Doug Mitchell
EF2
Add a zero or two to that.
Add a zero or two to that.
Robert Dewey
EF5
Same here. As far as I'm concerned, certain aspects of GW are still just a theory. I certainly wouldn't gamble any amount of money on the RUC 3 hour forecast, let alone a forecast looking out to 175,000 hours.
I may not have all of the meteorolical numerical facts, but there are so many proposed theories regarding GW that it would be almost impossible for the model to take them all into account and spit out results for each one.
Kevin Scharfenberg
Rdewey- do you even understand the meaning of the word "theory" as it relates to science? I get a huge kick whenever anyone downplays climate change research (or evolution or gravity) because it's "just a theory"!
GCMs aren't designed to "initialize" the current state of the climate in the same sense a 12Z weather model initializes off the 12Z observations. I can't speak for Trenberth, but I think that's what he's trying to say. If I kicked one off today and had it start on 1 January 1998 it might not capture the significant El Nino going on at that time, but might more resemble the climate state in 1994 or 2001, whereas a strong El Nino comes along in the model in 2000 instead. It might be 2 years off, but in the long-run averages these things are supposed to simulate (order 20 year+ spans), the timing errors average out. And the trend in the data is what's most important, not the exact numbers, anyway.
Linearity is used out of simplicity of very complex processes (so that this sort of thing can be done at all). However these assumptions did an incredibly good job of simulating the anthropogenically-forced upswing in temperatures over the long term average from 1860-2000 as shown in the IPCC report. That is to say, it's an important thing to know (that people often forget) when interpreting the results but the assumption has a good track record when looking in hindcast mode. And another thing, what makes you assume the linearity assumption biases the results on the "warm" side? Wouldn't there be an equal chance the bias would be "cold"? Or perhaps the bias is random. Thought of that?
So let me ask you a question. Do you think mesoscale modeling is also "faith" and not science? If not, show me one single case where a synoptic model, from which the mesoscale models are initialized, managed to initialize the mesoscale state of the atmosphere completely correctly. Just one. If that's the case, how on earth do mesoscale models work when the larger initialization model doesn't match reality on the mesoscale and there are insufficient observations available on that scale?
I'm sure the mesoscale forecasters at "First Church of SPC" are anxious for your answer.
GCMs aren't designed to "initialize" the current state of the climate in the same sense a 12Z weather model initializes off the 12Z observations. I can't speak for Trenberth, but I think that's what he's trying to say. If I kicked one off today and had it start on 1 January 1998 it might not capture the significant El Nino going on at that time, but might more resemble the climate state in 1994 or 2001, whereas a strong El Nino comes along in the model in 2000 instead. It might be 2 years off, but in the long-run averages these things are supposed to simulate (order 20 year+ spans), the timing errors average out. And the trend in the data is what's most important, not the exact numbers, anyway.
Linearity is used out of simplicity of very complex processes (so that this sort of thing can be done at all). However these assumptions did an incredibly good job of simulating the anthropogenically-forced upswing in temperatures over the long term average from 1860-2000 as shown in the IPCC report. That is to say, it's an important thing to know (that people often forget) when interpreting the results but the assumption has a good track record when looking in hindcast mode. And another thing, what makes you assume the linearity assumption biases the results on the "warm" side? Wouldn't there be an equal chance the bias would be "cold"? Or perhaps the bias is random. Thought of that?
So let me ask you a question. Do you think mesoscale modeling is also "faith" and not science? If not, show me one single case where a synoptic model, from which the mesoscale models are initialized, managed to initialize the mesoscale state of the atmosphere completely correctly. Just one. If that's the case, how on earth do mesoscale models work when the larger initialization model doesn't match reality on the mesoscale and there are insufficient observations available on that scale?
I'm sure the mesoscale forecasters at "First Church of SPC" are anxious for your answer.
Mike Smith
1. The meso models are run in forecast mode and we have a track record as to their ability to forecast the future. There is no such record of the GCM's and the GW crowd does not seem to want to run them in verifiable time ranges (1-2 years) because everyone realizes they will likely bust horribly. In 100 years, no one currently old enough to read their research will be around to verify their forecasts.
2. The RUC bombed on the Greensburg Tornado -- big time. So, no, I don't believe every mesoscale forecast. And, I would never bet $100,000,000,000 on the RUC or any other model.
3. The RUC's 3-hour CAPE forecast for Greensburg from 00Z (the tornado hit Greensburg 2 hours 55min. later) should be a cautionary tale for anyone who thinks CAPE can be accurately discerned at 25 years by a GCM.
I do give the the RUC's creators credit for putting it out there so other scientists can verify the results for themselves. That is more than we can say for some others: Carter (2007) examined evidence on the predictive validity of the general circulation models (GCMs) used by the IPCC scientists. He found that while the models included some basic principles of physics, scientists had to make “educated guesses†about the values of many parameters because knowledge about the physical processes of the earth’s climate is incomplete. In practice, the GCMs failed to predict recent global average temperatures as accurately as simple curve-fitting approaches (Carter 2007, pp. 64 – 65) and also forecast greater warming at higher altitudes when the opposite has been the case (p. 64). Further, individual GCMs produce widely different forecasts from the same initial conditions and minor changes in parameters can result in forecasts of global cooling (Essex and McKitrick, 2002). Interestingly, modeling results that project global cooling are often rejected as “outliers†or “obviously wrong†(e.g., Stainforth et al., 2005)
When a scientist puts the results out for all to see, he/she practicing science.
When a scientist trucates the results because a GCM forecast of cooling "couldn't" be correct, he/she is saying "trust me." Faith.
Kevin Scharfenberg
I've stated numerous times how the GCMs have been verified in hindcast mode (runs starting from ~1860, as in the IPCC reports), yet you don't respond. Instead you just keep stating misinformation that they "have no track record". I also posted Hansen's 1988 simulation which has been proven remarkably correct on the timescale they are intended to be verified, 10+ years. Since you refuse to understand or acknowledge the 1-2 year time scale in variations in climate are NOT what are being modeled by GCMs, but the multi-decadal trends, there is just no point trying to discuss it with you further. You are being dogmatic and unreponsive to argument. You're welcome to try to shoot down the entire method in the literature however (good luck with that!)
The RUC does not predict tornadoes. That said, since it busted "big time", shouldn't we just stop using it based on your reasoning? Of course the "GW crowd" doesn't rely on a single run of a single model, and it's dishonest to imply they do, but a series of model runs with different starting conditions using multiple models to determine a range of possible climate scenarios. I'm getting tired of typing that out over and over without response, too.
Yet again, I find myself repeating points you refuse to respond to. The models do not attempt to correctly forecast the value of CAPE at a given specific date and time, but long term trends.
Carter was referring to a paper by Loehle, but it turns out their "simple curve-fitting approach" was fatally flawed because it violated basic principles of the Nyquist-Shannon sampling theorem. (Comments on “Climate change: detection and attribution of trends from long-term geologic data†by C. Loehle [Ecological Modelling 171 (4) (2004), 433–450]). Debunked...soooo sorry!
Carter used the wrong data when claiming that the troposphere is not warming. In both his 2007 paper and his congressional testimony, he presented a plot of Christy & Spencer’s MSU analysis for channel 2, the so-called “Mid-Mroposphere†product (called TMT). It’s long been clear and documented in the literature that this product includes some of the well known cooling trend from the stratosphere, thus the TMT has an incorrect cooling trend. The C & S “Lower Troposphere†analysis introduced in 1992 (called TLT) was intended to address this problem. So this notion has been debunked for 15 years (another swing and miss).
And, when the vast majority show warming...?
Without actually seeing which models are being thrown out and why, I cannot comment (it's perfectly legitimate to throw something out if some problem can be documented), but given Carter's history (shown above) of badly misrepresenting and misinterpreting data (repeatedly, even after debunking), I have trouble trusting anything he has to say at this point.
Doug Mitchell
EF2
I don't know anything about ignoring modeling data but you can bet your sweet cheeks the IPCC report routinely ignored scientific papers that would've conflicted with their pre-drawn conclusion.
The best part is Steve Bloom's rebuttal in the comments which basically states "Pointing out uncertainties in our science only delays the politicians from taking necessary action!"
Sound familiar? (Criticizing the war in Iraq only emboldens the terrorists!!!!)
The best part is Steve Bloom's rebuttal in the comments which basically states "Pointing out uncertainties in our science only delays the politicians from taking necessary action!"
Sound familiar? (Criticizing the war in Iraq only emboldens the terrorists!!!!)
Patrick Marsh
EF5
- Joined
- Apr 18, 2006
- Messages
- 783
I'm just curious...
Mike and Kevin, what are your positions on Global Warming?
Mike and Kevin, what are your positions on Global Warming?
Kevin Scharfenberg
Patrick,
I have no reason to doubt the AMS, NOAA, IPCC, and the vast majority of scientists who study the issue that there is overwhelming evidence that the Earth is in an anamolous period of warming. While there is certainly natural forcings that may explain some of the warming, the vast majority of it cannot be explained without considering anthropogenic CO2. I think there is no reason to believe this warming will not continue with the ongoing anthropogenic increases in CO2. I also think the changes to climate are already very evident, with considerable changes in Arctic ice cover and the vast majority of glaciars retreading worldwide, among many other signals.
I believe this is a good time for society to decide how to mitigate against the societal impacts that are likely to result. Money spent moving toward clean energy sources would not only help in lowering anthropogenic CO2, but help pollution in general AND will help us get away from foreign oil that causes all sorts of upheaval. I think it's money well spent even without the greenhouse argument. Whereas ignoring the problem will likely cause greater upheaval (and associated monetary problems) with mass migration, poverty and famine due to drought, etc.
The pre-drawn conclusion bit reveals your own bias, but I can't blame the IPCC for rejecting some of the pseudo-science that's out there.
Here we go with the blog comments again. Steve Bloom clearly doesn't notice the blaringly obvious: The IPCC reports due quantitatively and qualitatively outline the uncertainty involved.
I have no reason to doubt the AMS, NOAA, IPCC, and the vast majority of scientists who study the issue that there is overwhelming evidence that the Earth is in an anamolous period of warming. While there is certainly natural forcings that may explain some of the warming, the vast majority of it cannot be explained without considering anthropogenic CO2. I think there is no reason to believe this warming will not continue with the ongoing anthropogenic increases in CO2. I also think the changes to climate are already very evident, with considerable changes in Arctic ice cover and the vast majority of glaciars retreading worldwide, among many other signals.
I believe this is a good time for society to decide how to mitigate against the societal impacts that are likely to result. Money spent moving toward clean energy sources would not only help in lowering anthropogenic CO2, but help pollution in general AND will help us get away from foreign oil that causes all sorts of upheaval. I think it's money well spent even without the greenhouse argument. Whereas ignoring the problem will likely cause greater upheaval (and associated monetary problems) with mass migration, poverty and famine due to drought, etc.
The pre-drawn conclusion bit reveals your own bias, but I can't blame the IPCC for rejecting some of the pseudo-science that's out there.
Here we go with the blog comments again. Steve Bloom clearly doesn't notice the blaringly obvious: The IPCC reports due quantitatively and qualitatively outline the uncertainty involved.
Mike Smith
Kevin,
It should be perfectly obvious that I reject the technique of extrapolating hindcasts to the future.
Let me suggest you re-read some of your own posts: Linearity is used out of simplicity of very complex processes (so that this sort of thing can be done at all). But the sun-ocean-atmosphere system is extremely complex. Until GCM's are able to show some skill (averaged temperature and precipitation) on a regional scale at shorter time periods, there is no reason to believe they have skill at longer time periods.
As I mentioned in a previous post, "model the past then apply to future" financial models (which are forecasting a far simpler system than sun-ocean-atmosphere) have yet to show skill in forecast mode. Last November 16th, this "mathematical" forecast was published: www.wallstreetwindow.com/drupal/node/1125 . It forecast the Dow was going to reach 16,000 this year. Unless something very surprising happens, another financial model will have bitten the dust -- big time. I can show literally dozens and dozens of these. Because these can be easily verified they demonstrate, to me, that the technique of "pastcast verification" is fatally flawed.
Think about this: If you were in your 60's and walked into the hospital with chest pains, would you want the emergency room to measure your blood pressure, your CBC (chemical blood count), your pulse and your blood oxygen and apply them to your prognosis? Would you then want them to take similar readings at hourly intervals after they apply treatment to determine whether you are, indeed, responding properly?
Or, would you rather have them treat you based on a model derived from studying heart attacks in 60-year olds over the past 20 years without learning your "initial conditions"? I suspect I know the answer.
You might wish to take a look at this: www.forecastingprinciples.com/Public_Policy/WarmAudit31.pdf
as well as this...
http://climatesci.colorado.edu/2006...ate-science-to-canadian-policymakers-part-ii/
which includes the following passage:
“So people who make the statement that we can’t predict the weather even, let’s say, ten days in advance, so how can we possibly predict the climate a century in advance are talking about apples and oranges.” [testimony by Dr. Ian Rutherford]
This is a clear misrepresentation of weather and climate modeling. Climate models include weather processes as a subset of the model. Even in the context of claiming that “Climate is the statistics of weather where you do a lot of averaging over time”, Dr. Rutherford is not correct. When we talk about the weather today, we still use statistics such as the daily average temperature. With multi-decadal mean temperatures, we are just referring to an different (longer) statistical averaging time.
Moreover, to characterize climate as a boundary value problem ignores peer reviewed papers which illustrate that climate predition is very much an initial value problem. Just one example is
Claussen, M., C. Kubatzki, V. Brovkin, A. Ganopolski, P. Hoelzmann, H.-J. Pachur, Simulation of an abrupt change in Saharan vegetation in the mid-Holocene, Geophys. Res. Lett., 26(14), 2037-2040, 10.1029/1999GL900494, 1999.
Further examples are discussed in
Rial, J., R.A. Pielke Sr., M. Beniston, M. Claussen, J. Canadell, P. Cox, H. Held, N. de Noblet-Ducoudre, R. Prinn, J. Reynolds, and J.D. Salas, 2004: Nonlinearities, feedbacks and critical thresholds within the Earth’s climate system. Climatic Change, 65, 11-38.
and
Pielke, R.A., 1998: Climate prediction as an initial value problem. Bull. Amer. Meteor. Soc., 79, 2743-2746.
Meteorologists that came before you and I started with barotropic models then baroclinic models, etc., always verifying against the real atmosphere. As the graphics from the Climate Science article show, there has been gradual improvement in those forecasts. These are measured results. Measuring the results of forecasts in a standardized manner is science. One researcher is then able to build on the next and make a more accurate model and the process repeats.
We don't know which, if any, of the CGM's are any good because this type of systematic evalution is not made as both the forecasting principles paper and climate science article demonstrate. The model developers just have "faith" that their tweaks produce positive results.
I was exposed, during the 1970's, to some modellers who were (true story) shocked the verification statistics on the initial LFM II were worse than the LFM I. When presented with actual verification stats, their reply was the verification couldn't be correct because the LFM II's "physics are better."! We keep tweaking the physics of the GCM's but we don't know if we are making them more or less skillful because there are no forecasts to validate them against. There is none of the "building block" scientific process that has made numerical modeling for weather forecasting so successful.
I like the scientific process a whole lot. There may be a time when GCMs can make skillful, validated forecasts after meteorologically realistic initialization (as Trenberth calls for, read his whole piece). At that time, I become intensely interested in the results 1, 10 and 20 years in the future.
It should be perfectly obvious that I reject the technique of extrapolating hindcasts to the future.
Let me suggest you re-read some of your own posts: Linearity is used out of simplicity of very complex processes (so that this sort of thing can be done at all). But the sun-ocean-atmosphere system is extremely complex. Until GCM's are able to show some skill (averaged temperature and precipitation) on a regional scale at shorter time periods, there is no reason to believe they have skill at longer time periods.
As I mentioned in a previous post, "model the past then apply to future" financial models (which are forecasting a far simpler system than sun-ocean-atmosphere) have yet to show skill in forecast mode. Last November 16th, this "mathematical" forecast was published: www.wallstreetwindow.com/drupal/node/1125 . It forecast the Dow was going to reach 16,000 this year. Unless something very surprising happens, another financial model will have bitten the dust -- big time. I can show literally dozens and dozens of these. Because these can be easily verified they demonstrate, to me, that the technique of "pastcast verification" is fatally flawed.
Think about this: If you were in your 60's and walked into the hospital with chest pains, would you want the emergency room to measure your blood pressure, your CBC (chemical blood count), your pulse and your blood oxygen and apply them to your prognosis? Would you then want them to take similar readings at hourly intervals after they apply treatment to determine whether you are, indeed, responding properly?
Or, would you rather have them treat you based on a model derived from studying heart attacks in 60-year olds over the past 20 years without learning your "initial conditions"? I suspect I know the answer.
You might wish to take a look at this: www.forecastingprinciples.com/Public_Policy/WarmAudit31.pdf
as well as this...
http://climatesci.colorado.edu/2006...ate-science-to-canadian-policymakers-part-ii/
which includes the following passage:
“So people who make the statement that we can’t predict the weather even, let’s say, ten days in advance, so how can we possibly predict the climate a century in advance are talking about apples and oranges.” [testimony by Dr. Ian Rutherford]
This is a clear misrepresentation of weather and climate modeling. Climate models include weather processes as a subset of the model. Even in the context of claiming that “Climate is the statistics of weather where you do a lot of averaging over time”, Dr. Rutherford is not correct. When we talk about the weather today, we still use statistics such as the daily average temperature. With multi-decadal mean temperatures, we are just referring to an different (longer) statistical averaging time.
Moreover, to characterize climate as a boundary value problem ignores peer reviewed papers which illustrate that climate predition is very much an initial value problem. Just one example is
Claussen, M., C. Kubatzki, V. Brovkin, A. Ganopolski, P. Hoelzmann, H.-J. Pachur, Simulation of an abrupt change in Saharan vegetation in the mid-Holocene, Geophys. Res. Lett., 26(14), 2037-2040, 10.1029/1999GL900494, 1999.
Further examples are discussed in
Rial, J., R.A. Pielke Sr., M. Beniston, M. Claussen, J. Canadell, P. Cox, H. Held, N. de Noblet-Ducoudre, R. Prinn, J. Reynolds, and J.D. Salas, 2004: Nonlinearities, feedbacks and critical thresholds within the Earth’s climate system. Climatic Change, 65, 11-38.
and
Pielke, R.A., 1998: Climate prediction as an initial value problem. Bull. Amer. Meteor. Soc., 79, 2743-2746.
Meteorologists that came before you and I started with barotropic models then baroclinic models, etc., always verifying against the real atmosphere. As the graphics from the Climate Science article show, there has been gradual improvement in those forecasts. These are measured results. Measuring the results of forecasts in a standardized manner is science. One researcher is then able to build on the next and make a more accurate model and the process repeats.
We don't know which, if any, of the CGM's are any good because this type of systematic evalution is not made as both the forecasting principles paper and climate science article demonstrate. The model developers just have "faith" that their tweaks produce positive results.
I was exposed, during the 1970's, to some modellers who were (true story) shocked the verification statistics on the initial LFM II were worse than the LFM I. When presented with actual verification stats, their reply was the verification couldn't be correct because the LFM II's "physics are better."! We keep tweaking the physics of the GCM's but we don't know if we are making them more or less skillful because there are no forecasts to validate them against. There is none of the "building block" scientific process that has made numerical modeling for weather forecasting so successful.
I like the scientific process a whole lot. There may be a time when GCMs can make skillful, validated forecasts after meteorologically realistic initialization (as Trenberth calls for, read his whole piece). At that time, I become intensely interested in the results 1, 10 and 20 years in the future.
Last edited by a moderator:
Doug Mitchell
EF2
So you're going to tell me that every single paper listed here and here in addition to my first link is pseudo-science? Jesus Christ, would you listen to yourself? You're claiming all that research is pseudo-science and I'm the one with the bias?
And for Steve Bloom's blog comment, it epitomizes the entire the entire attitude towards "deniers" or "skeptics". Even if our concerns are legit and valid, we're to be ostracized and in this case, we're TO BLAME for the perceived consequences of AGW.
Mike Smith
I believe the earth is clearly warmer than it was in 1975.
We don't know how much because there are demonstrated problems in the various measuring networks and the IPCC's (Hansen's) failure to release his computer code so that other scientists can attempt to replicate his results.
Papers have been published that indicate the combination of solar plus cosmic rays (decreased cloud cover) based on experimentation may account for the warming so far observed. There are papers that seem to support that hypothesis and others that seem to reject it.
GW is being observed on Mars, Triton and Pluto.
Based on the three items above, while we know the earth is warming we don't know how much and since we don't, it is extremely difficult to validate the "CO2 as main culprit" hypothesis.
There is anecdotal evidence for warming due to diminishing ice at the North Pole (which is not occurring in Antarctica) and some glaciers receeding. The former could be explained by a darkening of the snow cover due to increased pollution without an increase in temperature (change in albedo). In the case of glaciers, the TINY number actually studied may or may not be representative of the earth's glaciers as a whole.
We just don't know enough to even speculate intelligently about regional drought, hurricane strength, etc.
Hope that answers your question as to my overall thoughts on this subject.
Thanks for asking.
Mike
We don't know how much because there are demonstrated problems in the various measuring networks and the IPCC's (Hansen's) failure to release his computer code so that other scientists can attempt to replicate his results.
Papers have been published that indicate the combination of solar plus cosmic rays (decreased cloud cover) based on experimentation may account for the warming so far observed. There are papers that seem to support that hypothesis and others that seem to reject it.
GW is being observed on Mars, Triton and Pluto.
Based on the three items above, while we know the earth is warming we don't know how much and since we don't, it is extremely difficult to validate the "CO2 as main culprit" hypothesis.
There is anecdotal evidence for warming due to diminishing ice at the North Pole (which is not occurring in Antarctica) and some glaciers receeding. The former could be explained by a darkening of the snow cover due to increased pollution without an increase in temperature (change in albedo). In the case of glaciers, the TINY number actually studied may or may not be representative of the earth's glaciers as a whole.
We just don't know enough to even speculate intelligently about regional drought, hurricane strength, etc.
Hope that answers your question as to my overall thoughts on this subject.
Thanks for asking.
Mike
Last edited by a moderator:
Kevin Scharfenberg
I have to admit with your snide remark about the IPCC "pre-drawn conclusion" tipping off your own prejudice, I didn't even bother going to the link. Now that I do, it looks like the blogger is complaining that not every related paper made it into a certain IPCC report chapter. Without actually going through each paper, I can tell you there is such a large volume in the literature now that the IPCC chapters, which are overview in nature, can't possibly include every last reference out there, even the really good ones. But there are many reasonable, non-conspiratorial ways an overview document might not cite a related work:
a) It's covered in another chapter/section
b) It is out of date
c) It has been debunked
d) It was covered in a previous IPCC report
e) It's redundant to other works cited
f) It's too new an idea to be thoroughly vetted
...and so on.
I simply don't have the time (or inclination) to go through every work cited on the blog and try to figure out why it did or did not get cited. But generally, given how these "conspiracy theories" always turn out in the end, this too will turn out to be much ado about nothing.
