Utility of drone-based thermal imaging for mapping river temperature heterogeneity

Climate change will alter river temperature regimes, modifying the thermal suitability of rivers for temperature-sensitive fishes. As a result, there is an increasing need to map and monitor river temperature patterns in order to understand how fluvial ecosystems may respond to future climate change. Researchers have generally accomplished this task using temperature loggers (which provide temporally-continuous data, but with a limited spatial resolution) or by means of thermal infrared (TIR) imaging surveys (which generate spatially-explicit temperature data but only for a single snapshot in time). Recent advances in small unoccupied airborne systems (sUAS) and miniaturised TIR cameras present an opportunity to redress this space-time disconnect, enabling the improved understanding of river temperature patterns through the provision of spatially-continuous, multi-temporal data. However, information regarding the utility of drone-based TIR for surveying rivers is limited. Here, we present the results of a series of drone-based TIR surveys conducted with a view to understanding their suitability for characterising river temperature heterogeneity. Our findings reveal that drone-based TIR is able to clearly reveal the location and extent of discrete thermal inputs to rivers (eg. cool water refuges). However, the TIR imagery also suffers from substantial drift-induced bias rendering it unsuitable for the characterisation of larger-scale river temperature patterns. We conduct a statistical analysis of the causes of this drift, and find that metrics related to drone flight characteristics and environmental conditions (eg. prevailing meteorology) explain 66% of the variance in TIR-derived temperature drift. Our findings shed important light on the advantages and limitations of drone-based TIR for mapping river temperature patterns, and provide useful information for flight planning. We also discuss potential future developments to enable the extraction of robust temperature data from drone-based TIR solutions.