Skip to main content
ARS Home » Research » Publications at this Location » Publication #189748

Title: DISTRIBUTED HYDROLOGICAL MODELING OF TOTAL DISSOLVED PHOSPHORUS TRANSPORT IN AN AGRICULTURAL LANDSCAPE. PART I: DISTRIBUTED RUNOFF GENERATION

Author
item GERARD-MARCHANT, PIERRE
item Hively, Wells - Dean
item STEENHUIS, TAMMO

Submitted to: Hydrology and Earth System Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/8/2005
Publication Date: 8/22/2005
Citation: Gerard-Marchant, P., Hively, W.D., Steenhuis, T.S. 2005. Distributed hydrological modeling of total dissolved phosphorus transport in an agricultural landscape, Part I: Distributed runoff generation. Hydrology and Earth Systems Sciences. 2:1537-1579.

Interpretive Summary: Hydrological modeling of the agricultural landscape can assist in the evaluation of non-point source pollution sources, and the improved management practices that can be used to reduce the loss of agricultural nutrients from the farm landscape. This study applied a fully-distributed hydrological model (Soil Moisture Distribution and Routing model) to a small agricultrual watershed in the headwaters of the New York City water supply system. The model, which relied upon readily available input datasets including topography, soil type, vegetation type, and climate, performed well, succeeding in predicting the spatial distribution of landscape areas that frequently produced runoff. Results can be used to identify critical areas within the landscape, where the risk of runoff is high and excess nutrient application should be avoided.

Technical Abstract: Successful implementation of best management practices for reducing non-point source (NPS) pollution requires knowledge of the location of saturated areas that produce runoff. A physically-based, fully-distributed, GIS-integrated model, the Soil Moisture Distribution and Routing (SMDR) model was developed to simulate the hydrologic behavior of small rural upland watersheds with shallow soils and steep to moderate slopes. The model assumes that gravity is the only driving force of water and that most overland flow occurs as saturation excess. The model uses available soil and climatic data, and requires little calibration. The SMDR model was used to simulate runoff production on a 164-ha farm watershed in Delaware County, New York, in the headwaters of New York City water supply. Apart from land use, distributed input parameters were derived from readily available data. Simulated hydrographs compared reasonably with observed flows at the watershed outlet over a eight year simulation period, and peak timing and intensities were well reproduced. Using off-site weather input data produced occasional missed event peaks. Simulated soil moisture distribution agreed well with observed hydrological features and followed the same spatial trend as observed soil moisture contents sampled on four transects. Model accuracy improved when input variables were calibrated within the range of SSURGO-available parameters. The model will be a useful planning tool for reducing NPS pollution from farms in landscapes similar to the Northeastern US.