Characteristics and controls of vegetation and diversity changes monitored with an unmanned aerial vehicle (UAV) in the foreland of the Urumqi Glacier No. 1, Tianshan, China

Exposed surfaces following glacial retreat are ideal field laboratories for studying primary vegetation succession. Many related studies based on ground sampling methods have been performed worldwide in proglacial zones, but studies on species diversity and vegetation succession using aerial photography have been rare. In this study, we investigated soil organic carbon (SOC), total nitrogen (TN), plant species diversity, and fractional vegetation cover (FVC) along a chronosequence within the foreland of Urumqi Glacier No. 1 by combining field sampling and aerial photography. We then analysed soil development and vegetation succession along distance (distance from glacier terminus) and time (terrain age) gradients as well as the relationships between topographic and environmental variables (aspect, slope, SOC, and TN), distance, time, and species distributions. The results indicated that: (1) plant diversity and FVC showed increasing trends with increases in distance and terrain age, whereas soil nutrient content varied nonlinearly; (2) Silene gonosperma, Leontopodium leontopodioides, and Saussurea gnaphalodes were the dominant species in the early, transient, and later succession stages, respectively. Cancrinia chrysocephala occurred in all stages and had a high abundance in the early and later stages; and (3) the relationships of FVC with soil nutrient content were nonlinear. Moreover, distance and site age played important roles in species distribution. These findings confirm that distance and terrain age positively affect vegetation succession. The increase in FVC facilitated the accumulation of soil nutrition, but this trend was affected by the rapid growth of plants. Caryophyllaceae and Asteraceae were the most common plants during the succession stages, and the former tended to colonise in the early succession stage. We conclude that the UAV-based method exhibits a high application potential for assessing vegetation dynamics in glacier forelands, which has a significance for long-term and repeated monitoring on the process of vegetation colonisation and succession in deglaciated areas.