Collin, A.; James, D.; Lesacher, M., and Feunteun, E., 2024. Ultra-high resolution temperature mapping of the muddy, sandy, and rocky coasts using UAV-scaled SuperDove imagery. In: Phillips, M.R.; Al-Naemi, S., and Duarte, C.M. (eds.), Coastlines under Global Change: Proceedings from the International Coastal Symposium (ICS) 2024 (Doha, Qatar). Journal of Coastal Research, Special Issue No. 113, pp. 529-533. Charlotte (North Carolina), ISSN 0749-0208.

Worldwide coastal interfaces undergo anthropogenic pressures at an unprecedented rate, shaping their geomorphological and ecological responses in their finer scale. The adequate monitoring of their structures and dynamics therefore necessitates very high spatial, temporal and spectral resolutions. The study sheds light on the fusion of the 1-day multispectral spaceborne SuperDove imagery with standard unmanned airborne vehicle (UAV) imagery to (1) produce an innovative bottom-of-atmosphere surface reflectance imagery at 0.04 m pixel size provided with 8 optical bands (visible and near-infrared), then (2) map the temperature over a coastal system endowed with muddy, sandy and rocky environments, distributed in 12 habitats. The UAV-scaled SuperDove dataset was successfully modelled by linear regressions based on the UAV blue-green-red and digital surface model (DSM) data. The temperature was better assessed by the UAV-scaled SuperDove imagery joint with the UAV-based DSM (R²_test=0.67, RMSE=0.99°C), than the standard UAV blue-green-red imagery and the DSM dataset (R²_test=0.58, RMSE=1.11°C). Both models consisted of a 2-layer neural network (perceptron) featured with 5 neurons and 3 neurons. The rocky environment revealed colder than the muddy and sandy environments, which was discussed in the context of climate change.