September 07, 2005
SOME KATRINA LESSONS: We're going to see a plethora of commissions and inquiries (most about as useful and non-partisan as the 9/11 Commission), but here are a few lessons that seem solid enough to go with now:
1. Don't build your city below sea level: If you do, sooner or later it will flood. Better levees, pumps, etc. will put that day off, but not prevent it.
2. Order evacuations early: You hate to have false alarms, but as Brendan Loy noted earlier, even 48 hours in advance is really too late if you want to get everyone out.
3. Have -- and use -- a plan for evacuating people who can't get out on their own: New Orleans apparently had a plan, but didn't use it. All those flooded buses could have gotten people out. Except that there would have had to have been somewhere to take them, so:
4. Have an emergency relocation plan: Cities should have designated places, far enough away to be safe, but close enough to be accessible, to evacuate people to. Of course, this takes coordination, so:
5. Make critical infrastructure survivable: I think that one of the key failures was the collapse of the New Orleans Police Department's radio system. Here's the story on why:
Tusa said the police department’s citywide 800 MHz radio system functioned well during and immediately after the hurricane hit New Orleans, but since then natural gas service to the prime downtown transmitter site was disrupted and the generator was out. Transmitter sites for the police radio system “are also underwater with the rising water and [are] now disabled,” Tusa said.
Owners of the sites that housed police radio transmitters would not allow installation of liquefied petroleum gas tanks as a backup to piped gas, meaning generators did not have any fuel when the main lines were cut, Tusa said.
Radio repair technicians attempting to enter the city were turned away by the state police, even though they had letters from the city police authorizing their access, Tusa said.
This is absurd, and I'm pretty sure it's the major factor leading to the disintegration of the New Orleans Police Department. That sort of gear should be survivable -- and there should also be a backup plan for how to get messages back and forth if the radios go out anyway: Messengers, broadcasts on commercial radio, etc.
(There should be a separate post-disaster communications plan for survivors, too -- so that they can locate relatives and let people know they're alive).
Other crucial infrastructure should be hardened as much as possible, too. There's only so much you should do, but disaster survivability should be considered at every stage of design, procurement, and construction.
6. Stock supplies and prepare facilities: The Superdome didn't have adequate food, water, and toilet facilities, even though everybody knew it was going to be a shelter of last resort. The Convention Center was worse. All public buildings that might be used for refugees should be ready. We used to stock fallout shelters that way; we could do it again.
7. Be realistic: Here's what the Los Angeles Fire Department tells people about an earthquake aftermath:
To those of us who live and work in the Greater Los Angeles area, earthquakes and other natural emergencies are a reality. In order to deal with this situation, emergency preparedness must become a way of life. In the event of a major earthquake or disaster, freeways and surface streets may be impassable and public services could be interrupted or taxed beyond their limits. Therefore, everyone must know how to provide for their own needs for an extended period of time, whether at work, home, or on the road.
That's just how it is. People need to be encouraged to do this. Whenever I say this, I get responses along the lines of "poor people can't afford to stockpile food." But here's a family survival kit for $50 and it's pretty good. Most poor people in America can afford food (that's why so many poor people are fat). They do have other problems that make preparation less likely, though (if you're the kind of person who thinks ahead and prepares for emergencies, you're much less likely to be poor to begin with) and local authorities have to be ready -- see the stockpile advice above.
8. Put somebody in charge: Politicians and bureaucrats thrive on diffusion of responsibility, because it helps them escape blame (as they're trying to do in the fingerpointing orgy that's going on now). Somebody needs to be clearly in charge. Right now it's mostly state governors, but this needs to be made inescapably plain, regardless of where it is. I don't agree with Mickey Kaus that we should ignore federalism and just put the President, or the FEMA Director, in charge and empower them to override state and local officials, but even that would be better than leaving no one in charge.
There's much more to be done on this topic, but it awaits clearer information on who dropped what balls when. However, it's worth noting that structural problems are always soluble when the people involved are willing to cooperate, and that no structure works well when it's staffed by idiots or people who don't take the problem seriously. Which raises another point:
9. Make people care: Actually, Katrina may have done this. Most people -- and politicians are worse, if anything -- have short time horizons. Disasters are things that just don't happen, until they do. Planning for them is ignored, or even looked down on, often by the very same people who are making after-the-fact criticisms that there wasn't enough planning. People usually get better after a big disaster, for a while. Beyond that, voters and pundits need to treat the subject with the importance it deserves instead of -- as is more typical -- treating it as the silly obsession of a few paranoid types.
I'm sure there's a lot more to be learned, but this is a start. If you think I've missed something important, send me an email.
UPDATE: Aaron Taylor emails: "I'd add: Err on the side of overwhelming law enforcement presence."
Yes, and show zero tolerance for truly lawless activity. The "broken windows" theory applies in spades, I think, when windows are already broken . . . .
And reader Deena Bevis emails:
Clear chains of command are definitely essential, but so is oversight/accountability. New Orleans didn't have any of that until it failed. We need a system that tells us if someone in that chain of command is failing to complete their responsibilities, and we need to know that BEFORE something happens.
Basically: States and the feds should grade each other on disaster preparedness, and those reports ought to be public.
I'm afraid log-rolling and backscratching might interfere, but it's a thought.
ANOTHER UPDATE: A reader emails:
I read your post on lessons learned from Hurricane Katrina. I am a nuclear engineer working at a Midwest nuclear plant. We are required to have emergency plans. They are relatively detailed and many aspects are regulated. This includes communications, getting information to the public, recommendations to take shelter or evacuate, and coordination with federal, state, and local authorities. We are required to perform drills and the Nuclear Regulatory Commission grades us annually on one of them.
I'm not sure how well all of these steps would scale up to a large geographical area or the legality of grading states and localities on how well they execute their emergency plans. But people are acting as if this is an entirely new concept and it isn't. Why do we need another worthless commission to tell us what we already know and some of us already do.
Jim Hogue writes:
Maybe it's time to put that little Civil Defense logo (or something similar) back on AM/FM radios so people will know exactly where to tune in the event of an emergency?
And speaking of "In the event of an emergency" I haven't heard anything about the Emergency Alert System in relation to Katrina. Was it on? Did it work? Did it provide any useful information? I would think that a system that's been tested weekly since the 50's would have been pretty reliable.
Beats me. Emily Bennett has another communications question:
I find myself wondering if passenger cars equipped with OnStar could be used for communications in an emergency situation. OnStar constantly advertises its ability to get emergency personnel to its subscribers, and it seems to me that an ambitious FEMA or Homeland Security employee might begin talks with the OnStar folks to see if OnStar equipped vehicles could help manage evacuation traffic flows, provide communications to rescue personnel, and assist some of Bill Whittle's sheepdogs.
Probably not enough bandwidth and switching ability, but I could be wrong.
Reader Jay Johnson emails:
Having made it through the F4 tornado that blew through Jackson, TN in 2003 relatively unscathed, brought the importance of having an emergency kit such as that to light for the missus and me. We did go to our friendly, neigborhood "Everything's a Buck" store, and stocked up on things like cheap canned meals (beef stew, soup), dry foods, matches, water, batteries, cheap flashlights, copies of important papers, a change of clothes, a sealed container with purely emergency cash, some rudimentary tools (hammer, phillips and flat screwdriver, adjustable wrench, and a couple of pocket knives), and cheap first aid kits. It doesn't cost much, and an ounce of prevention is worth the extra peace of mind that comes from it.
Of course, nothing can completely prepare you for such an event, but everyone should do something to prepare for their short term survival in this spot.
Indeed. Reader Jeanne Jackson makes a point that seems trivial but isn't, in light of experience:
One important item you missed is providing evacuation plans for citizens with pets. One reason many people remained behind in New Orleans was that the emergency shelters barred pets, as did the buses, etc. for transporting evacuees. For many pet owners, especially childless and/or older people, pets are surrogate children. It is cruel, heartless, and unnecessary to insist, as a condition of rescue, that one's beloved dog or cat be abandoned to its fate. Were I to be told I must abandon my dogs in order to get out of a life-threatening situation, I, too, might choose to remain behind and take my chances.
I think you should leave the dog behind. But lots of people feel differently, and evacuation plans should recognize that.
ANOTHER UPDATE: Reader Harvey Schneider has some excellent advice for individuals:
One of the things my family has done is designated a contact person in the event of a catastrophe. My entire family (Mother, 2 Brothers, 2 Sisters in law, sister, brother in law, 4 nephews and 2 nieces) lives in Orange County California. We have designated a family friend in Phoenix Arizona as the person for everybody to contact as soon as possible and to leave any messages regarding health or other vital matters. Also, in the event that the entire area is unlivable, we have agreed to meet at our friends house in Phoenix and make further plans from there.
Good point. Meanwhile, Jim McMahon writes:
I would add another player to your list of lessons learned - listen to the insurance companies.
They assess and manage risk for a living. Since insurance rates are based on risk of loss, the easiest scorecard available to judge how well any particular area is prepared is the cost of commercial and homeowners insurance.
For disaster mitigation purposes, I would suggest an expanded system similar to car crash test ratings, where the disaster risk of a community or neighborhood is scored based on the lessons you list, including:
The inherent environmental risk of the area (flood plains, forest fire fuel, earthquake susceptibility)
Preparedness of the local & state gov - budgets, experienced people, drills, publicity
Individual preparedness based on site inspections, nature & number of voluntary associations, etc.
The costs of maintaining and replacing infrastructure.
These ratings can then be used as promotional tools by highly rated areas, and as cattle prods during elections. I would expect that most of the components of these ratings are already compiled and ready to use, needing only the imprimatur of the Fed.
Then FEMA's job becomes one of ongoing improvement of ratings in high-risk areas. They can grant tax breaks for real estate developments and local policies that improve the risks, and restrict the use of highway funds to prevent the construction of the Foghorn Leghorn Memorial Bike Path, Library and Fan Cub until the essential infrastructures are at their target ratings.
I'm sure some wags & wastrels will have issues with this, but show me another low-government way to honestly rate how well the lessons have been learned.
Well, insurance companies do have money riding on the outcome, which encourages honesty.
Christopher Johnson has more thoughts:
I would only add that churches ought to be urged to stockpile both food and other emergency supplies for people who don't have or can't immediately access their own. Many churches are very strong buildings, they're in just about every neighborhood and people aren't afraid of them. I know that if I needed immediate help to get me through the next few days after losing everything I owned, I would much more likely to try a church for help than to take my chances with a government bureaucracy.
MORE: Reader Jeff Cook emails:
5. "Make critical infrastructure survivable: I think that one of the key failures was the collapse of the New Orleans Police Department's radio system."
No, sorry. The collapse was in incident command.
It is axiomatic (lesson #1) that the first thing to fail in ANY emergency is communication. The NOPD incident command training should have taught them this. There is no way to assure that radio communications will continue after winds and power outages. This is the kind of thinking that has Blanco and Nagin in their bunkers giving orders and then wondering why they weren't carried out instantly. No one was listening. My experience has been that even seasoned dispatchers, who may or may not have power and transmitting ability themselves, have a hard time keeping channels clear in an emergency. I've seen communication break down during DRILLS.
This is why you need AT LEAST 72 hours notice for evacuation and why the NOPD should have default posting positions and "runners' assigned in the event of communications failure. There is no fail-safe communications system and there never will be. If they harden this technology for floods and hurricanes, will they survive an EMP? a nuclear device? Well, dammit why not??!! Who throws the switch from natural gas to lpg? How long does lpg in the tank last? Who refuels them? Are the refuelers available during a hurricane? Is it in our response plan to throw the switch? Is the switch thrower a police employee or a private contractor? Do they know their responsibility? Is the switch thrower even still employed? Answers to these questions can never be known for any extended period, especially when elected officials try to be the incident commanders. What can be known is that communications fail. Always.
Plan, plan, plan, practice, review, plan, plan, plan, ad nauseum.
They also appear to have forgotten lesson #2. "It is always easier to scale back than to scale up once the emergency has begun."
I've heard the words "incident command" and "unified command" exactly once each in the mainstream media since the blame-laying began. That tells me that all the really knowledgeable people are too busy to comment right now, and haven't been interviewed yet.
Finally bear in mind that emergency response and incident command is very, very, very difficult even in the best circumstances, which never exist.