Drivers of the dynamics in net primary productivity across ecological zones on the Mongolian Plateau

Understanding the drivers and mechanisms of the dynamics in grassland productivity is prerequisite for studying effective resource institutions and policies that can be used to govern grassland resources sustainably. We present a diagnostic analysis of the major drivers of the dynamics in grassland net primary productivity (NPP) across ecological zones on the Mongolian Plateau. We estimated a spatial panel data model for NPP (1986-2009) as a function of climatic and socioeconomic variables. Static and dynamic spatial panel models were estimated in each of the sub-regions, which were classified based on rural livelihoods and ecological models of grassland dynamics, to identify the major drivers of NPP dynamics. The statistical modeling results indicated that the major drivers of NPP dynamics vary across the six sub-regions. Grain output was the major predictor of NPP dynamics in the farming and farming-grazing zones of Inner Mongolia. Precipitation and livestock populations both had significantly positive relationships with NPP in the two grazing zones of Inner Mongolia. However, in Mongolia, livestock populations was the only significant predictor of NPP in the grazing zone with relatively stable climate, and precipitation was the only significant predictor of NPP in the grazing zone with highly variable climate. Human land-use activities and livestock management behaviors and the bidirectional causal relationships between livestock populations and NPP could explain the positive relationships between livestock population and grassland NPP. The heterogeneous drivers of NPP dynamics across space indicated the necessity of diverse resource polices and institutions for sustainable governance of grassland resources.