Wind stress effects on drone-based thermal infrared surface velocimetry measurements of tidal flow in an estuary

We evaluate the effect of surface wind stress on remote velocimetry measurements of tidal flow by comparing these measurements to the bulk flow velocity measured by a co-located acoustic velocity profiler in a tidal channel. The remote velocity measurements are made with a thermal imager mounted on a drone hovering directly over the acoustic measurement location. Drones are a useful platform to support a variety of cameras and sensors for capturing images that can be used to infer surface velocities. Drone-mounted thermal infrared microbolometer cameras are a lower-cost infrared imaging solution that can detect subtle temperature patterns which naturally occur at the surface of many flows. These thermal patterns are used as signals for pattern-tracking to produce velocity measurements across the observed water surface. Drone flights were conducted at Carpinteria Salt Marsh Reserve (California, USA). Wind speed and direction relative to the flow direction caused the drone-based surface velocimetry measurements to deviate from in-channel surface-extrapolated acoustic velocity measurements. Drone-based velocity measurements were slower than in-channel velocity measurements when the parallel wind stress direction was opposite the tidal flow, while drone-based velocity measurements were faster than in-channel velocity measurements when the parallel wind stress and tidal flow were in the same direction. The effect of wind stress on remote surface velocimetry measurements is relatively unstudied, and herein we quantify this effect by comparing image-derived estimates to in-channel velocity measurements. This experiment also demonstrates the feasibility of drone-based thermal surface velocimetry measurements in an estuary.