Global Warming = Record snow?

Great Debates
81

brazil: It is impossible to have a civilized discussion with someone who ignores two published, peer-reviewed reports clearly showing how the temperature has evolved in time as compared to the IPCC projections over the last 17 years, dismissing them as historical re-writes, and then claim he knows better by looking at graphs that have scales that do not allow one to see the requisite detail for the period. It is just utter and complete garbage and not worthy of Great Debates.

If you are going to accuse reputable scientific sources as committing fraud and substitute in their place your own interpretation of a graph that does not have sufficient detail to allow you to draw the conclusions that you do, then it is really pointless having this sort of discussion. Frankly, you should be posting in another forum.

That graph has a y-axis of 6 C and, according to this graph the width of the shaded region in 2000 is about 0.1 C. You simply can’t see it on the scale of that graph. Furthermore, if you look at the slope of the shaded region boundaries in 2007 according to this graph , it will look like they extrapolate back to zero around the year 2000 but in reality they are curved so that the width remains nonzero between 1990 and 2000.

And, no, the projections did not start in 2000 because the IPCC convention is to start the runs in 1990. Already by 2000, there is a spread in the model results.

There are so many wrong statements in these few sentences that it is hard to analyze them all:

(1) Temperatures have not been “essentially flat for the last 8 or 10 year”. If you look at the 8-year trend obtained for a least-squares fit of a line through the data, you get a trend of +0.12 C per decade for the Hadcrut3 data and +0.25 C per decade for the NASA GISS Land+Ocean data. If you look over 9 years, the trends are even larger. If you look over 10 years, they shrink down again because of the warmth of 1998, but the trends still range from +0.05 C per decade for the Hadcrut3 data to +0.18 C per decade for the NASA GISS meteorlogical stations data. Of course, all of these trends tend to have large error bars because of the relatively short span of time, which is why the bounce around quite a bit just based on including another year of data. If you go to longer periods than 10 years, the trends start to increase again and the Hadcrut3 and NASA results converge so that by 15 years, the Hadcrut3 data trend is +0.22 C per decade and the NASA GISS trend is +0.26 C per decade.

(2) The IPCC does not predict the short-term variability, which is dominated by other effects and is very sensitive to initial conditions. Nonetheless, in individual runs of climate models, comparable short-term variability is seen. See for example, here.

(3) As the IPCC AR4 figure and the Science article show you, in fact the smoothed average global temperatures remain well within the IPCC envelope and are closer to the high end than the low end.

82

By the way, it is impossible for me to tell which of links to Science require a subscription. If that one isn’t working for some people, you might want to see if this one does or you can look at Fig. 1 in this PDF of the paper.

83

Whoops! That latter trend is for the NASA GISS Land+Ocean data. (I originally did the trends also for the NASA GISS meterological stations data too, which have even larger positives trends than the Land+Ocean data, but I decided to take them out since it can be argued that they are not representative of the full surface of the globe but only the land regions. The quoted number of +0.18 C is the correct one for the NASA GISS Land+Ocean data.)

84

You gotta be kidding me. I used the chart that came straight from your precious IPCC. Then I used a chart that you yourself linked to; and you yourself claimed it was more detailed.

Too bad those charts don’t show what you wish they show.

Funny, but they look pretty flat to me. At least to my x-ray vision. :rolleyes:

In that case, they should have made the gray range bigger. Oops, indeed.

85

I think, Jshore you are not hearing what he said: “they look pretty flat to me.”

I look at the charts you linked to, and read this, and his entire point of view begins to be understandable. But, you are right, not much point in arguing.

Tris

86

Lol. And I guess you didn’t hear what I said when I asked you for a cite for your claim that “Global weather systems are more energetic now than they were thirty years ago”

And by the way, here’s a nice NPR story from 2006 about how global warming is threatening the American ski industry:

Aspen Resort Fights to Save Endangered Snow

Guess they hadn’t received the memo that Oceania is actually at war with Eastasia.

87

Look, you tried to use a chart that went over 1100 years to look at something over a 10 year interval. I suggested that you might be able to see more detail on one of their charts that just went over 110 years.

But, the point isn’t whether or not those charts show what I wish they show. I didn’t wish them to show anything, as they can’t weigh in on the issue of how the instrumental temps compare to the projections since they don’t show the instrumental temperatures. For that, you need to look at the IPCC AR4 Figure 1.1 or the figure in the Science article I linked to. I only showed you that other graph from the IPCC TAR so that maybe it would help convince you that the IPCC AR4 and that Science paper did not commit fraud as you basically accused them of doing (based on zero evidence). However, if you would prefer to just believe that the graphs are frauds because you don’t like the conclusions, there is nothing I can do to stop you and it seems like a huge waste of time to try.

Well, that plot was designed by Joe D’Aleo to fool you…and it seems to have worked very well. First of all, it conveniently cuts off all of the data before 1998. Then, it chooses the relative scale for CO2 and temperature such that in order to follow that rise in CO2 over the last 10 years, temperatures would have to have risen 0.8 C. You might ask who has predicted that temperatures over the last 10 years would rise by 0.8 C…Certainly not the IPCC, which has estimated they would rise an average about 0.2 C per decade over the next few decades.

If you scaled that plot reasonably, or just noted that a 0.2 C rise over the decade is roughly what is expected, then you can sort of visualize how the graph is too noisy to conclude whether or not such a rise has occurred or not…In other words, the trend over 10 years just has too large error bars. [As I noted in a previous post, the trend in the Hadcrut3 data over the last decade can be computed to be +0.05 C per decade with very large error bars. Interestingly, if you look over the last 8 or 9 years or anything greater than 10 years, you get a larger trend than for exactly 10 years…which is why they cherrypicked that particular length of time, no doubt. Note that they also cherrypicked which data sets they used, conveniently ignoring the NASA GISS data set that would have shown a +0.18 C per decade trend, almost exactly what is expected…although again still with big error bars. Still, this is the least of their sins and I am almost willing to forgive them that. Otherwise, it is sort of like giving a speeding ticket to a mass murderer.]

If you scale things reasonably and look over a longer time period, then you get the graph shown here. See here for more discussion.

I don’t see why since as the AR 4 report Fig. 1.1 and the figure in that Science magazine article show, the smoothed global temperatures lie squarely within that gray range since the start of the modeling runs in 1990, and are closer to the top than the bottom.

88

By the way, here is a PDF file of that Science article for those who cannot access the figure because you have to be subscribed to access it (if indeed you do).

89

The instrumental temperatures are available elsewhere. See below.

:confused: The question is about temperature data from the last 8 to 10 years. Thus, temps before 1998 are irrelevant.

I realize that you would like to average in earlier temps. It’s a lot easier to make “predictions” after the fact.

:confused: The claim was about temperatures, not about CO2 levels.

90

I know that they are available elsewhere and, as the IPCC AR4 and the Science paper I have linked to show, the smoothed global temperatures since 1990 remain within the envelope of the IPCC projections, generally running closer to the top end than the bottom end.

However, the point is that trends over such short time intervals have huge error bars, showing more about the noise than the signal. As I have noted, the trend that you compute over such short intervals depends strongly on the length of the interval (Do I choose 8 years or 10 years?) or which data set you choose (do I choose NASA GISS or HADCRUT3?). Do you want to show a large trend for a period ending in 2007? Choose the NASA GISS 8-year trend which is 0.25 C per decade. Do you want to show a small trend? Choose the Hadcrut 3 ten-year trend which is only 0.05 C per decade.

It is only once you go to longer periods of time that you get more robust trends. In particular, the IPCC projections start in 1990, which now gives us 17 years of temperatures to compare to. Yes, it is true that the TAR was put out in 2001 (I think mainly being prepared during 2000) so that much of the post-1990 data was available before it was released. However, as noted in that Science article, it is hard to understand how this data could have significantly influenced the modeling given the nature of the models (not empirical but physically-based), the long development time, and the noise in the instrumental temperature record that you seem to believe that they may have tried to fit to.

Furthermore, even if you want to discount the pre-2000 data, the IPCC projections are still based on aligning the (smoothed) instrumental and model-predicted temperatures in 1990 and projecting from there and thus there is an unambiguous way to align the envelope for the IPCC projections with the instrumental temperature record…And the actual smoothed global temperatures remain within that envelope, still closer to the top end than the bottom (although not as close to the top end as they were in 2000). Clearly, after more time has elapsed we will have an even better assessment.

It is usually considered bad form in Great Debates to quote one sentence out of context and pretend that you don’t understand why this is being said. The entire context of that is:

Is there something about that entire explanation that you don’t understand?

91

Fine. You seem to be arguing that no graph can be used to compare predictions to temperatures unless the graph contains both things. Conveniently, this makes it impossible to assess the IPCC predictions by looking at the graphs from the time when those predictions were made. In any event your argument is silly. One can look at the predictions from one graph, and the instrumental record from another.

Sure, and the IPCC prediction would appear to contain such error bars (actually an error range). Unfortunately for you, we’re at the very edge of that error range, if not outside it already.

:confused: I have no idea what your point is or why you are discussing the representation of CO2 levels on the graph that I linked to.

Looks to me like you are simply trying to distract from the my point – which is that temps have been essentially flat for the last 8 to 10 years. For those of us with x-ray vision, anyway. :rolleyes:

Yes, I don’t understand why you bother to attack strawmen. I will concede for the sake of argument that the data in the graph I linked to covers too short a period to nail down a statistically significant trend.

It doesn’t affect my basic point.

92

Yes, in principle I suppose that one can although you seem unable to do it correctly since you keep claiming to get a result that is in contradiction to the result shown in the IPCC AR4 report or that Science article and there is no reason to believe that you are doing it right and they are doing it wrong.

Not according to the two published works that have actually made the comparison. Just according to someone who is speaking out of ignorance and bias.

If by “essentially flat” you mean “within a range of error that includes flat but also includes slopes larger than the mean IPCC projection”, then, yes, it has been “essentially flat”, but alas that doesn’t really prove anything.

Well, yes, it totally undermines your basic point. Your point seems to be that the data seem to be in contradiction to the IPCC projection. The actual facts are that the data for the smoothed global temperature for the entire period 1990 to 2007 are within the envelope of the IPCC projection, running closer to the top end than the bottom end.

And, as for the trend over the last 8-10 years, the statistics are simply not sufficient to draw any strong conclusions because it is simply too short a time interval and the error bars are still too large.

93

No it doesn’t. Given that you are speaking from complete ignorance of elementary statistical inference, I’m not surprised that you can’t see this.

Same problem. As I pointed out in the other thread, you don’t have a basic grasp of statistical inference. Please stop pretending.

94

Wouldn’t an overall rise in global mean temperature mean higher evaporation-more clouds+more snow in Winter?
This sounds reasonable; but snowfall in any one region, is highly variable. Where I live (NE), you never know what winter will bring. Some years, Boston is buried in snow-other years, we go through the winter with little or no snow.
I would begin to accept “global warming”, when we see double and triple vaerage snowfalls in a NE winter.

95

I don’t think anyone really knows. Which is part of the reason why alarmists must make their predictions after the fact.

I would begin to accept global warming if the alarmists were able to make specific and interesting predictions well in advance.

Unfortunately, they are unable to do this. Instead, they (many of them, anyway) look at weather after the fact and suggest or imply that it may be related to global warming. Which doesn’t prove they are wrong, but it’s not very confidence-inspiring.

96

My wife watches TV Globo, and the reports from Fall (in Santa Catarina, southern Brazil) are most interesting…record low temperatures! Frosts (not seen in generations). Seems to me the Southern Hemisphere is cooling!

97

A gradually warming climate does not mean that there will not occasionally be record low temperatures in certain places. The Southern Hemisphere is warming although not as fast as the Northern Hemisphere (presumably mainly due to the fact that there is more water and less than land in the Southern than in the Northern Hemisphere).

There a plenty of predictions being made “well in advance”. Just look at the IPCC report. Of course, the problem with predictions well in advance is it takes a long time to verify them…but that is the nature of the beast.

98

By “well in advance” I obviously mean sufficiently far in advance to preclude any sort of past-posting.

Right, and that’s the nature of many kinds of charlatans. They make difficult-to-verify predictions. Which doesn’t prove that the alarmists are fraudsters, but it doesn’t inspire much confidence.

99

Here’s an interesting article that illustrates the point I’ve been trying to make:

So, let’s suppose that temperatures are flat or cooling for the next 10 years. Is that evidence against the CAGW hypothesis? Apparently not, if the article I linked to is to be believed.

As I mentioned earlier, here’s how things usually work with an honest hypothesis:

  1. The hypothesis makes interesting predictions about the world.

  2. If those predictions come true, it’s evidence in favor of the hypothesis.

  3. If those predictions don’t come true, it’s evidence against the hypothesis.

With CAGW (and other shaky theories), step 3 is often omitted.