NAS report: "There is an apparent plateau in our ability to make accurate seasonal forecasts"

I have not had time to read this report, but I have gathered both the short form and long form version of the report and placed links in the University of Miami press release for WUWT readers to have a look and comment. – Anthony

Report Issued Today Examines Improving Long-Term Climate Forecasts

MIAMI — September 8, 2010 — Operational forecasting centers produce climate predictions that provide input for important decisions regarding water management, agriculture, and energy.  “Assessment of Intraseasonal to Interannual Climate Prediction and Predictability”, a new report from the National Research Council/National Academy of Sciences, examines current capabilities for making climate predictions — such as seasonal hurricane or longer-term drought forecasts — and identifies opportunities for improvement.

(Here’s a links to: the short form brief: here – long form complete report: NAS_12878 – by chapters here – Source here at NAS – Anthony)

The report finds that operational forecast centers could increase the value of forecasts for officials and researchers by modifying procedures for archiving and disseminating information.  In addition, improvements over the next several years to observational capabilities, statistical and dynamical models, and data assimilation systems should permit the forecast systems centers to better understand key processes — such as variables in the atmosphere, ocean, or land surfaces — that could help improve forecasts.

“There is an apparent plateau in our ability to make accurate seasonal forecasts; this report provides a road map to enable both scientific research and technical advancements to go beyond this plateau and produce seasonal forecasts that are of greater societal relevance,” said Ben Kirtman, University of Miami Meteorology and Physical Oceanography professor and co-author of the report.  “One of the more daunting challenges in this regard is figuring out how to seamlessly transition research results into real improvements in operational forecasts.”

UM Scientist Ben Kirtman Helps Co-Author Nat’l Academy of Sciences Document

Ben Kirtman

About the University of Miami and its Rosenstiel School

The University of Miami’s mission is to educate and nurture students, to create knowledge, and to provide service to our community and beyond. Committed to excellence and proud of the diversity of our University family, we strive to develop future leaders of our nation and the world. Founded in the 1940’s, the Rosenstiel School of Marine & Atmospheric Science has grown into one of the world’s premier marine and atmospheric research institutions. Offering dynamic interdisciplinary academics, the Rosenstiel School is dedicated to helping communities to better understand the planet, participating in the establishment of environmental policies, and aiding in the improvement of society and quality of life. For more information, please visit www.rsmas.miami.edu

Advertisements

42 thoughts on “NAS report: "There is an apparent plateau in our ability to make accurate seasonal forecasts"

  1. Well, they could start improving their forecasts by reading the Old Farmers Almanac (who seem to do a better job than most…) and toss out the AGW / CO2 driven models that are substantially wrong. Then look out the window for a few decades and learn what real weather is like…

  2. Reports like this would be greatly improved by tossing out most of the jargon.
    “There is an apparent plateau in our ability to make accurate seasonal forecasts; this report provides a road map to enable both scientific research and technical advancements to go beyond this plateau and produce seasonal forecasts that are of greater societal relevance,” said Ben Kirtman, University of Miami Meteorology and Physical Oceanography professor and co-author of the report. “One of the more daunting challenges in this regard is figuring out how to seamlessly transition research results into real improvements in operational forecasts.”
    Translation: “We hope to improve the accuracy of weather forecasts”

  3. I have read this once, rather quickly, and it will repay very careful reading. It strikes me as a well written, well thought out and objective account of the problems in climatic forecasting with proper emphasis on limitations and problems with current techniques.

  4. A curious awakening. Looking for more funding? Deflecting the failure of the AGW meme? Part of a political out reach? Interest in real science, I think they have forgotten.

  5. EM Smith
    I was saying just the same thing yesterday to some co-workers. The Farmer’s Almanac is predicting a very cold winter this year. Anyone taking bets on the FA’s accuracy this year?

  6. Does anyone know what factors are actually used in seasonal forecasts?? current ocean temperatures and jet stream location for example could be relevant but I’m not sure what correlations have ever been found looking ahead a few months.

  7. I think all medium/long-term forecasts would be greatly improved by tossing out most of the pseudo-science which underpins it.
    We need to have much better data and develop a better knowledge of how deterministic chaos inherent in the system drives the oscillations of changing weather regimes.
    This is a hard problem which needs a different approach before it can be solved.

  8. The report is very interesting and has some good information about the predictability of weather and climate. But, in the end, the pitch is for more research (=Climate Ca$h) so as to “improve” the forecasts, even though they don’t know what the upper limit of the “improvement” will be:
    [from p. 13 of the report]
    “It is not possible to quantify the true limit of predictability for the climate system.”

  9. It’s refreshing to read about some objective research that doesn’t blame CAGW for climate change. I would like to see the development of a programmed statistical tool that can be used on multi-variant, non-linear data; such as least squares perpendicular to a changing slope. Time and space resolution are the limiting factors for observational models. The accuracy of “what if” assumptions limit the predictability of mechanistic models.

  10. That’s a kind way of saying:
    We know very little about weather/climate, and no where near enough to predict it…

  11. “A plateau” a bit of an understatement … it was very obvious to me when I looked at the Met Office world climate prediction that they were almost no better than random guesses (taking last year’s value and guessing the next would be the same).
    Moreover, there was a continued bias in the results … a bias that just happened to be equal to the presumed global warming they expected each year!!!!! Thankfully,after 9 abysmal yearly forecasts, the UK Met Office finally had the good grace to admit they couldn’t forecast next year’s world climate.
    But what about the decadal forecasts? Of course they don’t like to admit, that with higher levels of noise in the climate signal over longer periods, it is even more difficult to get the decade predictions right, but on the basis of the yearly forecasts I can make this prediction:-
    PEOPLE LIKE THE MET OFFICE WILL CONTINUE TO TELL THE WORLD THEY CAN PREDICT THE CLIMATE ONE YEAR AHEAD AND GET IT ABYSMALLY WRONG (AND BY CHANCE RIGHT OCCAISIONALLY) AND STILL SAY THEY CAN PREDICT IT FOR UP TO 9 DECADES

  12. Be nice!
    This is an important new humble position and tone from the climate tower of power.
    If this is a important step towards reality away from climate fantasy, its a good one!
    Maybe they start consulting Pierce Corbin and Bastardi to take a giant leap in forecasting capabilities.

  13. Eddie says:
    September 9, 2010 at 5:32 am
    Anyone taking bets on the FA’s accuracy this year?

    The OFA is hard to beat, as well as the plants & animals that say “Early Winter”.
    I would not put my money on AGW/C02 based predictions, or any mono-coded computer system that only agrees with itself/is capable of predicting only warmer/drier conditions from here until the end of time.

  14. Has there been an audit of Accuweather and Weather Action done to see whether they are significantly better at it? And any other set up I don’t know about who does similar??

  15. Why do they feel a need to forecast next years weather anyway?
    Except perhaps it saves them from having to go out and find a proper job in the real world.

  16. Rhys Jaggar says:
    September 9, 2010 at 7:20 am
    ==============================
    Rhys, this site rates them:
    http://www.forecastadvisor.com/
    If forecasting has reached a “plateau”, that means this is a good as it gets.
    If this is as good as it gets, you might as well just look out the window….

  17. Seasonal??? Here in Sarasota Fl. the early morning forecast for the day is wrong by mid-afternoon about as often as not. Big wrong. Like 6 degrees F for the max., or thunderstorms that don’t materialize, or storms that do materialize but weren’t predicted, or overcast instead of sun.

  18. WOJ says:
    September 9, 2010 at 7:26 am
    Why do they feel a need to forecast next years weather anyway?

    The insurance industry needs it in order to set rates. Farmers for choosing which crop to plant. Etc.

  19. Rhys Jaggar says:
    September 9, 2010 at 7:20 am
    “Has there been an audit of Accuweather and Weather Action done to see whether they are significantly better at it? And any other set up I don’t know about who does similar??”
    I would guess the audit is ongoing and natural 😉
    The farmers, supermarkets and clothiers would not remain long-term customers if the results fell below, say, 50% accuracy (a coin flip such as the Met Office “perform”). Human nature, if not harsh economic reality, would not allow them to recommend such services to their friends either.

  20. “Well, they could start improving their forecasts by reading the Old Farmers Almanac ”
    Heh. Indeed.

  21. There are two questions forecasting climatologists must answer. First, are we measuring all of the things we need to measure in order to make skillful long term forecasts? Second, bearing in mind the sensitivity to initial conditions characteristic of chaotic systems, are our measurements sufficiently precise for skillful long range forecasts? The obvious answer to the first question is “We don’t know”. For the second question, “Probably not” would be an honest response. Garbage in, garbage out.

  22. I hope they do not start messing with the Met Oficc seasonal forecasts , we really do rely on them in Britain , if they say cold we know it will be warm , warm we know it will be cold etc etc , they still use the code they used to confuse the Germans in the last war , devilishly cunning and only really confuses foreigners !!

  23. “Well, they could start improving their forecasts by reading the Old Farmers Almanac (who seem to do a better job than most…) ”
    Has anybody else noticed that market-driven long-term forecasts tend to be more accurate than those from any government agency? It seems almost like the government agencies have some ulterior motive…

  24. From what I can see, there should a number of uplifts in temperature this Autumn, the mid and end of September, the mid and end of October, so roughly every 2 weeks, giving plenty of above normal conditions. November I expect to be very mild, and amongst the warmest on record.

  25. Part of thier mission is”to create knowledge”? I think a school mission should be to pass on knowledge or to discover knowledge. “Creating” knowledge sounds fishy to me.

  26. I started a comprehensive response to this paper, but by the time I got through detailing some of the thoughts and opinions that I thought they might find helpful, I had ranted too much. So I just started a new blog post over on my site, where I am still reworking ideas over for clarity.
    I will respond again later (with links) when I get it more polished. Too [****] long to drop complete into someone else’s blog respectfully.

  27. latitude says:
    September 9, 2010 at 7:32 am
    Rhys, this site rates them:
    http://www.forecastadvisor.com/

    That’s a good site. Just from a brief straw poll, it seems the further east you go, the less accurate the weather forecasts. It’s that “land” thing, messing stuff up.

  28. “there should a number of uplifts in temperature this Autumn, the mid and end of September, the mid and end of October,”
    Not sure where from the script. Here in central MN we’ve had an early start to fall since the 1st of Sept. Blocking Highs in Canada this winter will again result from the negative PDO and La Nina and we’ll see coldening most likely with a jet stream sweeping in around them from mid-Alaska.

  29. From the paper: One of the most important boundary conditions for an atmospheric model is sea surface temperature (SST). Variations in SST can heat or cool the atmosphere, influence the rainfall patterns, and thus change the atmospheric circulation.
    Yes and Roy Spencer would be happy to supply his thoughts.

  30. NAS report: “There is an apparent plateau in our ability to make accurate seasonal forecasts”
    I am not surprised. Climate “research” has been focused on how to prove CO2 regulates the climate for thirty or more years so politicians have an excuse to tax us. As long as there is no real research allowed into the peer reviewed journals, how the heck do you expect to progress instead of “plateau”
    This report gives a good indication that the corruption in the area of climate research is finally taking its toll in a concrete manner. It was only a matter of time before the pixie dust settled.

  31. Figure 1.4 from complete report, “Predictions from various statistical and dynamical models”, sums the report up rather nicely.
    They can’t do it and they don’t know why.

  32. Recap of their attitudes that lead to the problems they have now, and offer of assistance if they want it. Posted to own blog due to length, and I may be adding more soon.
    http://research.aerology.com/improving-long-term-forecasting/
    “However they seem to leave out the effects of the interactions of the Earth Moon system, as if the sun was just the overhead light that glares on their monitors, and the moon was not to be considered at all, due to past superstitions about allowing “astrology” into the sciences.
    Systematic blocking of all grant fund applications for the study of Lunar cycles, since about 1950, has suppressed the study of the real tidal forces, and the various orbital resonance effects of the moon on the weather. Hence the parts that they are missing in solving the weather puzzle are almost all answered by looking at the timing of the effects of the moon on the weather.
    Many of the cyclic periods, that they have a full pocket of acronyms for, are derived from the orbital dynamics of the Sun, moon, and planetary interactions affecting dynamic ocean / atmosphere oscillations, below I have pulled sections of their paper with comments and advice I could give them to fix the problems they are having.”

  33. Z says:
    September 9, 2010 at 2:20 pm
    latitude says:
    September 9, 2010 at 7:32 am
    Rhys, this site rates them:
    http://www.forecastadvisor.com/
    That’s a good site. Just from a brief straw poll, it seems the further east you go, the less accurate the weather forecasts. It’s that “land” thing, messing stuff up.

    I did the same thing, popped in a few cities and towns I’m familiar with around the country. The part that made me laugh the hardest was my home town had a highest accuracy rate of 66% (1/3rd or more inaccurate forecasts!) while LA was almost 90%.
    It’s almost like those areas where you can just guess “warm and sunny” every day and be right most of the time are easy… unlike New England, where if you don’t like the weather, wait a minute.

  34. gary gulrud says:
    September 9, 2010 at 2:45 pm
    “there should a number of uplifts in temperature this Autumn, the mid and end of September, the mid and end of October,”
    Not sure where from the script.
    …………………………………………………………………………
    As I am working with predicting short term changes in the solar signal, this has a global impact, and would be apparent in the majority of land regions that are not subject to incursions of colder polar air, by local jet stream patterns and movements;
    http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/regional_monitoring/
    Since mid January I have had no particular uncertainties, and forecasts for temperature deviations from normals for most regions (and the resultant changes in volume of precipitation) have been thoroughly on target.

  35. Not too long ago many of the most intelligent minds on the planet were trying to make gold out of lead. Today, they’re trying to predict climate by the century and having just as much luck at it. Before you can push the envelope, you must first know where you are, and where it is.
    PS: Human Progress is made via hard slogging work and is measured in inches over centuries, nothing of value is achieved via TV guest interview spots, or NYT articles, or by giving testimony to politicians, or by speeches to your likeminded backers. Mann-kind has its fools, super-fools and mega-fools and they’re all too busy to slog.

  36. David Wright says:
    September 9, 2010 at 5:25 am
    Reports like this would be greatly improved by tossing out most of the jargon.
    Reminds me of this prize won by the UK Met Office a couple of years ago.

    An attempt by the Met Office to justify the inaccuracy of its weather forecasts has been picked out as one of this year’s worst examples of gobbledegook, at the annual Plain English Campaign awards.
    By Matthew Moore
    Last Updated: 6:14PM GMT 09 Dec 2008
    The national weather forecaster was shamed for posting message on its website which explained in 45 words what could have been said in three: “We’re not sure.”
    “Seasonal forecasts indicate how slowly-varying large-scale climate influences make particular seasonal conditions more likely than others. Random, unpredictable factors (‘chaos’) also partly determine year-to-year variations, and these will sometimes override large-scale influences. Such uncertainty makes a probabilistic format, as used here, advisable for seasonal forecasts,” it read.

    Some might also be amused by this, thinks that there are those of a sceptical POV that think that “pollution” escapes to space through holes:
    “Prince Charles baffled by ‘extraordinary’ climate change scepticism
    The Prince of Wales said he found the views of climate change sceptics ”extraordinary”.”
    http://www.telegraph.co.uk/news/newstopics/theroyalfamily/7993756/Prince-Charles-baffled-by-extraordinary-climate-change-scepticism.html#dsq-content

Comments are closed.