Convective Meteorology (Mesoscale Dynamics) Seminar
Satellite Data Assimilation, Verification, and Applications
Dr. Thomas Jones
Friday, February 8th
Cooperative Institute for Mesoscale Meteorological Studies,
University of Oklahoma, and NOAA/National Severe Storms Laboratory,
Norman, Oklahoma 73072
Satellite data assimilation is a vital tool in numerical weather prediction (NWP) and has been shown to significantly improve forecast skill in global models. Recently, increased emphasis has been placed on satellite data assimilation into convection allowing models to improve forecasts of high impact weather. One such application is the Warn-on-Forecast project which is designed to provide probabilistic short-term (0-3 h) forecast guidance for high impact weather events such as tornadoes, hail, high winds, flash flooding, and landfalling tropical cyclones. A prototype system has been developed to test a rapidly updating, ensemble-based data assimilation and forecasting system in a real-time environment beginning in 2016. This system assimilates a multitude of data types including conventional data from surface observations (parameters are temperature, dewpoint, wind velocity and pressure); WSR-88D radar reflectivity, Level 2 Doppler radial velocity, and finally several forms of satellite observations.
Satellite data assimilation is a challenging process and many different types of satellites and sensors exist, each with specific applications. In particular, satellite data often come in two forms: retrievals such as temperature, humidity, winds and cloud properties, and raw radiances that require a radiative transfer model to assimilate. Various satellite data products are discussed and their applicability to convection allowing models assessed. For the WoF system, we focus on cloud water path retrievals, water vapor channel radiances, and atmospheric motion vectors from the GOES-16 satellite. Assimilating cloud water path observations have increased forecast skill compared to not assimilating these data by significantly improving cloud analyses and forecast of convective initiation. The assimilation of GOES-16 clear-sky water vapor channel radiances into the WoF system has recently been tested with initial results indicating neutral to positive improvements in forecast skill. The primary goal being to improve the mid-tropospheric moisture environment within the model, allowing for better forecasts of developing and ongoing convection. Atmospheric motion vectors represent wind speed and direction retrievals calculated by tracking cloud objects and assimilating these data improves the overall dynamical analysis within the model. These improvements translate into a more accurate analysis of boundaries and convergence zones, increasing the forecast skill for convection.
In addition, the impacts of cloud microphysics scheme on satellite data assimilation are discussed and object-based verification of satellite objects has shown that modifications were required to the NSSL 2-moment cloud microphysics scheme to accurately represent upper-level clouds. Many applications of satellite data assimilation into the WoF system exist, and one such example is forecasts of land-falling tropical cyclones. For hurricanes Harvey and Irma in 2017, comparison with radar observations and storm reports indicate that the WoF system often identified the areas corresponding to locations specific high impact weather events 1–3 hours prior to their occurrence. Given the success of these initial tests, it is hoped that the system can be applied to future tropical systems in real time and provide forecasters with a new tool for short-term guidance of the multi-hazard environment of tropical cyclones.