Summer

1
Temporal and Spatial Variability of Chlorophyll-a
Versus Particulate Carbon and Nitrogen
Relationships in San Francisco Bay Importance
for Estimating Biomass and Ecosystem
Modeling. Kleckner, A.E., Parker, A.E., Lew, K.,
Hogue, V.E., Wilkerson, F. Romberg Tiburon
Center, San Francisco State University, Tiburon,
CA 94920
Introduction In making accurate estimates of the
amount of phytoplankton biomass, both
chlorophyll-a and carbon measurements are
critical. The use of CChl-a as a model variable
is based on the simplicity with which
chlorophyll-a concentrations are experimentally
determined (Zonneveld 1998). However,
chlorophyll-a is a weak measure of phytoplankton
biomass, as cellular chlorophyll content may
change due to photoacclimation. The knowledge of
CChl-a allows one to translate chlorophyll-a
measurements to the more valuable phytoplankton
carbon measure (Geider et al., 1997). Several
investigators have used relationships of CChl-a,
including for San Francisco Estuary (Cloern et
al., 1995). Cloern developed a model as a
function of temperature, daily irradiance, and
nutrient-limited growth rate. Chang et al.
(2003) examined CChl-a in the coastal waters of
the East China Sea. Simple regressions of C vs.
Chl-a showed variability in CChl-a of 13-93
based on depth. These findings suggest
significant variability of CChl-a, therefore we
investigated spatial and temporal variability in
CChl-a for central San Francisco Bay.
TEMPORAL VARIABILITY (for central San Francisco
Bay)
SPATIAL VARIABILITY (Spring only)
POC, PON, and Chl-a concentrations over time
Central SF Bay
San Pablo Bay
Suisun Bay
POC vs. Chl-a
POC vs. PON
Summer
y -17.90x 146.26 R2 0.19
y 6.50x 60.35 R2 0.29

• Concentrations of chl-a varied between 1.5 and
  4.1 µg L-1.
• POC concentrations were more variable than chl-a
  values between 31 and 136 µmol L-1.
• No significant correlation between POC and chl-a.
• PON concentrations between 0 and 12 µmol L-1.
• No significant correlation between POC and PON.

y 8.40x 107.87 R2 0.19
y -5.86x 192.10 R2 0.04
y 22.15x 64.61 R2 0.57

• Central SF Bay chl-a concentrations varied
  between 3.6 and 7.4 µg L-1.
• POC concentrations fell between 117 and 201 µmol
  L-1.
• No significant correlation between POC and chl-a.

• Suisun Bay chl-a concentrations varied between
  1.6 and 5.1 µg L-1.
• POC concentrations fell between 116 and 241 µmol
  L-1.
• No significant correlation between POC and chl-a.

• San Pablo Bay chl-a concentrations varied between
  3.1 and 14.0 µg L-1.
• POC concentrations fell between 85 and 412 µmol
  L-1.
• POC and chl-a are significantly correlated, (R2
  0.57).

Fall
Study Site
y 10.14x 71.48 R2 0.87

• Concentrations of chl-a varied between 2.3 and
  15.8 µg L-1.
• POC concentrations between 89 and 198 µmol L-1.
• POC and chl-a are significantly correlated, (R2
  0.84).
• PON concentrations between 3 and 12 µmol L-1.
• PON and chl-a were significantly correlated, (R2
  0.76).
• POC and PON were also significantly correlated,
  (R2 0.87).

Winter

• Conclusions
• There are considerable seasonal and spatial
  variations in concentrations of POC and Chl-a.
• This variability should be considered when
  applying ratios of CChl-a to estuarine ecosystem
  models.
• Using ratios of total Chl-a total C, or POCPON
  gives a misleading ratio as it includes a large
  amount of detrital carbon.
• Detrital carbon contribution is indicated by the
  intercept of regression slopes.
• Modelers that want a CN ratio for San Francisco
  Bay, or want POC values from Chl-a measurements,
  may find that the most appropriate value is to be
  calculated after subtracting the detrital carbon
  component.
• An alternate estimate of POCChl-a may be
  obtained from the slopes of linear regressions,
  (See figures) rather than just using a directly
  calculated ratio. (See tables)

• Chl-a concentrations varied between 1.4 and 5.0
  µg L-1.
• POC concentrations between 70 and 189 µmol L-1
  with a pronounced peak of 380 µmol L-1 in early
  Feb. (possibly a freshwater pulse).
• No significant correlation between POC and chl-a.
• PON concentrations between 1 and 62 µmol L-1.
• POC and PON were significantly correlated, (R2
  0.79).

y -16.16x 164.67 R2 0.05
y 2.41x 96.91 R2 0.79
Samples collected by bucket from Central Bay
June16-July 16, 2004 (Summer), September
1-October 1, 2004 (Fall), February 2-March 1,
2005 (Winter), and March 17-April 8, 2005
(Spring) and from shipboard at Suisun and San
Pablo Bays March 16-April 20, 2005.
Methods Surface water samples (100ml) were
analyzed for chlorophyll-a after filtration onto
GF/F filters using in vitro fluorometry with
phaeophytin correction (Arar and Collins, 1992)
using a Turner Designs fluorometer calibrated
with Turner Designs commercial chlorophyll.
100ml samples for particulate carbon and
particulate nitrogen measurements (POC and PON)
were collected onto precombusted (450C for 4
hours) GF/F filters and analyzed using a PDZ
Europa 20/20 GC-MS system. Linear regressions
were constructed for each season.
Spring
y 3.58x 96.39 R2 0.64
y 2.12x 101.40 R2 0.18

• Chl-a concentrations varied between 2.0 and 9.3
  µg L-1.
• POC concentrations between 100 and 137 µmol L-1.
• Spring POC and chl-a are significantly
  correlated, (R2 0.64).
• PON concentrations between 2 and 9 µmol L-1.
• No significant correlation between POC and PON.

Table of calculated ratios S.E.M. for Central
Bay
Arar, Elizabeth J. and Collins, Gary B. In Vitro
Determination of Chlorophyll a and Pheophytin a
in Marine and Freshwater Algae by Fluorescence.
U.S. Environmental Protection Agency, Method
445.0 September 1997. Chang, Jeng et al.,
Cross-shelf Variation in Carbon to Chlorophyll a
Ratios in the East China Sea, summer 1998 Deep
Sea Research II 50(2003) 1237-1247. Cloern et
al., Geider et al., Dynamic model of
phytoplankton growth and acclimation responses
of the balanced growth rate and the chlorophyll
acarbon ratio to light, nutrient-limitation and
temperature Marine Ecology Progress Series 148
187-200. Zonneveld, C. A cell based model for
the chlorophyll a to carbon ratio in
phytoplankton Ecological Modelling 113(1998)
55-70.
Acknowledgements This study was carried out as
part of a Summer undergraduate independent
research study by A.E.K. (SFSU course Biol 699).
We thank the University of Southern California
Sea Grant Program for their financial support.