012 Vol. 3 No. 2
Living in southern California it is hard to ignore our closest neighbor - the Pacific Ocean.
The beaches offer us a weekend getaway from cluttered cubicles and hectic schedules.
Maintaining and caring for our beaches have become a source of pride for some seaside towns
with organized beach cleanups, the banning of plastic bags and other wasteful materials. While
these bans and cleanups are beneficial, please remember that a healthy ocean relies on more
than just the items we humans are able to see. The microscopic organisms living in our water
have a major influence on the ecology of the oceans and our impact on them is often not
readily noticeable. However, their impact on us is noticeable when one species suddenly
dominates or some microbial constituents produce toxins that affect marine life. These
episodes of stimulated productivity are termed harmful algal blooms or HABs.
Microscopic organisms tend to get a bad reputation. Clever advertising has spurred
obsessions over unseen bacteria and viruses that may threaten our loved ones’ health, boosting
the sales of anything labeled “anti-bacterial” or “anti-microbial”. However, the ocean food web
is anchored by a host of harmless microbes, and theproduction of thousands of protists,
bacteria and viruses ultimately determine the success of our fisheries. The interaction among
these three groups is termed the “microbial loop” as their associations are not linear but tightly
coupled to one another.
The term protist refers to microscopic eukaryotic single-celled organisms. Eukaryotic
cells are typified by the presence of a nucleus and other membrane organelles and the ability to
170
reproduce sexually. Within the protistan group there are autotrophic organisms, also known as
phytoplankton and unicellular algae, which utilize photosynthesis to produce their own food,
heterotrophic organisms that consume autotrophs, and mixotrophic organisms that can switch
between autotrophy and heterotrophy depending on conditions. The work performed in Dave
Caron’s laboratory at the University of Southern California is focused on understanding the
complex ecological interactions present in the protistan fraction of the microbial loop.
The growth of the protistan community can be stimulated in multiple ways. Direct nutrient
input will stimulate photosynthesis, thereby increasing the number of organisms capable of
autotrophy. Stimulation of the portion of the protistan community capable of heterotrophy is a
bit more complicated as the preferred prey item will need to be stimulated first. Heterotrophs
that consume bacteria will flourish when bacterial production is increased and heterotrophs
consuming other protists will follow any increase in abundance of their prey.
The normal protistan community will host a myriad of different species at any given
time, with the dominant members changing as the input of nutrients, light availability and prey
abundance fluctuates over time. Harmful algal blooms (HABs) occur when the microbial loop is
somehow disrupted. In a water body that hadpreviously hosted a few hundred species of
protists simultaneously, one species suddenly dominates and will come close to comprising
100% of the protistan community present.
The autotrophic diatom, Pseudo-nitzschia, produces a neurotoxin called domoic acid
(DA) that is capable of causing illness in higher animals. When ingested by humans via
contaminated seafood, DA causes a syndrome termed amnesic shellfish poisoning with
symptoms ranging from nausea to loss of memory and even death. Thanks to successful
171
monitoring of seafood products by state health departments DA has not caused a human death
since the first identified outbreak in Prince Edward Island, Canada, in which 3 people died and
over 100 were sickened in 1987. DA amounts within cells are low compared to the
concentrations necessary to sicken humans or wildlife; it becomes a health threat when it is
concentrated by bioaccumulation through the food web. A single cell containing DA cannot
sicken us, but when shellfish and fish capable of consuming Pseudo-nitzschia feed, they
concentrate the toxin in their bodies. When marine birds and mammals consume the
concentrated toxin, they become ill. The effect of this toxin on marine food webs plays out
almost every spring on our television sets when the local news reports increases in the amount
of California sea lion strandings and in marine bird illnesses. The negative impacts DA
production has on the local ecosystem led to classification of Pseudo-nitzschia as a HAB
organism. Unlike other HAB organisms that may reach significantly high cell densities and
compose almost the entire protistan community, Pseudo-nitzschia does not have to be a
dominant member of the community to pose a risk.
Most HAB research has been limited to describing an event once it has begun, that is,
after reports of discolored water or sickened animals have reached scientists. However,
identifying conditions conducive to HAB development is essential for understanding HABs.
Pseudo-nitzschia is a common member of the protistan community in southern California and it
is present for most of the year, but DA is typically only detected in the spring. Stimulation of
Pseudo-nitzschia growth does not necessarily stimulate DA production. The work I have
performed at USC in completion of my doctoral thesis has dealt with understanding DA
production by Pseudo-nitzschia in southern California. We use a combination of laboratory
172
experiments and open-ocean sampling to investigate the factors that lead to DA generation.
Experiments with laboratory cultures can help determine which factors are important to DA
production and subsequently impact the design of expensive open-ocean sampling and HAB
monitoring programs in our area.
DA is a small molecule, 33% of which is composed of nitrogen. Nitrogen availability
directly impacts an organism’s growth; it is a component of many essential molecules (i.e.,
amino acids) and is needed for basic cell metabolism. Nitrogen availability is one of the main
factors that can limit autotrophic growth in the ocean.
When Pseudo-nitzschia uses available nitrogen to make DA, that nitrogen is not
available for other cell functions. To sustain autotrophic growth, the availability of carbon,
nitrogen, phosphate and silicate must be present in a molar ratio of 106 carbon: 16 nitrogen: 1
phosphorus: 16 silicate (Redfield ratios). When nitrogen is in short supply, the nitrogen to
phosphate ratio will be under 16:1 and nitrogen is said to be the limiting nutrient. Once
nitrogen availability has been increased and the ratio of nitrogen to phosphorous is 16:1,
neither is limiting.
The first hypothesis behind DA production concerned nitrogen availability in excess of
concentrations necessary for growth, hence ratios of nitrogen to phosphorous will be higher
than 16:1 and nitrogen to silicate ratios higher than 1:1. Phosphorous and silicate
concentrations would be the elements limiting Pseudo-nitzschia growth in this situation.
Laboratory experiments confirmed that in times of phosphate and silicate limitation DA
production increased, and the amount of increase in DA production differed according to the
nitrogen source available.
173
However, ocean samples we collected in the Los Angeles area in which DA was
measurable did not correspond to limited phosphate and silicate concentrations. In 2008 our
laboratory at USC joined four other universities in southern California as a part of the Southern
California Coastal Ocean Observing System (SCCOOS) HAB monitoring program. Weekly
samples are collected from Newport Pier and analyzed for protistan community composition,
chlorophyll concentrations (which are a measure of total autotrophic protist growth), nutrient
and DA concentrations. The dataset of weekly samples analyzed for DA over the course of a few
years for one location showed a pattern of DA presence corres-ponding to upwelling events.
Upwelling is a phenomenon in which deep water is brought to the surface. The process
occurs when persistent downcoast winds move surface waters offshore, away from the coast.
Deep, cold, nutrient-rich waters from offshore replace the surface waters. The combination of
persistent downcoast winds and a sharp decrease in coastal water temperature help scientists
identify an upwelling event. In the week prior to an upwelling event DA was undetectable
although Pseudo-nitzschia was present in the Newport Pier sample. However, DA was
detectable after an upwelling event.
Correlation of upwelling events and DA detection at Newport Pier is important, but the
primary research goal is to identify specific conditions that stimulate DA production. We are
now exploring the component that drives the correspondence of upwelling events and the
presence of DA – is it the introduction of nutrients in upwelled water that stimulates Pseudo-
nitzschia to produce DA or are DA-producing Pseudo-nitzschia cells present in the upwelled
water? Upwelled water will have nutrients proportioned close to the ideal Redfield ratio.
Answering these questions require us to monitor not only what is occurring at coastal locations
174
such as Newport Pier but also analysis of samples taken offshore, below the surface, where the
upwelled water originates.
The primary oceanographic tool for sampling offshore and at specific depths is a CTD
rosette. It is an instrument in which a carousel of bottles is placed in a circle around a series of
sensors on the interior. The majority of rosettes are outfitted with Conductivity, Temperature
and Depth (CTD) sensors that inform scientists in near-real time of temperature, salinity
(derived from the conductivity and temperature), and depth of the water as the carousel is
lowered. A sensor for chlorophyll fluorescence (such as the CMS maintains at Two Harbors) can
also be included on the CTD, allowing scientists on the boat to see the relative chlorophyll
concentration present in the water column as the rosette travels down. Chlorophyll
concentration is a proxy for autotrophic protistan abundance and since Pseudo-nitzschia is
autotrophic it will most likely be present in locations where autotrophic cell abundance is high.
Elevated chlorophyll concentrations are used to identify sample locations ideal for our research
objective. As the rosette is brought back aboard the ship it will pause at the depth identified
during the downcast as an area with high chlorophyll concentration and one or more bottles
will sample water from that depth. Niskin bottles are plastic tubes in which the top and bottom
of the bottle can be pulled open by a rope. The top and bottom are left open as the rosette of
bottles is lowered through the water column. A scientist on board the ship can press a button
that releases the hold on the rope closes the bottle. The now closed Niskin bottle contains
water from the specific depth we suspect the Pseudo-nitzschia inhabits. Once on board the
sample is collected and filtered onto glass-fiber filters for analysis.
175
Times in which upwelling events are likely to occur can be identified by weather
forecasts and there are specific seasons in which upwelling favorable winds can occur. In the
southern California area, upwelling events are most likely to occur in spring; this is helpful for
scheduling the sampling but there are still several months over which to spread the data
collection effort.
The time and monetary commitment involved in maintaining consistent monitoring and
sampling programs is prohibitive and heavily impacts the number and location of samples
collected. Narrowing down the specific time period in which the offshore sampling is most
effective allows scientists to focus their efforts (and money) to the collection and analysis of
samples that will be best able to meet their research objective.
In 2011, the SCCOOS sponsored the development of a Community HAB Watch Program
with additional support from the Centers for Ocean Sciences Education Excellence (COSEE)
West and scientific expertise from our laboratory at USC. The program involves eleven informal
science centers in southern California, spanning the coast from the Ocean Institute in Dana
Point to the Ty Warner Sea Center in Santa Barbara and including sites far offshore such as
Anacapa Island, maintained by the Channel Islands NationalMarine Sanctuary. Each of the
centers was provided with training HAB monitoring. The development of this program allowed
for a stronger connection between university researchers and the science centers, armed the
centers with current and comprehensive information on HABs present in southern California for
them to pass on to visitors and provided researchers with priceless information crucial to
expanding our knowledge of the timing and spatial components of HABs. Expanding the
coverage of coastline monitoring sites and community involvement becomes implicitly
176
important and essential to assisting scientists in meeting their research goals. It allows
information to be collected in more areas than the scientists can realistically visit on a regular
basis and provides comparative data on the spatial component of blooms. For example, in 2011
there was a significant bloom of the dinoflagellate Lingulodinium polyedrum identified in Dana
Point by the Ocean Institute and reported to USC researchers through the Community
HABWatch program. At Newport pier, the monitoring location maintained by USC,
Lingulodinium was not present and we would not have known about the bloom just a short
distance away at Dana Point without the Community HABWatch program. Results from the
HABWatch program can be obtained from www.sccoos.org.
Abstract (if available)
Abstract
The term harmful algal bloom (HAB) is used to describe any bloom of microalgae that has a detrimental impact to the local ecosystem and/or economy. The impacts of a HAB to an ecosystem can include death or injury to local wildlife through the production of toxins, decreased oxygen concentrations, physical damage, decreased light availability or food web disturbance. The economic impacts can be reduction in tourism, human illness, reduced fishing effort or interruption of desalination plant operations. The occurrence and intensity of HABs have been increasing globally during the past few decades, whether this increase can be attributed to enhanced awareness and monitoring, or to a dramatic upswing in the development of HAB events remains unresolved. ❧ A variety of HAB-forming species of microalgae occur in southern California, and several of these species are known to produce potent neurotoxins. The impact of algal toxin presence on both the intake and reverse osmosis (RO) desalination process and whether or not the naturally occurring algal toxins can pass through the RO membrane and into the desalination product was addressed through bench-scale RO experiments and monitoring for algal toxins at a pilot RO desalination plant. Concentrations exceeding maximal values previously reported during natural blooms were used in the laboratory experiments, with treatments comprised of 50 µg/L of domoic acid (DA), 2 µg/L of saxitoxin (STX) and 20 µg/L of brevetoxin (PbTx). None of the algal toxins used in the bench-scale experiments were detectable in the desalinated product water. Monitoring for intracellular and extracellular toxin concentrations of DA, STX, PbTx and okadaic acid (OA) within the intake and desalinated water from a pilot RO desalination plant in El Segundo, CA, was conducted from 2005 to 2009. During the five-year monitoring period, DA and STX were detected sporadically in the intake waters but never in the desalinated water. PbTx and OA were not detected in either the intake or desalinated water. The results of this study demonstrate the potential for HAB toxins to be inducted into coastal RO intake facilities, and the ability of typical RO operations to effectively remove these toxins. ❧ The ability to accurately and rapidly identify an emerging HAB event is of high importance. Monitoring of HAB species and other pertinent chemical/physical parameters at two piers in southern California, Newport and Redondo Beach, was used to investigate the development of a site-specific bloom definition for identifying emerging DA events. The neurotoxin DA is produced by the chain forming diatom Pseudo-nitzschia, and it is the most common HAB organism in southern California. Emphasis was given to abundances of the P. seriata size category of Pseudo-nitzschia due to the prevalence of this size class in the region. P. seriata bloom thresholds were established for each location based on deviations from their respective long-term mean abundances, allowing the identification of major and minor blooms. Sixty five percent of blooms identified at Newport Beach coincided with measurable DA concentrations, while 36% of blooms at Redondo Beach coincided with measurable DA. Bloom definitions allowed for increased specificity in multiple regression analysis of environmental forcing factors significant to the presence of DA and P. seriata. The strongest relationship identified was between P.seriata abundances two weeks following upwelling events at Newport Beach. ❧ Blooms of Pseudo-nitzschia can develop at depth in offshore waters not encompassed by coastal HAB monitoring programs. California sea lions are predominately associated with DA mortality events on the US west coast undoubtedly due to their large population sizes and overlapping distribution with Pseud-nitzschia. Quantifying the amount of DA in these animals and correlating this information with the presence of DA in phytoplankton and the local food web has become a research focus for many scientists. However differences in materials, equipment, technical capability, budgets and objectives of the various groups and/or agencies involved in this work have influenced the DA quantification platforms employed. The performance of two commercially available enzyme-linked immunosorbent assays for the analysis of DA in a spectrum of California sea lion body fluids was compared to the results obtained with liquid chromatography-mass spectrometry of the same samples. The results indicated differences among these approaches, presumably owing to matrix effects (particularly urine) and antibody reactivities. This information implies that care should be taken in attempting to compare datasets generated using different analytical platforms and interpreting the results of published studies. ❧ The Orange County Sanitation District diverted flow of secondarily treated effluent from a discharge pipe located 8.0 km offshore at 60 m depth to a pipe located 1.6 km from shore at 17 m depth for three weeks in September of 2012. Two incubation experiments were performed to examine the influence of treated effluent at various dilutions on natural, coastal phytoplankton communities, the first initiated a week prior to the diversion ('Pre-Diversion') and the second initiated a week after the start of the diversion ('Mid-Diversion'). The overall community response observed in both experiments following effluent addition was an increase in diatom and picoeukaryote abundances, a decrease in picophotocyanobacteria and a dramatic increase in heterotrophic bacteria abundance. The 1:10 effluent additions yielded significant increases in chlorophyll a concentrations, although the Pre-Diversion 1:10 experiments exhibited a lag in response to effluent addition. The DA producing diatom Pseudo-nitzschia was present throughout both experiments, however DA production was only detected in the Mid-Diversion experiment. The highest concentration of DA measured, 0.42 ± 0.057 µg/L coincided with phosphate and silicate concentrations below the detection limit of the method, suggesting limitation by these macronutrients.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Phytoplankton bloom initiation in the Southern California Bight: a multi-year local and regional analysis
PDF
Examining potential triggers of algal blooms and harmful algae in the Southern California bight region
PDF
B-vitamins and trace metals in the Pacific Ocean: ambient distribution and biological impacts
PDF
Future impacts of warming and other global change variables on phytoplankton communities of coastal Antarctica and California
Asset Metadata
Creator
Seubert, Erica Lee
(author)
Core Title
Distribution and impact of algal blooms leading to domoic acid events in southern California
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Marine and Environmental Biology
Publication Date
08/06/2013
Defense Date
05/22/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
brevetoxin,California sea lion,desalination,domoic acid,ELISA,HAB monitoring,harmful algal blooms,OAI-PMH Harvest,okadaic acid,Pseudo-nitzschia,saxitoxin
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Caron, David A. (
committee chair
), Hutchins, David A. (
committee member
), Sañudo-Wilhelmy, Sergio A. (
committee member
), Sukhatme, Gaurav S. (
committee member
)
Creator Email
erica.seubert@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-318562
Unique identifier
UC11293554
Identifier
etd-SeubertEri-1977.pdf (filename),usctheses-c3-318562 (legacy record id)
Legacy Identifier
etd-SeubertEri-1977.pdf
Dmrecord
318562
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Seubert, Erica Lee
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 a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
brevetoxin
California sea lion
desalination
domoic acid
ELISA
HAB monitoring
harmful algal blooms
okadaic acid
Pseudo-nitzschia
saxitoxin