ARTICLE IN PRESS

Continental Shelf Research 28 (2008) 2584–2593

Contents lists available at ScienceDirect

Continental Shelf Research

journal homepage: www.elsevier.com/locate/csr

Abiotic controls of potentially harmful algal blooms

in Santa Monica Bay, California
R.F. Shipe a,, A. Leinweber b, N. Gruber b,1
a
UCLA Department of Ecology and Evolutionary Biology, 621 Charles Young Drive South, Los Angeles, CA 90095, USA
b
UCLA Institute of Geophysics and Planetary Physics, Los Angeles, USA

a r t i c l e in f o a b s t r a c t

Article history: Despite the increasing occurrence of harmful phytoplankton blooms along the North American west
Received 9 April 2007 coast, records of phytoplankton populations and related environmental conditions are uncommon. In
Received in revised form this study, twice monthly measurements in the upper 50 m are used to assess physico-chemical
15 April 2008
conditions contributing to the growth of potentially harmful bloom taxa over two annual cycles
Accepted 11 August 2008
(2004–2005) in the Santa Monica Bay, California. Results were compared to the predictions of the
Available online 22 August 2008
Intaglio model [Smayda, T.J., Reynolds, C.S., 2001. Community assembly in marine phytoplankton:
Keywords: application of recent models to harmful dinoflagellate blooms. Journal of Plankton Research 23,
Algal blooms 447–461.] of phytoplankton community assembly. Potentially harmful taxa were present in every
Phytoplankton
surface sample and were numerically dominant during the largest observed blooms, contributing up to
Ecological associations
92% of the total phytoplankton abundance 45 mm. Large interannual variation was observed in the
Coastal oceanography
dominant taxa and bloom seasonality; Pseudo-nitzschia sp. dominated blooms in early 2004 (February
Regional Index: and April), whereas Prorocentrum micans and Lingulodinium polyedrum blooms occurred in May and
USA September of 2005, respectively. The Pseudo-nitzschia sp. blooms were associated with elevated nitrate,
California dissolved silicon and phosphate concentrations throughout the euphotic zone; the first bloom followed
Santa Monica Bay a strong upwelling and the second occurred during the onset of seasonal stratification. In contrast, the
blooms of P. micans were associated with highly stratified, low nutrient waters. Multivariate analysis
supports the roles of temperature, mixed-layer depth and nutrient concentrations as primary controls of
bloom growth, following the conceptual Intaglio model. The strong presence of potentially harmful
bloom species in the Santa Monica Bay during this study appears unusual in comparison to limited
studies over the last several decades.
& 2008 Elsevier Ltd. All rights reserved.

1. Introduction these taxa can be extensive; for example, a bloom of Pseudo-

In recent decades, the occurrence of harmful algal blooms
(HABs) has increased in global coastal waters (Glibert et al., 2005)
as well as along the North American west coast (Horner et al.,
1997). These events can cause harm to marine life (e.g. Work et al.,
1993; Scholin et al., 2000) and to human health, with associated
detrimental economic impacts (Anderson et al., 1993). Common
bloom-forming taxa in CA coastal waters include toxigenic
diatoms of the genus Pseudo-nitzschia and the red tide forming
armored dinoflagellates, Lingulodinium polyedrum (formerly
known as Gonyaulax polyedra) and Prorocentrum micans (Allen,
1941, review by Horner et al., 1997). Accumulations of cells of
 Corresponding author. Tel.: +1 310 7944903.
E-mail addresses: rshipe@ucla.edu (R.F. Shipe), leinweber@igpp.ucla.edu
(A. Leinweber), nicolas.gruber@env.ethz.ch (N. Gruber).
1
Now at Institute of Biogeochemistry and Pollutant Dynamics, ETH Zürich,
Zürich, Switzerland.
nitzschia extended from southern California to Washington in fall
1991 (Villac et al., 1993) and a L. polyedrum bloom in 1995
extended from the Baja peninsula to the Monterey Bay (Kudela
and Cochlan, 2000). Although the mechanisms of bloom forma-
tion are important to understand from the perspectives of ecology
and water quality management, blooms in the Southern California
Bight basins have generally been sampled opportunistically.
Therefore, relationships between harmful blooms and annual
cycles of oceanographic conditions are poorly understood.
An ecological theory relating coastal marine phytoplankton
species assemblages to abiotic conditions was first developed by
Margalef (1978), Margalef et al. (1979) and has been refined by
Smayda and Reynolds (2001) in the Intaglio model. According to
this conceptual model, nutrient accessibility and energy (a
combination of mixing depth and irradiance) are the principal
abiotic factors that structure assemblages. Diatom dominance is
expected under ‘‘temperate’’ conditions, which are characterized
by deep mixing and high nutrient concentrations. Nine different

0278-4343/$ - see front matter & 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.csr.2008.08.003
 ARTICLE IN PRESS

R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593 2585

harmful dinoflagellate types occupy the remaining matrix of
nutrients and energy. According to this model, P. micans is
classified as ‘‘Type II’’ and is a summer bloom taxa, expected to
occur under somewhat elevated nutrient conditions. L. polyedrum
is classified as ‘‘type V’’ and is expected to bloom under lower
nutrient concentrations, in more shallow mixed layers. Although
the Intaglio model makes predictions of broad taxonomic groups
based on qualitative nutrient and energy conditions, it is a useful
construct for evaluation of mechanisms controlling community
structure, particularly those giving rise to HAB.
In this study, we present a record of harmful bloom taxa and
test these predictions of the Intaglio model by relating their
abundances to environmental conditions using time-series mea-
surements in the Santa Monica Bay (SMB), CA over two annual
cycles. The SMB is a partially enclosed inshore basin located
adjacent to the Los Angeles metropolitan area and circulation
patterns and seasonal and interannual hydrographic variability
are well described (Hickey, 1992; Hickey et al., 2003; Nezlin et al.,
2004). Strong local upwelling events associated with temperature
minima occur in the winter and spring (Reid et al., 1958; Hickey,
1992; Gruber et al., in prep.), whereas stratification develops over
the summer and autumn. The results of upwelling events,
particularly those that occur to the west, can be advected
horizontally into the SMB. Although phytoplankton dynamics
are likely regulated by a combination of local forcing and larger
scale circulation and transport, we focus here on macronutrient
concentrations and local hydrographic conditions, which
have been shown to control biological production in this region
(Eppley and Renger, 1988; Nezlin and Li, 2003). Results
are evaluated in the context of the Intaglio model, which also
suggests that nutrient and energy conditions control the phyto-
plankton assemblage at the level of broad taxonomic groups.
Further, for harmful bloom-forming taxa, the Intaglio model
explicitly predicts that there will be increases in both nutrient
concentrations and mixing that support L. polyedrum, P. micans
and, Pseudo-nitzschia spp., respectively (Smayda and Reynolds,
2001, 2003).
2. Methods
Time-series measurements of potentially harmful phytoplank-
ton taxa and environmental conditions were conducted at the
Santa Monica Bay Observatory (SMBO) interdisciplinary mooring
site: 33155.900 N, 118142.937 W. This station is part of the
Southern California Coastal Ocean Observing System (SCCOOS,
http://www.sccoos.org/) and forthcoming publications will ad-
dress diurnal studies of atmospheric and oceanic dissolved
inorganic carbon and longer term phytoplankton dynamics. We
report here data from 54 sampling dates between January 2004
and December 2005 (Tables 1 and 2). Cruises were generally
conducted on a twice monthly basis, with an additional cruise
during a large phytoplankton bloom in February 2004. Sampling
of the upper water column aboard the R/V Sea World UCLA began
at the mooring site at approximately 9:30 a.m. local time.
Temperature, pressure and conductivity were measured during
the downcast of a Seabird SBE9/11+conductivity temperature
depth (CTD) profiler. During routine maintenance of the Seabird
profiler, between May 24, 2005 and October 25, 2005, an Applied
Microsystems MicroCTD unit was used. Mixed-layer depth was
defined as the depth at which potential density changed by
X0.01 kg m 3 over 3 m (Brainerd and Gregg, 1995). On some dates,
heavy precipitation resulted in a large density gradient over the
first 2–3 m. Near-surface temperatures were also measured at 1 h
intervals using a SeaBird SBE9 attached to the moored buoy,
beginning in April 2005. Niskin bottle samples were collected at 6

Table 1
Cruise number, sampling date in 2004, mixed-layer depth (m), total cell abundance
(cells l 1) and percent contribution of Pseudonitzschia spp., Lingulodinium poly- Table 2
edrum and Prorocentrum micans to total phytoplankton abundance in surface As in Table 1 but for 2005
waters (1 m)
Date mld Cell % Pseudo. % L. % P.
Date mld Cell % Pseudo. % L. % P. abundance spp. polyedrum micans
(m) abundance spp. polyedrum micans
26 4 Jan 68 570 2 15 0
1 20 Jan 29 10,400 3 1 0 27 19 Jan 36 50,500 12 0 o1
2 5 Feb 21 5000 30 16 0 28 25 Jan 12 4360 5 1 1
3 13 Feb 35 214,000 63 1 o1 29 1 Feb 32 3650 5 1 0
4 17 Feb – 73,000 33 1 0 30 15 Feb 31 11,800 o1 1 1
5 2 Mar 38 19,500 16 1 3 31 1 Mar – – – – –
6 16 Mar 11 7160 37 15 1 32 15 Mar 9 12,400 0 2 49
7 30 Mar 9 169,000 71 1 6 33 22 Mar 12 – – – –
8 14 Apr 10 141,000 27 1 1 34 12 Apr 13 – – – –
9 27 Apr 12 181,000 56 o1 o1 35 15 Apr – – – – –
10 11 May 9 39,600 5 19 2 36 21 Apr 12 6510 5 1 7
11 25 May 20 32,400 1 50 0 37 26 Apr 13 760 0 0 37
12 8 Jun 11 24,000 19 38 0 38 10 May 5 50,400 o1 2 86
13 22 Jun 15 37,400 3 38 0 39 24 May 8 81,300 0 3 91
14 6 Jul 9 2040 8 48 0 40 7 Jun 10 2980 5 2 3
15 20 Jul – 3970 1 1 0 41 21 Jun 9 4950 o1 3 27
16 10 Aug 8 3820 3 15 8 42 6 Jul – 6460 4 o1 1
17 24 Aug 5 750 0 3 6 43 21 Jul 5 23,200 o1 0 3
18 7 Sept 6 2190 0 14 1 44 2 Aug 5 34,900 0 52 8
19 21 Sept 17 14,000 13 1 0 45 16 Aug 6 9680 0 7 1
20 5 Oct 15 17,000 o1 23 0 46 30 Aug 8 9540 o1 32 2
21 19 Oct 17 360 0 15 0 47 13 Sept 9 3,59,900 0 90 2
22 2 Nov 24 1250 0 18 9 48 27 Sept 11 43,300 o1 54 1
23 16 Nov 9 1078 2 7 0 49 11 Oct 10 10,810 0 57 2
24 30 Nov 22 3740 1 31 6 50 25 Oct 13 13,900 0 19 1
25 14 Dec 34 210 0 o1 0 51 8 Nov 9 8010 0 36 0
52 22 Nov 18 23,900 o1 47 1
A percentage of ‘‘o1’’ indicates that cells were present, whereas ‘‘0’’ indicates that 53 6 Dec 23 6020 2 6 1
no cells of this taxa were observed. ‘‘–’’ indicates parameters that were not 54 20 Dec – – – – –
measured.
 ARTICLE IN PRESS

2586 R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593

depths, alternating between 1, 5, 15, 25, 35, 50 m and 1, 10, 20, 30,
40, 50 m (samples from approximately 1 m hereafter referred to as
surface samples). Water samples for determination of dissolved
silicon concentrations, cell counts and chlorophyll a concentra-
tions were drained into polycarbonate bottles and stored in
coolers for transport back to the laboratory. Water samples for
determination of phosphate, nitrate and nitrite concentra-
tions were drained into 30 ml polypropylene tubes and frozen at
14 1C for up to 5 months before analysis at the Scripps Insti-
tute of Oceanography analytical laboratory. Detection limits
were 0.02 mM for phosphate, 0.2 mM for nitrate and 0.02 mM for
nitrite.
For phytoplankton counts, a 1000 ml sample was filtered onto a
47 mm diameter 3 mm Nuclepore brand polycarbonate filter and
resuspended in 10 ml of the filtered seawater with 1 ml of a
neutral buffered formaldehyde solution (final concentration 4.5%
by volume). Within 4 months, phytoplankton (diameter 45 mm)
were enumerated using light and phase contrast illumination on
an Olympus BX60 microscope using a 20  objective lens. Cells
were identified to the lowest taxonomic grouping possible
following Tomas (1996). Thus, armored dinoflagellates and
diatoms including the potentially harmful taxa of interest were
most easily identified. Pseudo-nitzschia species cannot be distin-
guished to species without electron microscopic methods and are,
therefore, classified here as Pseudo-nitzschia spp. Most of the
observed cells were of the P. pungens/multiseries group, having
pointed ends and chains overlapping in a stepwise formation.
Cells were enumerated in at least three transects through a
Sedgewick-Rafter chamber and at least 100 cells per sample and
means for each sample are presented here. Differences in cell
abundances between transects were generally less than 15% for
abundant species.

Fig. 1. Time-series contours of physical parameters (a) temperature (1C), (b) salinity and (c) sigma-t (kg m 3) at the SMBO mooring (33155.900 N, 118142.937 W). White line
on potential density plot indicates depth of the mixed layer. Dark and light salinity contour lines indicate intervals of 1.0 and 0.1, respectively.
 ARTICLE IN PRESS
R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593
For phytoplankton biomass, chlorophyll a concentrations were
determined on 1000 ml samples filtered onto Whatman GF/F
filters. Chlorophyll a was extracted in 90% ethanol and determined
spectrophotometrically based on the methods of Ryther and
Fig. 2. Comparison of near-surface temperature measurements at SMBO. Solid line
is daily mean temperature from moored instrumentation and points denote data
collected during time-series cruises.
Fig. 3. Time series of nutrient concentrations (a) nitrate (mM), (b) dissolved silicon (mM) and (c) phosphate (mM). The contour interval for nitrate and dissolved silicon plots
is 2 mM, with an additional dotted line for 1 mM.
2587
Yentsch (1957) and Odum et al. (1958). Chlorophyll a standards
(Anacystis nidulans, Sigma Chemical) were prepared in ethanol and
analyzed annually.
To evaluate environmental conditions corresponding to
blooms of potentially harmful bloom forming taxa, correlation
analysis was performed with and without a lag of 2 weeks for
log-transformed surface abundances of each HAB taxa and
salinity, density, temperature, nitrate, phosphate, nitrite, dissolved
silicon and mixed-layer depth. Further, multivariate analysis
was done on the data set comprising the 46 sampling dates
with complete data for salinity, density, temperature, nitrate,
phosphate, nitrite, dissolved silicon and mixed-layer depth.
A value of zero was used when concentrations were below
detection limits. Principal components analysis (PCA) arranges
samples along orthogonal axes and was based on a correlation
matrix (since the variables had very different variances) with no
rotation. All statistical analyses were performed using Minitab
version 14.
 ARTICLE IN PRESS
2588 R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593
Fig. 4. Time series of cumulative abundances of potentially harmful bloom species and other diatoms and dinoflagellates in near-surface waters at the SMBO mooring.
3. Results
3.1. Physicochemical factors
Upper water column physical conditions showed strong
seasonal cycles during the 2-year-study. In the winter, the surface
water column was fairly well mixed and near-surface tempera-
tures ranged from 13 to 15 1C (Fig. 1a). Surface warming in the
spring and summer increased stratification, which persisted into
the autumn, and annual surface temperature maxima were
419 1C. Salinities roughly varied with temperature except during
precipitation events in the winter and spring that resulted in
salinities as low as 29.64 in the upper meter (Fig. 1b). The surface
mixed-layer depth was generally between 5 and 20 m in the
summer and autumn and as deep as 63 m in the winter (Tables 1
and 2).
Surface temperatures measured via CTD during time series
cruises and near-surface temperatures from moored instrumenta-
tion corresponded well and indicate that the time-series cruises
sampled most of the temperature maxima during 2005 (Fig. 2).
However, lower temperature events of short duration were missed
by the time-series cruises in the spring and summer. A power
spectrum of hourly sea-surface temperatures from the SMBO
mooring indicates the strongest peak at 1 cycle/day and higher
frequencies (12 h) (not shown). Frequencies lower than 1 day
(e.g. 7 day, 14 day) are not dominant, indicating that twice
monthly cruises would not coincide with any systematic tem-
perature variation.
Macronutrient concentrations were generally inversely corre-
lated with temperature as expected; low concentrations were
observed in the winter and maxima occurred in the spring and
summer (Fig. 3). Nitrate concentrations were higher than those of
nitrite (not shown), yet concentrations of each were often
depleted below 0.50 mM over the mixed layer and as deep as
45 m (Fig. 3a). Dissolved silicon concentrations ranged from 0.10
to 28 mM in the upper 50 m, and mixed-layer concentrations were
generally lower than 2 mM (Fig. 3b). Phosphate concentrations
ranged from 0.10 to 2.80 mM and mixed-layer concentrations were
generally lower than 0.40 mM (Fig. 3c). These surface water
concentrations of N, P and Si indicate that at Redfield ratios,
nitrate would be potentially limiting to phytoplankton growth.
3.2. Phytoplankton data
Potentially harmful taxa were present in every surface sample
and were often numerically dominant among the 45 mm
phytoplankton. In 2004, near-surface blooms were dominated
(hereafter signifying a contribution of 450% of total cell
abundance) by diatoms. Pseudo-nitzschia spp. and Chaetoceros
spp. were the dominant genera, and total cell abundances were
often 4150,000 cells l 1 in February and April (Fig. 4). In contrast,
armored dinoflagellates dominated blooms in 2005. Following a
small bloom of the diatom Thalassiosira sp. in January 2005,
P. micans reached concentrations of 474,000 cells l 1 in May
2005. Later, in September 2005, the highest observed cell
abundance during the time series (359,900 cells l 1) occurred
during a bloom dominated by L. polyedrum.
Time series of chlorophyll a concentrations integrated from the
surface to 50 m generally revealed similar trends as total
phytoplankton abundances. This correlation as well as vertical
profiles that show similar community composition with depth
(Figs. 6–8) suggests that phytoplankton abundances from the
surface represent the dominant phytoplankton community on
sampling dates. The data indicate that the largest phytoplankton
blooms in 2004 occurred on February 13 and April 27 (Fig. 5);
these dates also showed maxima in cell abundances. A subsurface
bloom of smaller magnitude on July 6 2004 was missed in near-
surface cell counts; however, qualitative sample inspection
showed a similar community composition. In 2005, the two
highest chlorophyll a maxima on April 27 and May 11 occurred at
depth and are not indicated by near-surface cell abundances.
However, events in both May 2005 and September 2005 indicated
both elevated cell abundances and chlorophyll a concentrations.
 ARTICLE IN PRESS
R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593
As a next step, the vertical distribution of the major blooms is
investigated. In February 2004, a diatom bloom was observed in
the upper mixed layer (Fig. 6). Hydrographic conditions and the
phytoplankton taxonomic composition were similar over three
successive sample dates, and nutrient concentrations decreased
with time. Thus, it is likely that the same bloom was sampled on
successive dates. On February 5, prior to the bloom, the diatoms
Chaetoceros spp. and Thalassiosira sp. dominated the assemblage
and cell abundances were less than 50,000 cells l 1. At this time,
nitrate and dissolved silicon concentrations in the upper mixed
layer were greater than 2.0 and 4.0 mM, respectively. Cell abun-
dances were much higher on the following two sampling dates
and nutrient concentrations progressively decreased too0.2 mM
nitrate and o1.0 mM dissolved silicon. Pseudo-nitzschia spp.
Fig. 5. Time series of chlorophyll a concentrations integrated from the surface to a
depth of 50 m on sampling dates in 2004 and 2005.
Fig. 6. Vertical profiles of cell abundances and nitrate and dissolved silicon concentrations before and during the February 2004 phytoplankton bloom; (a and d) 5 February,
(b and e) 13 February and (c and f) 17 February.
2589
comprised 462% of the total cell abundance at all 6 sample
depths on 13 February; other diatoms again included Chaetoceros
spp. and Thalassiosira sp. On February 17, Chaetoceros spp. was the
dominant taxa, although Pseudo-nitzschia spp. contributed nearly
30% of the total cell abundances at all sample depths.
In April 2004, a diatom bloom was comprised of the same taxa,
but hydrological conditions and the vertical distribution of
phytoplankton differed. High abundances of Pseudo-nitzschia spp.
were observed in surface waters on March 30 (Fig. 7). On April 14,
Chaetoceros spp. dominated at the surface and at 5 m whereas
Pseudo-nitzschia spp. dominated at 15, 25, 35 and 50 m. On April 27,
Pseudo-nitzschia spp. dominated at all sample depths, and there
was a dramatic cell abundance maxima at 20 m. Over the course
of these three dates, nutrient concentrations were progressively
depleted over the upper 20 m and the water column remained
stratified, with surface temperatures of 14–15 1C.
The last major bloom during the study occurred in September
2005. On August 30, cell abundances were fairly evenly dis-
tributed throughout the upper 50 m (Fig. 8). Approximately 30% of
the cells at each depth were L. polyedrum, 30% were Thalassiosira
spp., and 15–20% were Protoperidinium spp. On September 13,
L. polyedrum comprised 90% of the total cell abundance (359,000
cells l 1) and were observed primarily in surface waters. Although
present at lower concentrations on 27 September, L. polyedrum
remained the dominant taxa in the upper 30 m. The diatoms
Thalassiosira spp. and Chaetoceros spp. were present at lower but
significant quantities. The upper column was temperature
 ARTICLE IN PRESS
2590 R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593
Fig. 7. Vertical profiles as in Fig. 5 but on (a and d) 30 March, (b and e) 14 April and (c and f) 27 April 2004.
stratified throughout the bloom, and surface temperatures
dropped from 19.2 to 15.0 1C. However, nitrate concentrations
were below the detection limit throughout the mixed layer
(8–11 m) on all three sampling dates.
3.3. Statistical analyses
Results of the correlation analysis of log transformed cell
counts and abiotic factors were most significant without any lag
time and suggest different environmental controls of each
potentially harmful bloom taxa (Table 3). Pseudo-nitzschia spp.
abundances are significantly inversely correlated with tempera-
ture and positively correlated with potential density. P. micans
abundances are inversely correlated with mixed-layer depth.
Finally, L. polyedrum abundances are not significantly correlated
with any of the 8 measured variables.
Results of multivariate analysis of abiotic variables indicate
that the first two principal components of physico-chemical
variability had eigenvalues greater than 1 and combine to explain
61% of sample variance. The first principal component (PC1) is
positively correlated to nitrate, phosphate and nitrite concentra-
tions and mixed-layer depth and is negatively correlated to
temperature and salinity (Table 4). PC1 accounts for 39% of the
sample variance and a positive score represents upwelling
conditions. The second principal component (PC2) accounts for
22% of the sample variance and represents the effect of non-
diatom productivity in shallow mixed layers. Surface-dissolved
silicon and nitrate concentrations and salinity are positively
correlated with PC2.
In order to elucidate possible relationships between bloom
taxa and environmental conditions, sample dates have been
identified as those when total cell abundances were o1 104
cells l 1 or X1 104 cells l 1, and those dominated by a potentially
harmful taxa on the biplot of PC1 and PC2 (Fig. 9). Low cell
abundances were generally observed in the third quadrant of the
biplot (supporting nutrients as a limiting factor). Although blooms
were often dominated by potentially harmful taxa, sample dates
dominated by Pseudo-nitzschia spp. and P. micans do not cluster on
the biplot, indicating variation in environmental conditions.
However, clustering of the points on the biplot indicates that
the environmental conditions during 4 of the 5 L. polyedrum
blooms were similar.
4. Discussion
Potentially harmful phytoplankton taxa had a strong presence
in the SMB during the 2-year-study. Results indicate that at least
one of the three bloom taxa, Pseudo-nitzschia spp., P. micans and
L. polyedrum was present in every sample. Further, the 4
phytoplankton blooms with the highest near-surface cell abun-
dances consisted of these taxa. Although there are no previous
records of phytoplankton assemblages in this bay, the observed
abundances of P. micans and L. polyedrum appear exceptional
when compared to a 20 year record of dinoflagellates sampled
daily at pier stations in the nearby Santa Barbara and San Diego
Basins (Allen, 1941). In that study, P. micans and L. polyedrum were
two of the most common dinoflagellates observed at both sites
during weekly sampling between 1920 and 1939. However, these
 ARTICLE IN PRESS

R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593 2591

Fig. 8. Vertical profiles as in Fig. 5 but on (a and d) 30 August, (b and e) 13 September and (c and f) 27 September 2005.

Table 3 species generally occurred at very low abundances and were even
Correlation coefficients between log-transformed abundances of phytoplankton absent for entire years. High abundances of L. polyedrum
taxa (cells l 1) and physico-chemical variables measured in near-surface waters on
comprising ‘‘red tides’’ only occurred during 2 of the 20 years
n sampling dates
(1932 and 1939) at the Scripps pier (San Diego Basin).
Pseudo-nitzschia Prorocentrum Lingulodinium n The Intaglio model qualitatively predicts the nutrient and
spp. micans polyedrum mixing conditions that select for specific groups of marine
phytoplankton (Smayda and Reynolds, 2001). Observations from
Temperature 0.51 0.07 0.16 48
Salinity 0.10 0.22 0.15 49
the SMB can be compared to this conceptual model as predictions
Density 0.44 0.22 0.06 48 are made for diatoms in general, and specifically for the harmful
Nitrate 0.15 0.04 0.02 47 dinoflagellate taxa observed during 2004 and 2005. According to
Phosphate 0.25 0.12 0.28 47 the model, L. polyedrum, P. micans and Pseudo-nitzschia spp.,
Nitrite 0.26 0.10 0.03 47
respectively, should occur along a gradient of increasing nutrient
Dissolved silicon 0.00 0.17 0.06 48
Mixed-layer 0.17 0.36 0.28 46 concentrations and mixed-layer depths. Results of the multi-
depth variate analysis are generally consistent with this prediction. The
first principal component of variation in the abiotic data
Bold values are significant at the 0.05 level. corresponds to increasing nitrate concentrations and mixed-layer
depths. As predicted, blooms dominated by L. polyedrum, P. micans
and Pseudo-nitzschia spp. generally follow in this order along
increasing values of this PC1 axis (Fig. 9). This trend fits with the
Table 4
Correlation coefficients between the first two principal components and 7 physico-
expectation that nitrate would be the limiting nutrient in this
chemical variables study, as ratios with phosphate and dissolved silicon were much
lower than Redfield ratios. Nitrate has also previously been shown
PC1 PC2 to be the limiting nutrient for phytoplankton production in
southern California Bight waters (Eppley et al., 1979) and in the
Temperature 0.50 0.14
Salinity 0.21 0.38 SMB (Eppley and Renger, 1988). Temperature was also negatively
Nitrate 0.42 0.49 correlated with the PC1 and with Pseudo-nitzschia spp. abun-
Phosphate 0.37 0.17 dances; this covariate is not unexpected as it is associated with
Nitrite 0.37 0.02
upwelling conditions.
Dissolved silicon 0.15 0.63
Mixed-layer depth 0.48 0.42
Deviations from the predicted associations of taxa with environ-
mental conditions, such as Pseudo-nitzschia spp. dominance during
 ARTICLE IN PRESS

2592 R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593

Fig. 9. Biplot of the first two principal components of variability in abiotic variables for 49 stations (numbered points) with loadings (vectors) for the seven labeled
variables. Stations at which integrated chlorophyll a concentrations are X40 mg m 2 are indicated by solid points and concentrations p40 mg m 2 are indicated by open
points. Blooms dominated by potentially harmful phytoplankton species are indicated by shading around points.

relatively low nitrate conditions on April 27, 2004 (Fig. 7) reflect
bloom evolution and nutrient uptake. This highlights one
limitation of synoptic sampling and undersampling of phyto-
plankton blooms; certainly biomass outlasts the conditions
originally supporting the selection of these taxa and changes in
the phytoplankton community assembly process can be missed
with a sampling interval of 2 weeks.
The specific associations between harmful bloom taxa and
environmental conditions are of interest in the Southern Califor-
nia bight region which is adjacent to large metropolitan areas and
yet harmful bloom ecology is not well studied. The association of
Pseudo-nitzschia spp. with high macronutrient concentrations as
observed in the SMB also occurs at the Scripps pier (Lange et al.,
1994). However, a unique relationship between environmental
conditions and Pseudo-nitzschia spp. blooms in the SCB remains
elusive as there was large interannual variability both in this
study and during a 4-year-study at Scripps pier (Lange et al.,
1994). As in the SMB, Pseudo-nitzschia spp. (identified as Nitzschia)
blooms in the Monterey Bay are also associated with both
upwelling and stratified conditions (Bolin and Abbot, 1963; Buck
et al., 1992). Diatoms of the genus Pseudo-nitzschia have clearly
adapted to diverse environmental conditions and further work
will likely benefit from identification to lower taxonomic levels.
L. polyedrum was most abundant in near-surface waters under
more stratified and lower nutrient conditions during 2003 and
2004, with scores less than 1 and generally less than 0.5 on the
PC1 axis, conforming to the predictions of the Intaglio model.
A similar association with stratification and oligotrophy was also
observed in the San Diego Basin during the summers of 1964,
1965 and winter 1966 (Holmes et al., 1967). However, in contrast
to that study, our data did not support the association of
L. polyedrum with nutrient rich runoff. The L. polyedrum bloom
in September 2005 was sighted all along the southern California
coastline, including the San Pedro Bay, located just to the south of
the SMB (Moorthi et al., 2006). Upper mixed-layer nitrate
concentrations wereo0.20 mM during this bloom and ammonia
concentrations were similarly low (unpublished results). It is
possible that the cells were obtaining nitrogen from dissolved
organic sources or by vertical migration to deeper waters, as
previously observed in the Baja California waters (Dugdale, 1979).
The possibility that migration behavior affected the distribution of
cells during this bloom is supported by the surface accumulation
of cells on September 13, 2005, despite a 9 m mixed layer. Food
web implications of L. polyedrum blooms such as importance as a
food source to larval fish (Scura and Jerde, 1977) or the support of
longer food chains (Walsh et al., 1974) have not been documented
in the Southern California bight basin. However, similar roles are
possible based on the high bloom abundances observed in the
SMB and similar hydrographic conditions and food web structure
to those coastal studies.
P. micans is a cosmopolitan taxa that forms red tides (e.g. Cassic,
1981; Fukuyo et al., 1989) and it has been associated with human
gastrointestinal illness (e.g. Kat, 1979; Horstman, 1981), although
its toxicity is unconfirmed. Yet its presence in the southern
California Bight appears to be intermittent and its ecology is poorly
described. Springtime P. micans blooms in La Jolla Bay in 1964 and
1965 were associated with a steep shallow thermocline similar to
that seen during the spring 2005 bloom in SMB. In the 20-year time
series of Allen (1941), P. micans was often present in summer
months, but it never reached bloom concentrations.
 ARTICLE IN PRESS
R.F. Shipe et al. / Continental Shelf Research 28 (2008) 2584–2593
Upwelling favorable conditions can occur locally within the
SMB as well as over a larger region of the Southern California
Bight. Hence, the observed water masses and associated phyto-
plankton populations are generally a mixture of locally upwelled
waters and laterally advected waters from the north in the Santa
Barbara Basin or Point Conception region (Hickey, 1992).
In conclusion, this study suggests a strong presence of
potentially harmful bloom taxa in the SMB. The blooms of toxic
taxa Pseudo-nitzschia spp., P. micans and L. polyedrum are
generally associated with increasing gradients of nutrients and
mixing, as expected by the Intaglio model. Further, blooms here
are not isolated but part of a larger phenomenon and observations
made here may apply over large spatial scales, as during the
September 2005 event. Despite the complex hydrography of the
region, the results support primary controls of temperature
stratification and nitrate concentrations.
Acknowledgements
This project was funded by the UCLA Department of Ecology
and Evolutionary Biology (RS) and by the US National Science
Foundation award EAR 04-04405 (AL and NG). We thank Captain
Willie McCarthy, Dennis Weyrauch and Levanto Schachter for
assistance in seagoing expeditions. Technical assistance was
provided by Jessica Curtaz, Rebecca Rooke, Alina Corcoran,
Christopher Lin, Jacqueline Ng and Heather Coleman.
References
Allen, W.E., 1941. Twenty years’ statistical studies of marine plankton dinofla-
gellates of southern California. American Midland Naturalist 26, 603–635.
Anderson, D.M., Galloway, S.B., Joseph, J.D., 1993. Marine Biotoxins and Harmful
Algae: A National Plan. Woods Hole Oceanographic Institute, Woods Hole,
Massachusetts.
Bolin, R.L., Abbot, P., 1963. Studies on the marine climate and phytoplankton of the
central coastal area of California, 1954–1960. California Cooperative Oceanic
Fisheries Investigations Reports 9, 23–45.
Brainerd, K.E., Gregg, M.C., 1995. Surface mixed and mixing layer depths. Deep-Sea
Research, Part I 42, 1521–1542.
Buck, K.R., Uttal-Cooke, L., Pilskaln, C.H., Roelke, D.L., Villac, M.C., Fryxell, G.A.,
Cifuentes, L., Chavez, F.P., 1992. Autecology of the diatom Pseudonitzschia
australis, a domoic aid producer, from monterey bay, California. Marine Ecology
Progress Series 84, 293–302.
Cassic, V., 1981. Non-toxic blooms of Prorocentrum micans (Dinophyceae) in the
Karamea Bight. New Zealand Journal of Marine and Freshwater Research 15,
181–184.
Dugdale, R.C., 1979. Primary nutrients and red tides in upwelling regions. In:
Taylor, D.L., Seliger, H.H. (Eds.), Toxic Dinoflagellate Blooms. Elsevier,
Amsterdam, pp. 257–262.
Eppley, R.W., Renger, E.H., Harrison, W.G., Cullen, J.J., 1979. Ammonium distribution
in Southern California coastal waters and its role in the growth of
phytoplankton. Limnology and Oceanography 24, 495–509.
Eppley, R.W., Renger, E.H., 1988. Nanomolar increase in surface layer nitrate
concentration following a small wind event. Deep-Sea Research 35, 1119–1125.
Fukuyo, Y.K., Yoshida, K., Ogata, T., Ishimaru, T., Kodama, M., Pholpunthin, P.,
Wisessang, S., Phanichyakarn, V., Piyakarnchana, T., 1989. Suspected causative
dinoflagellates of paralytic shellfish poisoning in the Gulf of Thailand. In:
Okachi, T., Anderson, D.M., Nemoto, T. (Eds.), Red Tides: Biology, Environmental
Science, and Toxicology. Elsevier, New York, NY, pp. 403–406.
Glibert, P.M., Anderson, D.A., Gentien, P., Graneli, E., Sellner, K.G., 2005. The global
complex phenomena of harmful algal blooms. Oceanography 18, 136–147.
2593
Hickey, B.M., 1992. Circulation over the Santa Monica-san Pedro Basin and shelf.
Progress in Oceanography 30, 37–115.
Hickey, B.M., Dobbins, E.L., Allen, S.E., 2003. Local and remote forcing of currents
and temperature in the central Southern California bight. Journal of
Geophysical Research 108, 3081.
Holmes, R.W., Williams, P.M., Eppley, R.W., 1967. Red water in La Jolla Bay,
1964–1966. Limnology and Oceanography 12, 503–512.
Horner, R.A., Garrison, D.L., Plumley, F.G., 1997. Harmful algal blooms and red tide
problems on the US West Coast. Limnology and Oceanography 42, 1076–1088.
Horstman, D.A., 1981. Reported red water outbreaks and their effects on fauna of
the West and South Coasts of South Africa 1959–1980. Fisheries Bulletin of
South Africa 15, 71–88.
Kat, M., 1979. The occurrence of Prorocentrum species and coincidental gastro-
intestinal illness of mussel consumers. In: Taylor, D.L., Seliger, H.H. (Eds.), Toxic
Dinoflagellate Blooms. Elsevier, North Holland.
Kudela, R.M., Cochlan, W.P., 2000. Nitrogen and carbon uptake kinetics and the
influence of irradiance for a red tide bloom off southern California. Aquatic
Microbial Ecology 21, 31–47.
Lange, C.B., Reid, F.M.H., Vernet, M., 1994. Temporal distribution of the potentially
toxic diatom Pseudonitzschia australis at a coastal site in Southern California.
Marine Ecology Progress Series 104, 309–312.
Margalef, R., 1978. Life-forms of phytoplankton as survival alternatives in an
unstable environment. Oceanologica. Acta 1, 193–197.
Margalef, R., Estrada, M., Blasco, D., 1979. Functional morphology of organisms
involved in red tides, as adapted to decaying turbulence. In: Taylor, D.L., Seliger,
H.H. (Eds.), Toxic Dinoflagellate Blooms. Elsevier, North Holland.
Moorthi, S.D., Countway, P.D., Stauffer, B.A., Caron, D.A., 2006. Use of quantitative
real-time PCR to investigate the dynamics of the red tide dinoflagellate
Lingulodinium polyedrum. Microbial Ecology 52, 136–150.
Nezlin, N.P., Li, B.L., 2003. Time-series analysis of remote-sensed chlorophyll and
environmental factors in the Santa Monica-san pedro basin off Southern
California. Journal of Marine Systems 39, 185–202.
Nezlin, N.P., Oram, J.J., DiGiacomo, P.M., Gruber, N., 2004. Sub-seasonal to
interannual variations of sea surface temperature, salinity, oxygen anomaly,
and transmissivity in Santa Monica Bay, California from 1987 to 1997.
Continental Shelf Research 24, 1053–1082.
Odum, H.T., McConnell, W., Abbot, W., 1958. The Chlorophyll a of Communities,
Vol. 5. Publication of the Institute of Marine Science of the University of Texas,
pp. 65–96.
Reid, J.L., Roden, G.L., Whyllie, J.G., 1958. Studies of the California current system.
California Cooperative Oceanic Fisheries Investigations Progress Report 1,
27–57.
Ryther, J.H., Yentsch, C.S., 1957. The estimation of phytoplankton production in the
ocean from chlorophyll and light data. Limnology and Oceanography 2,
281–286.
Scholin, C.A., Gulland, F., Doucette, G.J., Benson, S., Busman, M., Chavez, F.P.,
Cordaro, J., DeLong, R., Vogelaere, A.D., Harvey, J., Haulena, M., Lefebvre, K.,
Lipscomb, T., Loscutoff, S., Lowenstine, L.J., Miller, P.E., McLellan, W.A., Moeller,
P.D.R., Powell, C.L., Rowles, T., Silvagni, P., Silver, M., Spraker, T., Trainer, V.,
Dolah, F.M.V., 2000. Mortality of sea lions along the central California coast
linked to a toxic diatom bloom. Nature 403, 80–84.
Scura, E.D., Jerde, C.W., 1977. Various species of phytoplankton as food for larval
northern anchovy, Engraulis mordax and relative nutritional value of the
dinoflagellates Gymnodinium splendens and Gonyaulax polyedra. Fisheries
Bulletin 75, 577–583.
Smayda, T.J., Reynolds, C.S., 2001. Community assembly in marine phytoplankton:
application of recent models to harmful dinoflagellate blooms. Journal of
Plankton Research 23, 447–461.
Smayda, T.J., Reynolds, C.S., 2003. Strategies of marine dinoflagellate survival and
some rules of assembly. Journal of Sea Research 49, 95–106.
Tomas, C.R., 1996. Identifying Marine Diatoms and Dinoflagellates. Academic Press,
Inc., San Diego.
Villac, M.C., Roelke, D.L., Villareal, T.A., Fryxell, G.A., 1993. Comparison of two
domoic acid-producing diatoms: a review. Hydrobiologia. 269/270, 213–224.
Walsh, J.J., Kelley, J.C., Whitledge, T.E., MacIssac, J.J., 1974. Spin-up of the Baja
California upwelling ecosystem. Limnology and Oceanography 19, 553–572.
Work, T.M., Beale, A.M., Fritz, L., Quilliam, M.A., Silver, M., Buck, K., Wright, J.L.C.,
1993. Domoic acid intoxication of pelicans and cormorants in Santa Cruz,
California. In: Smayda, T.J., Shimizu, Y. (Eds.), Toxic Phytoplankton Blooms in
the Sea: Proceedings of the 5th International Conference on Toxic Phytoplank-
ton Blooms. Elsevier, Amsterdam, pp. 643–649.