1999 Progress Report: Time Series Analysis and Modeling Ecological Risk

EPA Grant Number: R825433C058
Subproject: this is subproject number 058 , established and managed by the Center Director under grant R825433
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: EERC - Center for Ecological Health Research (Cal Davis)
Center Director: Rolston, Dennis E.
Title: Time Series Analysis and Modeling Ecological Risk
Investigators: Botsford, Louis , Quinn, James
Current Investigators: Botsford, Louis , Jassby, Alan D. , Quinn, James
Institution: University of California - Davis
EPA Project Officer: Hahn, Intaek
Project Period: October 1, 1996 through September 30, 2000
Project Period Covered by this Report: October 1, 1998 through September 30, 1999
RFA: Exploratory Environmental Research Centers (1992) RFA Text |  Recipients Lists
Research Category: Center for Ecological Health Research , Targeted Research


To derive demographic and trophic-dynamic rates from existing time series measurements to project risks to populations.

Progress Summary:

Predicting effects of multiple stresses on biological endpoints, particularly populations of sensitive species, in a complex, spatially and temporally varying environment, challenges both the best available parameter estimation techniques and current risk assessment models. Over the past several years we have made considerable advances in deriving the demographic and tropho-dynamic consequences of time series measurements and other information in the rivers, Bay and Delta, and in using those rates to project risks to populations. Results have been instrumental in establishing the current regulatory and monitoring regime managed by the Interagency Ecological Program, which plans and coordinates management and monitoring of the Bay-Delta system.

Past activities have developed a variety of techniques to provide confidence intervals on risk calculations from population time-series for species in spatially complex environments, and for testing (and rejecting) some sensitivity analyses (e.g., elasticity calculations) frequently applied to this kind of time-series data.

In the past year, we have worked more specifically on populations in the Bay-Delta watershed, particularly the striped bass population due to the rapid decrease in abundance of this important sport-fishery species. Population models have been linked to the transport and water quality models of the Rivers/Delta using methods developed in Dr. Orlob's project. The ability to embed population models within the circulation and water quality fields via simulated Lagrangian drifters (ie, particles) allows us to study the populations in a spatially-explicit modeling environment. As such, the populations can be tranported within the system in a manner that allows us to assess numerically the the effects of spatially variable stresses on population. We have also developed submodels to permit us to do the same kinds of analyses with trophic models.

We have been preparing for the availability of spatially-explicit fields of pesticide stressors and of model runs which mimic the CALFED management alternatives. In discussions with Drs. Hinton and Orlob and their associates, we have laid the groundwork for incorporating into our population models the impact of toxic compounds upon the fish populations in the Rivers/Delta. We assisted in designing the water management alternatives that mimic the CALFED alternatives and look forward to the availability of these runs. The ability to run our population models embedded in the water transport and quality models being developed by Dr. Orlob's group will undoubtedly provide insight into the spatial effects of both pesticide stressors and proposed CALFED water management changes-individually and in combination-that would be difficult to see outside of this integrated modeling framework.

Another major effort during the past year was to determine how life history differences, such as different maturity schedules, should be accounted for in the assessment of risk for endangered California salmon stocks. Specifically, we assessed how differences in life histories contribute the relative jeopardy in Pacific salmon, using coho (Oncorhynchus kisutch) and chinook salmon (Oncorhynchus tshawytscha) as examples. The characteristics of semelparity and anadromy in Pacific salmon lead to a unique form of a Leslie matrix, and a special definition of population extinction respectively. We showed that for obligate semelparous species, the later the age of maturity the lower the probability of extinction over a fixed time period, because of a shift in time scale and an increase in the number of cohorts. For indeterminate semelparous populations, as the distribution of age of maturity becomes broader, the probability of extinction first increases then declines. This complex behavior is a result of two counter balancing effects: cohort coupling and age class diversification. Age class diversification, increased spawning at multiple ages, causes the probability of the first reproductive failure to decrease. However cohort coupling, linkage between the first reproductive failure of a cohort and subsequent failures, causes all cohorts to go extinct nearly simultaneously. Existing analytical approximations for probabilities of extinctions do not accurately predict behavior of Pacific salmon. Semelparity violates the assumption of weak ergodicity implicit in that model, and the yearly changes in population abundance are likely larger than the small changes assumed in the diffusion approximation. Although density dependent recruitment has a minimal effect on obligate semelparous populations its effect on indeterminate semelparous populations can be dramatic. These results provide managers with some key indicators of the relative jeopardy of salmon stocks with different life histories.

Future Activities:

In the coming year, we will continue our assessment of how various aspects of population dynamics should be integrated into risk assessment of California fish species, especially those on the Sacramento and San Joaquin Rivers. This will include additional general analyses, and also assessment of specific salmon runs.

Both this year and next year, along with funding from CEHR, the approach is being extended with support from other sources (eg, an NSF/EPA Watershed Grant). As we proceed, we will further incorporate spatial variability in environmental stresses and the population responses at several trophic levels (e.g., phytoplankton, zooplankton, and three species of fish in the Delta; nutrients, multiple species of plankton, and clams in the lower Bay) in order to provide better methods for decision making.

It has become clear that the risk assessments needed for effective environmental decision may stretch the capabilities of even powerful simulation models applied to intensive monitoring data. As a result, the Core researchers (Botsford with fish and fisheries populations, and Quinn in terrestrial monitoring studies) are increasingly designing monitoring regimes to support more powerful analyses. Most of the risk assessment work in the past year applies data developed for invasive species in the GIS and biodiversity project (IV.1) discussed above. In the riparian realm, Arundo (giant reed) and tamerisk (saltcedar) are predominant invaders of most lowland zones studied by CEHR. Techniques for assessing risk of invasion derive both from changes in current distributions (see IV.1) and suitability models derived from a combination of regression from occurrence data and biophysical modeling (e.g. the Australian GARP model). Risk assessments for Arundo are being calibrated, and measures potentially applicable to Asian mitten crabs are being assembled (in conjunction with Dr. Grosholz in DESP). Integrating these capabilities and developing techniques for optimizing monitoring designs will be an ongoing effort.

Supplemental Keywords:

RFA, Scientific Discipline, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, Hydrology, Ecology, State, Monitoring/Modeling, Ecological Risk Assessment, Ecology and Ecosystems, aquatic ecosystem, environmental monitoring, risk assessment model, modeling, fisheries, multiple stressors, time series analysis, ecological risk, ecosystem health, environmental stress, California (CA), hydrologic modeling, ecological models

Progress and Final Reports:

Original Abstract
  • 1997
  • 1998
  • Final Report

  • Main Center Abstract and Reports:

    R825433    EERC - Center for Ecological Health Research (Cal Davis)

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R825433C001 Potential for Long-Term Degradation of Wetland Water Quality Due to Natural Discharge of Polluted Groundwater
    R825433C002 Sacramento River Watershed
    R825433C003 Endocrine Disruption in Fish and Birds
    R825433C004 Biomarkers of Exposure and Deleterious Effect: A Laboratory and Field Investigation
    R825433C005 Fish Developmental Toxicity/Recruitment
    R825433C006 Resolving Multiple Stressors by Biochemical Indicator Patterns and their Linkages to Adverse Effects on Benthic Invertebrate Patterns
    R825433C007 Environmental Chemistry of Bioavailability in Sediments and Water Column
    R825433C008 Reproduction of Birds and mammals in a terrestrial-aquatic interface
    R825433C009 Modeling Ecosystems Under Combined Stress
    R825433C010 Mercury Uptake by Fish
    R825433C011 Clear Lake Watershed
    R825433C012 The Role of Fishes as Transporters of Mercury
    R825433C013 Wetlands Restoration
    R825433C014 Wildlife Bioaccumulation and Effects
    R825433C015 Microbiology of Mercury Methylation in Sediments
    R825433C016 Hg and Fe Biogeochemistry
    R825433C017 Water Motions and Material Transport
    R825433C018 Economic Impacts of Multiple Stresses
    R825433C019 The History of Anthropogenic Effects
    R825433C020 Wetland Restoration
    R825433C021 Sierra Nevada Watershed Project
    R825433C022 Regional Transport of Air Pollutants and Exposure of Sierra Nevada Forests to Ozone
    R825433C023 Biomarkers of Ozone Damage to Sierra Nevada Vegetation
    R825433C024 Effects of Air Pollution on Water Quality: Emission of MTBE and Other Pollutants From Motorized Watercraft
    R825433C025 Regional Movement of Toxics
    R825433C026 Effect of Photochemical Reactions in Fog Drops and Aerosol Particles on the Fate of Atmospheric Chemicals in the Central Valley
    R825433C027 Source Load Modeling for Sediment in Mountainous Watersheds
    R825433C028 Stress of Increased Sediment Loading on Lake and Stream Function
    R825433C029 Watershed Response to Natural and Anthropogenic Stress: Lake Tahoe Nutrient Budget
    R825433C030 Mercury Distribution and Cycling in Sierra Nevada Waterbodies
    R825433C031 Pre-contact Forest Structure
    R825433C032 Identification and distribution of pest complexes in relation to late seral/old growth forest structure in the Lake Tahoe watershed
    R825433C033 Subalpine Marsh Plant Communities as Early Indicators of Ecosystem Stress
    R825433C034 Regional Hydrogeology and Contaminant Transport in a Sierra Nevada Ecosystem
    R825433C035 Border Rivers Watershed
    R825433C036 Toxicity Studies
    R825433C037 Watershed Assessment
    R825433C038 Microbiological Processes in Sediments
    R825433C039 Analytical and Biomarkers Core
    R825433C040 Organic Analysis
    R825433C041 Inorganic Analysis
    R825433C042 Immunoassay and Serum Markers
    R825433C043 Sensitive Biomarkers to Detect Biochemical Changes Indicating Multiple Stresses Including Chemically Induced Stresses
    R825433C044 Molecular, Cellular and Animal Biomarkers of Exposure and Effect
    R825433C045 Microbial Community Assays
    R825433C046 Cumulative and Integrative Biochemical Indicators
    R825433C047 Mercury and Iron Biogeochemistry
    R825433C048 Transport and Fate Core
    R825433C049 Role of Hydrogeologic Processes in Alpine Ecosystem Health
    R825433C050 Regional Hydrologic Modeling With Emphasis on Watershed-Scale Environmental Stresses
    R825433C051 Development of Pollutant Fate and Transport Models for Use in Terrestrial Ecosystem Exposure Assessment
    R825433C052 Pesticide Transport in Subsurface and Surface Water Systems
    R825433C053 Currents in Clear Lake
    R825433C054 Data Integration and Decision Support Core
    R825433C055 Spatial Patterns and Biodiversity
    R825433C056 Modeling Transport in Aquatic Systems
    R825433C057 Spatial and Temporal Trends in Water Quality
    R825433C058 Time Series Analysis and Modeling Ecological Risk
    R825433C059 WWW/Outreach
    R825433C060 Economic Effects of Multiple Stresses
    R825433C061 Effects of Nutrients on Algal Growth
    R825433C062 Nutrient Loading
    R825433C063 Subalpine Wetlands as Early Indicators of Ecosystem Stress
    R825433C064 Chlorinated Hydrocarbons
    R825433C065 Sierra Ozone Studies
    R825433C066 Assessment of Multiple Stresses on Soil Microbial Communities
    R825433C067 Terrestrial - Agriculture
    R825433C069 Molecular Epidemiology Core
    R825433C070 Serum Markers of Environmental Stress
    R825433C071 Development of Sensitive Biomarkers Based on Chemically Induced Changes in Expressions of Oncogenes
    R825433C072 Molecular Monitoring of Microbial Populations
    R825433C073 Aquatic - Rivers and Estuaries
    R825433C074 Border Rivers - Toxicity Studies