As a follow up to our 4/19 story about ash cloud modeling:
Ash cloud models – overrated? A word on Post Normal Science by Dr. Jerome Ravetz
I offer below a compendium of articles from Benny’ Pieser’s CCNET and GWPF of the UK
Above:
The Met Office has been blamed for triggering the “unnecessary” six-day closure of British airspace which has cost airlines, passengers and the economy more than £1.5 billion.–The Telegraph 19 April, 2010
…
Global airlines have lost about $1.7bn of revenue as a result of the disruptions caused by the Icelandic volcanic eruption, a body has said. Giovanni Bisignani, chief executive of IATA, criticised governments for the haste with which they closed airspace, and called on them to provide compensation to the airlines. “Airspace was being closed based on theoretical models, not on facts. Test flights by our members showed that the models were wrong.” —BBC, 21 April 2010
…
We are becoming a risk-averse society and that is dangerous. You cannot run a national economy on the precautionary principle; indeed, the sound position is to embrace as much risk as possible. Societies that embrace risk, such as the United States and recently the UK, tend to thrive, while those that seek to minimise risk, such as Britain during the 1970s, tend to wither. Financial capital is now fleeing Britain, heading to the Far East. A long queue of companies is chasing the money, including our own Prudential, which is floating a business on the Chinese stock market. The true venturers are over there, not in Britain. —The Times, 21 April 2010
…
Rational decisions have to be taken on the basis of some empirical understanding of the risks involved, and on the balance between risk and reward (or the cost of avoiding risk). Exposing the nonsense and muddle of the so-called precautionary principle is an essential part of the GWPF’s declared mission ‘to bring reason, integrity and balance to a climate debate that has become seriously unbalanced, irrationally alarmist, and all too often depressingly intolerant’. If the argument now raging over the policy response to the volcanic ash clouds assists in achieving this, it will demonstrate that ash clouds, too, have a silver lining. —The Global Warming Policy Foundation, 20 April 2010
Glorious vindication over computer modelling and AGW. I’m glad you guys picked up on this. Of course, no pro-warmers will see this as a general issue with GW modelling…
Global airlines have lost about $1.7bn of revenue as a result of the disruptions caused by the Icelandic volcanic eruption, a body has said. Giovanni Bisignani, chief executive of IATA, criticised governments for the haste with which they closed airspace, and called on them to provide compensation to the airlines. “Airspace was being closed based on theoretical models, not on facts. Test flights by our members showed that the models were wrong.”
The same models that have been successfully used by the VACCs around the world to notify airlines which routes to fly to avoid volcanic plumes! The models that in this case were proved correct by test flights. Note that 5 days after the start of the ban airlines flew in the regions where the models indicated the plume was dissipating, they didn’t test flight over the Atlantic where the models and satellites indicated fresh, more concentrated plumes. It’s a load of self-serving rubbish, in an attempt to extract some money from the goverment!
If they believe what they’re saying have all the VACCs shut down or ignore their reports world-wide and then take the heat when a few hundred people die in an avoidable crash.
REPLY: Phil you are full of bollocks. Observation trumps modeling any day of the week, twice on Sundays. In weather forecasting, I rely on models every day, I also rely on observations.
If the forecast model called for, “high winds and severe thunderstorms today”, and I give my evening forecast and we have in the area nothing but “partly cloudy” I surely am not going to tell the viewers “stay indoors”.
Observation trumps modeling. Model outputs aren’t worth crap if the observation and the forecast don’t match. The issue is overreliance on models and not enough observation. – Anthony
Packard Bell withdrew from the American market in 1999, but has remained successful in Europe under the direction of NEC.
Many of you may remember the top seller of PC’s was Packard bell 10-15 years ago. It was of incredibly inferior quality, but it was cheaper than better names. With NEC, what could go wrong building a SUPER duper computor for The Met Office. Why am I not surprised?
Couldn’t The Met Office buy some little off the shelf Apple and use it to miscalculate?
@steveta_uk
Their models were proven wrong after only days of modelling, proven by satellite imagery. That is enough in itself.
Stop. The 1.7 billion pound loss is a loss to the top line. The people that missed flights still have tickets. The savings comes in fuel. There is no fuel burned to speak of when planes are grounded. They may have lost 20% of that higher figure. There was not a loss of 1.7 billion profit.
Some good businesses even have insurance called business interuption insurance.
Can’t believe I’m saying this: Nikita Kruschiev and the boys at the Ol’Kremlin had it right I guess – The West is kaputski! I guess he knew it was going to be only a matter of time before all those little spoiled baby boomer commiecrat boys and girls grew up and became good little socialists and ruined everything. Thanks Met, and all you little EU Mets, you achieved what the Ol’USSR couldn’t.
Just to compensate, UK has its brand new, and respected of course by us, Merlin Wizard, in Prof.Corbyn. The only difference is that he does not wears the expected attire, but his noble figure resembles the one of his predecessor in astral science.☺
That it was the Met Office aside, I don’t have a problem with computer models indicating a potential problem of this sort. I have a problem with using computer models alone to establish a problem and act on that basis the way they did. What should have happened after the Met Office got their results was to immediately send up some aircraft to verify conditions and to monitor the conditions that way.
That they didn’t is what is appalling. Some folks in government, other than at the Met Office, and rather high on the ladder, ought to be fired for it. As for the airlines, if I were a shareholder, I’d want a few heads to roll, too, just because they didn’t do so either.
stevengoddard (10:11:42) :
If a plane had of gone down, everyone would be screaming for their heads. The Met Office is in a no-win situation, and chose “better safe than sorry.”
No the aviation authorities made that choice, based on the data provided by the VAAC. The Met Office VAAC issued a report based on the data (it collaborates with the Icelandic Met Office and Nordic Volcano Institute to determine the initial condition for the calculations). There was nothing wrong with the predictions, it’s just that some with a financial interest didn’t like the decision based on them and decided to blame the messenger. If the airline execs really believe what they’re saying they should ask governments worldwide to ‘shut down all 9 VAACs since their reports are useless’ and take potluck about avoiding plumes in the future.
The airlines wont recover a penny because of the standard it would set for when AGW are found not have been error & the associated financial damages inflicted by “bad models”. The airlines’ loss would be a drop in the bucket by comparison.
And yet if a plane had gone down, it would have been a 5bln loss and 16 days down instead of 6. Oh, and a few hundred lives…
Is there any better way to try *guess* what the ash is going to do?
Scientific or not, forecasting is still a best guess, regardless of how powerful the computers are.
wsbriggs (10:12:02) :
This should cause a great deal of rethinking on the subject of computer-based weather modeling. I’ve got a strong suspicion that the code developed by SUNY Stony Brook in the early 2000s using GPU acceleration was substantially superior to this. Los Alamos also did codes for dispersal of aerosols prior to the Salt Lake City Olympics. Both sets of code were tested to verify the results.
Non tested code shows arrogance of the worst sort.
No, your arrogance in making statements of this sort is far worse. The codes used by the VAACs have been and are tested.
Did they get it wrong? Probably.
Did they get it wrong given the available data at the time? Maybe not.
I think it’ll take a very good look from an outside party to answer the second question appropriately.
-Scott
“World Airways Jet Grounded In Europe, Possible Engine Damage From Ash”
http://www.businessinsider.com/world-airways-jet-grounded-in-europe-possible-engine-damage-2010-4
I wouldn’t have expected the German government to fall for the total ineptitude of the British computer simulation craze without immediately checking the real world situation. But they did – and checked after 3 days. Our traffic minister Ramsauer is a total dork. You can’t have people like that run an industrialized nation.
Only the Greens could have performed worse.
Ayn Rand presaged the risk adverse nature of the government bureaucrat in Atlas Shrugged. It is being played out in the use of the precautionary principle on an ever increasing scale by government. They are afraid to make a risk/reward judgment since they can always hide behind the lack of risk engendered by the precautionary principle. We, as a civil society, cannot continue to thrive with a government that acts this way. Initiative will be stifled, scientific inquiry will be suppressed, and civilization will wither.
We need to break this cycle now for the future of our children and our grandchildren. We need to take civil society back from the clutches of public servants whose allegiance is only to their government employer and not for the good of the people.
Remember Lorenz (1963):
“When our results concerning the instability of nonperiodic flow are applied to the atmosphere, which is ostensibly nonperiodic, they indicate that prediction of the sufficiently distant future is impossible by any method, unless the present conditions are known exactly. In view of the inevitable inaccuracy and incompleteness of weather observations, precise very-long-range forecasting would seem to be non-existent.”
http://eapsweb.mit.edu/research/Lorenz/Deterministic_63.pdf
However, a numerical simulation of Navier-Stokes equations a couple of days or up to a week into the future may give accurate results, in particular with the power of modern supercomputers. Thus, I would expect that in this particular case the UK Met Office forecast may be useful. The irony is that the UK Met Office is able to do such useful very-short-range forecasts due to extensive funding as a result of the (non existent) manmade climate crisis.
@stevengoddard (10:11:42) :
The point is the Met Office consistently get it wrong, and fail to ever admit their flaws. If this was a one-off, we would give them the benefit of the doubt, knowing that they would learn and improve.
From my essay Precautionary Principle — Philosophical Implications
Or “to many idiots, to few tigers” (retreadresources.com/blog)
This all sounds like a good idea. One does not need to be 100% certain if human health or irreversible damage appears to be the result of something, to restrict that something. Put another way, in the absence of definitive science to the contrary, if in doubt ban or restrict it. If our hypothesis is: X is bad for me or others or perhaps other animals, it should not be used or ingested or what ever. We do not prove the hypothesis, we falsify it. Therefore if it can not be falsified, it should be banished. I have grossly oversimplified here. In most situations common sense generally identifies what action, if any needs be taken, at least on a personal level. Sometimes a good case can be made on a much broader scale such as societal or international levels. If one looks up the subject and reads through the literature, in general terms, the ideas expressed are probably acceptable to most. So are the general rules of conduct found in most societies, philosophies and religions. However, as expressed in the old aphorism, “the devil is in the details.”…..
This whole situation is nothing more then those who are charged with taking some responsibility are loath to do so. They will not do, so they want only the benefits of office and refuse to accept any possible risk. What then do we pay them for? If it is to do nothing that even suggests judgment, they are grossly over paid. If it is to use judgment they are also over paid, for they have not done so.
It’s easy to be wise after the event, and because of the circumstances, the VACC’s involved in this emergency did the right thing.
The models used to track the ash are designed to provide airlines which routes to fly which are free of volcanic plumes. Unfortunately, because of deterministic the deterministic chaos inherent in our turbulent atmosphere, they cannot give information on the density of the ash.
Test flight where done early to test for ash, which they found thus proving the predictions made by the models were correct.
In this situation I would much rather be someone on the ground wishing I was in the air, rather than be someone in the air wishing I was on the ground!
Rick (10:58:15) :
Is there any better way to try *guess* what the ash is going to do?
Scientific or not, forecasting is still a best guess, regardless of how powerful the computers are.
It’s far better than a guess, turbulent transport of particle fields is very well characterized and very well modelled. The agreement between the calculations and the satellite images illustrate this very well.
There is no reasonable proximity measure to be applied in grounding an entire continent’s worth of air traffic over a plume of ash that didn’t live up to a computer’s prediction.
This is what sensor’s are for.
When it’s freezing outside, you know what to wear according to the temp & wind.
Now the MET has itself caught in it’s own post-normal web.
There is probably a tolerable level of ash that planes can fly through where there is not too much damage or risk.
But we didn’t really know what that level is. How do you test such a thing when you are putting expensive planes and lives at risk. Even if there is a tolerable level, there is still going to be some damage and a reduction in the life-time of the engines and planes.
But there is less ash in the air now (there is still some over the English Channel and Italy according to Modis) and the volcano was less active yesterday and today.
Nice 24 hour animation of the volcano based on one of the webcams from astrograph (you might want to turn your sound down – astrograph has an eclectic taste in music).
briggs says April 21, 2010 at 10:12 am
What kind of GPU Acceleration was SUNY using in the early 2000s? Before CUDA in 2006, everything was slow, low precision and lacking in scatter capability. Both ATI and Nvidia were calculating intrinsics using lookup tables until, a couple of years ago. You couldn’t do IEEE32 math prior to that.
Lots of crapola flying about, including Anthony’s response to Phil (and I say this as a fan of WUWT and Anthony’s efforts).
1) Operationally useful ash observations amount to diddly squat. Lack of a signature on satellite is not evidence of absence. Damaging ash concentrations are entirely possible at levels well below what can be seen on satellite.
2) “Cope and avoid” has worked successfully for years. Most volcanic eruptions are short burps of ash lasting a couple of days at most. The long-duration ones are lower in altitude and/or less in volume. Having a sustained eruption affect most of European airspace is unprecedented. You can’t “cope and avoid” most of Europe for a week, thus the re-thinking that occurred Sun/Mon.
It’s like how aircraft treat thunderstorms. If there’s a few out in open airspace, you go around or you orbit until it moves off the airport – picking your way through the weak spots with no radar is a loser bet. But what do you do if there are thunderstorms everywhere? Well, you hope you have a radar on board. Right now we may need an “ash radar” if we’re going to pick our way through the thin spots. There isn’t one ready for wide service.
3) In addition to having practically no useful data on where the ash is/isn’t and how thick it is, there is little data on what is and is not safe. The “safe” level varies much by engine design – hotter, high-performance engines (fighters) have less tolerance than other engines. In the broad spectrum of engine designs in use by commercial airlines, what one engine can eat safely might kill another – we don’t effing know.
4) Air sampling by running it through a jet engine to see if any bits stick to the insides is not very good data. Not really what I would call an “observation” – one wouldn’t say with certainty the sky is cloud-free with an old-style automated “CLR BLO 120” report.
We have a lot of “known unknowns” at this point. Bashing the Met Office on their ash plume forecasts, when there’s a long list of other worthwhile bashes to be had, is foolish.
REPLY: I agree, there’s plenty to bash at the Met Office. Good summary on your part. My issue is overreliance on models and not enough observation. Rapidfire sat imagery may provide a substitute for “ash radar”. Anthony