Investigating approaches for the assimilation of future SWOT observations into uncertain global hydrological models

Abstract:

Global estimates of river dynamics are necessary to manage water resources, mainly in developing countries where in-situ observation is limited. Remote sensors such as nadir altimeters can complement ground data. However current altimeters miss a large number of continental surface water bodies, what is expected to be surpassed by the future Surface Water and Ocean Topography (SWOT) mission. SWOT would be able to observe the seasonality of large portion of rivers and lakes globally and to provide two-dimensional maps of water elevation for rivers with width greater than 100 m. In this research we describe a modelling framework that prescribe optimal releases for large scale basin. Hence, we can depict how SWOT data might be used to the benefit global hydrological modelling. To perform the modelling framework the Observing System Simulation Experiment (OSSE) so-called "twin experiment" was implemented. The forcing and parameters of the model setup were perturbed to jointly achieve the uncertainties of global hydrological models (GHMs). This scenario is expected to be where the SWOT community will mainly assess the future SWOT data. The SWOT-like observations of discharge (Q), water surface elevation (WSE) and flooded water extent (FWE) were used to recover the “truth” model through the ensemble Kalman filter (EnKF) scheme in a large scale hydrologic and hydrodynamic model (MGB). The results indicate that considering the expected errors of the SWOT observations, DA of these variables has a large impact in the improvements of hydrologic and hydrodynamic simulations in global and continental scales what implies a promissory usefulness by SWOT scientific community. SWOT-like discharge can be able to improve simulations by using hydrological models in global scale in approximately ~40% for the reduction of errors of daily discharge. Moreover, when multiple-variable DA or anomalies of WSE DA approaches were implemented, the results were even more promising. Considering these results, the approaches developed in this research can frame the possibilities of SWOT for the global hydrological community. Finally, attendant proxies such as using multiple ongoing remote sensing products can be currently implemented in GHMs with DA techniques.