Predicting Transport of Soil Phosphorus in Landscape in Response to Manure Application.
Yongsheng Feng, Univ of Alberta, Dept of Renewable Resources, Edmonton, AB T6G2E3, Canada and Xiaomei Li, Alberta Research Council, 250 Karl Clark Road, Edmonton, AB T6N 1E4, Canada.
The Soil P-Export Model (SPEM) was developed as a tool to predict transport of soil P in agricultural landscape and its relation with manure application. There are a huge number of variables that affect the movement of P from soils to surface water. We will never have perfect knowledge of them for every single point in a watershed. Thus, even in a detailed distributed model where details of the spatial variability of the soil properties and processes are described, insufficient information of the necessary parameters will inevitably introduce uncertainties and errors. The approach used by SPEM attempts to strike a balance between representing the spatial variability of the soil processes and at the same time keeping the input requirements at a manageable level. The SPEM uses a probabilistic approach in dealing with spatial variability of the soil processes that control P export. This approach allows for the modelling of the spatial feature of the soil P processes, which cannot be easily dealt with lumped models. At the same time, this approach avoids the explicit modelling of detailed spatial distributions of the soil properties and processes, thus significantly reduce the amount of input requirement with minimum loss of information. The SPEM model was validated using the data generated from the field experimental plots. The prediction from the model also compared with a data set generated from over 58 of soils tested in lab using a rain simulator. The correlation between dissolved phosphate concentration in runoff (C, mg L-1) and soil available phosphate concentrate (P, mg kg-1) generated from this database was: C=0.0031P -0.047 (R2=0.93). The predicted relation was: C=0.0034 P. It also successfully predicted the water runoff from a small 25 km2 watershed.