Recent Advances in Global Measurement and Application of River Widths

Friday, 18 December 2015: 10:20
3011 (Moscone West)
Tamlin Pavelsky, University of North Carolina, Chapel Hill, NC, United States and George H Allen, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States
Among variables relevant to river form and discharge that can be observed from space, river width is perhaps the simplest to measure. Width can be extracted directly from optical or radar imagery, and application of remotely sensed widths to problems in hydrology, fluvial geomorphology, and ecology dates back more than two decades. Despite this long heritage, until very recently remotely sensed width measurements have largely been made on an ad-hoc basis for individual studies over relatively small regions. Global studies that required river widths have largely relied on estimates from downstream hydraulic geometry relationships with basin area, which inevitably simplify width variability and may, in practice, underestimate the fraction of wide rivers and the total river surface area in many basins.

Over the last two years, multiple new regional- and global-scale, satellite-derived river width datasets have been developed that have substantially improved our global understanding of river form. These datasets include the Global Width Database for Large Rivers (GWD-LR), which provides width measurements for rivers wider than ~180 m, and all rivers wider than ~300 m, based on the SRTM water mask and the Global River Widths from Landsat (GRWL), which provides measurements for rivers as narrow as 30 m and all rivers wider than ~100 m. Several regional-scale datasets have also been developed. These datasets will facilitate improvements to regional and global scale hydrodynamic models, will provide more robust information on global river surface area for gas flux studies, and constitute novel information on global patterns of fluvial geomorphology. These datasets represent the beginning, not the end, of global river width measurements, however, as in the future multitemporal width measurements can be combined with recently developed algorithms to estimate river discharge for many rivers, globally.