Page Banner

United States Department of Agriculture

Agricultural Research Service

2008 Annual Report

1a.Objectives (from AD-416)
The objectives of this project are to support a national assessment of the environmental effects of USDA conservation programs by providing detailed findings for a few intensively studied watersheds and to improve the performance of models to be used in the assessment. Specific objectives are:.
1)Develop and implement a data system to organize water, soil, management, and socio-economic watershed data;.
2)Quantify water quality, water quantity, soil quality, and ecosystem effects of conservation practices at the watershed scale;.
3)Validate models and quantify uncertainties of model prediction;.
4)Develop and apply policy-planning tools to aid selection and placement of conservation practices to optimize profits, environmental quality, and conservation practice efficiency; and.
5)Develop regional watershed models that quantify environmental outcomes of conservation practices.

1b.Approach (from AD-416)
The general approach is the acquisition, analysis, and interpretation of data from 14 ARS Benchmark Watersheds and the testing and evaluation of models for the national assessment. Conservation practices are being applied on the 14 watersheds. Development and testing of watershed models will be associated with the 14 watersheds. The watersheds provide a cross-section of climate, soils, land use, topography, and crops across major production regions of the U.S. The research will be carefully coordinated. Six multi-location teams will guide the research, with a specific team being responsible for each of the five objectives and a sixth team providing quality assurance guidelines for the other teams. This multi-location project will be affiliated with the following location-specific projects: 1265-13610-026-00D, 1902-13000-010-00D, 3602-12000-011-00D, 3602-12220-NEW-00D, 3604-13000-007-00D, 3622-12130-003-00D, 3625-12130-003-00D, 3625-13000-008-00D, 5358-21410-002-00D, 5368-13000-006-00D, 5402-13660-006-00D, 6206-13610-005-00D, 6218-13000-009-00D, 6408-13000-017-00D, 6408-13660-005-00D, 6602-13000-020-00D, 6602-13000-021-00D.

3.Progress Report

This multi-location project serves to coordinate among the multiple individual Conservation Effects Assessment Projects (CEAP). Specific progress is reported in the individual, location-based progress reports. An emphasis during FY08 was to contribute 23 scientific papers toward a special CEAP issue of the Journal of Soil & Water Conservation, to be published in Nov-Dec issue. Below is a brief summary of progress by objective. Performance is monitored through corresponding accountability in location project plans, and through ties from project milestones to individual scientist performance appraisals. Additionally, NRCS and other stakeholders attend the annual CEAP meeting to view results, interact, and provide input on their needs. Extensive interactions among scientists and stakeholders occur at local levels. Objective 1. As of September 2008, six watersheds were substantially represented in STEWARDS. Data incorporation was enhanced by error-checking to retain data accuracy during transfer among watershed and database personnel. STEWARDS database system training included a webinar for data and scientific staff and interaction during data preparation. Operational staff is in place. Suggestions identified during the beta phase were incorporated into the operational version of STEWARDS, which went online to the public 18 September. Objective 2. Assessments of water quality and conservation practices are continuing in all watersheds, and progress is according to schedule. Samples for soil quality assessments were done in several locations, and analyses are in progress. Rapid geomorphic assessments and sediment-source tracking are also continuing in some watersheds. Objective 3. SWAT has been calibrated and validated for 8 watersheds in 6 states, with work begun in 2 others. The landscape version of SWAT has begun testing on Little River (GA) and a linkage between SWAT and REMM has been done by Canadian collaborators. SWAT-APEX integration has been done for Leon (TX). AnnAGNPS has been calibrated and validated for 3 MS watersheds that have ephemeral gully and channel sources, as well as the Choptank (MD) and St. Joseph (IN) watersheds. Objective 4. We have adapted an indexing method from economics to evaluate water quality, in which all metrics (e.g., sediment, nitrates, nitrite) are weighted and aggregated. Rather than weighting by expert opinion or by statistical methods, we assign weights based on observed relationships in the data. The index calculator can be integrated into the multi-objective genetic algorithm. Objective 5. The Enhanced Object Modeling System (OMS) was delivered to NRCS. A prototype OMS-based regionalized watershed model was built and tested. Team 6. ARS Data Quality Assurance completed a second year in the NAPT proficiency testing program. Team 6 established an agreement to use an existing database for comparative analyses of different methods used in soil and water analysis.

1. Data management enhances watershed research capacity. Interdisciplinary research across natural and social sciences to address challenges in water resource management requires comprehensive and long-term data. As part of USDA’s Conservation Effects Assessment Project (CEAP), STEWARDS (Sustaining the Earth’s Watersheds, Agricultural Research Data System) was developed to compile, document and provide access to data from ARS research watersheds, which represent one of the largest research watershed data collections in the world, with many of the watersheds offering decades of data to address issues of climate variability and global change. Major contributions include: a paired data table and data definition table programming methodology, increased visibility for watersheds by publishing of metadata, the ARS Methods Catalog which provides transparency and can serve as a resource for other researchers, training opportunities and expanded networks for watershed data managers, and a template for watershed data management, in addition to the primary contribution - STEWARDS populated with data from multiple watersheds. This data system represents a move forward for hydrologic and environmental research by providing access to a multitude of data needed to support complex analyses. Challenges remain in data preparation and upload, including providing incentives and scientific credit to scientists who contribute to open source data systems. Anticipated impacts include increased productivity and collaborative opportunities for individual scientists, watershed teams, and the ARS water resources program and better accountability at the agency level for investment in long-term watershed research. (NP211, Problem Area 1)

2. Enhanced Object Modeling System (OMS) Delivered to NRCS. Modular frameworks for model development like OMS are well-suited for comprehensive projects like CEAP that require complex simulation component technology integrated into a common, collaborative, and flexible system. In order to provide such a system, the OMS application programming interface (API) was enhanced to allow flexible simulation management, self-contained model run-time execution (i.e., OMS-based models can be run without full installation of the OMS framework), and improved data file handling for input parameter sets and simulation output based on the generic CSV format. OMS was also extended to transfer and retrieve science modules to and from the OMS component library residing under the USDA Colab project management environment, and a prototype OMS component that allows visualization and manipulation of geospatial data was developed using NASA World Wind geospatial technology. The OMS was officially transferred to the NRCS on February 26, 2008, and will streamline development of customized, modular field to watershed agricultural system models to be delivered to the NRCS to address regional soil and water conservation and water quality needs. A journal paper describing space-time data structures in OMS is currently under preparation. [Contributes to Problem Area #1, Effectiveness of Conservation Practices, Product #5 of NP 201 Action Plan (FY 2006-2010).]

3. OMS-Based Prototype Regionalized Watershed Model. In addition to needed science base improvements for ARS watershed models, a flexible modular modeling configuration is needed to efficiently build a watershed model customized to regional processes, concerns, and issues. Java-based simulation modules (80+ representing interception, snow processes, soil water balance, lateral flow and ground water movement, and runoff concentration/flood routing in channels) from the European J2000 modular process-oriented hydrological system were integrated under OMS to form a prototype regionalized watershed model. The resulting prototype model is unique in allowing for conjunctive stream flow and groundwater interaction, carried out by hydrological response units (HRUs) which are connected by a lateral routing scheme to simulate lateral water transport processes. This allows a fully distributed hydrological modeling of river basins. The prototype model under OMS was tested against the J2000 model running under the Jena Adaptable Modelling System (JAMS) for reference watersheds in the United States and Europe in order to verify the accuracy of the OMS implementation. Formal evaluation of the prototype for stream flow is being performed using data from an ARS CEAP Watershed in Indiana (Cedar Creek Watershed). [Contributes to Problem Area #1, Effectiveness of Conservation Practices, Product #5 and Problem Area #3, Drainage Water Management Systems, Product #4 of NP 211 Action Plan (FY 2006 – 2010)]

4. Spatial density of rainfall measurement affects uncertainty in prediction of loads. Results from the Little River Experimental Watershed near Tifton, GA indicate that annual rainfall and corresponding streamflow were sensitive to subwatershed delineation. As a result, the thiessen-method or the centroid-method with sufficient subwatersheds is recommended for SWAT simulation of a watershed with high spatial variability of rainfall. Output uncertainties increased as the gauge-density decreased. Total phosphorus (TP) was the most sensitive to the changes in raingauge densities, followed by sediment, total nitrogen (TN), and hydrology. Seasonal variations in simulated hydrology and water quality were higher during summer and fall compared to spring and winter. These seasonal and temporal variations according to gauge-density scenarios can be attributed to the rainfall patterns within the watershed. Proposed methods and results will be helpful to prepare multiple raingauge input for SWAT and to understand the possible spatial and temporal uncertainties of SWAT output at other watersheds. [Contributes to NP 211, Water Availability and Watershed Management, Problem Area 1, Effectiveness of Conservation Practices.]

5. Uncertainty in SWAT predictions of pollutant loads varies by pollutant. Results from the Little River Experimental Watershed near Tifton, GA indicate that using the SWAT model to predict pollutant load reductions in response to conservation management practices provides different levels of uncertainty depending on the pollutant of interest. SWAT was calibrated to an eight-year period from 1979 to 1986 and validated against two different 9-year simulation periods from1987 to1995 and from 1996 to 2004. Typical crop rotations were assigned for each simulation period based on long-term changes in dominant crops within the surrounding counties. Typical agronomic schedules for both conventional and conservation crop management practices were defined for each major crop. Impacts of crop management practice on hydrology and water quality were evaluated by considering changes from conventional to conservation management practices on crop areas. Estimated pollutant load reduction rates accompanying increases in conservation management practices up to 30% of watershed area were 60.3% for sediment, 39.3 % for total P, and 17.2 % for total N while changes in total streamflow were as small as 1.0 %. However, estimated rates of reduction were very sensitive to select model parameters, especially FILTERW and curve number (CN). The uncertainty in predicted pollutant reduction due to changes in these two input parameters alone was highest in TP (+/- 68%) followed by TN (+/- 45%) and sediment (+/- 19%). [Contributes to NP 211, Water Availability and Watershed Management, Problem Area 1, Effectiveness of Conservation Practices.]

6. Conservation practice effects on sediment load. Knowledge of the effect of agricultural management practices on sediment yield has been derived predominantly from studies on plots or field-sized watersheds which may not be representative of the scales and complexity that typically exist on larger watersheds. Effects of enrolling erodible lands in the Conservation Reserve Program (CRP) and in-stream grade stabilization structures were evaluated using measured rainfall, runoff, and sediment concentration data and model simulations on the 2132 ha Goodwin Creek Experimental Watershed in north Mississippi. The combined effect of the grade control structures and the change of crop lands to a CRP-state (reducing cultivated land from 26 to 8%) was to reduce sediment yields by 78% near the outlet of the watershed. This study provided a quantitative assessment of the effects of converting crop lands to a CRP-state and of grade control structures in the channels of the watershed. This type of information is needed by watershed managers seeking to reduce sediment loads and is useful to evaluate the performance of watershed simulation models. [Contributes to NP 211, Water Availability and Water Management National Program 211, Problem Area 1--Effectiveness of Conservation Practices, Product 2.]

5.Significant Activities that Support Special Target Populations

6.Technology Transfer

Number of Web Sites Managed1

Last Modified: 10/10/2015
Footer Content Back to Top of Page