UK Met Office maps of 2070-2100 weather

[Mike Smith]

From the EPA's web site:

How do we know that natural sources are not responsible for ozone depletion?
While it is true that volcanoes and oceans release large amounts of chlorine, the chlorine from these sources is easily dissolved in water and washes out of the atmosphere in rain. In contrast, CFCs are not broken down in the lower atmosphere and do not dissolve in water. The chlorine in these human-made molecules does reach the stratosphere. Measurements show that the increase in stratospheric chlorine since 1985 matches the amount released from CFCs and other ozone-depleting substances produced and released by human activities.
Back to top.


What is being done about ozone depletion?
In 1978, the use of CFC propellants in spray cans was banned in the U.S. In the 1980s, the Antarctic "ozone hole" appeared and an international science assessment more strongly linked the release of CFCs and ozone depletion. It became evident that a stronger worldwide response was needed. In 1987, the Montreal Protocol was signed and the signatory nations committed themselves to a reduction in the use of CFCs and other ozone-depleting substances.


So, while the amount of CFC in the atmosphere has been going down, the ozone hole has reached record levels.


I don't think the writer of this (among many others) article thought the ozone hole would reach record levels in 2006:

Earth's ozone depletion is finally slowing
30 July 2003
NewScientist.com news service
Gaia Vince


Almost 30 years after it was first reported that pollutants were destroying the Earth's protective ozone layer, there is clear evidence that the global CFC ban has had an impact.

For the first time, it has been shown that the rate of ozone depletion in the upper stratosphere - 35 to 45 kilometres up - is slowing down. "This is the beginning of a recovery of the ozone layer," says Michael Newchurch, at the University of Alabama in Huntsville, who led the new research.

Protecting the earth's ozone layer is a noble goal. And, on balance, banning CFC's may turn out to be a good thing -- the jury is still out considering that most thought the record size of the ozone hole was reached in September, 2000 and it would shrink from there.

All I am saying is that the skeptics (Fred Singer for one) about human effects and solutions regarding the ozone layer have turned out to, at minimum, have had a valid point.

Jeff, the original study that NCAR replied upon came from the Max Planck solar institute in Germany and was released August 2, 2004, and NCAR accurately reported on it in the passage you cite. Interestingly, there is another Planck study released October 28, 2004, that says the sun is burning at its brightest in 8,000 years -- a study that a lot of global warming advocates seem to have ignored. This study says how much of the current warming of the earth is due to the hotter sun is an "open question." And, given that we do not understand the radiative transfer between clouds and the non-cloudy atmosphere, I would say that calling it an "open question" is putting it mildly.

The atmosphere is behaving in some ways that were unexpected five years ago (record ozone hole and cooler oceans). Given that global warming appears to be a far more complex problem than the ozone hole, healthy scientific skepticism seems in order.

That said, this is a great example of a civil and informative scientific discussion.

Mike

 

[Mike Smith]

The sun is the "warmest" in 8,000 years. We both agree that radiative transfer is not well understood. We both agree the ocean interface is not well understood. And, I think we agree (correct me if I am wrong) that if we ran the climate models and asked them to forecast the mean conditions from Day 150 to Day 180 they would probably fail far more often than they would succeed (otherwise, why doesn't NOAA use them in their extended outlooks?).

Given the above, I don't understand the basis for the following:

In the meantime, basic physics suggests anthropogenic climate change is certainly a valid claim, which is decreasingly becoming susceptible to being proved wrong."

Chris

I'm sincerely interested in how you come to the conclusion that anthropogenic climate change is "decreasingly ... susceptible to being proved wrong.

Look forward to your response.

Mike

 

[Mike Smith]

Chris,

Thanks so much for a thoughtful response. You bring up some good points that I will try to answer.

If the sun is the warmest it is in 8000 years, how does this explain the "rapid" increase in global temperature over the last couple of decades?

Quoting from the October, 2004, Planck study: solar activity has remained on a roughly constant (high) level since about 1980. The maximum in solar activity pretty much corresponds to the rise in temperatures as plotted by the IPCC.

Do you think the article I linked above, which offers a pretty compelling explanation, is completely bogus, and then, if so, why?

Very good question. That particular article escaped by attention, so I did a quick read. One of things I noted is that this study, like so many others, relies on what I call "fudging." Starting on p. 1646, they discuss that the model did not handle the Arctic Oscillation well. So, "adjustments" are made. I realize this is commonly done in modeling but is it good science in the context of predictive science?

Remember, we are attempting to validate a forecast of future conditions using the past. If we have to fudge to get the right answers, why do we believe the "unfudged" model makes an accurate forecast 70 or 100 years into the future? That does not scientifically follow in my view.

The fact that a typical climate model wouldn't probably get an ENSO or an MJO (if it can even successfully simulate that ) exactly right in 150 days implies that we should apply such skepticism in our interpretation of the output 150 years out, although as models begin to resolve convection (edit: and improve ocean coupling), these uncertainties decrease. Bingo! We all agree that the models can't get even average conditions right at 150 days so we must be skeptical at 75 or 150 years. I don't doubt the models will get better. When they can get the 20th Century right without fudging and can make consistently accurate 6 month or one year forecasts call me -- I will be intensely interested in the results.

Finally, However, it should be pointed out the British data shows a leveling in world temperatures since about 2000 as opposed to the IPCC's steady rise. Why the difference? Is that difference significant? Maybe.

You have probably seen that the Russian Academy of Sciences recently predicted major global cooling starting in the next few years (starting 2012-2015 and peaking in 2055-2060) and then global warming in the early 21st Century. They compare the magnitude of this cooling to "The Little Ice Age." The Russian Academy adds, The Kyoto initatives to save the planet should be put off until better times... The global temperature maximum has been reached on earth and the temperature will decline to a climatic minimum even without the Kyoto protocols, Ahdusamatov said.

His contention that the global maximum has been reached is consistent with the British data. Bill Gray is also calling for major cooling to begin in the next few years. The recent, and unexpected, ocean cooling be a sign they are corrrect (I have no opinion on this).

For one moment, humor me and lets assume the Russian Academy and Bill Gray are correct and major cooling is imminent. Now superimpose former Vice President Gore's rhetoric about warming being "settled science" and "This is not a political issue. This is a moral issue -- it affects the survival of human civilization," Gore said in an hour-long speech at the New York University School of Law. "Put simply, it is wrong to destroy the habitability of our planet and ruin the prospects of every generation that follows ours."

Now, assuming this major cooling starts 2012 and by 2020 the canals in the Netherlands are frozen (as the Russians predict) and the U.S. suffering through one bone-chilling winter after another. Do you think no one will remember that global warming was "settled science"? Do you think science in general and atmospheric science in particular will have one shred of credibility left? Then, when happens when there is a true atmospheric crisis that we can predict? Will we be the science that cried "wolf"??!

I am older (54) than most of the people who post on StormTrack. I do it because I see a lot of intelligent people and I hope some of my perspective might be beneficial. I readily admit I am a skeptic about long range forecasting models and long range forecasts in general. I was a professional meteorologist in the late 1970's when "global cooling" was settled science.

Let me leave you with two quotes from "Time" magazine...

1974: "Areas of Baffin Island in the Canadian Arctic, for example, were once totally free of any snow in summer; now they are covered year round."

2006: "Late last year, for example, researchers analyzed data from Canadian and European satellites and found that the Greenland ice sheet is not only melting, but doing so faster and faster, with 53 cubic miles draining away into the sea last year alone, compared to 23 cubic miles in 1996."

Do we really want to risk our credibility and literally all the money in the world ($1,000,000,000,000) on the current generation of meteorological models? Is that good science? Is that good policy? Could that money be invested in ways that are more likely to have a positive return?

All of us much reach our own conclusions. I just want those conclusions to be based on sober readings of the best science possible with open eyes based on the limits of contemporary predictive geoscience.

Respectfully,
Mike

 

[Mike Smith]

Chris,

They are not talking about the thermohaline circulation, but changes in solar energy. See below. Have a good weekend.

Mike




Forget Kyoto— global cooling is on the way
[25.08.06 19:31]

Russian scientist predicts global cooling, a cooling of the Earth which could involve glaciation.

Global cooling could develop on Earth in 50 years and have serious consequences before it is replaced by a period of warming in the early 22nd century, a Russian scientist said Friday.


Global cooling - though never widely supported - is a theory postulating an overwhelming cooling of the Earth which could involve glaciation.

"On the basis of our [solar emission] research, we developed a scenario of a global cooling of the Earth's climate by the middle of this century and the beginning of a regular 200-year-long cycle of the climate's global warming at the start of the 22nd century," said the head of the space research sector of the Russian Academy of Sciences' astronomical observatory.

RIA Novosti reports that Khabibullo Abdusamatov said he and his colleagues had concluded that a period of global cooling similar to one seen in the late 17th century - when canals froze in the Netherlands and people had to leave their dwellings in Greenland - could start in 2012-2015 and reach its peak in 2055-2060.

He said he believed the future climate change would have very serious consequences and that authorities should start preparing for them today because "climate cooling is connected with changing temperatures, especially for northern countries."

"The Kyoto initiatives to save the planet from the greenhouse effect should be put off until better times," he said, referring to an international treaty on climate change targeting greenhouse gas emissions.

"The global temperature maximum has been reached on Earth, and Earth's global temperature will decline to a climatic minimum even without the Kyoto protocol," Abdusamatov said.

 

[Scott Olson]

Just how large this influence is, is subject to further investigation. However, it is also clear that since about 1980, while the total solar radiation, its ultraviolet component, and the cosmic ray intensity all exhibit the 11-year solar periodicity, there has otherwise been no significant increase in their values. In contrast, the Earth has warmed up considerably within this time period. This means that the Sun is not the cause of the present global warming.

http://www.maxplanck.de/english/ill...tion/pressReleases/2004/pressRelease20040802/

That Russian claim seems to be at odds with other studies that have been published and reviewed. Even the IPCC took a look at this issue and decided that solar output would not explain the recent warming even if it were to be amplified by a feedback mechanism. Now, It certainly does look like increased solar forcing has played a role but the degree of that role is the question and I think based on what is currently understood, it would not account for the level of warming.

Do we really want to risk our credibility and literally all the money in the world ($1,000,000,000,000)

I have no idea where you came up with that number but I'd be skeptical at any attempt to put a price tag on significantly curbing C02 emissions. Investing in cleaner technologies and better energy sources isn't going to tear the worlds economy apart. So let's say that number is true, tnat number only represents less then half of single years US budget. That's an insignificant number compared to climatic impacts that WILL and ARE resulting as a result of inaction.

 

[Mike Smith]

Scott,

As I pointed out in Post #16, the results you quote are from the earlier study. The later study (which, as I pointed out, seems be ignored by global warming proponents) says the effect of the "warmer" sun on the earth's climate the last 20 years is an "open question." To my knowledge, no one questions that the drop in temperature leading to the Little Ice Age was caused by changes in solar energy as was the subsequent warmup (which some believe we are still experiencing). If the sun could cause that much warming for 300+ years, why can't 20 years of the "warmest" sun in the last 8,000 years cause the incremental warming since about 1980?

I suspect that if we currently had the climate of the 17th Century (Little Ice Age) and a major warming was forecast there would be plenty of people who would be fearful of it. The change in climate from the 17th Century to early 21st Century was major -- yet somehow the world not only survived but prospered.

You seem to have a great deal of faith in unverified 70 year model forecasts. You also seem to have a great deal of faith in the IPCC process in spite of resignations from reputable meteorologists (Chris Landsea) and findings of bias (British House of Lords). The British data shows that global temperatures have been flat since 1998. Moreover, there are credible charges of bias in the IPCC's temperature statistics since that time. All of this can be found with just a Google search.

I point this out because of your comment That's an insignificant number compared to climatic impacts that WILL and ARE resulting as a result of inaction. I just don't see that the current uncertainties lead to that conclusion.

Thanks for your thoughts.

Mike

 

[David Wolfson]

I'm no expert on solar climatology, but I do know this: the accurate measurement of solar insolation impinging on the Earth is difficult and only really possible from space in the last twenty years or so. There is no way scientists could say positively that "the sun is the 'warmest' in 8,000 years", if they mean by that solar output.

In fact they can only be drawing inferences from measurements and natural evidence taken on the earth's surface. So the correct assertion is that "the most solar radiation in 8,000 years is reaching the earth's surface".

The amount of insolation reaching the surface is, of course, dependent on solar output; but it's also dependent on photochemical properties of the atmosphere that reflect and/or absorb solar energy before it reaches the surface. Atmospheric reflection and absorbtion are something like two orders of magnitude greater than the alleged solar variability.

Given this, IMHO the Russian hypothesis, if anything, is more cause for concern. It means that for whatever reasons there's more solar energy reaching the earth's surface than before, but those reasons are most likely related to atmospheric effects that may not reverse or may themselves be a consequence of anthropogenic warming. FWIW.

 

[Mike Smith]

David,

Thank you for your thoughts.

You are, of course, correct that they have to use a form of proxy data to estimate solar flux back 8,000 years. When I was in dynamic meteorologgy class, circa 1973, when we were learning the equations of the atmosphere, lower case "c" was the "solar constant." It was thought then that the amount of solar energy reaching the earth was constant on the time scale of decades and centuries. We now know that is not the case.

That said, climatologists have to use proxy data to estimate the earth's climate prior to about 1880-1900. There is a lot of room for error with both types of proxy data.

This is my main point in this thread. A lot of "settled science" and definitive statements about the future state of the atmosphere are thrown about in the global warming debate that seem to have a rather tenuous scientific basis.

The classic scientific method says that hypothesis should be proven though careful experimentation with repeatable results. If you accept that definition of scientific method, please consider the following....

1. The models cannot reproduce known conditions in the 20th Century without "fudging."
2. The models cannot make any kind of reliable forecast at 6 months or 12 months, which, to me, calls into question their ability to accurately forecast at 60 years.
3. We have no theory as to why the sun's energy varies, let alone a method of forecasting those changes.
4. We don't understand many atmospheric and atmospheric-oceanic feedback processes.

Given 1-4, how can we make definitive statements?

5. We want exclude (Kyoto) the two largest polluting nations (China and India, see last week's "Wall Street Journal" article on the level of pollution in those nations) yet somehow control greenhouse gasses at tremendous costs (trillions or even one quadrillion dollars depending on who is doing the estimating). Even the proponents of Kyoto (see 1998 IPPC document) believe it would lower earth's temperature by 0.03degC by 2050. I do not believe, even by 2050, we will be able to detect a change in worldwide temperature of 0.03 deg. C which makes the Kyoto process a scientific challenge to measure (assuming its adopted).

6. Medicine says, "First, do no harm." If the Russians and Dr. Gray are correct that the real danger is cooling due to solar influences and if proponents of global warming are correct (one to one relationship between temperature and greenhouse gasses), are we doing the wrong thing by spending money to eliminate the greenhouse gasses? Are those gasses an "insurance policy" against catastrophic cooling? I don't know. And, no one else does.

If one could clean up the atmosphere for $500,000 and guarantee no negative economic effects it would be a slam dunk decision, regardless of solar forecasts as that would probably be a positive cost-benefit decision. That is not what we are talking about. We are talking HUGE sums of money and potential serious economic disruption.

The CFC industry was an $8 billion (1980 dollars) enterprise prior to the ban so that we could "fix" the ozone hole. People lost their jobs. As of today, it is certainly not clear whether that $8 billion loss and the concurrent economic disruption will be a net positive as the ozone hole is larger than ever (see posts above). Doesn't this suggest we need to be very cautious about the strategy of overturning industries based on atmospheric processes that are not fully understood?

Can we really say, based on today's models, that investment/economic disruption at least 100 times that large will be a net positive? It seems an awfully big bet based on rather flimsy evidence.

Perhaps I have a different perspective, one of business, where we know there is a limit to resources and we have to make cost-benefit-based decisions. It is hard for me to see that either the science or economics tip in favor of making this bet.

Again, thanks for your thoughtful comments. I certainly would like to learn your response to these items.

Mike

 

[Scott Olson]

Mike, there really isn't anything left to say to your points since Jeff, Aaron and Chris have all had excellent responses, I reccomend re-reading those posts because it has already been covered. Anyways, I think we can agree to disagree here.

Scott,

As I pointed out in Post #16, the results you quote are from the earlier study. The later study (which, as I pointed out, seems be ignored by global warming proponents) says the effect of the "warmer" sun on the earth's climate the last 20 years is an "open question." To my knowledge, no one questions that the drop in temperature leading to the Little Ice Age was caused by changes in solar energy as was the subsequent warmup (which some believe we are still experiencing). If the sun could cause that much warming for 300+ years, why can't 20 years of the "warmest" sun in the last 8,000 years cause the incremental warming since about 1980?

I suspect that if we currently had the climate of the 17th Century (Little Ice Age) and a major warming was forecast there would be plenty of people who would be fearful of it. The change in climate from the 17th Century to early 21st Century was major -- yet somehow the world not only survived but prospered.

You seem to have a great deal of faith in unverified 70 year model forecasts. You also seem to have a great deal of faith in the IPCC process in spite of resignations from reputable meteorologists (Chris Landsea) and findings of bias (British House of Lords). The British data shows that global temperatures have been flat since 1998. Moreover, there are credible charges of bias in the IPCC's temperature statistics since that time. All of this can be found with just a Google search.

I point this out because of your comment That's an insignificant number compared to climatic impacts that WILL and ARE resulting as a result of inaction. I just don't see that the current uncertainties lead to that conclusion.

Thanks for your thoughts.

Mike

 

[David Wolfson]

Good discussion! I'm a transportation modeler, which means that I've been involved in the details of "mode choice" models. These models attempt to forecast what transport modes travelers use when they travel, i.e. single-occupant auto, multi-occupant auto, bus, bicycle, light rail, park-and-ride, etc., etc. We do this to evaluate the "cost/benefit" of expensive capital projects and prioritize them. And then there're air quality conformity models....

To make matters short, there's not much someone can tell me about the baffling difficulty of projective forecasting. My models parameterize matters of social taste and preference that can be wildly divergent and whose characteristics are opaque to physical law. O.k., there're my creds, FWTW.

Point 1
Model calibration is necessary and unavoidable whenever there's any significant non-deterministic component in the model. Some components need it a lot, while others don't need much or any. Regardless, only in the case of very simple models does the responsible modeler calibrate the components to replicate the objective, i.e. the answer. Rather components are separately calibrated to better replicate measured and (hopefully) well-understood parametric quantities.

Then and only then do you look at the model result vs. Truth. If you get an answer that's sufficiently close to the Truth, you heave a sigh and sneak out of town before an anomaly surfaces. When the inevitable anomaly does surface, you evaluate the model structure theoretically to figure out if there're any components you want to try changing -- remove, add, or re-formulate. Then you calibrate the components again, cross your calloused fingers, and re-validate ad nauseum.

What's done is not "fudging".

Point 2
Basic laws of statistics, especially as derived from the Laws of Large Numbers (e.g. the Central Limit Theorem), are pretty darn well understood in theory and application by those who design models.

It's entirely unsurprising that the long-term behavior of a variant phenomenon is better understood, statistically speaking, than its shorter-term behavior.

Point 3
I think they do indeed have theories regarding variations in Earth-incident solar output and models for forecasting same. Obvious ones are the Milenkovich Cycle, the assymetry of the Earth's orbit itself, and the noted solar magnetic cycle. I'm sure there are some astrophysical ones as well. Might perturbations come up they haven't accounted for? See Point 1 above....

Point 4
No doubt at least somewhat true -- and thank goodness! That's what keeps creative scientists' paychecks coming in and value-added expert contractors such as yourself or myself(?) in business. If it weren't the case then weather forecasts would be built into a Casio watch and all the meteorologists would be scraping by selling chase videos, part-timing at Wal-Mart, or both!

... As Scott and Chris have already said while I've been hacking my way through this post....

 

[Mike Smith]

Chris,

Great post, thanks! I believe that anyone who happens to read this thread will see several well-thought-out perspectives on climate change. There is only one final thought I wish to make:

But I feel these parameterized modeling studies have greatly enhanced our knowledge of how supercells work. Likewise, I deduce global climate models, even as parameterized, are increasing our knowledge of the climate system.

I agree with the above. Certainly, mesoscale models and climate models have the potential to increase our knowledge in these areas. That said, would we want to rely on mesoscale models as our sole forecast tool for storm chasing? I think not. I believe the current generation of climate models is quite useful for learning about climate but I am not confident in their ability to predict the state-of-the-climate in 70 years.

Again, great series of posts...
Mike

 

[Mike Smith]

David,

Another great post!

I agree about the law of large numbers up to a point. What I am attempting to point out is a priori versus a posteri "forecasting." The fact is that ACTUAL world temperatures through the end of 2005 are well below the coldest scenario forecast by the models cited in the 1998 IPCC document. When verified (comparing model forecasts versus actual conditions), none of the models did very well. This is why I believe verification of the models' forecasts is important to establish their validity.

When trying to forecast the future by matching to the past, take, for example, the "correlation" between the winner of the Super Bowl and the performance of the Dow Jones Industrial average the following year. If you are not familiar with it, see here: www.bargaineering.com/articles/super-bowl-stock-market-indicator.html

Of course, if one went back to 1970's Green Bay v. Kansas City game, no one would have predicted that you could accurately forecast the stock market based on the game's winner. Yet, by examining past performance the correlation appears to be there.

Here is an example of an academic stock market prediction based on a model correlated to the past: www.eurekalert.org/pub_releases/2002-12/uoc--smc121402.php and note the explanation of the "physics" (implying certainty) in the press release. As you will see, he forecasts a stock market "crash" in 2003. While the market had a down year, there was no crash and unlike his forecast, the market ended 2004 with a small net gain over 2002. (A frequent joke on Wall Street is that these models have forecast "eleven of the last five recessions.")

Time after time stock market models based on past market conditions fail when used to forecast the future. Forecasting the Dow Jones Industrials, in theory, is far less complex than the maps produced forecasting climate 2070-2100. The Dow is a single parameter that goes up or down. The maps represented temperature and precipitation and the geographic distribution of each.

I am certain you refine your predictive transportation model based on data collected AFTER the model runs, a form of verification. If the model is wrong by a large margin the first week, what is the chance it gets Thanksgiving holiday traffic right in 2008? I know nothing about transportation models but my instinct says the odds are not very high of an accurate forecast for the distant future if the near term forecast fails by a large margin (as the 1998 forecasts published by the IPCC did) if you get my analogy.

Again, great information and great additional perspective. I am learning a lot from this thread.

Mike

 

[David Wolfson]

Actually, Mike, historically the travel models' objectives have been to forecast some flavor of representative "average" aggregate behavior -- not micro-scale temporal behavior at some future horizon. TTI the social wheel is turning and the answers decision-makers are interested in are less and less what the historical aggregate travel models are designed to forecast and more in line with operational considerations such as your example.

Micro-scale operational traffic modeling is a coming thing, done quite differently -- structured around physical phenomena more like operational met models. Interesting stuff.

AFA correlation (auto- and otherwise)... well you live with it. It just means your model component isn't as significant to the answer as you'd like to think it is. There're some real beauts in travel models. For example -- one of my favorites: many travel model formulations use trips = f(auto ownership, household income, household size). You think perhaps that number of vehicles owned, household income, and household size (i.e. number of income-earners) might be correlated....?

 

[Mike Smith]

Please go to: http://climatesci.atmos.colostate.edu/ and look at A New Paper That Evaluates the Accuracy Of The IPCC Models to Assess Regional Weather Patterns

Two important quotes: “The dynamic accuracy of GCM models have not been adequately tested. Such models need to be used to predict short-term weather changes since skill at such scales is essential if the models are to demonstrate a numerical fidelity in simulating wave-wave- interactions. If the GCMs have insufficient spatial resolution or physics to forecast weather as accurately as current operational weather numerical prediction models, what confidence should be placed on their skill at predicting long-term climate change? Long-term climate is made up of [an accumulation] of day-to-day weather.â€￾

...and...

Clearly, the multi-decadal climate models need to be tested on their ability to replicate the statistics of regional and planetary atmospheric circulation features. If they cannot perform well on these climate metrics, then any information that is communicated to policymakers on regional climate change in response to human- and natural-climate forcings and feedbacks is very misleading.

 

[David Wolfson]

Results of the Oxford distributed computing Climate Change Experiment project are scheduled for release: http://today.reuters.com/news/articlenews.aspx?type=domesticNews&storyID=2007-01-19T190953Z_01_L19437366_RTRUKOC_0_US-CLIMATE-WARMING-BRITAIN.xml&pageNumber=1&imageid=∩=&sz=13&WTModLoc =NewsArt-C1-ArticlePage1

The main site is: http://www.bbc.co.uk/sn/climateexperiment/

The methodology of this study is pretty strong and involves the generation of large numbers of alternative models statistically tested with back-casting. Holy cow, Batman! The Amazon rainforest becomes a desert -- soon?! Tell me that isn't scary.