Search

Conservation planners need simple yet accurate tools to predict sediment and nutrient losses from agricultural fields to guide conservation practice implementation and increase cost-effectiveness. The Texas Best Management Practice Evaluation Tool (TBET), which serves as an input/output interpreter and vastly simplified interface for the Soil and Water Assessment Tool (SWAT), was developed to predict mean annual runoff, sediment, nitrogen (N), and phosphorus (P) losses from agricultural fields in Texas under a variety of management scenarios and conservation practices. The Texas Best Management Practice Evaluation Tool was developed to provide a scientific basis for decision making, evaluation, and reporting in state and federal conservation programs. A rigorous, data-intensive calibration and validation process was employed in TBET development. First, hydrology outputs were calibrated with basin scale streamflow data from 20 US Geological Survey sites. Then, runoff, sediment, total N, and total P outputs were calibrated and validated with measured field scale data (260+ site years). Even without calibration,TBET was able to make "good" to "very good" predictions of mean annual runoff and total P losses according to commonly applied model evaluation methods. Calibrated TBET predictions of sediment and total N losses were not as accurate but were still "satisfactory." These much-improved results relative to those of often applied tools such as P Indices are very encouraging since the design goals of simple operation and accurate predictions for diverse Texas conditions with only readily available inputs were certainly met. Thus, TBET meets an important need by providing accurate, science-based estimates of conservation practice benefits at the field scale. [ABSTRACT FROM AUTHOR]

The application of poultry litter to agricultural land has become a topic of interest for policy makers due to public concern about its effects on water quality. The Soil and Water Assessment Tool (SWAT) version 2005 is designed to assess nonpoint and point sources of pollution. In this study, six subwatersheds in Texas (HUC-8; 12070101) are used to evaluate the model's ability to simulate water quality at a small scale. Each of these subwatersheds randomly received poultry litter rates of 0.0 to 13.4 Mg ha-1. Monthly and daily data from 2002 were used for calibration purposes, while 2000, 2001, 2003, and 2004 were used for validation. The SCS runoff curve number for moisture condition II (CN2) and the soil evaporation compensation factor (ESCO) parameters were found to be more sensitive than the surface runoff lag time (SURLAG) and initial soil water content expressed as a fraction of field capacity (FFCB). The monthly and daily runoff model simulations for the six subwatersheds resulted in calibration Nash-Sutcliffe efficiency (NSE) values of 0.59 and 0.53 and validation NSE values 0.82 and 0.80, respectively. The monthly and daily R² runoff values for the six subwatersheds resulted in calibration values of at least 0.60 and 0.53 and validation R² values of 0.86 and 0.81, respectively. The observed trends included SWAT's overestimation of runoff in the dry periods and underestimation in the wet periods. The monthly NSE and R² values for sediment and nutrient losses were generally above 0.4 and 0.5, respectively. Paired t-tests for the monthly manually adjusted parameter simulation of sediment, organic N and P, NO3-IV, and soluble P for the 2000-2004 period losses showed that their respective SWAT means were not significantly different from the measured values (α = 0.05), except for NO3-N losses for the Y10 subwatershed (p-value 0.042). The control subwatershed's measured and simulated water quality results were significantly different (α = 0.05) from the treated subwatersheds, most likely due to the amount of inorganic N present. Almost all of the subwatersheds that had poultry litter applied resulted in higher sediment, organic N, organic P, and soluble P losses than the control subwatershed upon averaging thee monthly validation values. High NO3-N losses may have been a function of poultry litter and commercial fertilizers being applied before a large rainfall event occurred. The subwatersheds that received smaller amounts of commercial fertilizer and/or poultry litter lost more sediment, organic N, and organic P than the subwatersheds that received the higher litter and/or fertilizer treatments. Overall, the SWAT simulated the hydrology and the water quality constituents at the subwatershed scale more adequately when all of the data were used to simulate the model, as evidenced by statistical measures. [ABSTRACT FROM AUTHOR]

Phosphorus (P) indices were developed to address nonpoint source P losses from agricultural fields; however, only limited information on P index performance at the field- and watershed-scale is available. Evaluation of P indices is necessary to provide the basis of modification and improvement of their usefulness as P management tools. In this study, the ability of the Texas P index to estimate P loss potential was evaluated by comparison with measured annual P loads over three years on four new pasture and six new cultivated litter application sites in the Texas Blackland Prairie. The Arkansas and Iowa P indices were also evaluated. The Texas and Iowa versions were able to provide reasonable estimates of P loss potential as illustrated with significant linear relationships (p <0.01) between P index values and measured annual P loads. In general, the P index values, Mehlich 3 soil test P, and poultry litter application rate were better correlated with dissolved P concentrations and loads (r² ranged from 0.12 to 0.91) than with total P and particulate P loads (r² ranged from 0.00-0.31). A major source of error in P index load estimations was their inability to capture variability in annual soil erosion. This source of error was dramatically reduced by using measured erosion instead of estimated annual average erosion (average r² values increased from 0.24 to 0.58). While these results illustrate a potential for the P Indices to make relative P loss assessments, research on incentives to prevent buildup of soil P levels, linkages between P levels in soils and receiving waters, and other important issues related to the use of P indices is warranted. [ABSTRACT FROM AUTHOR]