Impacts of Land-Use Patterns on the Interannual Variability of Ecosystem Carbon Sinks: A Case Study of Hubei Province, China

Abstract

Land-use changes are widely recognized as important drivers of terrestrial carbon sinks. However, how changes in land-use patterns affect the interannual variability of net ecosystem productivity (NEP) remains unclear. This study analyzed the impacts of land-use patterns on NEP in Hubei Province from 2001 to 2025, utilizing MODIS NPP data, the China Land Cover Dataset (CLCD), and meteorological observations. Linear and nonlinear trends of NEP were examined using the Mann–Kendall (M-K) trend test and Ensemble Empirical Mode Decomposition (EEMD), respectively. Four landscape metrics, the contagion index (CONTAG), largest patch index (LPI), patch density (PD), and Patch Richness (PR), were calculated to characterize land-use pattern changes, and Pearson correlation analysis was employed to explore their interannual relationships with NEP. The results showed that the linear increasing trend of NEP masked widespread nonlinear shifts: while the M-K trend test indicated significant increases in 40.6% of the area, EEMD revealed that only 11.91% exhibited true monotonic increases, whereas 44.8% presented an increasing-to-decreasing trend. These shifts were predominantly concentrated in the Jianghan Plain, Northeast Hubei, and eastern Northwest Hubei, where cropland-to-impervious conversion was relatively intense and landscape metrics exhibited relatively pronounced nonlinear changes. LPI and CONTAG were significantly negatively correlated with NEP in these areas, which was consistent with the fragmentation of contiguous landscapes by urban expansion. In contrast, PD showed contrasting correlation directions. It was positively correlated with NEP in the plain areas, but negatively correlated in the mountainous areas, suggesting divergent associations of landscape fragmentation under different landscape matrices. PR exhibited the weakest correlations with NEP overall. This study highlights that nonlinear trend analysis is essential for accurately assessing carbon sink dynamics and that landscape patterns exert spatially heterogeneous effects on NEP, providing a scientific basis for differentiated land-use management strategies for carbon sequestration in Hubei Province.