Mission Plan :

During the late morning of 20 August, NHC classified Hurricane Bill as a major hurricane with maximum surface winds of 105 kt and a MSLP of 949 mb. Bill was located near 23°N, 63° W, or about 550 miles to the north of our base in Barbados and was forecast to strengthen during the next 24 h to a 120 kt storm while moving to the NW. Hurricane Bill appeared on satellite as a classic West Indies hurricane with a large, clear eye surrounded by a thick CDO (Fig. 1). The hurricane was, however, imbedded within a relatively dry air mass and arc clouds were seen propagating away from Bill's core to the west and northwest. There was an absence of rainbands on Bill's south and west sides.

This was to be the last P3 mission in Bill and was designed to be a Tail Doppler Radar (TDR) mission with a rotating figure 4 pattern with 105 nmi radial legs. The P3 radar system (LF and Tail) was inoperable for much of the previous flight and AOC engineers were working on the system before and during the pre-flight pilots briefing. There was hope that the system could be repaired before a planned takeoff at 2000 UTC but that it might go down again during the flight. If the radar was broken, then we needed an alternative to the TDR mission.

The NOAA G-IV and an Air Force C-130 had been tasked by NHC to conduct a synoptic surveillance mission in the environment of Bill with a takeoff time of 1730 Z. The flight tracks for these surveillance missions would provide excellent coverage of the environmental air surrounding Bill, including a circumnavigation just outside of Bill's core by the G-IV (Fig. 2).

During the Pilot's briefing we discussed an alternative flight plan in case the radar system was inoperable. This plan included a figure 4 with 105 nmi radial legs, starting from the SE and ending up on the east side of Bill. We then would attempt a circumnavigation within the core of Bill and following what appeared to be a pronounced dry slot as viewed with microwave imagery. Dropsonde would be released at the end points, RMW and center of the figure 4 legs as well as at regular intervals during the circumnavigation of the dry tongue.

It turned out that the radar system would be inoperable for the flight and we could go with our alternative plan to study the affect of the dry air on Bill.

Figure 1: Visible satellite image of Hurricane Bill at 1945 UTC, right before N43's planned takeoff.

Figure 2: Planned track of the NOAA G-IV surveillance pattern around Hurricane Bill.

Mission Summary :

Our planned takeoff was delayed by a combination of work done on the radar and data systems problems. We finally took off at 2123 UTC and headed north to our IP, 105 nmi SE of Bill's eye. During the ferry to the storm, we noticed an absence of convection and a pronounced layer of SAL air below our flight level of about 600 mb (Fig. 3).

Figure 3. Photograph of SAL dust below our 600 mb level on ferry south of Bill.

At 2337 UTC we descended to 10 kft and reached our IP at 2334 UTC to begin our figure 4 pattern. We also released our first dropsonde here and the data from that sonde showed temperature inversions just below 950 mb and 800 mb. Saturated air was confined to the boundary layer and substantial drying was present above the inversion at 800 mb (Fig. 4).

We proceeded west and released sondes at 75 nmi radius, the SE RMW, and the eye which we reached at 2358 UTC. The eye sonde was typical of a strong hurricane with low-level moisture, a strong inversion and drying above 850 mb. The sonde reported a MSLP of 946 mb, slightly lower than estimated from satellite. Maximum flight-level winds on our inbound pass were 105 kt and the SFMR and sonde reported surface winds of about 85 kt. The flight-level winds were stronger on our outbound pass to the NW, about 115 kt and the SFMR and a sonde measured surface winds of about 95 kt, about 10 kt lower than expected based on the central pressure and the NHC advisory.

We reached our NW point, 105 nmi NW of Bill's eye at 0023 UTC and headed south for our second and last pass though Bill, from west to east. The west point was reached at 0038 UTC and we were in the eye again at 0104 UTC. We did not release a drop in the eye on this pass. The end of our figure 4 was at 0130 UTC, 105 nmi east and we climbed to 15,000 ft to begin our circumnavigation around Bill, trying to stay in the dry slot outside of Bill's eyewall and adjacent rainbands.

We uploaded microwave and water vapor imagery during the flight to help navigate downwind in the dry tongue. The nose radar was also useful in the navigation. Examples of these images that we used are in Fig. 5. We quickly flew downwind at about 100 nmi radius from Bill's eye, starting from the east side and ending up on the southwest side.

This maneuver took about 1 h and 15 minutes and we released 9 dropsondes along the way at intervals of about 30° of arc. The dropsonde data agreed with what was observed on the microwave imagery in that the driest and most stable air was found on Bill's east side and got progressively more unstable and moist as we worked our way to the north and west side.

Our last dropsonde was released at 0244 UTC at 23.5°N, 64.4°W, about 100 nmi south of Bill's eye. We expected to find dry air here but the sonde had a moist and somewhat unstable sounding. We most likely released that drop in or adjacent to a stratiform rainband. I would suspect that the drops from the G-IV at a greater radius will likely show much drier and more stable air.

From this south point we ferried back to our base in Barbados and we landed at 0412 UTC. We released a total of 20 dropsondes during the mission.

Figure 4: Skew-T diagram from first dropsonde, 105 nmi SE of Bill's center.

Figure 5: Microwave and water vapor imagery used during the flight to navigate inside of the dry slot outside of Bill's eyewall and inside of the spiral rainbands.

Problems :

The flight-level, SFMR, and dropsonde systems worked well for most of the mission. AOC engineers had to fix and reboot the data system computer prior to the flight, which delayed our takeoff, but it worked fine afterwards. The radar system was inoperable during the whole flight. Two of the dropsondes were fast falls and of these had no launch detect.

Michael Black
8/31/09