ABSTRACT
Vegetation responses to climate change are typically nonlinear with varied time effects, yet current research lacks comprehensiveness and precise definitions, hindering a deeper understanding of the underlying mechanisms. This study focuses on the mountain-type Qilian Mountain National Park (QMNP), investigating the characteristics and patterns of these nonlinear time effects using a generalized additive model (GAM) based on MODIS-NDVI, growing season temperature, and precipitation data. The results show that 1) The time effects of climate change on vegetation exhibit significant spatial variations, differing across vegetation types and topographic conditions. Accounting for optimal time effects can increase the explanatory power of climate on vegetation change by 6.8 %. Precipitation responses are mainly characterized by time-lag and time-accumulation effects, notably in meadows and steppes, while temperature responses are largely cumulative, especially in steppes. The altitude and slope significantly influence the pattern of vegetation response to climate, particularly in areas with high altitudes and steep slopes. 2) There is a significant nonlinear relationship between vegetation growth and both precipitation and temperature, with the nonlinear relationship between precipitation and vegetation being stronger than that with temperature, particularly in the western and central regions of the park. Different vegetation types exhibit significant variations in their response to climate change, with deserts and steppes being more sensitive to precipitation. 3) Precipitation is the primary driver of vegetation change in the QMNP, particularly for high-elevation vegetation and herbaceous vegetation. The complex temporal patterns of vegetation response to climate change in the QMNP not only deepen the understanding of the intricate relationship between regional vegetation and climate variability but also provide a methodological reference for global studies on vegetation responses to climate change.
ABSTRACT
Using spatial autocorrelation methods, we explored the spatial and temporal differences in the response of soil conservation (SC) and water conservation (WC) to ecosystem fragmentation during 1990 to 2019 in Qilian Mountain National Park (QMNP) of China. We found that the degree of ecosystem fragmentation decreased over the past 30 years, improving the WC and SC capacity of the ecosystems. However, the relationship among them varied temporally and exhibited a variety of spatial patterns. The correlation between fragmentation and WC increases year by year, and the correlation with SC weakens. There is a mismatch between park-level and regional autocorrelation between fragmentation and WC and SC. The spatial relationships between fragmentation and WC and SC in the QMNP show "high-high" and "low-low" patterns in its eastern and western sections, respectively. This heterogeneity is related to the differences in ecosystem composition, especially in ecosystem WC and SC capacity, and the characteristics of ecosystem fragmentation in the east-west direction of the QMNP.
