Interconnected hydrologic extreme drivers and impacts depicted by remote sensing data assimilation

Abstract

Abstract Hydrologic extremes often involve a complex interplay of several processes. For example, flood events can have a cascade of impacts, such as saturated soils and suppressed vegetation growth. Accurate representation of such interconnected processes while accounting for associated triggering factors and subsequent impacts of flood events is difficult to achieve with conceptual hydrological models alone. In this study, we use the 2019 flood in the Northern Mississippi and Missouri Basins, which caused a series of hydrologic disturbances, as an example of such a flood event. This event began with above-average precipitation combined with anomalously high snowmelt in spring 2019. This series of anomalies resulted in above normal soil moisture that prevented crops from being planted over much of the corn belt region. In the present study, we demonstrate that incorporating remote sensing information within a hydrologic modeling system adds substantial value in representing the processes that lead to the 2019 flood event and the resulting agricultural disturbances. This remote sensing data infusion improves the accuracy of soil moisture and snowmelt estimates by up to 16% and 24%, respectively, and it also improves the representation of vegetation anomalies relative to the reference crop fraction anomalies.