The Blackboard

Lucia,
Good to have you back. Happy Easter.

Thanks! By the way, our weather is lovely right now. My daffodils are blooming, as are several other pretty things!

Since it is the start of the spring season, let me be the first one to throw out the obligatory “cherry pick” for the 2000 and 2001 trend lines.

There, I scooped that one too.

Play ball!

Anthony–
Of course, it is widely understood that the only correct start years for testing IPCC models are either:
1) Sufficiently long ago to ensure that more than 1/2 the data used in the test the projections were known before SRES used to drive the projections were disseminated to the public
or
2) 1999.

Lucia,
Weather is getting better here too [Gulf Islands, BC, Canada]. That said, since the winter 2006-2007 there has been a cooling noticed by many gardeners, progressing by things being late by about one week per year/season. This winter/spring daffs were 3 weeks late, cherry blossoms and hummingbirds likewise, and significantly the herring spawn which happened a week or so ago, 3 weeks late. I know, it’s only weather, but the short term “trend” appears firmly down.

Welcome back!

And thanks for posting the following!

“Do remember: showing trends does not necessarily suggest any particular interpretation. Sometimes statistics are just descriptive. “

The weather hasn’t gotten better yet in Seattle, well briefly, but it didn’t stick. Last week we broke a streak of 140 straight days with a high temp < 60F, as measured at Boeing Field. Now we’re back to highs in the lower 50’s with rain.

Maybe by end of next week we may see 60 again.

Lucia,

You forgot to put the 1979-2000 trend line on your RSS graph, so I’ve rectified that for you.

Here it is:
http://deepclimate.files.wordp.....trends.gif

Notice that the trend to the end of 2000 was 0.13 deg/decade, a little lower than the trend up until present (which stands at 0.16 deg/decade as you noted).

So how can this be? How can the overall linear trend up until the present be higher than the trends in both contiguous sub-periods?

That is part 1 of your assignment. Part 2 is to explain why it matters. Good luck!

Deepclimate–
Is the effect you show supposed to be a surprise? This sort of thing happens all the time. So… why do you think it “matters”? And in what sense does it “matter”? That is to say: What question do you think we are supposed to be even asking ourselves?

Lucia, pay no attention to deepclimate. I believe he is cherry picker activist along the lines of tamino. The bottom line is that the earth is NOT continuing to warm. The best measure is the Ocean heat content which has not accumulated any heat since 2003.

Deepclimate,

Your homework is to compute the trend from 1180BP till present.

You may go here:

http://www.worldclimatereport......re-record/

to find where to access the data.

Please show your work and explain why it matters.

I think that it’s safe to infer from Deep Climate’s home work assignment that he/she is not a mathematician.

As Lucia says, this kind of effect is not unusual. You can see from the graph that the cyan points on the right are higher overall than the red points and so if you add them in they will pull the line up a bit on the right, increasing the slope. Obviously if temperatures continue to decline this will change.

A more interesting question, to me at least, is what is the minimum number of measurements required to create this effect? I think four points might be enough. I might play with this if I get time.

But the main question is why is this telling us anything about anything? Without a hypothesis this is just putting some straight lines over some points on a graph; it’s interesting and might give us some ideas for a hypothesis: a sinusoidal temperature variation with a period of 60 years perhaps?

Welcome back. I have not had my first cup of tea this fine but cool Easter morning. So, my comment is limited to that I hope you and yours have a glorious, and if like my family, a wonderful chocolate bash.

BobB–
FWIW: I think warming will resume. I just think the current trend is inconsistent with a specific set of predictions of warming. The currently “underlying warming” is probably less than projected.

The graphs in this particular post are only meant to describe three particular trends. Deep climate wants me to guess the point he is making and make it for him. The only “meaning” I see in the quite common phenomena he describes is this: Because many points near 2000 happened to fall below the beg-2000 and beg-now trendlines, and the temperatures did not simply decide to plunge both very rapidly and nearly monotonically, the longer trend line ends up with a higher trend.

I could show a similar thing by showing that the trend from 2001-now is more negative than both the trend from 2001-Dec 2009 and the trend from Jan 2009-Feb 2009. I could find other break points. (I’m pretty sure I could move the break point back to around October and show this
‘amazing’ phenomenon which is guaranteed to happen with noisy data.)

I doubt if deep climate would illustrate the 2001-2009 phenomena. If he did I doubt he would ascribe the same explanation to the one he ascribes to the longer trend line.

Lucia, yes the Earth will continue to follow the trajectory of recovery from the last ice age. But there are some who believe that the PDO superimposes a warming -cooling trend on top of the Natural warming trend:

http://icecap.us/images/uploads/akasofu.jpg

Bob B,

Thats a decidedly… odd graph. I’m quite curious who figured out that the “trajectory of recovery from the last ice age” turns out to be around 1 degree C per century, especially since a 1 degree rise corresponds with the roughly 1 degree forcing per doubling CO2 that (most) everyone knows will happen. Not to put words in Lucia’s mouth, but I believe her position is that:
.
1) CO2 and other anthropogenic GHGs are warming the Earth’s surface,
2) a doubling of CO2 ceteris paribus would very likely increase global temps by ~1.2 degrees C due to basic radiative absorption physics, and
3) there is considerable uncertainty in the magnitude of both positive and negative feedbacks (as well as natural variability in the climate system) that may lead to either a low or high climate sensitivity, but the divergence between observed and modeled temperatures suggests that (in Lucia’s opinion) climate sensitivity may be on the low side and/or natural variability may be underestimated.
.
Its a good starting point for some genuine discussion on a topic that all to often is dominated by the strawmen on both sides of the isle.

Bob B,
I’d be interested to see the sources for that graph too.

IIRC the only temperature measurement series that we have that goes back to the LIA is the CET series which showed a warming trend of around a quarter of a degree C per century when I did on OLS fit to it from 1650 to 1950.

I’m not suggesting that there is a great deal of value in looking at the CET. I’m just curious to know where the steeper line in the Bob B curve comes from.

The graph come from Dr Syun Akasofu
(2009 International Conference on Climate Change, New York, March 2009).

You can download his presentation here:

http://www.heartland.org/event.....dings.html

You can get Akasofu’s stuff from http://people.iarc.uaf.edu/~sakasofu/ but he does tend to use huge pdf files, so don’t click on links there unless you have a fast connection!

Looking at the graph in the context of the rest of the presentation seems to confirm that this is simply 120 years of historical data extrapolated forward and backward by 100 years. At least the lines are dotted! I guess the motto is, why extrapolate one way when you can extrapolate in both directions?

Zeke–
That’s about right. I’d add that I’m open to discussing any and all reasons for the divergence between the models and the data. However, I am don’t believe the explanation for the temperature increase during the 20th century is rebound from the glacial period or even the little ice age. I admit I can’t prove that theory wrong, but it strikes me as less plausible than the theory that it was mostly caused by greenhouse gases.

Maybe extrapolating from a reconstruction? I would think one could use proxy data and draw a line to the present day?

http://www.worldclimatereport......re-record/

Bob B,

Eyeballing your graph of Craig Loehle’s proxy reconstruction I see a rise of roughly 0.25C per century between 1600 and 1900. This rate of increase matches the CET that I mentioned earlier but is a long way from the extrapolations that you showed in a prior link. This quarter of a degree per century might explain between quarter and a third of the warming in the 20th century. What do you think explains the rest?

I think the present cold is a sign of a changing regime:
Up to 2008 the 1997-2007troposphere trend was very similar with the development of the 1972-1983 El Nino, compared with the nino34 index.
The 1983 el nino was maked by El Chichon.
The comparable 1983 El nino did not materialise in 2008/9.
see my “forecast”, based on historic nino34 extrapolation.
http://home.casema.nl/errenwij.....recast.gif

Chris, Surface station data is pure crap. The UHI effect on surface stations accounts for 1/2 the remainder and data manipulation from the keepers of that data account for some. Without the crappy surface data we really don’t know what the small recent warming over the last 100yrs is do we???

Pure crap? I don’t think it’s quite that bad! Up to 50% of the land surface data is perhaps due to UHI contamination but that is less than half the total surface. While we’re engaging in guesses, 15% to 20% might be a more reasonable guess for the extent of UHI contamination.

I’m not the one showing graphs that purport to show how much it has warmed over the last 100 or 400 years. And if we don’t know how much warming there has been, how do we know it has been small?

I’m not for or against AGW but I am against overconfidence on both sides of the argument.

May I add a new term to the AGW debate lexicon: Quaverer

A Quaverer is somebody who once was certain about the accuracy of the IPCC’s predictions but is now wavering. I could have used Waverer but I think their voice is more trembling as they bit down on some humble pie. And anyway in the tradition of this debate the name needs to be deliberately insulting.

Here are a couple from today’s Sydney Morning Herald (one of the main Australian broadsheets and usually home to AGW certainty)

Marian Wilkinson, Environment Editor writes about how “the rise in extreme temperatures in Australia because of climate change may not be quite as big as some feared”, after removal of the more obviously in error models so that “When the poorly performing models were removed, the predicted rise in temperature extremes was between 2 and 3 degrees by the end of the century rather than 3 and 5 degrees as the modelling had predicted, according to the study published in the journal Geophysical Research Letters”.

http://www.smh.com.au/environm.....-a0w7.html

Paul Sheehan an opinion columnist, who is usually quite certain, writes about Ian Plimer’s (well know professor of geology and AGW skeptic) new book Heaven And Earth. He starts by saying; “What I am about to write questions much of what I have written in this space, in numerous columns, over the past five years. Perhaps what I have written can withstand this questioning. Perhaps not. The greater question is, am I – and you – capable of questioning our own orthodoxies and intellectual habits?”

And ends with; “Heaven And Earth is an evidence-based attack on conformity and orthodoxy, including my own, and a reminder to respect informed dissent and beware of ideology subverting evidence.”
http://www.smh.com.au/opinion/.....ml?page=-1

Surely the correct term for these quaverers is heretics, while us disbelievers and doubters are merely pagans and agnostics!

Andrew Kennett: It is interesting that coming down to (a still unlikely) 2-3 degrees/century would be considered a big deal.

The lukewarmist position has always been the most scientifically grounded. In contrast, the world of magically positive feedbacks has always been more speculative and ideologically driven.

It has been warming for 300 years. In the last 50 it has warmed only slightly faster. This would seem to indicate to a non-ideologue that on the whole, a doubling of greenhouse gases would contribute about or slightly less than the (unamplified) 1 to 1.5 degrees predicted by the simple physics of the matter. And (not to get all Panglossian) this warming is happening at a time when we are probably due for a temperature dip (if the drop in solar activity is as meaningful as some say it is). So far from disastrous, it may be happening at just the right time.

The more interesting mental gymnastic will be how former alarmists claim to still have been smarter and more correct than the lukewarmists even after the alarmist scenario collapses into (largely fortuitous) modest warming. My guess is that the operant meme will be one of moral superiority: the demonstrated willingness to save the planet (even if based on a false premise combined with utterly ruinous plans of action) is the true mark of the modern Illuminati. Being factually correct does not excuse being unenlightened.

Deep Climate,

Now plot the trend from 1979-1997

My point is that all measures of long term trends are higher now than they were ending in 2000 or 2001 (while the short-term trends have fluctuated wildly).

RSS:
http://deepclimate.files.wordp.....trends.gif

NASA – GISTemp:
http://deepclimate.files.wordp.....rends1.gif

The 80-99 baseline metric is interesting. It is based on IPCC metric for projection to 2011-2030 which uses a 20-year moving average (see chapter 10 Executive Summary). In GISTemp, nine years into the 31-year projection period, it stands at 0.17 deg C per decade.

So let me rephrase my original question to Lucia. Before I brought this to your attention, were you already aware that all the long term trend measures of global temperature change were higher now than they were at the start point of your analysis, whether it be 2000 or 2001?

DeepClimate

were you already aware that all the long term trend measures of global temperature change were higher now than they were at the start point of your analysis, whether it be 2000 or 2001?

All? I guess I’d have to ask you to define “all” since I can’t claim to be familiar with “all” long term trend measures. However, if you mean was I aware that the trend from 1979-2000 is lower the trend from 1979-now, the answer is yes.

I’m not sure what you mean by short term trends fluctuating “wildly”. Shorter term trend are nearly always more variable than longer term trends. This is widely recognized around here. Is there some other point you think automatically follows from this factoid?

Generally speaking if you wish to be understood, it is better to ask question directly and make points directly. If find that spouting factoids without stating the point you think they support is often just a waste of time.

I was referring to the long-term trends in the graphs I provided. You state that you were already aware that the trend from 1979 forward is higher now than it was in 2001. Fine. I’m suggesting that it is relevant piece of information to display in your graphs. Apparently, you disagree.

Two other referenced long term trends are:
a) 20-year linear trend
b) Decadal average difference (between latest and previous decade)

These trends are also higher now, in both data sets referenced, than they were in 2001. Were you aware of that?

To me, the point should be obvious. If long term trends have actually increased somewhat during the 2000s (and they have, despite a slight dip in 2008), then statements or inferences to the effect that global warming has “stopped” or “stalled” are unfounded.

Finally, the IPCC metric (20-year moving average, baselined to 1980-99) for its projection of surface temperature increase to 2030 stands at 0.17 deg C /decade in both HadleyCRU and GISTemp as of the end of 2008. This is hardly an abject failure of the models as some have claimed.

I trust that my position and questions are clear to you now.

DeepClimate–
I could provides a zillion possible trends:

I provide the longest term one because it’s longest.
I provided the since 2001 because that’s the trend I always discuss. I provide the trend since 2000 because there have been some who suggest that it’s important to use ‘00′ dates for some reason of they feel is important.

I don’t think it’s particularly important to add the trend from start to 2001 nor the one from start to 2000. In my opinion, the very commonly occurring artifact you discuss means practically nothing. But if you wish to show add these extra lines to graphs at your blog, or mention them, I have no objections.

If your concern is this:

…then statements or inferences to the effect that global warming has “stopped” or “stalled” are unfounded.

I have never suggested global warming has “stopped” or “stalled” (whatever the latter might mean.) The recent negative trends are not evidence that global warming has stopped. I’ve said this many times.

I have stated that, as far as I can determine the recent trends are evidence the current underlying trend must be lower than projected by the IPCC. As they have projected higher trends than observed in the past, I don’t see how your preferred graphs would contradict anything I have claimed.

George Tobin [12741]

Unless of course, the “temperature dip” you refer to is actually a absolute drop in temperature following the absence of statistically significant warming over the past decade.

As a child I was taught that it is wrong to lie, even for a good cause. It seems to me that to derive a sense of moral superiority from lying for a “good cause” is hardly a sign of illumination. If my preference for sticking to verifiable data and facts over lying to demonstrate my commitment to “save the earth” means I’m a non-illuminated clod, so be it. I sleep better that way.

Deep Climate-do some significance tests on those to trends, and then come back and tell me we are warming faster than we thought since 2000. There is essentially no difference there.

Deep Climate

It might be helpful to all of us on this thread if you could remove your hand from the lower portion of the thermometer you are holding….

DeepClimate

then statements or inferences to the effect that global warming has “stopped” or “stalled” are unfounded.

I don’t know what the word “stalled” would even mean in this context. However, I have always maintained that global warming has not “stopped”. The recent negative trends are consistent with warming.

I don’t know why you think the other graphs somehow needed to be in this post. Possibly, you think every blog post every where must at all times only show that particular information. But I don’t see why that should be so, nor do I see how doing so would reduce the number of people who believe global warming has stopped. (Or “stalled” whatever “stalling” would mean.)

The recent trends are low. As you can see by the uncertainty intervals that I show on my graphs, the current trends do not exclude “underlying warming” right now, if we believe the weather noise is AR1 and normally distributed. (AR1 may not apply, other statistical methods are possible– but this is the assumption in Santer. )

I recognize that you want to make this point your way with graphs you think are spiffier, but I prefer to make the same point my way with graphs I think spiffier.

Tetris–
Behave.

Lucia [12751]
My apologies. Having to stoke our wood stove on April 13 [!] must have got to me.

Lucia: You have expressed your views on global warming several times, and my own are close to yours. In any case I was wondering what would make you change your views, either way? Do you wait for 30 years to see what transpires? I ask myself this question and have no really strong answer. Maybe other posters would express their views on this. You see, many areas in science are easily falsified. All it would take is one verified counter-example to disprove a theory. This does not appear to be the case with climate science. Anyway, just wondering.

Jack

jack mosevich [12754]

I look forward to reading Lucia’s reply to your question.

As far as the issue of science is concerned, the overarching problem we are faced with is that all assurances to the contrary notwithstanding, “Climate Science” does not qualify as science [as e.g. , geology, astronomy, physics, chemistry or biology].

This is so for several fundamental reasons. As much as its proponents would like us to believe this, it is impossible to provide verifiable proof for the original hypothesis that “Man-made CO2 emissions are causing an increase in global temperatures”. We can not correct for multi-variate natural fluctuations in the climate system and isolate the key variable. Reality is that there is not even a statistically significant correlation between rising CO2 concentrations and global temperatures, let alone the basis for a demonstrable causal relationship. On top of that, as posited the hypothesis is in fact not falsifiable and therefore, lies outside the realm of the scientific method.

When it became evident some 5-6 years ago that real world data did not provide support for the original hypothesis, several things happened. The cherry picking and “adjusting” and “re-calibrating” of the data intensified and the hypothesis was modified. Not to conform to the verifiable data mind you, as required by the fundamentals of the scientific method, but instead it was morphed to state that “man-made GHGs are causing global climate change”. That is gibberish, because the scientific methods teaches us that if it is impossible to prove that variable “a” causes a change in variable “z”, then it follows that is impossible to prove that variables “a”, “b”, “c”…..”f” amalgamated are causing a change in variables “g” through “z”. Ensemble “g” through “z”, it might be noted in passing, happens to be the most complex, non-linear, chaotic system known to man. Whatever that change is, it’s not science.

Also quite remarkable is that a large number of key “peer reviewed” iconic papers that are held up as “proof” of AGW/ACC were published without the inclusion of indispensable information, that in any chemistry or biology paper would be found in the “materials and methods” section. The absence of this information [source data, algorithms, etc.] is makes it impossible for anyone else to verify the research and its conclusions. That is not science, but an insult to our intelligence and the accepted norm in what passes for “Climate Science”. If a paper of the kind produced recently by Steig, Mann, et. al. purporting to show a warming trend in the Antarctic were submitted in the life sciences, it would be rejected out of hand, both because it had an empty materials and methods section and because its conclusions were entirely based on statistical extrapolations. As Trenberth acerbically noted when the paper was published: “you can’t make data where there is none”. But in “Climate Science” you can. Just ask Mann, Jones, Amman, Wahl, Hansen and the list goes on. Not only can you, and get away with it, it secures your next grant.

There are many more examples of this travesty, and until it is subjected to a rigorous clean up, it is not difficult to argue that “Climate Science” as currently practiced is pseudo “science” on par with tarot, astrology and homeopathy.

Jack and Tetris–
I’ve actually answered that question in comments from time to time. The hardest part of answering is that there are an infinite number of things that could happen that would change my mind. So, I then need to sift through the border line plausible ones. To give an answer that places the treshold at ‘just the right place’ while giving answers that don’t sound weaselly requires some analysis on hypothetical things.

However, to start:

1) If weather events that are extremely improbably under the hypothesis of warming where to occur, I would decide warming isn’t happening. What might these be:

If we don’t experience a volcanic eruption, and the annual average global mean surface temperature drops 2 whole degrees C next year, I would doubt continued warming. (Yes, even if it’s “just one year”.) Needless to say, this is so dramatic, I would also begin to wonder what the heck is going on!

But… that’s the sort of vapid answer I know won’t satisfy anyone. Let’s face it: Everyone would stop believing it’s warming Chicago ended up under an ice sheet in 10 years. But clearly, rational people would require less evidence of ‘non-warming’.

So, to be more fair:
If the trend over the next 30 years ends up being zero warming while present levels of CO2 are maintained, no super-huge mega-volcanos larger than Krakatoa start erupting and we can’t plausibly pin the blame on aerosols, I’m pretty sure any calculations I did would convince me that ghg’s don’t have much warming effect.

What if it’s 20 years? I don’t know. I’d have to look at the observations to see if it turns out “weather noise” is larger than I currently think based on historic measurements.

Not withstanding what other people think, I do not believe that we can have a 20 year long trendless state if the “underlying warming” is 2 C/century. (Note: Underlying warming is not precisely the same as “long term trend”. The underlying warming is the typical or average rate we would expect if we had an infinite number of honest to goodness real earth’s could subject them to similar forcings and find the average rate of change in temperature at any instant. This value will be a function of the forcings.)

I would have to do computations to be absolutely sure, but I suspect we 20 year long zero trend with is not consistent with 1 C/century warming. But…. I’m not sure. (And once again, if the current period is,for some reason, ridiculously calm “weather”, then I could change my mind.)

GISS has just been updated. It is up 47 which is higher than Feb and lower than Jan

Jack,

I´m not sure that the question of AGW will be answered satisfactorily however long we wait.If natural temperature variations can have frequency components that stretch beyond centuries one cannot say that any current trend will continue up or down. Until recently natural variation has been largely dismissed on the grounds that the amplitude is small and the frequency is high enough that it can be called weather noise. It is supposed to average out over fairly short time periods.

The current hiatus is a very good thing as it shows that natural variation is not so easily ignored. With luck, we will see much more investigation into the causes of natural variation and this may allow us to see what residual is left that needs some further explanation such as increased CO2.

I am moving slowly away from the idea that we know what increased CO2 will do to climate/temperature. I started off very much like Lucia in thinking that the “blanketing” effect would have to lead to a rise in surface temperatures if the OLR was to be brought back into balance. In essence it is treated as an energy balance problem which can be modelled by something like “Lumpy”. Bit by bit I have come to think that the supermodels are really just like a stirred Lumpy with some spatial and temporal averaging.

I do accept that a radiative energy balance is required in the long term but the critical point is that this balance must happen at or near the top of the atmosphere. I also accept that radiation codes like Modtran can probably calculate the change in OLR with a change in CO2 if all vertical temperatures and atmospheric compositions, including clouds, are known. Spatial and temporal averaging is a challenge given that the vertical structure is always changing in time and place.

What is not clear is whether the changed vertical profile that is needed to restore the OLR involves a change in the surface temperature. This involves the lapse rate and that is not just a radiation calculation. Convection, latent heat and changed cloud cover all come into play. Until the mysteries of the lapse rate are properly understood we just cannot say what the effects of a change in OLR might be.

What will convince me one way or the other is more science rather than just a longer sequence of temperature measurements.

As Lubos pointed out this past March is the coolest since 2000.

Tetris and Lucia: Thanks for your replies. I think it is unfortunate that Hansen and others have not discussed this topic just for the purpose of scientific integrity and intellectual curriosity.etc. I guess I worry that so many climate researchers seem to believe in the models and in our understanding of climate to the point that an orthodoxy has developed. I could of course be wrong on the latter point; maybe they do acknowledge doubts but not very publicly.

DC: the RSS overall trend has been decreasing over the last year or so that I’ve been following it.
As Bob says, your cherry-picked data manipulations, selections of meaningless ‘trends’ and unpleasant tone are very reminiscent of tamino.
To be fair I also think Lucia’s fig 1 is pretty meaningless.
You don’t need to draw any lines, straight or polynomial, to see that the warming seems to have levelled out.

[...] Deep Climate challenged me to look at longer trends, I thought I might as well do look at them. So, I [...]

Paul–
I didn’t suggest great meaning in those particular lines. I show them not because they tell use “more” than the graph without the lines. I show them because I periodically discuss those specific trends, so it’s useful to have them there when other discussions spring up.

Lucia,

http://tamino.wordpress.com/2008/01/31/you-bet/ was a reasonable approach to determining if no warming or the long-term trend is more likely. As far as being convinced that -no- warming is occuring, thats much harder!

Zeke–

My main objection for that as a method is that it’s slower than necessary while also setting things up as either/or.

Who wins the bet doesn’t help with other questions like: when do we decide if warming slowed down a bit? Of if it’s at the IPCC projected level?

Still, certainly, if there is no warming, the no-warmer would have a good chance of winning that but it will take at least two years for anyone to win.

Im willing to say that temperatures will decline for quite a while maybe 2-10 years and AGW will be officially terminated within 2-3 years thank you. Ol soldiers never die they only fade away LOL

[...] In comment 12745, Deep introduced the notion that we could learn something by noticing that 20 year average trends and trends based have been increasing. I pointed out that I was aware they are increasing, but at a rate less than forecast by the models underlying projections in the AR4. [...]