• NMQ-FLASH-LANDSLIDE (NFL): Utilization of NASA Earth Observations into a NOAA Coupled Flood and Landslide Prediction System over the US
Program Disasters Program
PI / Institution Jonathan Gourley / NOAA/National Severe Storms Laboratory
Start Date January 1, 2013
End Date December 31, 2013
  • Summary

    Background: When data from different platforms are used synergistically, their strengths can be capitalized upon by minimizing errors from individual platforms. As an example, integration of NASAs vertical profiles of reflectivity from TRMM Precipitation Radar (PR) have been demonstrated to improve surface rainfall rate estimates when integrated with NEXRAD-based National Mosaic and Quantitative precipitation estimates (NMQ; http://nmq.ou.edu), especially over the complex terrain of the intermountain West. These will enable us to further improve upon the current NMQ into a truly Multi-Radar and Multi-Sensor (MRMS) system. The MRMS products provide a unique potential to force high-resolution models to predict impending flash floods and landslides. Moreover, space-based observations of surface inundation, stream discharge, and soil moisture are well suited to be assimilated into these application-oriented models to correct state variables and produce more reliable forecasts. Objective and Method: The overarching objective of this proposal is to build an integrated Earth-observing and modeling system to predict and map, for the first time, flash floods and landslides over the US at unprecedented spatiotemporal resolution (250 m/2.5 min) for real-time decision making by operational forecasters in the US National Weather Service (NWS). The specific remote sensing data to be used include multi-radar precipitation from NOAAs NMQ system improved with TRMM/GPM PR data, SRTM and HydroSHEDs, surface inundation from ASTER and MODIS, and future SMAP soil moisture and SWOT altimetry. The high-resolution, multisensor NMQ will then be used to force a state-of-the-art Ensemble Framework for Flash Flood Forecasting (EF5) system called NMQ-Flooded Locations And Simulated Hydrographs (NMQ-FLASH: http://eos.ou.edu/USA_Flood.html). The first-year feasibility study (Stage 1) will implement the NMQ-FLASH system in real-time over the Arkansas-Red basins in south-central US and also within the Colorado River basins in the West. The latter region poses more challenges to precipitation estimation from NEXRAD due to terrain blockages and will thus highlight the strengths of space-borne data from NASA. Forecasters at local forecast offices and project collaborators at the River Forecast Centers (RFCs) will be engaged to provide feedback on product utility, optimization, and accuracy. In the subsequent three years (Stage 2), the NMQ-FLASH system will be expanded to cover the conterminous US, will incorporate nascent polarimetric NEXRAD network and spaceborne GPM dual-frequency PR observations, will be supplemented with an ensemble Kalman filter for data assimilation, and coupled to the SLope-Infiltration-Distributed Equilibrium (SLIDE) landslide model. Ultimately, this project will merge and fully utilize NASA, NOAA, and USGS observations from space and ground in order to feed an end-to-end, coupled modeling system called NMQ-FLASH-Landslide (NFL) providing detailed maps of the specific times, locations, and magnitudes of impending flash floods and landslides, as well as maps for post-disaster assessment, at unprecedented, high resolution over the conterminous US. Relevance: The proposed NFLs relevance to the NASA Earth Science Applications objectives is paramount in that the project addresses the first two priorities for natural hazard forecasting (i.e., floods and landslides), as well as providing a unique opportunity to guide post-disaster assessments, a key responsibility of the partnering agency, the NWS. The project not only incorporates the models and data produced and managed by disparate organizations but brings people together (i.e., NOAA, NASA and University researchers; NWS national flash flood program managers and end-users). In this sense, the project is multifaceted and collaborative in its use of data, models, and personnel.