A Flight into Hurricane Luis
September 8, 1995

Jack Beven

I. Introduction

One of my desires while working at the National Hurricane Center (NHC) has been to fly a reconnaissance plane into a hurricane. I'd never been able to do this up to now. A major problem with this desire is my job itself. While the National Oceanographic and Atmospheric Administration's (NOAA) P-3 research planes (four-engine propeller aircraft) were based in Miami for many years, I was never able to fly one into a storm. I was needed at the office every time a storm came close enough for the planes to fly.

I had a chance to fly a typhoon during the Tropical Cyclone Motion (TCM) 90 experiment. The National Aeronautics and Space Administration (NASA) DC-8 jet fled from Okinawa to Guam to avoid Supertyphoon Flo. The plane took several experiment people through the storm the next day. Unfortunately, I came down with a mild case of food poisoning that day (from a vending machine cheeseburger), so I passed on the trip. I thought it was a poor idea to fly in a typhoon with a weak stomach!

During the fall of 1995, a great twist of fortune gave me a chance to fly into Hurricane Luis. Here is the story of the circumstances and the flight.

II. Luis comes along...

For many years, the "Hurricane Hunters" have flown Atlantic hurricanes and returned vital information on location and intensity. There were two squadrons, an active duty unit and a reservist unit. The active duty unit was disbanded several years back, leaving the reservists (the 53rd Weather Reconnaissance Squadron) as the current crop of Hurricane Hunters. These people are very busy whenever a hurricane moves into flying range, and the active 1995 season made them busier than ever.

The Hurricane Hunters are based at Keesler Air Force Base in Biloxi, Mississippi. However, Biloxi is not always a convenient location to fly into hurricanes. When the storms are far out in the Atlantic, the planes fly from the U.S. air base on Antigua (or used to in 1995). They also use bases in Georgia and the Carolinas for storms further north.

Hurricane Luis was a large and powerful Cape Verde hurricane that began in the central tropical Atlantic in late August. The storm tracked west until September 5 when it moved into the Leeward Islands. It then packed maximum sustained winds of 125-140 mph. Luis wreaked havoc on the islands from Antigua to St. Martin, with lesser damage from Martinique to Puerto Rico. It moved northwest into the Atlantic north of Puerto Rico on September 6, and this motion continued on Thursday, September 7.

The Hurricane Hunters were quite busy in Luis. They started almost continuous flights when it was 300 miles east of the Leeward Islands. A major complication was that Luis aimed for Antigua, which forced the Hurricane Hunters to flee to Barbados. This was also the base for the P-3's flying research missions into Luis. (The move was a wise one, as Luis clobbered the Antigua base and it was unusable for the rest of the season.) Flights continued as Luis moved into the Atlantic north of Puerto Rico.

By September 7, the storm was too far away from Barbados for convenient flights. John Pavone, NHC's coordinator of the aircraft, arranged for the next several missions to fly from Homestead Air Reserve Base (ARB), Florida. This is 12 miles from my apartment of that time. Simultaneously, the NOAA planes returned to Miami after completing their experiments.

III. The chance comes!

I enter the story at this point. Thursday the 7th was the last of five day shifts for me, and I looked forward to a scheduled three-day weekend. I thought about a possible hurricane flight, as Luis appeared to be no threat to land for at least the first two days of my weekend. This meant I wouldn't be needed to work overtime. I hoped the P-3's would fly over the weekend and that I could hitch a ride on one of them.

During Thursday morning, I called Hugh Willoughby (now director of the Hurricane Research Division (HRD)) to see if the P-3's were flying over the weekend. He said they weren't. That closed the P-3 door. Later on, I was checking the latest aircraft data in John Pavone's office. I don't remember exactly how the conversation turned that way, but John mentioned that the Hurricane Hunters were flying Luis from Homestead on Friday morning. He then asked me if I wanted to go.

I must admit I was GREATLY surprised. The Hurricane Hunters usually didn't fly out of Homestead, so this was unexpected. I told John that I would very much like to go, and he started putting the paperwork together for the Friday morning flight. By mid-afternoon, everything was in place.

Several of my co-workers were also interested in flying Luis. One of them, Fiona Horsfall (now a Ph.D. meteorologist and computer specialist), wound up on the same flight as myself. Two others arranged to fly on Saturday morning. Fiona and I arranged to drive to Homestead together. We would arrive for the pre-flight briefing at 3 AM with take off scheduled for 5 AM.

I had one piece of business to take care of before I could prepare for the flight. I was dining that evening with a friend from Australia (Liz Ritchie, whom I met at TCM-90) who was at HRD for research. (She had flown Luis on the P-3's a few days earlier.) A cloudburst and puddles a foot deep hampered the cross-town trip to Key Biscayne. Fortunately, the rain missed Key Biscayne, and dinner went off well.

After dinner, I went home to prepare for the flight and to try to sleep. The preparations included packing a light meal, my camera, and my jacket. The meal was necessary as the mission was to last eleven hours, and the Homestead base kitchen was closed. The jacket was necessary because the Hurricane Hunter WC-130's are somewhat lacking in creature comforts like heating and air-conditioning. The camera, well, that's self-explanatory!

Trying to sleep turned out to be futile, as I was way too keyed-up. I twisted and turned for several hours until it was time to leave. Fiona picked me up shortly after 2 AM and we headed for Homestead.

IV. The journey begins

We arrived at Homestead just before 3 AM and immediately went to the briefing room at Base Operations. This turned out to be a crowded place. First, there was the aircraft crew of six. Second, there was a ground support crew of three or four. Then there were Fiona and myself. In addition, there were six other press people coming with us. These included two Miami TV people and four other reporters and cameramen. Finally, there was a Homestead ARB media relations man who was coming along to ride herd on the civilians.

The briefers told us this would be a normal mission with four eye penetrations and seven to eight hours spent in the storm. Luis was turning more north with time, so we would intercept the storm about 600 miles east of Miami. We also heard that the previous mission had found the lowest pressure yet measured in Luis, and that they had gotten bounced around by turbulence.

The turbulence news made some civilians a little nervous about airsickness, so the crew mentioned that airsickness bags were available. They also had Dramamine, but they recommended against taking it unless absolutely necessary. The reason was that Dramamine would knock you out and that you would miss the excitement of the flight. (I was a little apprehensive about airsickness, but turbulence makes me more nervous than nauseous. Thus, I wasn't too worried.) The crew also told us they would provide earplugs, as the plane wasn't insulated against sound. One final comment was that a portable toilet was aboard if needed. After the briefing and some last rest-room stops, we headed to the plane.

[Small HLuis1]
Picture 1

(11K JPG)
[Small HLuis2]
Picture 2

(16K JPG)

The WC-130 is a four-engine propeller aircraft. Pictures 1 (11K JPG) and 2 (16K JPG) show the plane from outside. (These were taken after we landed.) You can see the main entry door, as well as the picture window just aft of the door. The WC-130 has two such windows, and most of my pictures were taken there. The rest of the windows outside the cockpit are like the small portholes just forward of the propeller. Most of the windows are on the flight deck (the cockpit), so the pilots had the best view. The oddly-shaped black nose contains a weather radar antenna.

[Small HLuis3]
Picture 3

(51K JPG)

Picture 3 (51K JPG) shows the interior (the cargo compartment) from back to front. Yes, that's a bunk attached to the roof with someone sleeping in it! The red straps on either side of the aisle are canvas seats, while other red straps hanging to the floor are seat belts. The computer console is the dropsonde operator's position (more on this later), and the gray things hanging just left of that are the life jackets (in case of a water landing). The large object in front of the computer console is a fuel tank. (No smoking!)

The Air Force is thorough with pre-flight checks, so it was 15 minutes or so from the time the engines started until we took off just before 5 AM. The flight to Luis would take about two hours at 18,000 ft. This gave me time to get acquainted with the crew and their duties, and to occasionally listen to them on the internal communications system.

[Small HLuis4]
Picture 4

(36K JPG)

Five of the six crew members normally are on the flight deck, while the sixth is at the dropsonde position. Picture 4 (36K JPG) shows the pilot and co-pilot's position taken from the flight engineer's position. They are watching the radar display, which is showing the rain pattern associated with the storm. The co-pilot is reading a sheet that tells how to estimate surface wind speeds from the appearance of the ocean surface. This is rather important, as the plane would be flying at 8,000-10,000 ft in the storm and the winds there usually differ from those at the surface.

[Small HLuis5]
Picture 5

(45K JPG)

Picture 5 (45K JPG) shows the navigator's position. He is looking at a chart of the plane's position and giving the pilot the proper course to fly. He also has a radar display in front of him. The WC-130 has a computer navigation system, so he always knows where the plane is.

[Small HLuis6]
Picture 6

(44K JPG)

Picture 6 (44K JPG) shows the airborne weather officer's position. He monitors the data flow from the automated instruments and relays the information back to NHC. The instruments return data every thirty seconds, thus giving a detailed look at conditions along the flight path. Both the raw and coded data (on the data sheets) are sent back to the office. The WC-130 has a satellite communications system that keeps it in constant contact with NHC. This allows the crew to pass along comments or ask for directions if necessary. In addition, the on-board computers allow the crew to do a great deal of data processing and display on the aircraft before transmitting it.

[Small HLuis7]
Picture 7

(44K JPG)

Picture 7 (44K JPG) is a close-up of the dropsonde operator's console. A dropsonde is an expendable instrument package launched from the plane to return weather data as it falls. On a hurricane flight, it is typically dropped in the eye to get surface pressure data. The sonde is launched from the gray tube as the plane crosses the center of the eye. The operator then monitors the data and processes it before sending it back to NHC. The console also can show the 30 second observations, so I spent much time there watching the data flow.

[Small HLuis8]
Picture 8

(24K JPG)

It was dark at takeoff, but dawn came before we reached the storm. I looked out the window and saw very large swells on the ocean surface (Picture 8 [24K JPG]). Luis had a large circulation that was an efficient swell generator, and these swells were moving from the storm toward the Bahamas. I had two reactions when I first saw the swells: First, I didn't want to have to land in them, and second, the hurricane is that way (meaning straight ahead)!

V. Into the storm!

When we were 150 nautical miles from the center, we descended to 10,000 ft for the run to the eye. From here, we would follow the 700 millibar (20.67 in) pressure surface through the storm. This meant we would descend to near 8,000 ft as we entered the eye, then climb to near 10,000 ft as we left. Even this far out on Luis' weak side, the winds were already 40 mph. The winds and clouds increased as we approached the eye, and we encountered bands of rain with some turbulence. (Throughout the flight the strongest turbulence occurred in the bands of showers and thunderstorms.) Soon the eye appeared on the radar, at which time the crew made us all strap down for the eyewall penetration. (The eyewall is the ring of thunderstorms surrounding the eye containing the highest winds and heaviest rains.)

The press had the picture window seats on this approach, and several other people had the communications headsets. This left me strapped into my seat looking over my shoulder through a porthole. There was little to see. The sky was various shades of gray with bursts of rain, and we were getting continuously bumped around. Then the turbulence died away, and I saw people start getting up. I asked Fiona (who had been listening to the crew during the approach) if we were there yet. She said yes, so I hurried back to the dropsonde operator's porthole. It was about 7:15 AM EDT and we were in the eye of Luis.

[Small HLuis9]
Picture 9

(17K JPG)

My first view of the eye (Picture 9 [17K JPG]) was somewhat disappointing. The eye was cloud-filled both above and below, and I had trouble seeing the eyewall in the distance. It looked nothing like the pictures I'd seen in books. Things changed little as we flew along (and the view out the porthole wasn't that good), so I decided not to take any more pictures of this eye crossing.

Suddenly, a loud bang occurred! The dropsonde operator had fired the sonde from its tube, and I was standing right next to it! The bang was audible all over the cargo compartment, and several people came back to ask what had happened! I alternated between watching the data flow on the computer and looking out the window until we had to strap in for leaving the eye.

The outbound leg to the east was rough. There were numerous rain bands and frequent turbulence all the way to the 150 nautical mile point, and we had to stay seated the entire time. It was rough enough that I didn't notice the plane turning northwest to fly to the north side of the storm. There was nothing to see, so I managed to grab a little sleep during this time. (I found out later that the flight-level winds were 100+ mph this entire leg!)

[Small LuisFT]
Flight Track

(26K GIF)

Finally, things settled down enough so we could move around. I went to check on the data collected during the pass. (This is a detailed vortex data message. This message is typical of what the planes send back when they locate the storm center.) It turned out that Luis had two rings of thunderstorms encircling the center, with the outer ring 50 nautical miles in diameter and the partial inner ring 20 nautical miles across. This is a condition know as concentric eyewalls, and it is usually the sign of a steady-state or weakening storm. This accounted for how sloppy the eye looked. The dropsonde reported a central pressure of 941 millibars or 27.79 in, which was up about 5 millibars from the end of the previous flight. (Normal atmospheric pressure is about 1013 millibars, while Hurricane Andrew had a 922 millibar pressure when it hit Homestead. Hurricane Frederic had a 943 millibar pressure when it hit the Gulf coast in 1979.) The flight level winds had reached 100 mph on the way in and 145 mph on the way out. (The 145 mph wind is marked in purple on the flight track (26K GIF), as it was the highest we saw on the flight.) Luis was weakening, but it still packed a wallop!

The northeast and north sides weren't as turbulent as the east side, and we were able to move around and look out the windows. The ocean, when we could see it, was VERY rough. After about an hour, we approached the eye from the north and it was time to strap down again.

[Small HLuis10]
Picture 10

(24K JPG)

This time I sat at the picture window. At first, all I could see was featureless gray as we flew through the eyewall. This was accompanied by frequent bumping. The turbulence then ceased, yet still nothing was visible out the window. Finally, about two minutes later the clouds and fog cleared and I could see the eye. I then took Picture 10 (24K JPG) looking back at the eyewall. The eye's low clouds are in the lower part of the picture, with the hazy edge of the eyewall elsewhere. Again, it was not a picture-perfect eye.

(Note: My memory of when I took pictures 11-13 is a little hazy. I may be assigning them to the wrong eye passage.)

[Small HLuis11]
Picture 11

(27K JPG)

Picture 11 (27K JPG) shows the typical scene inside the eye: overcast low clouds. Most hurricane eyes have low clouds, but Luis' was filled with them. (It also had its fair share of higher clouds.) The eye was so large that it couldn't clear itself out. Obviously little ocean was visible inside the eye.

After the bumps in the south eyewall, I again checked the collected data. This time we had flown through 115-120 mph winds on both sides of the eye. The minimum pressure was 942 millibars (27.82 in). The position indicated a north-northwest motion since the first fix.

The south side of Luis lacked shower activity outside the eyewall, and once we almost flew out from under the high-level overcast. Despite this, flight-level winds were still 75 mph when we turned northeast. After the turn, I went and took a short nap. I woke up in time to listen to the next approach on the communications headset. This involved the weather officer telling the pilot to make slight course corrections to keep the plane perpendicular to the wind interspersed with comments on the wind speed. The navigation officer chimed in with occasional comments on the radar patterns. This time through the radar indicated that the inner eyewall seen earlier was gone. This left a single 45 nautical mile wide eye. The approach from the east was not as rough as the previous outbound leg, and as soon as we cleared the eyewall I got up and headed for the picture window.

[Small HLuis12]
Picture 12

(30K JPG)

Picture 12 (30K JPG) shows a good view of the eyewall. The bottom of the wall is at the top of the dark band running through the central part of the picture, with the top of the wall faintly visible about a quarter way down from the top. Low clouds fill the eye between the plane and the eyewall, but this time a break in the overcast is letting sunshine through. This only happened a few times while we were in the eye.

I should mention here one thing I found a little strange: it never got all that dark outside the plane. Either the eyewall was not dense enough to make it that dark, or being 1-2 miles up makes a lot of difference in how dark it looks. Given how dark it looked under the eyewall, I think it's the latter.

After the west eyewall, it was time for another data check. This time we had measured 130 mph winds entering the eye and 105 mph exiting. The pressure was 943 millibars (27.85 inches) and the location showed a continued north-northwest motion. Luis was behaving about what the forecasters at NHC were expecting.

The next hour was rather smooth, as the southwest side of Luis again had only a little shower activity away from the eyewall. (The winds remained 75-85 mph for much of this time, though.) This gave me a chance to eat lunch and to spend time on the flight deck. (That's when I took the flight deck pictures.) Then I went back to the window for the final eye penetration. By this time, most of the reporters were either asleep or too tired to fight for the window, so Fiona and I had it to ourselves.

[Small HLuis13]
Picture 13

(23K JPG)

Picture 13 (23K JPG) shows the eye just after coming through the eyewall. The base of the is just above the dark band in the center, and the top is again faintly visible in the upper part. The striations in the low clouds show the winds are still quite strong at this point.

[Small HLuis14]
Picture 14

(31K JPG)
[Small LuisNA]

(149K JPG)

Picture 14 (31K JPG) is the only one where I could catch some clear skies. The top of the eyewall and a patch of blue skies are visible in the upper right hand corner of the picture, while the darkness below the base of the wall is seen in the lower left. Low clouds are visible in the lower right, while middle-level clouds (above us but below the top of the eyewall) are in the upper left. (The NASA GOES image [149K JPG] is a special false-color multispectral image taken about the time of our last pass through the eye. This and other similar images can be found at the NASA Remote Sensing Data Program Image Catalog.)

The reconnaissance coordinator at NHC let us cut our trip slightly short by letting turn in the eye and exit to the southwest instead of the north. This would save us time going back to Homestead. A data check after leaving the eye showed that the maximum winds on the south side were about 110 mph, and the minimum pressure was 944 millibars. The pressure rise showed a steady slow weakening while we were there. Luis continued to track north-northwest on a course that would take it between the southeastern United States and Bermuda. NHC anticipated it would eventually turn northeast away from the U.S., and the data we sent back helped reinforce this idea.

[Small HLuis15]
Picture 15

(10K JPG)

I got an excellent view of the ocean as we were leaving Luis, so I took Picture 15 (10K JPG). Things are pretty badly churned up, and it is obviously not a good place to be in a small boat. ("Small craft should remain in port!") At this time, the flight-level winds were 60-70 mph and the surface winds were probably 45-55 mph. Even though Luis could offer much worse, I thought it was a fitting last sight.

VI. Homeward bound

Once we got out to 150 nautical miles, we climbed to 18,000 ft for the trip home. The skies cleared, and we had beautiful views of the Bahamas as we flew over them. The swells we had seen at dawn were creating fantastic surf along the outer Bahamas. (Although we couldn't see surfers from our altitude, I'm sure they were there.) By this time, everyone was tired and we just wanted to get back to Homestead. Thus, I alternated between views at the picture window and the position data at the dropsonde console for the next hour-and-a-half.

Finally we got close enough to Homestead to have to strap down for landing. That turned out to be a little tricky, as a thunderstorm with gusty crosswinds was over the runway. After some cautious maneuvering, we managed to land successfully.

We returned to Base Operations to turn in the equipment we had borrowed. Before we left, the Miami TV crew grabbed Fiona and myself for quick interviews, which got edited from the report that actually made it on the air. Fiona drove me home through the rain, and I gratefully made my way into my apartment and to my bed. It had been a long, tiring, and VERY exciting day!

VII. Epilog

Luis moved north through the open Atlantic on Saturday the 9th, with my two other colleagues getting to fly it. The storm then turned northeast and clipped Newfoundland before losing tropical characteristics.

Another flying opportunity came about a week later. Hurricane Marilyn moved through the Virgin Islands into the Atlantic, and the Hurricane Hunters again came to Homestead to fly it. I didn't go this time, but four more of my colleagues got to fly it and share my experience. From the pictures and tales they brought back, it looks as if they had just as interesting of a trip as I did.

I don't know when events will come about so I can fly a hurricane again. However, the next time the chance comes, I'm going to jump at it!

Note: All pictures on this page except the flight track and the NASA image are © Jack Beven. Please ask my permission before using them publicly.

Back to Jack's Home Page

Valid HTML 4.01!