Status and Future of Global Flood and Landslide Nowcasts and Forecasts Using Satellite Precipitation Observations (Invited)

The advent of quasi-global, real-time precipitation analyses has lead to the reality of running global hydrological models and algorithms for the estimation of the occurrence of floods and rain-induced landslides. These calculations provide information useful to national and international agencies in understanding the intensity, timeline and impact on populations of these significant hazard events. The quality of such applied hydrological estimations should improve with time due to continuation and improvement of multi-satellite precipitation observations through the Global Precipitation Measurement (GPM) program and the further development of the models and algorithms. This talk will summarize the results from the NASA-based, real-time flood and landslide nowcasts and forecasts and describe directions for improving results going into the GPM era. Global flood and landslide estimation systems have been running in real-time at 0.25° latitude/longitude resolution using multi-satellite rainfall analyses for several years, with results available through the TRMM website (trmm.gsfc.nasa.gov). Published evaluations of the current system indicate useful skill in comparison with global event inventories. The evaluations indicate higher skill for larger rainfall systems (e.g., tropical cyclone landfall vs. flash flood). This result is reasonable considering the resolution of the rainfall information (0.25° and 3-hr) and the resolution of the current models/algorithms (0.25°). Improvements over the next few years will include 1) better precipitation analyses utilizing space-time interpolations that maintain accurate intensity distributions, 2) improved rain estimation for shallow, orographic rainfall systems and some types of monsoon rainfall, 3) higher resolution landslide algorithms with combined physical/empirical approaches, 4) higher resolution flood models with accurate routing and regional calibration, and 5) use of satellite soil moisture for more accurate pre-conditions. In addition, the satellite rainfall and surface observations will be integrated with regional atmospheric models to provide enhanced information for the hydrologic calculations. One example of an experimental step in the use of atmospheric models is the linking of the satellite rainfall to global numerical rainfall forecasts to extend the usefulness of the flood and landslide forecasts. In this example of potential enhancement the model rainfall is adjusted by the satellite observations to provide improved rainfall amounts and flood forecasts. In addition recent results with an improved global hydrological model running at 1/8th degree resolution will be discussed that are shown to produce more realistic evolution of flooding events and more detailed information. These improvements are a few steps in a continuous pathway to simultaneously improve satellite rainfall estimates and their applications to increase the accuracy of hazard estimation and forecasts on a global basis.