8/20/2007

New Op-ed: Does Government Weather Forecasting Endanger Lives?

. . . . With all the blame still going around about Hurricane Katrina’s devastation, one fact has missed getting much attention: private weather forecasting companies predicted the threat to New Orleans well before the National Weather Service. In fact, AccuWeather issued a forecast that the hurricane would hit New Orleans 12 hours earlier than the government service.

This is hardly something new. Private companies with a lot at stake would often rather pay for private forecasts than rely on the “free” forecasts from the government. Hugh Connett, the president of Bridgeline, a gas pipeline company in Louisiana, claims that the government’s hurricane forecasts are too imprecise. He says that private companies such as AccuWeather do it better, because they give more accurate predictions and provide hour-by-hour forecasts of a storm’s path. . . .


UPDATE: For those interested, links for the sources have been imbedded in the article. Simply click on the link above.

UPDATE 2: Please note that a correction was made in the piece. The portion of a sentence in the second paragraph stating that there were 5 hurricanes predicted to hit the US has been cut.

Labels: , ,

37 Comments:

Anonymous Anonymous said...

http://www.stormeyes.org/tornado/blog/archives/2005/09/archives/accusux.jpg

8/21/2007 12:12 AM  
Blogger Atascosa Lookout said...

I disregard all weather forecasting. Public and private. How can one take seriously an entity that creates or embraces formulas that are only intended to exaggerate extremes? They are all full of Hot Air!

8/21/2007 9:59 AM  
Anonymous Anonymous said...

Sounds like you have an agenda against government agencies in general. The fact of the matter with regard to hurricane forecasting is that AccuWeather has continually of the past several years put lives and properties at risk by issuing forecast tracks well out scientific reasoning. They pick these outlying forecast tracks on the off chance they might pan out and they can claim to have outperformed the NHC. Thereby gaining notoriety and increased business and profits. Tell me whose doing this country a disservice.

8/21/2007 10:22 AM  
Anonymous Anonymous said...

Sounds like you have an agenda against government agencies in general. The fact of the matter with regard to hurricane forecasting is that AccuWeather has continually of the past several years put lives and properties at risk by issuing forecast tracks well out scientific reasoning. They pick these outlying forecast tracks on the off chance they might pan out and they can claim to have outperformed the NHC. Thereby gaining notoriety and increased business and profits. Tell me whose doing this country a disservice.

8/21/2007 10:27 AM  
Anonymous Anonymous said...

It is a grave tragedy that your article was published at Fox News without rebuttal. That such an article with baseless lies could go unchallenged is a great disservice to our country, our tax-paying citizens, and the tax-based institution endowed to protect them. I ask what is your agenda beyond the privatization of the weather service? If you nobly believed in the supposed wonders of pure capitalism, couldn't you achieve presenting your view without lying, and worse, using propaganda to achieve your presentation to the public? Shame on you!

It's kind of pointless to refute your argument in this relatively private court of your blog, since you must know that you are misleading the public and it may serve little purpose than to inform you of what you already know, but I will present a few contradictions in the hope that some of the public will read your blog.

First, "tophe" above aptly points out one of the blaring contradictions in your article, which I'm sure you were aware of previously. A shame that couldn't be provided to all the misled Fox viewers. The NHC did an outstanding job of forecasting Katrina, and if you didn't somehow know that, you now do. They obviously did better than your hero, Accuweather.

Second, the National Hurricane Center does not issue seasonal weather forecasts, last I checked. Further NOAA, an entity which does make seasonal forecasts, does not forecast seasonal landfall numbers. Your private industries often do, however, and make egregious forecasts (e.g., Accu-weather's infamous New York/carolinas devestation forecast for the year 2006. Wow!)

Next, it is simply not true that reconaissance data was held back. There was a computer glitch in the problem stated in your editorial, and that has been publicly released. Do some research, or, if you already knew that, shame on you once again!

The fact that your editorial is presented to the Fox News crowd is assinine. The fact that you pretend to be serious shows that you are a calculating propagandist, but perhaps a poor one, because so many of your arguments are so weak when compared with real data.

I also am disturbed by your wish to privatize the weather service. What, now only privileged people will get to know when danger is headed their way? Or, I'm sure a for-profit business is most concerned with the welfare of others. We won't go there. Let's first make sure you get your facts right before we even start to dabble in the fallacies of your philosophical reasoning for privatizing successful government institutions.

8/21/2007 11:50 AM  
Blogger John Lott said...

Dear Scott Krentz (or Isohume):

I am sorry that you so strongly dislike Accuweather, but the ForecastAdvisor.com or Forecast Watch both list that Accuweather does a better job than the National Weather Service.

8/21/2007 2:42 PM  
Blogger Unknown said...

In your article, you state that "AccuWeather issued a forecast that the hurricane would hit New Orleans 12 hours earlier than the government service."

Take a look at the link that was provided above by a previous poster. It shows a forecast that was issued by AccuWeather and NHC. The two forecasts were issued 2 hours apart.

The NHC track shows a predicted landfall in the New Orleans and Gulfport, MS area. AccuWx's forecast shows a predicted landfall in the Florida Panhandle. Tell me, which forecast verified?

While I agree, there are probably some areas of the NWS operation that border on "competition," I don't feel that private forecasting firms should feel threatened.

If I'm a business whose operation depends on weather, I would: 1)have my own division that forecasts for me, if I could afford it; 2)pay a private firm to forecast for me. Either way, you end up with a customized forecast with more information and resources that are dedicated to you.

The government forecasts are meant to be for the general public. As a member of the general public, why would I want to pay twice for a weather forecast? My tax dollars already pay for the development, operation, and maintenance of weather satellites, Doppler radar, upper-air observations, automated surface observing stations, and ongoing research. I have as much right to have access to that data as the private companies do. When I want just a general forecast for my area, why should I pay a private firm when I'm already paying the government to do it?

If I wanted something that was more customized to me, giving me an hour-by-hour updated sent to my PDA or cell phone, then yes, I would pay a private firm for that. I just don't have a need for it. There are several people that do have a need/want it (i.e., frequent travellers, people who work/recreate outdoors). Private weather firms will always be able to make a business of providing more detailed and customized information quicker and more often than their competitors.

This situation is very similar to that regarding the Postal Service, UPS, FedEx, DHL, etc.

8/21/2007 3:57 PM  
Blogger John Lott said...

Dear Chris:

Please take a look at the link that I provide in my piece. Just click on the part of the article that I quote from and it will take you to the links that I provide for my sources. I can only assume that you haven't seen the sources that I relied on.

"private forecasting firms should feel threatened." Well, some of the private forecasting firms surely feel threatened. Whether it is on purpose or not by the NWS is not the point. In any case, I am not sure why all forecasts shouldn't be privately provided. What is the motivation for having the government do it? Why should some people pay for the services that others receive?

Thank you.

8/21/2007 5:37 PM  
Anonymous Anonymous said...

Mr Lott,
Would you too have all the data collected by the NWS be done by private industry? And what if the NWS was to begin selling all the data that she collected and that was used by private industry, you would be private forecasting industries dripping life flies.

I'm not really sure how much hurricane related data, satelite pictures, aircraft flythourghs, ect... that private industry pays for but look at how much money they make using all the FREE data they have access to provided by the NWS and the NHC.

Lets have Accuweather launch the next weather satelite or fund the
next year of upper air soundings and the material support and the electronic maintanence support and
see how quickly they yelp.

Forecasting is just tip or visual part of the NWS. Will your next attact be against the watch and warning proccess? I'm sure you can find some one who does it better and that will assume the liabity for a missed warning.

'Cause once you begin being paid for a 'correct' forecast or warning and it's not correct I'd love to be the lawyer to take down the private industry forecasters.

8/21/2007 6:10 PM  
Anonymous Anonymous said...

http://jumper1.net/pictures/hurrisk.jpg

Your second paragraph could just as easily have begun:

"Despite dire predictions from AccuWeather ...".

On July 28, 2005 (approximately one month before hurricane Katrina), an AccuWeather meteorologist was quoted in "Insurance Journal" ranking the Carolinas, Florida, and New England as the top three locations "for storm activity during ... August".

Somehow I begin to think that long range predictions of forecasting aren't their forte, either.

8/21/2007 6:23 PM  
Anonymous Anonymous said...

Don't you think it would be appropriate to actually obtain the forecasts for Hurricane Katrina and compare them yourself... instead of taking a quote from someone who works for AccuWeather (the company that you are comparing the NHC to). The link posted in the original post refutes your piece. That is really poor journalism.

Makes me wonder how much of your article really is true... or correctly researched??? Probably not much.

8/21/2007 6:32 PM  
Blogger John Lott said...

Dear Joe:

I didn't argue that the NWS should be abolished in my piece, but I did argue that they shouldn't be competing against private companies for what the private companies were providing. I didn't have data on that particular question. But since you raise the issue, my guess is that private companies would probably do a better job getting the data together. Why do you think that non-profit organizations such as the government would have the same incentive to get things right? Do you think that things would be better if the government gathered the news? You would only have to have one reporter at each news site. You wouldn't have to deal with all that duplication in effort.

I am also not sure where your threat of legal action against the private companies comes from. Would you approve of also making it possible to sue the government NWS whenever it makes a mistake in making forecasts.

Dear Cody:

The comparisons that I made between the different weather services was regarding data where we have a lot of observations (daily forecasts over a year or more). Anecdotal discussions aren't very useful for the reason that you imply, but if Accuweather makes more mistakes than other services, it will go out of business. If the government NWS makes mistakes, will it go out of business. The data that I discuss that makes systematic comparisons on temperature and other measures shows that the government does much more poorly in making forecasts. Do you disagree with this much more systematic data?

Dear Anonymous:

I did use the data from two services that compare the accuracy of forecasts. Do you have a problem with their data? If so, what is the problem?

8/21/2007 7:07 PM  
Anonymous Anonymous said...

I have not read all the rebuttals to your article...but wish to remind you that the government is the key keeper and provider of observation systems like satellites, radar and the weather station networks which run all the monitoring systems and numerical models that BOTH private and government forecasting agencies depend on. So I think you should be watching what you are saying Mr. Lott...you will not be making many friends in the scientific and forecasting community with these sort of un-verified accusations

8/21/2007 7:31 PM  
Anonymous Anonymous said...

You my friend are a world class blowhole. Shame on you.

8/21/2007 8:15 PM  
Blogger John Lott said...

Dear Last Anonymous:

"the government is the key keeper and provider of observation systems like satellites, radar and the weather station networks which run all the monitoring systems and numerical models that BOTH private and government forecasting agencies depend on."

-- True, but there is no reason why the government has to be the keeper of all this. Is there? If so, possibly you could explain it to me.

8/21/2007 8:27 PM  
Anonymous Anonymous said...

Mr. Lott,

As a libertarian I cannot express adequately how firmly I believe in less government interference in free market processes. Your "agenda" against government agencies, as Scott Krentz puts it, is a healthy one based on sound evidence of general government performance in many areas and in the midst of a society where “self-responsibility” is a lost concept. As a meteorologist, however, I see how generalizing operational meteorology as you do can be very dangerous. Your headline itself is completely inappropriate.

I simply do not have the time to refute each and every claim you make in your article. The Katrina claims are simply inaccurate, as others have clearly illustrated. You are basing your findings on research whose data and methodologies are poorly documented and are meant to serve the private sector. I found reading the various claims from the sources you reference less than satisfying as many of these reports would fail to stand up to the scrutiny of basic scientific review processes.

Businesses pay for personalized forecasts when they collaborate with private sector entities. The NWS main mission is to support the taxpayers (i.e., general public; businesses pass corporate taxes to consumers and thus do not pay taxes) by fulfilling a Constitutional duty to protect the lives of U.S. citizens. (It is unusual, in this day and age, to find a government agency actually focused on a mission that adheres to the Constitution.) I advise you to spend some time at various NWS offices and get to know the agency a little better. I will wager you'd be pleasantly surprised with the performance and customer satisfaction of the local NWS offices as many of these offices truly go above and beyond to protect the lives and property of American citizens and advance the current state of operational meteorology. Based on my many experiences with the private sector, I cannot imagine a world where the NWS hands over its critical weather support to profit-based private businesses. Citizen lives would most certainly be in grave danger.

The collection of data must be standardized and centralized in order to maintain the integrity of the data and various resources that rely on that data. I should not have to elaborate on the reasons why a non-standardized warning process would work against the public interest. I would love to see Accuweather install thousands of weather stations and hundreds of military-grade radars across the country and maintain them. The truth of the matter is, simply, companies like Accuweather want it all. . .they want the freedom to bash the NWS operational process while enjoying the low cost of their services as they piggyback off the enormous costs endured by the U.S. taxpayer to install, maintain, and enhance equipment critical to each and every one of the private sector’s personalized services. They are able to make sensationalized predictions with no consequence to the general public to which there is no charge for basic forecasts and guidance while NWS forecasts and warnings are under significant public and political scrutiny. Without the NWS, the state of U.S. operational meteorology would descend into chaos as multiple companies pursue independent data collection, quality control, forecast, watch, warning and verification programs. Scientific research, integrity, public safety, etc. would be seriously undermined.

I don't regard your article as dishonest; I understand and agree with your general economic philosophies. However, I do feel the article is ignorant to the processes of operational meteorology and the danger of placing data collection, quality control, hazardous weather forecasting, watches, and warnings in the hands of a wealth of private entities whose primary responsibility is to earn a profit. You simply cannot place the Constitutional duty of the government to protect the lives of its citizens in the hands of free market enterprises without risking the same endangerment you speak of in your headline. To think there is no consequence to the NWS for poor products and services is simply wrong.

The current environment of co-existing private and public sectors is actually working quite well, where the NWS maintains its mission of public protection and the private industry continues to offer fee-based, highly-personalized and specific services to businesses that require it. Fortunately, with or without your best wishes, it will remain this way for quite some time.

Best of luck in your many other battles against government waste.

8/21/2007 10:24 PM  
Anonymous Anonymous said...

Satellites cost tens of millions of dollars to develop, maintain, and disseminate. The cost would far outweigh the profit that a private forecasting company could generate from it. However, they are the primary source of improved medium range weather forecasts throughout the globe. You cannot put a price on the safety of others, be it the general public or the military operations throughout the world that depend on accurate forecasts.

8/21/2007 10:45 PM  
Anonymous Anonymous said...

John...

I dont hate AccuWeather...I was merely pointing out a fact regarding their hurricane forecasting.

Also...I looked up my town on ForecastAdvisor.com and apparently the NWS outperformed AccuWeather overall for the last month and the last year. In fact...the NWS was rated # 1 in both categories, while AccuWeather was rated last for the month and second to last for the year.

8/21/2007 10:52 PM  
Anonymous Anonymous said...

folks

Note the name. I am NOT anonymous.
My web site WXRISK.COM and is one the premier leading private weather forecasting firms in the nation. It is VERY well known.

And for the record MY forecast for the hurricane season of 2006 was for 13 named storms. My forecast for DEAN was issued back on AUG 10.

First note how pathetic Mr Lott's answers have been!! He makes claims about Accu weather SUPPOSED superior Katrina forecast... then when some poster provided REAL data from the forecast in question Mr Lott simply falls back to his
'sources'

ForecastAdvisor.com or Forecast Watch are Both private weather COMPANIES that -- get this make make Money off of other private weather companies that are trying to prove that private weather sector companies are superior.

Somehow Mr Lott thinks that these two sources are an unbiased source. This is the equivalent of asking a Madam in a whore house if getting STDs is a serious problem.

Even so the sources he cites present NO data that support Mr Lotts claims that Accu weather forecast was 12 hrs earlier and better than NHC's.

Note that Mr Lott wont look at the ACTUAL data of the Katrina forecast per se. Perhaps Mr Lott a Researcher at U of MD does not know how to read Latitude and Longitude?

Mr Lott and I talked on the phone Tuesday afternoon. It turns out that Mr Lott cannot seem to recall what he actually wrote in his own essay

In the conversation we had I raised the issue of Mr Lott asserting the NWS/ Hurricane Center (TPC) had ...in 2006 forecast numerous East coast landfilling Hurricanes.

Mr Lott told me in no uncertain terms that "I never hinted or said anything like that. The term East coast does NOT appear in my essay.

Since I could not find that specific sentence -- RE east coast hurricanes -- I presumed that Mr Lott was familiar with with his own writing.

Turns out Mr Lott is in fact unfamiliar with his own work... or Mr Lott is a LIAR.

Oppps. Nice try John. This sentence IS in your essay.

Despite dire predictions from the National Hurricane Center, no hurricanes hit the U.S. last year...

That sentence DOES imply that NHC did predict numerous East coast hurricane strikes ( landfalls) in the 2006 season.

That is a lie

Ironically it was Accu weather --with whom Mr Lott seems obsessed with -- that did in fact PUBLICLY state the 2006 hurricane season was going to feature numerous East coast hurricanes.

Yet here is Mr Lott telling readers that The hurricane center screwed up the 2006 season forecast --

Yes The Hurricane center's 2006 forecast was wrong. But so was MOST of the Private weather sector seasonal hurricane forecasts!

Mr Lott tells readers that the hurricane center DID forecast Numerous east coast strikes. That is 100% wrong.

Then Mr Lott cites Accu weather as NOT making those types of blunders but somehow over looking the KEY fact that is was ACCU weather that in fact predicted the numerous east coast hurricane strikes!!!

Opps.

Mr Lott claim/s to be libertarian.
His argument seems to be IF NWS shuts down the "free data" and say Accu weather --OR any other private company-- got possession of that data that would help private sector forecasters.

Mr Lott never gets around to explaining HOW having all the data owned or controlled by Accu Blunder would that benefit ME.

what a uninformed fool

8/21/2007 11:38 PM  
Blogger MAX POWER said...

Wow, to claim that Accuwx did a better job with Katrina is a whopper of a lie. Based on the linked image, and my own memory watching the forecasts as Katrina approached, Accuwx was calling for a Florida Panhandle landfall well into Friday evening 8/26/05 for several hours after the NHC had shifted their forecast west to include Biloxi/New Orleans. And as others have pointed out, Accuwx has a history of terrible seasonal predictions such as a high threat of hurricane landfalls on the US Northeast coast.

Meanwhile, many of us are still waiting for Mr. Lott to back up his claim with evidence that NOAA/NWS/NHC had ever forecast 5 hurricanes would make landfall in 2006. Crickets are chirping.

There's plenty of room for the private sector to provide value-added weather information to specific clients, sectors, and the public. Trashing NWS efforts with falsehoods is not useful to anyone.

8/22/2007 12:15 AM  
Blogger John Lott said...

Dear Anonymous:

Tens of millions of dollars is not a lot for a private company. How much do you think communication satellites cost? How much do you think some buildings cost? Or factories? To develop new drugs on average cost about $1 billion. What do you think is at stake in terms of losses from bad weather? Why is it obvious that these things will not pay for themselves?

Dear Isohume:

Thanks. It doesn't make much sense to take anyone city or town. The question is what is the average over a large number of them.

8/22/2007 12:21 AM  
Blogger Unknown said...

Mr. Lott:

You are correct. I did not see the links in your article because I read your article on Fox News. However, now that I have read them, I will respond.

I am troubled by the following statement:

"AccuWeather Inc. senior meteorologist Michael Steinberg said emergency managers and the public could have been given an earlier warning of Katrina's threat to New Orleans. He said the private company had issued forecasts nearly 12 hours earlier than the hurricane center warning that Katrina was aiming at the area."

This is very misleading. What I believe Mr. Steinberg is saying is that AccuWx issued their forecast for a New Orleans/Gulfport landfall 12 hours before NHC issued a Hurricane Warning for the area.

By 24 hours from landfall, AccuWx probably did predict a New Orleans/Gulfport impact. The eventual track was fairly certain by then. However, Hurricane Warnings (24 hours) and Hurricane Watches (36 hours) are not issued much further out in time due to inherent uncertainties regarding forecast tracks. I mean, let's face it, it's not exactly a good idea to evacuate 500,000-1,000,000 people from an area when a relatively large amount of doubt still exists in the eventually track.

However, Mr. Steinberg's statement implies that AccuWx outforecasted NHC regarding Katrina's landfall point.
That is simply not the case.
Just because (according to Mr. Steinberg) AccuWx forecasted the correct landfall prior to the official Hurricane Warning issuance, does not mean the NHC was not already forecasting the same landfall several DAYS before AccuWx did.

Like I said, very misleading.

The forecasting for Hurricane Katrina was a huge success for NHC.

Thank you for your time.

8/22/2007 3:03 AM  
Blogger Unknown said...

I find it amazing how excited people are getting about this issue. When you talk about bigger issues like gun control and crime there is not nearly such a large response.

8/22/2007 7:58 AM  
Anonymous Anonymous said...

John...I'm happy to hear you say this as I fully agree. One city can not possibly account for the overall forecast accuracy. For your article...what locations were used to calculate the average?

8/22/2007 10:52 AM  
Blogger Don said...

Dr. Lott,

Your piece contains some serious errors and material omissions that imperil the validity of its arguments:

1) You write of the National Hurricane Center, "This year they are again predicting as many as 10 hurricanes, up to five of them hitting the U.S." The National Hurricane Center does not predict hits or landfalls in its seasonal forecast. For what it is worth, the NHC's seasonal forecast can be found at: http://www.noaanews.noaa.gov/stories2007/s2864.htm. Nowhere is there any mention of hurricane hits or landfalls.

2) You repeat an old allegation that the National Hurricane Center deliberately withholds vital data from private forecasters. You state, "Unfortunately, these aren’t the only problems with the National Weather Service. The service has been accused in the past of withholding government aircraft reconnaissance of hurricanes for up to 11 hours before releasing the information to private companies or the public." However, the National Hurricane Center's website provides an explanation that addressed the allegation that resulted from a misunderstanding of how the reconnaissance and surveillance process works. That explanation can be found at: http://www.nhc.noaa.gov/aboutrecon.shtml.

3) You cite figures from Forecast Watch and Forecast Advisor. What appears to be two sources is really a single source. Forecast Watch and Forecast Advisor are services offered by Intellovations, LLC.

4) You fail to mention that the Forecast Watch evaluations are very different from the independent assessments offered by Moody's and Standard & Poor's. This is a material omission. Forecast Watch provides ratings to clients who pay it. On its website, Forecast Watch states:

You have always competed with the NWS, but now the competitive landscape is getting more crowded, as technology improves and customers become more demanding. We can help you differentiate yourself in the new marketplace, as well as work with you to develop solutions to help you with performance analysis and model improvement. We can provide you with pdf reports for your sales and marketing teams, and provide technical data for your meteorological staff and management.

With the large amount of forecast and actual weather data captured by ForecastWatch.com, there are plenty of opportunities for data mining and quality verification. Intellovations can work with you to develop a quality assurance program that will help you improve your presentation quality which will only make your customers happier...

We proudly count AccuWeather and CustomWeather as satisfied clients, and we'd love to add your company to the list.


There is an inherent conflict in which companies who benefit from paying clients offer competitive evaluations for those clients. Their incentive is to demonstrate improvements produced by their services and this can readily be accomplished when the ratings function is not segregated from the client service function. Certainly, your article makes no mention of the kind of robust internal controls that would be necessary to minimize such a risk.

In contrast, Moody's and Standard & Poor's are wholly independent of the companies they evaluate. For more on Forecast Watch's methodology: http://www.forecastwatch.com/services/.

These deficiencies have nothing to do with opinion, which anyone can and should be free to express. Instead, they have to do with facts and material omissions that compromise the arguments that rest on that information.

8/22/2007 11:04 AM  
Anonymous Anonymous said...

Mr Lott...

Have been following the debate on several forums...

Still have not heard an explanation as to how a company which is paid by private forecast firms to evaluate the accuracy of their forecasts can be considered an independent and objective source for evaluating the alleged massive superiority of private forecasting over NWS forecasts which your article states.

Looking forward to your response...

Jon in Jersey

8/22/2007 12:54 PM  
Anonymous Anonymous said...

I tend to look at all weather forecasting in the same light...they hold themselves out to be the "experts", thay all make forecasts, they are a vast majority of the time wrong, and they continue to portray themselves as the experts. (Just look at the projected "cones" for Hurricane Dean since the beginning...notice how the northern edge became well south of the earlier southern edges? Now as for Accuweather, every once in a while I'll track their forecasts for my local area over their 15 day forecast period...and I've emailed them about this, basically asking them "why bother" because their forecasts change so dramatcally over that period of time - just as a quick example in my area in PA I've been watching their forecast for this Friday through Monday...to date, the forecast high has varied from 79 to 92, 78 to 92, 79 to 90, and 74 to 85 for each day respectively, and the forecast has varied from "rainy" to "cloudy" to "mostly sunny" for each day - I can't help but wonder exactly what the "ACCU" means in their name!

8/22/2007 2:00 PM  
Blogger John Lott said...

Dear DH Tolleris:

"ForecastAdvisor.com or Forecast Watch are Both private weather COMPANIES that -- get this make make Money off of other private weather companies that are trying to prove that private weather sector companies are superior."

-- Both sources were very clear on their methodology. Do you have a specific problem with what they did? What they did was very simple? How would you have done it differently?

"Mr Lott and I talked on the phone Tuesday afternoon. It turns out that Mr Lott cannot seem to recall what he actually wrote in his own essay

In the conversation we had I raised the issue of Mr Lott asserting the NWS/ Hurricane Center (TPC) had ...in 2006 forecast numerous East coast landfilling Hurricanes.

Mr Lott told me in no uncertain terms that "I never hinted or said anything like that. The term East coast does NOT appear in my essay.

Since I could not find that specific sentence -- RE east coast hurricanes -- I presumed that Mr Lott was familiar with with his own writing.

Turns out Mr Lott is in fact unfamiliar with his own work... or Mr Lott is a LIAR."


-- Please point to where in the piece that I was talking about estimates of East coast Hurricanes? I did a search on my piece and the term "East" is not in it. The sentence "Despite dire predictions from the National Hurricane Center, no hurricanes hit the U.S. last year... " does not contain the word "East."

?Mr Lott claim/s to be libertarian.
His argument seems to be IF NWS shuts down the "free data" and say Accu weather --OR any other private company-- got possession of that data that would help private sector forecasters."


-- DH has a financial interest in this debate because he relies on the NWS providing him free data. That is fine. I am sure that some companies benefit relative to others, but there is no obvious reason why it is necessary for others to subsidize his data.

Dear Kevin:

Please follow the link in my piece on the website to the Associated Press story that I rely on for the statement reguarding hurricane Katrina. I didn't say 5 hurricanes in 2006. What I said was that they predicted 5 would hit the US in 2007.

Dear Chris:

The Associated Press seemed like a reputable source. Given that it is possible that he was misquoted, I have talked to Michael Steinberg and confirmed the statement that was made to the AP. Your statement however seems contradictory in that on the one hand you say that it was well known by that time but on the other hand you say it is "not issued much further out in time due to inherent uncertainties regarding forecast tracks." Steinberg said that he thought that the piece was very good.

Dear isohume:

The discussion for ForecastAdvisor.com looked at the largest 50 cities.

Dear Don:

I haven't looked at Moody's ratings, but I will try to find them. I wish that you would provide a concrete discussion of the weaknesses of the Forecast Watch and Forecast Advisor data.

8/22/2007 2:13 PM  
Anonymous Anonymous said...

Mr Lott

Could you point me to the part of the forecastadvisor.com study that concludes that the NWS forecasts "endangers lives"???

Jon in Jersey

8/22/2007 2:34 PM  
Blogger Don said...

Dr. Lott,

As I noted, I've withdrawn my 3rd and 4th posts given that Intellovation's founder explained his company's approach. The first two points concerning the NHC's seasonal forecast and explanation of the reconnaissance and surveillance process still stand.

8/22/2007 2:36 PM  
Anonymous Anonymous said...

you wrote:

[b]Please follow the link in my piece on the website to the Associated Press story that I rely on for the statement reguarding hurricane Katrina. I didn't say 5 hurricanes in 2006. What I said was that they predicted 5 would hit the US in 2007.[/b]

you've been asked multiple times and you have not answered this question: who is "they" in this case? thank you in advance, Trixie

8/22/2007 3:35 PM  
Blogger John Lott said...

Dear Trixie:

"Who is 'They'?" I believe that you must mean: who are they? I am sorry that you haven't been able to follow the explicit answers that I have given to this question. NOAA predicted that 5 hurricanes would hit the US.

8/22/2007 4:27 PM  
Anonymous Anonymous said...

Dr. Lott,

You are not an expert when it comes to the science of Meteorology. Please stick to economics and stay out of a science that you know next to nothing about. The entire meteorological community disagrees with your statement. You are giving UMD a bad name right now, which is a shame because they have a very reputable atmospheric sciences department.

Joe Bastardi, who works with AccuWeather, was touting last week that Hurricane Dean was going to go near southern Florida. The National Hurricane Center never had this track and they consistently sent Dean on a southern track into the Caribbean and eventually into the Yucatan Peninsula. I assume that you have been following Dean enough to know where it actually went.

Seasonal forecasts are one thing. They usually have very low skill and most agencies do not get it right. But when meteorologists from Accu-Weather totally miss 72 to 96 hour forecasts of actual hurricanes (like Katrina, Ophielia, and Rita to name a few (which they said was going into Houston)), and the fact that the NHC had some of the best forecast tracks (http://www.nhc.noaa.gov/pdf/TCR-AL122005_Katrina.pdf Go to page 13 for error stats), your arguement does not hold. The Department of Homeland Security actually used AccuWeather's mis-forecast of Katrina because the former head of DHS (Tom Ridge from PA-- where AccuWX is from) used their forecasts on a frequent basis. Did this lead to the delay in action by DHS? Answer that question for me.

You consistently point to this website that does verification of temperatures of the 50 biggest cities. First of all, you are taking 50 point locations out of a massive geographical area and your taking these forecasts over a limited time domain. Are you telling me that you are basing your thesis on this? That's terrible research and you know it.

Please stick to what you know best. You've insulted a private meteorologist that respects the job of the many government meteorologists. Many private forecasters respect the job that the government does in terms of weather forecasting.

8/22/2007 4:27 PM  
Blogger John Lott said...

Dear Most Recent Anonymous:

Despite your claim about everyone in the "meteorological community" disagreeing with me, I have gotten numerous calls from people in that community who do agree with me.

You consistently point to this website that does verification of temperatures of the 50 biggest cities. First of all, you are taking 50 point locations out of a massive geographical area and your taking these forecasts over a limited time domain. Are you telling me that you are basing your thesis on this? That's terrible research and you know it.

It was at least a large sample. If you have some other comparison, please provide it. Simply claiming that the comparison data that I have provided is unfair is not very useful.

Did this lead to the delay in action by DHS? Answer that question for me.

Obviously I don't know the answer to this, but it is strange that you find daily data for 50 cities (18,250 observations =365*50) is to small and not representative enough.

8/22/2007 7:23 PM  
Blogger John Lott said...

Discussion closed.

8/22/2007 7:26 PM  
Anonymous Anonymous said...

You are a social scientist and not a meteorologist. A one year supply of data is not sufficient for comparative analysis of accuarcy in the field of meteorology. You are an economist and thus should be familiar with the workings of Edward Lorenz and other basic mathematical theories that formulate basic meteorological prinicples. Please respect the field of economics and we will respect the field of meteorology

8/23/2007 12:28 AM  
Blogger John Lott said...

"A one year supply of data is not sufficient for comparative analysis of accuarcy in the field of meteorology."

There are 365 daily reports in a year. That seems like enough data to me to get a statistically significant difference in prediction rates. I cited two sources of data from Forecast Watch and Forecast Advisor. If you have another source, please provide it.

"You are an economist and thus should be familiar with the workings of Edward Lorenz and other basic mathematical theories that formulate basic meteorological prinicples. Please respect the field of economics and we will respect the field of meteorology."

This is a silly argument. I know statistics. You just argue from authority without actually explaining anything. I will tell you what. If you respect economics, does that mean that you will be unable to properly comment on whether the service should be publicly or privately provided? After all, that is an economics question. RIght?

8/23/2007 1:04 AM  

Post a Comment

<< Home