Integrating lateral inflows into SWOT mission river discharge algorithm

Abstract

Estimating river discharge from observed surface water extents and elevations along rivers wider than 50-100 meters is central to the NASA and CNES Surface Water and Ocean Topography (SWOT) mission. Although near-global in its coverage, SWOT will not observe tributaries smaller than 50-100 meters in width draining laterally into observable river reaches. This is problematic for discharge algorithms that use the Mass Conserved Flow Law Inversion (McFLI) approach, which relies on mass conservation across neighboring reaches to compute discharge. However, unobserved tributaries and neglected overland flow cause violations of mass conservation. We quantify the effect of lateral inflows on the performance of estimated discharges along the mainstem of the Muskingum River using MetroMan, a McFLI discharge algorithm, for the period July 1-20, 2013, selected for evaluation because it captures the rise and fall of a large storm event. Three scenarios are considered: (1) disregarding lateral inflows, (2) providing MetroMan with true lateral inflows based on measured USGS streamflow, and (3) providing MetroMan with lateral inflows derived from NASA's North American Land Dada Assimilation System (NLDAS) and the Hillslope River Routing (HRR) model. The scenarios were further expanded to account for a range of lateral inflow magnitudes and variations in the distribution of lateral loading. Results show that when observed lateral inflows are included in MetroMan, the derived discharges have a relative root-mean-square error (rRMSE) of 23% vs. 360% when lateral inflows are neglected. In general, discharge retrievals were considerably degraded once unaccounted lateral inflows exceeded 9% of the average discharge in the domain. More importantly, when MetroMan uses simulated lateral inflows derived from USGS gauges, which at peak flow conditions have percent errors as high as 93%, discharge retrieval performance is maintained (rRMSE = 17%). This point is especially encouraging for potential SWOT data users.