Can’t they just make up their mind….
Scientist predicts ‘mini Ice Age’
ST. PETERSBURG, Russia, Feb. 7 (UPI) — A Russian astronomer has predicted that Earth will experience a “mini Ice Age” in the middle of this century, caused by low solar activity.
Khabibullo Abdusamatov of the Pulkovo Astronomic Observatory in St. Petersburg said Monday that temperatures will begin falling six or seven years from now, when global warming caused by increased solar activity in the 20th century reaches its peak, RIA Novosti reported.
The coldest period will occur 15 to 20 years after a major solar output decline between 2035 and 2045, Abdusamatov said.
Dramatic changes in the earth’s surface temperatures are an ordinary phenomenon, not an anomaly, he said, and result from variations in the sun’s energy output and ultraviolet radiation.
The Northern Hemisphere’s most recent cool-down period occurred between 1645 and 1705. The resulting period, known as the Little Ice Age, left canals in the Netherlands frozen solid and forced people in Greenland to abandon their houses to glaciers, the scientist said.Yeah whatever.
At least this one has more science behind it than the global warming hoax. That the output from the sun is cyclical has been known for hundreds of years. Any HAM radio operator can tell you that, but the global warming hustlers never want to talk about it.
I just hope I can burn enough fossil fuels before I die to counteract this ice age stuff.
Hi ed-
I think we’re starting to debate around the point as to humidity- in fact, I think I’ve forgotten what the original point is! I think you’re exactly right in your observation that humidity is local. But it can have global consequences as we see with the hydrologic cycle.
If you think that the climate is too complex to model, what to you think of the models which have correctly predicted climate variations?
Denny Crane,
I don’t dispute how much ice is in Juneau or how much is in the artic ice cap. I was just using the figures others were providing to raise a question about those figures. Now you say:
And it was stated before this is all within the city limits of Juneau, so we can infer from that information that Juneau is that large. Someone else can Google it, I just raise the question.
Also nuclear submarines are designed to break through the Artic Ice cap, obviously the part of the cap that’s floating. I don’t know exactly what thickness they can break through, but it’s less then 50 feet and more like 15 feet.
echibby,
The best information is what I linked to before, which is a summary of research papers, some of which looked at 20 state-of-the-art GCMs to see how well they predicted one of the largest of earth’s atmospheric phenomena – the tropical Indian monsoon. The results are that they don’t work. The concern is that these GCMs are the major source of evidence behind the Kyoto treaty. Bush is wise to hold off supporting Kyoto until there’s more data.
Climate Models (Inadequacies: Precipitation) – Summary
Mac Lorry-
Thanks for the link. It’s interesting that the authors of the piece point out that a set of self-described “20 state-of-the-art GCMs”failed to predict weather trends. GCM stands for Global Climate Model- these aren’t used to predict local weather variations, only global climate trends. So are these original models actually GCMs? The article is unclear. The authors of the article seem to equate inability to predict/model Indian monsoon trends with inability to predict/model global climate trends. Conveniently, they ignore models that have correctly predicted global climate trends, such as Hansen’s 1992 study I cited earlier.
I also stumbled upon another site: a blog written by climate scientists. Here’s what they have to say about climate models:
http://www.realclimate.org/index.php?p=240#ClimateModelling
Thanks for the dialog- I hadn’t looked into global modeling much before this.
echibby,
Thanks for the link. There’s lots of information on that site, but I note that the date at the top is 1 Dec 2004. This is a fast moving subject and like many people, I expect on-line sources to be more up to date. I’m not saying the content is invalid, just that I’m disappointed it’s been static for more than a year.
Computer models work by starting with a set of parameters that describe the atmosphere as it is now or was at some point. Then predict what those parameters will be at some set period into the future by applying the current understanding of how the atmosphere works. Those predicted parameters become the input for the next cycle. The faster the computer the shorter the time period can be and the greater the number of parameters that can be used. Some of the most powerful computers in the world are used to predict weather in this way.
From your own experience you know that the prediction for tomorrow is more accurate then the prediction for two days into the future, which is more accurate then the predictions for next week. Flaws in the model accumulate so that each cycle is less accurate than the one before it. Obviously, there can be some random occurrences where the prediction for tomorrow is wrong while the prediction for next Friday is correct. To obfuscate matters further predictions are given in percentage chance, such that there’s a 50% chance of rain next Friday. Then if it rains or not the predication is only 50% wrong.
The GISS global-climate model’s success in estimating Pinatubo’s climate impact shows that science has a reasonable understanding of how massive quantities of sulfur dioxide in the upper atmosphere alters the climate. The forcing in such a case is so great as to overwhelm normal processes.
Last summer computer models were used to predict the track and strength of hurricanes in the Gulf of Mexico. Some models were correct on what would happen in the next 24 hours, but then were wrong in following 24 hour predictions. You can’t judge a model’s accuracy on it’s success or failure of any given prediction. Similarly the success of estimating Pinatubo’s climate impact doesn’t translate to meaning that computer models in general are accurate.
The question the studies about the many GCM’s was trying to answer is how accurate are their predictions. One means is to measure their predictions of major atmospheric phenomena. These models work the same way the weather models work, by projecting what the atmosphere will be a short time into the future, then using those projected parameters as input for the next cycle. If a model can’t predict one of the largest cyclical atmospheric phenomenas on Earth, then we know the inputs to the next cycle are wrong. Being that errors accumulate with each cycle, what value are such models in predicting theclimate decades into the future?
There’s a real simple principle at work here. Science can’t model something it doesn’t understand. Alternately, if the model’s predictions are wrong, is shows that science doesn’t understand what they are modeling. The answer is more research and faster computers.
Mac Lorry-
It seems we’re in agreement on the need for more research and faster computers!
Re: the realclimate website- go back and check the actual entries. The index was created in Dec 2004, and the entries in the index (under climate modelling, at least) range across 2005, the most recent being Nov 30. The user comments are even more recent. If you check out the home page of realclimate, the most recent entry is Feb 9, 2006. So the site itself is relatively up to date- the authors presumably don’t update every subpage on a frequent basis.
Regarding climate modelling- there is a difference between climate and weather. As far as I can tell, no one generating GCMs will use them to predict whether or not there will be rain next Friday. Climate models cannot predict what a hurricane will do- this is a discrete, small event in the global climate cycle. I agree with you- they are next to useless to predict what the monsoon will be like this year, next year, or further into the future. They’re not designed to do this. But climate models can be used to make a hypothesis about the long term climate (not weather!).
It’s like comparing apples and oranges- modelling a hurricane, or a monsoon, which is a fast dynamic event on the global scale, is much different than modelling long term, large geographic trends. I found this site that actually lets you analyze the data from a number of climate models:
http://www-pcmdi.llnl.gov/ipcc/about_ipcc.php
Thanks,
echibby
Mac Lorry-
I just remembered this. Last September (coincidentaly right after Katrina- the paper was accepted long before), a study was published in Science:
Science Webster et al. 309 (5742): 1844
The abstract is:
“We examined the number of tropical cyclones and cyclone days as well as tropical cyclone intensity over the past 35 years, in an environment of increasing sea surface temperature. A large increase was seen in the number and proportion of hurricanes reaching categories 4 and 5. The largest increase occurred in the North Pacific, Indian, and Southwest Pacific Oceans, and the smallest percentage increase occurred in the North Atlantic Ocean. These increases have taken place while the number of cyclones and cyclone days has decreased in all basins except the North Atlantic during the past decade.”
What is/was controversial about the paper was that a number of leading climatologists didn’t accept the conclusions, saying the data did not support the thesis of the paper. Essentially, a lot of counterarguments focused on the fact that the science was not there yet that enabled a projection of microevents (hurricane intensities and numbers) from a macroeffect (global sea surface temperatures).
I bring this up to illustrate that climatologists, who generally agree on global warming having an anthropogenic cause, still have violent disagreements on projecting microevents such as hurricanes.
echibby,
I’m not an expert on weather and climate computer modeling, but I do have some experience with computer modeling for other types of non-linear problems where there’s no pure mathematical means of arriving at a solution. These models all have some basic principles in common as I wrote about in my prior post on this topic.
Computer models take a set of initial conditions, project what the values of these conditions will be a short time into the future, then use these projected values as the input to the next iteration. If you’re into calculus you may recognize this process as being a form of numerical integration. These fundamentals are common to all the weather and climate computer models regardless of their specific purpose. The difference is in the details of what parameters are used to define the atmosphere, how many data points are used, how the projections are made, and the time period being covered in each iteration. The problem that plagues all such models is that any errors in the projected values accumulate with each integration the model goes through. It’s the classic garbage in garbage out problem.
The weather, and thus, the climate is a chaotic system. From Chaos theory we know that small influences accumulate over time to produce large effects. This is sometimes refereed to as the butterfly effect, which says that a butterfly flapping it’s wings in the tropics will result in a storm in some other location in the future that wouldn’t have occurred if the butterfly never existed. For that reason, no GCM with so course a data set that it ignores one of earth’s largest atmospheric phenomena would have any scientific credibility.
While not specifically designed to predict the Indian monsoon, the fact is that the 20 GCMs in the study do predict the Indian monsoon, just not accurately. The degree to which the Indian monsoon effects world climate is the degree to which all the GCMs in the study introduce errors into subsequent iterations. As these errors accumulate, the predictions of the GCMs become increasingly inaccurate.
I’m not disputing the earth is in a warming period, just the cause of it. Growing volumes of data show the earth has expensed two complete warming and cooling cycles since Roman times, all before human produced CO2 could have had any influence on the climate. While still in the minority, several scientists have pointed to the cyclical changes in the Sun’s energy output as the primary cause of these warming and cooling cycles, including the Russian scientist who’s predicting a cooling in mid-century.
The site I linked to before has been accumulating data supporting Sun cycles as the driving force behind climate change and also data showing the world will be better off with higher atmospheric CO2 concentrations during the warm period than it would be with lower concentrations.
The conclusion is that warming will continue regardless of what humankind does, apart from nuclear war, and that we should be moving people and infrastructure out of coastal low-lands. It would also be smart to harden the infrastructure that can’t be moved to withstand hurricanes, regardless.
Hi Mac Lorry-
The question you’re trying to answer is not what effect monsoons or hurricanes have on global climate, but what effect these events have on global weather. GCMs make projections of the former, not the latter. If we were speaking of Global Weather Models (not sure if they exist or not), that would be a different matter. I believe what concerns you is that a GCM that doesn’t include weather events will be inaccurate.
As a matter of terminology, climatologists use models to make projections, not predictions. Climate projections deal with averages- what will be the average temperature in five years? Ten years? And beyond… And how are these averages influenced by macro trends, such as water vapor, cloud cover, humidity, solar variance, etc? You correctly point out that sun cycles affect the global climate. Solar variation is already accounted for in current GCMs.
As terrifying as a hurricane or monsoon is, these events do little to influence macro trends. These are weather events. An event like Pinatubo, which perturbs the atmospheric dynamics to a large degree, does influence the macroscale. This is a climate event. A very concise summary of the differences between climate and weather can be found here:
http://atoc.colorado.edu/~fasullo/pjw_class/weathervsclimate.html
We both agree that the earth’s average temperature is increasing. We disagree on the causes. There is a wealth of data extending back tens of thousands of years) wich correlates temperature and atmospheric composition. Accurate climate models (they do exist!) can be tested against historical data, and they’re pretty good at this. Obviously, these models aren’t used at all in examining weather events over the previous millennia (such as hurricanes, blizzards, etc).
I understand your reluctance to buy into a climate model which ignores local events, but these have very little effect on the global climate. Of more concern is getting accurate historical data on CO2 composition, temperature, etc. Researchers have gotten very good at this, which allows them to test the robustness of GCMs. I think we can agree that a) global temperatures are rising and b) CO2 levels are rising due to the actions of humanity. Where we disagree is if the two are correlated at all. Climatologists overall think that they are. This is due to the models, which have shown themselves robust at projecting the climate (not weather).
echibby,
I understand what you are saying about weather not being the same as climate, but trends in weather define the climate, not the other way around. Project the trends and you project the climate. GCMs project shifting patterns in perception, not just temperature, as the two are tightly coupled. In order to project trends in precipitation you have to model trends in weather patterns, as it’s the weather that produces the precipitation.
The GCMs in the study all make projections about the precipitation produced by the Indian monsoon or else they couldn’t have been in the study. Given the cyclical feedback nature of all these models, any inaccuracy in any part of the projected data set accumulates, and because of the interconnectedness of the atmosphere, any error spreads through the entire data set. The further into the future a projection is made the less accurate it becomes.
It’s true that the GCMs have been tweaked so that using past data they can predict the current climate. Unfortunatly, this has not been accomplished through a fundamental understanding of how the climate works, but through the introduction of numerious correction parameters. The same technique of using past data to develop a model until it can accurately predict the present has been used in attempts to predict the stock market. Even though the rewards are obvious, no such model is accurate enough to beat the average rise of the market over time. The conclusion is that the technique of using past data to tune a computer model is of little use. The only valid test is to measure how well a model predicts intermediary conditions such as the precipitation produced by the Indian monsoon.
The role of the solar cycles is just starting to be understood. Here’s a small excerpt from a piece published on the CO2 Science website as a result of reviewing three scientific studies linking the sun to historical climate changes.
Solar-Powered Millennial-Scale Climatic Change
I’m not sure how much of this you want to read, but you have been kind enough to provide links, so I’ll do the same.
More Evidence of a Solar-Climate Link
A Pair of Two-Millennia-Long Climatic Records
Here’s a quote from Climate Model Inadequacies (Radiation) – Summary
I hope I got the links to the right pages.
Theres growing evidence that the current crop of GCMs are fatally flawed and it’s the Sun that caused the historical climate oscillations and is doing so again. Other studies demonstrate that C02 concentrations lag warming and cooling cycles, and thus, are cause by climate change rather than cause climate change.
Hi Mac Lorry-
Thanks for the links. I actually don’t see any thing to dispute about the analysis of CO2 Science- the papers they summarize discuss the historical effect of solar variation on temperatures. All the models already take this into account. CO2 Science then makes the claim that this alone can account for current warming trends- I don’t think the original papers claimed this, but it’s difficult to know without going back and reading the primary literature.
I’d like to call a halt to our exchange- I don’t think we’re going to sway each other. You’ve given me a lot to think about and digest- thanks! I’m definitely going to go back to the primary literature and read up on this. It’s very easy to selectively cite in order to prove a point either way. You’re obviously a thoughtful person and I hope you can do the same.
Thanks,
echibby