IMPACT ON HURRICANE TRACK AND INTENSITY FORECASTS OF GPS DROPWINDSONDE OBSERVATIONS FROM THE SYNOPTIC SURVEILLANCE MISSION INTO TROPICAL STORM BONNIE ON
20 AUGUST, 1998.

Sim D. Aberson

Hurricane Research Division
Atlantic Oceanographic and Meteorological Laboratories
National Oceanic and Atmospheric Administration

Click here for N42RF catalog of sonde drops.
Click here for N42RF zipped post-processed sonde data.
Click here for N43RF catalog of sonde drops.
Click here for N43RF zipped post-processed sonde data.
Click here for N49RF catalog of sonde drops.
Click here for N49RF zipped post-processed sonde data.

1. Synoptic situation

Tropical Storm Bonnie developed from an easterly wave midway between the Cape Verde Islands and the Leeward Islands on 19 August, 1998. It moved westward, strengthening slowly. Due to the potential threat to the Virgin Islands and Puerto Rico, a synoptic surveillance mission was tasked for nominal time 21 August 1998 0000 UTC. At that time, Bonnie was embedded in the westerlies to the south of the subtropical ridge about 150 km northeast of Antigua, and a strong trough was located to just east of the U. S. east coast, which could impact Bonnie's track ( Fig. 1).

2. General Assessment of dropwindsonde impact

A. GFDL model

Figure 2 shows the GFDL forecast tracks for Tropical Storm Bonnie, and Table 1 shows the errors and impact of the synoptic surveillance mission. The results are mixed, so that through 24 h, the no dropwindsonde control qualitatively provides a better forecast than that with the dropwindsonde data. After 24 h, the dropwindsonde data qualitatively improve the forecast. The upper-tropospheric data do not substantially change the forecast track.

B. VICBAR

Figure 3 shows the VICBAR forecast tracks for Tropical Storm Bonnie, and Table 2 shows the errors and impact of the synoptic surveillance mission. The results are mixed, though the track with the dropwindsonde data is qualitatively better than that without late in the forecast. The upper-tropospheric data degraded most of the forecast, suggesting that Bonnie was steered by a layer shallower than that used in VICBAR.

C. GSM

Figure 4 shows the GSM forecast tracks for Tropical Storm Bonnie, and Table 3 shows the errors and impact of the synoptic surveillance mission. The dropwindsonde data had a mixed impact on the GSM forecast track. The forecast was qualitatively and quantitatively improved in the critical 36 to 72 h range with the dropwindsonde data. The upper-tropospheric data had little impact on the forecast track.

D. Intensity

Table 4 shows the GFDL intensity forecast errors and impact of the synoptic surveillance mission. The dropwindsonde data had a mixed impact, and the upper-tropospheric data seems to have only helped at 72 and 96 h. Tuleya and Lord (1997) also showed modest improvements to GFDL model intensity forecasts in the HRD synoptic flow cases.

3. Targeting

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). One large perturbation corresponds to the circulation of Bonnie itself. The other, centered just east of the North Carolina coast, represents the axis of the deep-layer trough moving toward the east. Both features were relatively well-sampled by the synoptic surveillance missions.

Two sets of model runs have been performed. The first, the TG run, includes the dropwindsonde data taken within and around both Bonnie and the trough, or the majority of sondes. The other, the NT run, includes the complement of the first (the three dropwindsondes in the eastern Bahamas, the four dropwindsondes in the northeastern Caribbean sea, the three dropwindsondes from about 30°N 55°W to 30°N 60°W, and the three from about 25°N 70°W to 27°N 73°W). Results are shown in Table 1, Table 2, Table 3, and Table 4 and Fig. 2, Fig. 3, and Fig. 4. The TG run provided better forecasts than the run including all the dropwindsonde data at all forecast times in the GSM except 36 to 48 h. The TG run provided better forecasts than the run including all the dropwindsonde data from 84 to 96 h in VBAR, and after 72 h in GFDL.

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 difference is to the northeast of Bonnie. Other large differences define the western extent of the subtropical ridge to the north of Bonnie, a third to the southeast of Bonnie, and another representing differences in the trough off the United States east coast. The lack of dropwindsonde data in the area to the southeast of Bonnie suggests that the data assimilation has allowed the dropwindsonde data from surrounding areas to influence the initial conditions there. Since this area is a region in which small perturbations are expected to grow, this may have ultimately harmed the forecast. Figure 7 shows that, by 24 h into the forecast, the differences between the forecasts with and without the dropwindsonde data in these four regions have amplified. The largest perturbation is the due to the slight difference in location of Bonnie in the different forecasts. The perturbation in the trough has moved eastward and amplified rapidly. The next largest perturbation is the result of the initial differences to the southeast of Bonnie, which may be suspect. This large difference may be the reason that the early forecasts with the dropwindsonde data are further to the south than the ones without the data, causing degraded early forecasts in all three models. The large perturbation initially to the northeast of Bonnie is currently located to the north, and that representing the subtropical ridge axis remains, stretching along 30°N. Though almost all the dropwindsonde data were obtained in regions of likely perturbation growth, the results seem to confirm that the largest (smallest) ensemble perturbations correspond to amplifying (decaying) modes in the model.

Two further model runs have been performed: T1 refers to model runs with the dropwindsonde in and around the target region representing Bonnie, and T2 corresponds to a similar run with data pertaining to the trough to the north. Track forecasts from these runs are shown in Figures 8, 9, and 10 for the three models. The forecasts with the data surrounding Bonnie closely resemble those with all the targeted dropwindsonde data (the TG forecasts), although they are slightly faster. The forecasts with the data surrounding the trough the north closely resemble those without any dropwindsonde data, though these forecasts are slower, and only have substantial impact after four days. This suggests that the second target region is not important in the ultimate track forecast of Bonnie until the mid-range. Unfortunately, the data surrounding Bonnie initially pushed the forecasts to the south resulting in forecast degradation.

4. Conclusion

The dropwindsonde data obtained during the synoptic surveillance mission for Tropical Storm Bonnie at nominal time 21 August 1998 0000 UTC has provided mixed results. The MRF ensemble forecasting system suggested that data surrounding Bonnie and in the trough east of the North Carolina coast would have the greatest impact on the Bonnie forecast. The greatest impact has been shown to be from the data surrounding Bonnie, whereas only slight impact is seen before four days from the data in the northern feature. The impact of the data in and surrounding Bonnie shows a maximum in a relatively large area to the southeast of Bonnie in which no dropwindsondes data were available, suggesting that the data assimilation filled the gap with surrounding data. This asymmetry caused the forecast track to be further south than may have otherwise occurred, and caused the early forecast track to be degraded versus those with no dropwindsonde data included.

Return to 1998 assessment page
Return to HRD home page