Click here for N42RF
catalog of sonde drops.
Click here for N42RF 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 Leward 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, and a follow-on mission was ordered for the following day. At that time, Bonnie was embedded in the westerlies to the south of the subtropical ridge about 250 km north of western Puerto Rico. A trough was located to the east of the U. S. east coast, seemingly bypassing Bonnie, and the subtropical ridge axis was strengthening from the Carolina coast east-southeastward (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 mainly positive. The dropwindsonde data improved the forecasts at all times except between 24 and 48 h when both forecasts were excellent. The upper-tropospheric data improved the forecasts at all times except between 12 and 36 h, again when all the forecasts were excellent.
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 mainly positive, with the dropwindsonde data improving the forecasts at all times except at 36 and 48 h. The model runs without the dropwindsonde data are qualitatively better than those including the additional data through 48 h, since the dropwindsonde data allowed the storm to move in a more westerly direction. The upper-tropospheric data degraded most of the forecast, suggesting that Bonnie was steered by a layer shallower than that used in VICBAR.
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 mainly positive impact on the GSM forecast track, with degradation only through 24 h and at 72 h, when both forecasts were excellent. The forecast without the dropwindsonde data seemed qualitative better than that including the extra data, but the dropwindsonde data substantially improved the speed forecast. The upper-tropospheric data had positive impact only at 72 and 84 h.
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 36 h, 48 h, and 96 h onward. Tuleya and Lord (1997) 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). Only seven of the ensemble runs were available at this initial time. One large perturbation corresponds to the circulation of Bonnie itself. Another, was located in the trough, just to the east and northeast of Bermuda, and another associated with another vorticity maximum just downstream of the first. A small perturbation was located in an area of diffluent flow just to the north of a center of the subtropical ridge to the northeast of Bonnie. The maximum in the Gulf of Mexico was associated with Tropical Storm Charley. Due to air traffic control constraints, and because the second P3 aircraft was unavailable due to mechanical failure, only the region around Bonnie itself was well-sampled.
Two sets of model runs have been performed. The first, the TG run, includes the dropwindsonde data taken within and around Bonnie, or all the dropwindsondes in the southernmost flight, those extending from about 23°N 60°W to 25°N 64°W, those from 29°N 67°W to 30°N 70°W, and those from 28°N 75°W to 23°N 76°W. The other, the NT run, includes the complement of the TG run. Results are shown in Table 1, Table 2, Table 3, Table 4 and Fig.2, Fig.3, 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 through 72 h. The TG run provided better forecasts than the run including all the dropwindsonde data after 84 h in VBAR, and before 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 differences are at the center of Bonnie, in Bonnie's outer circulation to the northwest of the center, and in the axis of the subtropical ridge to the northwest of the center. Other difference maxima were in Bonnie's outer circulation to the east of the center, and in the trough to the north and northwest of Bermuda. The last of these maxima occur in an area in which dropwindsonde data were not obtained, suggesting 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 decay, and since the area is relatively far removed from Bonnie, this may not have 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 the three initial maxima in the storm circulation have amplified. The largest difference results from the joining of the maxima initially to the southeast of Bonnie and that over Bonnie's center. Another maxima located between Jamaica and Cuba is the result of the initial difference to the northwest of Bonnie. The difference maximum along 29N results from the initial difference in the subtropical ridge axis, and has decayed, as has the possibly spurious initial difference which has propagated southeastward in the trough. These results seem to confirm that the largest (smallest) ensemble perturbations correspond to amplifying (decaying) modes in the model.
The dropwindsonde data obtained during the synoptic surveillance mission for Tropical Storm Bonnie at nominal time 22 August 1998 0000 UTC has provided mainly positive results. The MRF ensemble forecasting system suggested that data surrounding Bonnie and in the trough east of the United States east coast would have the greatest impact on the Bonnie forecast. Due to logistical difficulties, only the first of these two regions was sampled during the synoptic surveillance mission. The impact of the data north and west of Bermuda shows a maximum in an area in which no dropwindsonde data were available, suggesting that the data assimilation filled the gap with surrounding data. Since this impact maximum was located in an area in which model differences are expected to decay, and far away from the circulation of Bonnie, this difference had no impact on the track forecasts.
|GFAL Error (km)|
|GFP3 Error (km)|
|GFTG Error (km)|
|12||59.||43.||( 27%)||33.||( 44%)||41.||( 31%)|
|24||49.||85.||( -73%)||65.||( -33%)||75.||(-53%)|
|36||75.||112.||( -49%)||62.||( 17%)||82.||( -9%)|
|72||140.||74.||( 47%)||164.||( -17%)||97.||( 31%)|
|84||264.||148.||( 44%)||254.||( 4%)||173.||( 34%)|
|96||453.||305.||( 33%)||439.||( 3%)||345.||( 24%)|
|120||1306.||1087.||( 17%)||1302.||( 0%)||1141.||( 13%)|
|VBAL Error (km)|
|VBP3 Error (km)|
|VBTG Error (km)|
|12||35.||23.||( 34%)||23.||( 34%)||30.||( 14%)|
|24||127.||125.||( 2%)||120.||( 6%)||125.||( 2%)|
|36||210.||225.||( -7%)||225.||( -7%)||230.||(-10%)|
|48||250.||266.||( -6%)||261.||( -4%)||281.||(-12%)|
|72||189.||155.||( 18%)||155.||( 18%)||163.||( 14%)|
|84||219.||59.||( 73%)||45.||( 79%)||69.||( 68%)|
|96||406.||192.||( 53%)||168.||( 59%)||184.||( 55%)|
|108||706.||412.||( 42%)||367.||( 48%)||403.||( 43%)|
|120||1203.||744.||( 38%)||664.||( 11%)||741.||( 38%)|
|GSAL Error (km)|
|GSP3 Error (km)|
|GSTG Error (km)|
|24||49.||53.||( -8%)||53.||( -8%)||39.||( 20%)|
|36||136.||82.||( 40%)||72.||( 47%)||84.||( 38%)|
|48||240.||137.||( 43%)||122.||( 51%)||150.||( 38%)|
|72||256.||263.||( -3%)||267.||( -4%)||246.||( 4%)|
|84||381.||351.||( 8%)||367.||( 4%)||328.||( 14%)|
|96||602.||596.||( 1%)||554.||( 8%)||518.||( 14%)|
|108||1023.||983.||( 4%)||912.||( 11%)||871.||( 15%)|
|120||1577.||1530.||( 3%)||1365.||( 13%)|
|GFAL Error (kn)|
|GFP3 Error (kn)|
|GFTG Error (kn)|
|24||12||4||( 67%)||3||( 75%)||4||( 67%)|
|36||21||11||( 48%)||16||( 24%)||13||( 38%)|
|48||11||6||( 45%)||15||( -36%)||14||( -27%)|
|72||5||8||( -60%)||2||( 60%)||6||( -20%)|
|84||4||5||( -25%)||3||( 25%)||1||( 75%)|
|96||7||4||( 43%)||5||( 29%)||5||( 29%)|
|120||16||14||( 13%)||15||( 6%)||17||( -6%)|
Return to 1998 assessment page
Return to HRD home page