Error Analysis of Satellite Precipitation-Driven Modeling of Complex Terrain Flood Events

Abstract:

The error characteristics of satellite precipitation driven flood event simulations over mountainous basins are evaluated in this study for eight different global satellite products. A methodology is devised to match the observed records of the flood events with the corresponding satellite and reference rainfall and runoff simulations. The flood events are sorted according to flood type (i.e. rain flood and flash flood) and basin’s antecedent conditions represented by the event’s runoff-to-precipitation ratio. The satellite precipitation products and runoff simulations are evaluated based on systematic and random error metrics applied on the matched event pairs and basin scale event properties (i.e. cumulative volume, timing and shape). Overall satellite-driven event runoff exhibits better error metrics than the satellite precipitation. Better error metrics are also shown for the rain flood events relative to the flash flood events. The event timing and shape from satellite-derived precipitation agreed well with the reference; the cumulative volume is mostly underestimated. In terms of error propagation, the study shows dampening effect in both systematic and random error components of the satellite-driven runoff time series relative to the satellite-retrieved event precipitation. This error dampening effect is less pronounced for the flash flood events and the rain flood events with high runoff coefficients. This study provides for a first time flood event characteristics of the satellite precipitation error propagation in flood modeling, which has implications on the Global Precipitation Measurement application in mountain flood hydrology.