Over-reliance on computer models causes a forecast for extreme heat to be retracted

From the “heat on the computer brain” department (h/t to Joe D’Aleo)

Portugal’s Met Office has retracted its prediction that temperatures in the country could reach 50ºC — the hottest ever recorded on mainland Europe — this week, drastically revising the forecast down by 10 degrees.

In a statement, the IPMA said forecasts published to its website and app on Tuesday had been “overestimated” for the region between Melides and Vila Nova de Milfontes, and in particular the city of Sines, where temperatures were predicted to reach 50ºC on Thursday and 46°C on Friday.

Facebook/ Comentadores

The forecasts were the result of a “statistical method” applied to numeric models, it said.

The IPMA said the actual temperatures expected for these days were between 40°C and 42°C.

“This situation arises from the exceptional temperature episode above the normal values ​​predicted for the beginning of August, after a long period of below-normal temperatures for the time of year,” the statement said.

Full story here

101 thoughts on “Over-reliance on computer models causes a forecast for extreme heat to be retracted

    • Evidently they’re aiming for the mean, so actual low temperatures cause high predictions. But they’re lying, because you know with 100% certainty that if it had been a hot summer so far, that they would be predicting more hot temperatures and not cool temperatures.

    • It sounds like they had below normal temps followed by above normal, so their “model” Drew an increasing line between them and assumed the increase would continue.

      • I see maximum for this week at 41 C. And close to Beja I see 42 C in Spain, so perhaps just over the border in Portugal it may reach that as well.

        Recently (July) is was unusually cold (20-25) whereas in Spain it was 30+ (as well as even 30(+) in the NW EU in many places). Rest of the year was below average.

      • Yes, claiming a record of 47c/48c when higher temperatures were recorded many decades ago is still fake news.

        Normal exposure was good enough for any station and still is now.

        Currently highs so far today have not been above 44c in Spain or Portugal.

  1. Sounds like Portugal’s Met office has the same level of credibility as the Australian Weather Bureau. Constantly creating alarm on hot extreme forecasting that doesn’t eventuate. I never see this outrageous forecasting for cold temperatures, it doesn’t fit the left alarmist narrative.

    • Same in the US of course. This past winter when we were having a period of below 0F (-18C) and sometimes reaching below -10F (-23C), the forecast was always for about 10F warmer. One day I saw the high temperature was below the forecast low.

      • Joe Bastardi points out that NWS forecasts in the US are often well above what actually occurs (but strangely, never below). The forecasts are from models. If models are that far off just a few days in the future, how much use are they for decades in the future? Rhetorical of course.

    • Sounds like you’re “constantly” making stuff up. The Australian BOM have are almost universally regarded with great respect for their integrity and skill. If you have any examples of “creating alarm on hot extreme forecasting that doesn’t eventuate” then what are they?

      • The same “universally regarded” AWB that took a record LOW temperature out of the figures because it didn’t fit the narrative.

        • Don’t you just love the drive by empty quips by these religionists? Hahaha I love it! They are caricatures of jokes of themselves.
          What an embarrassing lot. Must get old being a foregone conclusion

          • “drive by empty quips’. I like that, I think I’ll use it, or versions thereof…drive by insults, drive by lies, drive by histrionics…

          • I think you mean “oh yes, having asked for evidence and been shown it, I now graciously accept I was wrong”.

            Try that and you might get a thumbs up.

          • Evidence of “Constantly creating alarm on hot extreme forecasting that doesn’t eventuate”?

            I think you have either a comprehension problem, or a subterranean-low bar for evidence.

          • Logical Fallacy Alert! (Moving the goalposts)

            That’s not what you asked for, RyanS. You said, and I quote, “If you have any examples of ‘creating alarm on hot extreme forecasting that doesn’t eventuate’ then what are they?” The word “constantly” was not part of your request.

            I daresay it’s YOU who has a comprehension problem. You can’t even comprehend your OWN words.

      • How about homogenisation?
        “The Forum noted that the extent to which the development of the ACORN-SAT dataset from
        the raw data could be automated was likely to be limited, and that the process might better be
        described as a supervised process in which the roles of metadata and other information required some level of expertise and operator intervention. The Forum investigated the nature of the operator intervention required and the bases on which such decisions are made and concluded that very detailed instructions from the Bureau are likely to be necessary for an end-user who wishes to reproduce the ACORN-SAT findings. Some such details are provided in Centre for Australian Weather and Climate Research (CAWCR) technical reports (e.g. use of 40 best-correlated sites for adjustments, thresholds for adjustment, and so on); however, the Forum concluded that it is likely to remain the case that several choices within the adjustment process remain a matter of expert judgment and appropriate disciplinary knowledge. ”

        So in essence their adjustments can’t be replicated. If it can’t be replicated it isn’t science.

          • No it is what I equate with ” (T)the Australian BOM have are almost universally regarded with great respect for their integrity and skill”. Isn’t science about integrity, skill, replicability?

      • As I suspected, flimsy nothing burgers to back up the statement: “Constantly creating alarm…”

        Drive-by arm-waving and one link to an IPA shill… well done.

        • So you like to shoot down the messenger, not the message? Got it.

          Any proof of this “IPA shill? Any proof of anything?

          • Jennifer Maharosy is not mentioned as being on the payroll. She is mentioned as an editor of “Climate Change, the facts 2017”.
            “As a result of the work of Dr Maharosy, the Bureau of Meteorology announced in August a review of its procedures for the measuring and recording of temperature data in Australia”.

            That’s what has your tits in a bow? They reviewed their methodology?

            Of course you may be happy with the “data”. About 7.7 million sq Km. About 750 weather stations. Average grid size 10,200 sq Km. 😉

          • Uh huh, that is the only “evidence” cited to back up Wallaby Geoff’s arm-waving, ie none. So yet again we have….just arm-waving, opinion, stuff I heard at the pub, something I read on the internet – spouted out like it’s common knowledge.

            An echo-chamber filled arm-waving ignoramuses. Thats what you end up with when bs like that goes unchallenged. All too often…

          • Thank you for your elucidation. Your citing of climate facts that refute the point. Well done. /sarc

      • Record-challenging in southern Europe
        Ben Domensino, 2 August 2018

        Another heat record may be broken in the northern hemisphere this week as temperatures soar in southern Europe.

        A plume of hot air being carried north from Africa will push temperatures to record-challenging heights across Europe’s Iberian Peninsula in the coming days.

        According to forecast models, the highest temperatures are likely to occur over inland areas of Portugal on Saturday, where some locations are expected to reach the high forties and could nudge 50 degrees. Temepratures could also reach the mid-forties in western Spain.

        Portugal’s highest temperature on record prior to this event was 47.4 degrees in 2003, according to the Portuguese Institution for Sea and Atmosphere (IPMA). The highest temperature on record in continental Europe was 48.0 degrees in Athens, Greece during 1977, according to the World Meteorological Organisation.

        Parts of Portugal and Spain have been placed on red alert for extreme heat in the coming days. This is the highest category on Meteoalarm’s four-tiered weather alerting scale, indicating that weather conditions could be life-threatening over a large area.

        The impending heat in Portugal and Spain follows a number of other heat-related records in the Northern Hemisphere this summer. This includes new maximum temperature records in Finland (33.5C), Japan (41.1C) and Algeria (51.3C) during July

    • Forecasts for WX outside the norm one might think also disrupts energy production. If you were an energy producer/distributor and was told that a surge of HVAC usage would occur, do you start ramping up production by taking plants to full power, plan on buying energy elsewhere on the spot market, or just let rolling brown or blackouts occur?

    • Actually, not all weather offices look for breaking records. Our local weather office (PDX) is very leary of forecasting record high (or low) temperatures until they have no choice, and then they look for a reason for it not to happen. If you aren’t leary the chances you will get ‘burned.’ For instance, last year it was looking that we would hit or nearly hit the all-time record of 107F (42C) They posted in their forecast discussion that the possibility existed but they didn’t forecast for it. Guess what? Smoke from forest fires moved in and the high was only 91F (33C). Things like that and cool marine air moving in from the ocean faster than expected happen too often for our weather office to get too carried away with forecasting record highs especially all-time record highs.

  2. It sounds like computers out perform humans in most aspects of weather forecasting. link

    Having said the above, I sure wouldn’t bet my life on next week’s weather forecast, no matter how it was done.

    • If the weather forecast is from the Weather Channel, I wouldn’t bet on tomorrow’s forecast being correct.

      • I tracked the Weather Network and Environment Canada’s 5 day forecast accuracy for 3 years. The Median error was 2.1°C (Environment Canada too low, Weather Network too high). However, there were many forecasts that were way off. That was about 10 years ago, I should look at that data again.

    • Not many people grasp the reality…..until they have a computer that accurately predicts the weather…they will never be able to predict climate

      • Look at how much computers have developed since Edward Lorenz wrote his pioneering weather models in the 1960s. They’re a million times more powerful now than they were then. Now consider how much weather forecasting has progressed. Not that much.

        Lorenz pointed out that accurate long term forecasts weren’t even a theoretical possibility. Nobody proved him wrong. Yet, the weather forecasters seem to think bigger better computers are the answer to their problems.

        • The power is irrelevant to the problem. Adding more lines to an Excel model of anything doesn’t automatically make it more able to predict the future. The only thing that improves forecasts is better inputs. We should be putting all our efforts into the inputs, not the models or the computers.

          • The issue with the models is that regardless of how good the inputs are, the Earth’s atmosphere is a chaotic system and no amount of accurate inputs will allow a computer to accurately calculate the astronomical number of variables needed.

            Even assuming we could know every single starting variable for all the necessary inputs in our atmosphere AND we could know every single process taking place, we still wouldn’t be able to actually model it beyond a couple days at most (and that’s being really generous).

            Of course, that’s all well and good but there’s no way to know enough variables to do so because “enough” variables to accurately model it would mean knowing scattering patters for the radiation entering the atmosphere, exact positioning and density for all clouds, ect.

            We’re basically at a point where we know enough that we should realize we’ll never know enough to model the weather and by extension the climate.

            Our best bet for predictions is to look at pattern recognition where they pick a few major indicators (jet stream, ENSO cycle, solar cycle, ect) and just make predictions based on previous similar patterns. It’s still not going to be very accurate but probably better than attempting to model something that inherently can’t be modeled.

      • For that matter, while their climate rules can’t replicate the past ice ages from first principles, no way are they gonna replicate any future climate.

      • Not at all. The two are forecasting entirely different things. I can be pretty accurate forecasting the number of matches Chelsea will lose this season but hopelessly inaccurate which ones.

        And if I get the weather too hot in one place and too cold in another, my global temperature may still be correct.

        • This is what we refer to as a blind squirrel finding a nut. You still wouldn’t bet on it surviving the winter.

  3. Insufficiently characterized and unwieldy is the recipe for chaos and scientific divergence. He needs to specify a constraint for the frame of reference that improves the hypothesis’s accurate representation of reality as observed.

    • “Nassim Nicholas Taleb writes about “Monte Carlo generators” in his 2001 book Fooled by Randomness as a real instance of the reverse Turing test: a human can be declared unintelligent if his or her writing cannot be told apart from a generated one.” link

    • “… that improves the hypothesis’s accurate representation of reality as observed.”

      While sounding technically sophisticated, this is junk speak. My old mentor would have thrown me out of his office if I’d ever given him a hypothesis revision justification like that. Science is about clear communication. Communication that obscures easy comprehension is usually trying to pull a fast one on the reader.

      Adjusting the hypothesis is certainly okay in some approaches, especially early in new studies with novel observations, but it has limits in informing as to the natural truth being sought. When we find the hypothesis consistently being re-written (“improved”) with each new observation, like adding or adjusting new epicycles to an astronomical model, as climate science seems it must with CO2 CAGW, then it is likely a failed paradigm underpinning the hypothesis.

  4. Many pundits have been describing the recently warm spell, which is nothing special, as “as hot as hell”.

    I’ve been to Luxor in southern Egypt in July, and it was very hot (a real ~50C), but that was only as hot as Heck.

    Hell, that place where warmist fraudsters and scam artists will ultimately reside, is considerably hotter.

    We visited the remarkable temples and tombs of Luxor and area in the very early mornings, and returned to our hotel by noon to sit in the pool and drink quantities of a skunky Egyptian beer, misnamed Stella.

    One very large fellow luxuriated in the pool like a basking hippo, and loudly bellowed “STELLA!”, like Brando in “A Streetcar Named Desire” every time he wanted a cold one. It was very funny, at least the first few times. You had to be there…

    • We had to read (the script), listen to (on LP), and watch (on VHS) “A Streetcar Named Desire” in high school. I always yell “STELLA!” whenever near someone named Stanley or Stella, or someone orders a Stella beer. It makes me laugh.


      Stella, the only beer I have found I don’t like. Probably goes back to my 20’s when I was on my one and only 20 pint bender and the last beer I had was a Stella before being violently sick. Possibly a bit unfair on Stella but 40 years on I still can’t stand the stuff.

      • Hi HotScot,

        To be fair, Egyptian Stella is a much different brew from Belgian Stella Artois. The latter is drinkable, whereas the former was shipped and stored warm, and was served cold but was already skunky – it smelled of skunk as soon as it started to warm up, which in 50C heat was almost instantaneous.

        Drinking Egyptian Stella was kind of like a fire drill – the waiter would run the cold bottle out to the pool, open it beside you, and you would drain it in one gulp. If you did so in two gulps, it had already started to stink.

        The only thing worse than Egyptian beer was their red wine – absolute paint remover. Some other North African wines, for example some Tunisian varieties, were actually not that bad.

        • I notice Heineken, tho pretty good otherwise, has a skunky aftertaste.

          I’ve pretty much left lagers in favor of the IPAs (especially the double IPAs), tho some might say you have to learn to like them.

  5. …. daily weather isn’t but the climate change might be ….
    … solar activity collapsed in July, with the ‘classic’ sunspot count (Wolf SSN) reduced to just one point (1.1) while the new SIDC reconstructed number is at 1.6
    Composite graph is here
    SC24 is nearing what might be the start of a prolong minimum (possible late start of SC25 too) but even a ‘dead cat bounce’ from these levels is unlikely.

  6. Can a coastline even reach 50C? I would think those temperature extremes could only be reached by inland areas.

  7. Portugal is a small country, less than 100 miles from East to West at the widest point and little more than 50 miles across in the South. Going from South to North in less than 100 miles there is 32C to 50C and then back down to 42C.
    In accountancy before publishing forecasts it is normal to sense-check the data. Simple checks could save a lot of embarrassment. However, lack of checking by the experts justifies a large public relations team.

    Na contabilidade antes de publicar as previsões, é normal verificar os dados. Verificações simples podem poupar muito constrangimento. No entanto, a falta de verificação pelos especialistas justifica uma grande equipe de relações públicas.


  8. I’ve been noticing NOAA seems to be doing the same things in the US forecasts for maximum temps.
    The actual daily maximums are not even within 5 F of where they were forecast like 4-5 days earlier. By day 2 or 3 they are coming down, but still high.

    Are they purposefully trashing their forecasting skill to sell propaganda?

    • Rhetorical question I presume.

      I recall in late June, we were told that the temperature would be 102F in 5 days. Eventually, it was 90F. Even the day before they had it at 96F.

      • Yeah, Rich, I saw the same thing. Here in rural western MD, the very highest summer temp is usually 89F to 91F or so, unless it becomes very dry. This summer is extraordinarily wet — nearly 30″ of rain since the first of May.

    • And you are expecting the forecasted high/low 3-5 days down the road to be accurate? You must be one of those climate modelers.

    • joelobryan

      The precautionary principle? If they predict far worse than expected they can’t be blamed for not warning the public if something extreme, and unusual, does happen.

    • Yes, Joel — seen it myself. Some have suggested NWS and NOAA are slowly, incrementally adding more & more CO2 sensitivity into the weather models. Maybe warmer forecasts will convince the credulous that it’s getting warmer…..

    • USA NWS weather forecasts tend to run ‘way too hot. Here is one example:



      One of my friends and co-authors is Joe d’Aleo, an American Weather Forecaster who was the Founding Chief Meteorologist for the Weather Channel.

      The National Weather Service (NWS) of the USA forecast a warm winter for 2014-15 and Joe told me in October 2014 that the NWS forecast was seriously incorrect, and that the next winter would be particularly cold and snowy, especially in the populous Northeast. This was the second consecutive year that the NWS has made a highly incorrect (excessively warm) Winter forecast, in Joe’s opinion – and he and his colleagues at WeatherBell have an outstanding track record of accurate forecasts.

      Joe and I had been working together on a paper on Excess Winter Mortality, and I suggested to Joe that this false “warm winter” NWS forecast was dangerous, especially if the country and its people were unprepared. Joe agreed, but did not know how to tackle the problem.

      I proposed an approach, and we prepared a presentation for my friend at the US Energy Information Administration (EIA). At the EIA’s request, Joe then prepared his own Winter Forecast by month and by region, and the EIA re-ran their Winter Energy Demand calculations. Using Joe’s forecast, the EIA projected 11% more total winter energy required for the entire USA than the “warm” NWS forecast had projected. That is an awful lot of energy – mostly oil, natural gas and coal.

      After that brutally cold and snowy winter, a back-analysis showed that the actual winter energy used was 10% more than the projection using the NWS weather forecast, and just 1% less than that using Joe’s forecast.

      I’m not sure if we saved any lives, but I still think we did a good deed.

      Regards, Allan

  9. 40 C is equivalent to 544 W/m^2
    50 C is equivalent to 617 W/m^2, or about 73 W/m^2 of error

    This is a lot of error for a model that must be conserving energy at some point …

      • Phillip,
        With the BBC, I think that is – more usually – ‘dangerously average.’
        Or ‘disastrously’ average’ . . . . .
        Threats to knife/fork ratio, traffic light sequences, and – possibly – three o’clock.

        It certainly is worse than they think!

        [Mods: the above has a hint of /SARC, actually, on average.]

    • Obviously not a like for like cmparison. 59C would have smashed the warmest temperature for anywhere by several degrees. But the 2004 report is talking about the temperature of the ground, it says that air temperatures were above 40C.

  10. My understanding is they were using the US GFS rather that the euro model. The US model has a warm bias of anywhere between 7 and 10 degrees.

  11. https://nspires.nasaprs.com/external/viewrepositorydocument/cmdocumentid=611944/solicitationId=%7BE2CB9318-72CB-C51A-6962-013E762AE713%7D/viewSolicitationDocument=1/FULL%20ROSES%202018%20Amend21.pdf

    The following is part of the scope of the program on NASA wanting to study the earth’s global water system. Of course NASA has to make things complicated and combine energy with it. So they called it NEWS (NASA Energy and Water Cycle Study. It started in 2003 and they havent had much success trying to prove that there is any more water vapour in the atmosphere or that the water cycle is changing. BUT they keep trying. Of course all you skeptics realize where this is going. If they can find changes in the water cycle; of course they will tie this back to Mr. CO2. Mr CO2 has had many close calls with the IPCC police but NASA WANTS TO LOCK HIM UP FOR GOOD. So have a look at one of their latest requests for solicitation of proposal which closes in September 2018.

    “Appendix A-22 NASA Energy and Water Cycle Study”

    “Description of Solicited Research”

    “1) “To answer a long standing motivational question to the NEWS community, “Is the Water
    Cycle Accelerating?”
    2) we must be able to elaborate on and respond to the related
    questions of “How and Why is the Water Cycle changing?””


    Here we go again. The IPCC did not do the null hypothesis on CO2 and now NASA are following in the now time honoured tradition of 97% of climate scientists by again throwing away even the possibility of a null hypothesis before the study commences. The proper scientific method is to do the null hypothesis on the statement “is the Water Cycle changing” and then if after experimentation you can logically throw away the Null hypothesis, then you can proceed to the how and the why in another study. After that you can then proceed to a further study on the 1st statement “Is the Water
    Cycle Accelerating?”. I guess, to save time and money; NASA is just assuming it is changing because in their minds; who needs a null hypothesis anyway? It just throws roadblocks into the NASA scientific process.

    Of course NASA has to do their own thing. They are in the last phase of of a 15 year study(it will be lengthened because they are running years behind) on the hydrological processes of the globe. There are 2 other organizations doing the same thing. The World Climate Research Program (WCRP) has as its core mandate (The Global Energy and Water Exchanges (GEWEX) project) to study the same thing, as well as the US Climate Change Science organization also has a program with respect to the hydrological system; NASA as always has to duplicate or triplicate things.

    Funding for the The World Climate Research Program (WCRP)
    1) World Meteorological Organization (WMO)

    WMO is a specialized agency of the United Nations, which is dedicated to international cooperation and coordination on the state and behaviour of the Earth’s atmosphere, its interaction with the land and oceans, the weather and climate it produces, and the resulting distribution of water resources. WMO also hosts the WCRP Secretariat.

    2) Intergovernmental Oceanographic Commission of UNESCO (IOC-UNESCO)

    IOC-UNESCO is the United Nations marine science organization. The purpose of the Commission is to promote international cooperation and to coordinate programmes in research, services and capacity-building, in order to learn more about the nature and resources of the ocean and coastal areas and to apply that knowledge for the improvement of management, sustainable development, the protection of the marine environment, and the decision-making processes of its Member States. Find out more on the IOC-UNESCO website.

    3) The International Science Council (ISC)

    ISC was created in 2018 as the result of a merger between the International Council for Science (ICSU) (previously a sponsor of WCRP) and the International Social Science Council (ISSC). It is the only international non-governmental organization bringing together the natural and social sciences and the largest global science organization of its type with a vision to advance science as a global public good. You can find out more on the ISC website.

    4) Implementing Partners and Volunteer Contributors

    WCRP funding is provided by its co-sponsors (see above), as well as from the voluntary contributions of certain countries. It enables the Programme to initiate, encourage and stimulate high-priority climate system research. Our effort is substantially supported by joint sponsorship, matching funds/support and collaborative implementation by key partners who share the scientific interests of the WCRP community.

    Considering a financial contribution to WCRP in 2018?

    In 2017 the countries that provided additional financial support were:

    Australia, Austria, Canada, China, Finland, Germany, Japan, New Zealand, Russia, United States of America

  12. Humans have a bad habit of initially believing whatever they are told, then maybe scrutinizing it later. Thus, they believe an exaggeration of TEN degrees, even if it gets retracted. The seed is sown that there was this super hot day and people will then be susceptible to other lies such as a minor trace gas that comprises just .04% of the atmosphere is the driver of global temperatures/ climate/ weather/ whatever witch hunt of the day. The retraction covers for the ramifications of the lie (ie when the People find out they have been lied to and swindled for TRILLIONS being beaten to death with baseball bats.) Hey, they retracted the false prediction, so have covered themselves. The lie, however has persisted even though the retraction exonerates them of blame.

    We have seen such exaggerated predictions in Australia- super hot days and phantom cyclones that were just boring old rain depressions with damage being what you expect from some heavy rain. A couple of the super hot days were quietly retracted and there were no reports of damage and further hype related to the phantom cyclones. There was a recent prediction of some mega cyclone in the pacific. Looking at the satellite, there was no mega storm-little bit of rain, but nothing special. There were no reports of damage. How could anybody be that incompetent?

    Fake (climate) news has been occurring more frequently and around the world. A couple of years ago, there was a prediction of New York going to have a mega snowstorm. On the radar, it looked just like a rain depression we had in Sydney around the same time. It so happened that family members were there, in New York at the time. The report on the ground- a few inches of snow-what do you expect in the middle of winter in New York. It did not prevent hype, airport closure and the usual hyperventilating and hand wringing.

    Yes folks, we are going to cop more of these wild exaggerations of extreme weather. You do have to wonder how they can be 10 degrees Celsius off the money. Either gross incompetence or politically motivated lies. I am inclined to believe the latter.

    • Phillip

      Their reasons will probably include that of caution as there are so many holidaymakers from the UK pouring into the country over the next few weeks and “We must ensure they are prepared for the worst”.

      Like anyone who dares leave our sceptred isles can’t think for themselves.

  13. “This situation arises from the exceptional temperature episode above the normal values ​​predicted for the beginning of August, after a long period of below-normal temperatures for the time of year,”

    I haven’t seen anybody in the media comment on the long period of below normal temperatures?

    I wasn’t even aware that Portugal had suffered that.

  14. I just logged out of BoM weather pages on Elders weather
    theyre running that story of 50 or more panic!
    wonder if theyll adjust it?
    bet they will NOT mention it was models or an utter ballsup

  15. Funnily enough, the climate-change-doom-mongering BBC are going with it, despite the obvious inaccuracies. Even funnier, they have switched off comments for their agenda-fulfilling article, so I cannot even share a link to this one pointing out the (yet another) model-based overheating error…



  16. Even if it did break a record this week , so what ?! Isnt that just weather ? Ya know like when we have the coldest winter on record like we did a few years back. Or is it only attributed to Global Warming when its hot ?

Comments are closed.