Actions for Improving Hurricane Fred Forecast with METEOSAT-10 All Sky Radiances

Email RIS file
Bookmark
Report an Issue

Improving Hurricane Fred Forecast with METEOSAT-10 All Sky Radiances

Author
Wellington, Alisha
Published
[University Park, Pennsylvania] : Pennsylvania State University, 2022.
Physical Description
1 electronic document
Additional Creators
Chen, Xingchao and Schreyer Honors College
Access Online

Availability

I Want It
I Want It

Finding items...

Restrictions on Access
Open Access.
Summary
Hurricane Fred made landfall on Cape Verde August 31, 2015, as a category 1 hurricane. The hurricane originated from an African easterly wave that developed from a broad cyclonic rotation within the lower atmosphere off the coast of West Africa on August 29. The low-pressure system quickly intensified over the warm water of the Atlantic into a tropical storm and later a category 1 hurricane. Cape Verde and the eastern Atlantic Ocean are considered a data void region due to the lack of buoys and radars that are vital in providing meteorologists information on the state of the atmosphere. As a result, the track and intensification of Hurricane Fred were poorly forecasted. The main objective of this project was to improve the forecast of Hurricane Fred using the Weather Research and Forecasting (WRF) Model together with advanced data assimilation of all-sky infrared radiances from Meteosat-10. Without assimilation of satellite all-sky infrared radiances, biases in the WRF track, minimum surface pressure, and maximum surface wind forecasts existed when compared to the best track observations from the National Hurricane Center (NHC). When comparing the observed and WRF forecasted tracks, the WRF forecasted track was shifted to the north compared to the best track observations from 1800 UTC August 30 to 1200 UTC August 31. Regarding the WRF forecasted minimum surface pressure, the WRF forecast considerably underestimated the minimum surface pressure when compared to the best track observations. Similarly, the WRF forecasted maximum surface winds significantly underestimated the observations. Lastly, the WRF forecasted brightness temperatures were also quite different from those based on the Meteosat-10 radiances. Data assimilation (DA) served as a way for us to improve Hurricane Fred's track and intensity forecasts. DA is a statistical technique that combines information from model forecasts and observations to create the best estimate of the atmosphere. In this study, data assimilation was used to create more accurate forecasts of Hurricane Fred. To this end, the Penn State WRF Ensemble Kalman Filter (EnKF) system was used to assimilate conventional observations and all-sky infrared radiances from Meteosat-10 into the WRF model. With updated initial conditions via data assimilation, four deterministic forecasts were initialized: i)CONV 08/29 18Z, ii) CONV 08/30 00Z, iii) IR 08/29 18Z, and iv) IR 08/30 00Z. CONV represents experiments in which only conventional observations were assimilated, and IR represents experiments in which both conventional and all-sky infrared radiances were assimilated. Both the CONV and IR forecasts initiated at 1800 UTC August 29 were nearly spot on with the best track observations on August 31. When assessing the surface minimum pressure, the IR 08/30 00Z forecasts had much smaller absolute errors compared to the CONV forecasts. For the maximum wind speed forecasts, all four forecasts captured the highest maximum wind speed that occurred on August 31, with the IR forecasts having smaller absolute errors compared to the CONV forecasts. Satellite all-sky infrared radiance data assimilation improved the intensity forecast of Hurricane Fred, demonstrating potential value of this approach for hurricane forecasting in the eastern Atlantic Ocean.
Other Subject(s)
Genre(s)
Dissertation Note
B.S. Pennsylvania State University 2022.
Technical Details
The full text of the dissertation is available as an Adobe Acrobat .pdf file ; Adobe Acrobat Reader required to view the file.
View MARC record | catkey: 38042199