Monitoring coastal water flow dynamics using sub-daily high-resolution SkySat satellite and UAV-based imagery

Kasper Johansen; Aislinn F Dunne; Yu-Hsuan Tu; Burton H Jones; Matthew F McCabe

doi:10.1016/j.watres.2022.118531

Monitoring coastal water flow dynamics using sub-daily high-resolution SkySat satellite and UAV-based imagery

Water Res. 2022 Jul 1:219:118531. doi: 10.1016/j.watres.2022.118531. Epub 2022 May 2.

Authors

Kasper Johansen¹, Aislinn F Dunne², Yu-Hsuan Tu³, Burton H Jones², Matthew F McCabe³

Affiliations

¹ Hydrology, Agricultural and Land Observation, Water Desalination and Reuse Center, Biological and Environmental Science and Engineering Division, King Abdullah University of Science and Technology, Thuwal, 23955-6900, Saudi Arabia. Electronic address: kasper.johansen@kaust.edu.sa.
² Reef Ecology Lab, Red Sea Research Center, Biological and Environmental Science and Engineering Division, King Abdullah University of Science and Technology, Thuwal, 23955-6900, Saudi Arabia.
³ Hydrology, Agricultural and Land Observation, Water Desalination and Reuse Center, Biological and Environmental Science and Engineering Division, King Abdullah University of Science and Technology, Thuwal, 23955-6900, Saudi Arabia.

PMID: 35526428
DOI: 10.1016/j.watres.2022.118531

Abstract

Sub-daily tracking of dynamic features and events using high spatial resolution satellite imagery has only recently become possible, with advanced observational capabilities now available through tasking of satellite constellations. Here, we provide a first of its kind demonstration of using sub-daily 0.50 m resolution SkySat imagery to track coastal water flows, combining these data with object-based detection and a machine-learning approach to map the extent and concentration of two dye plumes. Coincident high-frequency unmanned aerial vehicle (UAV) imagery was also employed for quantitative modeling of dye concentration and evaluation of the sub-daily satellite-based dye tracking. Our results show that sub-daily SkySat imagery can track dye plume extent with low omission (8.73-16.05%) and commission errors (0.32-2.77%) and model dye concentration (coefficient of determination = 0.73; root mean square error = 28.68 ppb) with the assistance of high-frequency UAV data. The results also demonstrate the capabilities of using UAV imagery for scaling between field data and satellite imagery for tracking coastal water flow dynamics. This research has implications for monitoring of water flows and nutrient or pollution exchange, and it also demonstrates the capabilities of higher temporal resolution satellite data for delivering further insights into dynamic processes of coastal systems.

Keywords: Coastal water flows; Dye plume; Machine learning; SkySat; Sub-daily; Unmanned aerial vehicle (UAV).