The use of remote sensing indices to understand flood‐recharged soil moisture impacts on trees in semi‐arid floodplains: A review

Abstract

Remote sensing indices (RSIs) are essential for detection, mapping and monitoring of floods, soil moisture and tree characteristics. An understanding of the nexus between flood-recharged soil moisture (FRSM) and endangered biodiversity tree hotspots in semi-arid floodplains is critical. Specifically, the influence of FRSM on biophysical characteristics of keystone tree species with social, economic, ecological and environmental functions is crucial. However, the use of RSIs to understand the link between episodic floods, FRSM and trees, as well as their respective spatiotemporal dynamics within the semi-arid floodplains, remains largely unexplored. Hence, this study reviewed literature on the adoption of RSIs in understanding short-term phenological and long-term tree structural changes arising from flood-related infiltration. The reviewed literature shows the predominance of the normalised difference water index, the land surface water index and normalised difference vegetation index for flood, FRSM and tree mapping. The review also shows the emergence of SAR-based normalised difference flood index for understanding the impacts of FRSM on riparian trees. Although the review identified an increase in data availability, however, there are high costs of active commercial remote sensing data of superior quality and computing resources. Furthermore, available free data are incapable of penetrating canopies and have coarse spatial resolution constraining the use of RSIs to understand the impacts of FRSM on trees. In view of increasing free SAR data and cloudbased computing resources, application of emerging deep learning algorithms offers prospects for the use of RSIs to understand the impacts of FRSM on riparian trees.