The world is full of climate dashboards (and dashboards of dashboards), and so you might imagine that all datasets and comparisons are instantly available in whatever graphical form you like. Unfortunately, we often want graphics to emphasize a particular point or comparison, and generic graphs from the producers of the data often don’t have the same goal in mind. Dashboards that allow for more flexibility (like WoodForTrees) are useful, but aren’t as visually appealing as they could be. Thus, I find myself creating bespoke graphics of climate and climate model data all the time.
Some of these are maintained on the Climate model-observations comparison page but many of the graphs that I make (often to make a point on twitter) aren’t saved there and often their provenance is a bit obscure. Given that twitter will not last forever (though it might be around for slightly longer than a head of lettuce), it’s probably useful to have a spot to upload these graphics to, along with some explanation, to serve as a reference.
I have therefore created a couple of ‘pages’ (in wordpress speak) with fixed URLs where I will be curating relevant graphics I make (and findable at the bottom of the page under “DATA AND GRAPHICS”). The first is focused on the surface temperature records. I often update relevant graphics associated with this in early January (when we get another dot on the graphs), but there are associated graphs that I’ve made that don’t make it into those updates, so this is a place for them too. This includes the impacts of ENSO, comparisons across different platforms, or the impact of homogenization.
The second page is bit more eclectic. These are graphs that are relevant to some trope or talking point that often pops up, and my graphs are an attempt to provide context (usually), or to debunk it entirely. This is where you’ll find maps of where the climate is warming faster than the global average, time-series of river ice break-up dates, and an example of sensible scaling of CO2 changes and temperature.
To start with, I’m just going to upload some graphs I’ve made recently (with any updates that are needed), and I’ll add content as I make something new. If there are any other ideas (that aren’t too involved!), I’ll be happy to look at adding those too. Let me know if this is useful.
That’s great thanks Gavin.
I think a better graph to show the relationship between changing CO2 and changing T (the first one on the second page) would be:-
https://www.woodfortrees.org/plot/esrl-co2/isolate:60/mean:12/scale:0.2/plot/hadcrut4gl/isolate:60/mean:12/from:1958
KW: “,i>I think a better graph to show the relationship between changing CO2 and changing T ”
I don’t know about your: “better“, and about your: “show the relationship the changing CO2 and changing T “.Here is why:
Unless I missed something, the two variables shown on your graph seem to be …. short-term departures from … long term trends of Co2 and Temperature, respectively. If correct, then this graph removes the signal (of AGW) and leaves only the (short-term) noise in its stead.
So to borrow a phrase: this is a graph that only its parent could love … ;-) i.e. whoever plotted it, probably for a classic climate change denial: using the short-time noise to discredit climate (i.e. walking average over, say, 30 years) relationship between rising CO2 and rising T. .
So whoever recommended this plot to you as “ a better graph to show the relationship between changing CO2 and changing T” – tried to make a fool of you.
As George W. famously put it: “There’s an old saying in Tennessee — I know it’s in Texas, probably in Tennessee — that says, fool me once, shame on — shame on you. Fool me — you can’t get fooled again.“
Fascinating graph, Keith.
As Piotr says,, someone has tried to make a fool out of you. Notice that the first “processing step” is “Isolate(samples)”. If you look into WoodForTrees’ help menu, you’ll see that the “isolate” processing is explained as “Does the same running mean as ‘mean’, but then subtracts this from the raw data to leave the ‘noise'”
In other words, it removes the long-term trends. It’s a clever form of hiding the long-term relationship between CO2 and temperature.
It reminds me of the old math joke (yes, I know – “math joke:” is an oxymoron):
Taking differences,and then forgetting to add the constant back in when you sum or integrate is a sure way to fool yourself. Salby is the classic example of an “expert” that makes this mistake. Whether he fooled himself, or was fooled – doesn’t really matter. He’s usually believed by people that want to be fooled.
Piotr and Bob,
Noise is generally signal you don’t want. True noise, to me, is white. Clearly from my graph what you are calling noise has strong structure and is correlatable between phenomena. To me that is interesting and should not be removed by calendar-year averaging. If you are trying to understand the relationship between the change in two variables, then surely differentiation or departure from the mean are more useful tools.
My graph shows a much stronger structural correlation between CO2 and T than Gavin’s, which is the stated aim of his.
So what does mine show? When temperatures go up, CO2 goes up. When temperatures go down, CO2 goes down. Simple physics, we all know this and we all know why. Just open a warm soft drink to understand. I learnt this at uni in the early ‘80s and I would be surprised if anyone who reads realclimate would deny this relationship. What people choose to ignore is the lag, and this has important implications. The lag proves cause and effect, unless temperature in sentient.
If there is a strong positive feedback from CO2, especially one that is at least as much again as the actual ∆T from ∆CO2, then temperature would continue to rise due to the continuing increase in CO2. Clearly this is not happening.
As a similar example let’s look at the beginning of the current interglacial. Science tells us that as the earth started to warm, CO2 is released from water. The earth continued to warm until about 10KYO and then stopped, as did the CO2 increase. In the alternate reality of climate science, the initial temp increase bootstrapped the ∆CO2 to cause more warming and the CO2 becomes the cause, not the effect. In this reality, why did warming stop when there was still an increasing CO2 concentration?
Please do not think I am suggesting the increase in CO2 over the last 150 years is not due to human emissions
K Woollard: “Noise is generally signal you don’t want. True noise, to me, is white. Clearly from my graph what you are calling noise has strong structure and is correlatable between phenomena. ”
Since this blog and Gavin’s graph discuss the climate TRENDS in T as a function of CO2 – then whatever you get AFTER removing this trend IS the NOISE. Consequently, the “structure” of that noise is irrelevant to the discussion of the TREND, and certainly – a graph of the NOISE alone is not an IMPROVEMENT over the graph of the SIGNAL – despite your boasts:
KWoollard: I think a BETTER graph to show the relationship between changing CO2 and changing T (the first one on the second page) WOULD BE: [ and here you gave the link to a graph that … REMOVED the TREND]
Threfore your graph, is NOT “better” – quite the opposite – it aims to mislead – tries to deny the climatological trend (T as a function of CO2) by …. removing the trend and leaving only the residue. .
Further, Gavin’s graph has great societal relevance – since it us who emit CO2 and it is us, or our children, who would suffer the consequences. Your graph, on the other hand, is irrelevant and inconsequential – the tiny short-term wiggles in global temperature likely natural , having no real importance to us and teach us NOTHING about the climatological TREND.
KW: “The lag proves cause and effect”
… in tiny natural fluctuations around the mean, NOT the cause and effect of GW: different timescales and sources of CO2 – different cause and effect.
The same goes for your another denialist chestnut – the glaciation cycles. Denialists use to claim:
1) since climate change happened before humans, then humans can’t cause current climate change. That’s like saying that because wildfires happened long before humans, then humans can’t possibly start a wildfire”
2) “T is the cause and CO2 the effect and therefore human emissions pf Co2 cannot cause GW.. The misdirection here is in attributing massive change in global T to a tiny trigger (Milankovic) and ignoring feedbacks that AMPLIFIED MANYFOLD this tiny warming. And if CO2 is in a positive feedback with T, then by definition higher Co2 increases T. And if CO2 increased T during deglaciation, then it would increases it also OUTSIDE of the feedback, i.e. when we add CO2 from an external source (fossil fuels)
3) third denial myth that glaciation helps to debunk says: “we are too small to affect the climate“. Ice albedo and water vapour amplified the warming effect of rises in natural CO2 during deglaciation, AND they amplify the Human warming: more Co2, higher T, more ice melts and more water vapour – even warmer than by CO2 alone ….
In other words, these two natural feedbacks from glacial time will AMPLIFY effects of our action or inaction – if we increase CO2 conc. they will make warming larger than from Co2 alone, if we decrease CO2 they will make the cooling stronger than from the decrease in Co2 alone.
I read the discussion, I have a question, do the models take into account the climate trend where the earth has been cooling naturally since the 17th century and we have disrupted this process with CO2 emissions and somehow offset the natural cooling? And the human-caused greenhouse effect itself is underestimated as a result?
My own attempts to furnish the world with graphics illustrating the progress of AGW was prompted by the poor provision of such things on the internet back in the 2000s. Things have improved greatly since then with most (but not all) of the deficiencies now become history.
One of the problems I saw back then was the common appearance of graphics created to show some particular dramatic development in AGW, say the dramatic 2005 Atlantic hurricane season or the big 2007 melt season of Arctic sea ice. When these events became history and were then followed by less dramatic years, the continued existence of these old graphics on-line without easy access to more up-to-date ones (or even any access) was effectively the creation of cherry-picking, something normally associated with the denialists. Today, while such ancient graphics will appear when folk venture into the on-line world, data provision like that being announced in the OP here and institutions are providing web browser tools for the public, the likes of the NSIDC’s chArctic or JAXA’s Vishop (at time of writing, currently crashed as it occasionally does). (although in the case of polar ice, I still feel the big annual cycles of polar freeze/melt are obscuring the trends through time), those old favorites have become almost rare.
Excellent idea. Thanks, Gavin!
This helps bigly!
This is a very welcome undertaking. I am particularly interested in seeing graphs of SST (Global) along in comparison with Aerosol forcings.
Thanks in advance, and best regards.
Good show, Dr. Schmidt! Thanks very much.
You could also install MediaWiki, the software which is used at Wikipedia, perhaps to a subdomain like wiki.realclimate.org
https://m.mediawiki.org/wiki/MediaWiki
Installation is similar to WordPress, and it can be extended with different editors, templates, and so on. Same for the content scope, i.e. constraint to images, perhaps with dedicated pages for references and notes.
For those interested. Zac Labe on Mastodon has reams of excellent climate charts and graphs. Well worth a follow if you are.on Mastodon and feel a little dirty using Twitter these days..
His Mastodon handle is. @ZLabe@fediscience.org
Great work, thanks.
OT for here, but Zack Labe has been doing a superb job collecting and updating data on the cryosphere. [I heard that tfg and his dinosaurs ripped up NSIDC’s reporting, which is now infrequent.]
https://twitter.com/ZLabe/status/1571290249178783744
“Near Real Time Visualizations” –
https://zacklabe.com/arctic-sea-ice-figures/
The temperature graph should probably make it clear that it provides annual temperature anomalies rather than monthly temperature anomalies, to avoid confusion.
Perhaps a confidence interval to capture uncertainty in annual temperature anomaly estimates could be useful, but I suppose it might create confusion for some less-technical readers.
The graph with the unadjusted temperature anomalies could benefit from either a link to Zeke’s methodology or a brief description of the methodology used to produce the unadjusted temperature anomalies.
I wonder if there could be benefit from a graph that breaks down the impact of each type of adjustment (adjustments for types of SST measurements, exposure adjustments, urbanization adjustments, homogenization, etc.).
These graphs all seem to begin in the 1800s. I wonder if you could possibly redo the graph of the 800.000 yr relationship between Co2 and temperature. I do find creations all over the place, but they’re all a bit different. And I searched but strangely couldn’t find one on RC (maybe how I looked though). Anyway, something that lays the lines over each other, not above and below.
Regarding the temperature vs CO2 at the top of “the second page” would it not be more sensible to put CO2* on a logarithmic scale to better reflect the approximate relationship?
My version is here: https://southstcafe.neocities.org/climate/bestAndCO2.png
* Better might be a CO2 equivalent of all greenhouse gases
Mark, good idea, given that CO2 radiative forcing is approximately logarithmic – to plot it this way has more sense .I wish it were the standard way people plot Co2 vs T, not only for the future, but especially for the geological past.
Better might be a CO2 equivalent of all greenhouse gases
Well, I think the opposite might be the case – the fit might be worse. The observed warming
is increased by rising CO2 and other GHGs, but is cooled by aerosols and albedo increases – in fact the total cooling is comparable in size the warming by the non-CO2 GHGs:
https://upload.wikimedia.org/wikipedia/commons/thumb/a/a0/Physical_Drivers_of_climate_change.svg/450px-Physical_Drivers_of_climate_change.svg.png
– after two effects mostly cancelling each other out – what is left – a more or less the CO2 alone effect.
Now, if you include the other GHGs without subtracting equivalent cooling by aerosols – the fit should get actually get poorer, at least in the periods when the increase in non-CO2 GHGs and aerosols is significant.
Your logscale CO2 graph seems the wrong way around, it seems to show more slope at the higher numbers rather than lower slope
Can’t actually tell, as only two labels are given on the log side. One can’t assume the axis value.
You don’t know much about mathematics, Keith Woollard, do you?
First, you disproved the effect of global warming on the hydrological cycle by saying that in … some town in the Western Australia … there was no clear correlation between local temperature and local precipitation in that town.
Then you lectured Gavin Schmidt, Director of the NASA Goddard Institute for Space Studies, that you have a better graph on the climatological relationship between T and CO2:
K. Woolard: “I think a better graph to show the relationship between changing CO2 and changing T would be: https://www.woodfortrees.org…”
those promoting as “better” a graph that …removed the signal (long term climatological trends of T and Co2), and retained only … the residual noise. ;-)
I.e. classic denialist technique of “disproving” climate change trends with the short-term noise. And though me and Bob Loblaw pointed it out to you – you haven’t acknowledged either, instead, without missing a step, you moved on to … questioning the validity of graph by Mark B.:
K. Woolard: “ Your logscale CO2 graph seems the wrong way around, it seems to show more slope at the higher numbers rather than lower slope”
Which would be a valid point, IF CO2 have been increasing at a comparable rate since 1750. But as you know well, e.g. from the discussed graph – it doesn’t. Ppen Excel, and put in the following numbers:
year Co2 [ppm] Log(Co2)
1750 280 2.448
1950 310 2.491
2020 415 2.618
From that: Delta log(Co2)/Del(t):
1750 – 1950 = (2.491 – 2.448)/200 = 0.000215
1950 – 2020 = (2.618 -2.491)/70 = 0.001814
So much for your questioning the validity of the graph and/or integrity of Mark B. THERE IS “more slope at the higher numbers rather than [at the lower numbers]”. 8.5 TIMES more to be precise.
Hence, again, your presumption that it is the other people who are wrong and/or dishonest – is rooted in your own ignorance. Not a very endearing trait.
Higher slope compared to the linear scaled version
Keith Woollard: Jun 10. “ Higher slope compared to the linear scaled version”
I am … not sure about that. Here is what you ORIGINALLY wrote:
K. Woolard Jun 8.: “Your logscale CO2 graph seems the wrong way around, it seems to show more slope at the higher numbers rather than lower slope”
i.e.: – no reference to the linear scaled version and you seem to question the internal consistency (slopes at lower vs, at high values of CO2; “the wrong way around”). Internal consistency is WITHIN the same graph.
But even if if you meant slopes on two different – it does not make much sense. These are two very different plots: X-axes ranges are very different: Gavin’s 1870-2025 vs. Mark’s 1730-2030. Also Mark’s graph is much more “square-ish” than Gavins. As a result, the X-axis of Mark displays 4 times more X-compressed than Gavin’s. Which means that ONLY to a naive observer the same slope would APPEAR 4-times steeper than it actually is.
All of which would be irrelevant, if you have run a few numbers, to double-check whether your visual impression does not lead you astray. But, instead, you went on to question the validity/honesty of Mark’s graph, based solely on your … gut feeling that Mark’s slopes are “the wrong way around”.
I am sure there is a “teaching point” somewhere, but given our past encounters – I doubt you are interested.
I am probably wrong. When trying to plot two different relationships on the same graph, you have two options to change…. the offset and the scale. I suspect Mark B has gone with a larger scale and a smaller offset than Gavin
K Woollard. Jun 11: “ the offset and the scale. I suspect Mark B has gone with a larger scale and a smaller offset than Gavin .
Open the graphs side by side and you will see that the scales of T and CO2 axes are very similar. What’s different is X-axis: as I have already explained to you in the post to which you reply:
“X-axes ranges are very different: Gavin’s 1870-2025 vs. Mark’s 1730-2030. Also Mark’s graph is much more “square-ish” than Gavins. As a result, the X-axis of Mark is 4 times more X-compressed than Gavin’s. To a naive reader the slope would APPEAR 4-times steeper than it actually is.”
You aren’t really looking at the same thing as me Piotr, if you copare the slope towards the end of the record for T and CO2 Mark B’s graph has the CO2 becoming more steep in line with the T, whereas Gavin’s diverges
Keith Woollard 16 JUN: “You aren’t really looking at the same thing as me Piotr ”
Not surprisingly giving that you constantly move your goalposts, Mr. Woollard. Let’s recap:
1. KW 8 JUN: Your logscale CO2 graph seems the wrong way around, it seems to show more slope at the higher numbers rather than lower slope
i.e. you questioned the validity/honesty of Mark’s CO2 plot within his CO2 plot (the slope of CO2 on the right side of Mark’s graph being too steep compared to the left side of the same graph)
– I replied that it is what should be expected – a steeper slope when CO2 increases faster (the right side of Mark’s graph) than when it was increasing very slowly. (the left side)
2. KW. 11JUN – changes his tack – claiming now that he …didn’t criticize Mark’s Co2 graph within itself, but in comparison with …another graph (Gavin’s).
-To which I replied that the supposed discrepancy is because you merely EYEBALLED the two graphs and didn’t notice that Mark’s X-axis was 5-times more compressed. Which to a naive reader would make the slope APPEAR more steep.
3. KW 16 JUN: …recasts his original claim again: “You aren’t really looking at the
same thing as me Piotr. Mark B’s graph has the CO2 becoming more steep in line with
the T, whereas Gavin’s diverges””
So by saying originally that ” [Mark’s graph’s CO2 having too much slope] at the higher numbers rather than lower slope” you didn’t mean:
– the slopes of CO2 on two sides of Mark’s graph
– nor even the slopes of Co2 in two different graphs,
but different slopes of … T in those different graphs???
If so – then you must be the most inept communicator on this group. (And this is a group already containing the products of the precise and logical minds of T. Kalisz, M Shurly, JCM, KIA, and Victor…)
P.S. And what is this “ divergence” between CO2 and T in Gavin’s graph you now speak of? In which years?
Some years ago I grumped about bad availability of data and that graphs are only available as seperated graphs for the paleoclimatic timeline. (I think I did that in this forum.)
Different file formats were also a problem, so after trying around for a while I gave up creating my own graphs.
The graphs I had in mind… I think also were partially also available here on one of the graphics-pages.
Motivated by that anger (of graphs being only available cluttered around the web, different time scaling etc.), I did anonymous entries/changes of a climate page from Wikipedia, inserted links to the data sets (I think mentoned here on realclimate.org) and graphs in the hope that my suggestions were used to create a graph with all the data sets included… I also wanted to have the phase of human existence being somewhat emphasized, as it seems to be very unclear to most lay people, what time scales are used and what tiny fraction of that time was habitable by humans.
All my suggestions were removed/deleted, and I was quite pissed!
But recently I checked the situation again, and my suggestions (which data (include references), make ONE graph, etc.) were picked up, and a nice graph has been created.
This article contains the graph:
https://de.wikipedia.org/wiki/Globale_Erw%C3%A4rmung
Direct Link to the graph:
https://de.wikipedia.org/wiki/Datei:Temp-phanerozoic_combined-de.svg
It’s close to what I had in mind. The maginfier is quite a good idea, btw.
Just. wondering, could (or has one already been made?) a fine-tuned graph that juxtaposes prehistoric rising temperatures (to at least 800 kyr) and the extinctions of plants and animals be made? Just wondering which came first, the rising (and/or extreme lowering) of temps, then the extinctions, or the extinctions, then the radical changing of the temperature? Intuitively we think temps came first, but I’m wondering about the influence of life on it …
Some references I looked at:
https://www.countercurrents.org/glikson220210.htm
https://www.iflscience.com/mass-extinctions-and-climate-change-why-speed-rising-greenhouse-gases-matters-34640
https://images.theconversation.com/files/89469/original/image-20150723-22852-9nmf7j.png?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip
Sorry. What I mean is life, or it’s extirpation, influencing temps. Don’t know if I said that.
Me.Gavin, you have always proven up to the task of making sense of the complexities of human influence creating climate change. What you suggest with your new references makes the study of climate science easier to review. Citizen observers such as I need the helping hand you have extended more then you know. Thank you for that
Best to you and yours,
Lucien
A (set of) graphics that I would find interesting, is the comparison of regional and global temperatures over time.
Especially for certain regions and time periods this would be intersting.
If it’s easy to see the differences between regional and global temperatures, maybe some of those people (those reachably by arguments) who often pick the green grenland and the warm middle ages as “So you see – it has been warm already in the past, and that was not a problem! So we have no problem today!” would then comprehend that their claims are wrong.
Graphics might be the eye opener, while text is not.
I think it’s not by accident, that graphics are presented to the eye, and the “eye opener” might be taken literally here… it might not be convincing to the ideologists, but to those open for rational arguments.
A very good suggestion. But I would specify that it not be in the form of a small global map projection with fuzzy colors.
3-D plotting is easily available, make it interactive so you can scroll across a grid, yadda yadda.
And no, I’m not volunteering; hire some kid who isn’t working on a 15-year old computer.
Just some graphics that show the global average as well as the regional averages for certain areas/time periods would be fine.
I don’t think that over engineering with 3D and displaying all data for all periods and all regions would make sense. Just to have something to look at and explain the problem of regional vs. global averages would be great.
It’s much easier to explain things with the matching graph at hand. Without it, explanations become too elaborate and it’s not easy to start with that at all. And people might not be able to follow, if you only use text or speech.
With the graph at hand, explaining the meaning of each case is then simple. And to start with some selected regions and timeperiods makes sense, as they are often picked as (alleged) counter arguments. In these cases, 3D and complicated usage is rather counter productive. Just looking at the graph, without the need to work interactively is the right thing, imo.
Complicated matters must be explained easily enough for non-experts to follow.
If the fake-arguments are easy to understand, and the real arguments need too many pull-ups for the mind, then the fakers have an easy job.
Why do you (and GISStemp) base the temperature anomaly graph on a 1951-1980 baseline? That misses about 0.22 C warming above the 1850-1900 proxy for preindustrial aveage, understates the amount of warming due to human GHG emissions, and confuses the public when they read that the amount of warming is now 1.1 or 1.2 C above the predinustrial average at other sources.
Very helpful, thanks Gavin!
Just two days ago I’ve been asked for a graph of CO2 and temperature (similar to the one you have since 1880) since about 1000 years. Do you have one or do you know about one?
Thanks, Urs
[Response: No, but I could work on it… – gavin]
On another note: One of the best retorts at Held v. MT (youth seeking to enforce Montana’s Constitutional Right “to a clean and healthful environment” trial came from Professor Steve Running, who share the IPCC Nobel Prize, was asked words to the effect of “When the last Montana Legislature repealed the Montana Environmental Policy Act and prevented state agencies from evaluating the effect of climate change on energy projects, did that affect your testimony” explaining climate change? To which Running responded, “You can’t repeal physics!” Running and his lawyer are the best team I’ve seen in decades of law practice. Watch online at https://www.youthvgov.org/held-v-montana 9 am mountain time, June 12-23. Or get the posts when they are posted. Scientists here can learn a lot from Running–he didn’t miss a beat. Hope you will post Running’s testimony here when it becomes available with the video of exhibits. Another highlight came when the opposing lawyer objected to Running’s explanation of climate change on wildlife y say that foundation was lacking because he was not a wildlife biologist. Whereupon Running explained he coauthored the IPCC article’s picture that was showing on the screen which also appeared on the cover of an IPCC report.
Really helpful for teaching. Maybe put the two key links right at the top of your post for easier access? Or even directly on the home page of RealClimate. Thank you!
[Response: They are at the bottom under “Data and Graphics” – gavin]
Great move – I think we are all seeing now that Twitter is unlikely to be a reliable store of knowledge for much longer.
On that point, are you (or Real Climate) on Mastodon yet?