by Erik Larson
To be concerned, however, he first had to know that a cyclone even existed. All Halsey knew was that a nondescript tropical storm was at that moment arcing north into the U.S. mainland and soon would cross into the Atlantic.
To Halsey, it was fine, brisk day to be at sea.
STRAITS OF FLORIDA
A Matter of Divination
ON WEDNESDAY MORNING the storm rumbled into the Straits of Florida just north of Cuba and promptly confounded the Weather Bureau’s forecasters. Willis Moore and his professors believed the storm would now move north. To them, the storm appeared to have begun a long turn or “recurve” that would take it first into Florida, then drive it northeast toward an eventual exit into the Atlantic. No real evidence supported this projection. It was merely what the latest iterations of the Law of Storms predicted and what the bureau’s scientists expected based on the little that was known about tropical cyclones. In the age of certainty, at the gateway to the twentieth century, the expected was as good as fact. To turn was every storm’s destiny.
Shortly after noon on Wednesday the Central Office telegraphed a report to New Orleans that the storm “probably will be felt as far north as Norfolk by Thursday night and is likely to extend over the middle Atlantic and South New England states by Friday.” So far, the report said, “the storm has been attended only by heavy rains and winds of moderate force.”
The report contained some excellent news—the storm would “terminate the period of high temperature which has prevailed east of the Mississippi.”
IN HAVANA, WEDNESDAY, Julio Jover sent an 8:00 A.M. dispatch—by mail—to La Lucha: “We are today near the center of the low pressure area of the hurricane.”
Again, that dreadful word.
When William Stockman read Jover’s report, he surely laughed. He cut the report from the newspaper and affixed it to a special form designed by the Weather Bureau to help station chiefs collect praiseful articles from the nation’s newspapers and forward them to Moore. Stockman saw Jover’s report as further justification for the telegraph ban—it was another example of alarmist forecasting by the Cubans, who seemed to care more about drama and passion than science. Stockman did not consider the storm worthy of much further attention.
THE STORM AND its expanding cyclonic system now influenced a territory covering a million square miles of ocean and began to shape the weather in the southern United States. In Tampa, telegraph wires whistled. Winds reached twenty-eight miles per hour. In Key West, the barometer fell to 29.42 inches, the lowest level yet reported. The wind came from the northeast and accelerated to forty miles an hour, a true Beaufort gale.
Wednesday evening, however, the wind in Key West abruptly weakened. Its velocity dropped to six miles an hour, barely a breeze. Later that night, it began accelerating again, this time from the south.
The bureau’s forecasters believed the sudden easing of the wind and the attendant change in direction meant the center of the storm had passed over or near Key West, and saw this as confirmation of their belief that the storm would soon be traveling up the Atlantic coastline. Once again, they tailored fact to suit their expectations. They knew just enough to believe they had nothing to fear.
But the storm did not go north.
The bureau had missed the true meaning of the wind shift at Key West. Here was an area of calm immediately adjacent to a zone of gale-force wind, in a storm that had just crossed the great mass of Cuba without losing any of its size or energy or its ability to produce biblical volumes of rain. No one knew it at the time, but the conditions at Key West provided the clearest evidence yet that the storm’s architecture was changing.
At the storm’s center, centrifugal force had come to play—the same force that flings children off the rims of playground carousels. The winds spiraling toward the storm’s center now traveled at such a high rate of speed, they began to generate centrifugal force that sought to push them back out again. Where the inrushing and outpushing forces balanced, the winds began to form a circle, a gigantic carousel over the ocean.
This storm was about to open its eye.
THE NEXT MORNING, Thursday, at 6:00 A.M., William Stockman sent a dispatch that placed the storm 150 miles north, by east, of Key West.
It was a grave mistake, for it colored the expectations and perceptions of the bureau’s Central Office at a critical point in the storm’s journey. Stockman’s guess—and that’s all it was, a guess, armored in the certainty of the age—provided a framework into which Moore’s forecasters eagerly fitted other incoming observations.
Two hours later, the Central Office issued its 8:00 A.M. national weather map for Thursday, along with a prediction that “the storm will probably continue slowly northward and its effects will be felt as far as the lower portion of the middle Atlantic coast by Friday night.”
The Weather Bureau transmitted the map and its notes via the impossibly intricate web of telegraph wires that ran along every railroad right-of-way in the nation. The report caught the attention of fishermen in Long Branch, New Jersey, who cabled Washington: “Advise quick about storm unable decide about taking out nets.”
Moore liked messages like this. They showed that his efforts to increase the bureau’s credibility were beginning to pay off. The bureau’s own scientists had always believed that one day they would be able to make accurate long-range forecasts; the most enthusiastic hoped they might even learn to make rain and quash hail. It was the public that had always questioned the bureau’s competence. At last that skepticism was beginning to weaken. Many shippers, railroad agents, and cotton traders had grown as dependent on the bureau as the Galveston police had on electricity.
At 2:15 Thursday afternoon, Moore sent a reply to the Long Branch fishermen: “Not safe to leave nets in after tonight. Wind likely to increase from northeast beginning tomorrow morning.”
Moore’s telegram showed that the bureau was still convinced the storm was barreling north, bound ultimately for the Atlantic. The bureau had few hard facts about the storm, yet what is remarkable about its cables that day is the complete absence of doubt or qualification.
A week later, with Galveston in ruins, Cuba’s Julio Jover paid a visit to Colonel Dunwoody. Emboldened by disaster, Jover sought to confront Dunwoody on the telegraph ban, but the conversation expanded to include the efficacy of hurricane prediction.
As the interview gained heat, Dunwoody grew frustrated. He told Jover, “You had better go to the Belen College Observatory and there study a work which I wrote about meteorology—see if what I said about the prediction of cyclones is not a question of divination, as a cyclone has just occurred in Galveston which no meteorologist predicted.”
Jover, incredulous, paused a moment. He said, slowly, as one might address the inmate of an asylum: “That cyclone is the same one which passed over Cuba.”
“No sir,” Dunwoody snapped. “It cannot be; no cyclone ever can move from Florida to Galveston.”
KEY WEST
M Is for Missing
AT 7:00 A.M. Galveston time, Thursday, September 6, Joseph Cline made the station’s morning observations, coded them, and had a messenger carry the report to the Western Union office on the Strand, where it entered the great surge of weather details that crowded the nation’s telegraph lines that morning, and every morning. Joseph reported normal atmospheric pressure of 29.974 inches and a temperature of 80 degrees, markedly lower than the night before. The sky was clear and blue. Such fair weather must have been reassuring to Joseph and Isaac—the best evidence yet that the tropical storm was at that moment racing toward the Atlantic. Only much later, as meteorologists came to understand the strange physics of hurricanes, would such intervals of fair weather in the path of a tropical cyclone take on a more menacing cast.
In Washington, a legion of clerks at the Central Office processed the incoming blizzard of weather data and quickly constructed the morning’s national map, which the office then telegraphed back to every station in the country. Each station th
en added a local and regional forecast prepared by headquarters, set the map in type, and printed copies for distribution to newspapers, post offices, boards of trade, seamen’s taverns, and other public institutions.
The map that reached Erie, Pennsylvania, Thursday morning showed a vast low-pressure zone over the Pacific Coast. The base of the low stretched from Los Angeles to El Paso. From there, the low spread north to Spokane, Washington, and the Canadian border. But two high-pressure zones still held the rest of the nation’s weather in check. Temperatures again soared into the nineties in Cincinnati, Davenport, Green Bay, Louisville, Washington, and Chattanooga. Even in cool green La Crosse, Wisconsin, the temperature hit 94 degrees. A brief notation on the map read: “The tropical storm has moved from Key West to Tampa, Florida.”
In fact, the storm never did pass directly over southern Florida. Blocked by one of the high-pressure zones, it executed an abnormal left turn that put it on a course directly toward Galveston, eight hundred miles away across the superheated Gulf. The high pressure had caused a change in the seasonal pattern of winds sweeping off the Atlantic. Instead of blowing toward the northwest, these winds now blew mainly west, and carried the storm toward the Texas coast.
Only the storm’s outer bands reached Florida. The winds in Key West, Tampa, and Jupiter did reach gale force, but caused little damage other than knocking out the fragile telegraph link between Key West and points north.
Where the Thursday-morning weather map should have displayed temperatures for Key West, the Central Office inserted only the letter M, for missing.
GULF OF MEXICO
The Devil’s Voice
WHAT THE LOUISIANA’S thirty passengers must have thought Wednesday as the steamship passed the striking red-and-black storm flag at Port Eads, Louisiana, is anyone’s guess. For some passengers, no doubt, the prospect of a storm was an exciting one, just the thing to yield a good story to tell the friends and relatives who would meet the ship in New York the following week. Others took comfort in Captain Halsey’s obvious confidence. If there was any serious threat to the ship’s welfare, surely the captain would proceed no farther. A few passengers did not see the storm flag. They were seasick and already considered death an attractive option.
Once past the bar off Port Eads, the Louisiana accelerated. The muted booming of the ship’s steamplant became an even thrum. Smoke from her stack blew forward over the starboard rail in a long black smudge that flurried cinder upon the sea.
Captain Halsey ordered the decks cleared and hatches sealed, but the thought of turning back did not occur to him. He held the Louisiana to its southeastward course throughout the night, despite the rising wind and seas.
At 6:00 A.M. Thursday he checked the ship’s barometer and saw the mercury at 29.60 inches, nearly three-tenths below normal. The wind still came from the north-northeast, but at intervals circled until it came directly from the north.
The storm was a cyclone and by now Captain Halsey, veteran of so many such tropical storms, had to know it.
By ten that morning, the storm was much worse. The barometer dropped another third of an inch, to 29.25. The depth of the decline was troubling in itself, but the speed of the descent was what most captured Halsey’s attention. The first decline, to 29.60 inches, had taken all night. This latest had taken four hours.
Horizontal rain clattered against the bridge with the sound of bullets against armor. Wherever the wind gained entry, it spoke. It moaned among the cabins and corridors like Marley’s ghost. The hull flexed. Beams twisted. To the passengers, the ship seemed on the verge of disintegration.
At noon, Halsey ordered a sharp reduction in speed. He wanted only enough forward drive to let him maneuver and keep the ship’s bow pointed into the oncoming wind and waves.
The barometer continued sinking. At one o’clock, Halsey checked the glass and saw the mercury “had fallen to the remarkable figure of 28.75.” He had never seen a lower reading. He believed the Louisiana had arrived at the heart of the storm, for the wind now shifted wildly from one direction to another. “I do not like to speak of anything outside of the log record,” Halsey said, “but I think the wind was blowing at the rate of more than 100 miles an hour.”
Wave after wave washed the ship’s deck and thundered against the cabin ports. By now all thirty passengers were sick beyond fear. At one point a giant wave struck the ship from behind just as it slid into a valley between two other mountains of water. In an instant, the ship was buried bow to stern under tons of green sea and foam.
The Louisiana rose clear, her deck like the rim of Niagara Falls. Another wave caught the ship broadside and flushed seawater down her ventilation shafts into the engine room.
It was at this point that Halsey estimated the velocity of the wind at 150 miles an hour.
THE TRANSFORMATION WAS stunning: One moment a nondescript tropical storm, the next, a hurricane of an intensity no American alive had ever experienced. The storm did not grow through some gradual accretion of power; it exploded forth like something escaping from a cage. The Weather Bureau of 1900 had a code word for winds of 150 miles an hour—extreme—but no one in the bureau seriously expected to use it.
The storm had undergone an intensification known to late-twentieth-century hurricanologists as explosive deepening, but the Weather Bureau of Isaac’s time had no idea such a dramatic change could occur. As the twentieth century closed, hurricane experts still did not understand what caused it. There were theories, however. For a storm to grow so quickly, some researchers proposed, it had to encounter an additional atmospheric force—an upper-level vortex, perhaps, or a fast air-stream that somehow set the storm spinning more and more rapidly. Hugh Willoughby, head of the National Oceanic and Atmospheric Administration’s (NOAA’s) Hurricane Research Division, proposed that explosive deepening could be caused when a storm passed over the Loop Current, a branch of the Gulf Stream that propels warm water through the Straits of Florida.
The Loop may have been in place in the summer of 1900. The Gulf was hot to begin with because of ambient high temperatures and because so far in that season there had been no other hurricanes to roil and cool the waters. The Loop brought a deep channel of warmth that the wind and rough seas could not have cooled. If present, it would have been directly in the path of the storm of 1900 when it exited Cuba. “If a storm runs over the Loop,” Willoughby said, “it’s got essentially an infinite source of heat.”
No one knows whether crossing the Loop triggered the storm’s electric growth. What is certain, however, is that for the storm to have generated winds of the velocity reported by the Louisiana’s Captain Halsey, it had to have formed an open, circular core of extremely low pressure. Isaac and his peers in the Weather Bureau preferred to call this the focus, or center. They shunned the term eye, coined by the Spanish and used so freely by Spanish captains. It was too romantic, too anthropomorphic. In the age of scientific certainty, one could not allow one’s judgment to be clouded by mere poetry.
AT THE VERY center of the eye, the air is often utterly calm. Sailors throughout history have reported seeing stars at night, blue sky during the day. Often, however, the eye is neither clear nor cloudy, but filled with a liquid light that amplifies the stillness, as if the world were suddenly fused in wax. The sea, however, is anything but calm. Freed abruptly from the wind, waves from all quadrants of the eyewall converge at the center, where they collide and compound to form sudden mountains of undirected energy.
Sunlight playing in the eyes of cyclones produced colors that drove brave seamen to their knees. Captains reported olive-green clouds and a spectral blue light that stained sails and the faces of men until all seemed turned to ice. In 1912, the Reverend J. J. Williams of Black River, Jamaica, saw the sky begin to bleed. “Around the entire horizon was a ring of blood-red fire, shading away to a brilliant amber at the zenith. The sky, in fact (it was near the hour of sunset), formed one great fiery dome of reddish light that shone through the descending rain.”
The eyewall is an impossibly hostile realm where air flowing toward the center reaches its highest velocity. Observers trapped in a cyclone’s eye consistently reported hearing a great roar as the calm passed and the opposite eyewall approached. The frightened Malay crew of a ship off Sumatra called this chorus the Devil’s Voice. To Gilbert McQueen, commanding a ship bound for London, the eyewall sang its advance in “numberless voices, elevated to the highest tone of screaming.”
One of the strangest encounters with the eye was that of Capt. William Seymour, of Cork, Ireland, and his brigantine Judith and Esther, as the ship made for Jamaica in the summer of 1837. Seymour sailed into one of four hurricanes that scoured the Caribbean that summer within days of each other.
The storm knocked the ship onto her side three times, the third time just as the ship was leaving the eye. Once again the ship righted, but now something profoundly peculiar occurred that piqued great excitement among seekers of the Law of Storms. Lt. Col. William Reid wrote at once for more details.
Captain Seymour replied: “For nearly an hour we could not observe each other, or anything but merely the light; and most astonishing, every one of our finger-nails turned quite black, and remained so nearly five weeks afterwards.” He could not explain it. “Whether it was from the firm grasp we had on the rigging or rails I cannot tell, but my opinion is, that the whole was caused by an electric body in the element. Every one of the crew were affected in the same way.”
Such phenomena, however, were only sideshows to the most important feature of the eye, its plummeting pressure. Normal pressure at sea level is 29.92126 inches, or 14.6969 pounds per square inch. In the wall of the eye, spiraling and ascending winds lift air at over a million tons per second. As the air soars, pressure at the surface falls. Air within the eyewall rises with so much force it literally lifts the surface of the sea, one foot for each one inch of barometric decline. The lowest barometric reading ever recorded was 26.22 inches, during Hurricane Gilbert in 1988. Gilbert raised the level of the sea by over three feet.