A partial least squares (PLS) regression analysis was employed to identify the key soil variables
in surface layer. The analysis indicated that the metal parameters (Fe, Mn, Cd, and Pb) were the important
factors for diffuse N loading. However, the TN in the paddy rice was the principal predictor. With the soil
property information, the diffuse N yield from the two types of farmlands can be assessed in a simplified
way. The developed method in this paper is
flexible and efficient, which also is compatible with results of
the SWAT model.