Combining UAV and high-resolution image-based particle image velocimetry to monitor flow in lakes and rivers

Abstract:

Unmanned aerial vehicles (UAV) are increasingly being applied by the geophysical community for large-scale surveys of environments that are challenging to reach (e.g. volcanoes, glaciers). Some of these natural systems include large bodies of water such as oceans, lakes and rivers. These environments are highly dynamic and, even in normal conditions, their characterization requires long surveys involving the deployment of expensive equipment and large crews. Technical challenges, and thus costs, become prohibitive especially during extreme natural events that may pose a threat not only to the monitoring facilities but, more importantly, to the surrounding residential areas (e.g. river flooding, extreme waves). Traditional measurement techniques have difficulty or are unable to monitor some flow quantities (i.e. flow discharge and wave energy content) associated with these natural systems, especially during emergencies. Extreme hydrodynamic loads due to high flow discharges and/or highly energetic waves may tamper with traditional monitoring systems that are typically submerged or fixed in the vicinity of the water body.

Here, we propose to employ specially developed UAV systems to remotely and safely gain high-resolution images of the water surface. Such images are processed using robust large-scale particle image velocimetry (LSPIV) algorithms that are able to accurately resolve the complex two-dimensional flow field of the water surface over extensive areas. These systems can theoretically be deployed in a cost-effective way for long periods of time for continuous monitoring. Such monitoring is key to inform and calibrate predictive tools that can reconstruct potential emergency scenarios. In addition, during extreme natural events, data can be collected with no risk to human operators. Here, we discuss the concept and technology employed to render these measurement systems effective, and provide examples of applications that show the depth of the data that can be extracted from aerial images and thus the potential of this technology.