Land cover simulation and analysis for the Greater Bay Area of China in the context of the 2035 development plan

Abstract

This study focuses on the Guangdong-Hong Kong-Macao Greater Bay Area (GHM-GBA) for simulating Land Use Land Cover (LULC) considering the development plan for 2035. The research aims to quantify LULC change and simulate future LULC scenarios based on policy implications for 2035. In this study, Patch-Generating Land Use Simulation (PLUS) model was used to project future LULC under different scenarios. These scenarios include, Natural Increase Scenario (NIS), Ecological Conservation Scenario (ECS) and Urban Development Scenario (UDS). The China Land Cover Dataset (CLCD) from 2010–2022 along with eighteen driving factors for both historical and recent time periods and planning data were used to simulate the future succession of LULC patterns for 2035. This multifaceted methodology represents significant advancement over previous LULC simulation studies in GBA, which often relied on a more limited set of historical and development factors. The analysis revealed several key trends across the LULC categories. The simulation results (for 2022–2035) reveal that the cropland is expected to experience a modest increase of approximately 1.38%, indicating potential expansion of agricultural activities in future. However, the projections show declines in natural land covers, with forested areas decreasing by 2.44%, shrubland by 19.57%, grassland by 22.26%, water bodies by 9.15%, and barren land by 10.54%. Conversely, impervious surfaces are expected to increase by an average of 13.16%, suggesting urban development and infrastructure expansion. The findings provide valuable insights for regional environmental planning and sustainable development. The comparative analysis of the PLUS model’s performance across different policy scenarios can aid in improving LULC change projections. A significant contribution of this current study is the comparative analysis of the PLUS model’s performance under different policy-driven scenarios, which can aid in improving LULC simulations and projections.