Surface Temperatures Trends Through May:
Month 89 and counting!
Month 89 and counting!
Trends for the Global Means Surface temperature for five groups (GISS, HadCrut, NOAA/NCDC, UAH/MSU and RSS.) were calculated from Jan 2001-May 2008 using Ordinary Least Squares (OLS) using the method in Lee & Lund. to compute error bars, and Cochrane-Orcutt and compared to the IPCC AR4’s projected central tendency of 2C/century for the trend during the first few decades of this century.
The following results for mean trends and 95% confidence intervals were obtained:
- Ordinary Least Squares average of data sets: The temperature trend is -0.7 C/century ± 2.3C/century. This is inconsistent IPCC AR4 projection of 2C/century to a confidence of 95% and is considered falsified based on this specific test.
- Cochrane Orcutt, average of data sets: The temperature trend is -1.4 C/century ± 2.0 C/century. This is inconsistent with the IPCC AR4 projection of 2 C/century to a confidence of 95% and is considered falsified based on this specific test for an AR(1) process.
- OLS, individual data sets: All except GISS Land/Ocean result in negative trends. The maximum and minimum trends reported were 0.007 C/century and -1.28 C/century for GISS Land/Ocean and UAH MSU respectively. Based on this test, The IPCC AR4 2C/century projection is rejected to a confidence of 95% when compared to HadCrut, NOAA and RSS MSU data. It is not rejected based on comparison to GISS and UAH MSU.
- Cochrane-Orcutt, individual data sets: All individual data sets result in negative trends. The IPCC AR4 2C/century is falsified by each set individually.
- The null hypothesis of 0C/century cannot yet be excluded based on data collected since 2001. This, does not mean warming has stopped. It only means that the uncertainty in the trend is too large to exclude 0C/century based on data since 2001. Bar and Whiskers charts showing the range of trend falling inside the ±95% uncertainty intervals using selected start dates are discussed in Trends in Global Mean Surface Temperature: Bars and Whiskers Through May.
The OLS trends for the mean, and C-O trends for individual groups are compared to data in the figure immediately below:
Click for larger.Figure 1: The IPCC projected trend is illustrated in brown. The Cochrane - Orcutt trend for the average of all five data sets is illustrated in orange; ±95% confidence intervals illustrated in hazy orange. The OLS trend for the average of all five data sets is illustrated in lavender, with ±95% uncertainty bounds in hazy lavender. Individual data sets were fit using Cochrane-Orcutt, and shown.
Discussion of Figure 1
The individual weather data in figure 1 are scattered, and show non-monotonic variations as a function of time. This is expected for weather data; some bloggers like to refer to this scatter as “weather noise”. In the AR4, the IPCC projected a monontonically increasing level increase in the ensemble average of the weather, often called the climate trend. For the first 3 decades of the century, central tendency of the climate trend was projected to vary approximately linearly at a rate of 2C/century. This is illustrated in brown.
The best estimates for the linear trend consistent with the noisy weather data were computed using Cochrane-Orcutt (CO), illustrated in orange, and Ordinary Least Squares (OLS) illustrated in lavender.
Results for individual hypothesis tests
Some individual bloggers have expressed a strong preference for one particular data set or another. Like Atmoz, I prefer not to drop any widely used metric from consideration. However, because some individuals prefer to examine results for each individual group seperately, I also apply the technique to describe the current results of two hypothesis tests based on each individual measurement system.
The first hypothesis tested, treated as “null” is the IPCC’s projections of a 2C/century. Currently, this is rejected at p=95% under ordinary least squares (OLS) using data from 3 of the five services, but it is not rejected for UAH or GISS. The hypothesis is rejected against all 5 servies when tested using C-O fits.
The second hypothesis tested is the “denier’s hypothesis” of 0C/century. This hypothesis cannot be rejected using data starting in 2001. Given the strong rejection with historic data, and the large uncertainty in the determination of the trend, this “fail to reject” result is likely due to “type 2″ or “beta” error.
That is: The “fail to reject” is likely a false negative. False negatives, or failure to reject false results are the most common error when hypotheses are tested using noisy data.
Results for individual tests are tabulated below:
| Group | OLS Trend | Reject / Fail to Reject? | CO Trend | Reject / Fail to Reject? | ||
| (C/century) | 2C/century | 0 C/century | (C/century) | 2C/century | 0 C/century | |
| Average of 5 | -0.7 ± 2.3 |
Reject | Fail to reject | -1.4 ± 2.0 | Reject | Fail to reject |
| GISS | 0.0 ± 2.3 | Fail to Reject |
Fail to reject | -0.4 ± 2.0 | Reject | Fail to reject |
| HadCRUT | -1.2 ± 1.9 | Reject | Fail to reject | -1.6 ± 1.6 | Reject | Fail to reject |
| NOAA | -0.1 ± 1.7 | Reject | Fail to reject | -0.3 ± 1.5 | Reject | Fail to reject |
| RSS MSU | -1.3 ± 2.3 | Reject | Fail to reject | -2.1 ± 2.1 | Reject | Fail to reject |
| UAH MSU | -0.8 ± 3.6 | Fail to reject | Fail to reject | -2.0 ± 3.1 | Reject | Fail to reject |
The possibility of false positives
In the context of this test, rejecting a hypothesis when it is true is a false positive. All statistical test involve some assumptions, those underlying this test assume we can correct for red noise in the residuals to OLS using one of two methods: A) The method recommended in Lee&Lund or B) Cochrane-Orcutt, a well known statistical method for time series exhibiting red noise. If these methods are valid, and used to test data, we expect to incorrectly reject true hypotheses at p=95%, 5% of the time. (Note however, finding reject in February, March, April and May do not actually count separately, as the rejections themselves are correlated with each other, being largely based on the same data.)
Given the results we have found, the 2C/century projection for the first few decades of this century is not born out by the current data for weather. It appears inconsistent with underlying trends that could possibly describe the particular weather trajectory we have seen.
There are some caveats that have been raised in the blog-o-sphere. There has been some debate over methods to calculate uncertainty intervals and/or whether one can test hypotheses using short data sets. I have been examining a variety of possible reasons. I find:
- In August, 2007, in a post entitled “Garbage Is Forever”, Tamino used and defended the uses OLS adjusted for red noise to perform hypothesis tests using short data sets, going into some detail in the response to criticism by Carrick, where Tamino stated:
For a reasonable perspective on the application of linear regression in the presence of autocorrelated noise see Lee & Lund 2004, Biometrika 91, 240–245. Your claims that it’s “pretty crazy, from a statistics perspective” and “L2 is only reliable, when the unfit variability in the data looks like Gaussian white noise” raises serious doubts about your statistical sophistication.
Later posts, when this method began falsifying the IPCC AR4 projection of 2 C/century, Tamino appears to have changed his mind about the validity of this method possibly suggesting the uncertainty intervals are too high.
The results here simply show what anyone would obtain using this method: According to this method, the 2C/century is falsified. Meanwhile, re-application to the data since 2000 indicates there is no significant warming since 2000 as illustrated here.
- Gavin Schmidt suggested that “internal variability (weather!)” noise results in a standard error of 2.1C/century in 8 year trends; this is roughly twice the standard error obtained using the method of Lee & Lund, above. To determine if this magnitude of variability made any sense at all, I calculated the variability of 8 year trend in the full thermometer record including volcano eruptions, and measurement noise due to the “bucket-jet inlet transition. I also computed the variability during a relatively long historic period with no volcanic eruptions. A standard error of 2.1 C/century suggested by Gavin’s method exceeded both the variablity in the thermometer record for real earth including volcanic periods and that for periods without volcanic eruptions. (The standard error in 8 year trends computed during periods with no volcanic eruptions is approximately 0.9C/century, which is smaller than estimated for the current data).
I attribute the unphysically large spread in 8 year trends displayed by the climate models to the fact that the model runs include
a) different historical forcings, some including volcanic eruptions, some don’t. This results in variability in initial conditions across model runs that do not exist on the real earth
b) different forcings during any year in the 20th century; some include solar some don’t.
c) different parameterizations across models and
d) possibly, inability of some individual models to reproduce the actual characteristics of real-earth weather noise.This is discussed Distribution of 8 Year OLS Trends: What do the data say?
- Atmoz have suggested the flat trend is either to ENSO and JohnV suggested considering the effect of Solar Cycle. The issue of ENSO and remaining correlation in lagged residuals has been discussed in previous posts and the solar cycle is explained here.
- The variability of all 8 month trends that can be computed in the thermometer record is 1.9 C/century; computing starting with a set spected at 100 month intervals resulted in a standard error of 1.4 C/century. These represent the upper bound of standard errors that can be justified based on the empirical record. Variabiity includes features other than “weather noise”– for example, volcano eruptions, non-linear variations in forcing due to GHG’s, and measurement uncertainty, including the “jet transition to bucket inlet” noise. So, these represent the upper limit on variability in experimentally determined 8 year trends.
Those who adhere to these will conclude the current trends fall inside the uncertainty intervals for data. If the current measurement uncertainty is as large as experienced during the “bucket to jet inlet transition” associated with World War II, they are entirely correct.
After consideration of the various suggestions about uncertainty intervals, and the issues ENSO, solar cycles and other features, and considering the magnitude of the pre-existing trend I think over all the data indicate:
- It is quite likely the IPCC projection for an underlying climate trend of 2C/century exceeds the current underlying trend.
I cannot speculate on the reasons for the over estimate; they may include some combination of poor forecast of emissions when developing the SRES, to the effect of inaccurate initial conditions for the computations of the 20th century, to inaccuracy in GCMs themselves or other factors.
- It remains likely the warming experienced over the past century will resume.
While the 2C/century projection falsifies using both OLS and C-O, the flat trend is entirely consistent with the previously experienced warming trend. In fact, additional analysis (which I have not shown) would indicate the current trend is not inconsistent with the rate of warming seen during the late 90s. It is entirely possible natural factors, including volcanic eruptions depressing the temperature during the early 90s, caused a positive excursion in the 10 year trend during that period. Meanwhile, the PDO flip can be causing a negative excursion affecting the trend. These sorts of excursions from the mean trend are entrely consistent with historic data.
Warming remains consistent with the data. As the theory attributing the warming to GHG’s appears sound and predates the warming from the 80s and 90s, I confident it will resume.
What will happen during over the next few years
As Atmoz warns, we should expect to see the central tendency of trends move around over the next few years. What one might expect is that, going forward, we will see the trend slowly oscillate about the mean, but eventually the magnitude of the oscillation will decay.
One of my motives in blogging this is to show this oscillation and decay over time and to permit doubters to see the positive trend resumes.
I will now set off on the sort of rampant speculations permitted bloggers. When the next El Nino arrives, we will see a period where the trends go positive. Given trends from the 70s through 90s, and current trends, it seem plausible to me that, using the methods I describe here that that we will experience some 89 month trends with OLS trends of 3C/century - 4 C/century or even greater sometime during the next El Nino. At which point, someone will likely blog about that, the moment the 89 month trend occurs. 
This result will entirely consistent with the current findings. An OLS (or CO ) trend of 3-4 C/century is likely even if the true trend is less than 2C/century, and even if CO and OLS do give accurate uncertainty intervals.
What’s seems unlikely? I’d need to do more precise calculations to find a firm dividing line between consistent and inconsistent. For now, I’ll suggest that unless there is a) a stratospheric volcanic eruption, b) the much anticipated release of methane from the permafrost or c) a sudden revision in the method the agencies use to estimate GMST, I doubt we’ll see an 89 month trend greater than 4.6 C/century within the next five years. (I won’t go further because I have no idea what anyone is emitting into the atmosphere!)
Meanwhile, what is the magnitude of the trend for the first three decades of this century? That cannot be known with precision for man years. All I can say is: The current data strongly indicate the current underlying trend less than 2C/century, and likely less than 1.6 C/century!
Updates
Martin Ringo found a typo in my table. I corrected the RSS trend to match the spreadsheet.
Written by lucia.Comments
lucia
(Comment#3557)
June 24th, 2008 at 1:04 pm
Anthony– I think I’ll back some “anomaly cookies” tonight. I’m sure if I show now nice and yummy they are, I can get a blogger to bet, hoping to win some! I better go mow the lawn and then buy some ingredients. I need yellow, red and blue for trend lines and maybe some dots for the temperatures.
terry
(Comment#3558)
June 24th, 2008 at 1:16 pm
I’ll second Anthony, very, very nicely done! As I understand this: the trend as of now does not support the great heat catastrophe as reported by many, but also does not support the great cold catastrophe as speculated by other manys. At least, I hope I got it right!
lucia
(Comment#3559)
June 24th, 2008 at 1:46 pm
Terry– As for the issue of the cold catastrophe: I don’t think there is any evidence of a cold catastrophe. That said, based on what’s in this post, which is data after 2001, I couldn’t say the cold catastrophe is falsified. I need to look at data from other periods. But in the case of the cold catastrophe, that’s fair. The reason is, as far as a know, there is no specific concrete prediction of any cold catastrophe, when it starts or what it’s magnitude might be. So, in this case, it doesn’t make sense to ignore the existence of the warm trend prior to 2001. This “flat bit” is happening, but it’s not inconsistent with the previous quite distinct warming trend. So, from a purely empirical view, I would never suggest that the climate trend has changed. It much more likely the trend is the same as it ever was with high excursions during the 80s and 90s and a low one now. But, for that to be true, that trend is less than 2C/century. Maybe it’s 1.5 C/century, which they were projecting in the TAR. Or maybe it’s lower. We don’t really know and can’t know precisely based on this amount of data. And that even holds if we go back to the 90s! Even going back to the nineties, the uncertainty in the trend is still pretty large.
The data is inconsistent with the 2C/century projection of the AR4, which starts this century and is not retroactive. And, there is no earlier start date for analysis that makes sense to me. Starting in 2000 would means we are testing AR4 projections against data available before the TAR.
So, in my mind, 2C/century is inconsistent using all data available to test that projection. Period. No caveats.
Zeke
(Comment#3560)
June 24th, 2008 at 1:48 pm
Good post Lucia. The million dollar question is, of course: “I cannot speculate on the reasons for the over estimate; they may include some combination of poor forecast of emissions when developing the SRES, to the effect of inaccurate initial conditions for the computations of the 20th century, to inaccuracy in GCMs themselves or other factors.” As someone who works in the climate policy realm, my concern is that short term stagnation of temperatures that is consistent with natural forcings but not at odds with projections of long term trends could be used as a justification for continued inaction on limiting emissions. I do wish I were a tad more disinterested in the outcome of our little planetary experiment, as it certainly makes for quite the academic debate! One quick thing: your first caveat in this sentence doesn’t seem to fit: “What’s seems unlikely? I’d need to do more precise calculations to find a firm dividing line between consistent and inconsistent. For now, I’ll suggest that unless there is a) a stratospheric volcanic eruption, b) the much anticipated release of methane from the permafrost or c) a sudden revision in the method the agencies use to estimate GMST, I doubt we’ll see an 89 month trend greater than 4.6 C/century within the next five years.” Also, using the last eight years to project the underlying trend for the first three decades of this century is something of an exercise in futile speculation at this point :-p since a stratospheric volcanic eruption would certainly reduce the chance of seeing an 89 month trend greater than 4.6 C/century within the next five years.
Zeke
(Comment#3561)
June 24th, 2008 at 1:50 pm
Hmm, thats odd, the last two lines of my prior post appeared to have mysteriously switched order… It should read: since a stratospheric volcanic eruption would certainly reduce the chance of seeing an 89 month trend greater than 4.6 C/century within the next five years. Also, using the last eight years to project the underlying trend for the first three decades of this century is something of an exercise in futile speculation at this point :-p
lucia
(Comment#3564)
June 24th, 2008 at 2:51 pm
Zeke: One quick thing: your first caveat in this sentence doesn’t seem to fit: “What’s seems unlikely? I’d need to do more precise calculations to find a firm dividing line between consistent and inconsistent. For now, I’ll suggest that unless there is a) a stratospheric volcanic eruption, Zeke, the reason it seems odd related to the period. I was describing 89 monty trends, so this assumed people start at what-ever is “now” when they decide to do a fit, and go back. The way the “big up” would happen is this: After the eruption, temperature first go down. Later they go up. So, much later, if someone picks 89 months after the “minimum”, there could be a large “up” trend. That’s precisely what we see during the “volcanic” periods. The standard deviation of 8 year trends soars because you get big downtrend ending with the eruptoin snd then big up trends starting with the eruption. The guys at RC and William Connoley tried to highlight the “downs” as reasons why one would simply expect these big dips during a major run up. But, the excessive “ups” also happen. On this: As someone who works in the climate policy realm, my concern is that short term stagnation of temperatures that is consistent with natural forcings but not at odds with projections of long term trends could be used as a justification for continued inaction on limiting emissions. I do wish I were a tad more disinterested in the outcome of our little planetary experiment, as it certainly makes for quite the academic debate! I appreciate and share your concern. I have wanted a push toward nukes for a long time. I’m also disappointed that we can’t manage to get any off-shore wind generation. In comments at some blogs, there have been some who insist that my showing this must be related to a desire for inaction. The opposite is actually the case. I think the tendency of activists on what is clearly overselling models, the evidence for very high levels of warming etc. impedes our ability to find solutions and develop methods to reign in emissions. If one actually listens to many who criticize the IPCC or warming activists, quite a few would be willing to agree to reducing emissions. However, quite a few of the lukewarmers would insist on policies that take special cate to maintain industrial and economic capacity. The difficulty is these compromises almost certainly need to involve increasing nuclear energy, and really pushing for alternative energy. Really promoting alternative energy will involve doing things like building off large scale off shore wind installations, funding research into alternative energy generously. Taxing carbon and encouraging conservation alone is destined to fail. In my opinion, we are much more likely to develop adopt practical workable policies if we correctly assess the level of certainty, are even handed in assessing maximum and minimum likely levels of warming, and stop demonizing people for holding different opinions about whether the likely warming is 0 C/century or 3 C/century. I think both are unlikely, but demonizing people never works. (BTW: I’m not accusing you of doing this. I’m describing some of what I see at other blogs and in, oh, say interview of certain prominent people.)
Anthony Watts
(Comment#3565)
June 24th, 2008 at 2:52 pm
After reading this analysis, while I still have my strong doubts about GISS data preparation, I’m prepared to take an averaged approach rather than to discard it as an outlier.
A new view on GISS data, per Lucia « Watts Up With That?
(Pingback#3566)
June 24th, 2008 at 3:44 pm
[...] Here is her analysis: Surface Temperatures Trends Through May: Month 89 and counting! [...]
Tilo Reber (Comment#3572)
June 24th, 2008 at 4:57 pm
“When the next El Nino arrives, we will see a period where the trends go positive.” Completely agree. “Given trends from the 70s through 90s, and current trends, it seem plausible to me that, using the methods I describe here that that we will experience some 89 month trends with OLS trends of 3C/century - 4 C/century or even greater sometime during the next El Nino.” El Nino’s generally don’t go longer than a couple of years, and many last less that one year. A PDO could go for 89 months, but as I understand it we may recently have switched into a colder PDO cycle. If this is the case, we may continue in a relatively flat range for the next 10 to 15 years. My opinion is that the 70s to 90s temperature rises were outliers and that while we may see some similar intervals in the next 100 years, they will be the exception more than the rule. Take a look at the PDO effect over that time period. It shows an abnormally large number of El Ninos. In addition, we had three very large solar cycles. Also remember that the effect of CO2 is logarithmic and will thus have less effect as we add more. I’m willing to make a friendly wager that we do not get an 89 month period with a trend above 3C/century over the course of the next 20 years.
http://jisao.washington.edu/pdo/
lucia
(Comment#3573)
June 24th, 2008 at 5:37 pm
Tilo– So, we need a shorter wager. You’ve been blogging here right? And guest blogging at Anthony’s right? Plus, you use your, real name. I think we can probably negotiate a bet– provided you are willing to bet home made cookies and the bet has a short enough time span that we get to eat the cookies. If you’re willing to risk having to spend an hour over a hot stove and mail me cookies, we can try to think of some sort of bet.
I only bet cookies! Twenty years is too long– I could 70 before I eat cookies I might have a chance to win. Or, I might have to bake cookies when I’m 70. I could get Alzheimers and forget where the cookbooks are.
Tilo Reber (Comment#3574)
June 24th, 2008 at 6:28 pm
Lucia, I tried to get James Annan to take a 10,000 dollar bet on the decadal trend over the next 30 years starting in 2000. Above .2C per decade and he gets the money, below and I get it. He wouldn’t “bite”. Between friends, two dozen cookies would be the perfect bet. But I will have to get someone else to bake them for me if I loose. If I make them they will only be good for your dog or for mine. If we can come up with a shorter period so that we can reap the rewards of the bet without having to chew with false teeth, that would be great. My blog is actually just a place where I hang charts that I reference. It’s not intended to be an active site. Let me know. And I prefer brownies - with or without the cannabis. 
lucia
(Comment#3575)
June 24th, 2008 at 7:13 pm
Tilo— It’s interesting he wouldn’t bite. It does appear those who set out these challenges won’t take bets that are supposed to be even money according to the IPCC documents. Did you post the challenge to Annan your blog? Or did you contact him by email. I think, for the cookie bet, we should pick something meaningless. Your graph didn’t show GISS temp, and I gave you a hard time. So, I say in December when the data come out, we calculate the OLS trend based on the six months from June-Nov. for GISS only and for the three groups you used in your graph at Anthony’s blog. * If the trend for GISS comes out lower than the average of the three groups, you bake me a dozen cookies. I’ll make brownies, but no cannabis. Would walnuts do instead? Hopefully, you live in the US so postage and handling of food isn’t a consideration. Otherwise, if postage is exhorbitant, we may need to call this off agree to send cookies to some charity. (Anyway, imagine the climate impact of shipping brownies to Australian. It could make the world explode! )
* If the trend GISS trend is higher, I bake you a dozen cookies.
* In the event of an absolute tie… we send cookies to an agreed on third party.
* In the event that the agencies post major data corrections that affect data from Jan 2001-May 2008 sufficiently to change the “reject/ fail to reject” outcomes on the table in my current blog post, either one of us can weasel out. (Actually, I think in that case of these sorts of revisions, we should send cookies to Steve McIntyre, provided Canadian customs doesn’t have problems with cookies crossing the border.)
You may appoint a designated baker. I’ll bake them myself; brownies are easy.
Atmoz
(Comment#3576)
June 24th, 2008 at 8:37 pm
It’s interesting he wouldn’t bite. It does appear those who set out these challenges won’t take bets that are supposed to be even money according to the IPCC documents. How many people do you know that would risk money on a flip of a coin? A good bet is where all involved think they have the best chance of winning. In this particular case you have the IPCC projections of 2C/century which will represent Annan’s bet, but the other bettor must have a hypothesis as well. That could be that levels of CO2 have no affect of global mean surface temperature (an extremely silly supposition). In that case, your hypothesis would be 0C/century. The cutoff value on who was most right would then be the mean of the two hypotheses, or 1C/century. To consider how silly that bet would be to take, consider that over the next 10 years the OLS trend is 1.95C/century. This would be remarkably consistent with the IPCC projections, but Annan would lose. Who wants to be right, but lose the bet? If you actually want him to take the bet, make it reasonable. Perhaps your hypothesis would is that CO2 does contribute a modest amount of warming, but that there is no/few positive feedbacks. So the warming would be about 30% of IPCC projections. This is around 0.5C/century. So the bet would be made with a cutoff at 1.25C/century. 10 years is a short time to be betting over though. On those time scales, you still need to worry about noise (ENSO, volcanoes, etc). He may take the bet at 1.25C/century, but I would not be surprised if he didn’t.
pdm (Comment#3577)
June 24th, 2008 at 8:41 pm
lucia, the first : It remains likely the warming experienced over the past century will resume. The second: Meanwhile, what is the magnitude of the trend for the first three decades of this century? That cannot be known with precision for man years. All I can say is: The current data strongly indicate the current underlying trend less than 2C/century, and likely less than 1.6 C/century! There seems to be a contradiction here. I don’t see how your analysis shows that it “remains likely the warming experienced over the past century will resume.” I didn’t think your analysis was predictive in any way? Am I wrong there? BTW, I love your posts, I check them daily.
can you reconcile these two statements in your post:
While the 2C/century projection falsifies using both OLS and C-O, the flat trend is entirely consistent with the previously experienced warming trend. In fact, additional analysis (which I have not shown) would indicate the current trend is not inconsistent with the rate of warming seen during the late 90s.
lucia
(Comment#3578)
June 24th, 2008 at 8:55 pm
How many people do you know that would risk money on a flip of a coin? A good bet is where all involved think they have the best chance of winning. Sure. Or, alternatively, one takes fun bets. My main thoughts were: Annan could negotiate the start date to exclude data already in the pocket. That is entirely reasonable after all. Did he? I also agree the other better ought to have a hypothesis. Otherwise, you really can’t come up with something fair. (I’m using the other bettor having some sort of stand as a criteria for betting. Even with cookie bets.) The only reason I find Annan turning down the bet interesting is James does periodically offer bet challenges at his blog. Evidently, he’s sort of into this whole betting idea. So, I thought he might dicker over start dates of terms. I’m hoping Tilo takes me up on the brownie bet. The outcome is meaningless — but I think that’s kind of the fun of it. Plus, who can’t use brownies during the Holiday season?
lucia
(Comment#3579)
June 24th, 2008 at 9:15 pm
pdm: the first : It remains likely the warming experienced over the past century will resume. Falsifying 2C/century doesn’t mean 0C/century must be true. It simply means 2C/century is out of contention (from a probabilistic standpoint.) Since this warming has been happening, I strongly suspect it will resume. As the theory that GHG’s should cause at least some warming, I believe the warming over the whole century was caused, at least in part, but GHG’s. But of course, I realize there are some who disagree. The second contention: the data since 2001 show it’s likely less than 1.6 C/century. I picked that date before I was even familiar with that data, and I picked it based on publication dates of IPCC documents. So, even if there are people at some blogs (not you) who think I picked it for some other reason, I know I didn’t cherry pick it. So… based on the trend I found, I think the warming is likely less than 1.6 C/century. But, less than 1.6 C/century still leaves warming. If you read my conversation with Atmoz, I’d probably due cookie bets with bloggers– but the range I think is likely is between 1.0 C/century and 1.5 C/century over the long haul. Obviously, I think there is uncertainty. I’ll only bet cookies, and only about a dozen. (And then only with fun people who have blogs and who state their own positions so we can negotiate reasonably fair bets.
While the 2C/century projection falsifies using both OLS and C-O, the flat trend is entirely consistent with the previously experienced warming trend. In fact, additional analysis (which I have not shown) would indicate the current trend is not inconsistent with the rate of warming seen during the late 90s.
I think I’m going to have to show a separate t-test comparing the 1991-2001 trend to 2001-now. But, mostly, if you look at the bar and whiskers plot, below, you’ll see that there is no particular reason to suspect 1.0 C/century is “out” based on recent data. 
Ravalli County News » Blog Archive » IPCC Warming Projections Fail Statistical Tests
(Pingback#3580)
June 24th, 2008 at 10:01 pm
[...] Looking at data from January 2001 through May 2008, the IPCC projection of 2 degrees per century “is considered falsified…” [...]
Martin Ringo (Comment#3581)
June 24th, 2008 at 10:16 pm
There are a lot of “bets” being suggested these days. What is a fair bet — and I don’t mean that in an actuarial sense but in an equity sense — between a believer of the IPCC modeling and a skeptic of that same modeling? Let me suggest that it is not the 2.8 C average of the best estimates over the six scenarios in the AR4 or the 2.0 C value oft cited. I will suggest that it is the 1.4 C which is the next-to-lowest scenario low of the six scenarios. That is, throw out the lowest (B1) and take the next (A1T and B2). A believer in the modeling should be able to accept that as a criterion of belief/confidence. Of course, we will have to wait awhile, but let me ask another question? How do you tell who wins? Suppose we are using Lucia’s 5-series average (or at least my version of the same thing). Now post 2000 OLS trend estimate is -0.7 C per century, but if we used Cochrance Orcutt estimation we would get -1.4 C per century. That is a pretty big difference. But I could argue that the error term is Autoregressive Conditional Heteroskedastic and we should be using at least at AR1 with ARCH(2) estimation, which would give us a trend estimate of -1.6 C per century. Of course the problem with using more sophisticated estimates is that they require increasing levels of statistical knowledge to judge whether or not they are appropriate or simply chosen to push the estimate one way or another. Let me offer a simple way for simple two variable relationships. Take a sizable part of the beginning and end of the sample, say, one year for our 89 month sample. Take the average of the temperatures for each of the beginning and end subsamples and time variable, and then calculate the time trend as the linear slope of between the two points. Here the trend would be -1.3 C per century. (The technique is a very old-fashion, say around the late 50s or early 60s, version of robust regression, but it eliminates a good portion of the serial-correlation and heteroskedasticity problems. The original suggest was to divide the sample into three parts and connect the two ends. Using subsamples closer to the ends, improves the efficiency of the estimation although raises the risks or a relative lack of robustness.) Anyway the offer of that kind of bet with something akin to the determination of the result I suggest is in my opinion a fair challenge of the belief in the modeling. A bet on a trend anywhere below 2 C per century is — again in my opinion — asking for someone to take a sucker bet. They would lose with a trend of 1.9 C per century, but that would be, in essence, a “told you so” for the modelers’ point of view. There is uncertainty as to modeling predictions. True it is not the same thing as weather noise, but that doesn’t mean it isn’t real, at least with regard to the suggested bets.
lucia
(Comment#3582)
June 24th, 2008 at 10:27 pm
Anyway the offer of that kind of bet with something akin to the determination of the result I suggest is in my opinion a fair challenge of the belief in the modeling. A bet on a trend anywhere below 2 C per century is — again in my opinion — asking for someone to take a sucker bet. Do you mean a modeler offering a bet below 2 C/century is offering what they consider a suckers bet? Since the IPCC suggests the best estimate of the trend is 2C/century? I do think as just an offer out of the blue, someone who believes 2C/century should consider 2C/century the dividing line for a “fair bet” with even money. The only reason to make fair bets is for fun, as on average neither party profits from these. (If Tilo eventually agrees to the brownie bets, on average money will be wasted and postage. So, clearly, the only reason to do it is for fun.)
Martrin Ringo (Comment#3584)
June 25th, 2008 at 12:25 am
Lucia, If the bet is decided by above or below 2 C per century, then I believe the model believer is being played for the sucker, i.e. the model believer is the sucker. The IPCC scenarios (shown somewhere in AR4) are below: IPCC AR4 Scenarios (trends in degrees C per century) Average 2.80 1.67 4.45 The average of the “best” estimates is 2.8, but there are obvious differences. What I said was, “What is the low end trend that is consistent with a fundamental confidence in the modeling processes but at the same time recognizes that such modeling has some internal uncertainties?” I simply threw out the lowest value. If it is scenario related (i.e. a bunch of other estimates were close), the scenario is too nebulous for modeling, i.e. requires too much judgment about things which can be estimated. That criticism applies to the other estimates but I am only trying to find what is something that a model believer can have lots of confidence in. And my guess is the 1.4 C per century. If he needs a bigger range, then I would say from the evidence presented in the AR4 the guy does not have confidence. A bigger number indicates a foolishness in the light of evidence of model uncertainty. i.e. a sucker waiting to happen. A 1.4 C per century isn’t unfair to the model skeptic. It allows more than enough “luke warming.” Further, it current trends stay roughly the same or maybe a bit hotter, the trend estimates will statistically differ from 1.4 too. A skeptic also has to have confidence in his beliefs. Thus, I believe that people who have the courage of their convictions, should be betting, at least for fun, in the vicinity of mid way between 1 and 2 degrees C per century over a 30 year period. But, hey, if you can find a sucker, go for it.
Best Low High
B1 1.8 1.1 2.9
A1T 2.4 1.4 3.8
B2 2.4 1.4 3.8
A1B 2.8 1.7 4.4
A2 3.4 2 5.4
A1FI 4 2.4 6.4
Min 1.80 1.10 2.90
Max 4.00 2.40 6.40
Zeke
(Comment#3588)
June 25th, 2008 at 7:33 am
I always thought Tamino’s wager was a reasonably fair one, at least for those who believe that there is an equal chance we will see warming or cooling in the future (e.g. E[temp] = 0). http://tamino.wordpress.com/2008/01/31/you-bet/ Obviously its not a bet you could get a lukewarmer to take up, but lukewarmers are much harder to arbitrage :-p
Boris (Comment#3594)
June 25th, 2008 at 9:04 am
It’s interesting he wouldn’t bite. It does appear those who set out these challenges won’t take bets that are supposed to be even money according to the IPCC documents. Why would he bet on what he sees as a 50/50 proposition? It’s a guaranteed loser if the money is held in escrow because he couldn’t overcome the vig of lost interest. Would you bet 10K cookies on a cionflop to be made in 2030? Those cookies would be mighty stale. Annan has many bets with denialists out now. And the offers are open for more if the wager is profitable.
Tilo Reber (Comment#3595)
June 25th, 2008 at 9:05 am
Lucia: Okay, the bet is on. I am in Colorado. So postage should be cheap. But let’s make it 2 dozen, otherwise I won’t get any after my wife and daughter get hold of the package. My offer to Annan is in this thread: http://julesandjames.blogspot......l#comments “My main thoughts were: Annan could negotiate the start date to exclude data already in the pocket. That is entirely reasonable after all. Did he?” Annan did not offer a counter bet. He did say that he was working on his own number for a decadal trend. So I may make the offer again once he comes up with it. Annan seems to be mainly interested in bets where the odds are strongly stacked in his favor. Taking a 50/50 bet to demonstrate that he actually stands behind an AGW number is not for him. When it comes to defending those numbers he would rather rely on error bars and the “has not been falsified” excuse.
Tilo Reber (Comment#3596)
June 25th, 2008 at 9:09 am
Boris: Those are simple details to overcome, Boris. We could both put the money into a money market account now, and then the winner could take the entire value at the end of a bet.
“It’s a guaranteed loser if the money is held in escrow because he couldn’t overcome the vig of lost interest.”
lucia
(Comment#3597)
June 25th, 2008 at 9:28 am
Zeke– Tamino’s bet is in the vicinity of fair between someone who believes the previous trend on earth will hold and someone who believes in zero warming or actual cooling. But Tamino explicitly said he wouldn’t actually bet under any circumstances. So, it’s not a real offer to bet. But what if it were, I think it’s skewed a bit. Or more precisely, it’s fair to a very small subset of possible people. Tamino’s bet is also not based on the IPCC and GISS temp specifically. He took 1.8173 C/century of warming; that’s the 1975-now trend. The IPCC AR4 has a central tendency is 2 C/century. So, he gets a 0.2 C buffer on “consensus theory”, which Tamino has, at a minimum, not disavowed. If the bet offer were real, he is telling the other better to take 0 C/century. So, who is this “fair” to? I think the only group it’s fair to is “Cooling has already started coolers.” There are so few of these, the label isn’t even circulated. Since Tamino got a roughly 0.2 C/century buffer relative to the IPCC projections, these people have to believe their theory says the underlying cooling is -0.2 C/century and it’s already started. Otherwise, Tamino got a small edge in the bet, based on a) IPCC theory and b) what I would judge to be his rhetoric that gives the appearance he thinks the future warming rate will exceed the warming since 1975. (That is, I think he designed in a buffer relative to his rhetoric.) And to be truly fair, the other bettor has to be permitted anonymity. After all, this is a public bet. Bragging rights is an issue. The fact that one party doesn’t have to be known to the public lets them save face if they lose!
lucia
(Comment#3598)
June 25th, 2008 at 9:42 am
Tilo, I don’t find the idea of betting $10,000 on a coin toss to be attractive. I see you’ve already had this explained out to you on Deltoid. That is to say, he does only take bets he thinks give him a large edge. This is obviously sensible, from the point of view of betting large sums of money. I’d do the same thing. But it also illustrates why refusing to take a bet doesn’t mean you don’t believe what you are saying. With real money bets, everyone wants to profit, and that includes covering transaction costs. We are betting on what is pretty much a coin toss–but it’s two dozen brownies for you if you win. This is not only a coin toss in terms of who might win, but there isn’t even any saving or losing face involved. After all, whether GISS trend is higher or lower than the others over 6 months probably won’t cause either of us embarassment!
So Annan’s answer is:
If I win, I’ll take 2 dozen chocolate chip cookies.
lucia
(Comment#3599)
June 25th, 2008 at 9:57 am
Boris– Do you happen to know what bets Annan took? Are the conditions and bet levels posted? I’m just curious to read what who bit, and what conditions. I’ve only read the structure of a few bet offers.
Tilo Reber (Comment#3600)
June 25th, 2008 at 10:37 am
“But it also illustrates why refusing to take a bet doesn’t mean you don’t believe what you are saying.” Not making a counter offer does tend to indicate that. “With real money bets, everyone wants to profit” No, I would be willing to take a 50/50 bet with the only profit being that an AGW climate scientist has actually stood behind a prediction. The only catch is that I want a real number, not an error band. I think that the temptation for AGW climate scientist is to overblow their numbers, feeling that they will be saved by their error bands. Since so many of the AGW climate scientists seem to also be AGW political advocates, I want evidence that they are as willing stand behind their numbers with their own money as they seem to be willing to stand behind their numbers with trillions in public money. And since so many of the AGW climate scientists seem to be such passionate advocates and defenders of their predictions, I would think that they would actually welcome the opportunity to stand behind their numbers with real money of their own. If they think that they need a slight edge in order not to look foolish at taking a 50/50 bet, then a counter offer would seem to be the answer. Arranging the kind of sucker bets that Annan currently engages in don’t really prove to me that he believes what he preaches.
Fred Nieuwenhuis (Comment#3601)
June 25th, 2008 at 10:42 am
How about bets (for bragging rights) in the near term? GISS land/sea anomoly for June08? I’d say < 0.3 C.
Tilo Reber (Comment#3603)
June 25th, 2008 at 11:17 am
Lucia: Yes, I don’t actually have any strong feelings about what the outcome will be. But someone will be eating cookies, and that’s a good thing. I would think that another good short term bet would involve the Arctic sea ice area. Will it reach the same 3 million square kilometer anomaly this year that it reached last year? If one believes that the Ocean is a large heat sync that has absorbed a lot of energy over the past thirty years, and if one believes that it will take several decades to come to equilibrium, then one might believe that this years melt off will be as large or larger than last year, even if we have had a flat trend in surface temperature. I do believe that the ocean is a heat sync, but I don’t believe that the lag time to equilibrium is longer than a decade.
“This is not only a coin toss in terms of who might win, but there isn’t even any saving or losing face involved.”
lucia
(Comment#3604)
June 25th, 2008 at 11:33 am
Tilo-
No, I would be willing to take a 50/50 bet with the only profit being that an AGW climate scientist has actually stood behind a prediction.
If you’re willing to put down $10,000 on a 50/50, you have more loose change than I do!
But, conceptualy I can understand this.
In my view, bets do need to have hard numbers. Also, with respect to people who blog, cookies, bet to a non-anonymous third party, with the bet announced involves “face”. I think that’s just as important. But yes, you can’t bet error bands. You can only bet an actual concrete number.
We have bets in for summer. I need to tabulate them and post for reference soon. But, we have no cookies or money on those bets.
Fred– Even though the agencies haven’t reported, I think if we are going to set up short term bets for monthly anomalies, we need to place bets before a month starts. Most of June is gone. But, if people are willing, I might look into something that could let individuals log virtual bets for fictional quatloos. I’d have to figure out rules. Someone registers, gets 100 quatloos. You can bet between 0-5 quatloos on GISSTemp each month. Then, we track who is ahead and who has gone bust. It could be fun, and quatloos would fictional.
rex (Comment#3607)
June 25th, 2008 at 12:53 pm
Actually June looking quite cool
http://discover.itsc.uah.edu/amsutemps/
Has anyone here even contemplated the fact that temps may actually DECLINE over the next 10-100 years? Its called Climate Change
lucia
(Comment#3608)
June 25th, 2008 at 1:05 pm
Rex– It looks cool compared to 2007. I added other years, and maybe not so cool. I’ll wait for UAH to do the averaging instead of trying to eyeball that.
Fred Nieuwenhuis (Comment#3609)
June 25th, 2008 at 1:11 pm
As far as I’ve observed, predicting GIStemp values is a bit of a crapshoot regardless of where you are in the month.
lucia
(Comment#3610)
June 25th, 2008 at 1:35 pm
Fred–
If we want a real crapshoot, we predict how much May’s GISS temp will change when they report June’s results!
Hans Erren
(Comment#3611)
June 25th, 2008 at 1:38 pm
If you save your image as gif it won’t look so blurred.
Decrease colour depth to 256 colours, but do not use floyd-steinberg dithering.
lucia
(Comment#3612)
June 25th, 2008 at 1:47 pm
Lets see if the GIF looks better. The new version of wordpress seems to do something different.
Tim Curtin
(Comment#3613)
June 25th, 2008 at 5:56 pm
With all due respect to the inestimable Lucia, I see no cause to reduce CO2 emissions, as in a forthcoming paper I demonstrate how that would risk a major world wide famine far more damaging than 2-3 oC by 2100. My paper’s regression shows how rising atmospheric CO2 explains the growth in world food production since 1980 - without that extra CO2 the new crop varieties would have starved. The claimed damage from 2-3oC warming is nonsense when the world daily copes with variations much larger than that. Even if the whole world warmed to the average temperatures already normal in Dubai we would survive, as those Emiratis prove so handomely.
climatepatrol
(Comment#3618)
June 26th, 2008 at 6:46 am
@Lucia
Great post.
The current data strongly indicate the current underlying trend less than 2C/century, and likely less than 1.6 C/century!
When taking the average trend of five metrics of the entire satellite era, I arrive at a trend of 1.63°C per century only when using Giss “meteorological stations only”. This is the true outlier with a trend of 1,86°C per decade. Giss-Land-ocean seams okay to me when looking at the entire 29-year period. see superimposed graph here. But then, unlike you, I am not a mathematician. Excel did it all for me
Boris (Comment#3619)
June 26th, 2008 at 7:31 am
Here’s Annan’s betting summary:
lucia
(Comment#3620)
June 26th, 2008 at 7:46 am
Thanks Boris– On the other hand, Monbiot offered a nonsense bet. The guy he was talking to believes in slow warming, and Monbiot offered a bet that would only be taken by a person who believed in zero or cooling. Those willing to bet need to learn to give decent counter offers– ask people who offer nonsense bets to bet the central tendency from the IPCC! Well… I”ll need to look at the rest!
Wow, the bet offer from Kappenberger and Micheals was dangerous (for them!) They’d bet a statistically significant downward trend in 10 years? The honest to goodness downward trend would have to be … (I’m not going to calculate) between -2C/century to -3 C/century to get even odds of a statistically significant downward trend! Granted, cherry picking the start point pushes the odds a bit!
Bob B (Comment#3621)
June 26th, 2008 at 7:53 am
#3610–Lucia June is now tracking colder again: http://discover.itsc.uah.edu/amsutemps/ I bet for UAH and RSS June could be another cold month
Raphael (Comment#3623)
June 26th, 2008 at 8:35 am
Lucia, I thought that since this thread was a bit off topic talking about bets, that I would ask a few of my “stupid” questions. I have regular politcal discussions with my sister. She is a liberal, I am a conservative, so they can be quite interesting. Earlier this year, she brought up the topic of the threat of global warming and climate change. I knew little about the subject, and my initial response was “I haven’t thought about it.” I told her that the temperature on earth goes up, the temperature goes down, and it has been doing that for millions of years. She said, “What about the CO2 we are putting into the atmosphere causing increase?” I said, well, CO2 causes a delay in radiation leaving the atmosphere. An increase in co2 would increase this delay. But, I cut short the topic by saying I had no idea how much increase would occur, whether the increase would be a “threat”, and if any action the US could take would have a significant impact. That was then, and several months later, after much searching, I still do not have acceptable answers. I wanted to ask you, because I believe you to be a truth seeker and are a AGW beleiver. (or to translate, I believe I can get a rational explaination of the AGW believer’s views, which can be difficult to find on the internet without having to translate it from “sciencespeak”.) So I’ll just start with something which has been bothering me for a little while. Among the first things I found were the datasets from the Vostok ice core. I am not interested in CO2 leading temperature (I’ve seen enough of that debate), but rather the last 10000(or so) years of temperature data. Assume: The Vostok ice core shows some relationship between global temperature and CO2. Assume: Answer to previous question was yes. Assume: Answer to previsous question to be yes. Moving on to a few easier questions: When does a warming trend become a threat? (as per the “threat of global warming”)
Assume: CO2 accellerated warming is occuring.
Should the surface temperature dataset from Vostok station for the last 50 years show signs of warming?
Assume: current CO2 levels are significantly greater than any previous time during this interglacial.
Should a comparison between the ice core data set and the station data set show the station temperatures to be significantly warmer than icecore temperatures?
Assume: My calibrated eye ball fails to see warming in the station data which is inconstistant with the icecore data.
Should I have my eyeball recalibrated?
What is being threatened?
Which of those things being threatened should move the Federal Government to take signifcant action?
Will that action have a significant impact on climate?
Boris (Comment#3624)
June 26th, 2008 at 11:23 am
Those willing to bet need to learn to give decent counter offers Agreed. Tell that to Tilo.
Atmoz
(Comment#3625)
June 26th, 2008 at 11:32 am
Where are you finding ice core data from the last 50 years?
Raphael (Comment#3626)
June 26th, 2008 at 11:45 am
It isn’t Ice core data for the last 50 years, it was surface temperature data (1958-2008) from Vostok station. Which I found here. (I hope I remembered how to tag that properly)
Arthur Smith (Comment#3627)
June 26th, 2008 at 12:56 pm
Raphael - I’m no expert on this, but my understanding is that the ice core “temperature” time series is not a measure of temperature directly, and in particular not a measure of the local temperature in Antarctica, but rather a proxy chemical measurement (based on oxygen isotopes?) that is I believe primarily determined by average sea surface temperatures around the globe. So there would not be any expected correlation between the Vostok station temperature (which has indeed not warmed recently - central Antarctica remains very cold for several well-understood reasons) and the “ice core temperature” because they’re measuring different things.
Raphael (Comment#3628)
June 26th, 2008 at 1:29 pm
Arthur, Because isotopic fractions of the heavier oxygen-18 (18O) and deuterium (D) in snowfall are temperature-dependent and a strong spatial correlation exists between the annual mean temperature and the mean isotopic ratio (18O or δD) of precipitation, it is possible to derive ice-core climate records. J.R. Petit et.al Because I can’t imagine how something in snowfall could be dependant on a temperature anywhere except where the snow is falling, I infer that it is local temperature which is being measured via the proxy. But, I have a poor imagination and am open to alternative suggestions.
lucia
(Comment#3629)
June 26th, 2008 at 2:06 pm
Raphael– Should he surface temperature dataset from Vostok station for the last 50 years show signs of warming? In priciple yes. But, I don’t know the resolution for time in the cores, or whether or not they can actually get measurements at a surface or how many inches of core corespond to a year. So… since I don’t know much about it, for all I know, the measurements from cores never start at “now” and back. I know when they took cores from radioactive waste in million gallon tanks at Hanford, there were difficulties involved in the very top surface. So, I wouldn’t be surprised if you can’t get info for very recent times. (I looked to read answer– looks like Atmoz asked this “Where are you finding ice core data from the last 50 years? ” S Assume: Answer to previous question was yes. I’m not sure I know what you are asking. Why do you think the temperatures for the ice-cores temperatures to be cooler than the station data? Do you mean the nearby station data? Moving on to a few easier questions: When is a warming trend become a threat? (as per the “threat of global warming”) These aren’t easier questions for me! These questions are what constitutes the debate over climate change!
You are asking me a lot of questions I don’t know the answer to! But, there are a few people whom might be able to answer.
Assume: current CO2 levels are significantly greater than any previous time during this interglacial.
Should a comparison between the ice core data set and the station data set show the station temperatures to be significantly warmer than icecore temperatures?
What is being threatened?
Which of those things being threatened should move the Federal Government to take signifcant action?
Will that action have a significant impact on climate?
Nick Stokes (Comment#3630)
June 26th, 2008 at 2:34 pm
Because I can’t imagine how something in snowfall could be dependent on a temperature anywhere except where the snow is falling, I infer that it is local temperature which is being measured via the proxy. But, I have a poor imagination and am open to alternative suggestions.
Rafael, there’s a discussion of proxy temperature measurement here. Isotopic separation occurs when water evaporates or condenses. The main influence for ice is the temperature when the water last condensed. That’s reasonably local.
Raphael (Comment#3634)
June 26th, 2008 at 7:39 pm
Nicks, Thanks. I tend to try to make sure I understand at least the general concepts. But, fully realize that there can be major gaps in that understanding. I want to know enough about global warming to at least hold an intelligent conversation about it. Lucia, No problem if you don’t know the answers. This gives me something to do which requires some thought. I’d probably have more fun with it if I had any desire to relearn the math that has atropied to the point of non-existance. While I realized it wasn’t what you usually post on, I thought maybe you could give me enough insight put the thought to rest. Sorry if I wasn’t clear in the way I phrased the questions. For all my intelligence, I can be pretty dumb when it comes to getting the information out coherently. The idea went something like: CO2 before the beginning of the current interglacial (17000-22000 years ago) was ~180 to 190 ppm. The local temperature during this time was about -63C. Currently CO2 is around 386 ppm but temperatures at Vostok are still fluctuating around -55C But, I think I can disregard the Vostok questions for now. It made more sense before I took a look at the numbers for the CO2 icecore data set. I had assumed that the recent (last 23,000 years) CO2 dataset was as robust as the temperature dataset. But it’s not.(15 data points vs 400. bleh.) 7 data points during the steep post ice age warming, and 8 for the “flat top” doesn’t provide much information. On the surface I thought there was something rather interesting. The other questions I have because I have been brainwashed to believe that a particular threat level requires an appropriate force response. When I hear the word threat, I think what is the appropriate response to a threat at that level. *shrugs* Doing something for the sake of doing something? No thanks.
After the rapid climb out of the ice age (11,000 years ago) CO2 was sitting at about 263 ppm. Since that time, temperature fluctuated around -55C while CO2 declined a bit and then increased a bit more. The final CO2 from Vostok ice core sat around 284 ppm (~2300 years ago).
lucia
(Comment#3635)
June 26th, 2008 at 8:47 pm
Raphael– I wasn’t suggesting there was anything wrong with the questions. The are the questions discussed by everyone. I just don’t know the answers to all of them. They aren’t easy questions.
Tilo Reber
(Comment#3636)
June 27th, 2008 at 1:24 am
Lucia: Did you notice this quote by Annan. “but so far not a single sceptic has come forward to disagree with the standard (IPCC) view. Well, I came forward to disagree with the standard IPCC view. But apparently Annan doesn’t want to take his own advice of “taking on a few losing bets, just to avoid looking dishonest”. Not only won’t he take a loosing bet to avoid looking dishonest, he won’t take a 50/50 bet to avoid looking dishonest. In fact, he wouldn’t even make a counter offer to a 50/50 bet to avoid looking dishonest. Boris: “Agreed. Tell that to Tilo.” Boris, I made the original offer. Do you want me to make myself a counter offer? And I havn’t refused a reasonable counter offer, since Annan hasn’t made one. Let’s be honest here, Annan is only taking sucker bets and he is not taking any bets that require him to stand directly behind IPCC predictions.
Sooner or later the denialists might decide that it is worth taking on a few losing bets, just to avoid looking dishonest. I just hope I can get a slice of the cash when they do.”
Boris (Comment#3639)
June 27th, 2008 at 6:27 am
Tilo, You were pretty adamant about betting the IPCC 50/50 on the Deltoid thread. If you want to bet something along the lines of what Atmoz suggested upthread I think Annan might be interested. Why would not taking a 50/50 bet look dishonest? You seem to miss the whole point of wagering in this context. Each side should believe they have an advantage. If I say 0.2C/dec and you say 0.1C/dec, a fair bet would be 0.15C/dec. Then we’d each think the other was a sucker. If you think the bets are sucker bets, then we agree that those scientists peddling cooling or very little warming are full of it.
Raphael (Comment#3647)
June 27th, 2008 at 10:42 am
Lucia, There was a problem with my Vostok questions. I had assumed that a dataset I hadn’t seen was as robust as one I did see. This explains why I had never seen the question. There simply isn’t enough CO2 data for the recent (20000 years) past. Had I bothered to look instead of assuming, I would have went, “Oh, well that won’t work.” *shrugs* The other questions– I know there are nothing wrong with those questions. It is just the way my mind works when it comes to “threats”. My sister tells me I’ve been brainwashed to think that way. Brainwashing, conditioning, and training are all pretty much identical concepts. The differences involve intent and technique. Technically, I was trained to view threats a certain way. So when they occur, little thought is required to take action. I recognize that changes in climate can be a threat to national security. For an example, a potential of a modern European “little ice age”, prompted the creation of a plan for a military response to the destabilization caused by the event. I would rather spend 500 billion dollars to prevent the event which was quite likely to occur, than to spend 500 billion plus loss of life responding to the destablization. But, I wouldn’t want to spend a dime, if there wasn’t a reasonable chance of neutralizing the threat, or if the threat wasn’t quite likely. So, do you have any opinions on the questions?
Clark (Comment#3650)
June 27th, 2008 at 10:58 am
If “cap and trade” and other economy-destroying nonsense get enacted, we are all going to have to pay a lot more than $10,000 - no matter where the trend ends up. You just better hope they don’t put a special carbon tax on cookies!
MarkR (Comment#3665)
June 27th, 2008 at 9:52 pm
From CA. <blockquote cite=”Well done Kenneth. Now I would like to see Lucia include the raw unadjusted data from the CRN1 stations in her statistical tests of the probability of the IPCC forecasts. Maybe you could link your calculations and results to Watts database, and we can see the further stations coming in? I think the IPCC forecasts will be mathmatically improbable. Also, it may be that there has been no material recent warming in the US. The most accurately suveyed area in the world.”>
Boris (Comment#3677)
June 28th, 2008 at 9:00 am
“Also, it may be that there has been no material recent warming in the US. The most accurately suveyed area in the world.” If the US isn’t warming, no one is. We’re number one dammit!
lucia
(Comment#3683)
June 28th, 2008 at 11:36 am
MarkR: Now I would like to see Lucia include the raw unadjusted data from the CRN1 stations in her statistical tests of the probability of the IPCC forecasts. My impression is this would involve a huge amount of work on my part. I would need to get the data from each individual station, figure out “the lucia way” of finding bad data. (There will always be bad data at even the best station.) Then, I would need to figure out how to weight each station etc. This is the sort of thing that NOAA/NCDC, GISS and HadCrut assign to teams of people. The project may be do-able, but I’m afraid creating a “lucia” data based based on the stations Anthony’s volunteers have found to be good is more than I’m willing to take on. But… if someone were to form a group blog etc. to do it, that would be great.
MarkR (Comment#3687)
June 28th, 2008 at 1:22 pm
Hi Lucia I think Kenneth over at CA already has a spreadsheet of it for you.
Niche Modeling » Garnaut Report Due
(Pingback#3885)
July 3rd, 2008 at 11:06 pm
[...] attempt to revew the quality of evidence provided by existing work. The preface reports work that looks alot like lucia’s: The dissent took a curious turn in Australia in 2008, with much prominence being given to [...]
Nick Stokes (Comment#3888)
July 4th, 2008 at 1:31 am
Lucia, Following up on David Stockwell’s pingback, you might like to look at the time series analysis that he posted. It’s quite readable, even though not all that decisive. The main points I noted were:
1. There is an overall uptrend over the 150 years of data
2. They are strong on identifying a 1976 breakpoint with a stronger uptrend afterwards
3. They identify a ARIMA(0,1,2) model. That’s one step up from your C-O analysis, though I doubt if the reason for preferring second-order is very strong. Anyway, they agree that there is strong autocorrelation.
4. They applied a Jarques-Bera test, and gave no details, but it seemed to pass. Not surprising that they chose J-B, which comes out of their ANU department.
5. Garnaut, who requested the analysis, asked if the positive trend (no specific number) post-2000 was falsified. The report doesn’t say much, but apparently Garnaut says the answer is no.
Cassanders (Comment#3890)
July 4th, 2008 at 3:23 am
@Nick Stokes Cassanders
Sems to be a reasonable take, but how do you think the results would have looked if they redid their exercise and included the years from 2001(2?) up to now?
In Cod we trust
David Stockwell
(Comment#3892)
July 4th, 2008 at 4:22 am
Nick, On point 5, the question was: “Is there ANY (my emphasis) indication that there is a break in any trend present in the late 1990s, or at any other point?” Where BV08 is inadequate is in the test of ANY evidence of change in trend. To assert ‘no significant break’ first you have to establish the test actually has the power to detect a break. There are so few annual data points after 1998 that the power is very low, i.e. it would only detect a very severe downturn. I would have run a structural break test on monthly data from 1979 as see what turns up. With 120 points on each side of a potential break, who knows? When asked for ANY evidence, they should have looked a bit harder, don’t you think!
lucia
(Comment#3893)
July 4th, 2008 at 5:42 am
Nick- It’s just the current trend is not inconsistent with 2C/century or higher. CO falsifies 2 C/century projected by the IPCC. I’ll have to look at the specific report. But, genearlly, T tests to show whether two trends are consistent tell you whether or not they must be different. But in this case, they could be the same– it’s just the early trend would have to represent a “high” excursion above the underlying rate and the later trend would have to represent a “low” excursion. I may have to do the t-test for two means now. The problem is I need to double check precisely how to do this for trends.
Your take suggests that pdf is entirely consistent with what I’ve said in various blog posts. CO doesn’t falsify a positive trend either. I’ve pointed out 1,2, and 4. I’ve never looked at Arima (3). If you look at the “bar and whiskers” plots, there are plenty of pre-existing positive trends that are entirely consistent with the current weather– so that’s the same as his point 5.
lucia
(Comment#3900)
July 4th, 2008 at 9:08 am
Nick– They don’t examine the hypothesis I examined– which is whether or not the data are consistent with the IPCC projection of 2 C/century. I also don’t falsify a positive trend, and have shown numerous graphs showing the post 2001 data are I haven’t specifically looked at 1976. I’m not surprised there is a statistically break there. The volcanos started erupting in 1963, and 1976 is located in what would commonly be considered the time when the recovery from one of them was expected to start. Then, we start having the cooling aerosols drop out (due to the consequences of the clean air act.) There is much more scatter and autocorrelation during that period than recently. I think this large autocorrelation is likely a consequence of the volcanic eruptions which introduce a dust veil which persists and drops off. So, mostly, I don’t see how any of what they say contradicts a single thing I’ve written or found.
I had a quick look at the Breush Vahid article. As far as I can see, they do more-or-less similar test relative to mine (though there’s are more complicated.) They find more or less what I find.
The text appears to discuss whether or we can falsify a positive trend after 1998.
That doesn’t happen to be the year I look at for a break, because I don’t look for a break. I picked my year based on publication dates of the IPCC documents.
a) not inconsistent with the positive trend of about 0.6 C/century since 1900.
b) not inconsistent with the positive trend since 1990.
Nick Stokes (Comment#3904)
July 4th, 2008 at 4:08 pm
Lucia, I agree. Their approach is fairly similar to yours; it doesn’t address exactly the same questions, but its results don’t contradict yours. I thought the way they went about it was interesting. Cassanders, as far as I can tell, they’ve included data up to 2007. My reading of their treatment of the recent stuff is that Garnaut asked them the question about the last decade, and they just said they couldn’t find strong evidence of a break in trend in 1998.
Ian Castles (Comment#3905)
July 4th, 2008 at 6:49 pm
Lucia, Do you have any comment on B & V’s reliance on averages of calendar years rather than monthly data? For 2007, the temperature anomaly in January was between 0.4 and 0.5 C higher than in December, both for the CRUTEM and the GISS. Under a calendar year analysis, the results would have been exactly the same if the progression of GMST anomalies through 2007 had been in the opposite direction. Does this not raise a question about restricting the analysis to annual averages?
Yes, on my reading there is nothing in the B-V paper that contradicts anything that you’ve said. It is however surprising that Garnaut did not also ask the ANU statisticians for an assessment of the statement made in his interim report in February that “Global mean surface temperature increase since 1990 has been measured at 0.33 C, which is in the upper end of the range predicted by the IPCC in the Third Assessment Report.” This statement was sourced to the Rahmstorf et al (2007) paper that you (and a number of others) have strongly criticised - and it seems to be no less amenable to analysis by time series experts as the hypotheses that B & V were asked to test.
lucia
(Comment#3907)
July 4th, 2008 at 7:58 pm
Ian, Do you have any comment on B & V’s reliance on averages of calendar years rather than monthly data? I have no idea why they chose to do this. Apparently, they know enough statistics to deal with the autocorrelation, so I would have thought they’d use the available monthly data. Using the monthly data would give more total data points, and also permit them to include the recent downturn in the temperature. Using the most recent data would, presumably, provide important information. In particular, it would provide more recent data. This is important answer the question “Has the trend changed recently.” As it is, they’ve left out 5% of the data– and in particular, data that are important to the specific question being asked. That said, I too am surprised Garnaut didn’t ask the economists to consider the questions about the IPCC 2.0C/century projection and/or Rahmstorfs evaluation of the TAR projection. It seems to me, that is the more important question that should be addressed by professional statisticians. I know the questions you asked were related to data comparisons between the AR 4 projections and the TAR projections. I’ll look at this more on Monday, but right now, those questions were either not understood or ignored by Garnaut and so not addressed by the statisticians he hired. It’s the fourth of July, so I haven’t read the paper in enough detail to say anything more!
David Stockwell
(Comment#3909)
July 4th, 2008 at 8:38 pm
Even though they set up a straw man by looking at whether there is a trend or not, a significant result for even that is dubious using their approach. Three issues they did not address are: 1. They used a one-sided test to show a significant trend term in the unit root model. A two-sided test with at conventional 1.96 critical value would have made two of the series non-significant. A two sided tests should be used unless there is a very good reason for doing otherwise, as it could be argued that a one-sided tests was used simply as a device to make a conventionally non-significant difference significant. 2. They used a slide-and-eyeball test for showing that “the temperatures recorded in most of the last decade lie above the confidence band of forecasts produced by a model that does not allow for a warming trend.” But in a footnote they admit “The confidence bands in these graphs are calculated using the dynamic forecastoption in Eviews and they do not incorporate estimation uncertainty.”. A slight increase in uncertainty concerning the parameters and the start point would put all temperatures inside the confidence band. 3. Not using MSU RSS data is a big omission.
Gavin Schmidt Corrects for ENSO: IPCC Projections Still Falsify | The Blackboard
(Pingback#3914)
July 5th, 2008 at 11:19 am
[...] Blackboard: Solar Cycle. The issue of ENSO… Surface Temperatures … [...]
lucia
(Comment#3915)
July 5th, 2008 at 11:43 am
They used a one-sided test to show a significant trend term in the unit root model. Yes. I have avoided one-sided tests for this very reason. Many readers don’t realize that since I use two sided intervals, the tests are supposed to say there is s 2.5% chance a trend lies above the upper interval and a 2.5% chance it lies below the lower interval. (Now, I know you know it doesn’t mean precisely that. Matt Briggs discussed what it means more precisely at his blog. But, wording precisely what it means is difficult, since it really has to do with how often the statistician would be wrong if they applied this test methodology over and over and over etc. Also, it must always be understood that a statistical test result is contingent on the assumptions for the statitical model being correct. But these assumptions are always made.)
Nik (Comment#6814)
November 17th, 2008 at 3:09 pm
Do you or anybody have a record of average temperatures from month to month? There are many for yearly changes but trying to do any kind of research on this place without getting all the save the world stuff is hard. I just wanna learn about the place and its temperatures and I cant find any. Can you help me?
lucia
(Comment#6815)
November 17th, 2008 at 3:13 pm
Nik– Somebody somewhere has those, but I don’t! The services looking at climate change use anomalies. Still, compilations of average temperatures for various locations on earth exist– both on line in an a variety of references. The bet resource for you depends on what you want to know. Have you tried weather.com to look up the averages for your zip code? Those sorts of services do give simple weather averages.
Comments Closed: If you would like them re-opened, Contact Lucia
Anthony Watts (Comment#3556) June 24th, 2008 at 1:01 pm
Nicely done Lucia. You’ve earned a cookie