Spatio-temporal characteristics of the trade-off and synergy relationships among multiple ecosystem services in the Upper Reaches of Hanjiang River Basin
The research on the interactions among multiple ecosystem services (ES) is a hotspot. Most of the previous studies focused on the qualitative description of ES interactions, however, there have been relatively few studies on spatially explicit and quantitative assessment of ES interactions. In this paper, we mapped the ecosystem service of soil conservation (SC), net primary production (NPP) and water yield (WY) in the upper reaches of Hanjiang River Basin (URHR) based on the land use and land cover (LULC), NDVI, soil properties and observed climate data covering 2000- 2013. Moreover, we quantitatively assessed the variation characteristics of interactions among different ES with a spatio-temporal statistical framework by applying the partial correlation analysis at a pixel scale. The results are shown as follows: (1) From 2000 to 2013, the mean annual SC was 434.20 t·hm~(-2)·yr~(-1), and the mean annual WY was 250.34 mm. They also presented a rising tendency at the rate of 16.10 t· hm~(-2)·yr~(-1) and 3.79 mm·yr~(-1), respectively. However, the mean annual NPP was 854.11 gC·m~(-2)· yr~(-1), and presented a decreasing tendency at the rate of 8.54 gC·m~(-2)·yr~(-1). (2) Spatially, SC was high in the North-South mountain area, while it was low in the Middle valley region. Similarly, the NPP in the Middle valley region was lower than that of other regions. However, the WY increased from north to south. (3) The three pairwise ES presented different interactions. Both the interaction between SC and NPP and that between SC and WY presented as trade- off, accounting for 62.77% and 71.60% of the total area, respectively. On the contrary, the interaction between NPP and WY was prone to synergies, accounting for 62.89% of the total area. (4) Pairwise ES in different land cover types also presented a different interaction. As for woodland, wetland, cropland, artificial land and bare land, SC and NPP, as well as SC and WY both presented trade-off, while WY and NPP presented synergy. Specially, in grassland, all the three pairwise ES presented a trade- off relationship. Therefore, spatially explicit and quantitative assessment of ES interactions are more helpful for revealing the temporal nonlinear evolution, and the spatial heterogeneity of ES interactions. This analysis framework also contributes to the regional sustainable land management and the optimization of multiple ES conservation.