Hurricane forecasters say seasonal errors are hurting their credibility
The Miami Herald has an excellent article about the third consecutive high-profile failure of seasonal hurricane forecasts to closely approximate reality. (The forecasted storm totals were way too low in 2005, way too high in 2006, and substantially too high in 2007.) The article focuses, quite rightly IMHO, on the fear that these forecasting failures are lowering the public's confidence in the much more important -- and much more accurate -- operational forecasts regarding individual storms that the National Hurricane Center does such an excellent job with. I talked about this issue in my season wrap-up for Pajamas Media, and the Herald keys on it as well. Excerpt:
[G]iven the errors -- which can undermine faith in the entire hurricane warning system -- are these full-season forecasts doing more harm than good? [Yes. -ed.]
''The seasonal hurricane forecasters certainly have a lot of explaining to do,'' said Max Mayfield, former director of the National Hurricane Center. ...
Mayfield and virtually all hurricane researchers and forecasters, some of whom were skeptical years ago, now support the issuing of full-season predictions. [Why?? -ed.]
But many openly share concerns about the current system, focusing in particular on NOAA's tendency to subtly link the National Hurricane Center in West Miami-Dade County to the seasonal forecasts produced by [Gerry Bell, NOAA's lead seasonal forecaster]'s team, which is based in Maryland.
In fact, it is important to emphasize the distinction between the six-month seasonal forecasts and the real-time forecasts of an actual hurricane or other tropical system, which are called "operational forecasts.'' ...
Many [operational forecasters] worry ... that substantial errors in those full-season predictions can undermine faith in their generally accurate forecasts of actual storms.
They note that NOAA, parent agency of the hurricane center and Bell's team, often releases Bell's predictions during pre-season news conferences conducted at the hurricane center.
During other years, the hurricane center's director is ordered to participate in the pre-season news conference, wherever it might be held.
''NOAA has been using the good name of the National Hurricane Center, at least to some extent, to help promote the seasonal product and that's not the mission of operational hurricane forecasters,'' Mayfield said.
''In some areas, hurricane forecasters are losing credibility even though they are not the lead on this -- and that's always a concern,'' he said. "We don't want the credit for the seasonal forecasts.''
Bell said the differences between the two groups should be clear to the public by now. He said South Floridians and other residents of the hurricane zone should never disregard real-time forecasts, especially based on a misconception about the full-season predictions.
''There's no basis for those kinds of comments,'' Bell said, "especially if they're made by people who don't know what they're talking about.''
There might be "no basis" for comments linking operational forecasts and seasonal forecasts -- no valid basis, anyway -- but NOAA is setting itself for the inevitability that such comments will be made, with or without a "basis," when it releases its seasonal forecast with such a media splash and involves the NHC in that splash. So forgive me if I have little sympathy for the hue and cry that people "who don't know what they're talking about" are to blame for this. Ignorant people will always mouth off about things they don't understand, all the moreso when it suits a political agenda. NOAA is squarely to blame for giving them an easy opportunity to do so.
Philip Klotzbach, who issues the Colorado State forecast along with William Gray, "said long-range predictions satisfy the public's 'inherent curiosity'," according to the Herald. Well, he's a scientist, so he can do stuff simply for curiosity's sake if he wants to. But NOAA officials aren't just scientists, they're also policymakers, and they need to base their actions on sound policy judgments -- not just a desire to satisfy idle curiosity. It seems to me that these seasonal forecasts are indeed doing more harm than good, and NOAA should either stop issuing its own forecast or at least vastly scale back the media profile that it chooses to give that forecast. Don't call a press conference, don't do interviews, just quietly release the thing on the Internet (loaded with caveats) and satisfy the weather nerds' "curiosity" that way, without unintentionally (but foreseeably!) misinforming the public at large. And certainly, if you must make a media splash, don't involve the NHC operational forecasters in it, for heaven's sake.
It would also be a good idea to issue a press release, whenever anybody releases a seasonal forecast, reminding the media how generally pointless and useless these things are, that they're really just a curiosity, and that we ought to focus on what matters: preparing for big landfalling storms (which can happen in active and "inactive" seasons alike) and forecasting them accurately when they actually form.
Anyway, read the whole thing. And if anyone is tempted to turn this thread into a global-warming debate, please at least read my PJM piece first, if you haven't already. I address a lot of the obvious arguments there (like the old stand-by, "OMG If They Can't Even Forecast A Hurricane Season, Then How Can They Forecast The Climate In 100 Years?? Al Gore Suxxx!!") and I'd rather not repeat myself.
P.S. I will, however, repeat what meteorology Ph.D. student Charles Fenwick wrote back in August, because he made the point very well:
I don’t take too much interest in [seasonal forecasts] personally and don’t like how they are being pushed to the general public. They are a experimental works in progress and should be treated as such. I am most displeased with NOAA’s trumpeting of their forecasts. It gives the public the sense that these are operational forecasts that are on par with the other forecasts of the National Weather Service and that is definitely not the case. [One blog commenter, responding to a dire track forecast for an individual storm, asked], “Where are all the hurricanes the NHC had forecast for the last 2 years? just curious as to why we should panic over predictions that have little or no accuracy?” This shows the confusion that the hurricane season forecasts cause because the National Hurricane Center is not the agency that puts out the seasonal forecast and, as I just said, the seasonal forecasts do not have the same accuracy as the operational forecasts put out by the NHC. … [The seasonal] forecasts are most useful for people who have a stake in the macro-scale, namely insurance companies. They are of little value to individuals.
UPDATE: Welcome, InstaPundit readers!
I'm guessing the Instalanche makes it almost certain that a global-warming debate will break out, so I'll surrender to the inevitable and quote the key graf from my Pajamas Media article:
Ah, some say, but if scientists can’t even predict the number of hurricanes in a given year, how can they possibly predict the state of the Earth’s climate in 50 or 100 or 500 years? This argument is superficially compelling but fatally flawed, as it ignores the massive differences between the types of phenomena being predicted. Predicting the average temperature of the Earth in a given number of years is, at least arguably, much easier than predicting the precise number of hurricanes next year, or for that matter, the precise temperature (in tenths of a degree) at 4:57 PM tomorrow. The reason: the planet’s average temperature over a long period of time is not impacted by short-period, butterfly-effect type variables in the way that shorter-term forecasts are. This is not to say that long-term climate forecasts are necessarily accurate, but if they are to be proven inaccurate, it must be done some other way. The argument that “short-term and medium-term forecasts are flawed, therefore long-term forecasts must be even more flawed” simply does not pass muster.
In the Herald article, they make an even better analogy: early seasonal hurricane forecasts are like "predicting -- this past October -- the Marlins' precise win-loss record in 2008." Whereas, to extend that analogy, long-range climate-change forecasts are more like predicting that the Red Sox will continue to be successful over the course of the next decade. The latter prediction, based on long-range trends (like the amount of money a team spends, how good it is at evaluating talent, whether it's well-managed, etc.), is a lot more likely to be accurate than a prediction of an individual team's specific record in a specific year, even though the latter involves events that are more imminent at the time the forecast is made. The same thing applies to weather and climate predictions.
Just as it was both unsound and unwise for some global-warming advocates to hold up the 2005 hurricane season as proof of their position, it would be equally unsound and unwise for global-warming skeptics to hold up 2006 and 2007 as somehow disproving the existence of global warming. Such arguments are unsound because they confuse climate, which is comprised of long-term trends, with weather, which chronicles individual events. They are also unwise strategically because they are so vulnerable to attack when things — predictably — turn out differently in future years.
The heavy reliance on 2005 in certain quarters, which gave some lay observers the false impression that all hurricane seasons would henceforth be similar to the freakish ‘05 season, left global-warming advocates open to cynicism, criticism and rebuttal when 2006 and 2007 failed to live up to expectations. Similarly, a global-warming skeptic who claims today that 2007 disproves global warming is leaving himself open to the argument, if 2008 is an active season, that ‘08 proves global warming is real after all. The more honest (and strategically sound) course, for both sides, is to discuss global warming on its actual merits, and not obsess over minor year-to-year variations that tell us very little, if anything, about long-term trends.