Spatiotemporal evolution of the leaf area index and hydrothermal driving mechanisms in Gansu Province

Analysis of the spatiotemporal evolution of the Leaf Area Index (LAI) and its hydrothermal driving mechanisms is essential to provide a scientific basis for ecological barrier construction in arid and semi-arid regions. Based on multi-source remote sensing data from 1981 to 2018, we systematically analyzed the spatiotemporal dynamics of LAI and underlying hydrothermal mechanisms across different ecological zones in Gansu Province using Sen+MK trend analysis, coefficient of variation, Hurst exponent as well as sensitivity and contribution analyses. The results show that areas with low multi-year mean LAI (<0.8) accounted for 66.47% of the provincial vegetation area and were mainly concentrated in the central and eastern regions, with the highest mean value (4.19) being observed for the deciduous forests of the Qinling Mountains. Overall, 89.88% of Gansu province exhibited an increasing trend in LAI throughout the period examined, with 67.16% showing a significant increase characterized by an annual growth rate of 0.004. The most significant increase in LAI was observed in the Ordos Plateau grassland zone. Areas showing medium-to-high fluctuations in LAI covered 75.81% of the total, with the strongest fluctuation observed in the Qaidam Basin semidesert zone. In terms of persistence, 88.67% of the Gansu area showed potential for sustainable growth, with the Ordos grassland zone having the highest proportion (98.49%). Specific humidity was identified as the dominant factor driving LAI, with areas where its contribution rate exceeded 60% covering 80.77% of the vegetation area. Sensitivity to evapotranspiration and precipitation was also pronounced, with high-sensitivity areas mainly distributed in the southern part of Gansu. This study provides quantitative support for evaluating the potential for vegetation restoration and implementing differentiated management strategies in arid regions in northwestern China.