Eli Rabbet posted a link to a brief pamphlet recently published by the UK Met Office. The contents can also be viewed as individual web pages. When I clicked to visit the UK Met Offices Fact 2 page, I read the title “Temperatures are continuing to rise”, and saw this graph:
The main point of the graph is to show readers that “Temperatures are continuing to rise”.
Naturally, I immediately visited the current information on global surface temperatures available at the UK Met Office, where I saw this:
The copyright appears to be 2008. The technology for web pages permits quick updates. I wonder if a 2009 version will be published soon? I assume the discussion will be revised to explain the non-rise at the tail end of the graph.
(BTW: Given that the seven page document appears to be some sort of official document published by an important agency, there are some weird issues with the thing…. but let’s not nitpick.)
Maybe the MET will put this on a coffee mug? Or a tea mug, as it were.
Hm-m-m…
Maybe the MET is trying to divert attention from this forecast:
http://www.metoffice.gov.uk/corporate/pressoffice/2008/pr20080925.html
Lucia,
I like your use of the oh so diplomatic “non-rise”. As near as I can tell the second MET graph shows a global temperature drop after 2000 of 0.2C [from 0.5 down to 0.3] with a 95% confidence range. Since the IPCC tell us that this is as good as certainty, that is a significant decline I would have thought.
Eli Rabbet should really make an effort and actually read the material he sends around.
DougW–
The copyright on the pdf says ©Crown copyright 2008 08/0112.
I’m not sure when the Met office posted the page.
The item reads like something created to present to children in school. There are no citations. The sources for information are not provided, and are very ambiguous.
Example, figure 1 for Fact 3: says vostock ice core. So… what’s the source fro the green line to show present day? What’s the source for the projected trend using the red line? Now.. how about figure 1 in “Fact 5”. This is the figure Eli found a small bit to complain about. He thought people won’t know what the A2, B2 and B1 emissions are– and he’s right. But people also won’t know which of the many possible models is indicated by light blue line. And, since they start in 1990, why don’t they show how that projection compares to data? Are they worried it would “confused” 3rd graders might ask why the data are (or are not) below the projection? (Most projections are too high by 2007…. so unless they picked one of the few that is not, the 3rd graders might notice this.)
Oh well… the audience must be 3rd graders. So, I guess missing these things is ok. Not providing any references is probably fine too. After all, why should glossy pamphlets from the Met office be expected to meet the standards of the average blog post?
Could be , but I think they think of the rest of us as third graders.
And they cover the tail (their tails?) with the following:
BarryW
The La Nina “reasoning” comes from Phil Jones at the Hadley/Met’s affiliate at the University of East Anglia’s Climatic Research Unit [CRU]. I say “reasoning” in quotation marks because as Pielke Sr and others have accurately pointed out, La Nina conditions can not “mask” anything because La Ninas and El Ninos are an integral part of the atmospheric system.
And as far as treating their readers as 3rd graders, Pielke Sr has also convincingly shown the HadleyCRU to have become an unabashed advocacy-based organization in terms of how they present their AGW/ACC information. As they say: BS baffles brains. Biased BS even more so..
“Continuing to rise”
I wish
Well, it is true that if there is no underlying trend, the earth’s surface is warmer during El Ninos and cooler during La Ninas.
But the difficulty for the MetOffice is that it’s natural for smart-a-lecky third graders to do this:
1) Read the “facts 2” section.
2) Count the number of bars after 2000 and notice that 2008 is not on the graph.
3) Ask their parents or teacher to look at the up to date information.
4) See the more recent graph with the dashed line implying an actual drop in temperature.
So, how will words explaining slight slowing of the warming help the teacher confronted with the smart-a-lecky third grader, who is likely to keep saying “But the dashed line goes d_o_w_n.”
As written in their educational materials, the text of the “Facts 2” section appears to rely on the idea that the graphics will just speak for themselves. But now that the 2008 data are fully in, if we let the new, up to date graph speak for itself, it appears to to contradict the idea that temperatures continue to rise!
Given that the document was written in 2008, and the authors already knew 2008 was coming in cold, and they’d already had to fix their graphics once, they should have been more cautious writing their text. It might have been wise for the authors at the Met office writing the “Fact 2” section to acknowledge the possibility that, based on their understanding, their smoothed curves might show actual dips and still not contradict the idea there is warming.
But they didn’t include any such text. So…. now we have this poorly written “Fact 2” page which is just sitting there not explaining why the brief period of non-rising temperatures doesn’t contradict warming!
Lucia, It appears red line you asked about comes from the WRE Stabilization Profiles (1000). My guess is the current data comes from Mauna Loa. I wouldn’t recommend this document as an introduction to this subject. I can understand that to effectively communicate any branch of science to the public, it is necessary to dumb things down. You don’t throw undergrads material that’s appropriate for grad students. But it wouldn’t have killed the authors to add a comment or two on the distinction of weather and climate. Fact 2 is a bit too dumbed down. I’ve even seen an explaination of the greenhouse effect using units of W not W/m^2. I thought the author made a mistake so I asked him about it and he said he used watts to avoid confusing people.
Lucia,
I wrote to Hadley asking about the 2008 update to their graph when we were wondering what it would look like. The new version appeared shortly after 🙂 They even sent me an email to say it was done.
Did you notice that the educational version chose a very early period for the baseline? It makes the anomalies around .2ºC higher and most of the years appear to be above “normal”. This makes no difference to sophisticated readers like us but the visual impression definitely changes.
Chad—
Ahhh! Maybe so. I never doubted the red line came from somewhere. But it’s odd that the Met office would create what appears to be a glossy handout without giving any references. The level of ambiguity falls well below hobby-blog-commenter level!
One of the problems is that the document is simultaneously too dumbed down and not dumbed down. If it’s for 3rd graders — or even their teachers who may not be climate blog addicts, why do the authors use the cryptic “A1, B2, B1 and A1F ” type scenarios in one of their “Fact” discussions? If it’s for high-school students, why no references at all? How are they supposed to learn what those scenarios even mean? Oh well. . .
How does using incorrect units not confuse people?
Oh well.
Jorge—
Yes. I noticed the change in baseline. Again: This will confuse third graders — the group who I assume this is aimed to teach.
There was a discussion some time ago about telling a lie by saying something that is true. Since we don’t know what the temperature should be, each graph implies a different “truth”. Someone who is unsophisticated in the manipulation that goes on and what is really being expressed by the graph might draw the conclusion that the zero point on the y axis is what is the “correct” temperature. The first graph can be read as the twentieth century is much warmer than what the earth’s temperature should be , while the second can be read as the nineteenth was much colder than what the temperature should be.
I found it interesting that Fact 1 seems to have been toned down on the website compared with the pdf:
From the pdf
http://www.metoffice.gov.uk/climatechange/guide/downloads/Your%20Guide_Facts_The%20Big%20Picture_pdf.pdf
There is indisputable evidence that this observed global warming is predominantly caused by increases in greenhouse gas concentrations. Concentrations of carbon dioxide (CO2), created largely by the burning of fossil fuels, are now much higher, and increasing at a much faster rate, than at any time in the last 600,000 years. Because CO2 is a greenhouse gas, the increased concentrations have contributed to the recent warming and very likely most of the warming over the last 50 years.
So how did ‘indisputable evidence’ in the first sentence turn into ‘very likely’ in the last? – but on the website, it doesn’t say that at all:
http://www.metoffice.gov.uk/climatechange/guide/bigpicture/fact1.html
There is indisputable evidence from observations that the Earth is warming. Concentrations of CO2, created largely by the burning of fossil fuels, are now much higher, and increasing at a much faster rate, than at any time in the last 600,000 years. Because CO2 is a greenhouse gas, the increased concentrations have contributed to the recent warming and probably most of the warming over the last 50 years.
The copyright on the second chart says 2009 and it does appear to incllude 2008.
Norman–
Yes. The second chart is the current version and includes 2009. That’s, in fact, part of my point.
The current 2009 version is not included in the UK Met office’s online discussion of their “Fact 2: Temperatures are continuing to rise”. In their online discussion, the UK Met office uses the graph without the 2008 temperature data to explain “Temperatures are continuing to rise”.
That’s my point.
Ah, but, if I read the website right, the downward dip in the dotted line results from assuming that 2009 onwards are the same temperature as 2008. Given that the ENSO index (http://www.cdc.noaa.gov/enso/enso.mei_index.html) is unlikely to remain negative for several additional years, nor is the solar cycle likely to remain at the low point of the cycle (and there’s a 0.2 W/m2 peak-to-trough amplitude swing from normal solar cycles, which is about 12% of currently calculated GHG forcing), it seems unlikely that the next several years will be as cool as 2008.
Of course, this all depends on what your definition of “is” is. “continuing to rise” could be disproved by one cool year by one definition, or by taking the trend from 2001 to 2008 by another.
Personally, I would prefer the graph without the dotted line extrapolation… given that there isn’t a good way to “extend” our assumption about future years, let the data stand on its own.
Marcus–
Yes. The dotted line depends on some assumptions. The UK Met office also made assumptions in the first graph, but didn’t use a dotted line. They switched to the dotted lines when those assumptions began to show dips.
Many who comment here think it would be better if the UKMet just ended their smoothed line when they needed to begin using fake future data to create the smooth line. But… well.. they don’t!
I’m sort of waiting to see if the UK met office decides they need to ‘correct’ their assumptions for creating fake smooth lines (as they did last year when a stupendous dip first appeared sometimes after January.)
A deliberate distortion. Fear of muddying the AGW waters. Fear of confusing people – not just 3rd graders – but most everyone who hasn’t the foggiest clue about equilibrium climate sensitivity vs. short-term weather noise. Fear especially of confusing media people who prefer to provide ultra-simple messages for their readers. The sophisticated reader – the one who wants to know more – gets shafted. It happens every day at Real Climate. Someone looking for a real answer to a complex question, told to shut up and go learn their catechism.
Mmmm, gotta love the taste of advocacy.
Would it also be correct to say: “The recent more frequent La Nina conditions are due to cooling since 1998.”
“…recent slight slowing of the warming…” pretty much sums up the alarmist position.
It’s the scientific version of “a little bit pregnant” when faced with incovenient data…
Yes, the whiff of advocacy. How it pervades.
I don’t think “reeks” is spelled p-e-r-v-a-d-e-s
“Would it also be correct to say: “The recent more frequent La Nina conditions are due to cooling since 1998.—
I believe that the climate -> La Nina/El Nino link is still an active area of research. It would be extremely surprising if changes in global mean temperature didn’t impact ENSO in some way – but the question is whether we can predict a priori the direction or magnitude of that impact.
On the other hand, it seems pretty clear that ENSO impacts global mean surface temperature fairly significantly.
And then there’s the PDO and AMO and other poorly understood acronyms…
It took them till today (2nd March) to add the year-to-date point for 2009 to the 2nd graph on this page (CET), and till about 18th Feb to make the January data not provisional (yet February has been finalised already).
http://hadobs.metoffice.com/hadcet/cet_info_mean.html
Maybe there is a reluctance to present data which the maintainer believes is misleading (and granted the error bars from a single month are fairly high).
Ms. Lucia, Ms. Lucia, oooo, oo, ooooooo, ooo, ooooo, pick me, pick me!!!!,
uuhh, yes Ms. Lucia, why does the ifrst graph go all the way up to 1.0, and the second graph only go up to 0.6???
Aren’t they both from the same source data?, if so, how can the years after 2000 vary so much in temperature?—And, around 1940, the temperature on the first graph shifts up by fully two tenths of a degree, is this “anthropogenic” in origin?—And, if so, which anthropogene did it?
Thank you Ms. Lucia, in advance, for your clear exposition.
Gaelan–
The shift in temperatures you see has to do with the “anomaly method”. When using this, you pick a set of years to use as your baseline. You find the average temperature for those years and subtract.
In the figure used to explain “Fact 2”, they seem to have made the baseline the one year 1900. However, the standard HadCrut baseline is the 30 years from…. Hmm.. I can’t remember if it’s 1950-1980 or 1960-1990 or what. But it’s later in the century. Anyway, that’s the reason fro the major shift.
The “anomaly method” is a critical step not just in the temperature record but in the justification of the abilitiy of the GCMs to describe and predict climatic changes.
If the “monthly” temperature record was expressed in absolute terms it would vary by about 4C between Jan and Jul about a mean of about 14C.
Removal of this effect is a prime reason why anomalies are used.
In the case of the models it could be seen to have an additional role. It highlights the similarity between the model trends and the record without drawing attention to the inability of the model to produce the 14C mean value. In the CMIP2 archive this varies from ~11.5C to ~16.5C.
As I understand it, this discrepancy is considered unfortunate but not damning. Also, as I understand it, this and other failures of the models to get the climate (not just the weather) right does not prevent the models from describing the anomalies in the temperature record and predicting climate change.
Ironically it could be argued, (I am not saying any one does) that the failure of the models to describe climate accurately, each one doubtless in different ways, strengthens the argument that they can predict future temperature increases, as it is something that they all agree upon. I.E. no matter how bad the model, it will always predict rising temps with rising CO2.
Alex
I apologise if this is OT but I would reallly like to see the temperature records (we have more than one world it seems) expressed in absolute terms and comparison between them and the model outputs performed in absolute terms. It raise a few eyebrows and open a few eyes.
When one model indicates a temperature rise of 2.5C (to 16.5C) at so many decades hence, and another model says too late we have already got there, I find it a bit of a worry.
I do not know if the spread in the CMIP3 data is as broad as I have not seen the data expressed in absolute terms.
Alex
Alex–
The reason all AOGCMs predict rising temperaturea with rising CO2 is that you don’t need an AOGCM to predict this. All other things being equal, rising CO2 must result in at least some rise in temperature. So, even very poor models must predict some rise.
The question is: Are AOGCM’s any better at getting the sensitivity than simple 1-D radiative convective models? Are they any better at predicting temporal evolution (which required getting mixing in the ocean correct?)
The things the large AOGCM’s predict robustly (i.e. things they all agree on) are behaviors that can be predicted without the need for complex models. To use simpler examples: Suppose you have a 3ft deep pool whose floor is lined with a black rubber mat. Do you need a computer model to predict the black mat absorbs heat? Do you need a computer model to predict the hot water will rise, cool when it reaches the top and then the cool water will sink?
So, if 7 modelers ran codes and predicted this only, but all were wrong on the actual surface temperature of the black rubber mat, and not quite right on the details of the circulation pattern, how confident would you be that they could predict the temperature of the mat if the sun’s heat doubled? Etc.?
To some extent, the argument advanced for why we should believe climate models are “accurate” is similar to my pool with the black rubber mat hypothetical. Yes: CO2 will cause some warming. Yes. Models are sufficiently good to capture the qualitative features which we would know even without running a model. But are the models bang on accurate in forecasting the future? Who knows?
Alex– I can do that for the models. It will be in annual average temperature because that’s easiest to compare to the data.
Why does it seem like the reporst show a steady, straight-line increase in trace gas concentrations of carbon dioxide, while at the same time the global temperatures for the period since 2000 (approximately) show a decrease?
Or is reality not even a very poor model?
Catechism says this is the difference between equilibrium sensitivity vs. short-term (i.e. subdecadal) departures from equilibrium. It’s proof of more “warming in the pipe” as the planet increases in radiative imbalance. Catechism says such effects are not inconsistent with AOGCM predictions, that these are mere transients on the way to a deterministically warmer future.
Well said, lucia. When it comes to GHG forcing effects on GMT, the AOGMSs only say what the 1D models tell them to say.
Leif Svalgaard better not read this pamphlet as he would lose it at page 5. Isn’t that Lean’s TSI graph – the one that includes background TSI and was used to suggest TSI did have a reasonably large effect?