Looking to the Skies: Realising the Combined Potential of Drones and Thermal Infrared Imagery to Advance Hydrological Process Understanding in Headwaters

Abstract

In river systems, headwater networks contain the vast majority of the stream length. Thus, climate and land‐use change in headwaters have disproportionate impacts on downstream ecosystems and societies that rely on them. Despite decades of hydrological research, difficulties in observing hydrological properties across scales means that scientific knowledge of processes driving streamflow in headwaters remains limited. However, the recent emergence of two complementary technologies, drones and thermal infrared (TIR) remote sensing, has potential to collect data at scales and resolutions needed to advance hydrological process understanding in headwaters. In this commentary, we explain how drone‐based TIR can offer unique high‐resolution observations of surface connectivity and headwater network dynamics across multiple spatio‐temporal scales. We explore the current state‐of‐the‐art of drones and TIR imaging in the hydrological sciences, highlighting the potential benefits but also steps that will need to be taken to release these technologies' full potential. We finish by contending that drone‐based TIR is particularly well‐placed to bridge the current gap between field (point) observations and model simulations to provide the improved hydrological understanding needed for a changing world.

Ecohydrological interactions during drought

“…To achieve this, the ecohydrological community needs to leverage new measurement approaches, generate new data sets, and rely on novel modelling approaches (e.g., Wankmüller & Carminati, 2022) as well as taking advantage of longer term historical data to put ecohydrological drought in context (He et al, 2022). This includes—among others—relying on isotopic methods on specific ecohydrological contexts (e.g., Mattei et al, 2022), more widespread use of sensor technologies (Belmonte et al, 2022; Northup et al, 2022), novel remote sensing opportunities (Dugdale et al, 2022) and existing remotely sensed data (Mokhtar et al, 2022), and newly applied or developed statistical methods (Parolari & Paschalis, 2022), or machine learning/AI tools for processes that we currently do not incorporate well in physically‐based models of ecohydrological processes (Razavi et al, 2022).…”

Real‐Time Monitoring and Postprocessing of Thermal Infrared Video Images for Sampling and Mapping Groundwater Discharge

“…Although TIR cameras can only measure the water temperature at the surface, it is much less labor‐intensive for surveys covering a large spatial scale than direct observation using thermometers or fiber‐optic distributed temperature sensing (FO‐DTS) systems (Hare et al., 2015). For surveys of wide rivers, the spatial distributions of groundwater discharge zones and the stream temperature can be mapped by using a TIR camera mounted on an unmanned aerial vehicle (UAV), helicopter, or aircraft (Dugdale et al., 2022; Niwa, 2022; Rautio et al., 2015; Wilbur et al., 2020). Even for fifth‐order river sites, however, UAV‐based TIR sensing is less suitable for locating groundwater discharge points than handheld TIR imaging owing to their submeter scale and position beneath the riparian tree canopy (Briggs et al., 2022).…”

“…• Real-time thermal imaging helps find suitable sampling points of groundwater discharge differing in chemistry from stream water • Postprocessing of thermal video can easily map springs with thermal anomalies and the spatial distribution of the stream temperature • The water temperature of springs is useful for inferring their sources when multiple groundwater sources contribute to a stream and the stream temperature can be mapped by using a TIR camera mounted on an unmanned aerial vehicle (UAV), helicopter, or aircraft (Dugdale et al, 2022;Niwa, 2022;Rautio et al, 2015;Wilbur et al, 2020). Even for fifth-order river sites, however, UAV-based TIR sensing is less suitable for locating groundwater discharge points than handheld TIR imaging owing to their submeter scale and position beneath the riparian tree canopy (Briggs et al, 2022).…”

Harnessing hyperspectral imagery to map surface water presence and hyporheic flow properties of headwater stream networks

“…Recent methodological developments, such as the deployment of flow presence-absence sensors and drone surveys (e.g. Dugdale et al, 2022;Carbonneau et al, 2020;Zanetti et al, 2022), provide important constraints, but tend to be limited in space or time. Water presence can be detected in large open water (Wang et al, 2022) bodies or main stem river reaches with width greater than existing satellite imagery pixel resolutions (∼10-30 m pixel, Wang et al, 2022;Qin et al, 2021;Verma et al, 2021).…”