Mission Summary
Hurricane Bret
Landfall

Mission Briefing:

After two Extended Cyclone Dynamics flights in Hurricane Bret, it was time to capture Bret's landfall on the south Texas coast. The day before we drew up plans for two flights. The first flight would map out the wind field before the storm made landfall, including a rainband inflow module designed by Gary Barnes (U. Hawaii) that had been flown in Hurricane Bonnie (1998), and the second plane would follow up by mapping the onshore and offshore flow in the atmospheric boundary layer along the coast, while also measuring wave height and storm surge with Ed Walsh's Scanning Radar Altimeter (SRA).

On the morning of the 22 August, however, it was apparent that the storm would make landfall much earlier than we had hoped. Sam Houston redesigned the flight track for the first flight, combining the figure-4 wind field mapping with the coastal patrol (Fig. 1). Mike Black's soundings of the inner edge of the eye had been very successful the day before, so we decided to add two circles in the eye to the flight track. N43RF was picked to fly first, with a take off time of 1730 UTC, and N42RF was tasked for a possible follow-on flight that evening, if the storm had not moved too far inland. Texas Tech had deployed two towers, one at the Corpus Christ WSR-88D radar (KCRP) and the other at Kingsville Naval Air Station Annex (KAS, Fig. 2a)

Mission Synopsis:

N43RF left MacDill Air Force Base at 1738 UTC, and descended to 12,000' at our IP off the south Texas coast at 2013 UTC. The hurricane was 0.5° latitude south of the expected position, so the IP was adjusted to allow for an east-west run into the center (Fig. 2a). At 2032 UTC we dropped the first GPS sonde and AXBT ~20 km outside the east eyewall, where flight level winds were already 96 kt. The east eyewall sonde and AXBT drops at 2035 UTC had max winds of 120 kt, and sea-surface temperature (SST) of 26.6° C. Terry Lynch noted that the Step Frequency Microwave Radiometer (SFMR) winds agreed well with winds at the splash of the eyewall sonde. The aircraft then circled upwind in the eye and Mike selected the next three sondes to fall just on the inner edge of the eye at flight level. The aircraft exited west out of the eye at 2050 UTC, and the SFMR had winds of 62 m/s (124 kt) on the west side. We proceeded west to a point about 15 km inland (30 km west of the barrier islands) and at 2055 UTC slowly turned to head southeast. Two sondes were deployed over land (after consultation with AOC crew), west of Port Mansfield and near the Laguna Atacosa National Wildlife Refuge, and a third was dropped over the Intercoastal Waterway north of Port Isabel. These three sondes should provide information on the offshore flow. At 2108 UTC we dropped the third AXBT and turned to head north for the second leg of our figure-4. After passing through the eye at 2118 UTC and dropping the 4th AXBT, and then dropping the 5th AXBT and several sondes in the north eyewall, N43RF tracked north along the coast and completed the figure-4 at Port Aransas (PTAT2) at 2037 UTC.

From PTAT2 we flew south along Padre Island collecting SRA data (Fig. 2b). We deviated southeast to hit the eye again, and then at 2207 UTC we turned east of Brownsville (KBRO) to head for buoy 42020, where we dropped a sonde at 2230 UTC. Then we headed back to PTAT2. Now there was a vigorous rainband between CRP and PTAT2, so the pilots flew a circle around Corpus Christi Bay (Fig. 2c) and then back to the coast where we resumed another SRA run, adjusting the track at 2253 UTC to go through the eye. At 2314 UTC we turned east of KBRO to do a final pass through the eye. We spent about 10 min circling in the eye, to collect more inner edge soundings, and at 2344 UTC we made our last pass through the north eyewall in heavy rain, with 118 kts flight level winds. A lower fuselage reflectivity sweep (Fig. 3 ) shows how small the storm was. Our last sonde was dropped offshore at 2355 UTC, in the rainband that was over Corpus Christi. We continued along the coast, and at 0005 UTC passed on the eastern edge of the strongest reflectivity feature we saw during the whole flight, a rainband crossing the middle of Matagorda Island. There was a 13.8 m/s updraft in the cell, and the tail radar sweeps had narrow high reflectivity elements similar to the features observed in a rainband off Moorehead City, North Carolina, during research flights in Hurricane Bonnie. At 0009 UTC N43RF climbed to head home and landed at MacDill at 0253 UTC , 23 August 1999.

Evaluation:

Hurricane Bret was the most intense hurricane that we have flown the TC Windfields at Landfall Experiment in. Because the inner core was so small, and moved inland in a sparsely populated region we had considerable freedom to deploy sondes and to maneuver the aircraft. Figure 4 shows the distribution of mean boundary layer wind data from the sondes that was received at the Tropical Prediction Center in real time; this does not include 7 other sondes that were not transmitted. The SFMR software had some problems, so the real-time data were not available, but they will be re- processed to yield detailed estimates of the coastal winds. Initial reports form the Southern Region of the National Weather Service (NWS) are that both Brownsville and Corpus Christi radars successfully archived Level-II data. The Southern Region also archived 1-min data from the CRP, ALI, PIL, BRO, and HRL ASOS stations. The Texas Tech towers recorded data successfully. This will be an excellent data set to examine wind fields in a landfalling hurricane. In addition, the eyewall soundings will add to the data set that Mike Black is compiling to study variations in vertical profiles of winds in tropical cyclones.

Acknowledgements:

Barry Damiano, Dave Rathbun and the flight crew agreed to every change in the flight pattern that we requested (there were many). Terry Lynch kept the radar running, and Jim Barr and Dale Carpenter managed to keep with our evolving plans for GPS sonde drops.

Mike Black called most of the sonde drops, and also added his upwind inner-edge eye circle drops to the mix. Joe Cione kept track of the AXBT data, and John Gamache minded the radar. Sam Houston and Frank Marks made the initial flight tracks, and Pete Black and Barry Damiano helped us tune up the pattern just before the pre-flight briefing. Tim Crum, OSF, and Victor Murphy, Southern Region of the NWS, alerted the WSR-88D sites about our experiment. The Southern Region of the NWS also archived the 1 -min ASOS data. NWS staff at Corpus Christi and Brownsville were quite helpful in letting us know the status of the radar data soon after the storm.

Problems:

The workstation had a few problems with the sonde processing icons (since fixed). The tail radar had a bad AFC circuit, and the engineers had to continually reset the tail radar to restore the reflectivity. The Doppler data do not appear, at first glance, to be seriously affected. The lower fuselage reflectivity also seems low when compared with Brownsville WSR-88D dBZ images. Three sondes had no winds and another failed early. The SFMR winds transmitted in real time had problems, but the software has since been fixed by Eric Uhlhorn and AOC.

Peter Dodge
9/7/99

Return to Bret mission listing.
Return to Bret main page.