Nonlinear Effects of Landscape Patterns on Ecosystem Services at Multiple Scales Based on Gradient Boosting Decision Tree Models

Exploring the complex effects of landscape patterns on ecosystem services (ESs) has become increasingly important in offering scientific support for effective spatial planning and management. However, there is a particular lack research nonlinear ESs scale dependence. Taking Huainan City (in China) as case study, this study adopted InVEST model to estimate four key including carbon storage (CS), habitat quality (HQ), nitrogen export (NE), water yield (WY). Then, we calculated selected metrics at multiple scales. Furthermore, gradient boosting decision tree (GBDT) was developed investigate relative importance explaining their interrelation. The results indicated that most were significantly correlated with ESs. GBDT model, which can explore relationships, performed better than linear regression variations landscape-level Shannon’s diversity index (SHDI) contagion (CONTAG) class-level aggregation (AI) edge density (ED) variables influenced affected within certain range, varied scale.