Excessive agricultural phosphorus (P) has been a major contributor to non-point source pollution. North Carolina developed the Phosphorus Loss Assessment Tool (PLAT) to evaluate the potential P loss from agricultural fields to waterbodies via four components. Our overall goal was to evaluate the potential of using spatial data to estimate P loss without physically visiting fields since many PLAT required parameters occur in spatial formats. The objective of the first study was to assess the possibility of spatial implementation of PLAT and to compare the effect of scale on the PLAT numerical results and the associated categorical rankings. Since an important input parameter, the average annual soil loss determined by the Revised Universal Soil Loss Equation, is not directly available from field measurement, our objective in the second study was to assess the potential of obtaining RUSLE estimates, specifically the topography factor LS, through Digital Elevation Model data in a Geographic Information System environment. In the first study, two methods of whole field average (WFA) and grid average (GA) were used to compare the difference in modeling P loss at different scales. The same list of PLAT required parameters were prepared from soil test reports and spatial database at the coarse scale of whole agriculture field and the fine scale of 0.4-ha grid. Soil tolerance value was used to temporarily replace the soil loss data. In the second study, a widely used Arc Macro Language (AML) program for estimating RUSLE topographic factor LS was evaluated through two approaches of whole field (WF) and representative profile (RP) analysis on a North Carolina landscape. Watershed delineation technique was adopted to select the representative profiles based on the references of slope distributions and field subdivisions from NRCS water quality specialists. Results from the first study indicated that soluble and particulate P loss, which occupied 59.3% and 26.3% of the total P loss through WFA method, and 56.1% and 39.0% through GA method, were the major pathways. Leaching P loss from PLAT was negligible. Particulate P loss was sensitive to scale as verified by the 12.7% increase of proportion in total P loss. The difference of particulate P loss through two methods was significant (p < 0.05), but no difference of soluble P loss and P source effect was found on a 95% confidence level. The overall P loss potential through two methods exhibited no significant difference due to the neutralization effect of individual pathways. Results from the second study showed that the AML program alone was not suitable for calculating RUSLE topographic factor on a North Carolina landscape because of the significant underestimation (˜35% and ˜20% through WF and RP approach, respectively). The concept of representative profile indeed improved the estimation accuracy (˜15%), however, the linearity of the fitted line between field measured LS and GIS-aided LS estimate was not satisfactory. An adjustment factor was proposed rectifying the RUSLE-based AML program in order to approximate field measurements. This study demonstrated the potential of implementing PLAT model and the soil loss equation using spatial parameters derived from database instead of visiting the fields. The scale of modeling in estimating particulate P loss and RUSLE topographic factor LS was important and the adjustment factor was necessary to adapt the AML program application. The accuracy of model performance needed to be improved before claiming that GIS-aided PLAT modeling will provide a complete replacement for the field measurement.
【 预 览 】
附件列表
Files
Size
Format
View
Evaluation of the Phosporous Loss Assessment Tool (PLAT) and Revised Universal Soil Loss Equation (RUSLE) using Geospatial Information
PDF
download
878987797
028-85220240
