Satellite-Observed Spatio-Temporal Variation in Spring Leaf Phenology of Subtropical Forests across the Nanling Mountains in Southern China over 1999–2019

Abstract

Knowledge of spatio-temporal variation in vegetation phenology is essential for understanding environmental change in mountainous regions. In recent decades, satellite remote sensing has contributed to the understanding of vegetation phenology across the globe. However, vegetation phenology in subtropical mountains remains poorly understood, despite their important ecosystem functions and services. Here, we aim to characterize the spatio-temporal pattern of the start of the growing season (SOS), a typical spring leaf phenological metric, in subtropical forests across the Nanling Mountains (108–116° E, 24–27° N) in southern China. SOS was estimated from time series of GEOV2 leaf area index (LAI) data at 1 km spatial resolution during the period 1999–2019. We observed a slightly earlier regional mean SOS in the southern of the region (24–25° N) than those in the central and northern regions. We also observed spatially varying elevation gradients of the SOS. The SOS showed a change slope of −0.2 days/year (p = 0.21) at the regional scale over 1999–2019. In addition, approximately 22% of the analyzed forested pixels experienced a significantly earlier SOS (p < 0.1). Partial correlation analysis revealed that preseason air temperature was the most responsible climate factor controlling interannual variation in SOS for this region. Furthermore, impacts of air temperature on the SOS vary with forest types, with mixed forests showing a stronger correlation between the SOS and air temperature in spring and weaker in winter than those of evergreen broadleaf forests and open forests. This suggests the complication of the role of air temperature in regulating spring leaf phenology in subtropical forests.