Chapter One
The Perfect Storm
The perfect storm is as predictable as it is inexorable. Born in the Atlantic Ocean, it hits Puerto Rico and Hispaniola and Cuba, and it grows bigger as it moves through the warm waters of the Gulf of Mexico. Though there is plenty of time to flee, many residents along the Gulf Coast stay put. And just as predicted, this storm makes a straight track for the tiny camp town of Grand Isle, Louisiana, obliterates it, and moves north toward New Orleans.
The hurricane moves upriver for nearly sixty miles, leaving catastrophe in its wake. It passes right over New Orleans, and as it does, the storm tilts nearby Lake Pontchartrain like a teacup and dumps it into the city. A quick rush of brackish water drenches New Orleans and leaves it sitting in as much as twenty feet of water. And then the hurricane is gone, and everything lies in ruins.
The perfect storm is big enough to make New Orleans a certain kind of hell, but not so big that it makes first responders throw their hands up in despair. The floodwater is the worst of it—it collects in the lower parts of the city and takes weeks to pump out. As it sits, the water becomes a thick and fetid mash of household chemicals and dead things and gasoline that bubbles from the tanks of thousands of submerged automobiles and service stations. The water makes some people ill, but the worst is the complication it adds to the rescue efforts.
All told, the water and wind brought by the hurricane damage some 250,000 homes and turn a million residents into vagabonds, many of whom are now utterly dependent on the government for food and shelter. The storm kills tens of thousands of people outright and leaves the city virtually uninhabitable, downing all communications systems and paralyzing the infrastructure. After the storm passes, looting breaks out. And thousands of dazed and dying survivors sit on their roofs in the semitropical sun awaiting rescue. Though the tales of heroic rescue are numerous and inspiring, many people perish, waiting for help that doesn't come.
Some call Hurricane Katrina the perfect storm. It wasn't. The perfect storm, which the Federal Emergency Management Agency (FEMA) calls Hurricane Pam, exists only on a computer screen, the creation of a small federal contractor located in a nondescript office park on the outskirts of Baton Rouge. Developed in the spring of 2004 over a period of fifty-three days and at a cost of $800,000, Hurricane Pam is a low-tech affair, nothing more than a simulated computer storm surge that plays out on a monitor accompanied by a stack of descriptive documents that catalog the damage the storm wrought when it made its fictitious landfall.
Hurricane Pam is a training exercise, designed to get local and federal disaster responders thinking about how they might deal with the aftereffects of a catastrophic storm that hit New Orleans. Louisiana is lousy with emergency disaster plans, and its various government agencies have invested millions of dollars cooking them up. The city of New Orleans has one specifically for hurricanes, as do all of the parishes (counties) along the coast. Inland, the rest of the state's sixty-four parishes have created generic plans to deal with a wide variety of calamities, natural and man-made. Not to be outdone, about twenty state agencies have disaster plans. Some have several.
The federal government, as well, has dumped hurricane plans on the state over the years, and they are all hundreds of pages long, thick with appendixes and crammed with dense, jargon-filled prose. Most of them were cooked up in Washington by small teams of bureaucrats; a few were created without any local input at all. Most sit unread in disaster offices throughout southern Louisiana. In the office of Jesse St. Amant, the longtime emergency preparedness director for Plaquemines Parish, the collection of disaster plans on his long, low bookshelf stretches for several feet.
St. Amant's favorite federal plan is Response 95. Unveiled by FEMA in May 1995, the plan's debut was spoiled when a wandering rainstorm dumped twenty inches of water on the city of New Orleans as the exercise was taking place. This rainfall "of biblical proportions," as the local newspaper described it, swamped the city in waist-high stormwater. Though FEMA struggled mightily to fill a hotel ballroom downtown with local disaster planners, it was forced to cancel the event when the area's first responders phoned in hasty regrets. "I told FEMA that real life always trumps an exercise," St. Amant said with a chuckle.
As the head disaster planner for what is unarguably Louisiana's most vulnerable parish—Plaquemines juts eighty miles into the Gulf of Mexico and is completely surrounded by low-lying marsh—St. Amant believes Response 95 may have been the biggest bust of all, since any potential readership it might have achieved was washed out with the spring rain in New Orleans. But nearly all of the scores of state and FEMA training documents met a similar fate: The common practice among governmental bodies in Louisiana was to accept such studies without comment, agree to adopt them by unanimous vote, and store them on a shelf, along with the budget books and other effluvia of local bureaucracy. "Nobody ever actually reads them," St. Amant said.
But the Hurricane Pam scenario was a plan with a twist. Officially called the Southeast Louisiana Catastrophic Hurricane Plan, the Pam exercise made the readers—local emergency responders—authors as well. Instead of sitting first responders down in a ballroom and playing a cookie-cutter "wargame" scripted by some Washington contractor, Pam took a bottom-up approach, inviting the participants to take a crack at writing their own game plan for coping with the "Big One," down to the grittiest detail. Though guided by FEMA, the plan was created by the men who would wear the hip waders and man the flatboats, the medics and doctors who would operate the triage centers, and the cops and city workers who would be out on the street in a real disaster. Pam was what is known in the emergency response business as a planning exercise, where participants are fully briefed about a catastrophe and then draw up a blueprint for how they would cope. The rules were simple: Players can only make plans with the resources they possessed at the moment. For example, if a firehouse had five engines but two were always rotated out of service for maintenance, then the firefighters could only plan to respond with three. Anything else was unrealistic and destined to fail.
Although Pam was billed as part of a new drive by Washington for "catastrophic planning" in a bad new world of international terrorism, the impetus for the exercise was really Hurricane Georges, a rather puny hurricane when it hit the Gulf Coast in September 1998. Though small, Georges killed beyond expectations, taking some 600 lives as it rampaged through a procession of Caribbean islands before tacking into the Gulf of Mexico and taking dead aim at New Orleans. But just before making landfall, Georges defied forecasters and swung sharply to the east, veering into Biloxi, Mississippi, and causing scattered damage from New Orleans to Mobile.
If ever a city dodged a bullet, New Orleans did when Georges veered east. The hurricane prompted a massive and chaotic evacuation and would have been a serious killer had it hit New Orleans on the perfect path it had been taking. Georges made Louisiana disaster officials realize they were woefully unprepared for the "Big One" of Gulf legend. The city hadn't been hit square by a hurricane for more than a generation, and its first responders were beyond rusty.
Georges had laid bare the fundamental insufficiency of the state's emergency hurricane plan: There was no coordination between state agencies and parishes, as local governments followed their own plans for dealing with the storm. Some parishes called for mandatory evacuations while others did not, some parishes opened shelters while others failed to do so, and each parish had an idea for when it was proper to tell citizens to evacuate, which it didn't bother to communicate to any other parish. The result was gridlock—both literally and figuratively. There had to be a better way.
The father of Hurricane Pam may well have been a man named Colonel Michael L. Brown (no relation to the man with the almost identical name who headed FEMA in 2005), the former deputy director of emergency preparedness in Louisiana. In August 2000, Colonel Brown had written a twenty-page letter to James Lee Witt, the FEMA director at the time, requesting money for a plan that would simulate the effects of a massive hurricane hitting New Orleans and would help locals develop a "post-devastation" schematic for rescuing survivors and cleaning up. To this point, improbable as it may seem, the federal government had no plan on hand that specifically focused on dealing with the aftereffects of a catastrophic hurricane.
"We know that a Category 3, 4 or 5 hurricane striking the mouth of the Mississippi River at New Orleans would be a disaster of cataclysmic proportion," Brown wrote in his letter to Witt. After reciting the expected destruction—seventeen feet of water in New Orleans streets, up to 5,000 dead, as many as 300,000 people stranded in their homes—Brown got to the point: The state of Louisiana was in no position to respond to such a cataclysm, and that would likely mean death for many citizens. "In the aftermath of such a catastrophe, there will be an offshoot of life-threatening situations that will swiftly deplete our resources and absorb whatever limited time we might have to rescue those who survive," Brown said. "We believe that the level of response required to sustain, protect and rescue survivors during such post-hurricane devastation is well beyond what we conceptualize as ‘the worst-case scenario.'"
In quick and terse fashion, Brown sketched a rough outline of what was needed. There would be a scenario presented of a mock hurricane, a storm that would hit New Orleans in just the right spot to create catastrophe. With this scenario in hand, disaster officials would present it to first responders and stage a series of conversations, where they assessed the damage, mounted a coordinated search-and-rescue effort, identified shelters for evacuees, and moved down the line to recovery. And throughout the exercise, participants would be challenged to answer the practical questions of disaster response for themselves, instead of being handed the answers from on high. How would emergency responders communicate? How long would rescue teams work before taking a break and who would take their place? Where would they get boats, and from whom? Who would provide food and water? Brown's exercise aimed to answer these questions in the most minute and granular fashion. And the answers would come through discussion and argument with locals and state responders, not from Washington officials 1,100 miles away.
The FEMA brass liked the idea. They wrote a few letters, chewed on it a bit. And Brown's idea went nowhere.
The following year, a new administration set up in Washington, and Brown tried again, resending his letter to Witt's successor, Joseph Allbaugh, shortly before the 2001 hurricane season began. Only this time he sent it through the state's powerful U.S. senator, John Breaux. But again, nothing happened. Then came the terrorist attacks of September 11, 2001, and the danger of natural disasters got eclipsed by the threat of Islamic jihadist madness. Once more Brown's idea fell by the wayside.
After 9/11, the vogue in disaster management, not surprisingly, became terrorist events of the darkest sort that could be imagined: nuclear suitcases detonated on crowded streets, poison-gas attacks with crop dusters, mail sabotage with biological agents, suicide bombers targeting shopping malls—what were known as "low-probability, high-consequence" events. As concrete barriers went up around Washington office buildings, bureaucratic obstacles flew up around spending money on preparedness that wasn't connected to a shadowy terrorist threat. A congressionally mandated series of terrorist drills, carried out in desultory fashion before 9/11, now grew to stratospheric levels of grandness. Hurricane planning went out the window, unless a state was able to disguise it as terrorist training—as some states did.
The Department of Homeland Security, which was created by Congress in 2002 and swallowed FEMA along with about two dozen other federal agencies, started cooking up disaster exercises far more spectacular and lavish than the relatively modest plan Colonel Brown was championing down in Louisiana. They were sequels to the congressionally mandated Top Officials exercise, or TOPOFF, a $3 million, classic large-scale, military-style drill in May 2000 that had featured lavish pyrotechnics and complicated physical maneuvers. It called for staging a full-scale mock terrorist attack that unfolded simultaneously in three different cities. In the first city, Denver, terrorists staged a biological attack using a plague agent, while Portsmouth, New Hampshire, experienced an assault with mustard gas. Meanwhile, in Washington, D.C., terrorists exploded a bomb that was laced with nuclear material. The elaborate exercise, involving one thousand officials, was at the time one of the biggest ever conducted outside of the Pentagon. But it would look like a nickelodeon compared to the Technicolor extravaganza of its sequel, TOPOFF2, staged by the Department of Homeland Security in May 2003, when terror fears were at their zenith. Carried out with eight thousand state, local, and national officials and costing a whopping $16 million, TOPOFF2 put the dirty bomb in Seattle and socked Chicago with the biological plague. It featured live explosions, fake news segments, and the use of hundreds of "victims" moaning realistically in the streets and in hospital emergency rooms.
TOPOFF2 was a primarily a stage show. But it highlighted serious problems. An internal government report of the five-day drill said that the nation's emergency-response system was hampered by the failure of government agencies to share information, by uncertainty over the chain of command, and by confusing new government procedures. The upshot: America may be not much better prepared to deal with a big terrorist attack than it was before 9/11. "Fortunately, this was only a test," said the report. "However, if a real incident occurs before final procedures are established, such unnecessary confusion will be unacceptable." The report added that because the drill was simulated, "the full consequences of the confusion"—including the possibility of needless civilian deaths—"were not observed."
FEMA was a reluctant participant in the TOPOFF2 exercise. First, TOPOFF2 was run by a hated rival agency, the Office of Domestic Preparedness, which was also part of the new Homeland Security behemoth. But more important, FEMA officials viewed it as overscripted and too top-down in its approach to the response. Moreover, it didn't address the more likely scenario of a calamity caused by Mother Nature.
Later in 2003, FEMA began to revisit the subject of the New Orleans hurricane scenario, which many in the agency believed was desperately needed but still unfunded by the Department of Homeland Security. But this time, an official from the White House's Homeland Security Advisory Council sat in on these meetings, and he heard the agency's frustration over not having the resources to exercise a killer hurricane scenario. "He was astonished that as of that date we had not completed this type of plan and [he] promised to do what he could," said Sean Fontenot, a Louisiana disaster official. Four months later, FEMA got the money for its bottom-up disaster-planning experiment. The agency let a contract in May 2004, and began work immediately on what would become Hurricane Pam.
Oddly enough, the White House's interest in staging the exercise rose directly out of its obsession with Osama bin Laden.
In December 2003, just over two years after the twin attacks on New York and Washington, President Bush called on the Department of Homeland Security to draw up a list of priority concerns. The White House was eager to show that this lumbering department it had just created was capable of prioritizing and focusing on disasters that presented the greatest risk to the nation. The aim of the project was to identify the worst disasters that could happen and then come up with a plan of preparation.
The result was a series of "Planning Scenarios," a fifty-five-page list of fifteen doomsday events that could visit the United States and cause major fatalities. In keeping with the Bush administration's focus on terrorist attacks, twelve of the fifteen scenarios dealt with shadowy international groups bent on doing the nation harm. For the purposes of these scenarios, the shadowy group was called "the Universal Adversary," and it was imbued with capabilities that exceed those of many countries.
In Scenario 8, the Universal Adversary ruptures a train tanker carrying chlorine gas by detonating a small bomb underneath it. The resulting leak kills 17,500 people and hospitalizes 100,000 more. In Scenario 13, the same group uses anthrax to spike food simultaneously at a West Coast packing plant and at an orange juice factory in the South. After the food is distributed, some 300 people die. In Scenario 7, it releases sarin nerve gas in an office building, killing 6,000 people. And in Scenario 5, the Universal Adversary commandeers a crop duster and sprays a mustard gas mixture over a packed college football stadium.
Scenario 1 is the ultimate low-probability, high-consequence nightmare scenario. In this instance, the Universal Adversary builds and detonates a ten-kiloton nuclear device in Washington, D.C. The scenario begins with the theft of highly enriched uranium from a nuclear plant in the former Soviet Union. It ends Hollywood-style, with a mushroom cloud that blooms over the city and drifts slowly east-northeast. In this scenario, it is assumed that many first responders will die. "Decontamination, disposal and replacement of lost infrastructure will cost many billions of dollars," the planning document says. The scenario doesn't attempt to calculate the number of fatalities that will occur, but it does calculate that "an overall economic downturn, if not recession, is probable in the wake of the attack."
By contrast, Scenario 10 is less sexy yet far more probable. In this story line, a massive hurricane hits a major southern city, which also happens to be a popular tourist destination, just like New Orleans. A twenty-foot storm surge with accompanying rain overwhelms the local levee system and drowns low-lying areas. More than a thousand people die outright, in part due to a flawed evacuation scheme. Thousands more residents are left homeless. Local storm shelters are overwhelmed. A 95,000-ton oil tanker impales itself on a bridge and begins to leak. Municipal utilities fail, as do communications systems. Food and water are in short supply, and local 911 switchboards are overwhelmed by nuisance calls from people seeking lost pets.
"There are severe economic repercussions for the whole state and region," the planning document says. "The impact of closing the port ripples through the country. The loss of petrochemical supplies could raise prices and increase demand on foreign sources."
Guiding all of these scenarios was the Office of State and Local Government Coordination and Preparedness, a body within the Department of Homeland Security that doles out grant money to "detect, prevent and disrupt" terrorist attacks, but rarely deals with natural calamities. For the most part, these scenarios are simple script treatments. They do not describe a plan of response. They are simple "what if" story lines that would be familiar to any casual reader of airport potboiler novels.
The shortcomings of the scenarios deeply bothered Eric Tolbert, who was chief of FEMA's response division. Tolbert was an old-line disaster responder who had begun his career in the field as a paramedic and had worked his way up to being the state director of emergency management in his native North Carolina. Like many of FEMA's old guard, he believed the agency was spending too much time worrying about terrorist events—a relative rarity—and not enough time on the disasters that break out with far more regularity: the fires and floods and earthquakes that torment nearly every region of the country.
Tolbert and FEMA's new director, an Oklahoma lawyer named Michael D. Brown, had long been pressing the White House, Secretary of Homeland Security Tom Ridge, and Congress to spend money planning for real-world disasters—the natural ones that hit all the time and, most of all, the really big ones that hit rarely but with devastating consequences. In 2004, they got the money and permission to begin developing plans for the natural disaster scenarios on the national scenario list. The appropriation was a relative pittance—only a few million dollars out of Homeland Security's $31 billion overall budget. But it was a start.
And Tolbert had no doubt about where to start. "When I have a nightmare," he said, "it's a hurricane in New Orleans." And so the planning began.
Tolbert was hardly the only person having bad dreams about a storm hitting New Orleans. The city hadn't been hit dead-on by a hurricane since Betsy in 1965, and the last one to even seriously menace the city was Hurricane Camille in 1969, which hit the Mississippi shoreline about fifty miles east of New Orleans. In the last 120 years, the city has averaged one hurricane hit about every decade, and "with Camille hitting over 30 years ago, we are well overdue for a major one," Paul Trotter, the National Weather Service's local chief, told a reporter in May 2004.
Moreover, much of the city's vaunted levee system, mandated by Congress to provide no more than Category 3 hurricane protection, was untested. The U.S. Army Corps of Engineers, which was in charge of all New Orleans levees, had seen its local budget slashed repeatedly by the Bush administration—by some $80 million in 2005 alone. And millions of acres of surrounding wetlands—a natural buffer to storm surges—had melted away under intensive development by energy and shipping concerns.
In 2004, a University of New Orleans researcher predicted that even a hurricane of moderate strength would devastate the city, causing loss of life and property damage eclipsing that of even a major West Coast earthquake.
Hurricane Pam was intended to address a threat that was well known to everyone. FEMA brought together hundreds of people from dozens of federal agencies and the military to sit down with state troopers and school superintendents and volunteer firefighters to hammer out the details of response, drawing clear lines of authority and responsibility, and calculating just exactly what resources would be needed. Though the federal government paid for Hurricane Pam, the locals were the ones who brought the plan to life, hanging it with the minutiae of the first response, working through the logistics of evacuation and preparation, and crafting the government's initial reaction to a storm that everyone knew was coming.
Innovative Emergency Management Inc. (IEM), the Baton Rouge consulting company hired to build the computer models that brought Pam to life, deliberately avoided special effects. Rather, the company sought to turn disaster planning on its head by creating a believable scenario and then inviting the participants to grapple with its effects. The hurricane would be little more than a handout. The drama would unfold as disaster officials discussed how to cope. The officials would be segregated into groups to discuss smaller pieces of the disaster plan. Every day, leaders of these groups would deliver progress reports to top state and federal officials. All the sessions would be recorded and then transcribed and cleaned up. In the end, the free-flowing plan of action would appear in a three-ring binder, ready for review in advance of the next hurricane season.
Though low-tech was the guiding principle, a great deal of research went into making the Hurricane Pam scenario believable. The last thing the planners wanted was for the training sessions to break down into arguments over whether the hurricane as described would cause the damage as stated. IEM's technicians called in a levee expert from Louisiana State University and got him to help them design the perfect track for their New Orleans storm, one that would sink the city under a uniform, seventeen-foot storm surge that would reach deep inside southeast Louisiana and bring damage to thirteen parishes. The company also consulted with the National Weather Service until it got what it considered to be the perfect storm—a hurricane that approached from the southwest, tracked up the Mississippi River, and crossed to the east, passing directly over the city and soaking everything in rain.
Over fifty-three days, IEM built Hurricane Pam, focusing on the small details of an exercise that would involve as many as 270 disaster responders. Greg Peters, an IEM contract worker, noted that the company fussed relentlessly over its guest list. "You know, they'd sit around asking questions like ‘Do we need someone from the phone company to be here?'" he said. "The answer was yes and so they'd move on: ‘How about the Department of Transportation?' and so on."
But the center of the plan was always this perfect storm, a few "words and tables," as IEM's president and chief executive, Madhu Beriwal, put it—words and tables calculated to stand up to assault, should someone want to quibble with the scenario. Brad Tiffey, IEM's chief manager for the Hurricane Pam drill, worried endlessly about the storm's precise path. "It's funny how big of a difference five or ten miles can have on a hurricane's effect," he said. Ultimately Tiffey and his technical staff settled on a tried-and-true storm track, the one traveled by Hurricane Georges, the very real storm that had sparked the creation of Pam in the first place. The only difference was that this time the tempest did not veer to the east.
But Tiffey and his staff also recognized that Pam could not be too big or too overwhelming. "They wanted a plan to arise from all of this, not plot the Apocalypse," Peters said. So, as they fashioned their storm, IEM's technical staff departed from the Hollywood script but also from the traditional mind-set of disaster planners that the best way to practice is to present the absolute worst-case scenario. FEMA's original Pam disaster imagined a massive Category 5 hurricane bearing down on Louisiana with winds of 160 miles per hour and a twenty-foot storm surge. But the designers worried that in the face of such an unusually large hurricane, the locals "would throw up their hands and say, ‘We can't cope with this,'" said Madhu Beriwal. So IEM took a different tack; it created a smaller storm, a Category 3 storm, which wrought plenty of havoc but did not leave the city vaporized. Notably, Hurricane Pam did not breach the city's flood protection system, though its storm surge easily overtopped the levees and swamped many city neighborhoods.
There was a second reason for making Pam milder than a superstorm. It would help bust the myth that the "Big One" was the only storm worth fretting over. "We decided to let the participants see that a slow-moving Category 3 could be just as devastating as anything out there," said Tiffey. In truth, only three Category 5 hurricanes have hit the continental United States since records have been kept. Smaller hurricanes are far more common and often just as deadly. And so it became that Pam was no monster tempest but a slightly oversized Category 3 storm, with winds of 125 miles per hour. What set Pam apart was the soaking amount of rain it carried (up to twenty inches in some spots), its lumbering speed, and its unerring track toward the most populated city in Louisiana. In the scenario, the National Weather Service predicts New Orleans will be hit by Pam, and that's exactly what happens.
Outside of the scenario, the plan was a skeleton that the players would fill out themselves. There were fifteen sections to the plan, covering everything from rescue efforts and medical care to clearing the city's streets of water and debris and handling hazardous waste spills. For each section, there was to be a workshop where locals would consult the scenario, draft a list of supplies and manpower, and then fire questions at each other, just as Colonel Michael L. Brown had envisioned it.
As the locals talked out the problems and refined their techniques in the workshops, IEM stenographers would be on hand to refine the ideas and put them into serviceable prose. The result would be a full plan of action for the "Big One" as well as a plan that could be employed for smaller disasters as well. And the plan would be a living document, adding lessons learned as future hurricanes hit and undergoing continual refinement and revision. Ultimately, IEM intended to feed the plan into a hurricane simulation program to see exactly how effective it was, which would in turn spark more refinements.
But most important, the Hurricane Pam plan would be absorbed by the very people who needed it most. "A lot of times people don't read the plan," Beriwal said. "The intent of this planning exercise was to engage."
On July 16, 2004, the first day of the exercise, the Louisiana Emergency Operations Center started filling with disaster officials. The building added an element of realism to the drill—it is a near-perfect storm center, filled with computers and phones and big plasma screens, as well as a stage area from which reports can be delivered. A smaller cadre of local officials attended these pre-landfall sessions, laying out evacuation plans and readying the Louisiana Superdome as the city's shelter of last resort, while an official from the National Weather Service issued faux bulletins and flashed slides on a projection screen. IEM activated the joint command center, where state and federal officials took daily briefings and coordinated the overarching plan of response. One of the most unusual experiments in disaster planning had begun.
On the third day of the exercise, the room filled to capacity as Pam made landfall while about 270 emergency workers looked on. Representatives from the Pentagon, the Coast Guard, and FEMA mixed with National Guard officers, state disaster chiefs, and emergency personnel from dozens of parishes and the adjoining states of Arkansas and Texas. Pam made its predicted low-tech landfall: The model played out on a projection screen as IEM employees handed out documents describing the havoc Pam had wrought on its slow churn through Louisiana. And then it was over.
One by one, IEM officials read out the damage tabulations. In the scenario, Pam's storm surge crashes into the city and easily tops its floodwalls. The gush of water, coupled with driving rain, fills New Orleans and the surrounding suburbs like a cereal bowl. The floodwater turns the city into a grim Venice, swelling its lone evacuation center well beyond what officials can immediately handle. About 50,000 seek shelter in the Louisiana Superdome before the storm hits; many thousands more come afterward. In total, some 500,000 area residents are left at least temporarily homeless.
New Orleans being a city of ambivalent evacuees who have grown used to watching hurricanes veer from predicted paths at the last minute, IEM assumed that evacuation rates would be low. In the scenario, only 65 percent of the metropolitan area evacuates, including a mere 34 percent from the city itself. Nearly 600,000 people decide to ride out the storm by IEM's projections—not an unreasonable figure, given Louisiana's evacuation track record. But riding out the storm proves foolish: 61,290 people in the thirteen-parish area perish due to wind or water. About 20,000 of that number are in New Orleans.
Pam destroys 462,000 housing units and 4,020 businesses; thousands of other buildings are damaged. The water smothers the city's drainage system, knocking out 80 percent of the drainage pumps, including many of the older ones that run on an obsolete electrical current and will be difficult to repair. By IEM's calculations, many of the pumps will be out of commission for six months. It will take almost a month to rid city streets of floodwaters.
Some 30 million cubic yards of debris clutter the city, which is sure to tax the available landfills and slow the search-and-rescue efforts. Survivors sit on roofs and in attics all over town. Local officials will need to make more than 200,000 rescue runs, which they are incapable of doing on their own. Nearly 60,000 hungry people sit in government-approved shelters.
There was an overriding caveat built into the Hurricane Pam scenario: No matter the rhetoric about who is in charge of recovery or who has responsibility for carrying out certain tasks, local officials would almost certainly be unable to fend for themselves. "The response capabilities and resources of the local jurisdiction . . . may be insufficient and quickly overwhelmed," the disaster scenario said.
After the damage readouts, the 270 officials broke into groups that were based on their specialty. A group charged with draining water out of New Orleans included officials from the Army Corps of Engineers and the local levee board, while a search-and-rescue planning team included officials from the U.S. Coast Guard and the Louisiana Department of Wildlife and Fisheries. There was a group in charge of shepherding food, water, and generators. Another group discussed temporary housing, while yet another laid plans for triage centers to take care of the thousands of injured.
And so what unfolded was something quite different from the usual "war game" exercise that FEMA and the Pentagon had always used to stage their drills. War games are far more scripted, high-concept exercises often played on very realistic mock sets. But the low-tech Pam generated tremendous passion, perhaps because it had such a real-life application.
In the search-and-rescue session, for example, participants figured that they would need 308 boats, 800 body bags, 400 flashlights, 150 paddles, and 12 spare bilge plugs for small craft. "I was like, whoa, fellas, twelve boat plugs? We're getting down in the weeds here," said Jesse St. Amant, of Plaquemines Parish. But getting down in the weeds was precisely the point: Madhu Beriwal believed that a great failure of FEMA's previous disaster plans was that they didn't have enough detail, precisely because they were cooked up by people in Washington who have no clue as to what a boat plug is, for instance, or how many might be needed.
It was the same in other sessions. The supply distribution group calculated it would need about 40,000 volunteers to staff local shelters, while Army Corps officials calculated that they would need to be able to deliver 1.53 million gallons of water and 5.5 million pounds of ice to the area each day. The debris removal group determined it would take around two years to clear all the ruination. And a group charged with drying out the city estimated that it would take two months to complete.
Rather than joining any specific group, FEMA officials attended all of them to lend advice and offer technical help. IEM was in every room as well, to record the conversations for its written plan and to keep the participants talking.
IEM employees were also there to enforce the cardinal rule, as outlined by Colonel Brown. "No fairy dust," he said, and what he meant was this: If a job called for 300 boats, participants would have to find those boats and not just wish them to exist. If planners needed fifteen semitrucks to haul generators to New Orleans, they had to identify where they would get them, or at least make a realistic guess at the source. "They were supposed to plan with the resources that were available or that could presumably be brought in," said Beriwal. "They were not supposed to be thinking that magically 1,000 helicopters would show up and do this."
In retrospect, there was fairy dust, much of it dispensed by FEMA, though most participants didn't realize it at the time. Greg Peters said the conversation would often stutter to a halt in the breakout groups over some question over basic supply: Who would provide a daily supply of, say, bottled water for the thousands of expected evacuees or ensure that triage units would have ample supplies of bandages? "It happened a lot—the conversation would stop over something like generators or ice and a FEMA guy would say, ‘Look, don't worry about that, we've got contracts in place, you'll get your million gallons of water a day or whatever,' " Peters said. "It was almost like they were bragging."
Indeed, the Hurricane Pam plan is shot full of FEMA's promises: bedding on hand for 100,000, mobile communications centers, even video uplinks to establish teleconferences over great distances. Later, when FEMA was actually put to the test, it would have trouble supplying even the most basic requests—flashlights, for instance, appeared to be beyond the capability of the agency, let alone the staggering amount of water and ice specified by the Army Corps of Engineers.
Despite FEMA's empty promises, the Pam exercise could be considered at least a qualified success. That was due in large part to the fact that the locals had clearly read the plan, if not lived it. Peters and others said that the local officials got excited as they hammered out the details of their Pam response, carrying their discussions and arguments into the bars of Baton Rouge long after the sessions were over and rising in anger to defend a particular plan of action. "I know some of the search-and-rescue guys almost came to blows while I was there," Peters said.
Though Beriwal doesn't know whether fistfights broke out, she agrees with the premise that passion ruled. On the eighth and final day of the exercise, after the sessions were over and IEM workers huddled in the emergency operations center to begin talking over how to turn the drill into a document, Colonel Brown, the state's deputy disaster chief, ambled over to say his good-byes.
"He spoke to us about how meaningful and important the exercise was," Beriwal said, "and he was so overcome by emotion that he left the building. He didn't finish his sentence. He was crying."
Rather than an end unto itself, the Pam exercise was supposed to mark an initial dialogue, the start of a continuing discussion over how to cope with a disaster nearly everyone believed would soon befall New Orleans.Many of the elements of the Pam plan remain little more than bold-stroke sketches: The question of reentry to the city by returning residents was barely discussed, for example. Although follow-on workshops were held in November 2004 and July and August 2005, questions regarding temporary housing were never fully worked out. Local officials never got the opportunity to ask the hard questions about FEMA's prepositioned stockpiles of supplies. A discussion on the inevitable necessity of moving evacuees from short-term shelters such as the Superdome was given only the briefest of consideration in the Pam drill. And questions about securing the storm-wracked city from looters and lawbreakers went similarly unaddressed in the eight-day planning session. Even so, IEM did manage to cover some ground in the follow-on workshops. It nailed down details of temporary medical care, supply logistics, and temporary housing for storm victims.
But in a breathtaking display of penny-wise planning, FEMA canceled most of the follow-up sessions scheduled for the first half of 2005, claiming it was unable to come up with money for the modest travel expenses its own employees would incur to attend. FEMA officials have since said that the shortfall amounted to less than $15,000."Homeland took the money from us so we couldn't take any steps at all to address the gaps," FEMA director Brown recalled.
Pam remains a work in progress and very much incomplete. Only about a third of its fifteen sections have gotten any attention at all from workshop groups. But in subsequent disasters in Louisiana, the locals have excelled in the areas that were covered during the Pam workshops. That's particularly true of evacuations, which improved markedly in the state following the Hurricane Pam drill.
Of course not even Pam imagined the ultimate doomsday scenario—the utter collapse of New Orleans's flood-control system. Indeed, very few of the doomsday documents that sparked the plan's creation mention the possibility that a storm surge could topple the city's levees.
Copyright © 2006 by Christopher Cooper and Robert Block
