How well will the <scp>S</scp>urface <scp>W</scp>ater and <scp>O</scp>cean <scp>T</scp>opography (<scp>SWOT</scp>) mission observe global reservoirs?

TLDR

Accurate observations of global reservoir storage are critical to understand managed water resources, and the NASA SWOT satellite mission, with a 21‑day repeat orbit, is expected to greatly improve monitoring for bodies larger than 250 m². The study aims to quantify SWOT’s spatial and temporal measurement uncertainty for reservoirs to enable optimal use of the data and inform future global monitoring applications. The authors assess SWOT performance through a three‑pronged analysis of temporal aliasing, spatial‑property‑induced errors, and real‑reservoir tests using in‑situ data and SWOTsim simulations. Temporal errors are below 5 % for reservoirs under 10 km² and under 0.1 % for those over 100 km², while surface‑area errors stay under 5 % and height errors under 15 cm for bodies larger than 1 km² except for highly elliptical, mountainous, or low‑coverage cases, and six real‑reservoir tests confirm these findings.

Abstract

Abstract Accurate observations of global reservoir storage are critical to understand the availability of managed water resources. By enabling estimates of surface water area and height for reservoir sizes exceeding 250 m 2 at a maximum repeat orbit of up to 21 days, the NASA Surface Water and Ocean Topography (SWOT) satellite mission (anticipated launch date 2020) is expected to greatly improve upon existing reservoir monitoring capabilities. It is thus essential that spatial and temporal measurement uncertainty for water bodies is known a priori to maximize the utility of SWOT observations as the data are acquired. In this study, we evaluate SWOT reservoir observations using a three‐pronged approach that assesses temporal aliasing, errors due to specific reservoir spatial properties, and SWOT performance over actual reservoirs using a combination of in situ and simulated reservoir observations from the SWOTsim instrument simulator. Results indicate temporal errors to be less than 5% for the smallest reservoir sizes (&lt; 10 km 2 ) with errors less than 0.1% for larger sizes (&gt;100 km 2 ). Surface area and height errors were found to be minimal (area &lt;5%, height &lt;15 cm) above 1 km 2 unless the reservoir exhibited a strong elliptical shape with high aspect ratio oriented parallel to orbit, was set in mountainous terrain, or swath coverage fell below 30%. Experiments from six real reservoir test cases generally support these results. By providing a comprehensive blueprint on the observability of reservoirs from SWOT, this study will be have important implications for future applications of SWOT reservoir measurements in global monitoring systems and models.