Site Selection for Satellite-Based Stream Gaging in Alaska

Satellite-based stream gaging has valuable potential in Alaska (AK), where the current stream monitoring network is sparse, many locations that should be monitored are remote and hazardous, and the historic record of streamflow in many gaged rivers is relatively short (less than 20 years). The USGS, in cooperation with other federal and state agencies, is piloting operational satellite gaging sites at existing USGS stream gaging stations in AK so that ground truth data are available for methods development and accuracy assessment. The goal of the effort is to produce near-real-time stream gaging information and to extend historic gaging data based on Landsat and other satellite data.

Selection of sites for satellite gaging is based on multiple competing criteria related to the parameters being measured: stream surface altitude and area. Stream surface velocity is being explored as an observable parameter, but satellite imagery required for velocity determination is not yet readily available. Wider rivers enable more accurate estimates of stage and discharge, because satellite altimetry data represent an areal average over the footprint of the satellite signal on the water surface; minimum river width in the pilot is set at 100 m. Tidally influenced rivers are excluded from consideration because stage and area are not good indicators of discharge rates at tidal sites. Satellites used for altimetry, ICESat-2 and Jason-3, set an important limit to locations for satellite gaging because gaging is only possible at locations where satellite ground tracks cross a river of sufficient width. ICESat-2 and Jason-3 have footprints of approximately 17 m and 200 m, respectively. Satellite discharge estimates are more accurate if a few ground measurements of river discharge are available to remove, so river sites accessible by a ground crew are desirable. A further consideration is the morphology of the river and the nature of the stage-discharge or width-discharge relations. Physically constrained river reaches, which have large changes in stage at greater discharge rates, are better suited to use of stage-discharge relationships, whereas less constrained river reaches, which have large changes in width at greater discharge rates, are better suited to use of width-discharge relationships.