Another blow to the Gorons! Much of the AGW climate scare is based on computer models. Many people foolishly believe computer models are predictions or forecasts while in truth computer models only generate *what if* scenarios.

Climate computer models are extremely complex and *what if* permeates the models. The relationships between various factors are not well known and any algorithm that tries to express those relationships will necessarily be approximate.

My profession is writing software. Software is verified by testing. Whenever a change is made to a computer program, a full test regimen must also be done to ensure the program works as intended.

Climate models can't be tested like normal software. The only way to validate these models is to match their output with real-world observations. A long-term climate model must be tested against decades of data. Remember, each time a program is changed, the testing must restart. This makes long-term climate models practically unverifiable.

I have a question about climate models: How *open* are the models? Are the algorithms shared? Can independent researchers view the inner details of the models? More important, can independent researchers view the source code?

Reply: I do not honestly know. Anyone else?

The research clearly shows that you can actually slow down sea-level rise when you increase temperatures over Antarctica because snowfall increases, but warmer temperatures also have the potential to speed up sea-level rise due to enhanced melting along the edges of Antarctica.

Over the next century, whether the ice sheet grows from increased snowfall or shrinks due to more melt will depend on how much temperatures increase in Antarctica, and potentially on erosion at the ice sheet edge by the warmer ocean and rising sea level.

The current generation of climate models has improved over previous generations, but still leaves Antarctic surface temperature projections for the 21st century with a high degree of uncertainty.

Brett,

Your headline would have been equally valid had you left out the word "Antarctic."

Reply: Maybe.

Probably it's just me not paying attention , but has anybody - say in the past year or so - noticed any big time reports about the models having UNDERESTIMATED the actual observed warming ? After all , something like that is not the type of news that would generally go unnoticed or hidden , much less neglected by the AGW alarmists . I would expect to be hearing the Town Criers bellowing it out from the highest bell towers in the land . "See , " they would be saying . " we told you so ! This guy here made these model predictions , and now we're finding out it's actually Worse than the models predicted ! " Yet, as far as I'm aware , we have very little if anything of this nature. Is there any of you math wizards out there that could calculate the odds of this occuring naturally ? I personally would think that [To paraphrase a line from Starship Troopers ] - Statistically speaking , this should have happened accidentally by now . So I guess the question I have is , should the AGW Alarmist community be suprised if the general fence sitting Public sounds like the sceptics when they look at this and say " Geeez , now here's a big suprise , the models over estimated warming again ." But , then again , It's probably just me not paying attention !!

Brett,
I could be wrong, but I thought that the GCMs predicted that the Antartic would cool as the rest of the globe warms, and that this cooling is perfectly in line with over-all AGW.

Too many conflicting predictions and scenarios and explainations. Reply: No doubt.

"The study by researchers at Scripps Institution of Oceanography at UC San Diego provides strong evidence that a glacial ice cap, about half the size of the modern day glacial ice sheet, existed 91 million years ago during a period of intense global warming."

The above quote comes from a Jan 2008 story that I have attached below:
http://www.eurekalert.org/pub_releases/2008-01/uoc--awc011008.php

The author of the researtch report also states:
"Until now it was generally accepted that there were no large glaciers on the poles prior to the development of the Antarctic ice sheet about 33 million years ago," said Richard Norris, professor of paleobiology at Scripps Oceanography and co-author of the study. "This study demonstrates that even the super-warm climates of the Cretaceous Thermal Maximum were not warm enough to prevent ice growth."

Obviously, the climate model algorithms need to be corrected based on this new research finding. The a 2-inch sea level reduction only changes the potential sea level rise to the range of 5-21 inches.

What I find even more interesting is the following quote:
"The results from the study are consistent with other studies from Russia and New Jersey that show sea level fell by about 25-40 m (82-131 ft) at the same time that the ice sheets were growing during the Cretaceous period. Sea level is known to fall as water is removed from the oceans to build continental ice sheets; conversely, sea level rises as ice melts and returns to the sea."

This more significant potential decrease in sea levels would have significant implications and needs to be investigated further. Obviously the ice conditions of Antarctica do not follow simple correlations.

My Lord. Forecasters looking out the window!

What ever will they think of next.?

An interesting question: When does "global" mean "local"?

According to NASA, Antarctica is in a strong cooling trend and a strong warm trend - at the same time.

http://earthobservatory.nasa.gov/Newsroom/NewImages/Images/antarctic_temps.AVH1982-2004.jpg
http://earthobservatory.nasa.gov/Study/WilkinsIceSheet/images/wilkins_avh_2007.jpg

The solution is to give Antarctica anti-psychotic medication.

You headline would have been unusual if it would have read, "Models Get Prediction Right".

Models are wonderful toys for which we pay, through taxes & grants, millions of dollars. Scientists spend even more millions on computer time to play around with no measurable result.

Even if they cannot make reasonable predictions with the models, they will keep trying. They have to justify their expenditures.

How does ice melt at 60 below?

Quick IQ test for the alarmists -

The south pole receives no sunlight from March 20 to September 20. How much of the incoming UV is trapped by greenhouse gases?

1. Huge amounts which will cause Antarctica to melt and raise sea level 200 feet.

2. Enough to make the Penguins migrate to the Arctic - as reported by NBC.

3. Close to zero, because there isn't much incoming UV.

4. Enough to make the Penguins drop their eggs and produce tap-dancing chicks.

JEP,

They test the computer models by running them to predict past weather events and see how the results correlate. If it predicts past weather events accurately, then it is assumed that what the model predicts for the future is fairly accurate. Here's a Wiki entry that explains GCM's more accurately:

http://en.wikipedia.org/wiki/Global_climate_model

The computer models are not perfect, but they give us a starting point. If we didn't have computer models, we wouldn't be able to enjoy the many modern conveniences that we have today, including modern medicines. If we don't give scientists the time and money needed to perfect the computer models, then they will never get any better at predicting the weather.

Compare 1970's desktop computer technology with that of today and you can get my point.

Here's another link about climate models:

http://www.pnas.org/cgi/content/full/98/19/10529

An excerpt: Major progress in our understanding of climate processes in the past, present, and future has been made by the development of numerical models that simulate climate at an increasing level of detail. Recent breakthroughs in spatial coverage and temporal resolutions of systems recording today's climate and high-resolution reconstructions of past climate conditions from diverse archives using new proxies make it possible to validate climate models and thus improve their reliability for future predictions.

Wait a minute. WAIT A MINUTE! Do you people only read intop a reoprt what you want to? And then extrapolate whatever you want from the misreading of the information. Read this quote again from the article.

"The computer models are accurately simulating climate for the other six continents, according to the study."

That means (to me) that over the rest of the planet they are getting it right! And all you can focus on is the Antarctic? And the reason they got that wrong is because the ozone hole is larger? Brett it seems to me that the headline could just as well have been:

SCIENTISTS GET IT RIGHT OVER MAJORITY OF PLANET, BUT MISS THE ANTARCTIC DUE TO OZONE HOLE!

Reply: Too many words for a title.

jp,

No, none of the GCMs predict Antarctic cooling. They all predict warming, and lots of it in both poles due to positive feedbacks provided by additional water vapor in the atmosphere as a result of CO2 warming.

The only portion of Antarctica that is warming is the peninsula that holds McMerdo, and that sits outside the Antarctic circle, additionally, there is a strong suspicion that that warming is due to vulcanism along the coast.

Svensmark's GCR theory predicts Antarctic warming due to additional cloud cover (clouds have a higher albedo than ice) but that requires more water vapor than appears to be available in the area (Antarctica is the worlds largest desert) getting less than 2" of precipitation/yr).

jep is algorithm a contraction for
al gore rhythym

orientation of suns rays on earths surface changed by toplike wobble of earth now has mor direct rays in nh vs sh. polar regions more affected by this this balance of planet-this is just one of many natural factors keeping south pole colderrrrrrr

jep, Kansas USA:
How open are the models?
This issue has been discussed a length on the Climate Audit site. Steve has obtained some source code for some models over the years but the GISS code in perticular has been very difficult to get.
Perhaps a Query at Cliamteaudit.org would get you some up to date status.

bill-tb,

How does ice melt at 60 below?

It melts at that temperature due to the special physics that can only be attributed to anthropogenically-produced CO2.

Sorry, I should have added (/sarc off) at the bottom of the last post.

Hi kevinag,

If the modelers say they are doing a good job, it must be true.

If the facts don't fit the theory, change the facts
Albert Einstein

Gary:

I believe you missed my points.

1. To my knowledge, no computer models have done well in "back predicting" climate (sorry, I don't recall the correct term).

2. Without running the model into the future, it's impossible to know if the model really "back predicted" the past climate or if it was just fudged until it seemed to predict the past. (I'm not saying the researchers fudged their results, but fudged the algorithms until they received the correct results. )

3. Computer models are very valuable in engineering and medicine and have led to many wonderful discoveries. But the real world is the test for those models. A new, effective medicine derived from or through computer modeling is the proof of the model. Climate models may be useful in understanding the climate but only so much as they accurately reflect the real world.

I have a question about climate models: How *open* are the models? Are the algorithms shared? Can independent researchers view the inner details of the models? More important, can independent researchers view the source code?

It took me about fifteen seconds to find the fms: the flexible modeling system (published by NOAA).

From the what is fms link on that site:

FMS is a software framework for supporting the efficient development, construction, execution, and scientific interpretation of atmospheric, oceanic, and climate system models. FMS comprises the following:

1. A software infrastructure for constructing and running atmospheric, oceanic, and climate system models. This infrastructure includes software to handle parallelization, input and output, data exchange between various model grids, orchestration of the time stepping, makefiles, and simple sample run scripts. This infrastructure should largely insulate FMS users from machine-specific details.
2. A standardization of the interfaces between various component models.
3. Software for standardizing, coordinating, and improving diagnostic calculations of FMS-based models, and input data preparation for such models. Common preprocessing and post-processing software are included to the extent that the needed functionality cannot be adequately provided by available third-party software.
4. Contributed component models that are subjected to a rigorous software quality review and improvement process. The development and initial testing of these component models is largely a scientific question, and would not fall under FMS. The quality review and improvement process includes consideration of (A) compliance with FMS interface and documentation standards to ensure portability and inter-operability, (B) understandability (clarity and consistency of documentation, comments, interfaces, and code), and (C) general computational efficiency without algorithmic changes.
5. A standardized technique for version control and dissemination of the software and documentation.

FMS does not include the determination of model configurations, parameter settings, or the choice amongst various options. These decisions require scientific research. Similarly, the development of new component models is a scientific concern that is outside of the direct purview of FMS. Nonetheless, infrastructural changes to enable such developments are within the scope of FMS. The collaborative software review process of contributed models is therefore an essential facet of FMS.

It appears to me that the models themselves and their source code is readily available to anyone who chooses to use them. I am similarly under the impression that a variety of web-based sites are available that host the various models, some free and some not, for anyone who cares to use them.

Of course, it's far more satisfying to throw around questions and innuendo than to actually engage in the work yourselves.

My question for "jep, Kansas USA" is: how hard have looked for yourself?

- Has your ability download a particular model been blocked? I don't mean "have you read somebody on a blog somewhere who read somewhere..." -- I mean have you tried it yourself and been blocked?

- Have you tried to discern the algorithms, and been blocked?

- What "inner details" have you sought and been denied?

- What source code have you attempted to download, and been blocked from receiving?

Just got through reading THE CHILLING STARS by Svensmark and Calder. Their basic theory is that clouds caused by cosmic rays, and the changes in the shielding magnetic field of the sun rather than its luminosity, are a primary driver. Perhaps THE primary driver of planetary temperature. What is interesting is that the tendency of Antarctica to go in opposite directions to the rest of the world falls out naturally from the theory.

So, I guess the climate models get a B+ (6 out of 7 continents plus the Antarctic peninsula). Not perfect but hard to characterize as alarmist paranoia.

funny,

nowhere does it say. "the computer models were wrong" That admission might lead us to beleive that if the models could possibly be WRONG. AGW might not be happening because the models are used to prove AGW. Garbage in Garbage out

meanwhile, the sun is blank

http://www.spaceweather.com/

"No, none of the GCMs predict Antarctic cooling. They all predict warming, and lots of it in both poles due to positive feedbacks provided by additional water vapor in the atmosphere as a result of CO2 warming."

You better tell that to Spencer Wearts,Kirk Bryan, and Syukuro Manabe, as well as the folks at Real Climate:

http://tinyurl.com/48z936