Monday, August 29, 2011

Fallout from the storm...and a rebuttal

It's been a week and a half since my last post, as I've been on vacation in northern California during that time.  But what a week to miss--a strong hurricane moving up the east coast with lots of media hype surrounding this storm.  I wanted to just briefly cover some highlights of the fallout from this storm by looking at what other people are saying.

1) Irene caused/is causing major flooding across much of the northeast.
As seen here:

 Much of the northeast remains under flash flood watches or warnings.  It's not completely safe to be out there just yet.  In the coming days we'll see the true extent of the damage caused by the storm.  The highest rainfall amount (so far) as reported by the NWS was in Bunyan, NC with 15.66 inches of rain.  However, greater than 10 inches of storm total precipitation was also reported in Virginia, Maryland, Delaware, New Jersey and New York.

2) Irene brought near-record storm surge to the Chesapeake Bay region.
Surge heights of 7.5 or 7.63 feet were reported near Norfolk at their peak.  This is only a few inches shy of the 7.89 foot record set by Hurricane Isabel in 2003.  Since storm surge height is often used as one proxy for the "strength" of a storm, this indicates just how significant of a hurricane this was for the east coast.

3) Radar images along the coast showed excellent detail of the storm as it passed over.
This is just a subject that I happen pay attention to.  Some of the newly-updated dual-pol NEXRAD radars got glimpses of this storm.  Since I've really only seen dual-pol products applied to severe thunderstorms and snowstorms, it will be interesting to see what dual-pol can do for hurricanes.  If you have liked my previous radar interpretation blog posts, then I recommend this post by Patrick Marsh (everybody's favorite graduate student) about interpreting wind fields in the near-hurricane environment from doppler radar.

4) The hype surrounding this storm is generating a lot of press...about the press.
I'm told that the network news coverage of this event was pretty spectacular and omnipresent.  There was no escape from it.  Being on the road most of the time, I didn't get to witness it.  But, several news outlets are wondering if the media went too far--or did just fine.  I suppose it depends on your viewpoint.  I have a feeling that had the storm been worse than predictions that we'd then get complaints that there was not enough warning, we were caught completely off guard, and so on.  Actually, based on the numbers above, I don't think that, at least for the meteorological community's part, this was over-hyped.  By the media, perhaps.  But I think the meteorological community did well.

5) Budget cuts to the weather service?
Of course, in the midst of all this, a few articles did come out both supporting and criticizing the NWS.  Some articles, like this Huffington Post article, highlight the impeding budget cuts to the National Weather Service and NOAA satellite programs.  They argue that our aging weather satellite fleet would be in danger of falling apart over the next few years if the program lost money.  We depend on weather satellites for so much of our observations in meteorology today that it's almost unthinkable that we'd let such a valuable tool disappear for budget reasons.  As the article points out, we have almost no observers taking weather readings out on the open ocean.  But that's exactly where hurricanes form and develop, and exactly where we need the observations the most.  Satellites are our greatest tool to fill that immense data void and without their observations, our hurricane models (and even our everyday global weather models) would be reduced to shots in the dark.  The GOES-R series, due to be launched in the next few years, has significant technological upgrades and much higher resolution than our current satellites.  With this GOES-R series, we have replacements for our aging satellite fleet in the works.  It would be one of the most short-sighted and foolish things I can think of to cut the program off now.

6) Fox News's "Do we need a national weather service?" editorial
This article has been receiving a lot of buzz (at least among my friends and colleagues in the meteorological community) over the past few days.The article criticises the NWS as a "relic of America's past that has outlived [its] usefulness."  The editorial writer argues that private companies provide better forecasts than the NWS and that cutting the NWS's budget by $126 million to levels of almost a decade ago wouldn't have any serious effects.

I believe this position represents an extremely limited vision of what the National Weather Service does, limiting it to just "forecasts".  It's true that private companies can often provide better forecasts on a point-by-point basis than the National Weather Service (NWS).  There is a thriving private sector in meteorology that makes serious money off of specialized and localized forecasts.  These companies can run high-resolution models and conduct internal research to improve their forecasts.  All of this is definitely doable by the private sector and I'll even admit--the forecasting aspect of the NWS is probably its most expendable feature with respect to the private sector's abilities.

But let's consider some of the other roles of the NWS.  Let's take those high resolution models that private companies run.  Let's step back to the very first step of the forecasting process.  To do any sort of model, you need to have initial conditions.  We have to know what the atmosphere is like now (and how it has been recently) to be able to look into the future.  We know this by looking at observations.  How many types of observation platforms are managed by the NWS and its parent organization NOAA?
  • Weather satellites (GOES, POES, MODIS, and a suite of other specialized satellites)
  • The NEXRAD doppler radar network (jointly with the FAA and the DoD)
  • All upper-air sounding stations (weather balloons)
  • River level gauges
  • METAR observation stations at airports (jointly with the FAA)
And that's just a sample.  Almost our entire supply of upper-air observations and most of our surface observations come from the NWS or NOAA in some form or another.  Without these observations, no forecaster would be able to do predictions anywhere near as accurately as we do now.  No model would be able to get anything close to thunderstorm initiation or rain-snow dividing lines correct without this data.  We'd just maybe get the overall pattern correct over the next few days--sometimes.  All of our current weather prediction--be it public or private sector--depends on data from the NWS and NOAA.

One might argue that all of these data information services could be privatized.  We do have a couple examples that I know of where this has happened. The National Lightning Detection Network (NLDN) is administrated by the Vaisala Corporation.  Ever wonder why we don't see more lightning maps around, particularly in real-time?  Vaisala charges customers for access to lightning data in near-real-time, and those that buy it are prohibited from publicly releasing it until after a certain latency time (unless they pay more).  Another pay-for-data model has been adapted by the European Centre for Medium-Range Weather Forecasts.  I usually don't show many ECMWF model graphics and you won't find many detailed ECMWF model output images online either.  Once again--the ECMWF community charges people in non-member nations for their model output.  As such, even though they have one of the most advanced modeling systems around, we don't see much of their output online.  That is, unless you happen to belong to an organization that pays for this data.

My point is, we do have instances where private or near-private entities are controlling weather observations and model output.  In these cases, there is invariably a fee involved in accessing the data--that's how the private sector makes its money.

This is actually one of the things that worries me the most about eliminating the public sector from weather forecasting.  From the perspective of a researcher in meteorology, I enjoy and heavily rely on the plethora of free data that the NWS and NOAA provide.  You can go online and get both archived and real-time radar data whenever you need them.  You can look at output from all major weather models run by NOAA as soon as they are complete. You have access to weather observations at tens of thousands of locations through this free network.  I can see satellite images of anywhere in the country updated as frequently as every minute at times.  All of this for free.  Not only is the data free, but the mechanisms for producing, decoding or displaying the data from the NWS and NOAA are almost all open source--even the Radar Product Generator (the program that controls the doppler radars) and major weather models (like WRF) have all of their source code online, free to download and run. 

As I mentioned before, an observational network controlled by the private sector would virtually end all the free data and open source decoding.  Universities that are already cash-strapped would have to pay more money to get information that used to be free.  Cities and towns whose emergency managers relied on real-time radar and model updates to get emergency planning going would have to pay for access to this data.  Or, for a few dollars more, pay the company to have one of their expert meteorologists interpret the data for you.  Maybe we'd save some money by cutting spending for the NWS, NOAA and their programs.  But we'd probably end up paying it all right back to private companies in the end.

Which brings me to my last point--the accountability of the private sector.  While, as with any government organization, the politics will somehow find a way to infiltrate the system, I feel that the NWS really hasn't succombed to that as much.  Having worked in an NWS forecast office and for the NWS NEXRAD program, I have only seen the fringes of politics enter the business they do and never directly impacting the forecasting or warning process.  I feel that the NWS still provides an objective forecast and warning system.  The forecasters I know there are driven by their desire to get the forecast right--not by some other political or administrative agenda.

I also know several people who work for private sector meteorological companies.  They, too, possess a keen interest in getting the forecast right and that same thrill as NWS forecasters feel when things play out the way they thought they would.  I strongly support these forecasters and really respect the skill and passion they bring to their jobs.  It's not them that concerns me.

It's more the administration of these corporations that has me worried.  As any private-sector corporation, they are driven by profit.  With the way Wall Street has gone recently, we've seen that there are some darker sides of this system, even as robustly proven as it has been for our nation. We'd hope that higher profits are obtained by delivering a better product than your competitor.  But when your products can mobilize people and impact transportation or resource allocation, things get dicey.  Whole markets have developed that trade based on weather forecasts--people buy agricultural futures based on how they think the weather will impact crops.  A change in the forecast can cause the cancellation of hundreds of flights or the rerouting of dozens of container ships.  A change in the forecast can mean a change of millions of dollars.  So much of what we do depends on the weather.  Having a (hopefully) objectively-based, public-sector weather outlet that provides forecasts (and particularly warnings) without worrying about profit seems to make the most sense to me.  I'm not saying that the free-market economy would fail here--if a company makes bad forecasts, no one will trust it and it will fail.  But that strong link between the weather and our economy is something to think about when considering the role of private-sector weather companies in a country without a National Weather Service.

I usually try not to be opinionated (or voice my opinions) on these sorts of subjects, but I felt I should add my thoughts here.  After all, that's what a blog is good for, right?   If anyone else has any comments or thoughts, I'd love to hear them.

I also will do a write-up soon on my thoughts about the growing or shrinking weather-literacy of the American people.

1 comment:

  1. Another issue with private sector accountability is their lack of openness to independent verification. Accuweather's legal terms state that no one is allowed to archive their forecasts even just for verification purposes, and I am sure they'd sue anyone who did and published the results. A certain OKC news station would not allow their name to be used in a paper about OKC TV station forecast verification. The NWS, on the other hand, archives their NDFD grids and distributes them freely.