Remote Sensing of Riverbank Migration Using Particle Image Velocimetry
Austin J. Chadwick, Evan Greenberg, Vamsi Ganti
Abstract
Abstract Mobile river channels endanger human life and property and over centuries shape ecosystems, landscapes, and stratigraphy. Quantifying channel movements from remote sensing is difficult, in part due to the diversity of river mobility processes (e.g., channel migration, cutoffs, avulsion) and planform morphologies (e.g., meandering, braided). Here, we present a framework for quantifying riverbank migration from remote sensing that upscales recent methodological advances from laboratory flume studies utilizing particle image velocimetry (PIV). We apply PIV to image time series of 21 rivers worldwide, showing PIV ignores cutoff and avulsion processes by design and is well suited for tracking riverbank migration regardless of planform morphology. We show that PIV‐derived results for riverbank migration are consistent with published results from centerline‐ and bank‐based Lagrangian methods. Unlike existing methods, PIV offers a grid‐based Eulerian framework where defining channel centerlines is unnecessary and quantified uncertainty in riverbank positions is propagated into uncertainty in migration rates. PIV offers means to efficiently extract global patterns in riverbank migration from decades of satellite data, as well as investigate river response to climate change and human activities in our rapidly changing world.