What formal assurance methods have been applied to climate models?

Question

What formal assurance methods have been applied to climate models?

Answer 1

Refering to the criminally hacked emails is so 2009. If there is anything scientific to learn from the emails, Any way, what do the emails say about



1. CO2 is a greenhouse gas? http://www.epa.gov/climatechange/ghgemissions/gases.html

2. The laws of thermodynamics? http://physics.about.com/od/thermodynamics/a/lawthermo.htm

3. Atmospheric CO2 levels are increasing? http://co2now.org/

http://legacy.signonsandiego.com/uniontrib/20080327/images/co2.html

4. This CO2 is due to the combustion of fossil fuels? http://www.sciencedaily.com/releases/2009/11/091117133504.htm

5. The Earth's temperature is increasing? http://data.giss.nasa.gov/gistemp/graphs/

6. Natural factors which influence climate would be cooling the Earth if not for anthropogenic CO2? http://www.skepticalscience.com/global-warming-natural-cycle.htm







I hate to break it to you, but comparing the model output to historic data is the answer to your question.

Answer 2

This statement at the top here is a hardcore denier croc of BS, and invalidates your anyway bogus "question:"



"Global climate circulation models are our look into the future of global climate. They are the predictions for all sorts of climate metrics like temperature, precipitation, sea levels, etc. which are major factors on which we are proposing/basing climate policy."



1. Other than token and trial measures here and there, there is no climate policy to speak of, so "proposing/basing" is really "proposing."

2. There is no "we" that makes sense in this statement of yours. When it comes to possible public policies to reduce the future negative effects of AGW, a few people are proposing, a few are lying about science, and most are busy with other things. No brownie points to anyone chiming in to point out which of the "we"s you are part of.

3. Substantive proposals to combat human-caused climate change that have been taken under active and serious consideration by major governments in recent years have been based on the OVERALL scientific consensus (IPCC, etc.) and various mechanisms for reducing global greenhouse gas emissions (recommended by any sensible reading of that consensus). The computer models that simulate future scenarios are but one part of the overall science. The fact that carbon emissions from human sources are significantly impacting the global carbon cycle for centuries to come is a mainly matter of scientific understanding and observation, and only secondarily a function of computer plugging and chugging. How human-caused changes to global carbon disrupt ecosystems and how those disruptions have mostly negative effects on human economies is also understood based on scientific principles and observations (and economic statistics) much more than on computer simulations. Finally, when it comes to economic solutions to scientifically-comprehended challenges, e.g. possible measures to reduce the use of carbon fuels, the likely effects of economic incentives are largely a matter of empirical experience and common sense (when something is more expensive you use less of it, and switch to alternatives, etc.), not software codes.

Answer 3

Al second sentences is very good and I wish more alarmists would use that type of reasoning. His first sentence makes an assumption that they are out of context and I don't think he knows that. Clearly Harry is frustrated with Australian data and whether or not it is truely justified, I think most of us know why he is probably frustrated. When you get caught, it always easy to say you were taken out of context.



I looked at Pegminers link to see what it might show me. After selecting a couple links there is a study (of course backed by MODELS) which tries to blame the Little Ice Age on 4 volcanic eruptions yada yada yada. Models are useful and are getting better all the time but they still have their limitations and you can "prove" just about anything you want to with them. If you want to tweek them to demonstrate how a volcanic eruption can cause additional cooling, that shouldn't be too difficult. I don't think Pegminer would disagree with that.

Answer 4

This is probably the least unreasonable question I've seen you ask, although I can't see a point in asking it here. The codes are often made available to other scientists for testing and use. Perhaps not always, but I have seen discussion in the past from climate gate (probably from one of your more tedious questions) relating to not wanting to give the code to some denialists because they weren't scientists. So I doubt you would be given code if you requested it, but you could certainly ask them this question. I'm not sure why you would ask us, Trevor is the only climate scientist who is active here atm. You know where to find Dana, you could ask him. Or you could email James Hanson or Roy Spencer and ask them.

Answer 5

l very much doubt it has been done and will be done... to many people are making money on the climate change band wagon... What makes me laugh the most is celebrities raving on about climate change and how we should do this and do that... there carbon foot print in 1 year, would be more then mine in a life time... bloody hypocrites...

Answer 6

I really don't give a crap what models say... I live in Jeddah Saudi Arabia on coast of red sea Somewhere around Latitude 21.35' Longitude 39.10' This isn't my first year here I have lived here for past 30 years now. Its Summer, August, start of August temperatures are suppose to be around 45 max and 30 min. (thats in Deg C) I'm trying to lose 3 Kgs so I go for a walk every night, this isn't something new for me I would go for a walk every night even if its 35 Deg C and 80% humidity. The most amazing thing this winter is we have temperatures of 40 Deg C Max 26 Deg C Min And humidity of max 65% during night and 45% during day. This might be quite hot for few but I remember the days during every July, August and start to mid of September that it was impossible to breath due to humidity, just walking from car to home or office was a killer. I'm not really missing that but something else is missing, which is global warming. Frankly I would say enough of lies, even if you show high temperatures around the world just because you control the media and those scientists who are payed by you, you still can't tell me that every place in the world is showing signs of cooling other than you satellites? This is the biggest lie I've ever come across in my life. I actually have an excel sheet with me that I can use to predict temperatures around the world by tweaking simple parameters, I really can use it to show global cooling, heating or whatever you want. Computer models remain computer models, they can never replicate the real thing. I was reading interview of one of the designers for F22 Raptor, he said something weird that till today they use wind tunnel to do the final testing of airflow on the air craft, so someone asked him why don't you use computers for that? He said there aren't enough powerful computers available today to replicate the wind effect. This is Lockheed Martin/Boeing a company that does nothing but make aircrafts they say they cannot replicate wind tunnel and some people who can't even figure out tomorrows weather perfectly will tell me that their computer model could predict world temperature 4-5 years in advance when it has failed to predict that its freaking too cool for Summer here? Bunch of lies!

Answer 7

taking emails out of context is not proof. The models are tested against historical data (backward time). Publishing the results in scientific journals and having many trying to find things wrong is the way it works. That is the formal scientific process. If the theory holds, it is accepted.

Answer 8

Lol...no clue. I just love reading the alarmists responses.



"You mean those emails that were taken out of context? Like "Hiding the Decline" using "Mike's nature trick" where you swap out data that doesn't agree with your theory with data that does? Totally out of context. They were just trying to swap out data that didn't agree with their theory with data that did."



"You mean the climategate emails that were illegal hacked even though there was no proof the servers were hacked and it might be someone from inside the CRU? Squirrel!!!!!"

Answer 9

I used to work in commercial manufacturing, so I have some familiarity with ISO requirements, such as ISO 9000. Those requirements were no guarantee that things would be done correctly, but instead addressed whether or not a manufacturer had their own standards and followed them. If problems did arise then there was a plan in place for how to address them. One way to make compliance easier was to have an internal ISO plan that was intentionally "loose." This possibility existed because manufacturers wrote their own ISO plans. While they had utility, there is also a tremendous variance in what they cover and how faithfully they are applied. Companies also were required to hire their own ISO inspectors, and it always struck me as odd, for example, that the company that I worked for in California always brought in their ISO inspectors from Ireland because they found them easy to work with.



As for what checks and balances there are in the development of climate models, for one thing they are peer-reviewed and often the result of collaborations between many people and institutions. The ones in public use are scientifically validated. The Community Earth System Model is an example of one such climate model



http://www.cesm.ucar.edu/



You can, if you so desire, become trained on use of the model, download the source code, run it, etc. On the web site you can find information on the "governance" of the model, which is probably the closest thing to ISO certification. Model units are controlled by working groups under the guidance of a scientific steering committee. Because of peer review and the open source nature of the model, and the requirement for scientific validation, I would say that the model is at least as rigorously checked as (and probably much more than) most commercial software. The proprietary nature of commercial software means that much of the code is unavailable for public inspection.



EDIT for jim z: I agree that models may be used for insight, and within their limitations quantitative results may be obtained from them, but they do need validation. Whether or not the study you mentioned was a valid use of the model and whether the results were justified I don't know.



I was actually trying to answer the question that was asked--that was about software assurance in case you missed it (admittedly the question was obscured by all the "additional details" that even Ottawa Mike acknowledged were irrelevant).



EDIT for Ottawa Mike: As usual, I don't think you're really interested in the answer to your "question." This is not commercial software, so expecting it to have commercial software assurance testing is a bit silly. Most software does not go through the software assurance testing you're talking about--perhaps it's expected in the banking industry or transportation applications, but this are neither of those. If the software crashes nobody gets hurt, they just try to figure out what went wrong and re-run the simulation. There is also no need for security features like you might find in banking software--although with groups like the Heartland Institute or the hackers of the UEA emails running around, perhaps they DO need those security features.



Again, this code is open (unlike virtually ALL commercial software), so if you think there's a problem with it, you can analyze the code line-by-line and pass what you find along to the working groups. As is your wont, you have no substantive criticism so you sling mud and hope some if it sticks. It mostly sticks to you.