The impact of land-use and land-cover change on carbon sink: implications from the case of Guangdong Province, China.

Abstract

Carbon sink (CS) is crucial for achieving sustainable development and is closely related to land-use and land-cover change (LUCC). However, the understanding of how LUCC affects CS is challenging due to the complex interactions with other driving factors such as climate change. Guangdong Province, one of China's most urbanized and frequently human-affected areas, serves as a valuable case for studying these mechanisms to advance carbon neutrality goals. This study proposes an innovative land-cover transfer carbon sink matrix method to isolate the contribution of LUCC to CS while considering the lag effect of vegetation succession. Methodologically, the study treats the CS changes in areas where land-cover types remain stable as the combined effects of other driving factors, while the CS changes following LUCC and subsequent completion of vegetation succession are considered the influence of time lag effects. This approach serves as a reference for isolating the contribution of each type of LUCC to CS. By integrating an improved CASA model with a heterotrophic respiration empirical formula, this research systematically reveals the spatiotemporal evolution patterns of CS in Guangdong Province from 2012 to 2021. The results are as follows: (1) The average CS gradually increases and exhibits significant spatial differentiation, showing a gradient pattern as eastern rural areas > northern areas > western areas > eastern urbanized areas > central areas with rapid urbanization in central urban areas leading to a decline in CS; (2) The innovative land-cover carbon sink transfer matrix method reveals that LUCC contributed 22.61% (5.06 MtC) to the total increase in total CS through land type conversion, with ecological engineering driving a contribution rate of 83.96%; (3) The study quantifies the non-linear response relationship between land type conversion and changes in CSs, finding that the actual CS increment is 95.10 gC·m^-2 lower than the theoretical value, resulting in a relative error of 37.11%. The land-cover transfer carbon sink matrix proposed in this study provides methodological support and a decision-making basis for land optimization management and ecological compensation policy formulation in highly urbanized areas.