Remote sensing of tracer dye concentrations to support dispersion studies in river channels

Abstract

We evaluated the potential of remote sensing to enhance tracer experiments by providing spatially distributed information on visible dye concentration. During tests in an experimental facility and a large natural channel, we measured Rhodamine WT concentration and spectral reflectance. As an initial proof-of-concept at the River Experiment Center in Korea, a small unmanned aircraft system (sUAS) was used to acquire hyperspectral images of a sinuous outdoor flume. On the Kootenai River, field spectra were collected from a boat and hyperspectral images and high resolution aerial photographs were obtained from manned aircraft. We modified an Optimal Band Ratio Analysis algorithm to identify wavelength combinations that yielded strong correlations between a spectrally based quantity X and dye concentration C. Both the flume and field tests yielded very strong (R2 from 0.94 to 0.99) relationships between X and C across a broad range of visible wavelengths. On the Kootenai, we found that X vs. C relations derived from field spectra could be applied to hyperspectral images and that dye concentrations could be estimated nearly as reliably from three-band images as from hyperspectral data. These results imply that remote sensing could become a powerful tool for mapping dye patterns. Such a capability would advance our understanding of dispersion processes by enabling more rigorous testing of numerical flow models.