NHC Senior Hurricane Specialist Dr. Mike Brennan gave a seminar on “The Impact of Supplemental Dropwindsonde and Rawinsonde/Observations on Model Track Forecasts of Hurricane Irene (2011)”
As Hurricane Irene approached the east coast of the United States the NOAA G-IV jet flew 10 synoptic surveillance missions from 23-27 August, along with one surveillance mission flown by an Air Force Reserve C-130 aircraft on 23 August. Additionally, supplemental 0600 and 1800 UTC rawinsondes were launched beginning at 1800 UTC 22 August from upper-air stations in the southeastern United States, and expanded to include all of the continental United States from the Rocky Mountains eastward beginning at 0600 UTC 25 August.
Data denial studies were performed for the supplemental observations using the National Centers for Environmental Prediction (NCEP) Gridpoint Statistical Interpolation (GSI) data assimilation scheme. The NCEP Global Forecast System (GFS) and Hurricane Weather Research and Forecasting (HWRF) models were then run using the analyses that excluded the supplemental observations. Differences between the operational GFS and HWRF forecasts and the “data-denial” forecasts were used to quantify the impact of the supplemental observations on the track forecast of Irene in both models.
Preliminary results suggest a small but overall positive improvement due to both the dropwindsonde and supplementary radiosonde observations on the track of Irene in the GFS through 5 days. The dropwindsonde data showed the largest improvement in 2 to 3 day forecasts, while the supplemental rawinsondes showed the largest improvement at days 4 and 5. The supplemental data resulted in little change to the track of Irene in the HWRF model through day 3 and some degradation to the track at days 4 and 5.
The recorded presentation is available via ftp here.