Using unmanned aerial vehicles to quantify spatial patterns of dominant vegetation along an elevation gradient in the typical Gobi region in Xinjiang, Northwest China

Abstract

Elevation gradient is a key environmental factor influencing the distribution of plants in the Gobi region. It can indirectly regulate the distribution of plants based on terrain and varying soil moisture, nutrient distribution, and particle composition. Undertaking ground vegetation surveys to observe vegetation distribution over large scales in the Gobi region is challenging due to the harsh environment and manpower constraints. In addition, due to the sparse vegetation distribution in the region, the precision of vegetation parameters extracted from satellite remote sensing images is low. Using the spatial-point pattern analysis method based on the Unmanned aerial vehicles (UAV) near-ground remote sensing data, this study analysed the spatial patterns of dominant plant species in the typical Gobi area of south of the Tianshan Mountains, Xinjiang in northwestern China along an altitudinal gradient in an alluvial fan. The results showed that dominant plants in the study area were distributed in a patchy pattern at the top of the diluvial fan and in a banded pattern in the middle and bottom. The richness of dominant species decreased with a decrease in altitude, and they were aggregated at small scales and randomly or uniformly distributed at large scales. In addition, degree of aggregation and uniformity in dominant species increased with a decrease in altitude. Interspecific relationships at the top of alluvial fan were weak with almost no correlation, and the correlations of dominant species were highly positive at small scales and highly negative at large scales, with a decrease in altitude. Both correlation analysis and grey correlation analysis indicated that the altitude and the soil moisture were the main factors influencing the spatial distribution of vegetation significantly, while the habitat heterogeneity was the main factor influencing the interspecific relationships. According to the results, UAVs can provide high-resolution data suitable for analysing sparsely distributed vegetation in the Gobi region. The results of the present study enhance our understanding of the factors influencing the distribution of plants in Gobi plant communities and provide a theoretical basis for the better conservation of the vegetation resources in the fragile ecological ecosystem.