Click here for N49RF
transmitted sonde data.
Click here for N49RF HSA sonde data.
1. Synoptic situation
Hurricane Georges developed from an easterly wave that moved off the coast of Africa 13 September, and developed into a tropical depression two days later. The system moved westward, strengthening rapidly into a hurricane by 17 September. Due to the potential threat to the Gulf coastal states, a synoptic surveillance mission was tasked for nominal time 27 September 1998 0000 UTC. At that time, Georges was moving northwestward and was located in the northeastern Gulf of Mexico, to the south the subtropical ridge axis (Fig. 1). A cold low was located over northeastern Mexico. The weak circulations to the east of Bermuda were Hurricanes Karl and Jeanne.
2. General Assessment of dropwindsonde impact
A. GFDL model
Figure 2 shows the GFDL forecast tracks for Hurricane Georges, and Table 1 shows the errors and impact of the synoptic surveillance mission. The results are strongly positive, with improvements greater than 33% at all times except 72 h. The dropwindsonde data initially moved the forecast track to the left, but also slowed the track, allowing for recurvature further to the east than the run without the dropwindsonde data.
Georges made landfall 35.5 h into the forecast at Biloxi, MS. The forecast without the dropwindsonde date made its ultimate landfall 20 h into the forecast near Port Sulphur, LA, 112 km to the west. The forecast with the dropwindsonde data made landfall near St. Malo, LA, 82 km to the west, a large improvement. Both versions forecasted landfall at least 12 h before it occurred.
Figure 3 shows the VICBAR forecast tracks for Hurricane Georges, and Table 2 shows the errors and impact of the synoptic surveillance mission. The dropwindsonde data improved the forecast at 12 and 24 h, the two verification times before landfall. The dropwindsonde data pushed the storm further to the west away from the actual track, but the slower speed allowed for the improvements early in the forecast.
Georges made landfall 35.5 h into the forecast at Biloxi, MS. The forecast without the dropwindsonde date made landfall 21 h into the forecast at South Pass, LA, 160 km to the west. The forecast with the dropwindsonde data also made landfall at South Pass, LA, 160 km to the west, but at 23 h into the forecast. Both versions forecasted landfall about twelve hours before it occurred.
Figure 4shows the GSM forecast tracks for Hurricane Georges, and Table 3 shows the errors and impact of the synoptic surveillance mission. The dropwindsonde data had a large positive impact on the GSM forecast track through 72, and a mixed impact aftwards. Again, the dropwindsonde data pushed the forecast further to the left and made thestorm move slower than in the forecast without the additional data, allowing for a mainly positive impact.
Georges made landfall 35.5 h into the forecast at Biloxi, MS. The forecast without the dropwindsonde date made landfall 19 h into the forecast at Port Eads, LA, 157 km to the west. The forecast with the dropwindsonde data made landfall in the same place, but at 24 h into the forecast.
Table 4 shows the GFDL intensity forecast errors and impact of the synoptic surveillance mission. The dropwindsonde data had a positive impact at all times except 72 h. Tuleya and Lord (1998) also showed modest improvements to GFDL model intensity forecasts in the HRD synoptic flow cases.
The atmosphere has long been recognized as a chaotic system (Lorenz 1963), e.g. very small perturbations to initial conditions result in increasingly large differences in the evolution of the atmosphere with time. Since the exact state of the atmosphere can never be measured, all analyses contain errors whose magnitudes can only be estimated. An indeterminate number of initial conditions consistent with the observational data can therefore be used in numerical weather prediction, and single model runs at any synoptic time only give one possible solution to the evolution of the atmosphere. Many operational forecast centers around the world, therefore, now employ ensemble forecasting as a means of quantifying the uncertainty in the evolution of the atmospheric system. Small perturbations from a best "control" state are calculated and added to and subtracted from this control to allow for different integrations starting from theoretically equally likely initial states. These perturbations are designed so as to mimic the fastest growing modes in the model and to create the largest envelope of possibilities in the forecast. Therefore, they generally correspond to locations where large analysis errors will most impact the forecast. Those features corresponding to perturbations which most impact the tracks of tropical cyclones must be found, properly sampled, and thoroughly tested, to prove the efficacy of targeting techniques.
Figure 5 shows the variance of the size of the perturbations in the National Centers for Environmental Prediction (NCEP) global model ensemble forecasting system (Toth and Kalnay 1993). The maxima over the northeastern Gulf of Mexico is associated with Hurricane Georges. The large maxima over the Florida Straits is assocated with the anticyclone trailing behind the Hurricane. A maximum over the Great Lakes is associated with a mid-tropospheric vorticity maximum moving eastward, having little effect on the hurricane. Other smaller maxima over Guatemala and centered near Acapulco are associated with waves in the deep easterlies on the southern side of the subtropical ridge. The dropwindsonde data fully sampled only the first of these maxima, that associated with the hurricane itself.
An additional model run has been performed. The TG run includes the dropwindsonde data taken within and around Hurricane Georges (all except the first dropwindsondes of the mission), or just over two thirds of all the dropwindsondes released. Results are shown in Tables 1-4 and Figs. 2-4. The TG run provided better forecasts than the run including all the dropwindsonde data only at 12 h in the GFDL, at all times except 24 h in VBAR, and after 36 h in the GSM. The TG run improved the landfall forecast in the GSM only.
Figure 6 shows the difference in the 850 - 200 hPa averaged winds between the runs in which all the dropwindsonde data and none of the dropwindsonde data are included. The largest differences are north of the Yucatan peninsula with a smaller difference in the west central Gulf of Mexico. The differences extend a few hundred kilometers away from the location of the dropwindsonde data, and the largest difference appears to be centered between dropwindsonde locations. This may help to identify the cause of some of the mixed results in this case. The main impact seems to be the removal of a spurious vortex just to the south of the hurricane by the dropwindsonde data, and this difference is credited with allowing for the relatively large improvements at some time periods in some of the models. The largest difference is in a region in which the ensemble perturbations are large, and therefore the differences are expected to grow, though the second maximum may be expected to decay in time. Figure 7 shows that, by 24 h into the forecast, the largest difference between the forecasts with and without the dropwindsonde data has amplified and surrounded the hurricane, whereas the second maximum has decayed slightly and moved away from the hurricane. These results seem to confirm that the largest (smallest) ensemble perturbations correspond to amplifying (decaying) modes in the model. The results of the TG model runs confirm that a portion of the dropwindsondes have almost the same impact as all the dropwindsonde data.
The dropwindsonde data obtained during the synoptic surveillance mission for Hurricane Georges at nominal time 27 September 1998 0000 UTC has provided mixed, though mainly positive, results. The MRF ensemble forecasting system suggests, and the model runs confirm, that data around the hurricane itself have the greatest positive impact on the forecast. There seemed to be little problem with spreading of the data impact away from the location of the dropwindsondes in this case.
|GFAL Error (km)|
|GFTG Error (km)|
|12||51.||15.||( 71%)||10.||( 80%)|
|24||109.||40.||( 63%)||58.||( 47%)|
|36||163.||102.||( 37%)||120.||( 26%)|
|48||202.||134.||( 34%)||153.||( 24%)|
|72||350.||289.||( 17%)||312.||( 11%)|
|84||603.||397.||( 34%)||415.||( 31%)|
|96||956.||505.||( 47%)||505.||( 47%)|
|Landfall||112.||82.||( 27%)||97.||( 13%)|
|VBAL Error (km)|
|VBTG Error (km)|
|12||39.||31.||( 21%)||31.||( 21%)|
|24||97.||84.||( 13%)||93.||( 4%)|
|36||150.||160.||( -7%)||160.||( -7%)|
|Landfall||160.||160.||( 0%)||174.||( -9%)|
|GSAL Error (km)|
|GSTG Error (km)|
|12||81.||41.||( 49%)||54.||( 33%)|
|24||126.||71.||( 44%)||71.||( 44%)|
|36||200.||128.||( 36%)||137.||( 31%)|
|48||250.||206.||( 18%)||197.||( 21%)|
|72||564.||457.||( 19%)||427.||( 24%)|
|84||637.||672.||( -5%)||_||( )|
|96||743.||721.||( 3%)||_||( )|
|Landfall||157.||157.||( 0%)||146.||( 7%)|
|GFAL Error (kn)|
|GFTG Error (kn)|
|12||-17||-17||( 0%)||-16||( 6%)|
|24||-28||-24||( 14%)||-24||( 14%)|
|36||-28||-24||( 14%)||-25||( 11%)|
|48||-4||3||( 25%)||1||( 75%)|
|84||16||9||( 44%)||13||( 19%)|
|96||25||17||( 32%)||15||( 40%)|
Return to 1998 assessment page
Return to HRD home page