The distance decay of similarity in climate variation and vegetation dynamics

The negative relationship between similarity and distance has been revealed in many subjects in geography and ecology fields. This study aimed to illustrate the strength of the distance-decay relationship in variation of climate and vegetation, and to quantify the relationship. Solving this problem could help to test some model specifications based on the climate and vegetation time series on sample sites, to determine the distance function in the spatial interpolation technique for meteorological factors and vegetation dynamics, and to use the distance-decay perspective as a quantitative technique to adapt strategies for future climate change and vegetation dynamics. To achieve the study goal, we quantified variation similarity using mutual information (MI), which measured the dependence between two variables or time series. We carried out a distance-decay analysis of climate and NDVI variation similarities, assessed by the MI against the log-transformed geographical distances between meteorological stations. The results suggest that all station pairs shared some similarity in the processes of climate variation and vegetation dynamics, and the MI values showed a gradual decrease with the increase of distance. In addition, temperature, precipitation, and NDVI time series had different MI value ranges and distance-decay ratios due to various influential factors. The logarithmic distance-decay relationships are of potential usefulness to the study of community similarity and the neutral theory of biogeography. Our research provides an approach for analyzing spatial patterns in relation to dependence and synchronization that may inform future studies aiming to understand the distribution and spatial relationship of climate and vegetation changes.