Future weather warning systems will be driven by short-term ensemble forecasts of convective storms produced by frequently updated, storm-scale, data assimilation systems. Presently, these systems…

This is a grant awarded by the National Aeronautics and Space Administration (NASA) to the University Corporation for Atmospheric Research (UCAR) to demonstrate the value of using high-resolution satellite data from the next-generation GOES-R satellite to improve forecasts of severe convective storms. The $343,152.86 grant will fund research that aims to show how assimilating GOES-R atmospheric motion vectors and convective initiation products into a state-of-the-art storm-scale ensemble forecasting system can accelerate the "spinup" of convection and lead to better predictions of severe weather like tornadoes. The work is expected to enhance future severe weather forecasting and warning capabilities. This grant does not have a set-aside designation. UCAR, which manages the National Center for Atmospheric Research (NCAR), is a nonprofit consortium that conducts atmospheric and Earth system research under federal contracts and grants.

Future weather warning systems will be driven by short-term ensemble forecasts of convective storms produced by frequently updated, storm-scale, data assimilation systems. Presently, these systems rely heavily on radar networks to provide observations within convective storms. Next-generation satellite systems (e.g. goes-r) promise an additional source of convective-scale observations with unprecedented high spatial and temporal resolution. Assimilating satellite-based products show potential to vastly improve forecasts of convective development and changes in storm intensity over what is currently possible with radar data assimilation only, especially in regions where radar observation coverage is poor. Current approaches with storm-scale data assimilation rely on the development of high reflectivity storm echoes before observations are available to be assimilated, delaying the onset of storms within weather prediction models. This spinup delay reduces the value of forecasts with immature convective development in storm-scale analyses. Yet, recent work has shown great promise to considerably accelerate the spinup of convection through assimilation of derived satellite products, leading to much improved forecasts of convective storms. This will be especially important for providing accurate severe weather warnings in cases where severe weather, such as tornadoes, occur soon after initiation, as occurred in moore, OK on 20 may 2013. The proposed work will demonstrate the viability of using high spatial and temporal resolution near storm environment (NSE) atmospheric motion vectors (amvs) and convective initiation (CI) related products derived from current and future goes satellites, within a state-of-the-art storm-scale data assimilation and ensemble forecasting system, to improve predictions of CI and storm evolution. These efforts are expected to lead to future improvements in forecasts and warnings of severe convection. After convection has developed, a considerable challenge for forecasters is anticipating changes in storm behavior that may elevate the threat for local storm hazards such as large hail, tornadoes and damaging winds. As in convective initiation applications, changes in the near storm environment may be captured by frequently updating and high spatial density amvs. Further, total lightning flash rate has been identified as a prognostic variable that has a positive correlation with updraft intensity. The geostationary lightning mapper on the next generation goes-r satellite will provide essential monitoring of changes in lightning activity, with consistent coverage across a broad area, including areas where current lightning activity monitoring systems are unreliable and weather radar coverage is poor or nonexistent. The proposed work will demonstrate the value of these data to improve storm predictions. To clearly demonstrate the value of select goes r products to improve prediction of severe storms, the proposed work will use data denial experiments with a state of the art storm scale ensemble data assimilation system. Cycled high resolution analyses with or without the goes r products will be done for a range of severe storm events, including tornadic supercells, for both onvective initiation and rapid changes in storm intensity scenarios with an aim of capturing regional diversity within the data set. Verification of ensemble forecasts of high impact severe storms initialized from analyses with or without assimilation of goes r products will be used to assess the forecast value of goes r products for prediction of severe storms.