A Real-Time Hydrometeorology Research Testbed for Heavy Rainfall and Streamflow Prediction

There have been increasing demands for accurate hydrological prediction, as urban flooding resulting from heavy rainfall and other high-impact weather has resulted in significant societal and economical costs. Thanks to the improved observation networks, advancement of computer technology, and scientific and technological development of data assimilation, the past decade has seen improvement in quantitative precipitation forecasts (QPF) and streamflow predictions. The ultimate improvement in rainfall and streamflow prediction that can truly benefit society requires a fully coupled, seamless, end-to-end system that consists of high-resolution, quantitative precipitation estimation (QPE) observations, state-of-the-art QPF modeling systems with radar data assimilation, high-resolution, location-specific quantitative precipitation nowcasts (QPN), and advanced hydrological modeling. This paper describes an experimental fully-integrated, hydrometeorology-based, heavy rainfall and streamflow prediction testbed developed by NCAR's Short Term Explicit Prediction (STEP) research program that was run in real-time from 7 July – 31 August 2014 over the Rocky Mountain Front Range region. This region was chosen because of its high frequency in flash flood occurrence, the existence of three operational NEXRAD radars, dense surface stations and rain gauges, streamflow gauges, and NCAR's Earth Observing Laboratory's mesoscale observation network FRONT (Front Range Observation Network Testbed) that includes the NCAR S-Pol and Colorado State University CHILL S-band polarimetric research radars.

This fully-integrated system includes 1) quality-controlled, radar-based QPE (MRMS and NEXRAD polarimetric) and rain gauge QPE, 2) QPF from state-of-the art WRF 3DVar models with (and without) radar data assimilation and frequent 1 and 3 h update cycles, 3) radar echo extrapolation (TITAN) and precipitation accumulation nowcasts (QPN) from 10 min – 1 h from the heuristic-based AutoNowcaster/Trident system, 4) high resolution 4-D winds and buoyancy analyses from VDRAS, 5) streamflow prediction on a spatially-continuous 100m resolution grid from the WRF-Hydro coupled atmosphere and hydrology model, and 6) a near real-time performance evaluation of the above components using Model Evaluation Tools (MET) and the Method for Object-Based Diagnostic Evaluation (MODE) tool.

Examples of the real-time performance of this system for selected heavy rainfall (>1 inch) events from 2014 will be presented. Results show there is still much room for improvement in providing location-specific and time-specific prediction of heavy rainfall, flash floods and peak streamflow and providing useful products for the end-user. A second demonstration of the experimental system is planned for the summer of 2015 incorporating improvements to the system that will be discussed in the paper. Highlights from 2015 demonstration may also be included in the conference presentation.