Hatrack River
Home   |   About Orson Scott Card   |   News & Reviews   |   OSC Library   |   Forums   |   Contact   |   Links
Research Area   |   Writing Lessons   |   Writers Workshops   |   OSC at SVU   |   Calendar   |   Store
E-mail this page
Hatrack River Forum Post New Topic  Post A Reply
my profile login | register | search | faq | forum home

  next oldest topic   next newest topic
» Hatrack River Forum » Active Forums » Books, Films, Food and Culture » Another Global Warming Thread (Page 1)

  This topic comprises 2 pages: 1  2   
Author Topic: Another Global Warming Thread
Aros
Member
Member # 4873

 - posted      Profile for Aros           Edit/Delete Post   Reply With Quote 
I don't understand OSC's review on global warming. He links to an article signed by 16 scientists that purports that global warming isn't real. He then proceeds to gloat because he was right all along.

Correct me if I'm off-base, but I see the situation a bit differently. Global warming is a topic that is somewhat contested within the scientific community. The majority of scientists and scientific experimentation conclude that it is real, may be caused by man, and will be detrimental to current ecology. There is also a very vocal scientific minority that includes a few prominent members that discount all the global warming data, and they have a few valid points.

I would say that the conclusion to draw is that the science is somewhat inconclusive. But that doesn't alter the value proposition.

If the majority is right and the impact of continuing our current carbon escalation is detrimental (or catastrophic), world governments need to act now to decrease / reverse carbon footprints.

If the minority is right and global warming is natural and not detrimental, actions to decrease global emissions will create some level of market inefficiency. This will result in some level (projected to be minor) of currency devaluation and loss of market power for participating economies.

Bottom-line, assuming the science is inconclusive is the rational assumption. But isn't my cost-benefit analysis incontrovertible? Based on the uncertainty, we should definitely take steps toward carbon reduction. The costs are low and the potential benefits are high.

Am I missing something? Or is the political argument just some sort of strawman?

Posts: 1204 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
Orincoro
Member
Member # 8854

 - posted      Profile for Orincoro   Email Orincoro         Edit/Delete Post   Reply With Quote 
quote:
Global warming is a topic that is somewhat contested within the scientific community.
As some are so fond of saying.

Know what else is a hotly contested In the scientific community? Gravity.

So yeah. I love the mutually exclusive dual narratives from the deniers: "science is a religion! You don't buck any disagreement!" and "you disagree with each other, therefore you don't know!" any sense of proportion or reason is gone from this as an issue the public is aware of. Either way, it *is* the greatest lie of our time, whether it exists (it does by the way) or doesn't. Somebody, a lot of people, are deceiving us.

Posts: 9553 | Registered: Nov 2005  |  IP: Logged | Report this post to a Moderator
Scott R
Member
Member # 567

 - posted      Profile for Scott R   Email Scott R         Edit/Delete Post   Reply With Quote 
First, here's the article OSC refers to.

Now:

quote:
Based on the uncertainty, we should definitely take steps toward carbon reduction. The costs are low and the potential benefits are high.
It depends on what you mean by "taking steps." The article makes a point of contending that drastic steps would be unwise:

quote:
If elected officials feel compelled to "do something" about climate, we recommend supporting the excellent scientists who are increasing our understanding of climate with well-designed instruments on satellites, in the oceans and on land, and in the analysis of observational data. The better we understand climate, the better we can cope with its ever-changing nature, which has complicated human life throughout history. However, much of the huge private and government investment in climate is badly in need of critical review.

Every candidate should support rational measures to protect and improve our environment, but it makes no sense at all to back expensive programs that divert resources from real needs and are based on alarming but untenable claims of "incontrovertible" evidence.

There's a lot of subjectivity in the discussion.
Posts: 14497 | Registered: Dec 1999  |  IP: Logged | Report this post to a Moderator
Aros
Member
Member # 4873

 - posted      Profile for Aros           Edit/Delete Post   Reply With Quote 
There may be a lot of subjectivity in the article proper. My point is that a lot of people grasp onto the headline and conclude:
- Global warming isn't real and. . . .
- It's stupid to do anything about it.

Most people don't want to look at the actual scientific method. They have an opinion that is politically derived and will believe whatever credible experts that support their own predetermined position.

But then again, I am a scientist. What do I know.

Posts: 1204 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
Mucus
Member
Member # 9735

 - posted      Profile for Mucus           Edit/Delete Post   Reply With Quote 
Man, I'd like some "huge private and government investment in climate" first before I can get around to debating whether we should review and cut it.
Posts: 7455 | Registered: Sep 2006  |  IP: Logged | Report this post to a Moderator
Aros
Member
Member # 4873

 - posted      Profile for Aros           Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Orincoro:
[QUOTE]
Know what else is a hotly contested In the scientific community? Gravity.

This actually made me laugh. It is a very apt analogy.

The scientific community knows that both gravity and global warming exist. It's the specific causes and outcomes that are disputed.

A bit of an oversimplification, but elegant nonetheless.

Posts: 1204 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
Orincoro
Member
Member # 8854

 - posted      Profile for Orincoro   Email Orincoro         Edit/Delete Post   Reply With Quote 
The real problem is education. I got a first rate education, aside from my already being very smart. A lot of people get a really crappy education, and some of them are smart too- they're actually the most dangerous. Dumb people don't scare me, ignorant people do.


(edit to add: if anybody else is noticing a decline in my grammar or spelling, i dont have a brain tumor. I've been going to language school for 5 hours a day for the past month, so English is a little awkward at the end of the day. This will be continuing for the foreseeable future).

Posts: 9553 | Registered: Nov 2005  |  IP: Logged | Report this post to a Moderator
Aros
Member
Member # 4873

 - posted      Profile for Aros           Edit/Delete Post   Reply With Quote 
Good point. That's the whole problem with the Tea Party / Libertarianism. There are some great ideas out there. There are some really brilliant people. But it's the social-conservatism / intelligent design / anti-global-warming nutjobs that ruin it all. I know a few, and I think they epitomize the whole intelligent but uneducated model. And Ron Paul isn't exactly the most stable poster-boy.

On a side-note, I don't see why Libertarians view their doctrine so rigidly. A Democrat, for example, espouses social programs -- but not to the degree that the party advocates full-blown socialism. Why can't Libertarians advocate a more moderate form of their ideology? Unfortunately their ACTUAL political platform consists of legalizing all drugs and prostitution, coupled with shutting down the IRS. I'm just saying, there's a whole lot of middle ground where a realistic approach might prove effective.

Posts: 1204 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
twinky
Member
Member # 693

 - posted      Profile for twinky   Email twinky         Edit/Delete Post   Reply With Quote 
Public opinion on climate science is informed more by what politicians say than by what scientists say.
Posts: 10886 | Registered: Feb 2000  |  IP: Logged | Report this post to a Moderator
Dan_Frank
Member
Member # 8488

 - posted      Profile for Dan_Frank   Email Dan_Frank         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Aros:
Good point. That's the whole problem with the Tea Party / Libertarianism. There are some great ideas out there. There are some really brilliant people. But it's the social-conservatism / intelligent design / anti-global-warming nutjobs that ruin it all. I know a few, and I think they epitomize the whole intelligent but uneducated model. And Ron Paul isn't exactly the most stable poster-boy.

On a side-note, I don't see why Libertarians view their doctrine so rigidly. A Democrat, for example, espouses social programs -- but not to the degree that the party advocates full-blown socialism. Why can't Libertarians advocate a more moderate form of their ideology? Unfortunately their ACTUAL political platform consists of legalizing all drugs and prostitution, coupled with shutting down the IRS. I'm just saying, there's a whole lot of middle ground where a realistic approach might prove effective.

Libertarians, especially big-L Libertarians, tend to be more idealistic than pragmatic. They believe that violation of the Non-Aggression Principle (or whatever their flavor of it may be) is immoral, so they want to violate it by immediately, forcibly deconstructing the state so that nobody will violate it anymore.

Is... not the best plan. Ron Paul isn't the only nut.

That said, you're absolutely right. With a government as huge as ours, injecting a little bit of libertarianism into both parties wouldn't be a bad thing.

Some small-l libertarians do advocate for change on a pragmatic, piecemeal basis. Some of them actually still believe that there are more libertarian ideas they aren't necessarily focusing on that are also good and worth implementing, but they also recognize that smashing the state in the pursuit of progress is a terrible idea.

On the actual topic of global warming, I generally accept that it's happening, although I wonder how catastrophic it actually is. Every meaningful prediction of catastrophe that I've seen so far has been woefully wrong, but obviously that could change.

More importantly, though, what boggles my mind is how thoroughly green/environmentalist/luddites have largely hijacked the cause. I think the most egregious example of this is in regards to nuclear power.

Nuclear power has minimal carbon footprint and generates vastly more power than solar/wind/etc. And yet most green groups demonize it and try to stop nuclear advancement as much as possible. If global warming is so catastrophic that it's worth jeopardizing 90% of the country's energy sources, why wouldn't they embrace a powerful technology like nuclear as a viable replacement?

This speaks to a larger issue with the green movement that I have, which is that all of its proposed solutions are very backwards. Reducing energy usage, giving up the most efficient non-nuclear power sources, etc. I would much rather see serious effort being made to create new technology to counteract the carbon that is so dangerous, like designing a better plant. And yet that sort of thing isn't mainstream research right now.

David Deutsch gave a TED Talk years ago where he said some of this. I agree with it.

Posts: 3580 | Registered: Aug 2005  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Aros-

I think the costs of decreasing our carbon footprint to the levels recommended by the IPCC report will do more than result in minor currency devaluation and "loss of market power" for participating economies. Or, I guess, just that the "loss of market power" will be extensive. It'll potentially cost tens or hundreds of trillions of dollars to get to target CO2 levels, particularly if we focus solely or almost solely on renewables. On the other side, it's unclear how "catastrophic" the effects of global warming truly are.

I'm not going to try and advocate those positions (it feels almost as exhausting as stepping into a gay marriage debate), although I think I could probably find statements from at least semi-respectable economists saying as much. Really, I just don't think the cost-benefit analysis of action vs. inaction is quite as cut and dried as you indicate.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
Lyrhawn
Member
Member # 7039

 - posted      Profile for Lyrhawn   Email Lyrhawn         Edit/Delete Post   Reply With Quote 
There are dozens of beneficial effects of switching away from dirty power sources that have nothing to do with climate change. And yes, it will cost a lot of money...but as opposed to what? You don't think sustaining the current system is already costing trillions of dollars?
Posts: 21420 | Registered: Nov 2004  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Again, I'm not really interested in going into a lengthy discussion advocating the skeptical economic arguments surrounding climate change. There are definitely economists, including very prominent ones like Arrow, Solow, Sen, and Stiglitz, who believe the benefits of action outweigh the costs. And, if anything, economic opinion is shifting in this direction.

However, there's still significant diversity of opinion on the issue on the economics side, much more than there is on the science side, and people who believe that the supremacy of the "act now" stance is established are, IMO, premature.

For more on both sides of the economic debate, you could start with this Wikipedia article, particularly the section on the response to the study:

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

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
Aros
Member
Member # 4873

 - posted      Profile for Aros           Edit/Delete Post   Reply With Quote 
I do know that there are a lot of countries making a lot more effort than we are, without significant economic collateral damage. I don't know if they're meeting the IPCC report standards, however.

I remember hearing on NPR that we're the only first world nation, and the only large power outside of China and Russia, opposed to a lot of the UN climate initiatives.

Posts: 1204 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
Jeff C.
Member
Member # 12496

 - posted      Profile for Jeff C.           Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Aros:
But it's the social-conservatism / intelligent design / anti-global-warming nutjobs that ruin it all.

When you say Intelligent Design, do you mean the idea that the Big Bang happened and that some kind of god began the process (i.e. the hand that ignited the spark)? Or do you mean the idea that God made the Universe and everything in it instantly and that evolution never occurred (i.e. random crap popping up out of thin air)?

I ask because those are two very, very different things and I am not entirely clear on the terms associated with them. I've heard the term "Intelligent Design" tossed around a few different times, and people have taken it to mean a few different things.

Posts: 1322 | Registered: Feb 2011  |  IP: Logged | Report this post to a Moderator
Orincoro
Member
Member # 8854

 - posted      Profile for Orincoro   Email Orincoro         Edit/Delete Post   Reply With Quote 
"Intelligent Design" refers to a thinly veiled pseudoscientific claim that, basically, the universe is in some way controlled by God, and that humans would not exist without a supernatural force to create them, and that evidence of this exists, particularly regarding evolution. It is creationism.

It is, without any reservation, a huge lie and propaganda campaign against science education, by people who feel that science is wrong, and that it threatens religion or faith (in this case, concerning the Christian god). Most of the actual material claims of ID are either completely bogus, or are based on fundamental misunderstandings of scientific theories and their applications. Particularly, ID creationists (ID and Creationism are the same term), claim that biological systems exhibit "irreducible complexity," the phenomenon that exists in all complex life forms- that an appendage or cell or other biological device would appear to have no intermediate function that would cause it to evolve. For instance- a half developed finger has no function, or the complexity of the human eye, or ear, would not develop on its own.

These claims are outright bogus. In some cases the process is not fully understood, but there are loads of examples of he seemingly irreducible complexity is reduced in an observable way. Anyway, the claim is propaganda for Young earth creationism. That's about it.

Posts: 9553 | Registered: Nov 2005  |  IP: Logged | Report this post to a Moderator
TomDavidson
Member
Member # 124

 - posted      Profile for TomDavidson   Email TomDavidson         Edit/Delete Post   Reply With Quote 
quote:
However, there's still significant diversity of opinion on the issue on the economics side, much more than there is on the science side...
Well, to be fair, if you ask three economists a yes/no question, you'll get eight answers.
Posts: 36934 | Registered: May 1999  |  IP: Logged | Report this post to a Moderator
rivka
Member
Member # 4859

 - posted      Profile for rivka   Email rivka         Edit/Delete Post   Reply With Quote 
What if they're Jewish economists?

The mind boggles.

Posts: 32919 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
Thirty[url=http://online.wsj.com/article/SB10001424052970204740904577193270727472662.html] expert response to that Wall Street Opinion piece.

This response is signed by 38 top climate experts. It was authored by Keven Trenberth, whose views are horribly misrepresented by a deliberate strip quoting in the original article.

quote:
Do you consult your dentist about your heart condition? In science, as in any area, reputations are based on knowledge and expertise in a field and on published, peer-reviewed work. If you need surgery, you want a highly experienced expert in the field who has done a large number of the proposed operations.

You published "No Need to Panic About Global Warming" (op-ed, Jan. 27) on climate change by the climate-science equivalent of dentists practicing cardiology. While accomplished in their own fields, most of these authors have no expertise in climate science. The few authors who have such expertise are known to have extreme views that are out of step with nearly every other climate expert. This happens in nearly every field of science. For example, there is a retrovirus expert who does not accept that HIV causes AIDS.

quote:
Climate experts know that the long-term warming trend has not abated in the past decade. In fact, it was the warmest decade on record. Observations show unequivocally that our planet is getting hotter. And computer models have recently shown that during periods when there is a smaller increase of surface temperatures, warming is occurring elsewhere in the climate system, typically in the deep ocean. Such periods are a relatively common climate phenomenon, are consistent with our physical understanding of how the climate system works, and certainly do not invalidate our understanding of human-induced warming or the models used to simulate that warming.

Thus, climate experts also know what one of us, Kevin Trenberth, actually meant by the out-of-context, misrepresented quote used in the op-ed. Mr. Trenberth was lamenting the inadequacy of observing systems to fully monitor warming trends in the deep ocean and other aspects of the short-term variations that always occur, together with the long-term human-induced warming trend.


Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Rabbit-

As an expert, how long (in your opinion) would a period of no appreciable surface temperature rise need to be before climate scientists should begin to question their models' validity? It seems to me (from a naive Bayesian perspective) that at some point (30 years? 50 years? 100 years? 1000 years?) a short-term aberration would be sufficiently long to call into question the accuracy of the models themselves.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
TomDavidson
Member
Member # 124

 - posted      Profile for TomDavidson   Email TomDavidson         Edit/Delete Post   Reply With Quote 
quote:
As an expert, how long (in your opinion) would a period of no appreciable surface temperature rise need to be before climate scientists should begin to question their models' validity?
The question, I think, is which models would be considered invalid. If everything else is warming as predicted but surface temperature is not, models which attempt to describe changes in surface temperature are the ones which would be questioned.
Posts: 36934 | Registered: May 1999  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Maybe that's my naivete showing through. I would have thought there were integrated models, that you couldn't separate them from each other that way.
Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
scholarette
Member
Member # 11540

 - posted      Profile for scholarette           Edit/Delete Post   Reply With Quote 
Micheal Mann made a good statement regarding debate:
"Whats so unfortunate is that were stuck in this bad-faith debate in the public discourse, over whether the problem is actually real. This has prevented us from having the good-faith debate that is to be had, about what are the uncertainties, and what implications they have. There is just this amazing gulf between what the public thinks were debating, and what were actually debating at meetings and in the scientific literature."

Posts: 2223 | Registered: Mar 2008  |  IP: Logged | Report this post to a Moderator
Samprimary
Member
Member # 8561

 - posted      Profile for Samprimary   Email Samprimary         Edit/Delete Post   Reply With Quote 
quote:
This has prevented us from having the good-faith debate that is to be had, about what are the uncertainties, and what implications they have. There is just this amazing gulf between what the public thinks were debating, and what were actually debating at meetings and in the scientific literature."
It's uncanny; change the time a little bit and we're talking about evolutionary theory.
Posts: 14070 | Registered: Aug 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by SenojRetep:
Rabbit-

As an expert, how long (in your opinion) would a period of no appreciable surface temperature rise need to be before climate scientists should begin to question their models' validity? It seems to me (from a naive Bayesian perspective) that at some point (30 years? 50 years? 100 years? 1000 years?) a short-term aberration would be sufficiently long to call into question the accuracy of the models themselves.

Statistically, the normal year to year variation in the global surface temperatures is between 0.3 and 0.4 degrees centigrade. The predicted warming per decade is around 0.15 C. From an extremely simplistic analysis, that means we have to average over at least 12 years before we could conclude there was any significant trend in the temperatures and over a much longer period to conclude that there was no warming trend. That question is really moot at this time because, contrary to what is claimed by climate deniers, global average temperatures have been rising for the past decade. The argument only exists at all because 1998 was an anomalously hot year due to a very strong el-nino. None the less, the past decade is the warmest decade on record. Every year from 2001 - 2011 was hotter than every previous year since 1850 except 1998.

But the real answer to your question, is zero years. Scientists are continually questioning the validity of their models and working to improve them regardless of recent temperature trends. No one is waiting for 10, twenty or thirty years.

There are some parts of the models that we know with certainty are valid, things like conservation of matter and energy, the laws of radiant heat exchange, absorptivity of gases, phase equilibrium, gas laws, fluid mechanics, transport phenomena and so forth. And some parts the problem we know we don't understand very well. We aren't ever going to start doubting the validity of fundamental pieces of the model based on surface temperature trends. If surface temperature trends don't fit the models, it is almost certainly because there are factors we have not included in the models and not because we've got the basic physical laws wrong.

There are other pieces of the model that scientists know we don't understand very well, things like how forests and algae will respond to changes in the atmosphere, how ocean currents will change and how particles in the atmosphere scatter light and lead to cloud formation. Scientists are continually working to improve our understanding of these pieces.

We also know that one of the major sources of error in the models is in the approximations that must be made to solve the very complex dynamic mathematical systems. Efforts are under way continually to improve these as well. No one is waiting to see if current predictions prove accurate to question and refine these models. Its a continuous on going process.

But perhaps that isn't your real question. Maybe what you meant was "how long would a period of no appreciable surface temperature rise need to be before climate scientists should reject the hypothesis that humans are changing the global climate by adding green house gases to the atmosphere. There is not a simple answer to that question because the hypothesis itself is not simple. There is a really basic part of that theory which describes a fundamental relationship between green house gases and the earths temperature and there are more complex parts which try to predict dynamic processes in the air, ocean and biosphere of the planet.

Rejecting the basic theory would require a major revision of the fundamentals of physics. No one in science is arguing that the basic theory is unsound, even people like Richard Lindzen don't reject the basic theory. We know that the temperature of the earths surface is a result of radiant heat exchange with its surroundings. That is beyond question. The radiant heat balance is without any doubt influenced by the composition of infrared absorbing gases, like CO2, H2O and CH4 in the atmosphere. No scientist questions this. The levels of CO2 and CH4 in the atmosphere are unquestionably increasing and human activity is the cause. That is as certain as things ever get in science.

If current models don't accurately predict warming, it has to be because the energy is going some where other than the surface and not because energy isn't conserved. If the models don't accurately predict, then we have to figure out where the energy is going and figure out how to include that in the models to better predict the system dynamics. But the basic theory that changing the concentration of IR absorbing gases in the atmosphere will result in higher temperatures of the planet surface is not going to change unless we discover something that completely up ends the fundamental laws of thermodynamics.

Of course we don't know what we will find as we continue to study some of the big unknowns, like ocean dynamics, clouds, and plants, but we are able to see how sensitive the models are to those unknown factors. That sensitivity analysis is done all the time and is one of the reasons that there are error bars in the predictions. So far, nothing that has happened falls outside the range of what's been predicted. If it did, we'd have to do some serious revisions of the models. What we can say based on what we know with confidence right now is that its extremely unlikely that future research will tell us that the problem is just going to go away on its own.

[ March 07, 2012, 02:36 PM: Message edited by: The Rabbit ]

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by The Rabbit:
Perhaps that isn't your real question. Maybe what you meant was "how long would a period of no appreciable surface temperature rise need to be before climate scientists should reject the hypothesis that human are changing the global climate by adding green house gases to the atmosphere."

That's approximately what I was asking, if you switch "changing the global climate" to "changing the global surface temperature in a predictable way."

quote:
So far, nothing that has happened falls outside the range of what's been predicted. If it did, we'd have to do some serious revisions of the models.
If I look at a chart of predicted surface temperature increase over time like this one from IPCC 2007 for instance, it seems to indicate that under a moderate CO2 growth rate should expect about a 0.3C temperature anomaly from the 2000 baseline by 2012. The only error bars on the chart are out at 2100, but if we assume uncertainty grows exponentially one would expect very tight error bounds on the near-term predictions (on the order of +/- 0.05C). But this chart (or the one Lyrhawn linked from the other thread) seems to show a change from the 2000 baseline, even if you adjust for the El Nino and La Nina events, of about 0.075C. This is much less (in terms of standard deviation) than the change of 0.25-0.35C I would have predicted from the IPCC chart.

There are probably lots of reasonable explanations for this difference. But as someone with a technical research background outside the field of climate science, the difference between the prediction in the first chart and the realization in the second seems surprising.

<edit>Here's a somewhat dated chart from Roger Pielke, Jr. making a somewhat similar point.</edit>

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Samprimary:
quote:
This has prevented us from having the good-faith debate that is to be had, about what are the uncertainties, and what implications they have. There is just this amazing gulf between what the public thinks were debating, and what were actually debating at meetings and in the scientific literature."
It's uncanny; change the time a little bit and we're talking about evolutionary theory.
With one big exception. It doesn't really matter much whether Americans believe in evolution or not. It's not something that's caused by human activity or that is likely to be changed by political decisions. Society doesn't need to respond to evolution. Evolution won't stop because Texas school boards strike it from the curriculum. The worst thing that could happen if they teach "intelligent design" instead of science in America is that America might start falling behind in medicine and biotechnology.

In contrast, the best science out there says that global climate is being changed by humans. The best science out there says that if we don't make some major changes the consequences will be disastrous for humans and almost everything else that lives on the planet. The best science out there says this a matter of utmost urgency. W We have to act now, sooner if it were possible. We can't wait a generation until all the questions are answered and the skeptics are either persuaded or die. What Americans believe about climate science influences their willingness to make the changes that science says we have to make. What ordinary Americans believe about climate science matters enormously to everyone and everything on the planet.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
If I look at a chart of predicted surface temperature increase over time like this one from IPCC 2007 for instance, it seems to indicate that under a moderate CO2 growth rate should expect about a 0.3C temperature anomaly from the 2000 baseline by 2012. The only error bars on the chart are out at 2100, but if we assume uncertainty grows exponentially one would expect very tight error bounds on the near-term predictions (on the order of +/- 0.05C).
I don't know how you are reading that graph. The text from the same report says "For the next two decades a warming of about 0.2C per decade is projected for a range of SRES emissions scenarios." link.

The primary purpose of the graph is to show how changes in human emissions of greenhouse gases will most likely impact future climate change. For this reason, error bars were included only for the final point to avoid confusing the issue. The error bars are the result of a sensitivity analysis that accounts for many of the scientific uncertainties. Those uncertainties are independent of the human response and so they affect all the scenarios the same way. They don't actually influence the trend predicted for the different scenarios. In other words, even though the "error bars" for scenario B1 and A1B overlap, the models indicate that A1B will always result in a large temperature change than B1 regardless of the other uncertainties.

Error bars have been estimate for the full period. They don't grow exponentially, they are roughly a constant percent of the change at any given point. That expected percent error is actually higher for very short time periods because the models don't include the factors that cause short term variations (like el-ninos).

Using the predictions from the 2007 IPCC report we would expect that the period from 2000 - 2009 should have been on average 0.1 to 0.3 C warmer than the 1990 - 1999. The measured temperature record shows that the average temperature from 2000 to 2009 was in fact 0.200 C higher than the average temperature from 1990-1999. That's dead on. The uncertainty in the temperature measurements themselves is at least 0.05C.


quote:
But this chart (or the one Lyrhawn linked from the other thread) seems to show a change from the 2000 baseline, even if you adjust for the El Nino and La Nina events, of about 0.075C.
I don't know how you are reading either graph. There is a lot of year to year variability so you have to average over 5 to 10 years to expect to see any trends. If I interpolate the numbers from that graph, I get that the average temperature between 2000-and 2009 was 0.18C higher than the average temperature in the previous decade. If I do a running ten year average, there is a continuous increase. The average temperature anomaly from 2006 - 2011 is 0.1C higher than the average temperture from 2000-2005. The numbers reported in the scientific literature using the actual numbers rather than interpolating from a graph say that the decade from 2000-2009 was 0.200 degrees warmer than the previous decade.

quote:
This is much less (in terms of standard deviation) than the change of 0.25-0.35C I would have predicted from the IPCC chart.
Once again, I must conclude that the error is in your reading of the chart and not in the science. The scientific reports predicted ~0.250% warming for the past decade. The measurements show 0.2 degrees of warming. There is no discrepancy.

[ March 07, 2012, 08:33 PM: Message edited by: The Rabbit ]

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Thanks for the information on the errors. A 100% standard deviation (i.e. 50%) on the anomaly is much bigger than I would have expected. Why (if it's not too technical) does it end up being proportional to the expected value of the anomaly rather than growing exponentially with time. I would have thought there was some aggregate uncertainty in the models that would cause the standard deviation to grow super-linearly.

Given a 0.2C/decade growth, the IPCC would suggest that over the past 12 years we should expect a 0.24C change (or 0.12-0.36C at 90% confidence), which is different, but not that different that the 0.3C I eyeballed off the chart. The big difference from what I said above is really that the near-term uncertainty is much larger than I would have guessed from the IPCC chart.

That said, I don't know how you're seeing even a 0.2C differential anomaly from 2000 to 2012 in those charts. I would guess, based on the Wikipedia chart I linked above, an exponentially weighted average in 2000 to be about 0.5. The reported anomaly since then seems to be a relatively constant 0.575 (adjusting for El Ninos/La Ninas), which is how I got my estimate of a 0.075 temperature anomaly between 2000 and 2012.

What would you estimate the smoothed anomaly estimate to be in 2000 and 2012 based on the chart I linked (and how would you come to that estimate)?

<edit>Rereading your analysis above, are you getting an estimate of the anomaly in 2000 by uniformly weighting all the observations from 1991-2000, then getting an estimate for 2012 by uniformly weighting all the estimates from 2003-2012? Cause I can see how that would get you to a 0.2C anomaly differential (although it's certainly not the data smoothing method I would use).</edit>

[ March 07, 2012, 05:27 PM: Message edited by: SenojRetep ]

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
<edit>Here's a somewhat dated chart from Roger Pielke, Jr. making a somewhat similar point.</edit>
The graph linked appears to a homework assignment not a scientific publication. There no indication of what if any data was used to make the graph. Global average temperatures from 2000 to 2007 do not fall on a straight line. The "trend" you get depends very heavily on the window you use to average the data. If you average over the 5 years prior to 2001 and 2007 respectively, you get a temperature change of 0.09C (or 0.15 C per decade). If you average of over the previous 6 years you get 0.11C (0.18 C per decade) If you average over 10 years, you get 1.4C (0.23 C per decade) which is a bit higher than the IPCC 2007 prediction line he shows. It appears he cherry picked his data window to make a point since any other way of averaging the data is more consistent with the predicted trends. Even so, the IPCC 1990 report predicted a change between 0.1C per decade and 0.5C per decade. If he'd shown the range of the predictions rather than just the average, his (apparently) cherry picked data line would fall squarely in the range predicted 20 years ago.
Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
The "homework assignment" was tongue in cheek, but it was just a blog post, not a peer-reviewed paper.

I agree the window function is very important. But your expanding window function is illustrative; averaging (especially with uniform weightings) over long windows will tend to miss changes in the pattern. And considering only causal filtering (averaging over the past only) rather than non-causal smoothing (averaging over a window centered at the point of interest) will, in general, give you worse MSE. Sometimes you have to (like if you want to estimate the anomaly in 2011 and you don't have data for 2013 and beyond), but if you don't you shouldn't.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Thanks for the information on the errors. A 100% standard deviation (i.e. 50%) on the anomaly is much bigger than I would have expected. Why (if it's not too technical) does it end up being proportional to the expected value of the anomaly rather than growing exponentially with time. I would have thought there was some aggregate uncertainty in the models that would cause the standard deviation to grow super-linearly.
The absolute uncertainty does grow with time but the predicted temperature change also grows with time so the ratio of the two turns out to be nearly constant. It's a pretty common behavior for lots of systems. It's even more common for the relative uncertainty in measured or predicted values to decrease as the absolute value of the predicted value increases.

In climate modeling there is kind of a double whammy in predicting short range changes. First, off the models aren't designed to capture short term temperature fluctuations so they aren't going to predict accurately until the long term trend gets to be bigger than the short term fluctuations. Second, we there is also considerable uncertainty in our measurements of the global average temperature. Right now, that uncertainty in the temperature measurements could account for a significant fraction of differences between the measured and predicted values.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by SenojRetep:
The "homework assignment" was tongue in cheek, but it was just a blog post, not a peer-reviewed paper.

I agree the window function is very important. But your expanding window function is illustrative; averaging (especially with uniform weightings) over long windows will tend to miss changes in the pattern. And considering only causal filtering (averaging over the past only) rather than non-causal smoothing (averaging over a window centered at the point of interest) will, in general, give you worse MSE. Sometimes you have to (like if you want to estimate the anomaly in 2011 and you don't have data for 2013 and beyond), but if you don't you shouldn't.

I agree absolutely, but that doesn't change my conclusion. The short term fluctuations in temperature are too large to determine whether there has been any significant increase or decrease in the slope of the curve. Anyone who claims otherwise is either misinformed or lying.

What we do know with out question from the data is that during the past decade temperatures have been higher every single year than in any year (save one) prior to this decade.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
That said, I don't know how you're seeing even a 0.2C differential anomaly from 2000 to 2012 in those charts.
I read the number for each year from the graph and calculated the average anomaly for 1990 - 1999 and the average anomaly from 2000 - 2009. There is certainly some error in my reading of the graph but the numbers I got are reasonably close to the numbers reported elsewhere for the average anomaly for the two decades. 0.2 C is not the number I got from the graph, I got 0.18 C from the graph. 0.2 C is the number I got from the reported average values for the two decades. Like I said earlier, there is a lot of uncertainty in the measurements themselves, probably ~ 0.05C.
Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
<edit>Rereading your analysis above, are you getting an estimate of the anomaly in 2000 by uniformly weighting all the observations from 1991-2000, then getting an estimate for 2012 by uniformly weighting all the estimates from 2003-2012? Cause I can see how that would get you to a 0.2C anomaly differential (although it's certainly not the data smoothing method I would use).</edit>
I didn't use a weighted average. Given that the year to year variability is so much greater than the 10 year trend, I don't think it makes sense to use anything else. If you do, you will be inordinately influenced by spikes like the one in 1998. The weighting the data as you did will make it more sensitive to changes in the derivative but it will also make the derivative more sensitive to the noise.

Look more carefully at the data in [url= http://upload.wikimedia.org/wikipedia/commons/f/f9/Enso-global-temp-anomalies.png] in the graph you linked[/url]. There is not a consistent trend within any 10 year slice of the data. There is a fairly consistent upward trend from 1992 to 1998, but if you go from 1990 to 1996, there is no real trend. The trend is only evident if you look at the several decades. The data is simply to noisy to conclude that there has been any change in the derivative over the last decade. There are statistics we could use to put a confidence interval on it if you want me to try.

[ March 08, 2012, 09:05 AM: Message edited by: The Rabbit ]

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Rabbit-

I've thought a lot about this over the past couple of days, and I really think the analytical method you're using to estimate the process value is introducing lag effects. By using a uniformly weighted tail estimator you're introducing a 5 year lag into your estimates, giving a good estimate of the temperature differential between 1995 and 2006, but a relatively poor one between 2000 and 2011. It's also very sensitive to window length (as you pointed out in your response about Pielke's anlaysis); if you use a window shorter than 10, you get a significantly lower estimate, as you do if you use a window longer than 20. In all, depending on how you set your window, you could get any estimate between 0.06-0.21C (all of which are below the expected value of 0.22, several by more than a standard deviation).

I've mentioned the approach you're using to several people experienced with analysis of stochastic processes and the uniform response has been "that's not right" or "why would someone do that?"

Using my exponentially weighted average, depending on the decay rate, you get estimates between 0.07C-0.17C. I also ran a collective intelligence experiment, showing the chart (minus the potentially biasing information about what the observations represented) to twenty people, all with experience in this sort of analysis, most with post-graduate technical degrees. The consensus was that the change from 2000 to 2011 was between 0.05C-0.1C with a mean estimate of 0.75C.

I also ran the data using your uniform weighting technique but eliminating the lag (i.e. centering the window on the data). In this case, depending on window length, you get a range of estimates 0.03-0.11C (where, to deal with the fact that there's no future data after 2011 I've used a half-window smoothing, which isn't ideal but is justifiable).

I think you're being pretty glib that there's nothing surprising about the warming data from the last decade. There are several well-justified estimation methods which would find the temperature growth as being very surprising w.r.t. the IPCC model.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
Senoj, Yes I agree that doing a straight unweighted average will make it more difficult to see when there has been a change in the long term trend. But doing a weighted average as you are doing makes it much more likely that you will mistake short term oscillations for a change in the long term trend. Luckily, there is an alternative approach. We can do a statistical test to see there is a significant difference between the trend observed this decade and the trend observed over the previous decade. There is not. What you are trying to do is, in effect, is estimate whether a set of very noisy data has a non-zero second derivative. That just isn't something that can be done accurately without a lot more data.

No way of weighting the data can change the fact that we simply don't have enough data to know whether the difference between the temperature trend this last decade and the one before it are the result of short term natural oscillations or a change in the underlying rate of surface warming. We simply can not reject the hypothesis that the earth is continuing to warm at a rate consistent with the models based on the data we have.

But even if the data for this decade turns out to be inconsistent with the current model, it still doesn't tell us why. If the trend over the past few years does continue long enough to indicate a problem with the model, it doesn't tell us what that problem is or how that problem will effect the long term prediction. And no one in science is just waiting for another 5, 10 or 50 years to find out. We know where there are big knowledge gaps and uncertainties in the existing models. Scientists are working hard to fill those gaps. Other scientists are working hard to determine how sensitive the models are to those uncertainties and how much they will affect the long term trend.

The challenge is that society can't just wait until the scientist figure it all out. If the models are even remotely correct, we have to act now to make a difference. Regardless of the various uncertainties in the model, we know the basic premise is valid. Increasing the concentration of green house gases in the atmosphere will trap more heat in the planets atmosphere and oceans. This much is not debatable.

The precaution principle says that in the absence of scientific certainty, we should act in the way that has the least potential for harm. There is no question that harm that will be caused by climate change, if 97% of climate scientists have it right, will vastly outweigh the harm that will be caused by decreasing our green house gas emissions.

If your doctor told you there was a small mass in your child's abdomen that has an 80% chance of being cancer but it's too small to do a needle biopsy so they can't be sure without surgery. They could operate now and take it out or wait a few months to see if it gets bigger so they can do a biopsy. If its cancer and they operate now, your child has a 99% chance of a full recovery. If it's cancer, every month they wait your child's chance of survival drop by 20% and the time the child will need to spend in the hospital and on chemotherapy goes up. If they operate now and it's not cancer, you will have wasted a bunch of money and your child will have 3 or four days in unnecessary days in a hospital and a week or two recuperating. What would you do?

When we decide what to do about climate change, we are making that kind of decision right now for every child on the planet

[ March 09, 2012, 11:55 AM: Message edited by: The Rabbit ]

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
I think you're being pretty glib that there's nothing surprising about the warming data from the last decade. There are several well-justified estimation methods which would find the temperature growth as being very surprising w.r.t. the IPCC model.
What I find most surprising about the temperature data from the last decade is not the lack of a strong trend, that's easily explained by the normal variability. What I find most surprising about the temperatures from the last decade is its consistency. Every single year since 2001 has warmer than every year on record prior to 2001. Every year since 2001 was warmer than 149 of the past 150 years. The only year prior to 2001 that was warmer than the coldest year in the last decade, was 1998.

That isn't a result of the way I've smoothed the data to make a point. Those are the raw numbers.

If someone is being glib about the data, its you. You are ignoring the elephant in the room. You are arguing about whether the model correctly predict the slope of a noisy data set of a small time window and ignore the fact that every single year in that time window has broken the pre-1997 record high.

The only scientific theory that can explain why every single year in the past decade might be record setting hot is the theory that human emissions of green house gases are causing global warming. The models might not be getting the all details quite right yet, but they sure got the big picture.

No one has proposed anything else that can explain that big picture and if you are going to argue that we should ignore the explanation supported by 97% of climate scientists, you sure better have a theory that fits the data at least as well.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
King of Men
Member
Member # 6684

 - posted      Profile for King of Men   Email King of Men         Edit/Delete Post   Reply With Quote 
This may be a stupid question. You mentioned that a strong El Nino had made 1998 a very hot year. What is the mechanism for this? Where does the additional energy come from?
Posts: 10589 | Registered: Jul 2004  |  IP: Logged | Report this post to a Moderator
MattP
Member
Member # 10495

 - posted      Profile for MattP   Email MattP         Edit/Delete Post   Reply With Quote 
I believe it's caused by lower winds perturbing the ocean less, exposing less cold deep-ocean water to the surface, resulting in warmer ocean surfaces, resulting in warmer everything. Or something like that.
Posts: 3263 | Registered: May 2007  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
I agree, the uniformity of the data over the last decade has been remarkable. I've found that very surprising as well.

quote:
No way of weighting the data can change the fact that we simply don't have enough data to know whether the difference between the temperature trend this last decade and the one before it are the result of short term natural oscillations or a change in the underlying rate of surface warming. We simply can not reject the hypothesis that the earth is continuing to warm at a rate consistent with the models based on the data we have
This gets back to my original question, which is how much data would be required for the observed change in the trend to become statistically significant in your opinion. If observable temperature were flat for another 10 years, would that be significant? another 30? It seems to me that at some point you have to be able to reject the hypothesis that global surface temperature is growing linearly, even in the absence of a causal explanation.

Your policy concerns aren't immediately relevant to what I'm asking. I appreciate that the implications of global warming are significant, and optimal decision theory might dictate taking action to avoid even unlikely catastrophes (although I think the precaution principle you cite, if I understand it, is a rather naive way of looking at such decisions). But that's not my purpose in pursuing this particular conversation.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by King of Men:
This may be a stupid question. You mentioned that a strong El Nino had made 1998 a very hot year. What is the mechanism for this? Where does the additional energy come from?

The additional energy comes from the deep oceans, sort of. More accurately, the surface temperature differences observed in an El Nino/La Nina event are caused by changes in ocean currents that result in differences in the way energy is transported between the surface and the deep oceans. In an El Nino, more of energy stays at the surface and in a La/Nina more energy gets transported to the deep oceans. They are two halves of a cycle so that the result is that temperatures in the deep and the surface will oscillate about a constant point.

No one really understands what causes the changes in the ocean currents. They might be caused by changes in the trade winds, but then no one know what causes the changes in the trade winds.

Understanding how global warming will affect energy transport in oceans is one of the huge unknowns in predicting the rate that surface temperatures will change. The steady state solution (i.e ultimate temperature change after a long enough time) isn't affected much, but the rate at which we will approach that steady state depends a great deal on how fast the deep ocean warms up. If the deep ocean warms faster, the surface will warm slower. But don't expect that warming in the deep ocean is going to be a panacea, it's more likely to cause big problems and than solve them.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
This gets back to my original question, which is how much data would be required for the observed change in the trend to become statistically significant in your opinion. If observable temperature were flat for another 10 years, would that be significant? another 30? It seems to me that at some point you have to be able to reject the hypothesis that global surface temperature is growing linearly, even in the absence of a causal explanation.
The answer to that question depends on what the data actually looks like in the future. It matters both how much year to year variability there is and how much the long term trend changes. Playing with the annual temperatures for the last 30 years, it seems that you typically need at least 12-15 years before the trend in the data is statistically significant at the 95% level. So I'd guess that you'd need at least 20 years to determine that the trend had changed. That's only a guess and the answer really depends on how much the trend changes and what other things happen over that period. For example, if we have a major volcanic eruption like Mt Pinatubo, we'd expect global temperatures to drop about half a degree for a couple years, regardless of what's going on with global warming.
Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
Destineer
Member
Member # 821

 - posted      Profile for Destineer           Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Dan_Frank:
Nuclear power has minimal carbon footprint and generates vastly more power than solar/wind/etc. And yet most green groups demonize it and try to stop nuclear advancement as much as possible. If global warming is so catastrophic that it's worth jeopardizing 90% of the country's energy sources, why wouldn't they embrace a powerful technology like nuclear as a viable replacement?

This speaks to a larger issue with the green movement that I have, which is that all of its proposed solutions are very backwards. Reducing energy usage, giving up the most efficient non-nuclear power sources, etc. I would much rather see serious effort being made to create new technology to counteract the carbon that is so dangerous, like designing a better plant. And yet that sort of thing isn't mainstream research right now.

Dan, I agree with you that nuclear is (a large part of) the way to go, but after a number of disasters and near-disasters, including the very recent one in Japan, do you really find it so hard to understand why people would oppose nuclear energy?
Posts: 4503 | Registered: Mar 2000  |  IP: Logged | Report this post to a Moderator
SenojRetep
Member
Member # 8614

 - posted      Profile for SenojRetep   Email SenojRetep         Edit/Delete Post   Reply With Quote 
Thanks, Rabbit. I still disagree with you on the data analysis, but I appreciate your perspective.

FWIW, one of the participants in my collective intelligence experience, after I revealed what the time series was forwarded me a Ezra Klein blog post that links to another blog post that links to this paper. After reading it twice I'm still not confident of everything they're doing, but essentially they argue that by including exogenous variables related to El Nino events (MEI), volcanic events (AOD) and solar events (TSI), by averaging across the five pseudo-global temperature datasets they see mean temperature growth that is very consistent with the linear growth model. It seems to me that the biggest effect they introduce is accounting for TSI, which appears from their Figure 7 to have artificially inflated observations in the early part of the 2000s and artificially deflated temperatures in the later 2000s, for a total swing of about .1C. If the observations incorporate that compensatory factor, the change I've been arguing for (about .075C) becomes .175C, which is below expectation but not surprisingly so.

I don't know how exactly how the impacts shown in Figure 7 were calculated, but if they're accurate I would find that a persuasive reason why the observed temperature anomaly has not increased significantly over the last decade. The apparent cyclicality of the TSI also suggests that the next five-ten years should demonstrate a temperature increase much larger than the .017C/year estimated under their linear regression.

That said, because I don't know how these adjustments were arrived at I remain, to some degree, skeptical of them. I'm particularly troubled by the fact that they allowed a large amount of potential lagging and then chose lags independently for each dataset to optimize fit to the data. I also think their inclusion of second-order Fourier effects to deal with residual intra-year cycles is strange, since they appear to be aggregating to the year level anyway, so any residual cyclicality would manifest as a DC offset, which is already compensated for by their fusion algorithm. Which is just to say, there are a few things in their data analysis that I find squirrelly, which prevents me from whole-heartedly accepting their results.

Posts: 2831 | Registered: Sep 2005  |  IP: Logged | Report this post to a Moderator
King of Men
Member
Member # 6684

 - posted      Profile for King of Men   Email King of Men         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by The Rabbit:
quote:
Originally posted by King of Men:
This may be a stupid question. You mentioned that a strong El Nino had made 1998 a very hot year. What is the mechanism for this? Where does the additional energy come from?

The additional energy comes from the deep oceans, sort of.
I restate in my own words to check understanding: There is usually a heat sink in the deep oceans; under a strong El Nino the transfer to that heat sink is not as fast. Thus the 'extra' energy was there all along, but usually it doesn't pile up at the surface.
Posts: 10589 | Registered: Jul 2004  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Thanks, Rabbit. I still disagree with you on the data analysis, but I appreciate your perspective.
There isn't one right way to analyze the data. The best way to analyze the data depends on what you are trying to learn from the data. You are right that the trend over the last decade is a lot flatter than the trend over the previous decade. I'm right that this difference is not statistically significant. Which is more important depends on what you are trying to learn from the data.

If you are a scientist studying the climate, the difference in the temperature variability over the last decade is important to understand. There are a lot of things that influence the short term temperature cycles that aren't understand. They aren't "error", they are caused by something. Even though we don't have enough data to know whether the long term trend has actually changed, scientists should be concerned about finding a way to understand what's happening better.

But if what you are trying to learn from the data is whether or not global climate change is a problem that is likely to get worse in the future, you have to analyze the data in a different way. The people outside the scientific community who are talking about this data aren't doing it because they are trying to figure out how to explain the weather patterns better. They are using the data to justify inaction . They are using the data to argue that its premature to do anything about climate change. They are using it to argue that the models are bad and shouldn't be trusted. They are using it to argue that climate science is junk science. The data simply does not support those conclusions. I chose to analyze the data this way because I was asked to why the data wasn't reason to seriously doubt the models being used.

Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
The Rabbit
Member
Member # 671

 - posted      Profile for The Rabbit   Email The Rabbit         Edit/Delete Post   Reply With Quote 
quote:
Originally posted by King of Men:
quote:
Originally posted by The Rabbit:
quote:
Originally posted by King of Men:
This may be a stupid question. You mentioned that a strong El Nino had made 1998 a very hot year. What is the mechanism for this? Where does the additional energy come from?

The additional energy comes from the deep oceans, sort of.
I restate in my own words to check understanding: There is usually a heat sink in the deep oceans; under a strong El Nino the transfer to that heat sink is not as fast. Thus the 'extra' energy was there all along, but usually it doesn't pile up at the surface.
Pretty much. The oceans act as a heat sink during a La Nina and a heat source during an El Nino. The deep oceans aren't a true heat sink because a true heat sink has an infinite capacity to take up energy without changing. The deep oceans don't. If we had many decades of El Nino conditions, the temperature in the deep oceans would drop. If we had many decades of La Nina conditions the deep ocean temperatures would rise. Because we cycle fairly rapidly between the two, the deep ocean temperatures stay fairly constant.
Posts: 12590 | Registered: Jan 2000  |  IP: Logged | Report this post to a Moderator
natural_mystic
Member
Member # 11760

 - posted      Profile for natural_mystic           Edit/Delete Post   Reply With Quote 
Here's a response to the WSJ piece by one of the people whose work was cited in said piece.

http://www.nybooks.com/articles/archives/2012/mar/22/why-global-warming-skeptics-are-wrong/?pagination=false

Posts: 644 | Registered: Sep 2008  |  IP: Logged | Report this post to a Moderator
rivka
Member
Member # 4859

 - posted      Profile for rivka   Email rivka         Edit/Delete Post   Reply With Quote 
Thanks for sharing that. Very thorough article.
Posts: 32919 | Registered: Mar 2003  |  IP: Logged | Report this post to a Moderator
  This topic comprises 2 pages: 1  2   

Quick Reply
Message:

HTML is not enabled.
UBB Code is enabled.
UBB Code™ Images not permitted.
Instant Graemlins
   


Post New Topic  Post A Reply Close Topic   Feature Topic   Move Topic   Delete Topic next oldest topic   next newest topic
 - Printer-friendly view of this topic
Hop To:


Contact Us | Hatrack River Home Page

Copyright © 2008 Hatrack River Enterprises Inc. All rights reserved.
Reproduction in whole or in part without permission is prohibited.


Powered by Infopop Corporation
UBB.classic™ 6.7.2