Hurricane Camille: The Record From a Home Barometer

Chapters

1. 0:00 The Mississippi coast, 1969
2. 0:39 The question
3. 0:58 A double eyewall
4. 1:17 From Africa to the Gulf
5. 1:47 Flying into the eye
6. 2:54 901 millibars
7. 3:44 Where 190 mph came from
8. 4:30 Radar on film
9. 4:49 Landfall, 11 PM
10. 5:41 A 24.6-foot storm tide
11. 6:10 The wind instruments
12. 7:03 Our measurement
13. 8:01 A barometer in Bay St. Louis
14. 9:36 The 1969 record
15. 10:08 The 2014 reanalysis
16. 10:43 Checking the 901
17. 11:53 Our arithmetic
18. 12:32 How 900 mb was analyzed
19. 13:37 Where Camille stands
20. 14:22 The cost
21. 14:38 What the record supports
22. 15:02 Next: Tuscaloosa, 2011

Transcript

The Mississippi coast, 1969 0:00

This is the Mississippi coast, filmed from a Navy aircraft in the days after Hurricane Camille came ashore, in August 1969. Camille is one of only four Category Five hurricanes on record to hit the United States mainland. And it is the hurricane that was measured twice.

In 1969, the official record said the winds were a hundred and ninety miles an hour, and the pressure in the eye was nine hundred and nine millibars. In 2014, a team went back through every surviving piece of data. The wind came down, to a hundred and seventy-five. And the pressure went lower, to nine hundred.

The question 0:39

Weaker and stronger, at the same time. And the number at the center of both records was not read by any government instrument. It was read on a marine barometer, in the home of a boat dealer, in Bay Saint Louis. This is every number in Camille's record, and the instrument that made it.

A double eyewall 0:58

This is how one radar saw Camille on the afternoon before landfall, from the Naval Air Station at Pensacola. Two rings of cloud, one inside the other: a double eyewall. In 1969, almost nobody knew what that meant. It is the key to the first number that was wrong.

From Africa to the Gulf 1:17

Camille began as a tropical wave that left the coast of West Africa on August 5th. It did nothing until it reached the western Caribbean. On August 15th it crossed the western tip of Cuba as a strong Category Two.

Then it reached the warm water of the Gulf of Mexico. The Hurricane Center's season summary noted that it was apparent almost from the outset that Camille would be an explosively deepening storm. In about twenty-four hours, the pressure measured in its eye fell by fifty-eight millibars.

Flying into the eye 1:47

In 1969, the only way to measure the center of a hurricane at sea was to fly into it. Air Force and Navy reconnaissance aircraft crossed Camille's eye again and again. Their best instrument was dropped out of a hatch.

It was called a dropsonde. A small package of sensors falls on a parachute through the eye, and radios back pressure, temperature and humidity as it goes down. The last reading, at the sea surface, is the central pressure.

On the afternoon of August 16th, two dropsondes inside a circular eye ten nautical miles across both read nine hundred and eight millibars. That evening, the same flight measured nine hundred and five. Two instruments agreeing, a storm near its peak.

The satellites of 1969 saw the same storm from above. The pictures show a pinhole eye inside a smooth, round shield of cloud. Decades later, the reanalysis team would line those images up against Hurricane Wilma of 2005, one of the most intense storms ever measured, and find the two looked almost alike.

901 millibars 2:54

On Sunday afternoon, August 17th, an Air Force WC-130 went in one last time. Its dropsonde reported nine hundred and one millibars. The Weather Bureau's preliminary report called it the lowest pressure of record ever recorded by aircraft reconnaissance.

The nine hundred and one went everywhere. It was in the operational advisories, in the preliminary report, in the federal climate record, and in the Mariners Weather Log. For a short time, it was the lowest pressure any aircraft had ever measured in a hurricane.

This is the coded message the aircraft sent. In the middle of it, a single line: H, nine zero one. And the post-flight summary ends with the next fix, at midnight Greenwich time, aborted: engine number four had been shut down.

Where 190 mph came from 3:44

On that same flight, the crew looked down through a clear patch in the eye and saw the sea surface in deep furrows, running with the wind. From what the sea looked like, and from the nine hundred and one, they reported surface winds of a hundred and eighty knots.

The forecasters turned that into a landfall intensity of a hundred and sixty-five knots, a hundred and ninety miles an hour. The aircraft of 1969 could not accurately measure winds inside a hurricane. The hundred and ninety was a judgment, made from the look of the sea.

And it was the last look anyone got. No reconnaissance flight was inside the storm during the last several hours before it reached the coast. From then on, the hurricane would be measured only by what was waiting for it on land.

Radar on film 4:30

There was radar. The Weather Bureau station at New Orleans ran a WSR-57, with a camera that photographed the radar screen once a minute, on sixteen-millimeter film. But nobody could watch it as a loop. The images were seen live by the operator, and then filed away.

Landfall, 11 PM 4:49

At about eleven at night on Sunday, August 17th, the eye of Camille came ashore near Waveland, Mississippi. The eye was small, about ten nautical miles across. The strongest winds were packed into a ring only six to eight nautical miles from the center.

And the whole storm was compact. Its outermost closed isobar reached only about a hundred and fifty nautical miles from the center. Camille was a very intense hurricane that was also very small, and that combination decides almost everything that follows about how it was measured.

Even the track changed later. The original best track showed Camille brushing southeastern Louisiana near the mouth of the Mississippi River before it reached Mississippi. The radar fixes show the center stayed just offshore. That landfall was removed in 2014.

A 24.6-foot storm tide 5:41

In front of the eye, the sea came in. At Pass Christian, the storm tide reached twenty-four point six feet, higher than any storm tide of record before it. Nothing would approach it until Katrina, in 2005.

This is the coast around Gulfport after the water went back, filmed by Navy photographers. Ships were left sitting on dry land. Along the coast highway, US 90, the shore was stripped down to slabs and splinters.

The wind instruments 6:10

Now, the wind instruments. The Weather Service office in Mobile states it plainly: the hurricane destroyed all the wind-recording instruments in the landfall area. And there were very few of them to begin with.

The highest sustained wind anyone ever measured in Camille came from a drilling rig in the Gulf: a hundred and thirteen knots, and then the anemometer failed. On a recorder left running on a rig east of Boothville, a gust of a hundred and seventy-two miles an hour was traced on paper, before the paper jammed.

On land, the strongest wind anyone recorded came from far inland. At Columbia, Mississippi, about seventy-five miles from the coast, the instruments measured a fastest-mile wind of a hundred and four knots, and nine hundred and fifty-eight millibars, about four hours after landfall. On the coast itself, no anemometer survived to say what came first.

Our measurement 7:03

We wanted to know what the official instruments read at the coast that night. So we went to the federal archive of hourly weather observations, and pulled every station on that coast that was operating in August 1969. The archive lists a station at Bay Saint Louis, nine kilometers from the landfall point. For those two days, it holds no observations at all.

The nearest station with a full record is Keesler Air Force Base, at Biloxi, forty-eight kilometers east. It reported every hour through the storm. Its lowest pressure was nine hundred and eighty-six millibars, at one in the morning. Its strongest wind, seventy knots.

Seventy knots is a hurricane, but it is less than half of what Camille carried at its core. Keesler was simply too far away. In a storm whose strongest winds sat within eight nautical miles of the center, twenty-six nautical miles might as well be another state.

A barometer in Bay St. Louis 8:01

Which brings us to Charles Breath. He was a boat dealer and a mariner, and he kept a marine barometer in his house, on a bluff just west of the bridge in Bay Saint Louis. As Camille approached, he logged it every hour, and then every half hour.

He also had a decorative wall barometer, with a scale that stopped at twenty-eight inches. That night, its needle fell right off the end of the scale. On the marine barometer, the last entry in his log read twenty-six point eight five inches: nine hundred and nine millibars.

By then the roof was coming off. When the eye arrived, the wind stopped, and the family walked about five hundred feet to another house they owned, on higher ground. Breath carried the marine barometer with him.

In the second house, he read it again. It had dropped further, to twenty-six point seven zero inches. Nine hundred and four millibars. That reading never made it into the log.

After the storm, the barometer was taken to New Orleans and tested, and the Weather Bureau office there found it accurate. In exchange, Breath was given another marine barometer, one whose scale went down to twenty-six inches.

There were other readings that night. In Pass Christian, about seven nautical miles from the center, a man named James Cagle read his home aneroid in the eyewall: corrected for its error, nine hundred and nineteen millibars. Another reading, of eight hundred and ninety-seven, was later set aside: the barometer could not be found, and the number looked rounded.

The 1969 record 9:36

When the official record was written, it took the nine hundred and nine from the log, and the hundred and ninety from the aircraft. Nobody now knows why the nine hundred and four was left out. The best guess is that someone went by the log, where nine hundred and nine was the last entry.

And the nine hundred and one from the aircraft quietly disappeared. A footnote in the season summary, months later, said only that a check of the raw data showed it should be corrected to nine hundred and five. Nothing in the files explains why.

The 2014 reanalysis 10:08

For forty-five years, that was Camille. Then the Hurricane Center needed an answer to a simple question: which hurricane was the strongest ever to strike the United States? In 2014, a team that included Margie Kieper, Chris Landsea and Jack Beven reopened the files.

They found the radar film. Two consecutive frames every ten minutes, from late morning on August 17th until after landfall, were copied from the New Orleans radar film. For the first time, Camille was assembled into a moving radar loop, forty-five years late.

Checking the 901 10:43

Then they looked again at the nine hundred and one. A dropsonde does not just report the bottom. It reports how high each pressure level sits as it falls. And in this one, the level of eight hundred and fifty millibars sat at six hundred and ninety-two meters.

The coded message carried five levels on the way down. Seven hundred millibars, at two thousand three hundred and ninety meters, sixteen point six degrees. Eight hundred and fifty, at six hundred and ninety-two meters, twenty-eight degrees. And at the bottom, nine hundred and one, with no height at all.

That height is too high. Camille's own good dropsondes, the ones that read nine hundred and five and nine hundred and eight, put that level at five hundred and fifty-one and five hundred and eighty-six meters. In Hurricane Wilma, a true nine hundred and one sat at five hundred and sixteen.

The team estimated the real surface pressure three different ways. They got nine hundred and eighteen to nine hundred and nineteen, then nine hundred and twenty, then nine hundred and nineteen. The sonde may simply have stopped transmitting before it reached the sea.

Our arithmetic 11:53

We redid the arithmetic ourselves, with the standard equation that links the height of a pressure level to the pressure below it. From the sonde's seven hundred millibar level, we get nine hundred and nineteen. From its eight hundred and fifty millibar level, also nine hundred and nineteen. The sonde's own upper readings contradict its last one.

And the double eyewall explains why the real number was higher. That afternoon Camille was replacing its eyewall, a cycle that briefly weakens a hurricane. The reanalysis puts it at a hundred and thirty-five knots at that moment. The crew saw the sea at the very time the storm was off its peak.

How 900 mb was analyzed 12:32

For landfall, the team went back to Charles Breath. His nine hundred and four was taken three to four nautical miles from the center, on the eastern edge of the eye. The center itself had to be lower. They analyzed nine hundred millibars.

Nobody read nine hundred. It is an analysis, built on a reading made in a living room, and on where that living room sat inside the eye. In 1969, Nash Roberts had reached almost the same number, nine hundred and one, in a manuscript that was never published.

The wind they assigned at landfall: a hundred and fifty knots, a hundred and seventy-five miles an hour. Down from a hundred and sixty-five knots. The peak in the Gulf came down the same way, from a hundred and sixty-five to a hundred and fifty.

The peak moved, too. The new peak in the Gulf sits on August 16th and early on the 17th, at the moments of the nine hundred and eight and the nine hundred and five, the two readings where instruments agreed. Camille was strongest when it was measured best.

Where Camille stands 13:37

So where does Camille stand now? The Hurricane Center ranks it between two others, as the strongest hurricanes on record at landfall in the United States. Above it, the Labor Day hurricane of 1935, at eight hundred and ninety-two millibars. Below it, Andrew in 1992, at nine hundred and twenty-two. Same Category Five as in 1969. Less wind, and a deeper center.

And the Hurricane Center is open about the limits. It considers its best-track intensities accurate to within about ten percent, and says the uncertainty is higher for storms of Camille's era. The true peak wind on that coast was never measured, and now it never will be.

The cost 14:22

Camille killed two hundred and fifty-six people. A hundred and forty-three died on the Gulf Coast. A hundred and thirteen died two days later in Virginia, where the remnants dropped up to twenty-seven inches of rain in Nelson County, most of it in a few hours.

What the record supports 14:38

Here is what the record supports. The official instruments on that coast were destroyed, too far away, or not there at all. The aircraft's most extreme numbers were an estimate made by eye and a dropsonde that failed its own check. And the lowest trusted pressure measured on land in Camille came from a barometer in a boat dealer's house, checked by the Weather Bureau, and nearly forgotten for forty-five years.

Next: Tuscaloosa, 2011 15:02

Next, a tornado that radar followed across eighty miles of Alabama. On April 27th, 2011, it crossed Tuscaloosa and the edge of Birmingham, with peak winds estimated at a hundred and ninety miles an hour, on the ground for eighty point six eight miles. That is the next episode. Tuscaloosa, 2011.

Description and sources

On Sunday afternoon, 17 August 1969, an Air Force WC-130 dropsonde in Camille's eye reported 901 mb, and the crew, seeing the sea in deep furrows, reported surface winds of 180 knots; the forecasters set the landfall intensity at 165 knots, 190 mph. In Bay St. Louis, Charles Breath logged 26.85 inches (909 mb) on his marine barometer, then read 26.70 inches (904 mb) in the eye after walking to another house; the 909 became the record. The 2014 reanalysis (Kieper, Landsea and Beven, BAMS 2016) showed that the 901 sonde's own 850 mb height (692 m) implies about 919 mb, analyzed 900 mb at landfall from the 904 reading at the eastern edge of a 10-mile eye, and set the wind at 150 knots. Our measurement: in NCEI's hourly archive, the Bay St. Louis station has no observations for 17-18 August; Keesler AFB, 48 km east, recorded at most 70 knots and 986 mb; and the sonde's 700 and 850 mb levels both give 919 mb by the hypsometric equation. Camille killed 256 people, 143 on the Gulf Coast and 113 in the Virginia floods.

PRIMARY SOURCES

• NOAA National Hurricane Center, 'Hurricane Camille reanalysis', news release, 1 April 2014. https://www.nhc.noaa.gov/news/20140401_pa_reanalysisCamille.pdf
• Kieper, M. E., C. W. Landsea and J. L. Beven, 2016: A Reanalysis of Hurricane Camille. Bull. Amer. Meteor. Soc., 97, 367-384. doi:10.1175/BAMS-D-14-00137.1 https://repository.library.noaa.gov/view/noaa/15209
• NOAA AOML Hurricane Research Division, Atlantic hurricane database reanalysis, 1969 (metadata, Camille '901' dropsonde analysis, Camille landfall pressures). https://www.aoml.noaa.gov/hrd/hurdat/1969.html
• Simpson, R. H., A. L. Sugg and Staff, 1970: The Atlantic Hurricane Season of 1969. Monthly Weather Review, 98, 293-306. https://www.aoml.noaa.gov/hrd/hurdat/mwr_pdf/1969.pdf
• NOAA NHC, HURDAT2 Atlantic hurricane database, AL091969 (Camille). https://www.nhc.noaa.gov/data/hurdat/
• NWS Mobile, Hurricane Camille. https://www.weather.gov/mob/camille
• NWS Middle Atlantic River Forecast Center, Flood of 20-21 August 1969. https://www.weather.gov/media/marfc/Flood_Events_2016/1969/Aug20-21,1969.pdf
• NWS Jackson, Hurricane Camille (deaths and damage). https://www.weather.gov/jan/1969_08_17_hurricane_camille
• NOAA NCEI Global Hourly (ISD), stations 747686-13820 (Keesler AFB), 722138-13835 (Bay St. Louis), 999999-12958 (NAS New Orleans), 17-18 Aug 1969. https://www.ncei.noaa.gov/access/services/data/v1
• NWS Birmingham, Tuscaloosa-Birmingham tornado, 27 April 2011 (next episode). https://www.weather.gov/bmx/event_04272011tuscbirm

ARCHIVE PLATES: all 9 are credited on screen in the video and listed in this project's manifest.

Hook Echo reads a disaster through the instruments that recorded it: radar, sensor, survey and report, with the primary sources above.

About the material. Video: three federal films on the Internet Archive: "Camille and the Seabees" (U.S. Navy, Naval Photographic Center, 1971), "A Lady Called Camille" (U.S. Department of Agriculture, Motion Picture Service, 1971) and Army Signal Corps footage of Gulfport and Biloxi, August-September 1969 (National Archives 33218). Documents: the NAS Pensacola radar photograph, the Air Force reconnaissance messages and the 1970 Monthly Weather Review page, from NOAA's Camille reanalysis files. Satellite: ATS-III and NOAA imagery, public domain via Wikimedia Commons. The station records and the dropsonde arithmetic are our own measurement from NOAA's public data. Every diagram is drawn in code from the sources listed. Nothing in this video is AI-generated imagery.

#HookEcho #HurricaneCamille #Hurricane