Predicting Hurricanes: The Whirlwind of Controversy Surrounding Hurricane Alicia

Hurricanes will always be a way of life for many Texans. Young Texan schoolchildren learn about the Galveston Hurricane of 1900 in their classes; they hear their grandparents discuss Hurricane Carla. Every summer, a flurry of maps, supply lists, and a list of twenty-six names remind Texans that the season is upon them, and The Big One could be coming this year. Texans plan their lives around possible hurricanes, planning their weddings, vacations, and family events on the very real fact that a hurricane may decide to interrupt their good time. The middle of August, 1983, was no different; except this time, The Big One really was coming. On August 13, the residents of Houston, Texas and the surrounding areas had no idea their lives were going to be significantly impacted in the next five days. The residents, however, over the next five days, watched and waited as the Gulf of Mexico churned out one of the greatest weather-related disasters to ever hit the Houston/Galveston area.

Hurricane Alicia struck the area on the early morning of August 18, around 2AM.  Alicia is significant in her own right due to strength, destruction, and impact, but to the National Oceanic and Atmospheric Administration’s division called the National Weather Service, and their division of the National Hurricane Center, Alicia was a chance to test out their new landfall prediction system. The landfall prediction system, which was decades in the making but got its start with Alicia, significantly saved lives and the benefits of testing it during this Texas storm has made it possible to better predict where hurricanes will land in the future, possibly saving millions of lives. The prediction system of using meteorological data to guess landfalls was in the minds of meteorologists for at least a decade before the system debuted with Alicia.

The Monthly Weather Review, a journal published by the American Meteorological Society, had articles discussing what methods could be used for landfall prediction at least a decade before Alicia, if not more. Forecasts taken at the 24 hour mark and studied, according to Neumann in “Trends in Forecasting the Tracks of Atlantic Tropical Cyclones”, go all the way back to 1954 (2). Seldom does a year go by in the Monthly Weather Review during the 1970’s and early 1980’s without Neumann and Pelissier publishing a scholarly paper discussing the ins and outs of how a landfall prediction system could be achieved, what would be its pros and cons, and what it would mean for hurricane tracking and forecasting as a whole, and also the social impact. Considerations were made to every variable of the storm, including its placement in the ocean, and angle of the hurricane path. Neumann and Pelissier write “landfall forecasts are likely to be more accurate and accompanied by less overwarning around the Gulf of Mexico than elsewhere in the United States” (An Analysis of Atlantic Tropical Cyclone Forecast Errors, 1970-1979 1981).

The small group of scholars who assembled the working prediction system did not obviously expect such a storm as Alicia. Predicting something as unstable and unpredictable as a hurricane is naturally hard for the National Weather Service. Nevertheless, predictions, beginning with the first advisory regarding Hurricane Alicia are issued with every advisory. A hearing before the U.S. House Subcommittee on Natural Resources, Agricultural Research and Environment explains that the National Weather Service and their specific department of the National Hurricane Center issue “advisories (every six hours) and bulletins (between advisories when necessary) whenever a hurricane becomes a threat to the United States” (23). Once a hurricane makes landfall, it is said to “hit” the coast. The question of the landfall prediction is determining where exactly this will happen, which is, of course, hard. This difficulty can be seen in their definition of a “hit”. A landfall “hit” is when the eye makes landfall 65 miles from the point at which they forecast. For example, if a hurricane is forecast to hit Corpus Christi, TX, and the eye makes landfall not directly over Corpus Christi, but still within 65 miles, it is considered a “hit” by the National Weather Service.

When predicting a hurricane, many computer models are used to determine where the storm will go next. Many weather-watchers may remember these models being broadcast on the news during more recent hurricanes. When predictions began, there were significantly more models to look at, now there are about seven. When writing about landfall prediction when it was still in the planning stages, C.J. Neumann and Joseph M. Pelissier wrote in an article called “Models for the Prediction of Tropical Cyclone Motion over the North Atlantic” for the Monthly Weather Review that “typically, the availability of so many models leads to conflicts in the determination of an official forecast” (522). If prediction of a normal hurricane is not difficult enough, Hurricane Alicia was a storm that did not follow the rules. The hurricane exhibited several signs that are rare in most Atlantic hurricanes. Early on August 17th, the storm’s eye began to circulate in a looping motion, known as a cycloidal eye. “Hurricane Alicia Galveston and Houston, Texas August 17-18, 1983”, a report to the Congressional Committee on Natural Disasters explains that “cycloidal motions in hurricanes tracks are difficult to forecast and have only rarely been observed in the past” (p13). The report also says that the “long quasi-steady deepening from weak tropical disturbance to full hurricane strength is very unusual” (p18). The eye of Alicia also, according to the report, “exhibited a very unusual ‘double eye’ structure…just prior to landfall, and surprisingly again during the two hours after landfall” (21).

No matter what good or bad came out of Hurricane Alicia, there was a wealth of meteorological knowledge that was studied and gained from the storm. The Committee on Natural Disasters Report of the storm asserts “Alicia is one of the best documented hurricanes ever to effect the United States” (p3). While it was wobbly and unpredictable on water, on land, it was also a memory-maker. Hurricane Alicia is widely considered to be one of the most impactful storms to the Houston/Galveston area on record. The Army Corps of Engineers’ report on the storm, published shortly after Alicia, said that “in terms of damage and recovery costs… ALICIA was one of the most devastating storms of this century” (44). Alicia was quite dangerous in this sense, but she was more dangerous due to the fact she formed so close to land, experienced the rare hurricane signs mentioned above, and took an unpredictable northern turn less than two days before Alicia hit land. This twist in her path is perhaps what caused the most confusion and panic.

Alicia appeared to be heading mostly to the west/northwest until about 5pm on Tuesday, August 16th. Also at this time, the advisory was updated to announce that Alicia had reached hurricane strength. Starting at 5am Wednesday, about 18 hours before she hit landfall, Alicia took a northern turn that baffled meteorologists. Still, the shift at first was not dramatic enough to make anyone worry. The Natural Disaster Committee’s report states “anyone projecting the path of the storm…at 5 pm Wednesday would have expected Alicia to make landfall around Matagorda (80 miles southwest of downtown Galveston…)” (p124). As the turn became sharper and sharper until Alicia was basically heading due north, the residents and city officials began to realize that turning at this point was the worst thing Alicia could have done.

Neumann and Pelissier wrote in 1981 that “Experience [studying the path of hurricanes] has shown in most cases, an effective compromise between accuracy and timeliness is reached if warnings are issued about 18 h[ours] before landfall” and adding a crucial piece that “12 h[ours] of which occur during daylight” (1264). Hurricane Alicia followed this plan, but her sudden shift at just 18 h before landfall put the previous predictions in the dust and her rapid strengthening caught everyone by surprise. The shift put an already nervous Galveston in terrible danger, but forecasters still planned the hurricane to continue at a mostly west-northwest motion. The Committee on Natural Disasters’ Report states “there is no obvious explanation for this change in the storm’s track” (14). This sudden shift put unexpecting Galveston and Houston residents in a dangerous position—given the new forecast with Alicia’s track, and the strengthening from barely a Hurricane to a Category 3, residents had to seriously consider evacuation, it was nearly too late to evacuate.

It was only after the last landfall prediction had been made at 5am Wednesday that Alicia’s winds were clocked at 100 mph and the storm was deemed “dangerous”. The Natural Disaster Committee Report notes that “when Galveston officials reached the time for a decision, the storm was a weak hurricane, intensification to category 3 was not expected” and further notes about the landfall location that “the most likely landfall location was forecast to be well to the southwest of Galveston” (p 129). The report goes on to say “when the storm became dangerous Wednesday afternoon, it was too late to initiate a large-scale evacuation of the island, and the predicted path was still to the south anyway” (p 129). An anonymous editorial in the Houston Chronicle published shortly after the hurricane said “it cannot have helped to have had Alicia characterized officially as having the lowest possible potential for disaster during much of the period when people are deciding what to do”, and adds “we think the weather service unfortunately tends to undercut its position by indicating a degree of dangerousness” ("Quantifying Hurricanes is a Risky Business" 1983).

The Army Corps of Engineers emphatically states in their report of the storm “its course was erratic, making landfall location predictions difficult” (6). When it came time to predict Alicia, she was first issued an advisory by the National Weather Service on 5pm Monday. The hurricane had only a 9% chance of hitting Galveston, still the highest possibility on the chart, but only slightly higher than the chance of hitting Lew Iberia, Louisiana (7%). When the next forecast came out at 11am Tuesday, Galveston leaped to 17%, the highest percentage, but still relatively low, at least, nobody got terribly worried. Galveston’s percentage climbed with the next advisory, but still remained under 36%, what would cause most of the Houston/ Galveston residents to be skeptical. Robert Lansford, a state coordinator of the Texas Dept. of Emergency Management said, “Citizens are going to see the low percentages and say ‘why worry about it’ when they should be listening to local officials about what they should be doing” (Snyder 1983). Only the 5am Wednesday report put Galveston anywhere above 50%--a measly 51%, and then the National Weather Service announced suddenly at 11am Wednesday that their landfall “probabilities were being discontinued because preparations for evacuation should already have begun” (U.S. Committee on Natural Disasters 1984, 124).

Under the confusion of an unpredictable storm suddenly strongly forecast to hit the area, residents of the Houston/Galveston area paid close attention to the National Weather Service’s meteorologists and their local officials for advice of whether or not to evacuate. The report to the Congressional Committee on Natural Disasters touches on the readiness of the Houston media in the preparation for Alicia. “The Tuesday editions of all three major newspapers…carried front-page stories about Alicia,” (pg 125) and the report goes on to mention “all the television stations displayed the probability distributions along the coast and attempted varying degrees of explanation” (pg 126) but added that “The National Weather Service had hoped…that the media would simply report the “total” probability values for places rather than…for various time increments” (p 125). This is not, however, what the media did. They put up the raw data they received from the National Weather Service and with little or no explanation and this caused confusion and grief among the residents. An article about the new landfall prediction system in the Houston Chronicle shortly after the storm quoted Steve Huffman, the city manager of Galveston, saying the odds “caused more problems with citizens….This is the first storm that (people) had to deal with the odds system”. He goes on to describe the reactions of the citizens. “They didn’t understand what it meant. It added to the number of phone calls we had to answer and took up our time trying to explain.” (Alicia gave center the chance to test system of laying odds 1983).

With every update by the National Weather Service, new landfall predictions were being broadcast on television and circulated in newspapers. This caused confusion among the residents, who were still unsure whether or not to evacuate. Clouded with numbers they didn’t understand and a ticking clock to disaster that got closer every minute, the residents of the Houston/Galveston area looked to their city officials. City officials, busy watching the storm and answering worried calls, also were unsure about whether or not to call the evacuation, because of the relative low strength of the storm until the hours right before landfall and questions as to how much trust they should put in the new landfall predication system. City officials were stuck on the decision of where (or when) to draw the line between causing unnecessary, early fear and giving a warning that was advanced enough that city residents had enough time to properly prepare their houses and evacuate.

Earl Baker at Florida State University did a research study in 1995 in which he studied the effects of a county during the coming of a hypothetical hurricane. His findings after he surveyed the county’s population was that “by far the most important variable [in the residents’ decision to evacuate or not] is local officials’ advice or orders” (7). Some cities decided to go ahead and order evacuations, just to be safe, while others decided to wait, believing the storm was not dangerous enough, until the storm did become dangerous enough and it was too late to order evacuation. The Natural Disaster Committee tells in their report “Kemah officials…made no evacuation recommendations one way or the other”, probably due to uncertainty or lack of adequate concern for the damage Alicia would cause, “while nextdoor Seabrook sent vehicles through neighborhoods with loudspeakers urging evacuation” (129).

Within their minds, along with the landfall predictions for Alicia, was the memory of Hurricane Allen. Galveston was especially criticized with waiting too late to decide to evacuate, and then not being able to call an evacuation because of time. The Natural Disaster Committee’s Report explains that “one reason Galveston…officials were reluctant to play it safe and suggest an early evacuation was their belief that residents were unhappy about the early…evacuation officials had urged during Hurricane Allen.” During the threat of Allen, the report says, “65 percent of Galveston’s residents responded…making a ‘three-day weekend’….Allen came ashore south of Corpus Christi, making the evacuation of Galveston ‘unnecessary’” (p 132). Texas residents will also remember a similar incident with Houston residents not heeding evacuation orders during the next big hurricane to affect Houston, Hurricane Ike, because of their personal experiences with the traumatic evacuation of Hurricane Rita that turned out to be also unnecessary.

Because of this, the evacuation orders for the different towns along the Houston/ Galveston coast were mixed. Galveston never ordered an evacuation for the City of Galveston, on the Eastern side of the island, just the West Side of the island. Approximately 90% of the people left the West Side. Perhaps the evacuation would have been predicted to go smoother and require less time if Galveston had had an evacuation plan. Stephen Huffman, Galveston City Manager, speaking at the Subcommittee of Natural Resources hearing said “an overall evacuation plan was ‘desperately needed’” (61). Mayor E. Gus Manuel of Galveston was later criticized for not evacuating the island. An article in the Houston Chronicle in the days following Alicia introduces “Fletcher Harris, Jr., [Galveston’s] former Civil Defense communications coordinator” who ranted to the paper “I have never seen such a catastrophe of misunderstanding and stupidity” (Olafson 1983). Pielke and Pielke, in their book Hurricanes: Their Nature and Impact on Society, remember Hurricane Alicia as one of the memorable storms along with 1989’s Hugo and 1992’s Andrew, and adds into their work on Alicia “landfall at Galveston came as somewhat of a surprise as the storm was poorly forecasted” (24).

Another article in the Houston Chronicle soon after the hurricane, written by Joan Beck, tells that “Government officials also can disagree as to what safety precautions are necessary”. She uses the City of Galveston as the prime example of bad judgment: “Just before Alicia struck, for example, Texas Gov. Mark White requested that the 60,000 residents of Galveston be evacuated”. However, she reports, the warning was not heeded and “Galveston officials decided not to do so” (Beck 1983). Galveston officials, under the heading of Mr. Manuel, either chose to do nothing or were indecisive for so long, they ran out of time to make a decision that was effective. This confusion from new information and tough decisions about evacuation are most likely why, in the Subcommittee of Natural Resources’ Hearing, numerous reputable people and city leaders praise the system for saving lives, and then the report discreetly and quietly adds “Mayor E. Gus Manuel of Galveston was not so certain as the other witnesses of the current value of the new probability forecasts” (p 33). Manuel told the hearing “I am not so sure of your percentage deal as of yet” (52). Manuel also told the hearing “it takes us about 36 to 40 hours to evacuate the whole island.”

The Committee on Natural Disasters’ report states, however, “studies have calculated that evacuation of [Galveston] island has to begin as much as 26 hours before landfall” (129). When Mr. Andrews, a committee member, asked if there should be someone else appointed for the purpose of deciding when to evacuate, Mayor Manuel is recorded in saying “you do have that already, if you want. The Governor has that authority. In turn, he has appointed mayors to make that decision…” (56). Most of the people who were interviewed by the various committees and the newspapers afterwards were positive of the land-falling system. Dr. Neil Frank, a noted weather scholar and long-time head weatherman for KHOU-TV in Houston said of the system that “Hurricane Alicia was almost the nightmare we have come to fear in meteorology…” but says also that “the initial feedback [of the probability system] from Alicia suggests the probabilities were quite useful” (12). The Subcommittee on Natural Disasters’ report said that “if anything, the system’s ‘accuracy’…might contribute to overconfidence in the system’s ability to ‘predict’ where future storms will strike” (134) and “the system appeared to function well” (134).

In conclusion, the experience of Hurricane Alicia for the National Weather Service and their new landfall prediction system is one that is praised for the most part, but the system did have its critics. So many things in the Hurricane Alicia experience were against the system working—the sudden northerly twist about 18 hours before landfall, right as predictions stopped, the mishandling of the information by the media, the confusion of the public, and the nonchalant attitude by different cities in the Houston/Galveston area until it was too late to be strongly proactive. However, the system overcame all these things to be an effective tool that is now used along with every storm that threatens the United States coast since Alicia.

While strides have been made and models have been sharpened with the new inventions in technology, the overall system of landfall prediction, thanks to its beginnings with Alicia, is one that saves numerous lives with each storm and allows the public and city officials early warning and more understanding of the storm affecting their area.  

---

"Quantifying Hurricanes is a Risky Business." Editorial. Houston Chronicle, August 1983.

"Alicia gave center the chance to test system of laying odds." Houston: Houston Chronicle, August 18, 1983.

Army Corps of Engineers. Hurricane ALICIA 15-18 August 1983. Military Report, Gavleston: Corps of Engineers, 1983.

Baker, Earl J. "Public Response to Hurricane Probability Forecasts." Professional Geographer, May 1995: 137.

Beck, Joan. "Improving warning systems for storm seem to still be in the hit-or-miss stage." Houston Chronicle, August 1983.

Neumann, Charles J. "Trends in Forecasting the Tracks of Atlantic Tropical Cyclones." Bulletin of the American Meteorological Society, October 1981: 1473-1484.

Olafson, Steve. "Galveston Mayor gets blame, praise for hurricane role." Houston Chronicle, August 1983.

Pelissier, Charles J. Neumann and Joseph M. "An Analysis of Atlantic Tropical Cyclone Forecast Errors, 1970-1979." Monthly Weather Review, June 1981: 1248.

—. "Models for the Prediction of Tropical Cyclone Motion over the North Atlantic: An Operational Evaluation." Monthly Weather Review, March 1981: 522.

Pielke Jr., and Roger A Pielke Sr. "Huricanes: Their Nature and Impact on Society". New York: John Wiley and Sons, 1997.

U.S. Committee on Natural Disasters. Hurricane Alicia, Galveston and Houston, August 17-18, 1983.

U.S. Government Report, Washington, D.C.: Washington Academy Press, 1984.

U.S. House Subcommittee of Natural Resources. Hearing to the Subcommitee, U.S. House Subcommittee of Natural Resources, United States House of Representatives, Washington: U.S. Government Printing Office, 1984.