Journal des sciences de la Terre et du changement climatique

Abstrait

Modelling Phosphorus Export from Humid Subtropical Agricultural Fields: A Case Study Using the HSPF Model in the Mississippi Alluvial Plain

Diaz-Ramirez J, Martin JL and William HM

This research used extensive field data collected during the 1990’s by a consortium of federal, state, local, and university personnel in agricultural areas in the Mississippi Delta region to evaluate runoff, soil erosion, and nutrient losses. Agricultural phosphorus export modelling was performed on Deep Hollow watershed (11.29 ha) using the
Hydrological Simulation Program FORTRAN – HSPF to demonstrate the usefulness of the continuous processesbased model for a farm scale drainage area in the Mississippi Delta region. The main goal of this study was to evaluate the ability of HSPF to simulate storm, seasonal, and long-term runoff, sediment, and phosphorus export at the farm scale in Mississippi. The model was spatially divided into six hydrologic respond units based on similarity of land cover and soil characteristics. Rainfall and other climate time series were setup every 15-minutes and the model results were output at the same time step. Agricultural field operations (tillage, cultivation, harvest, fertilization) were incorporated into the HSPF model using the special actions module. Phosphorus simulations were accomplished using the AGCHEMPHOS module. Phosphorus adsorption and desorption were simulated by the Freundlich method. Model outputs: runoff, sediment load, dissolved & total phosphorus loads were evaluated against observed data from storm, seasonal, and long-term periods. Collected storm-event data at the drainage area outlet were available for runoff (69 events), suspended sediment load (40 events), and phosphorus loads (23 events) from 1997 to 1999. The model was calibrated using available data from 1998 to 1999 and validated using field data in 1999. All the parameter values used in this assessment were in the range provided by HSPF model developers or found in the literature. Simulated runoff values showed very good agreements to storm, monthly, and annual basis. In analyzing simulated sediment load export values, storm type steps yielded poor performance, but HSPF produced good seasonal and annual sediment load values. Simulated dissolved phosphorus (DP) loads were better at long-term time steps than storm-by-storm periods. HSPF total phosphorus (TP) loads were better at simulating long-term periods than storm-by-storm time steps. In conclusion, HSPF runoff algorithms are very robust at any time scales from storm to annual periods and sediment and phosphorus modules are very comprehensive for seasonal and annual time steps at the farm scale.