|Yuan, Yongping - UNIV OF MISSISSIPPI|
Submitted to: International Journal of Sediment Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: January 9, 2007
Publication Date: February 15, 2007
Citation: Yuan, Y., Bingner, R.L., Williams, R.G., Lowrance, R.R., Bosch, D.D., Sheridan, J.M. 2007. Integration of AnnAGNPS and REMM for watershed riparian buffer system assessment. International Journal of Sediment Research, 22(1): 60-69. Interpretive Summary: The development of management practices to reduce pollutants produced from agricultural fields entering stream systems is critical for improving overall watershed water quality. The implementation of riparian buffer areas along streams can be a management approach utilized throughout a watershed with a minimal impact on farming practices. The evaluation of the effectiveness of the buffers using conventional measurement techniques can be difficult because of the slow flow of water through the vegetation that comprises the buffer. In order to better assess the effectiveness of riparian buffers, computer simulation models have been developed to evaluate the upland contributions to the riparian areas as well as the effect of the riparian buffer areas. This study combines the simulation approach of an upland computer model and a riparian buffer system model and assesses the capability of the combined models to perform watershed evaluations. This study will also be used to determine the feasibility of developing a more comprehensive and integrated modeling approach which is needed to evaluate the impact of agricultural production practices along a riparian buffer system throughout a watershed system. By using the data of a monitored watershed in Georgia, the simulation approach was able to show a five-year sediment reduction of 93% due to the effect of the riparian buffer. Most of the soil in the study site was consisted of sand material, which the buffer completely trapped, as well as some silt and to a lesser extent clay material. Attached to the trapped sediment were nitrogen and phosphorus compounds that the models also were able to accurately reflect. Models are effective tools for evaluating the best management approach for any given watershed. By using these tools on any watershed in the United States, watershed managers will be able to use a systematic approach in determining the best practices available, including riparian buffers, to improve watershed water quality.
Technical Abstract: The USDA Annualized Agricultural Non-Point Source Pollution model (AnnAGNPS) has been developed to aid in the evaluation of watershed response to agricultural management practices. However, AnnAGNPS does not currently contain features to estimate the effect of riparian buffer systems (RBS) on water quality. The Riparian Ecosystem Management Model (REMM) has been developed by USDA-ARS to simulate the water quality impacts of riparian and other edge of field buffer systems. However, information on upland water and pollutant loading from fields to the RBS for REMM is normally lacking and preparing this information can be very time consuming. This can be improved by using AnnAGNPS to provide upland loadings to REMM. The development and integration of AnnAGNPS and riparian buffer technology from REMM would provide a powerful tool that can evaluate the impact of agricultural management practices as well as RBS at a watershed scale. In this study, AnnAGNPS was used to smulate water and pollutant loadings from upland field to riparian buffers at the Gibbs Farm in the Georgia coastal plain. These AnnAGNPS outputs were used as the inputs to REMM. REMM was used to simulate water and sediment movement along riparian buffers at the Gibbs Farm. AnnAGNPS simulated upland loadings were compared with observed upland loadings. REMM simulated runoff and sediment along riparian buffers wee compared with observed runoff and sediment. AnnAGNPS simulated upland loadings were compared with output from riparian buffer zones simulated by REMM to evaluate the overall function of RBS. AnnAGNPS simulataed annual runoff t the edge of the field is close to observed annual runoff except for one year (1993). Excluding sand, a good portion of which is believed to be removed from runoff one meter into a grass buffer, AGNPS predicted annual sediment yield is close to annual observed sediment yield. REMM simulated sediment along the RBS follows the pattern of observed sediment fairly well. More than 90% of the sediment from the edge of the field is removed after passing through the RBS.