Watersheds provide multiple ecosystem services. Ecosystem service assessment is a promising approach to investigate human-environment interaction at the watershed scale. The spatial characteristics of ecosystem services are closely related to land use statuses in human-dominated watersheds. This study aims to investigate the effects of land use on the spatial variations of ecosystem services at the Dianchi Lake watershed in Southwest China. We investigated the spatial variations of six ecosystem services-food supply, net primary productivity (NPP), habitat quality, evapotranspiration, water yield, and nitrogen retention. These services were selected based on their significance at the Dianchi Lake watershed and the availability of their data. The quantification of these services was based on modeling, value transference, and spatial analysis in combination with biophysical and socioeconomic data. Furthermore, we calculated the values of ecosystem services provided by different land use types and quantified the correlations between ecosystem service values and land use area proportions. The results show considerable spatial variations in the six ecosystem services associated with land use influences in the Dianchi Lake watershed. The cropland and forest land use types had predominantly positive influences on food productivity and NPP, respectively. The rural residential area and forest land use types reduced and enhanced habitat quality, respectively; these influences were identical to those of evapotranspiration. Urban area and rural residential area exerted significantly positive influences on water yield. In contrast, water yield was negatively correlated with forest area proportion. Finally, cropland and forest had significantly positive and negative influences, respectively, on nitrogen retention. Our study emphasizes the importance of consideration of the influences from land use composition and distribution on ecosystem services for managing the ecosystems of human-dominated watersheds.
Keywords: Human-environment interaction; InVEST; Maps; Modeling; Quantification; Spatial variations.