Bridging the gaps between data collection and prediction at thunderstorm scales

点击翻译按钮获取中文摘要

While weather forecasting in general is showing steady progress, the timeliness and quality of warnings for severe and damaging weather has not markedly improved in the past decade. This occurs even though the tools and approaches used (forecasting models, computers, etc) have become more sophisticated. We hence believe this lack of progress is due to the fact that the current state of the atmosphere, necessary to make any forecast, is not correctly assessed. If this is the case, then we need to perform more research work on how to better measure the atmosphere in conditions when and where severe weather can occur, as well as on how to improve how the measurements are being used to provide information to weather forecasting tools. Since storm formation and evolution is critically dependent on temperature and humidity patterns, and since those are poorly measured over large areas with existing instruments, particularly in cloudy or stormy conditions, it is on this task that we will put most of our efforts.******To advance this task over the next five years, we will undertake a multi-pronged research effort. Namely, we will:******- Improve the information obtained from existing instruments: Our focus will be on continuing the development of a radar-based technique to obtain information of the humidity of air close to the surface that generally feed storms;******- Test a new instrument: We will evaluate the measurements obtained by a new prototype instrument developed here and designed to get information on temperature and humidity in the lower atmosphere over large areas;******- Revise current approaches to use measurements to initialize forecast models: We will develop new ideas on how to better assimilate or take advantage of the information coming from measurements to obtain a more accurate and complete description of the state of the atmosphere. Specifically, we want to devise new ways to take the information we have about the present and the past and better infer the state of the atmosphere that caused them with a focus on retrieving information most relevant to its future state.******In parallel, we will continue the development of a simple yet effective approach to forecast precipitation by using the time evolution of radar echoes of precipitation to determine what these will do in the future.******This research will be primarily performed by undergraduate and graduate students under my supervision. Our hope is that, in combination, all these lines of inquiry will help eliminate the main obstacle to better forecasts of damaging weather such as floods, hail, and tornadoes.**