November 12, 2013

Deploy Resources on Forecast, NOT on Disaster

Haiyan represents another human tragedy caused by severe weather.   And it highlights again the weakness in the way mankind responds to such disasters.

Days before Haiyan's landfall, the forecast models showed the threat.   Uncertainty was low as the best models honed in on the solution.  Some warnings went out, but the major players waiedt until death and destruction occured before initiating a major response.  The storm hit, victims were injured and desperate, and help did not begin arriving in force until 3-5 days after the event.  Many die, others sicken, looting begins, and the situation deteriorates until the second week after the storm.

Sounds familiar?  This kind of scenario is SO familiar, from Katrina to Haiyan and a many storms in between.

We can do much better.   Mankind, and particularly the U.S., needs to deploy on forecast, not deploy on disaster.  We can do this now for the simple reason that weather forecasts are hugely better than even a decade ago.   Sandy was a good example of our increased prediction prowess, and there are many more.  So why begin to deploy relief  AFTER the disaster strikes, but before?  And have the capability to move in with massive resources immediately after the storm passes.

Consider Supertyphoon  Haiyan.   Forecasts for landfall on the Philippines of an intense typhoon were made days before.  The best numerical weather prediction forecasts had stabilized on a very threatening solution early in the week.   For example, here are the 48 and 96 hr forecasts of sea level pressure and 850 hPa (aournd 5000 ft) wind speed for 48 and 96 hr before landfall from the European Center models.   Both forecasts had the right position for a very strong storm.  The threat was clear.

To quote Dr. Jeff Master's WeatherUnderground blog on December 5th, three days before landfall:

"Both the GFS and European models predict that Haiyan will hit the central Philippines between 3 - 6 UTC on Friday, and Haiyan will likely be the most dangerous tropical cyclone to affect the Philippines this year.

With a high probability of a huge supertyphoon heading for the extremely vulnerable Philippine coast, the world waited when it should have acted.

The U.S. decided to deploy the carrier George Washington and its group, which was  in Hong Kong, on November 11th, more than TWO DAYS after the storm hit.   According to Pentagon information it should take 48-72 hrs to be on station.   They should have deployed in the days before the storm's landfall, waiting a safe distance away until they could move in quickly.  Deployment of other aid should have begun at a similar, pre-landfall time.

Better late than never.

In many ways, aircraft carriers are the perfect delivery system for help in such events.  They are stand-alone cities with huge resources.  Helicopters and VTOL aircraft that can get into the most difficult conditions.  Large medical facilities.  The capability to hold vast supplies of food, water, generators, and other necessary equipment.  Unmanned and manned aircraft that can reconnoiter the damage and provide critical information. 

Let me go even further.  Imagine if three of the U.S. carrier groups were given the primary missions of providing relief for major disasters.  Yes, they would retain substantial armaments to be available for the national defense, but they would be supplied with massive supplies and experienced personnel to deal with  large disasters.   To push this idea further, one might station these carrier groups in three positions that might be most useful for disasters: off of SE Asia, NW of Africa, and in the Caribbean (I put C's at the suggested positions).   With two or more days warning for major storms, they could be in position in time.   For earthquakes, tsunamis, and volcanic eruptions they would move immediately after disaster strikes. Europe and Australians can take care of themselves.

 Can you imagine how many lives such an approach could save?  Can you imagine the good will it would engender for the U.S.?  Instruments of war being used as instruments to protect life and property around the world.   Too idealistic?  Perhaps.  Possible to do?  Without a doubt.

But carriers or not, my basic idea is simple.  Weather forecasts (and ancillary forecasts like waves and water level) have gotten much better during the past decades.  For most weather disasters, the 2-3 days forecasts will be very, very good.  If meteorologists believe uncertainty is low, then the U.S. government, UN agencies, non-governmental organizations, and other countries should begin deployment no less than 48h BEFORE weather disasters strike.  Have the capability to rush in help and supplies as soon as severe weather clears.

There is a lot of talk about the world having to deal with more severe weather under global warming.   But the "Inconvenient Truth" is that mankind is unable to handle the disasters of today.  Better forecasts, greater resources for disaster relief, and predeployment could have huge positive impacts for our fellow humans faced by terrible natural disasters.


  1. I couldnt agree with your more. Deploy before not after. At worse, its an exercise. Humanitarian relief. We can do this...lets not wait. We can be the harbinger of good will...cheaper then other forms of aid.

  2. It does seem odd that we don't take this action, especially in this age "of an abundance of caution" ( phrase that makes me want to vomit every time I hear it). I support your idea. Thanks for making the case.

  3. Great ideas here! It probably makes too much sense for politicians though.

  4. Is it 2016 yet? I foresee a Mass write-in campaign in the future (pun intended)...

  5. Fact is this was done for Sandy thanks to collaboration between NOAA and FEMA. See the bottom of numbered page 4 into page 5 here. It would be a good model for the international community to anticipate pre-positioning resources when it's clear in advance a country's resources will be overwhelmed. Maybe there is a place for a "meteorologists without borders" group.

  6. This is the "warn on forecast" philosophy and it's long overdue. It seems only meteorologists understand how good NWP has gotten in the last decade. Instead of investing in incremental NWP improvement now, we should be devoting resources to convincing decision makers to act proactively.

  7. I don't like paying for the military but I would be happy for some of my tax dollars to be spent this way, with one caveat: rebuilding efforts MUST take into account rising sea levels and the increasing severity of storms. Build to a better code, don't have people living in shacks 2-3' ASL as many people in the subtropics do.

    If we can't build everyone decent housing -- and let's face it, we don't even have the will to try in the US -- at least build substantial storm shelter so that people can escape their flimsy housing to a place that will probably protect them.

    They're always going on about "three days, three ways" and encouraging US to prepare in advance of natural disasters; I've never understood why that doesn't extend to governments. Why on earth didn't the mayor of Tacloban -- poor man, he has clearly seen horrors since the storm -- encourage people to get the hell out and/or to move to appropriate shelter if they couldn't actually leave the city? Apparently, many people actually came TO Tacloban, thinking that the city had safer buildings wherever they were. So sad.

  8. I have read your blog about issues like this and the future of education etc. and clearly you are very passionate and call the government out on failures(which I agree are many). You certainly have lots of well-thought out ideas and are very articulate, so why aren't you running for public office? In most circles, the people who criticize the most are asked to step up and volunteer to work on the solutions. Please understand I mean no disrespect here, just an honest question.

  9. Before we get to disaster relief, we don't even know how strong the winds were in Haiyan. So, how can we accurately size the response in advance? For this reason, I suggest the nations of Asia combine to start a hurricane recon and forecast operation similar to the U.S.'s. This has the added value of improving the initialization of computer models for better forecasting worldwide when these storms occur.

    As to the main point of this posting: For years WeatherData, now AccuWeather Enterprise Solutions, has worked with insurance companies, emergency managers, and others to insure they deploy on forecast. This has worked VERY well in hurricanes and other disasters, including Katrina and Sandy. In the case of Sandy, our client Lowes had 400 truckloads of flashlights, generators, etc., delivered to stores in the path.

    With regard to the military, I have to agree with Rush Limbaugh on this one: The mission of the U.S. military is to break things and kill people. It is a misuse of the military for this type of purpose, no matter how noble the goal.

    That said, I agree that carriers are the perfect platform. How about the UN purchase some retired carriers or commission some new ones designed for this purpose? Japan, China, India, and other nations, as well as the U.S., can chip in.

  10. The military-industrial complex would not be well-served if the naval power were used to do good instead of firing a lot of ammo. How would we be able to spend trillions for the benefit of weapons suppliers if the world learned peace?

  11. Cliff, your answer is here:


Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

Finally! Rain West of the Cascade Crest

After a long dry period, western Washington and Oregon are finally getting some significant rain. The radar image at 7 AM shows a mass of pr...