Grantee Research Project Results
2008 Progress Report: Nonlinear and Threshold Responses to Environmental Stressors in Land-river Networks at Regional to Continental Scales
EPA Grant Number: R833261Title: Nonlinear and Threshold Responses to Environmental Stressors in Land-river Networks at Regional to Continental Scales
Investigators: Melillo, Jerry , Peterson, Bruce J. , Kicklighter, David Wesley
Current Investigators: Melillo, Jerry , Peterson, Bruce J. , Vörösmarty, Charles J. , Felzer, Benjamin S. , Kicklighter, David Wesley , McClelland, James , Wollheim, Wil
Institution: Marine Biological Laboratory
Current Institution: Marine Biological Laboratory , University of New Hampshire
EPA Project Officer: Packard, Benjamin H
Project Period: September 1, 2007 through August 31, 2010 (Extended to August 31, 2011)
Project Period Covered by this Report: November 1, 2007 through October 31,2008
Project Amount: $899,191
RFA: Nonlinear Responses to Global Change in Linked Aquatic and Terrestrial Ecosystems and Effects of Multiple Factors on Terrestrial Ecosystems: A Joint Research Solicitation- EPA, DOE (2005) RFA Text | Recipients Lists
Research Category: Climate Change , Aquatic Ecosystems
Objective:
Our objective in this research is to explore the relationships among environmental stresses, the nonlinear and threshold behaviors they cause within freshwater ecosystems of large drainage basins, and the ecosystem services provided by the streams and rivers of the basins. To do this we are refining our process-based aquatic model, the Aquatic Ecosystem Model (AEM) and testing its ability to simulate documented nonlinear and threshold responses to environmental stresses at a variety of spatial scales, from the river reach to the entire river network within a drainage basin. We will then couple the AEM with our improved terrestrial biogeochemistry model, the Terrestrial Ecosystem Model (TEM), thereby creating a new version of our Drainage Basin Model, which we will use for regional analyses of nonlinear and threshold behaviors in freshwater systems at large scales. A cartoon conceptualization of these couplings appears in Figure 1.
Figure 1. Conceptualization of the Drainage Basin Model. Nonpoint loading of solutes and particulates to river and stream ecosystems will first be predicted by TEM. The AEM then will use these simulated nonpoint loadings along with estimates of point sources as inputs into both local (within grid) and ultimately macro-scale (between grid) river networks. AEMg will simulate biological processing of organic matter and nutrients in low order streams within a grid cell computing the net export of constituents to higher order macro-level streams. AEMrc then will simulate additional biological processing of organic matter and nutrients in major river corridors to estimate the export of materials on their movement downstream to the river mouth. Water will be moved between stream reaches with the Water Transport Model (WTM). Potential 8 km networks are shown. The heavy black line on the map represents the Missouri and lower Mississippi Rivers.Progress Summary:
The coupling of the component models is being done in a modeling framework, FrAMES, specifically designed by us for land-water interaction studies. FrAMES is briefly described in Appendix 1 of this progress report.
OVERVIEW:
We have made significant progress in refining both the AEM and the TEM and we have begun to incorporate TEM into FrAMES. Our research to date has resulted in 5 publications and four presentations at national meetings.
AEM Progress: Our goal in this part of the research is to develop a version of AEM that will allow us to explore nonlinear responses to various climate and terrestrial drivers. We are developing AEM to have a moderate level of complexity so that it is compatible with the terrestrial ecosystem model (TEM), and will facilitate our exploration of coupled dynamics of terrestrial and aquatic ecosystems. We have been adding requires additional functionality to AEM to adequately represent key aquatic ecosystem drivers (temperature, light). Examples include quantification of water temperature and light energy inputs through the water column, both of which are influenced by aquatic processes and conditions as water is routed through the river network. The hydrology module, which predicts runoff, discharge, and channel hydraulics, has already been developed in FrAMES, and will provide the hydrologic conditions necessary to model ecosystem processes in both the terrestrial and aquatic environments. The terrestrial inputs of energy (carbon transfer from land to water, i.e. allochthonous inputs) and nutrients (nitrogen and phosphorus) will be driven by TEM, which has recently been integrated into FrAMES, but not yet fully coupled with AEM.
Here, we describe progress towards developing model functions that determine AEM drivers (Water Temperature, Light) and several ecosystem processes (Denitrification, Nitrogen Assimilation/Storage, benthic primary production). The full AEM will take the shape of the conceptual diagram in Figure 2. We also present early applications of AEM that explore non-linear dynamics across a variety of river networks. The river network perspective is central to the development of the AEM. Finally, we describe plans to further develop AEM (adding in respiration, remineralization, particulates, etc.). The ultimate goal is to understand how water quality and ecosystem health has changed over the last 50 years in response to changes in land cover, land use (e.g., fertilization) and climate, and how they might respond under future management and climate scenarios.
Figure 2. Conceptual model of the Aquatic Ecosystem Model (AEM) that is being implemented in FrAMES.
Water Temperature Module. Water temperature is an important water quality parameter in its own right (e.g., fish habitat), but also influences other water quality parameters (e.g., oxygen, nitrogen) via its influence over many biogeoechemical processes. Water temperature is regulated by the temperature of terrestrial runoff, and re-equilibration during routing. In FrAMES-AEM, temperature of terrestrial runoff is regulated by source of runoff (surface vs. groundwater), air temperature at time of runoff and groundwater recharge, and integration of groundwater temperature through time. Temperature equilibration in the river network uses the empirical re-equilibration model developed by Dingman (1974). Equilibrium temperature is a function of solar radiation and air temperature. The rate of re-equilibration is a function of wind speed and cloud cover. Water temperature (Tw) leaving a particular grid cell is thus modeled as:
Where Tw is the actual surface water temperature, To is the initial temperature, accounting for mixing of upstream inputs, local inputs, and channel storage. Te is equilibrium temperature, L is river length, p is density of water = 999.73 (g/m3), Cp = specific heat of water = 4.1922 kJ/g/C, Q is discharge (m3/s), w is channel width (m), k = energy exchange coefficient (kJ/m2/C/d).
Initial results are highly encouraging, suggesting this water temperature module can be applied across a wide range of biome types. Time series of predicted and observed temperature are comparable for a number of U.S. stream gages (e.g., Figure 3). Results indicate that accounting for re-equilibration during routing is critical for adequately predicting water temperature. Maps of water temperature reveal patterns consistent with expectation. Additional work that is required to fully model water temperature includes the incorporation of riparian shading influences, reservoir operating procedures, and irrigation withdrawals.
Figure 3. A) Time series of predicted and observed water temperature near the mouth of the Mississippi R. between 1970 and 1990. Predicted-mixing indicates predicted water temperature if no re-equilibration occurred during routing. B) Map showing number of days water temperature is greater than 25º C during 1990 in the Mississippi R. basin.
Light Module. Light inputs regulate primary production rates in aquatic systems, which in turn influences carbon inputs, nutrient uptake, and oxygen conditions. We have developed a light module that predicts light inputs to the benthic sediments, based on the reach-scale model described by Julian, et al. (2008). Light is modeled as a function of solar radiation, shading (a function of riparian cover, leaf area index, and stream width), reflectance at the water surface, water depth, and light extinction through the water column. Light extinction is a function of turbidity and DOC concentrations. The light module drives gross primary production described below. Riparian cover and leaf area index will ultimately be derived from TEM outputs, while stream width and depth are predicted by the hydrology module. DOC is currently predicted as a function of wetland percentage in the grid cell and runoff (driving the inputs), and a specified decay coefficient. DOC inputs will also ultimately be obtained from TEM in the fully coupled mode. Turbidity is currently specified. Figure 4 shows simulation results from a preliminary version of our light module – photosynthetically active radiation reaching the river bottom in January and June in the Mississippi River Basin.
Figure 4. Figure 4. Photosynthetically active radiation reaching the river bottom in January and June.
Ecosystem Modules. We have developed an initial set of ecosystem modules to predict benthic gross primary production (GPPben), denitrification and nitrogen assimilation. GPP is modeled based on Kirk (1994) and Ryther and Yentsch (1959) as:
Where CHLben is benthic chlorophyll, GPPmax is the maximum GPP rate at light saturation, Ed is light at depth, Ek is light at onset of saturation, DIN and DIP are inorganic N and P concentrations, and Ks_N and Ks_P are the half saturation constants for N and P. CHLben is estimated from GPP assuming a C:CHL of 50:1. CHL experiences mortality and sloughing. Light is modeled as described above. DIN is modeled as a function of inputs from terrestrial systems based on the predictions of Green, et al. (2004) and denitrification, assimilation, and remineralization. DIN inputs will ultimately be derived from TEM. Denitrification and assimilation rates are based on the non-linear efficiency loss model recently published by Mulholland, et al. (2008). An exploration of the non-linear consequences of this model at basin scales is described in the next section. Preliminary maps of benthic GPP throughout the Mississippi are shown in Figure 5. These results still need to be tested, and the model refined (e.g., better parameterization of algal sloughing and mortality is still needed).
Figure 5. Predicted GPP (g C / m2 / d) throughout the Mississippi River basin in January and June, given the light field shown in Figure 3.
Non-linear Dynamics. Denitrification and nitrate assimilation show declining uptake efficiency with increasing concentration (Mulholland, et al., 2008). That is, as concentrations increase, process rates increase at a slower rate, described by a power function. We have begun to explore the consequences of this efficiency loss model at river network scales, and in conjunction with hydrologic variability (Wollheim, et al., 2008). Early application was to a single river basin in northeastern MA, the 400 km2 Ipswich river basin. We have also applied the model across a wide variety of basins at mean annual time scales (Wollheim, et al., In preparation). The model indicates that a decline in river network removal efficiency has occurred between preindustrial and contemporary settings due to the large increase in N inputs combined with efficiency loss (Figure 6). Aquatic ecosystem services, therefore, are predicted to be less effective at controlling material fluxes through river basins as humans continue to perturb the N cycle and as terrestrial ecosystems saturate. In real world terms, the rate of export to the coastal ocean is predicted to have increased at a rate faster than inputs to aquatic systems. Results show that at a threshold DIN input level of 1,000 kg/km2/yr, river networks will be unable to denitrify more than 20% of inputs, regardless of their characteristics. However, comparison with observations at basin mouths suggest that high N load basins have higher than expected removal than expected based on surface water denitrification alone.
Figure 6. Predicted proportion of DIN inputs to river networks removed via aquatic denitrification as a function of the rate of DIN input, assuming the measured efficiency loss for the denitrification process (Mulholland, 2008). Considerable variability exists across basins because of variability in runoff conditions, temperatures, lake abundance, reservoir abundance (contemporary setting only), and distribution of N inputs. However, above a certain threshold of DIN inputs (~ 1,000 kg / km2 yr), no river networks are apparently capable of denitrifying more than 20% of inputs.
TEM PROGRESS: We have worked to develop an improved version that will allow us to better simulate land-water couplings in the face of ongoing environmental changes including climate change. We have also begun to incorporate TEM into FrAMES, which will facilitate the AEM-TEM coupling.
Developing and Validating TEM-Hydro - Consideration of the effects of climate change on hydrology over land surfaces and consequent land-water couplings often focuses on changing precipitation or temperature, and neglects the key role that vegetation plays in mediating the exchange of water vapor between the land and atmosphere. Elevated CO2 alone is likely to increase runoff due to reduced stomatal conductance and transpiration. Increases in photosynthetic rates due to higher CO2 may offset these reductions in stomatal conductance and transpiration. Because of the linkages between photosynthesis and transpiration, factors such as nitrogen limitation and ozone damage, which reduce photosynthesis, also have the capability to significantly impact the water cycle.
As part of our research on this project we have developed a new version of TEM, currently called TEM-Hydro. We have done this by incorporating into the TEM structure a more detailed biophysics, and the disaggregation of vegetation carbon and nitrogen pools. The calculation of evapotranspiration now employs the formulation of Shuttleworth and Wallace (1985) that explicitly considers both soil evaporation and plant transpiration. To determine canopy stomatal conductance, we employ the approach of Ball, et al. (1987). A conceptualization of the new coupling of the carbon, nitrogen and water cycles in TEM-Hydro appears
Figure 7.
Figure 7. Schematic showing the couplings of carbon, nitrogen and water cycle as represented in TEM-Hydro.
To date, we have validated the model for eastern deciduous forest ecosystems and done a set of climate simulations to explore the relative effects of future changes in atmospheric CO2 concentration, climate and nitrogen availability on evapotranspiration and runoff from forested drainages. With respect to the validation, TEM-Hydro captures the estimated mean annual evapotranspiration (Figure 8) in forested eastern U.S. basins as well as the observed seasonality and interannual variations of river discharge. A paper documenting this research is in press in the Journal of Geophysical Research.
Figure 8. Measured versus simulated evapotranspiration in forested watershed in the eastern US (Felzer, et al., 2009).
APPENDIX 1 - Framework for Aquatic Modeling of the Earth System
The Framework for Aquatic Modeling of the Earth System developed at the University of New Hampshire (former home institution of Prof. Vörösmarty and Dr. Fekete) and maintained at The City College of New York at CUNY, is the third generation modeling platform developed by the UNH/CCNY team to support the development of hydrological and land surface models. FrAMES separates the core model infrastructure functions (such as managing space and time, performing input/output, and model execution) from the science components such as model layout (the interaction between various processes) as well as the actual implementation of the processes themselves.
FrAMES modules define the processes occurring on single computational objects (grid cell, distinct point, etc.). Modules are implemented as pairs of function calls. Module-definition functions allow the configuration of the model layout by requesting variables for exchanging data values between individual modules and building a list of the actual module functions. FrAMES manages all the requested variables including I/O basic disaggregation, etc. FrAMES also provides rudimentary support for horizontal processes over multiple computational objects linked via object topology. The object topology (currently network tree or 2D grid) determines the available operators (route or derivatives). During model execution, the framework steps through time, updates input forcings, performs mathematical calculations on each grid cell, propagates variable outputs horizontally, and outputs all requested variables. FrAMES is surrounded by a collection of support tools that carry out the pre- and post-processing of the input and output variables and provides basic visualization and data sampling.
FrAMES is incorporated into our Global Rapid Integrated Monitoring system (Global-RIMS[1]). Global-RIMS provides the both data archive for a wide variety of input data (climate, land cover, topography, river networks, river discharge, etc.) and the publishing of the archive and the processed data through web services. FrAMES operates directly from the Global-RIMS data archive through web-based data services (OpenDAP) and produces outputs that are ready for publications through similar data services. By implementing the coupled TEM/DBM model in FrAMES not only simplifies the data management but allows us to publish our results through standard web interfaces provided by Global-RIMS.
Besides the coupled model development we started to extend the existing data archive in Global-RIMS with high resolution regional and global datasets (HydroSHEDS high resolution river network[2], ) to support our EPA effort.
[1] http://www.global-rims.net
[2] http://hydrosheds.cr.usgs.gov
Future Activities:
AEM Next Steps. We plan to continue development and testing of AEM modules to provide a suite of key ecosystem processes needed to model key water quality attributes and aquatic ecosystem condition through time (Figure 1). The major processes that we will be adding include respiration of allochthonous carbon inputs, particulate fluxes, and oxygen conditions. Over the next year, TEM will be fully incorporated into FrAMES, allowing us to route predicted carbon and nitrogen fluxes from terrestrial systems as a function of land use change and climate change/variability. Coupled model results will be tested against EPA and USGS water quality observations for the period of record.
TEM Next Steps. We plan to continue our integration of TEM into FrAMES, allowing us to route predicted carbon and nitrogen fluxes from terrestrial systems as a function of land use change and climate change/variability. Coupled model results will be tested against EPA and USGS water quality observations for the period of record.
References:
Journal Articles on this Report : 5 Displayed | Download in RIS Format
Other project views: | All 28 publications | 13 publications in selected types | All 13 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Alexander RB, Bohlke JK, Boyer EW, David MB, Harvey JW, Mulholland PJ, Seitzinger SP, Tobias CR, Tonitto C, Wollheim WM. Dynamic modeling of nitrogen losses in river networks unravels the coupled effects of hydrological and biogeochemical processes. Biogeochemistry 2009;93(1-2):91-116. |
R833261 (2008) R833261 (2009) R833261 (2010) R833261 (Final) R834187 (2010) R834187 (2011) R834187 (2012) R834187 (2013) R834187 (Final) |
Exit Exit Exit |
|
Felzer BS, Cronin TW, Melillo JM, Kicklighter DW, Schlosser CA. Importance of carbon-nitrogen interactions and ozone on ecosystem hydrology during the 21st century. Journal of Geophysical Research-Biogeosciences 2009;114(G1):G01020 (10 pp.). |
R833261 (2008) R833261 (2009) R833261 (2010) R833261 (Final) |
Exit Exit |
|
Harrison JA, Maranger RJ, Alexander RB, Giblin AE, Jacinthe P-A, Mayorga E, Seitzinger SP, Sobota DJ, Wollheim WM. The regional and global significance of nitrogen removal in lakes and reservoirs. Biogeochemistry 2009;93(1-2):143-157. |
R833261 (2008) R833261 (2009) R833261 (2010) R833261 (Final) R834187 (2010) R834187 (2011) R834187 (2012) R834187 (2013) R834187 (Final) |
Exit |
|
Wollheim WM, Vorosmarty CJ, Bouwman AF, Green P, Harrison J, Linder E, Peterson BJ, Seitzinger SP, Syvitski JPM. Global N removal by freshwater aquatic systems using a spatially distributed, within-basin approach. Global Biogeochemical Cycles 2008;22(2):GB2026. |
R833261 (2008) R833261 (2010) R833261 (Final) |
Exit Exit |
|
Wollheim WM, Peterson BJ, Thomas SM, Hopkinson CH, Vorosmarty CJ. Dynamics of N removal over annual time periods in a suburban river network. Journal of Geophysical Research-Biogeosciences 2008;113(G3):G03038. |
R833261 (2008) R833261 (2010) R833261 (Final) |
Exit Exit |
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.