The FY2013 Disaster Assistance Supplemental provided support for the project titled “Develop Improved Predictions of Inland Flooding”, which was led by the the late Prof. Peter Lamb, CIMMS Director, and myself. FLASH, being a component of the broader MRMS project, was transitioned to NCEP with MRMS v11 in November 2016 for experimental use. To date , over half of local NWS offices throughout the U.S. are receiving FLASH products into their operational AWIPS2 systems. During the week of July 2, FLASH products will be disseminated to all NWS offices in the lower 48 states via the Satellite Broadcast Network. This completes a major milestone for the team!
The success of FLASH was largely attributed to the number of graduate students, post-docs, and research scientists who contributed and the NWS forecasters who participated and provided valuable feedbacks during the HMT-Hydro testbed experiments from 2014-2016. Throughout the project, the team published 12 articles in peer-reviewed journals, 4 of which were in the Bulletin of the American Meteorological Society. The FLASH project resulted in 7 graduate students earning Ph.D.s, 1 earning an M.S., and supported numerous visiting student interns.
While we have reached this major milestone, we continue to develop. FLASH is being reframed into a fully probabilistic framework, called Pro-FLASH. This consideration of uncertainty in the modeling process accommodates forcings from ensemble precipitation forecasts. This can increase the lead time with the distributed hydrologic forecasts, but must consider the associated uncertainties with the forcings. We are also developing FLASH products for our outer-CONUS domains including Alaska, Hawaii, Guam, and the Caribbean. Lastly, we are forging relationships in the private sector so that FLASH can assist in operations related to transportation and beyond.
Optimizing Local Thresholds for Flash Flood Prediction
A student led project, Zachary Flamig and his advisor Jonathan Gourley, submitted FLASH (Flooded Locations And Simulated Hydrographs) – a project that leverages existing technologies, resources, infrastructure and relationships to address critical data, knowledge and assessment gaps that will enhance the effectiveness of flood response and mitigation strategies
The Thriving Earth Exchange (TEX) and Amazon Web Services (AWS) are collaborating to use cloud-computing and Earth and space science to advance solutions to community challenges related to natural resources, climate change and natural hazards.
$15,000 in grant funds to use AWS on-demand cloud services.
Free registration to attend the annual 2015 AGU Fall Meeting, an Earth and space science conference attracting over 24,000 attendees. Assistance and direction in highlighting the project to enable adaptation by other communities and increase broader impact.
Graduate student and PhD candidate Race Clark (OU/CIMMS) won 3rd place at the 2015 OU GIS Day poster contest held November 17, 2015 in Norman, OK. His poster was titled “Processing Topographical Data for Hydrological Modeling”, which presented a new software tool for speeding up topographical preprocessing for hydrological models. This software tool is primarily designed for use during international capacity building workshops. His coauthor on the poster is fellow graduate student and PhD candidate Zac Flamig (OU/CIMMS). Both students are advised by Drs. J.J. Gourley (NOAA/NSSL) and Yang Hong (OU Civil and Environmental Engineering).
Manabendra Saharia, Ph.D. student in CEES/ARRC, won 1st place in the graduate student poster contest at the Society of Environmental Journalism (SEJ) 25th Annual Conference. The conference was held in Norman, Oklahoma from October 7th-11th, 2015. Manab’s poster focused on his work to identify the basins in the US with the fastest rainfall-runoff response.
On the evening of May 23rd into the early morning of May 24th, 2015 the Blanco River in Texas experienced a significant flooding event resulting in the closure of I-35 north & southbound near the city of San Marcos, TX. There were also multiple high-water rescues in the area including helicopter rescues of victims trapped in houses. The Hays County jail in Texas was also evacuated due to threatening flood waters. The FLASH development system was running for this event producing products forecasting & monitoring the flooding in near real time.
Traditional rainfall based products including the rainfall accumulation and the ratio of rainfall to flash flood guidance highlight a large area northwest of San Marcos for flooding impacts. The storm system produced 6-8″ of rain over the region to the northwest resulting in exceedance of flash flood guidance by 160-180%.
The FLASH hydrologic model based products correctly show the heavy rainfall being concentrated into the rivers and the downstream flooding impacts in San Marcos resulting from the heavy rainfall to the northwest of the region.
This was a very heavy rainfall event that was captured well by MRMS radar-only estimates
The flash flood guidance product indicated significant exceedance over threshold values, but the areas of concern were limited to the headwaters of the Blancos river
The distributed hydrologic model forecasts better refined the threat area to the Blancos river and provided several hours of lead time
This event clearly highlights the need to account for overland flow and routing, displacing the impacts well downstream of the causative rainfall
The MRMS-FLASH system, running in a real-time demonstration mode for several years now, performed admirably for the recent flash flooding events in Oklahoma and Texas. The MRMS-FLASH system runs in a completely automated fashion ingesting quantitative precipitation estimates every 5-minutes from the MRMS mosaic of NEXRAD data. The FLASH hydrologic modeling system subsequently produces 6-hour forecast hydrographs for everywhere across the CONUS.
On evening of May 25th, 2015 the city of Houston, Texas experienced a serious flash flooding event. MRMS precipitation estimates indicate that the western portions of the city experienced 6-8″ of rainfall.
The ratio of precipitation to flash flood guidance shows exceedance values of 100-140% over this area, however some of the USGS stream gauges reporting flash flooding (blue triangles) and the flash flood local storm reports (brown circles) fall outside of this region. This product is important because it is equivalent to the main tool operational NWS forecasters have available for flash flood monitoring and prediction.
The Coupled Routing and Excess Storage (CREST) distributed hydrologic model, a part of the MRMS-FLASH hydrologic modeling suite, generates maps of streamflow and unit streamflow (cubic meters per second per square kilometer) every 15 minutes. Comparisons between the observations of flash flooding and the maps of unit streamflow show a good correspondence between areas of high unit streamflows and flash flooding.
The MRMS-FLASH suite of hydrologic models are able to correctly identify areas of flash flooding thanks to the incorporation of
Elizabeth Mintmire Argyle received the Student Presentation Award from the 10th Symposium on Societal Applications, held as a part of the 2015 AMS Annual Meeting in Phoenix, AZ in January 2015, for her oral presentation “Forecaster ‘Best Practices’ During Operations in the Hazardous Weather Testbed Hydrology Experiment 2014.” As part of the recognition, Ms. Argyle received a $200 cash award and a certificate from the American Meteorological Society. Ms. Argyle works as a Graduate Research Assistant for the Cooperative Institute for Mesoscale Meteorological Studies at NOAA’s National Severe Storms Laboratory, in Norman, OK. She is also a PhD candidate in Industrial and Systems Engineering at the University of Oklahoma. Her PhD advisors are Dr. JJ Gourley (NOAA/NSSL), Dr. Ziho Kang (OU), and Dr. Randa Shehab (OU). Funding for Ms. Argyle’s research was provided by NOAA/OAR/Office of Weather and Air Quality (OWAQ) under the NOAA cooperative agreement, NA11OAR4320072.