Comparison of the accuracy of two UAV‐mounted uncooled thermal infrared sensors in predicting river water temperature

Abstract

AbstractMeasurement by a unmanned aerial vehicles (UAV) equipped with thermal infrared sensors has great potential as a new technique for understanding the spatial heterogeneity of river water temperature; however, the measurement technique has not been established yet. Therefore, this study measured surface and water temperatures using two uncooled thermal infrared sensors, which are commercially available, for 7 months, to obtain knowledge to establish a method for predicting the river water temperature using uncooled thermal infrared sensors mounted on UAVs. The survey covered a 500 m section upstream from the confluence of the Inukai River in the Katsura River of the Yodo River system, in Japan. DJI Zenmuse XT2 and MicaSense Altum were mounted on a DJI Matrice 300 at the same time and took images seven times at intervals of approximately 1 month between June 2021 and January 2022. The water temperature and depth were measured at several locations within the study area within 1.5 hr after the UAV acquisition. Then, the relationship between the measured water temperature and the surface temperature extracted from the thermal infrared orthorectified mosaic image was analyzed. Altum was more suitable than XT2 for obtaining thermal infrared orthorectified mosaic images with less inconsistencies among courses in the surface temperature range, and the difference in water temperature was also smaller, making it suitable for understanding the river water temperature. To predict water temperature from surface temperature, actual measured data of water temperature is necessary, but I demonstrated that it is possible to measure the river water temperature in a spatially continuous manner.