EEMD disentangles climate-NPP dynamics in Southwest China's karst area: Soil moisture overtakes precipitation as a key driver of afforestation-induced carbon sinks

Zhang, Wen , Li, Yudi , Jiao, Limin , Zheng, Hui , Fu, Bolin

2025-12-01 INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 2025   145(卷), null(期), (null页)

查看原文

The vegetation greening trend resulting from afforestation in the Southwest China's karst (SCK) area has attracted significant research attention. Identifying the interannual variation in net primary productivity (NPP), its sustained reaction to climate change, and differences between new and old forests before and after afforestation is vital in assessing the effect of the climate on the carbon sink potential of ecosystems. Using monthly NPP and climate data for 1981-2019, ensemble empirical mode decomposition was employed for the extraction of the interannual signal, and new and old forests were compared based on changepoint detection and forest-age data. According to the results, (1) NPP change significantly increased in 1981-2019; after 2001, the NPP growth rate in new forests was 1.47 times that in the old forests, indicating greater carbon sink potential. (2) NPP was sensitive to climate change on an interannual scale, and the sensitivity increased further after 2001. (3) Additionally, the main determinant of NPP in old and new forests shifted from precipitation to soil surface water content after 2001,which closely aligns with the implementation of large-scale ecological restoration projects. The findings reveal that afforestation reinforces water constraints and shifts the dominant climatic control, providing a basis for evaluation of ecological restoration effectiveness in the SCK area and supporting incorporation of water and carbon trade-offs into afforestation planning and routine forest management. This study also provides methodological and case-based support for conducting climate sensitivity assessments of afforestation initiatives across different countries and regions.