Grantee Research Project Results
2001 Progress Report: Integrative Indicators of Ecosystem Condition and Stress across Multiple Trophic Levels in the San Francisco Estuary
EPA Grant Number: R827644Title: Integrative Indicators of Ecosystem Condition and Stress across Multiple Trophic Levels in the San Francisco Estuary
Investigators: Dugdale, Richard C. , Kimmerer, Wim , Arp, Alissa J. , Thompson, Janet K. , Bollens, Stephen Morgan , Wilkerson, Frances P. , Julian, David William
Institution: San Francisco State University , Romberg Tiburon Center , United States Geological Survey , University of Florida
Current Institution: San Francisco State University , U.S. Geological Survey - Sacramento , University of Florida
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1999 through September 30, 2002 (Extended to September 30, 2003)
Project Period Covered by this Report: October 1, 2000 through September 30, 2001
Project Amount: $881,062
RFA: Ecological Indicators (1999) RFA Text | Recipients Lists
Research Category: Aquatic Ecosystems , Ecological Indicators/Assessment/Restoration
Objective:
The objectives of this project are to:
- Establish the utility of the following potential indicators of ecosystem
condition at three stations along a salinity gradient in the San Francisco
Estuary:
- Nutrient status and productivity performance of phytoplankton including the relative contribution of diatoms to phytoplankton biomass and productivity
- Reproductive rates of common copepod (zooplankton) species
- Nutritional condition of larval Pacific herring using morphometric characters, some of which are sensitive to growth (food) (e.g., body weight) and some of which are relatively insensitive to growth (food) (e.g., eye diameter)
- Changes in benthic community structure and growth rate, condition, and glycogen content of key benthic organisms
- Expression of stress proteins in benthic bivalves, larval herring, and copepods
- Investigate relationships of indicators to variation in other physical (e.g., temperature, salinity, turbidity) and biological (e.g., introduced species, copepod prey for herring) parameters
- Assess the utility of the ecological indicators for use in other locations.
This research is being conducted on the open-water ecosystem of the San Francisco Estuary including the portion of the landscape from freshwater to marine ecosystems (i.e., Suisun, San Pablo, and Central Bays). These potential indicators represent key population and individual-level processes in a variety of trophic levels, and cover planktonic and benthic communities. All indicators are relatively simple to measure, significant to population ecology, expected to be sensitive to stress levels, and likely to be transportable to other aquatic habitats.
Progress Summary:
Nutrient/Phytoplankton Indicator Study
As in the first year of the study, we have measured temperature, salinity, nutrients, and chlorophyll every month during cruises that occupied three stations in San Francisco Bay along a salinity gradient (Suisun Bay: USGS Station 6, San Pablo Bay: USGS Station 13 and Central Bay: RTC Station XB-D. These stations also were sampled weekly during October 2000 and March-April 2001, at times when fall and spring blooms in chlorophyll were observed in the Year 1 data. Data collected in year 2 showed similar spatial trends to Year 1, with high nutrient concentrations that decreased going seaward. In both years, a clear seasonal cycle in water column properties occurred. Following cool, winter conditions, there is warming in the spring, accompanied by lower salinity water and increased silicate but decreased nitrate and ammonium, in part through dilution reducing the effects of treatment-plant effluent and agricultural sources. As stratification sets in, a spring bloom developed in each location, sometime in April, elevating the chlorophyll concentrations above the non-bloom average of about 1-2 µg/L, reaching 30 µg/L in April 2000 in Suisun Bay. In 2001 in San Pablo and Central Bays, the April peak in chlorophyll concentration was greater than that in April 2000 (16 vs. 15 µg/L, 13 vs. 7 µg/L). By May, the chlorophyll was reduced back to 1-2 µg/L. Preceding the chlorophyll peaks in spring, the ammonium values often dropped to near-detection levels, and nitrate decreased. The spring bloom did not occur in Suisun Bay in 2001. The probable cause was the pulse of ammonium that appeared in December 2001 with concentrations of 16 µM, perhaps the result of low rainfall and less dilution of effluent wastewater. In contrast to the nitrate decrease at bloom time in the previous year and to San Pablo and Central Bay in 2001, nitrate concentrations increased. During October 2000, similar trends to April 2000, were seen but with a smaller increase in chlorophyll and accompanying decreases in ammonium, nitrate, and silicate concentrations. The nutrient status at these locations indicates nonlimiting conditions for phytoplankton success.
In spring of 2000 and 2001, the cells making up the peak chlorophyll concentrations were larger cells, as shown by the fraction of chlorophyll contained in greater than 10 µm cells. Microscopic analysis showed many diatom chains and large numbers of the diatom Skeletonema costatum similar to the earlier (1980) bloom community structure data of Cole et al. (1986). The diatoms have an obligate requirement for silicate for growth, and the result of diatoms pulling down silicate was apparent as a decrease in silicate concentrations before the peak chlorophyll in the time series data. The relative contribution of diatoms to the biomass and production in these three locations is high during bloom periods.
Although we were able to carry out 15N labeled nitrate and ammonium uptake, and 13C fixation experiments at all locations on each cruise, not all the samples have been processed due to instrumentation difficulties with our mass spectrometer, which is being repaired. However, the nitrogen tracer data available for the spring bloom period in Central Bay support the inhibition of nitrate uptake by elevated levels of ammonium. When the ratios of nitrate uptake to ammonium uptake for pre and bloom conditions in 2000 are plotted versus ambient ammonium concentration, it is clear that nitrate uptake is low at high ammonium concentrations. Nitrate uptake rates exceed ammonium uptake (i.e., ratio greater than one) only when ammonium concentrations are less than 4 µM. Time series plots of ambient concentrations and uptake show that ambient ammonium decreased throughout San Francisco Bay reaching low levels (4 µM or less) by April, accompanied by low nitrate uptake (measured with 15N), which increased in April once the ambient ammonium has reached a certain level. The ambient nitrate decreases as the nitrate uptake rates increase. In the Bay, the kinetics of ammonium and nitrate uptake (i.e., uptake versus ambient concentration) has not been measured. Nor has the direct inhibition of nitrate uptake been observed by experiments with added ammonium. How the interaction of nitrate and ammonium determine the type of phytoplankton producers also is unclear. Typically, diatoms use nitrate more efficiently than ammonium, whereas dinoflagellates (typical agents of red tides) have been reported to have high affinity for regenerated forms of nitrogen, including ammonium and urea. How this influences the probability of noxious or harmful algal phytoplankton species is not known. The balance of ammonium and nitrate in an estuary and the influence of overenrichment with ammonium may be an important determinant. Although this was not a scientific goal for this component of the proposed research, relevant data that address these questions may emerge.
Zooplankton Indicator Study
To date, we have completed 26 cruises, including monthly cruises between November 1999, and September 2001, and several additional cruises during spring bloom periods. Copepod (zooplankton) reproductive rates were measured by collecting with gentle net tows, diluting the catch in surface bay water, and incubating individual females for 24 hours in 125 mL polycarbonate bottles filled with bay water with eggs strained out. Data have been worked up and analyzed through July 2001. We have been measuring copepod reproductive rates with excess food, determining egg development times, and reevaluating the methods used. On two occasions during an egg production experiment, we found poor survival and low egg production, which we tentatively attribute to a bloom of a toxic dinoflagellate. Otherwise, survival has been very high in our experiments, most often 100 percent. Reproductive rate has been quite variable, which we tentatively attribute to food limitation. No effect of salinity has been observed. The samples taken to determine egg ratios of copepods that carry eggs have been difficult to interpret because most of the eggs have been separated from the females, and for some species, the eggs cannot be identified to species.
Condition Indices of larval Pacific Herring (Clupea pallasi)
Our primary objective is to determine the "nutritional" condition of larval Pacific herring using morphometric characters, some of which are sensitive to growth (food) (e.g., body weight) and some of which are relatively insensitive to growth (food) (e.g., eye diameter). We also will relate variation in larval herring condition to variation in other physical (e.g., temperature, salinity) and biological (e.g., copepod prey) variables, and evaluate their utility as ecological indicators. Once a month between November 1999, and May 2001, we collected triplicate net tows at two stations (San Pablo Bay and Central Bay). We sorted, identified, and measured the anal body depth (mm), pectoral body depth (mm), eye diameter (mm), standard length (mm), head width (mm), and dry weight (mg) of 902 herring larvae. Nine indices of condition were derived for each larva. Using a multivariate data analysis technique, we were able to determine that three of the nine indices showed considerable variation between locations and years. All three of these indices involve BW as the "sensitive to growth" measurement. There appear to be correlations between variation in indices 1-3 and chlorophyll a concentration and copepod egg production rates. These morphometric indices appear to be useful for evaluating nutritional condition in herring larvae and may prove useful as indicators of ecosystem condition or health.
Benthic Indicators Study
Monthly collections of benthic community, glycogen, and condition samples continue to be collected at 1 central bay site, 2 San Pablo Bay sites, 4 Suisun Bay sites, and 1 Grizzly Bay site. Detailed analysis of the benthic community data at the Grizzly Bay site (collection from 1977-2000) has shown that the measure most likely to reveal stress in this estuarine community is unfortunately the most difficult to collect and analyze. These data have shown that the disruption of benthic community by exotic species is the major stress on that community as well as on the pelagic community. Furthermore, the benthic community composition data reflects these stresses quite well. There have been six major introductions of exotic species to the benthic community during the sampling period, and each has resulted in a change to the benthic community. With the most recent introduction, that of the physiologically (osmotically) tolerant bivalve Potamocorbula amurensis, we have seen a shift to a benthic community now dominated by a filter feeder that is successful through all seasons. This has resulted in a reduction of the phytoplankton bloom in this area as well as a reduction in higher trophic groups (well documented in the past reports). Our most recent data show that this bivalve is beginning to show some declines in biomass during some high freshwater flow years, which has resulted in the return of some intermittent, smaller, and shorter phytoplankton blooms than the historic blooms. These blooms occur during spring and last only 2-4 weeks, as opposed to the historic blooms that occurred for 4-5 months from summer through fall. It is unclear what affect these small blooms, with shorter durations and greatly different seasonal cycles will have on the pelagic community that depends on the phytoplankton as a food source.
Details of benthic community reflections of stress due to exotics. Historically, the benthic community reflected the seasonal and decadal changes in salinity, with a "wet" fresh water community during periods of low salinity and a "dry" euryhaline community present during sustained periods of low outflow through the estuary and elevated salinity in the Grizzly Bay region. Both the "fresh water" and "euryhaline" benthic communities were dominated (in terms of biomass) by filter feeding bivalves Corbicula fluminea (freshwater community) and Mya arenarea (euryhaline community), prior to Potamocorbula's introduction. The invasion of the euryhaline Asian clam, Poatmocorbula amurensis disrupted the "wet" and "dry" community response to variations in salinity, displacing the previously dominant bivalves as well as other members of the benthic community. The post-Potamocorbula invasion benthic community in Grizzly Bay is distinctly different from the pre-Potamocorbula invasion community and is less variable from year to year. The post-Potamocorbula community also is dominated by recent invaders (P. amurensis, 1986, Corophium alienense, 1973, cumacean Nippoleucon hinumensis, 1986, and spionid Marezelleria viridis, 1991). Abundance of all organisms other than these dominant invaders is significantly reduced. Food limitation and removal of pelagic larval forms prior to settlement are two mechanisms by which P. amurensis may have had such a remarkable effect on the benthic community in Grizzly Bay. When abundance of benthic organisms is pooled by functional group such as feeding type (suspension feeder, deposit feeder) or reproductive type (pelagic larva, no pelagic larvae), there is evidence that these two mechanisms may be structuring the postinvasion community. Though the abundance of most species is reduced since the P. amurensis invasion, organisms that suspension feed are proportionally more reduced than organisms that deposit feed, and organisms that reproduce via pelagic larval forms are proportionally more reduced than organisms that do not. Organisms from the pre-invasion community that both suspension feed and reproduce via pelagic larval forms are absent from the postinvasion community.
Use of condition and glycogen as stress indicators. Our analyses of these parameters are continuing. The seasonal condition and glycogen data at the Suisun and Grizzly Bay sites are consistent with reproductive cycles and show that Potamocorbula is surviving with little glycogen stores throughout most of the year. One encouraging development is what appears to be an increase in glycogen in the San Pablo Bay animals, possibly reflecting a less stressful environment. We will continue the analysis and hope that 3 years of data will be sufficient to conclusively determine if the spatial pattern is significant. We look forward to comparing our results with the stress protein data during the coming year.
Stress Protein Study
We now have successfully begun analyzing benthic invertebrate tissues of using two-dimensional gel electrophoresis (2DGE). Unlike the more familiar one-dimensional gel electrophoresis, 2DGE separates proteins by both isoelectric focusing (the first dimension) and by molecular weight (the second dimension). With this technique, the proteins in a single sample of tissue can theoretically be separated into 10,000 or more distinguishable "spots" on a gel. These can represent the majority of all proteins expressed by the cells in that tissue at the time of sampling, meaning that one can now analyze the "proteome" of the tissue as it existed at the time of sampling. Because exposure to ecological stressors (and other factors) can cause increased or decreased expression of many proteins, proteome analysis may be able to provide valuable information on the nature and magnitude of the stressor(s) and the overall physiological state of the animal.
As exciting as this sounds, 2DGE as applied to marine invertebrates is still in its infancy. For example, although 2DGE is increasingly being used for biomedical research on mammalian tissues, very few researchers have used 2-D gels for analyses of invertebrate tissues. Furthermore, no researchers to our knowledge have yet published any 2DGE work on marine invertebrates in which they have utilized isolated pH gradient (IPG) strips. IPG strips represent an important practical development because they have the potential to increase the standardization and reproducibility of results both within the same laboratory and between laboratories. Future application of 2DGE for biomonitoring will necessitate reliable, reproducible results that can be performed by technicians in different laboratories, so IPG strips may be a key development.
Nonetheless, we encountered several hurdles in developing 2DGE (with IPG strips) for benthic marine invertebrate tissues. First and foremost has been determining appropriate tissue homogenization protocols that (1) maximize the yield of tissue proteins, and (2) are applicable across a range of animal tissues. Any technique for preparing proteins for 2DGE will favor either cytoplasmic proteins or membrane proteins. The majority of important stress-responsive proteins thus far discovered in benthic invertebrates are cytoplasmic, so we now use a homogenization medium that maximizes the recovery of soluble proteins. To increase the ease and frequency with which IPG strips can be utilized, we homogenize tissues in rehydration buffer containing biolytes and the detergent CHAPS. For tissue disruption, we tested a variety of mechanical homogenization techniques, including a Polytron electric homogenizer, Dounce tissue grinders, Potter tissue grinders, and a nitrogen cavitation bomb. We are now satisfied that a standard glass tissue grinder followed by ultrasonication provides good protein yield with minimal denaturation.
Because so little is known about stress-responsive proteins in marine invertebrates, we could not be certain which proteins would be best to analyze. Therefore, to maximize the range of protein sizes that would be included in our studies, while still allowing the separation and visualization of as many proteins as possible, we have begun using 3-10 pH IPG strips (a fairly broad range) and 8-16 percent gradient gels. To reduce variability between analyses caused by irregularities in the gels, we use a multicaster to create up to 12 (roughly) identical gels at one time. Tissue samples are then run in duplicate or triplicate. Successful 2DGE gels created using the protocols described above, of the foot of a hardshell clam, Mercenaria mercenaria, from Cedar Key, Florida, and total tissue homogenate from the Asian clam, Potamocorbula amurensis, (from San Francisco Bay (site 6.1, collected May 22, 2001) are available (gel available but not shown).
An unexpected difficulty has been interference from RNA and DNA in the tissue samples. Although this normally is not a serious problem in 2DGE of mammalian tissues, the presence of large amounts of nucelotides in our samples makes it impossible to use silver staining to identify proteins in the gels, because this stain also labels nucleotides. This leaves us with three alternatives. First, we can add endonucleases to the homogenates prior to 2DGE. Unfortunately, this has not worked as well as expected. Second, we can change the homogenization medium to minimize RNA and DNA. We are trying to avoid this because it would require a major change to the protocols we have developed thus far. Our third alternative, and the one we are now developing, is the use of alternate protein staining techniques. Currently, we are using Coomassie staining to label proteins. Although this well-known staining technique is less sensitive than silver staining, it is specific for proteins (and therefore does not stain RNA and DNA). An added benefit is that most Coomassie staining protocols (including the protocol we use for 2DGE) produce very little toxic waste, especially in comparison to silver stains. The drawback is that this technique is not sufficiently sensitive for some low-abundance proteins. In later experiments, we can use fluorescent protein stains (such as SYPRO Orange) that have extremely high sensitivity to detect these proteins. At present, however, these fluorescent stains are prohibitively expensive and require specialized equipment for visualization.
At present, we have homogenized P. amurensis from all collections at sites 4.1, 6.1, 8.1, and 12.5 from February 6, 2001, to July 17, 2001. The homogenized samples are in storage in an ultra-cold freezer. Protein contents have been measured for all samples. We are now finalizing our protein staining and gel casting protocols in preparation for running the samples in parallel with eight electrophoresis chambers.
Future Activities:
Phytoplankton: Mass spectrometry of Year 1 and 2 samples will continue, so that analyses of the light-enhanced productivity experiments can be evaluated for use as a "stress" indicator. As proposed, until May 2002, we will continue monthly sampling and tracer experiments at the three sites in San Francisco Bay for physical variables (salinity, temperature, turbidity), nutrients, phytoplankton biomass, and composition and productivity.
Zooplankton: We have suspended sampling to enable us to take additional samples during spring 2002, in an attempt to capture the phytoplankton bloom. During that time, we will refine our estimates of food limitation, and make estimates of availability of alternative food that might be supporting high reproductive rates. Sampling will end in May 2002, to give us time to analyze samples and report results.
Larval Herring: Future activities include sampling for larval herring
in San Pablo Bay and Central Bay locations from November 2001 through May 2002.
Future research will involve further investigation of the relationship between
larval condition indices and physical and other biological indices in an attempt
to develop ecosystem-level indices of condition.
Benthic Studies: We will
complete our analyses of the Grizzly Bay benthic community and apply the same
techniques to the San Pablo Bay benthic community. As stated earlier (last
year), we will discontinue the Central Bay site in December 2001, but will
continue the others.
Stress Proteins: In the coming year, we will begin laboratory exposures of marine invertebrates to various environmental stressors to identify protein "spots" in 2DGE that are characteristically stress responsive (and perhaps even responsive to specific stressors). Likely candidates are inducible heat shock proteins, free radical scavengers (e.g., superoxide dismutase), metallothioneins, and metallothionein-like proteins. The presence and location of these proteins in the gels initially can be determined using specific antibodies, with later identification being performed by computer using relative position within the gel. We will use these new computer software packages, which have been specifically designed for proteomics analysis, to look for similar changes in protein expression between 2DGE of animals exposed to experimentally induced stressors and 2DGE of animals collected from the San Francisco Bay.
Journal Articles:
No journal articles submitted with this report: View all 53 publications for this projectSupplemental Keywords:
marine, estuary, ecological effects, organism, ecosystem, indicators, biology, aquatic, ecology, monitoring, analytical, western, pacific coast, California, CA, phytoplankton, zooplankton, benthos, morphometrics., RFA, Scientific Discipline, Water, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, Nutrients, Ecology, Ecosystem/Assessment/Indicators, Ecosystem Protection, Environmental Chemistry, State, Chemistry, Ecological Effects - Environmental Exposure & Risk, Microbiology, EPA Region, aquatic ecosystem, environmental monitoring, nutrient supply, EMAP, marine ecosystem, Region 9, stressors, trophic transfer, bioavailability, phytoplankton dynamics, ecosystem indicators, salinity, bioindicators, aquatic ecosystems, benthos-associated organisms, San Franciso Estuary, ecological indicators, California (CA), benthic nutrients, populationRelevant Websites:
http://userwww.sfsu.edu/~aarp/arp-lab.html Exit
(Arp)
http://userwww.sfsu.edu/~bioocean/research/epaherring/epaherring.html
Exit (Bollens)
http://userwww.sfsu.edu/~phytopl/
Exit (Dugdale/Wilkerson)
http://www.zoo.ufl.edu/julian/ Exit (Julian)
http://online.sfsu.edu/~kimmerer/research.htm
Exit (Kimmerer)
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.