The assessment of land use change (LUC) proposals aimed at improving environmental conditions requires information regarding the relationship between actions and outcomes. Yet such information is often limited in detail, especially in developing country contexts. This paper demonstrates an approach to the estimation of farm level biophysical relationships between proposed LUC activities and their impacts on soil erosion, air quality and biodiversity in the context of a payment for environmental services (PES) auction in Sichuan Province, China. Because data at the farm level are unavailable, the models are based on divergences of each farm's biophysical conditions from well-established regional averages. This paper focuses on the estimation of the farm-to-region divergence factors and the development of environmental outcome predictions that reflect the relative ecological significance of alternative LUC proposals. The strengths and weaknesses of the biophysical models are also discussed.