Monitoring the colonization of alluvial deposits using multitemporal UAV RGB‐imagery

Abstract

AbstractAims(a) Evaluate the potential of Unmanned Aerial Vehicle (UAV) technology for very high resolution monitoring vegetation dynamics. (b) Describe and explain the colonization pattern of dewatered alluvial deposits by vegetation during early successional stages at an intra‐annual scale.LocationSélune River, Normandy, France.MethodsWe assessed vegetation cover using models based on UAV imagery and field samples at very fine spatial (cm) and temporal (intra‐annual) scales. A UAV flight was conducted every two months from January to July 2015, and vegetation was measured during image acquisition phases. Vegetation cover was mapped for each image acquisition date (four UAV orthomosaics) using a nonlinear regression model (Support Vector Regression algorithm). Then, the maps of vegetation cover were compared to evaluate the colonization process.ResultsVegetation cover was predicted from UAV with high accuracy (mean correlation coefficient: 0.90). Analysis of the maps revealed that colonization of the alluvial deposits by vegetation was rapid in spring.ConclusionsThis study shows that intra‐annual vegetation dynamics on alluvial deposits is rapid and that the colonization pattern can be observed in early successional stages. Very high spatial resolution images acquired by UAV can be used to create detailed maps to evaluate vegetation cover development.