ABSTRACT
Ecosystem services (ESs) play a crucial connecting role between human well-being and natural ecosystems. Investigating ESs and their interrelationships can aid in the rational distribution of resources and benefits and inform planning decisions that align with the principles of ecological civilization. Nonetheless, our current understanding of these relationships remains limited; thus, further theoretical exploration is required. This study employs the InVEST model to assess the key ESs in Guangdong Province for 2000 and 2018 and applies the multi-scale geographically weighted regression (MGWR) method to identify the primary drivers of ES changes and capture trends in spatial variations. The results showed that (1) from 2000 to 2018, the total carbon storage (CS) and habitat quality (HQ) decreased while the water yield (WY) and net primary productivity (NPP) increased. These ESs also showed spatial differences, with higher values observed in the hilly and mountainous areas of the north compared with the coastal and plain areas of the south. (2) Although the spatial distribution of ES trade-off strength varied, the overall pattern remained consistent from 2000 to 2018. The pairwise trade-off strength of CS-WY and WY-HQ decreased significantly in the northern region of Guangdong due to low rainfall, while that of CS-HQ decreased significantly in the Pearl River delta as a result of urbanization. Cultivated and forested land displayed higher and lower levels of NPP and WY, respectively, with forested land exhibiting greater trade-off strength than the other land use types. (3) Evident spatial heterogeneity was observed in the properties and intensity of the correlations between driving factors and changes in ES trade-offs. Natural factors were the primary determinants of trade-offs among ESs. However, at a regional scale, the landscape index and socioeconomic factors tended to represent stronger drivers. Based on these findings, we suggest that ecological management should be adjusted based on the geographic scale. This study offers a valuable approach to understanding the relationship between ES trade-offs and their drivers in geographic space and serves as a reference for the sustainable provisioning of ESs both locally and globally.
