Climate II

[choff]

Here's links for the MWP being global from the S. Hemisphere, nice try troll. As stated, by your very own broad definition of what constitutes lying, your guilty as sin. You can keep on spinning everything to the contrary as propaganda if you like, but then saying your own sources aren't biased isn't going to cut it with anybody.

http://www.co2science.org/subject/m/sum ... africa.php

http://www.thegwpf.org/medieval-warm-pe ... h-america/

http://www.theregister.co.uk/2012/03/23 ... ge_global/

http://www.universe-galaxies-stars.com/ ... eriod.html

http://ruby.fgcu.edu/courses/twimberley ... Palmer.pdf

[Schneibster]

The reason you hate Michael Mann is because he proved the MWP wasn't as warm as today.

http://www.desmogblog.com/2013/03/10/cl ... hreats-fly

What really got the ire of the climate science denial industry and its cheerleaders was that this second study showed that modern day temperatures were likely hotter than they had been in the so-called Medieval Warm Period. This negated a key argument from sceptics - which they continue with today - that it's been warmer in the recent past before the industrial revolution caused the westernised world to fall in love with fossil fuels.

Incidentally, it was never a very convincing argument anyway. Even if it was warmer in the past, it doesn't challenge the multiple lines of evidence which point to burning fossil fuels and deforestation as the main cause of the rapid warming, ocean acidification and sea level rises we see now.

Now stop lying.

[choff]

Here's a link on CO2 from real peer reviewed experts, not trolls.

http://www.ecofascism.com/review15.html

and no, it's not hotter than the MWP, we still don't grow barley in Greenland, Michael Mann's been busted in Climategates I and II for lying. It's not surprising you cling to him as a source, birds of a feather.

[Schneibster]

Hahaha, "Ecofascism.com?"

Bet they're real objective, huh?

[Teahive]

Not to mention submarine propellers and aircraft turbines. Curiously neither the Chinese nor the Russians, neither of whom are good at computer models, do a very good job of aircraft turbines. Why do you suppose that is?

"Ideally you're able to run lots of experiments with different starting conditions and take high quality observational data."

We have high quality data from plenty of experiments for these cases to be highly confident the models are reasonably complete. That's not the case for climate models. You can certainly gain important insights from an unverified model, but you can't be certain to have taken every significant circumstance into account, especially not if you're predicting unprecedented change.

[hanelyp]

"Ideally you're able to run lots of experiments with different starting conditions and take high quality observational data."

I recall back when the X-30 project was active a fair amount of discussion to the effect that the computer fluid dynamic models they had for the job needed to be calibrated against physical observation in order to produce quality results. The computer models were not a complete substitute for wind tunnel and free flight models. But they allowed engineers to converge on good designs with fewer (expensive) physical tests. Then once you had what the computer said was an optimal design another physical test would validate the results.

[Schneibster]

Not to mention submarine propellers and aircraft turbines. Curiously neither the Chinese nor the Russians, neither of whom are good at computer models, do a very good job of aircraft turbines. Why do you suppose that is?

"Ideally you're able to run lots of experiments with different starting conditions and take high quality observational data."

We have high quality data from plenty of experiments for these cases to be highly confident the models are reasonably complete. That's not the case for climate models. You can certainly gain important insights from an unverified model, but you can't be certain to have taken every significant circumstance into account, especially not if you're predicting unprecedented change.

We have data to check our climate models against, too.

An actual climate.

Maybe you forgot.

[Schneibster]

Not to mention submarine propellers and aircraft turbines. Curiously neither the Chinese nor the Russians, neither of whom are good at computer models, do a very good job of aircraft turbines. Why do you suppose that is?

"Ideally you're able to run lots of experiments with different starting conditions and take high quality observational data."

We have high quality data from plenty of experiments for these cases to be highly confident the models are reasonably complete. That's not the case for climate models.

Actually, you're wrong. We've known how to work these problems for a long time; use the Navier-Stokes equations and use definite numerical simulation. It is, in case you weren't aware, a problem worth a million dollar prize from the Clay Institute to prove either or both (they're believed to be related) of the existence problem, that is, whether or not an N-S equation exists for every possible physical flow, and the smoothness problem, that is, whether there are or are not singularities within real flows that N-S equations cannot model.

But we know how to solve N-S problems, at least how to extract a definite numerical solution with analysis. And the fifty-foot wide propellers work, and don't cavitate.

The problem is that the less computing power and the less time, the less precise the simulation is. Quite frankly we haven't had the computer power until the last ten years to show the human genome. Did you expect numerical simulation to take any less? We've just finally put together a cosmological model that shows a universe that comes out looking like what we see. Do you seriously contend this is a less complex undertaking than the climate?

To put it in so many words, no it doesn't have to do with experiments. It has to do with computer power and with the algorithms of numerical simulation and the math of the Navier-Stokes equations.

Now I have to point out that heat flow through the Earth-system, and even moreso through several of the significant subsystems including both the atmosphere and the ocean, is modeled with the N-S equations. That is, with numerical simulation of fluid flow.

Just like propeller models.

The increased computer power makes them both possible, and the continuing evolution of computers now makes them less and less difficult year by year. If you can run a thousand runs in a year, and see how close they come to the reality, with two different programs, it's relatively easy to tell which one compares better with reality; when you can only have one run in a month it's a much more difficult proposition.

That's the truth.

You can certainly gain important insights from an unverified model,

They're not unverified models. They can simulate the climate accurately to date from hundreds of thousands of years back, if not millions.

I hope you didn't think they were just sitting on their thumbs, or making stuff up, for twenty years. Your knowledge of how accurate these models are seems mired in the 1980s.

but you can't be certain to have taken every significant circumstance into account, especially not if you're predicting unprecedented change.

It's not unprecedented. It's unprecedented in the existence of the human race. Back to the PETM, the Paleocene-Eocene Thermal Maximum.

[Teahive]

We have data to check our climate models against, too.

An actual climate.

Maybe you forgot.

"where data capture is lacking in both quality and quantity"

Maybe you forgot to read.

The climate data we have is nowhere near the quality and quantity any serious climate scientist would wish for. Besides, we have no data describing a case that is similar to the future predicted by these models since it has never happened since proper data recordings began. That such changes may have happened in the distant past is beside the point since we don't have quality data for the distant past.

We don't know what the reaction and feedback mechanisms of this massively complex system called Earth to significantly rising temperatures under the current circumstances might be. Pretending that these models are known to be complete is detrimental to science.

[choff]

From ecofascism.com

Like Gray, many sceptics take aim at the “models”. Hendrik Tennekes’s CV lists: research director of the Royal Netherlands Meteorological Institute; chair of the Scientific Advisory Committee of the European Center for Medium-Range Weather Forecasts, and Professor of Aerospace Engineering at Penn State. He says climate models do not replicate reality. His motto is, “No forecast is complete without a forecast of forecast skill.” Simplifications required to design models render them useless as predictive devices. Models can’t predict rain. A computer’s power is irrelevant if the information plugged into it is defective. Tennekes is emphatic: “No amount of improvement in the quality of the observation network or in the power of computers will improve the average useful forecast range by more than a few days.” He avows: “Blind adherence to the harebrained idea that climate models can generate ‘realistic’ simulations of climate is the principal reason why I remain a climate sceptic.’” (65)

Freeman Dyson (PhD, Cambridge, Physics) declares climate models: “do not begin to describe the real world” and “do a very poor job of describing the clouds, the dust, the chemistry, and the biology of fields, farms and forests.” Climate modellers compensate with “fudge factors” and: “There are many fudge factors concerned with processes such as snow melting and vegetation-growth that cannot be modelled in detail.” Dyson is a Professor at Princeton and a fellow of the American Physical Society and the Royal Society of London. He dismisses “the fluff about global warming” the dangers of which are “grossly exaggerated”, adding that “increase in CO2 in the atmosphere has other consequences that may be at least as important as global warming – increasing crop yields and growth of forests, for example”. (66)

Antonino Zichichi says IPCC computer simulations “are incoherent and invalid from a scientific point of view” and are loaded with enough approximations to model a flying elephant. He is Professor Emeritus of Physics at Bologna U, President of the 10,000-member World Federation of Scientists, and author of ten books and 800 papers. He complains the IPCC has conned people into thinking “science has understood all about climate...but it’s not this way.” To predict climate one needs models of past climates including past behaviour of aerosols but of these things we “know little or nothing”. According to Zichichi: “if we ever come up with a mathematical structure capable of describing the past of the solid and liquid surfaces of earth, and only then, it will be possible to confirm what is being advocated today.” He adds, “It is not possible to exclude the idea that climate changes can be due to natural causes.” He thinks solar activities caused most of the recent warming. (67)

The most damning indictment of computer climate simulations comes from Kevin Trenberth – a veteran IPCC official and leading contributor to their Fourth Assessment Report (2007). Trenberth’s confession in Nature magazine’s “The Climate Change Blog” warrants a long quote:
“None of the models used by IPCC are initialized to the observed state, and none of the climate states in the models correspond even remotely to the current observed climate...the state of the oceans, sea ice, and soil moisture has no relationship to the observed state at any time in any of the IPCC models. There is neither an El Nino sequence nor any Pacific Decadal Oscillation that replicates the recent past; yet these are critical modes of variability that affect Pacific rim countries and beyond. The Atlantic Multidecadal Oscillation, that may depend on the thermohaline circulation and thus ocean currents in the Atlantic, is not set up to match today’s state, yet it is a critical component of the Atlantic hurricanes, and it undoubtedly affects forecasts for the next decade from Brazil to Europe. Moreover, the starting climate state in several of the models may depart significantly from the real climate owing to model errors...regional climate change is impossible to deal with properly unless the models are initialized.” (68)

[Schneibster]

We have data to check our climate models against, too.

An actual climate.

Maybe you forgot.

"where data capture is lacking in both quality and quantity"

Maybe you forgot to read.

Maybe you forgot the Vostok Core.

The climate data we have is nowhere near the quality and quantity any serious climate scientist would wish for.

The data is nowhere near the quality and quantity a high energy particle physicist would wish for, either, but we sure have found out a lot.

You do know that the current approach to the color/strong nuclear force is numerical simulation, right?

Oh, and BTW, did you notice how the weather forecasts went from a day to a three days to five days to ten days? Yep, that's right, numerical simulation. And gee, despite all those meteorologists swearing numerical simulation doesn't work, there they are making their livings off it.

Besides, we have no data describing a case that is similar to the future predicted by these models since it has never happened since proper data recordings began.

Actually simulating merely the last 150 years is the easiest part. It's taking it back 650,000 years to the beginning of the Vostok Core, or a few million to the continental shelf cores, and making it match all that, and then do everything we've seen in tree rings for a couple thousand years, and then finally make it match the industrial era, that's hard. Not very many models can do that. In fact only a handful; and two of them are run by NASA, the GISS AOGCM, and by the British Climate Centre (hope I got that right), the HADCRUT AOGCM. Both of them have been busy recently incorporating ocean dynamics and cloud dynamics and ice dynamics. It's been interesting watching the code evolve.

That such changes may have happened in the distant past is beside the point since we don't have quality data for the distant past.

Unfortunately for your point we have data of lower quality, but of many different origins. For example, ice cores from the Antarctic, from all of the mountainous areas of the world, mud cores from the continental shelves and from lake bottoms, tree rings, fossil tree rings (remember the petrified forest? Yes, you can see rings... and you can also carbon date things back fifteen or twenty thousand years. Combine the two for more reach, and for the climate of a given few-hundred-year stretch of time back then).

"We don't have quality data" != "We don't have any data." And increasing data quality is something we've been doing since we invented science, slightly after we invented fire.

We don't know what the reaction and feedback mechanisms of this massively complex system called Earth to significantly rising temperatures under the current circumstances might be. Pretending that these models are known to be complete is detrimental to science.

They don't have to be complete to be accurate. In fact, very few of our models are really complete; most of them, however, are getting very very accurate these days.

[Teahive]

You do know that the current approach to the color/strong nuclear force is numerical simulation, right?

Completely beside the point, because confirming one computer model does not confirm others trying to model completely different phenomena.

If you think it's been interesting watching the code evolve, watch how much it will further evolve in the future.

[Schneibster]

You do know that the current approach to the color/strong nuclear force is numerical simulation, right?

Completely beside the point, because confirming one computer model does not confirm others trying to model completely different phenomena.

So you think the N-S equations work one way in computer models of climate, and another way in computer models of galaxy formation?

Heh, that's amusing. Do you know what "numerical consistency" is?

Let me give you a hint: it's why when you turn into the coffee shop, you don't turn into a penguin. It would violate consistency.

The universe is consistent.

Maybe you forgot.

If you think it's been interesting watching the code evolve, watch how much it will further evolve in the future.

For N-S numerical simulations? Not unless/until they can solve the existence and consistency problems. All they'll be able to do is more simulation runs faster.

[Teahive]

Building something right isn't the same as building the right thing.

Maybe you forgot.

[Schneibster]

Building something right isn't the same as building the right thing.

Maybe you forgot.

You're changing the subject.

The subject is climate. There's only one climate. Either your simulation describes it accurately, and the weather models made from it work, or it doesn't and they don't.

And you're dissing the people whose simulations do, and whose weather models support 10 day forecasts.

Most folks don't understand that meteorologists who broadcast the weather do so with a local understanding of the history of the weather that they use to season the main data, which is what comes from the huge numerical simulations the NOAA's NWS runs every hour, using the latest hourly satellite and transponder data. The meteorologists never see it; and the evidence suggests that they think the NWS comes up with it by witchcraft or by being lucky every hour on the hour for the last twenty years. Very occasionally they see a set of circumstances where they agreed with the model forecast under a particular set of circumstances and it was wrong, and they correct it. Maybe twice a year.