Grantee Research Project Results
Final Report: A Shallow-water Coastal Habitat Model for Regional Scale Evaluation of Management Decisions in the Virginian Province
EPA Grant Number: R830878Title: A Shallow-water Coastal Habitat Model for Regional Scale Evaluation of Management Decisions in the Virginian Province
Investigators: Gallegos, Charles L. , Weller, Donald E. , Jordan, Thomas E. , Neale, Patrick J. , Megonigal, J. P.
Institution: Smithsonian Environmental Research Center
EPA Project Officer: Packard, Benjamin H
Project Period: June 1, 2003 through September 30, 2007
Project Amount: $746,433
RFA: Developing Regional-Scale Stressor-Response Models for Use in Environmental Decision-making (2002) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Aquatic Ecosystems , Climate Change
Objective:
Management decisions to protect estuaries are being made in the context of unprecedented environmental changes. For example, increased ultraviolet (UV) radiation, especially the damaging UV-B, has been documented and is expected to continue even at temperate latitudes. The carbon dioxide concentration of the atmosphere rose by 30% in the 20th century and is continuing to climb at a rate of about 1% per year. The effects of CO2 and other greenhouse gasses on global climate change are highly uncertain, but alteration of rainfall and runoff patterns are considered likely. Interactions between altered flow regimes and changes in land use patterns will have consequences for the delivery of sediments and nutrients to estuaries. Projecting the effectiveness of management actions must proceed on the basis of predictions from mathematical models, since experimental manipulations cannot be made on the relevant scales. However, the effects of simultaneous, multiple stressors have not previously been incorporated into models of ecosystem processes.
Our modeling efforts focus on shallow tributary embayments and small tidal creeks of Chesapeake Bay, because the ecological importance of shallow systems far exceeds their volumetric contribution to the Bay. Their importance derives from the many hectares of potential habitat for submersed aquatic vegetation created by their highly indented shorelines, and from their roles as spawning and nursery grounds for finfish and as refuge habitat for juvenile fish and crabs.
The objective of the modeling efforts in the estuary is to predict the magnitude and trends of existing and emerging indicators of the ecological condition of critical shallow water habitats. The model incorporates response functions to interactions among stresses (specifically suspended sediments, nutrients, and elevated UV-B irradiance) that are expected to change with increasing population and development in the coastal zone. The end points for our model are those indicators being used as de-listing criteria for Chesapeake Bay, namely chlorophyll, water clarity (diffuse attenuation coefficient) and dissolved oxygen.
We represent shallow subestuaries as well-mixed compartments that receive and process inputs from their local watershed, and exchange materials at their seaward boundaries. Mass balance modeling techniques were employed for the model structure, with rate processes dependent upon interactions amongst stressors. For stressor interactions in the watershed, we considered interactions between climate-induced flow alteration with changes in land use, as they impact delivery of nutrients and sediments to the estuary. In the estuary we modeled interactions amongst nutrients, suspended solids, dissolved organic matter, and UV-B on plankton growth and light penetration. We used a Monte Carlo approach to facilitate investigation of subestuaries with diverse morphologies, and to predict cumulative distribution functions of the de-listing criteria for comparison with reference curves which are currently under development.
Summary/Accomplishments (Outputs/Outcomes):
Subestuary Delineation and Watershed Loadings
One primary objective was to analyze how the geographic variability in physical structure and human use among the linked watershed-subestuary systems of Chesapeake Bay affects estuarine response. We identified 128 Chesapeake Bay systems that fit our local watershedsubestuary paradigm and captured the boundaries of these systems in a GIS database (definitions of common acronyms are given in Table 1). Subestuary areas range from 0.1-101 km2and their associated local watershed areas range from 6-1664 km2, with NLCD land cover percentages ranging from 6-81% forest, 1-64% cropland, 2-38% grassland, and 0.3-89% developed land. We also analyzed digital shoreline and bathymetric data to calculate a number of subestuary metrics, including subestuary area and water volume, mouth width and area of vertical profile, proportion of shallow water (<= 2m) area, elongation ratio, fractal dimension, the ratio of subestuary perimeter to subestuary area, and ratio of local watershed area to subestuary volume.
The spatial and statistical analyses of subestuaries and their watersheds proved to be extremely valuable for evaluating extent to which biological responses in subestuaries (such as submerged aquatic vegetation abundance) can be related to the geographic properties of the subestuaries and their watersheds (Li et al. 2007). Watershed land use can affect submerged aquatic vegetation (SAV) by elevating nutrient and sediment loading to estuaries. Most studies of SAV habitat requirements have focused on proximal estuarine water quality parameters that directly affect light availability, but the effects of indirect factors, such as estuarine and watershed characteristics, on SAV abundance are less well quantified. We analyzed the effects of watershed use and estuarine characteristics on the spatial variation of SAV abundance among 101 shallow subestuaries of Chesapeake Bay during the period 1984-2003. Using mapped data from annual SAV surveys, we calculated SAV coverage for each subestuary in each year during 1984-2003 as a proportion of potential SAV habitat (the area <2 meter deep). The variation in SAV abundance among subestuaries was strongly linked with subestuary and watershed characteristics. A regression tree model indicated that 60% of the variance in SAV abundance could be explained by subestuary fractal dimension, mean tidal range, local watershed dominant land cover, watershed to subestuary area ratio, and mean wave height. Similar explanatory powers were found in wet and dry years, but different independent variables were used. Repeated-measures ANOVA with multiple-mean comparison showed that SAV abundance declined with the dominant watershed land cover in the order: forested, mixed-undisturbed, or mixed-developed > mixed-agricultural > agricultural > developed. Change-point analyses indicated strong threshold responses of SAV abundance to point source total nitrogen and phosphorus inputs, ratio of local watershed area to subestuary area, and septic system density in the local watershed. Future analyses would benefit from the development of more spatially and temporally extensive water quality data to explore the linkage of landscape characteristics to water quality and then in turn to SAV abundance.
Another project task was applying watershed simulation models to provide daily predictions of local watershed loading for driving the subestuary model and to predict how watershed loadings will respond to changes in land use and climate. Several watershed modeling tools were evaluated, including SWAT, the CBP Phase 5 HSPF model, SPARROW, GWLF, and AGNPS. We focused on GWLF model as the watershed modeling tool most suitable for this project. Model calibration was a significant challenge. We developed a novel watershed calibration framework that simultaneously applied a multi-objective global optimization criterion to calibration data from several neighboring watersheds. The framework was used to calibrate 16 hydrological and nutrient parameters for 4 subwatersheds of the Rhode River subestuary. When calibrated to a single watershed, GWLF gave reasonable predictions for monthly and annual streamflow and total nitrogen (TN), but success for total phosphorus (TP) varied among watersheds. However, parameter estimates varied widely among watersheds. Calibrating model parameters using multiple watersheds reduced the relative mean absolute error of the model prediction by 3.9-7.6% for monthly streamflow, 1.9-6.5% for monthly TN, and 1.1-7.5% for monthly TP, respectively.
We applied the calibrated models to explore how climate and land cover change would interact to affect material discharges. When both climate change and land cover change were considered, GWLF predicted a slight decrease in average annual stream flow and slight increases in average annual TN, TP, dissolved inorganic nitrogen (DIN), and dissolved inorganic phosphorus (DIP) discharge in 2030 compared with a baseline of 2000. Among four study watersheds, some annual discharges of water, TN, TP, DIN and DIP in 2030 decreased while others increased. By 2095, all stream flow and nutrient discharges were predicted to increase. The relative increase and relative annual increase rate for TN or DIN were greater than for TP or DIP; and the relative increase and relative annual increase rate for all nutrients were greater than for stream flow. By 2095, annual discharges of water, TN, TP, DIN and DIP are predicted to increase by 18%, 36%, 33%, 36% and 32%, respectively. The GWLF model results predict that the influence of climate change will be much greater than that of land cover change (0.6-4.0% increase in urban development in 2030).
Subestuary Modeling
We constructed an ecosystem/water quality subestuary model having three segments with different volumetric dimension between the mouth and at the head of a subestuary. The watershed is the upstream boundary where multiple stressors, such as inorganic nutrients, sediments, and dissolved organic matter, are discharged. The downstream boundary is the Chesapeake Bay or a major Bay tributary, such as the Potomac or James River. Each subestuary segment has multiple ecological components (21 state variables), including dissolved inorganic nitrogen and phosphorus in the water column, dissolved inorganic nitrogen and phosphorus in a phytoplankton cell (cell quota), size-fractionated phytoplankton (3 size classes) and zooplankton (3 size classes), dissolved organic nitrogen and phosphorus, bacterial and detrital nitrogen and phosphorus, and dissolved oxygen. The coding of the physical exchange (with biological sources and sinks eliminated) and biological state equations (with physical exchange eliminated) were tested to ensure that the model conserves mass.
The model was implemented and validated using historical data sets that have been collected from the Rhode River subestuary. For example, reported exchange coefficients between the segments (Jordan et al. 1991), were tested using salinity measurements collected from November 12, 1999 to July 16, 2002 and found to be satisfactory for conserving salt. Test simulations for validation purpose were also done based on biological and chemical data sets collected from March 22 through October 4, 2000. Simulations have been conducted for different sets of parameters and coefficients and produced promising results in capturing a phytoplankton bloom in early summer of 2000. The results clearly showed that the local input of nutrients played an important role in the growth of the phytoplankton, although the system was strongly driven by exchange at the downstream boundary.
Using DIN, DIP, DON and DOP concentrations predicted by a climate change model and GWLF as upstream boundary conditions, the Subestuary Ecosystem Responses Function (SERF) calibrated to the Rhode River was run for period 2024-2036 to simulate dynamics of standard water quality indicators (chlorophyll a, dissolved oxygen and diffuse attenuation coefficient) under the future climate change scenarios. Climate change was found to have a significant effect on simulated phytoplankton biomass in the upstream receiving water segment only. The response of segments closer to the subestuary mouth are governed by downstream boundary conditions, which used historical data sets treated as constants for the entire simulation period in this analysis. Phytoplankton biomass in the lower segments would undoubtedly respond to increased concentrations of limiting nutrient at the downstream boundary; but due to complex biological interactions in the main Bay, predicting future trends of nutrient concentrations at the mouths of subestuaries was beyond the scope of this project.
For exploring regional variability among subestuaries with differing geomorphic characteristics, we constructed a reduced complexity, nitrogen-based model consisting of 3 state variables (dissolved nutrient, phytoplankton biomass, and zooplankton biomass) in 3 subestuary segments. Mixing and exchange at the watershed and down-estuary boundaries, nutrient release from the bottom, phytoplankton growth, and zooplankton grazing were treated similarly as in the full model, except that biotic compartments were not size-fractionated. We performed a Monte Carlo simulation with the reduced complexity subestuary model to better understand the range of responses amongst subestuaries receiving similar inputs. We ran 1000 simulations, drawing exchange coefficients from a lognormal distribution scaled to covary with the ratio of the mouth width to subestuary volume. Simulated annual average and peak chlorophyll concentrations showed strong negative correlation with the logarithm of the exchange coefficient.
Incorporation of model watershed inputs generated for the 2024-2036 scenario into the reduced complexity subestuary model had a negligible effect on simulated biomass, increasing annual average chlorophyll concentration by +0.3 mg m-3, and peak concentrations by an average of +2.2 mg m-3.
We conclude that the morphometric diversity among subestuaries of Chesapeake Bay is expected to lead to variations in physical conditions among subestuaries, most importantly the rate of exchange at the down-estuary boundary. These differences have an overriding effect on simulated peak chlorophyll concentrations and temporal dynamics. The simulations here did not incorporate climate change-related differences in mainstem boundary concentrations or variations in biological parameters. These added variations would be expected to further confound detection of a climate change signal due to local watersheds.
Climate Change and Stressor Interactions
Ultraviolet (UV) radiation is a significant stressor in aquatic environments and can inhibit primary productivity of phytoplankton. The effects of UV depend on many factors, including phytoplankton community composition and acclimation status. Using spectrally resolved biological weighting functions (BWFs), we determined sensitivity of photosynthesis and acclimation to UV in a common estuarine diatom, Thalassiosira pseudonana (Hustedt) Hasle et Heimdal, and a cryptomonad, Cryptomonas sp. (Litchman and Neale 2005). Cryptomonas sp. grown under high PAR (photosynthetically active radiation) (250 μmol quanta m-2s-1) was significantly more sensitive to photoinhibition in the UV-B part of the spectrum (280 to 320 nm) than T. pseudonana under high PAR. Growth under low irradiance (25 μmol quanta m-2s-1) increased sensitivity of T. pseudonana. After a week-long exposure to moderate UV radiation, sensitivity of Cryptomonas sp. declined, while sensitivity of T. pseudonana did not change. Growth rates and chlorophyll a-specific absorption decreased in both species. Based on the BWFs obtained in this study, we predict 11 to 26% UV inhibition of depth-integrated primary production by these species under summer conditions in a shallow, turbid temperate estuary. This effect is sufficiently large that it merits incorporation into models of primary production in shallow estuaries.
Simulation of light attenuation as an indicator of water clarity was an important component of the project. Light attenuation in shallow subestuaries is affected by 3 water constituents: phytoplankton, suspended solids, and colored dissolved organic matter. Of these three components, virtually nothing is known about the watershed loading of CDOM around Chesapeake Bay, and this uncertainty limited our ability to model light attenuation in subestuaries. To fill this data need, we initiated a sampling program to determine CDOM concentrations in feeder streams to a number of subestuaries of the Bay and a high marsh.
The role of tidal marshes as a source of dissolved organic carbon (DOC) and CDOM for adjacent estuarine waters was studied in the Rhode River subestuary of the Chesapeake Bay (Tzortziou et al. 2007, 2008). There was a net DOC export from the marsh to the estuary during seasons of both low and high marsh plant biomass. Optical analysis demonstrated that, in addition to contributing to the carbon budgets, the marsh had a strong influence on the estuary’s CDOM dynamics. Marsh-exported CDOM had optical properties that were consistently and markedly different from those of CDOM in the adjacent estuary. Our findings illustrate the importance of tidal marshes as sources of optically and chemically distinctive dissolved organic compounds, and their influence on CDOM dynamics, DOC budgets, and, thus, photochemical and biogeochemical processes, in adjacent estuarine ecosystems.
DOC and dissolved inorganic carbon (DIC) were measured in a C3-dominated and a C4dominated marsh community that had been exposed to ambient conditions or elevated CO2 since 1987. In the C3-dominated community, concentrations of DOC were typically higher in the elevated CO2 treatment, with this effect being significant at both the 30- and 75-cm depths (P=0.02 and P=0.07, respectively). In contrast, there was no stimulatory effect of elevated CO2 on DOC in the C4-dominated community. Overall, these results are consistent with other data suggesting that a portion of the excess carbon fixed in response to elevated CO2 in the C3dominated community is being made available for heterotrophic microbial activity as DOC.
Table 1. Definitions of commonly used acronyms.
GIS |
Geographic Information System |
NLCD |
National Land Cover Dataset |
SWAT |
Soil and Water Assessment Tool |
CBP |
Chesapeake Bay Program |
HSPF |
Hydrologic Simulation Program—FORTRAN |
SPARROW |
SPAtially Referenced Regressions on Watershed Attributes |
GWLF |
Generalized Watershed Loading Functions |
AGNPS |
AGricultural Non-Point Source Pollution Model |
Journal Articles on this Report : 4 Displayed | Download in RIS Format
Other project views: | All 23 publications | 4 publications in selected types | All 4 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Li X, Weller DE, Gallegos CL, Jordan TE, Kim H-C. Effects of watershed and estuarine characteristics on the abundance of submerged aquatic vegetation in Chesapeake Bay subestuaries. Estuaries and Coasts 2007;30(5):840-854. |
R830878 (Final) |
Exit Exit |
|
Litchman E, Neale PJ. UV effects on photosynthesis, growth and acclimation of an estuarine diatom and cryptomonad. Marine Ecology Progress Series 2005;300:53-62. |
R830878 (2003) R830878 (2004) R830878 (Final) |
Exit Exit |
|
Tzortziou M, Osburn CL, Neale PJ. Photobleaching of dissolved organic material from a tidal marsh-estuarine system of the Chesapeake Bay. Photochemistry and Photobiology 2007;83(4):782-792. |
R830878 (Final) |
Exit Exit |
|
Tzortziou M, Neale PJ, Osburn CL, Megonigal JP, Maie N, Jaffe R. Tidal marshes as a source of optically and chemically distinctive colored dissolved organic matter in the Chesapeake Bay. Limnology and Oceanography 2008;53(1):148-159. |
R830878 (Final) |
Exit |
Supplemental Keywords:
Chesapeake Bay; modeling; water quality; dissolved oxygen; water clarity; chlorophyll; subestuaries; UV radiation; wetlands; suspended sediments,, RFA, Scientific Discipline, Air, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, climate change, State, Air Pollution Effects, Monitoring/Modeling, Regional/Scaling, Environmental Monitoring, Ecological Risk Assessment, Atmosphere, anthropogenic stress, coastal ecosystem, aquatic species vulnerability, biodiversity, environmental measurement, ecosystem assessment, meteorology, climatic influence, Virginia (VA), global change, anthropogenic, climate models, UV radiation, greenhouse gases, environmental stress, coastal ecosystems, plankton, water quality, ecological models, climate model, Global Climate Change, land use, regional anthropogenic stresses, atmospheric chemistry, stressor response model, climate variabilityProgress 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.