|Ascough Ii, James|
Submitted to: American Geophysical Union
Publication Type: Abstract Only
Publication Acceptance Date: 10/24/2010
Publication Date: 12/15/2010
Citation: Ascough II, J.C., David, O., Heathman, G.C., Smith, D.R., Green, T.R., Krause, P., Kipka, H., Fink, M. 2010. Spatially-distributed stream flow and nutrient dynamics simulations using the component-based AgroEcoSystem-Watershed (AgES-W)Model. American Geophysical Union. Fall Meeting, December 13-17, 2010, San Francisco, California. Paper No. H43F-1329. Interpretive Summary:
Technical Abstract: The Object Modeling System 3 (OMS3), currently being developed by the USDA-ARS Agricultural Systems Research Unit and Colorado State University (Fort Collins, CO), provides a component-based environmental modeling framework which allows the implementation of single- or multi-process modules that can be developed and applied as custom-tailored model configurations. OMS3 as a “lightweight” modeling framework contains four primary foundations: modeling resources (e.g., components) annotated with modeling metadata; domain specific knowledge bases and ontologies; tools for calibration, sensitivity analysis, and model optimization; and methods for model integration and performance scalability. The core is able to manage modeling resources and development tools for model and simulation creation, execution, evaluation, and documentation. OMS3 is based on the Java platform but is highly interoperable with C, C++, and FORTRAN on all major operating systems and architectures. The ARS Conservation Effects Assessment Project (CEAP) Watershed Assessment Study (WAS) Project Plan provides detailed descriptions of ongoing research studies at 14 benchmark watersheds in the United States. In order to satisfy the requirements of CEAP WAS Objective 5 (“develop and verify regional watershed models that quantify environmental outcomes of conservation practices in major agricultural regions”), a new watershed model development approach was initiated to take advantage of OMS3 modeling framework capabilities. Specific objectives of this study were to: 1) disaggregate and refactor various agroecosystem models (e.g., J2K-S, SWAT, WEPP) and implement hydrological, N dynamics, and crop growth science components under OMS3, 2) assemble a new modular watershed scale model for fully-distributed transfer of water and N loading between land units and stream channels, and 3) evaluate the accuracy and applicability of the modular watershed model for estimating stream flow and N dynamics. The Cedar Creek watershed (CCW) in northeastern Indiana, USA was selected for application of the OMS3-based AgroEcoSystem-Watershed (AgES-W) model. AgES-W performance for stream flow and N loading was assessed using Nash-Sutcliffe model efficiency (ENS) and percent bias (PBIAS) model evaluation statistics. Comparisons of daily and average monthly simulated and observed stream flow and N loads for the 1997-2005 simulation period resulted in PBIAS and ENS values that were similar or better than those reported in the literature for SWAT stream flow and N loading predictions at a similar scale. The results show that the AgES-W model was able to reproduce the hydrological and N dynamics of the CCW with sufficient quality, and should serve as a foundation upon which to better quantify additional water quality indicators (e.g., sediment transport and P dynamics) at the watershed scale.