The goals of this research are to improve the initialization of convective storms in high-resolution (1 ~ 4 km) numerical models by assimilating dual-polarimetric Doppler ("dual-pol") radar observations, and likewise to improve understanding of the evolution of convective storms using numerical models and data assimilation. The objectives of this research include: 1) to develop a Doppler radar data assimilation methodology for ice-phase microphysical processes included in the "radar forward model", and to understand the general role of microphysics processes in convective storms and the storm related precipitation, using actual (i.e. as observed by ARMOR) case studies; 2) to understand the information content of dual-polarimetric radar observations and their relationship to model physics uncertainty; 3) to develop a technique to assimilate observations uniquely from the dual-polarimetric Doppler radar, investigate effective strategies for performing radar data assimilation, and evaluate the impacts of each dual-polarimetric variable in the model's [the Weather Research and Forecasting (WRF) model] initial condition and numerical prediction.
Intellectual Merits The research activities will improve skill in Doppler radar data assimilation with ice-phase forward model, a better understanding of uncertainty in model microphysical parameters and dual-polarimetric radar observations, and subsequently significant advances in dual-polarimetric Doppler radar data assimilation techniques. The research will take advantage of the availability of the ground-based C-band Advanced Radar for Meteorological and Operational Research (ARMOR) dual-polarimetric Doppler observations, as well as the unique suite of in-situ and remote sensing observational instruments currently operating in the University of Alabama in Huntsville (UAHuntsville)/National Space Science and Technology center (NSSTC; in particular, an X-band mobile Doppler radar). Dual-polarimetric measurements collected by the ARMOR radar will be assimilated into the WRF with ice-phase microphysical processes embedded into the data assimilation package. These data will also be used to examine and constrain the uncertainties in the WRF model's microphysical parameterization. Observations from the NSSTC instruments, high-resolution numerical simulations, and data assimilation results for select case events of convective storms will be used to conduct the aforementioned research activities. This research will be one of the first of its kind using dual-polarimetric Doppler radar observations with the WRF model and the 3DVAR data assimilation system for real case studies.
Broader Impact The research has the potential to improve operational forecasts of convective weather. In particular, the researchers will contribute to improving understanding of the evolution of convective storms and how to enhance short-term forecasts of convective precipitation via the assimilation of Doppler radar observations. The data assimilation techniques developed herein will be readily transferable to other research institutes, as well as the operational community, for improvements in the current prediction of convective storms. Given the upgrade of the WSR-88D radar network that is currently underway, research on the assimilation of dual-polarimetric radar data needs to move to the forefront so that new insights on the optimal methods for utilizing these data may be developed. The new findings will be communicated through conference presentations, peer-reviewed journal articles, and in educational materials for enhancing course curriculum. Collaboration with other university scientists and graduate students will further extend this research to the larger academic and educational community.
Determining the location, timing, and amount of rainfall is one of the most challenging problems in weather forecasting. Small changes in the interior of clouds and cloud systems can have large effects on where and when rain and snow falls at the surface. The processes that control how rain develops inside of clouds are hidden from view of most observing systems. Computer model predictions of clouds are particularly sensitive to these processes, especially in the case of convective storms (thunderstorms). There have been recent developments in radar technology that have promise for improving forecasts of rain produced by thunderstorms. Upgrades to the NOAA/NWS radar network now allow for much improved estimates of the rain content inside of clouds. However, the amount of ice in these storms (at heights where the air temperature is below freezing) is still poorly known, and it is not clear how much information the new radar observations can provide. Much of the rain that falls from thunderstorms begins as ice and snow high in the cloud, and because of this, predictions of rain will need to account for the amount of snow and ice in the middle and upper portions of storms. This project examined how much information the new dual-polarization radar system can provide on the snow and ice content in deep convective storms. It found that, as long as there was no rain present with the ice and snow, the radar provides very strong observations of snow and ice content. However, if the snow and ice is mixed with rain or if the number of snow and ice particles (and their density) is uncertain, then this can cause problems for radar observations of deep convective clouds. The results of this project will help to improve the use of dual-polarization radar data in forecasts of convective storms and will also improve measures of ice and snow in winter storm systems.
