| Summary: | Understanding the spatial associations of ecosystem services (ESs) driven by socio-ecological factors is essential for regional ecosystem management and effective resource allocation. Yet previous studies of mechanisms driving ESs have focused mainly on the supply or demand for ESs, rarely consider their threshold effects and supply–demand ratios. Taking the upper reach of Ganjiang watershed (URGW) in southern China as an example, the supply, demand, and supply–demand ratio of ESs (ESDRs) were quantified for grain production (GP), water yield (WY), soil retention (SR), carbon sequestration (CS), and green space recreation (GR). Correlation analysis and random forest models were used to investigate the spatial associations of five ESDRs and the relative importance and marginal effects of 9 socio-ecological factors upon them. At the watershed scale, only SR incurred a supply–demand deficit (ESDR < 0), while the other four ESs (GP, WY, CS, and GR) met the internal demand (ESDR > 0). When considered across 188 sub-watersheds, all ESs had different supply–demand deficits. The trade-offs in ESDRs mainly derived from two ESs (GP, SR) and the other three ESs (WY, CS, GR). The supply–demand ratio of SR was mainly determined by topography (i.e., slope gradient), while land use (i.e., proportions of cropland and forest land areas) played a decisive role in the supply–demand ratios of GP and WY. The supply–demand ratios of GR and CS showed similar spatial distribution, driven mainly by the normalized difference vegetation index and human population density. Specific socio-ecological factors (e.g., slope gradient and proportion of forest land area) had significant threshold effects on the ESDRs. Accordingly, research on the mechanisms driving ESs should consider ESs’ supply and demand in tandem, and pay particular attention to the potential threshold effects of socio-ecological factors.
|
|---|