A few years ago there was a cottage industry among some contrarian journalists, making use of the fact that 1998 was a particularly hot year (by the standards of its period) to cast doubt on the global warming trend. Ummmm, where did I see this? . . . Here, I found it! It was a post by Stephen Dubner on the Freakonomics blog, entitled, “A Headline That Will Make Global-Warming Activists Apoplectic,” and continuing:
The BBC is responsible. The article, by the climate correspondent Paul Hudson, is called “What Happened to Global Warming?” Highlights:
For the last 11 years we have not observed any increase in global temperatures. And our climate models did not forecast it, even though man-made carbon dioxide, the gas thought to be responsible for warming our planet, has continued to rise. So what on Earth is going on?
And:
According to research conducted by Professor Don Easterbrook from Western Washington University last November, the oceans and global temperatures are correlated. . . . Professor Easterbrook says: “The PDO cool mode has replaced the warm mode in the Pacific Ocean, virtually assuring us of about 30 years of global cooling.”
Let the shouting begin. Will Paul Hudson be drummed out of the circle of environmental journalists? Look what happened here, when Al Gore was challenged by a particularly feisty questioner at a conference of environmental journalists.
We have a chapter in SuperFreakonomics about global warming and it too will likely produce a lot of shouting, name-calling, and accusations ranging from idiocy to venality. It is curious that the global-warming arena is so rife with shrillness and ridicule. Where does this shrillness come from? . . .
No shrillness here. Professor Don Easterbrook from Western Washington University seems to have screwed up his calculations somewhere, but that happens. And Dubner did not make this claim himself; he merely featured a news article that featured this particular guy and treated him like an expert. Actually, Dubner and his co-author Levitt also wrote, “we believe that rising global temperatures are a man-made phenomenon and that global warming is an important issue to solve,” so I could never quite figure out in their blog he was highlighting an obscure scientist who was claiming that we were virtually assured of 30 years of cooling.
Anyway, we all make mistakes; what’s important is to learn from them. I hope Dubner and his Freaknomics colleagues learn from this particular prediction that went awry. Remember, back in 2009 when Dubner was writing about “A Headline That Will Make Global-Warming Activists Apoplectic,” and Don Easterbrook was “virtually assuring us of about 30 years of global cooling,” the actual climate-science experts were telling us that things would be getting hotter. The experts were pointing out that oft-repeated claims such as “For the last 11 years we have not observed any increase in global temperatures . . .” were pivoting off the single data point of 1998, but Dubner and Levitt didn’t want to hear it. Fiddling while the planet burns, one might say.
It’s not that the experts are always right, but it can make sense to listen to their reasoning instead of going on about apoplectic activists, feisty questioners, and shrillness.
Freakonomists getting outflanked
The media landscape has changed since 2005 (when the first edition of Freakonomics came out), 2009 (when they ran that ridiculous post pushing climate-change denial), and 2018 (when the above post appeared; I updated it in 2021 with further discussion, and here’s the news from 2023).
Back in the day, Steven Levitt was a “rogue economist,” a genial rebel who held a mix of political opinions (for example, in 2008 thinking Obama would be “the greatest president in history” while pooh-poohing concerns about recession at the time), along with some soft contrarianism (most notoriously claiming that drunk walking was worse than drunk driving, but also various little things like saying that voting in a presidential election is not so smart). Basically, he was positioning himself as being a little more playful and creative than the usual economics professor. A rogue relative to a stable norm.
I wonder how the Freakonomics team feels now, in an era of quasi-academic celebrities such as Dr. Oz and Jordan Peterson, and podcasters like Joe Rogan who push all sorts of conspiracy theories, and not just nutty but hey-why-not ideas such as UFO’s and space aliens but more dangerous positions such as vaccine denial.
Being a contrarian’s all fun and games when you’re defining yourself relative to a reasonable center, maybe not so much when you’re surrounded by crazies.
For example, what were Levitt and Dubner thinking back in 2009 when they published that credulous article featuring an eccentric climate change denier? I can’t know what they were thinking, but I suspect it was something like: “Hey, this guy deserves a hearing. And, in any case, we’re stirring things up. Conversation and debate are good things. Those global-warming activists are so shrill. Let’s make them apoplectic—that’ll be fun!”
The point is, this was all taking place in a media environment where climate change denial was marginalized. So they could run ridiculous pieces like the above-linked post without being concerned of having bad effects. They were just joking around, taking the piss, setting up boring Al Gore as a foil for “a particularly feisty questioner,” promoting a fringe character such as Professor Don Easterbrook from Western Washington University (he who told us in 2009 that climatic conditions are “virtually assuring us of about 30 years of global cooling”) secure in the belief that no one would take this claim seriously. Just a poke in the eye at humorless liberals, that’s all.
Recall that, around the same time, Levitt and Dubner also wrote, “we believe that rising global temperatures are a man-made phenomenon and that global warming is an important issue to solve” (see also here) so my take on the whole episode is that they felt ok promoting a fringe climate-change denier without concern that they could be upsetting the larger consensus. They got to have the fun of being edgy by promoting the prediction of “30 years of global cooling” without ever actually believing that ridiculous claim.
Nowadays, though, things are getting out of control, both with the climate and with extremists and wild takes in news and social media and in politics, and I imagine that it’s more difficult for the Freakonomics team to feel comfortable as rogues. They no longer have a stable center to push against.
A political science perspective
In political science we sometimes talk about proximity or directional voting. In proximity voting, you choose the party or candidate closest to us in policy preferences; in directional voting, you choose the party or candidate whose position is most extreme relative to the center while being in the same general direction as ours (to be precise, if we consider your position and each party’s position as vectors in a multidimensional space, you’d choose the party to maximize the dot product of your position and the party’s position, with that dot product being defined relative to some zero position in the center of the political spectrum). The rationale for proximity voting is obvious; the rationale for directional voting is that your vote has only a very small impact, which you can maximize by pushing the polity as far as you can in the desired direction.
There is a logic to directional voting; the problem arises when many people do it, with the result that extreme parties get real influence and even attain political power in a country.
Some examples of directional voting, or directional position-taking, include Levitt and Dubner pushing climate-change denial, people who should know better on the right supporting election denial in 2020, or, on the other side, center-leftists supporting police defunding, presumably following the reasoning that the police would not be defunded and the pressure to defund would merely cause police funding to decrease. Once you think to look, you can find this sort of political behavior all the time: a way to oppose the party in power is to support its fiercest opponents, even if you would not ever want those opponents to be in power either.
But . . . directional voting falls apart when the center does not hold.
In my home, the left-leaning feminists take it as a given that global warming is a fact and society needs to do something now. The graph and any associated commentary, however, has me puzzled. I think I am supposed to see that the y axis from 1998 for a bunch of years is sort of constant rather than sharply rising as it does after 2014. On the other hand, 2017 is lower than 2016 and I am informed that since then, 2017, the last data point, much has happened around the globe.
Naturally, being an American, degrees C does not come easily and I have no feeling whatever as to how in the world “Annual Global Surface Temperature, Relative to Late 19th Century Average” could be (and was) calculated.
To my suspicious mind, the true evidence regarding the increase in global warming comes from the entities who insist it is not taking place and therefore, nothing needs to be done.
The graph came from an old NY Times article with the headline “2017 was one of the hottest years on record. And that was without an El Nino.” I assume Andrew used it because it was easy to find, but the headline points to the thing that the Freakonomics post ignored. That is, you can think of the globally averaged temperature as the sum of naturally occurring variation and greenhouse gas-driven increase, so a downward trend in the naturally occurring variation can cancel the greenhouse gas increase for a few years. El Nino is part of the naturally occurring variation that makes for warmer temperatures.
“you can think of the globally averaged temperature as the sum of naturally occurring variation and greenhouse gas-driven increase”
We certainly have been strongly encouraged to think about it that way ever since global temperatures flattened out a bit after 1998, which never meant anything anyway despite the endless bickering. But there was and still is zero evidence that the climate can be decomposed to a two-parameter model this way.
If we peel back the first layer of this conceptual onion, we ask “what is ‘natural variation'” in this context beyond just a squishy placeholder variable representing every aspect of the climate that we don’t understand? And what does “greenhouse gas warming” mean when the CO2 band is fully saturated in the upper atmosphere and the meager measurement data we have from satellites refute Arrhenius’ most basic claims about shortwave/longwave absorption?
Nonzero evidence: https://www.carbonbrief.org/interactive-much-el-nino-affect-global-temperature/
The most prominent mode of natural variability is El Niño. It’s responsible for +/- 0.15C swings in global temperature. Remove that effect, and the global temperature change is less variable. Remove other known effects, and it starts looking rather smooth.
I addressed the Arrhenius criticism a couple of months ago and won’t repeat it here.
All those climate scientists don’t know nuffink!
“The most prominent mode of natural variability is El Niño.”
The most prominent mode of natural variability yet discovered is the Milankovitch cycles, which are thought to have produced climate fluctuations at least two orders of magnitude greater than El Nino.
As for “the Arrhenius criticism,” I’m still awaiting the first reasoned discussion of the CERES data that doesn’t involve one side simply shrugging off the discrepancies between measurement and theory. It would be a big problem in any other field.
The contribution of Milankovitch cycle variation to Earth temperature variation during the last 150 years is negligible and irrelevant to discussions of contemporary temperature change. If we were interested in contributions to temperature change over the last several thousands of years then it would be relevant (absent manmade global warming, especially since the industrial age, Milankovitch cycles should be causing a very small slow cooling right now).
We’re still awaiting for you to get to the point. Rather than drip feed insinuations about supposed flaws in satellite measures why not simply say what you consider the specific issues are?
“We’re still awaiting for you to get to the point. Rather than drip feed insinuations about supposed flaws in satellite measures why not simply say what you consider the specific issues are?”
I’ve already done that on this blog.
The Arrhenius model of CO2-induced global warming provides a perfectly plausible mechanism by which longwave radiation decreases from additional CO2, thereby heating the atmosphere. For folks who might not be familiar with the science, this has always been considered the single root cause of the predicted runaway global warming (AGW).
Up until recently, there was zero direct evidence that this was happening as described, although some aspects of the indirect evidence were supportive. There are a few ways the theory could be directly challenged with measurement data but there has been little interest in doing so. Despite this unfortunate disinterest, data from the CERES satellites finally gives us one way to compare the Arrhenius theory with actual measurements. If CERES data showed what Arrhenius predicted, that the recent heating temporally correlates to a reduction of outgoing longwave radiation, I would have been the first one to proclaim (FWIW) that AGW is finally supported by direct measurements, a huge hurdle to clear. But just the opposite happened, the CERES data shows that the recent warming was caused by attenuation of outgoing shortwave radiation.
This disconnect does not mean AGW is wrong, but it absolutely means that the process is more complicated than Arrhenius – or any of the climate scientists who have sworn allegiance to his theory – envisioned. Critics have argued for decades that we need to understand the water vapor cycle better than we do now to predict future climate, and those critics got a boost from this data. So again, nothing has been disproved, but we are definitely outside the realm of normal science when the basic physics conflict with the only data we have, and all we get is a collective shrug from the community. The Arrhenius model died very quietly and is now a zombie but I seem to be the only one keeping score.
Matt –
> This disconnect does not mean AGW is wrong, but it absolutely means that the process is more complicated than Arrhenius – or any of the climate scientists who have sworn allegiance to his theory – envisioned. Critics have argued for decades that we need to understand the water vapor cycle better than we do now to predict future climate, and those critics got a boost from this data. So again, nothing has been disproved, but we are definitely outside the realm of normal science when the basic physics conflict with the only data we have, and all we get is a collective shrug from the community. The Arrhenius model died very quietly and is now a zombie but I seem to be the only one keeping score.
I’m not qualified to argue the science with you, but there’s a bit of a red herring in that comment on that a notable % of “critics” say – as differentiated a but from what you write – that they don’t doubt that ACO2 emissions materially affect the climate, they only disagree with the “most likely” range as projected by “mainstream” climate scientists. As I’m sure you know, prominent “skeptics” like Curry say that there’s no reasonable argument about the “basic physics” (well, except when they contradict themselves in arguments about the basic physics).
So yeah, at the end you self describe as an N=1, but you also align yourself with a generalizes “critics” and I think you kinda have to pick one or the other.
“…predicted runaway global warming”?? I don’t think anyone predicted that tho I guess it depends on what you mean by “runaway”!
Your “Arrhenius model isn’t really correct – or at least atmospheric science/radiative physics doesn’t support that view. It would be more correct to say that enhanced CO2 reduces the ability of longwave IR (LWIR) to escape from the atmosphere to space to maintain radiative balance with incoming solar radiation. The effect is not to reduce LWIR escape but to move the average LWIR emission to higher altitude. Since it’s colder higher up, the atmosphere warms up right down to the surface to effectively push the temperature where LWIR emission occurs, to the temperature required for radiative balance.
That may or may not have been what Arrhenius thought (you could find out!) but it would be extremely foolish to assert that we can only explain things in terms known to Arrhenius while neglecting the 100 years of knowledge accrued in the meantime!
That’s not really correct either I think. It was demonstrated quite a while ago that under CO2-induced warming there is likely to be a temporary short period of reduced LWIR emission, but quite quickly under conditions of radiative imbalance the outgoing LWIR remains much the same (though average emission from a higher altitude) and the warming is largely due to increased absorption of shortwave solar radiation. That may well not be what Arrhenius envisaged but we don’t have to pretend that we can only know what Arrhenius worked out over 100 years ago.
Anyway, see for example:
Trenberth KE, Fasullo JT (2009) Global warming due to increasing absorbed solar radiation. Geophys Res Lett 36(7) L07706.
Donohoe et al. (2014) Shortwave and longwave radiative contributions to global warming under increasing CO2 Proc. Natl. Acad. Sci USA 111, 16700-16705
OK but you seem just to be venting your predjudices. Of course the climate system and global warming is more complicated than Arrhenius thought (including especially the influence of clouds), but the more recent fundamental predictions of climate science have broadly been found to be correct. Putting Arrhenius aside and moving forward just to 1967, Manabe (physics Nobel 2 years ago) and Wetherald calculated the thermal structure of the atmosphere under greenhouse warming including cooling of the stratosphere. That’s been verified by satellite measures as for example in a very recent paper:
Santer et al. (2023) Exceptional stratospheric contribution to human fingerprints on atmospheric temperature. Proc. Natl. Acad. Sci. 120, Vol. 120 e2300758120
And though it’s extremely difficult to use CERES data to directly measure the Earth’s Energy Imbalance (EEI) (the imbalance is less than 1 W.m-2 but the measurement uncertainty in CERES just from calibration is around 2 W.m-2 for LWIR), the energy imbalance can be measured relatively reliably from ocean heat measures since most of the excess thermal energy from radiative imbalance accumulates there and there’s a network of temperature sensors (ARGO) to measure this. Again, this more robust measure of ocean warming puts constraints on the energy imbalance (0.5 – 0.75 W.m-2 IIRC) that are consistent with expectations. See e.g.:
Schuckmann et al. (2023) Heat stored in the Earth system 1960–2020: where does the energy go? Earth System Science Data 15, 1675–1709.
(I haven’t put links in case the blog doesn’t like them but all papers can easily be googled and should be open access.)
I don’t work with climate data, so I am always puzzled with these graphs of “average global surface temperature”. How meaningful is an average temperature over the entire globe? It seems there is so much information left out. Are the stations that produce the data evenly dispersed over the surface of the earth? Or does the data come from some sort of sensor on satellites? Isn’t there likely vast variation in terms of climate change depending on the region in the world? For example, it’s obvious that glaciers are melting away in many (or most?) areas, but is the temperature change in these alpine or polar areas changing at the same rate as say tropical parts of the world, or does a single climate region drive most of this ‘average global’ change? Also, why do we care about the 19th century as opposed to the 18th, 17th, etc? Just because that’s the only data we have? If you plotted the number of temperature stations along the same axis would they be dramatically lower in the 19th century? Is that data very reliable? If some of the ‘data’ is via some model using a proxy that attempts to estimate past temperatures, then shouldn’t the data points on the graph in the past have error bars?
Is this type of graph actually used by climate scientists, or is it mainly used to try to move public opinion / policy? Just curious. Whenever I see these graphs they seem to provide more questions than information.
You’ve raised a bunch of good questions. A lot of them are probably answered better elsewhere. I will say that the average global surface temperature is not quite a “measurement” it’s more a model inference from a lot of measurements. These days there are temperature stations, satellites, weather balloons, and soforth, all of those get integrated together to produce an estimated average. In the past like in 1900 the reconstruction will be coming from different data, obviously no satellites or electronic sensor networks etc. There is certainly controversy over how this stuff is done. Tree rings and isotopes incorporated in coral reefs and lots of other proxies get involved as you go farther back in time.
I haven’t worked with this data directly but a woman who was in grad school with me did estimates based on coral isotopes using Stan back in like 2015 so that’s my tiny bit of knowledge.
Interesting. I was wondering if it was something like that…..actually I faintly remember something about tree rings on this blog.
https://www.stat.columbia.edu/~gelman/research/published/ClimRecon.pdf
Yes, all of these other things are plotted and analyzed, with error bars. Obviously, a single graph is just one distillation of a lot of information.
You can download lots of actual data yourself at https://berkeleyearth.org/ . Here’s one map: https://berkeleyearth.org/june-2023-temperature-update/
At some point the distillate is too distilled. Thanks for the links!
There are a number of groups that produce global temperature estimates. Their methodologies, in general, are available. General discussions are below, with the details in the papers that are linked.
GISS: https://data.giss.nasa.gov/gistemp/
Berkeley: https://berkeleyearth.org/methodology/
How meaningful is any average? It’s obviously not as complete as looking at the time series at every location on the planet, but it’s a good, easy to interpret number. There are a number of ways to deal with station density, quality, etc. The Arctic is warming much faster than the rest of the planet and regional analyses show that.
The baseline to compare to is not that important. We tend to look at anomalies from whatever chosen baseline. GISS uses 1951-1980 because it was the last 30 year period ending in 0 when they started their work and the standard climatological period from the World Meteorological Organization is 30 years with the last year ending in 0. In the US, for climate normals, we update every decade. WMO will do that globally now, although in the past, they’ve only updated every 30 years. For trend purposes, the baseline is unimportant. Many groups use the 19th century as a pre-industrial value, but that’s because there’s more directly observed temperature information then than in earlier centuries.
You can find versions of the simple graph with error bars. (For instance, Berkeley has a text file with monthly values with uncertainty.)
The graph is used by climate scientists because it’s a simple summary of information.
Cool! Thanks! The graphs at the links are much better. I don’t think I would use a plot with points and no error bars as in the blog post, because it seems pretty deceptive (not in a malicious sense, but in the sense of being such a simple summary that). At least when I think of points, I think of observed data or statistics of observed data, not unobserved parameters.
jd –
Coming from a total non-expert, so take it with a grain of salt.
> Isn’t there likely vast variation in terms of climate change depending on the region in the world?
Is that question rhetorical? Anyway, yes, there are significant regional differences in the rate of warming.
That in itself, however, doesn’t render an averaged global rate of warming meaningless. An averaged global rate of warming tells you something, something largely policy-related (particularly in terms of mitigation), even if it doesn’t directly inform as to the implications of climate change at the regional level (and thus how regions should best consider adaptation)
> but is the temperature change in these alpine or polar areas changing at the same rate as say tropical parts of the world,
No. And related information is pretty easily available if you’re interested in the details.
> Also, why do we care about the 19th century as opposed to the 18th, 17th, etc? Just because that’s the only data we have?
I’m not sure I understand the question. There’s data related to temps for far back before the 17th century, but obviously there are fewer directly measured temps as you go back in time (as the technology was less prevalent, or didn’t exist).
> If you plotted the number of temperature stations along the same axis would they be dramatically lower in the 19th century?
There’s missing data and imputation and kriging and smoothing and a ton of stuff I don’t understand (Matt Skaggs comments here often, he might explain). But from what I do understand, global temps are based on a more complicated process than merely averaging and plotting the data from existing weather stations.
>.. then shouldn’t the data points on the graph in the past have error bars?
I think it’s a fair guess that the statistical experts who calculate the rate of temperature rise account for increased uncertainty with the data as you go back in time. Do you really think they wouldn’t?
> Is this type of graph actually used by climate scientists, or is it mainly used to try to move public opinion / policy?
What do you mean by “used” by climate scientists? How would you imagine a scientist might use such a graph? The graphs are outputs from modeling projections. The projections have a function of informing further projections related to specific outcomes. And they also function function to inform the public and policy makers,
> Just curious. Whenever I see these graphs they seem to provide more questions than information.
Since you wonder about this whenever you see such graphs, you might want to engage in some actual research rather than just asking questions on a non climate-focused blog. The information is easily available.
Apologies to those who gave more informed answers that posted as I was writing mine.
The spatial variation of average annual temperatures is much larger than the spatial variation of annual temperature anomalies (departures from a long-term temporal average). Consider Mauna Loa: temperatures on its slopes vary by several tens of degrees K, but if one part is above normal in a given year, it’s highly likely that another part is also above normal.
A former colleague and coworkers once calculated that, with fewer than 100 reasonably-placed stations, you could calculate the global average temperature anomaly sufficiently accuracy for most applications, including measuring global climate change. We have many more than that, although in a few parts of the globe the spacing is barely adequate.
The graph is used by climate scientists because it is the best available measure of the atmospheric manifestation of global warming, and is extremely useful for comparing the magnitude of climate change in models against observations.
Most global temperature data sets begin in the mid to late 1800s because too many parts of the globe were unsampled before that. The reference period for calculating anomalies is arbitrary and it’s easy to convert from one to the other by adding or subtracting a few tenths of a degree. The most common choice for the anomalies themselves is something like 1961-1990 when there were lots of stations so that regional temperatures were well constrained, but if you’re trying to express the total magnitude of global warming you should transform the global average anomalies to an earlier reference period.
Various sources of uncertainty include how interpolation is performed, how changes in instrumentation is handled, what happened at unsampled locations, etc. HadCRUT5 generates an ensemble of 100 different global temperature analyses designed to sample these sources of uncertainty. Errors have temporal correlation so just plotting confidence intervals doesn’t really express the uncertainty properly.
IIRC, the latest IPCC report estimated the global temperature rise since the late 19th century at around 1.13 C +/- 0.2 C.
jd –
I just so happened to niaf now stumble across this, which mentions how a chart (from BEST) of temperatures displays greater uncertainty as you go back in time, as well as discusses some issues related to some of your other questions:
https://skepticalscience.com/frank_propagation_uncertainty.html
This reminded me to update that temperature by president chart from a few years ago. Warming during Trump was flat, and Biden is currently 8th most warming out of 26 presidents since records began:
https://i.ibb.co/6scjsnY/temp-By-Pres2023.png
The reason they thought there was a “pause” in 2009 is because warming was near flat during the Bush years after rising fast during Clinton. Then, of course, there was the very strong warming during Obama (only FDR oversaw more).
Why does this pattern exist? I have no idea, but it does. There is consistently warming during democrat presidencies, while under republicans it can go either way to net out near zero.
> Why does this pattern exist?
Just asking questions?
What pattern? Looking at such short term trends in the data (i.e., the four years of Trump’s administration) is looking at noise.
As for a putative “pause.”
It was a short term slow-down in the rate of increase of surface air temperatures in comparison to a much longer term increase in surface air temperatures, and without consideration of Ocean Heat Content (in which, IIRC, there wasn’t a short-term slowdown in the longer-term rate of warming) – which comprises much more of the climate system energy.
The “pause” in global warming” was, and remains, a pretty vacuous concept.
I am looking at the pattern in all the data available, ie from 1880-2023. It is quite obvious in that chart. Warming always* occurs under democrat presidents, while it is split ~50-50 under republicans.
* It used to really be 100% of the time, but for some reason the values from decades ago is still changing. So now Kennedy oversaw a slight decrease.
Such short-term chunks of data are meaningless. You’re looking at coincidences and just asking questions about why coincidences happen.
If you’re interested in knowing why records of previous warming are changed, it’s because the methodologies develop over time; I can’t give you a technical answer but it’s pretty easy to find that information
It is one of those coincidences that happen ~100% of the time and can apparently be used to predict the future.*
Anomalies like this indicate there is something fundamentally wrong with how we are thinking about this data.** Here it is, with data source links in the comments at top:
https://pastebin.com/JdaPRAEg
* At least so far, we will see what the results are by the end of Biden’s presidency, it would take quite a bit of cooling over the next 1.5 years to get negative though. Or perhaps he will be reelected.
** I now updated it with all the most recent gistemp values. For the image posted above I only appended the last few years, while there were once again small changes in the values it is negligible.
Anoneuoid –
Your chunking of time is completely arbitrary.
But feel free to propose a hypothesis for the causal mechanism by which temperature readings around the globe would vary as the result of who’s the president of the US.
It will be interesting, no doubt.
> Anomalies like this indicate there is something fundamentally wrong with how we are thinking about this data.
What do you mean “we” Kemosabe.
Ordinarily I’d think that this is a function of Poe’s law, but with you its often hard for me to tell.
I actually looked into that data (about presidential party and global temperature changes) awhile back. As I recall, there is something called the Southern Oscillation (I’m sure someone can speak to that – with its proper name – more intelligently), but the correlation between party president and global temperature change falls apart when you include the Southern Oscillation in the model. I didn’t bother with any further analysis (mainly because it is not an area I have any expertise in), but that pattern certainly seemed like something – until, then, it didn’t.
Sorry, I went back and looked at our lengthy 2019 exchange about exactly this topic. I don’t think it makes sense to repeat that whole discussion now. Plenty was said and nobody appears to have changed their minds.
I didn’t go over the original comments again, but sure. Two more presidencies fitting this empirical model may not be enough to convince you.
How many in a row would it take to convince you there is some relationship here?
Could someone link that previous discussion?
The idea that you can meaningfully distill four year trends in global temps, let alone find a casual connection in those trends to the party of the US president during those four ear trends, seems about as goofy as connecting the activity of bowling to significant levels of mortality.
Oh.
Wait.
42
So essentially it is impossible for you to believe this correlation exists. It would take longer than a human lifespan (I know 42 is a joke).
I mean this is just the first analysis I did. If I wanted to travel the garden of forking paths, I would exclude presidents with “irregular” terms. Pretty much all the outliers (Ford, Kennedy, Truman, Johnson) fall into that category. Then Nixon and FDR who fit the pattern but are not crucial.
That leaves Woodrow Wilson as the only exception, otherwise the distributions wouldn’t overlap at all. That is too p-hacky for my tastes, but there are millions of published articles that follow a similar procedure.
Joshua
If you were seriously wanting to revisit that tedious discussion, you can find it at
https://statmodeling.stat.columbia.edu/2019/02/11/global-warming-blame-the-democrats/
Dale –
Thanks. I didn’t bother trying to even parse the statistical minutiae, but it was fascinating to realize that one possible explanation could be that the partisan identify of US presidents can have a differential causal effect on volcanic eruptions. 😉
Reminds me of a blog guest post at Judith Curry’s website a while back that tied rates of global warming to stick market performance.
There’s a rule of thumb to describe this phenomenon: ABC (Anything But Carbon)
My guess is that it is one of those coincidences where one thing correlates with some other thing with no causal connection, which is noticed by the average person once a month according to Littlewood’s Law of Miracles. But if there is an causation it could as easily be that temperature changes affect people’s voting patterns as that Presidents affect temperature changes.
I also note that over USA history, the Democrat and Republican parties have exchanged policies quite a bit, e.g. on racial equity.
I’d guess this as well. We are missing some strong agricultural/economic signal that precedes a warming episode. It then shows up in which party gets the presidency. In general, this is an old idea: https://en.wikipedia.org/wiki/Sunspots_(economics)
Anyway, we don’t need an explanation to accept that something is true. And history has shown that the vast majority of explanations for things that do get widely accepted end up being wrong. I mean when it came to gravity Newton said “hypothesis non fingo”.
How long did people understand the sun was very likely to rise in the morning without the right explanation for why?
It also reminds me of the people who denied covid was seasonal in 2021 because they wanted to say southern states got more covid because they were republican or something. Actually, it was summer (and hot) so people there congregate inside like they do in the winter in the north. Then when winter came we got the huge surge in cases in the northern states. I wonder if the CDC et al are still denying the obvious seasonality. I know Fauci doesn’t.
Another (obvious, but relevant) confounding factor is the long lag between climate causes and effects, such that if we went to net-zero carbon emissions tomorrow, temperatures would continue to rise for decades. So that any changes occurring during Democrat presidencies were largely baked in from previous years.
As for explanations not requiring explanations (I kid), it seems to me Newton knew a couple things which overcame the aversion to action at a distance: 1) the English colonies and explorations were spread out enough so that Newton knew that apples all around the world fell straight to the ground, that is, toward the center of the Earth; and 2) that a object, such as a milk pail, when swung in an arc around oneself and released, travels in a straight line tangent to the arc at the release point, as its natural behavior, but while constrained to the arc, feels a radial force. Those known facts provided the start for his explanation of the moon’s orbit. I don’t have similar explanatory facts about the temperature-USA President correlation.
As someone mentioned, the Real Climate blog may well have such explanatory facts.
That is a great example. Over 300 years later the equivalence of gravitational and inertial mass remains a “coincidence” that requires explanation:
https://en.wikipedia.org/wiki/Equivalence_principle#Explanations
Good thing Newton didn’t reject it because he didn’t have an explanation.
That causation has the advantage of at least being theoretically plausible.
Except temperature trends in four-year intervals seems like just noise to me (maybe I should be a little more conditional in my language given folks like John N. G. are out and about).
That would even apply even if the directionality were reversed, imo, if you’re considering agricultural trends in a country so large and so climatically diverse, and even if you’re trying to say that there’s an effect independent of the president’s partisan identity (I.E., could result in a change of party but not a larger trend favoring demz or pubz on the whole).
But add in partisan identity and sorry, can’t see how this isn’t merely a kind of apophenia.
If you want to limit it to the period when warming became a political issue, that starts with the WMO warning in 1976:
https://allouryesterdays.info/2022/06/21/june-22-1976-times-reports-worlds-temperature-likely-to-rise/
So start with Carter. Since then there have been four each of Democrat and Republican administrations. Essentially all the warming has been under Democrat presidents. I really don’t see any way of p-hacking this correlation away. Instead it gets stronger.
This is actually a case where NHST could apply. Apparently you really believe warming under democrats and republicans are samples from the same distribution. But when we compare since 1976 (when it became a political issue) we see there was ~100x more warming when a Democrat was president:
> d = c(.188, .212, .355, .17)
> r = c(0.079, -0.124, 0.059, -.004)
>
> t.test(d, r, var.equal = F)
Welch Two Sample t-test
data: d and r
t = 3.6789, df = 5.9604, p-value = 0.01047
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
0.07635637 0.38114363
sample estimates:
mean of x mean of y
0.23125 0.00250
>
> sum(d)
[1] 0.925
> sum(r)
[1] 0.01
That is a vitamin C for scurvy level correlation.
What is most interesting about this to me though, is figuring out what kind of reasoning can be applied to exclude this without rejecting 99+% of what gets published in the scientific literature.
Climate change didn’t become a particularly significant partisan issue until the late 80’s at earliest. Even as recently as 2000 the Republican candidate for prez saw addressing climate change as an expedient political platform.
You’d be better off running with economic policy adoption as the party-associated mechanistic driver, but of course AFAIC, that’s still bullshit anyway because it’s scientifically meaningless to chop climate trends into 4- (or even 8-year) intervals. Repeatable climate drivers (as opposed to one-offs like volcanic activity) are not visible on four-year time scales, and even if they were they’d be swamped by “natural” drivers like El Ninos and the like. Not even skeptics have proposed anything on a short-term time scale like what you’re suggesting. The kinds of variables affected by something like economic policy IN ONE COUNTY on the entire planet (albeit a large and influential one) as differentiated by the relatively minor policy differences between Dem and Pub presidents don’t translate into the climate, let alone on that kind of time scale. The global. climate is analogous in this context to the proverbialcean liner, with a lot of dynamics/momentum (like energy) stored in the system. It takes a long time for it to maneuver in response to exogenous variables, particularly ones om the scale of differences in policy between Dems and Pubs as US presidents. And that’s true for the much more meaningful driver of climate, ocean hear content, in particular.
I’d even say there’s prolly no signal in the *economy* as the result of party-associated policy differences of the US president (as isolated from other factors like the balance of Congress or other exogenous economic forces) on that kind of short-term scale.
Of course I’m sure that’s arguable, and economists do argue about that. But if that’s not even clear, then the idea you could see a signal in temp trends looks on a par with bowling ally deaths.
https://insideclimatenews.org/news/03032023/jimmy-carter-climate-change/
So nope, it started in the late 1970s. Ie with Carter.
Of course you can isolate discrete little climate-adjacent elements to draw some kind of a facile conclusion about a political signal on a national scale, but that’s just poor analysis.
In terms of a SIGNIFICANT partisan signal on a national scale, to the extent that it would signal some kind of impact on the freaking global climate, let alone in four-year intervals – I think you need to do better than that.
I mean seriously, trying to work from that one aspect of Carter’s campaign to am almost immediate impact on the global climate? How can you be serious. I mean looking aside all the other problems with your argumentzfhe idea that it’s a strong national political signal in terms of global warming is just unsupported.
For example:
Through legislation he signed in 1990, Bush started the National Climate Assessment, a sweeping study documenting climate change’s impacts on the United States.
[…]
“We know that the future of the Earth must not be compromised,” Bush told the U.N. Intergovernmental Panel on Climate Change in 1990. “We bear a sacred trust in our tenancy here and a covenant with those most precious to us—our children and theirs.”
with those .
I am genuinely interested in what reasoning can reject the existence of this correlation without throwing out the vast majority of research that has been published in the last 70-80 years.
As said earlier, I don’t know why this ~100x more warming under democrats exists. I guess I shouldn’t have speculated at all, since it distracted from the point. It is simply a fact to be explained about our world. It is also a fact that global warming first became a political issue in the late 1970s, I don’t see where you dispute this.
What is your explanation for this fact? The usual method to check if it is a coincidence is a low p-value using a model that assumes no correlation.
Are you also willing to reject all the studies about eg whatever gene is associated with a disease that use the same method to come to a conclusion?
>What is most interesting about this to me though, is figuring out what kind of reasoning can be applied to exclude this without rejecting 99+% of what gets published in the scientific literature.
I probably shouldn’t get involved in this at all, but what data are you using in these analyses?
GISTEMP. Check the comments at the top here:
https://pastebin.com/JdaPRAEg
The code is in Andrew’s 2019 post “Global warming? Blame the Democrats.”
One link per post to avoid the spam filter…
Here is the R dput for the yearly data:
https://pastebin.com/vEWhJkTy
Ok, I see by linking back to the old discussion that I already made my main point… Use a Fourier Series or Chebyshev polynomial with a moderate number of terms, and then evaluate the situation
https://statmodeling.stat.columbia.edu/2019/02/11/global-warming-blame-the-democrats/#comment-967288
Leads to this 25 term ortho-poly I assume: https://i.ibb.co/YbyJN3p/temp-By-Pres.png
That shows that since Nixon was elected all presidents have presided over an overall increase in global temperature. The typical size of the slope was about similar for all presidents, the size of the variation from time to time is consistent with things like El Nino, and there’s nothing to see here. Linear slopes fit to each president’s term is a very noisy way to estimate what’s going on in a trend, and the fact that it leads to some correlation is not particularly interesting from a scientific perspective.
Answering the bigger scientific question, in general people misunderstand how to estimate rates of change in timeseries, and it is far far better to fit a smoothing function through data and then estimate derivatives from the smoothed function than it is to estimate derivatives from short runs of very noisy data. Richard Hamming discusses this in several books he wrote, including things like “Digital Filters” and “Numerical Methods for Scientists and Engineers” if I remember correctly. We should reject a lot of time-series analysis that does what you did by breaking down times into short windows and fitting a line as this is a method that amplifies any measurement/modeling noise there might be.
So the answer is, yes, we should reject a lot of the scientific literature, and all for the same reason, which is that timeseries analysis often really sucks and many scientists don’t have the mathematical background to understand why, and the kinds of things they are likely to do such as fit a line to some noisy window of data is exactly the kind of thing that will enhance noise in the results leading to spurious ideas that wouldn’t appear with a more mathematically natural analysis.
Anoneuoid –
You say you don’t care WHY a supposed correlation exists.
The you say that a partisan differentiation in policies re climate change began during Carter’s campaign.
You give one lame campaign comparison with no attempt to even quantify what that means in actual policy differences, a political signal in public opinion, etc.
And when I point out Bush’s rhetoric about the importance of addressing climate change along with an associated policy initiative…
…you say you don’t care WHY a supposed correlation exists.
Pick an argument and stick with it, please.
For all the interesting things you have to say about NHST and other stuff, you’re p-hacking your logic.
And then you go back to it, again!
> It is also a fact that global warming first became a political issue in the late 1970s, I don’t see where you dispute this.
The elevant aspect, with respect to the question of reverse direction of causality, is when there was a significant partisan signal so as to make a material difference in anything.
I say it wasn’t during Carter’s administration, which you state is “A FACT,” grounded by pointing to an aspect of a difference in platform between Carter and Reagan which is climate adjacent.
You show no meaningful difference in public opinion. You show no systematic difference in policy implementation, which is a conjecture that’s directly contradicted by Bush’s political rhetoric, not to mention Bush’s policies, not to mention McCain’s political rhetoric.
Your determination of what is an isn’t a ‘fact’ is rhetorical gamesmanship.
It looks like you’re hiding behind “a political issue” to ignore that you can’t demonstrate any kind of consistent partisan signal either in public opinion or in administration policies, in association with party ID.
Mott and Baily?
OK, I’ll drop it now for the sake of my sanity and to not further try Andrew’s patience.
It’s a coincidence.
Back in the 1970s someone pointed out that in years that a team from the original American Football Conference won the Super Bowl, the stock market went up. At the time that was 100% right, but based on “only” something like a dozen years. As far as I know, nobody took it seriously ….but that didn’t stop it from continuing to work. I think the rule that an AFC victory meant an increasing stock market was right more than 90% of the time all the way through 2000 or so, something like a 35-year string.
There are a lot of phenomena in the world. 1% coincidences happen all the time; so fo 0.1% coincidences.
The moon is almost exactly the angular size of the sun. It’s so close that even the slight eccentricity of the orbits of the moon around the earth, and the earth around the sun, can lead to a total eclipse or an annular eclipse. It doesn’t really need an explanation.
I think it’s great that Anoneuoid raised this issue though. It’s s fun fact and I think it’s a good way to get people to think about issues like the difference between the mean temperature anomaly (which we don’t know precisely) and the estimated mean temperature anomaly; and the amount of year-to-year variation compared to the average annual change in the anomaly. These are good things to understand.
Phil, I think you are being too generous to Anonyd00d. Interesting temperature anomalies could have been brought up and discussed without the accompanying wacky, trolling (or mentally ill) conspiracy theories, which is mostly what Anonyd00d focuses on. There are so many better ways to have these discussions and make arguments. He may just be a sick person. Conversations with him go only one way–it’s like watching what happens to Wile E. Coyote–drop as many A.C.M.E. anvils on him as you wish, but he will keep going. How many times has Andrew either told him to take a break from posting or to knock it off? Let’s see if Anonyd00d can hold to the edict of 2 comments per post, in perpetuity; it could be a golden era for the blog.
Unanon,
Well, you’re entitled to your opinion.
The world is full of wackos who believe odd things, including conspiracy theories. Some of them are just idiots, or idiot-adjacent. Others, like Anon, are smart enough that, by being selective, they can build an edifice of facts to support their beliefs. I find that kind of interesting, and I sometimes find it instructive to figure out exactly what he has right and wrong.
In this particular case, I hope he’s so convinced that this pattern will continue that he’ll accept a wager. A “vitamin C and scurvy” level of certainty, that should be good for, what, 50:1 odds or something. I’m in.
This was the original blog post: https://statmodeling.stat.columbia.edu/2019/02/11/global-warming-blame-the-democrats/
I think a Bayesian estimate (including errors bars) of the anomaly using a whole variety of sources would be interesting to see. I’m sure someone has done it, but I haven’t seen it.
I always get a little concerned with how well to trust the climate data. Correcting for the urban heat effect, for instance, or a few years back when they revised all the satellite temperature data to account for drift. It’s a complicated enough subject that I doubt one person could be an expert on it all.
Regardless, I think the debate surrounding climate change is more complicated than just getting opponents to admit that it is happening. Even if you grant that it is happening, the debate comes to whether the costs of mitigating it are worth it and can you coordinate countries around the world to agree to mitigate it.
Realclimate.org is one place to get your questions answered by people who know what they are talking about. The reports of the International Panel on Climate Change is another, although generally more has been learned by the time the reports come out. The panel has addressed the question of whether it is worth the costs to cut greenhouse gas emissions (it is, in spades).
I learned the basic physics of global warming back in 1971, more than 50 years ago, and I still have a hard time wrapping my head around the slow and inadequate response to it. If ever something showed that people are not the rational actors that economics assumes, this is it.
Unfortunately many of the actors are “oh-so-rational”. If one’s livelihood, corporate strategies and so on depend on maintaining a sense of uncertainty about the science and its implications it is perfectly rational to support the efforts of misrepresentation that has been a feature of the response to the science especially through the 1990’s-2010’s. It’s almost impossible to maintain that nowadays and perhaps the corporate and political (in some quarters!) response has shifted away from misrepesentation towards greenwashing.
Apart from that, it’s a horrlbly difficult problem and the other “oh-so-rational” response relates to the reluctance to offset present losses in (perceived) wellbeing against future gains. It seems to me that the problems assoiated with global warming are widely recognised and are being more honestly addressed, and so we’re left with the fact that it’s not really possible to move towards the sustainable economies that are actually the only viable future for humans, at a rate that’s faster than the changes in our infrastructures and economies can realistically accommodate.
So we can’t get where we need to go fast enough? Isn’t that because we didn’t start soon enough? Suppose we started back in 1978, when Mercer’s article about the threat from the West Antartic ice sheet (https://www.nature.com/articles/271321a0) came out? Even a modest carbon tax back then would have made a big difference by now. At the time, I thought that article would get politicians’ attention, but I was wrong.
Hi John,
I didn’t say “…we can’t get where we need to go fast enough..” I said we can’t really get there “…at a rate that’s faster than the changes in our infrastructures and economies can realistically accommodate.” That’s always a real world limitation.
But yes, if we’d had a rational sense of urgency 30 years ago things might be a whole lot different and we could be further down the path towards sustainable economies.
John, Chris:
The problem is that the proposed solutions have always been unrealistic from a social and an economic standpoint and still are because the climate crowd has always been tackling the wrong problem.
The climate crowd has always viewed the problem as: how do we reduce CO2 emissions? The **actual** problem for the overwhelming majority of the population is: how do we reduce CO2 emissions **without negative economic effects**.
Tragically many on the left **WANT** negative economic effects, so they **insist** on policies that will create them. The left hates development and wants it stopped. Even now, “climate is crying” Gov Inslee (and PE teacher Sen Murray) of WA want to remove hydroelectric dams to restore fish runs on the Snake River. Say what you want about the emotional value of fish, but the **economic** value of those fish will never come close to the cost of removing the dams, much less the power lost from the damns, the transportation lost from the removal of locks, or the alternative uses of the $25B. Same with the Pebble Mine project in Bristol Bay: the value of the minerals in the project outweigh the value of the Bristol Bay fishery by at least an order of magnitude and possibly two orders of magnitude. But the objective of Murray and Inslee here is to stop development, period, no rational argument to be brokered. No amount of human suffering is too much to save the fish!!!
Of course Inslee and Murray have contrived a “study” that claims glowing economic benefits from dam removal. Forgive me while I choke. And of course the EPA contrived to block the Pebble Mine.
These economic policies have direct negative impacts: the current navigation system on the Snake supports low cost barge transportation of bulk food commodities. What happens when that transport shifts to rail? Price goes up. Food prices go up. People’s quality of life is negatively impacted. This is one small example but almost all CO2-regulating climate policy has a similar effects: it drives up energy costs which in turn drive up the cost of necessities and drive down the quality of life. Same for Pebble mine: housing costs high? Part of that is the price of copper. Part of that is the cost of lumber. Despite the housing boom in Inslee’s WA state, lumber production there is less than half what it was in 1970. Hmmm…I have no idea why housing prices are so high!!! Why ever are people living in the street??
Reality: climate and environmental policies have major negative effects. When the climate crowd lined up on the “stop growth” environmental side of politics, they sealed their own fate. They have no one but themselves to blame.
Chipmunk –
> climate and environmental policies have major negative effects.
I’ll just speak to climate policies.
Please detail how you account for negative (and positive) externalities in your calculus. I’m particularly curious how you figure in factors such as carbon health impact of particulates (from burning fossil fuels), the geo-political and associated economic costs of keeping oil flowing, etc.
I ask because I have yet to find anyone who claims they know there’s a net cost ACO2 mitigation policies who takes seriously an accounting for negative externalities. I’m hoping you’ll be the exception
Thanks in advance.
“Please detail how you account for negative (and positive) externalities in your calculus. ”
If you believe that can be done, you’re mistaken. You’ve invested far too much faith and far too little knowledge in your beliefs.
The general guide to the relative benefits of any activity is cost. On the whole it’s likely that most negative externalities are offset by positives (recognized and accounted for or not) – which is why so much behavior with supposed negative externalties continues unabated – including the production and use of fossil fuels. But even that might have waned by now had environmentalists not blocked nukes. Energy is not an option for modern life, and the more it costs the less of modern life will exist.
Whatever the case you can readily observe the cost of environmental policies, can’t you? Widespread homelessness and exploding housing costs, for one. These can be linked directly to both environmental policy in general and climate policy in particular – for example the preservation of US standing timber as a carbon sink, rather than using US timber for US housing; and the general US policy of restricting resource development and importing those resources over long distances when in fact they’re readily available at equal or less cost locally.
Developed countries have already invested substantially to reduce particulate matter with a massive benefit. It’s clear that, without that investment, there would likely have been a substantial impact on life expectancy. However to claim that developed countries must stop producing particulate matter altogether because of the health impact is preposterous. Life expectancy in developed countries is more than double what it was in 1870 and still rising. That is happening **because** of, not **in spite of** the wealth our economy creates.
Question for you: how many life years of how much relative quality are gained by the trade that generates the particulate matter?
chipmunk
I should know better, but here goes:
“On the whole it’s likely that most negative externalities are offset by positives (recognized and accounted for or not) – which is why so much behavior with supposed negative externalties continues unabated – including the production and use of fossil fuels.”
Your ability to reach such a conclusion for such a complex matter is unrivaled. I can certainly agree that virtually all activities with negative externalities also have positive spillovers. But both are difficult to quantify and the conclusion that they offset each other (or worse, that the benefits exceed the costs) is based solely on your beliefs. The fact that these activities continue is poor evidence that the benefits exceed the costs – the literature on transactions costs is relevant here. Further, your insistence on one dimensional measures of costs and benefits is a belief, not a natural law.
Again, I don’t think it is worth continuing this discussion as neither of us is likely to be convinced to change our minds and I don’t want to try Andrew’s patience. So, I’ll let you have the last word.
Chipmunk –
> If you believe that can be done, you’re mistaken. You’ve invested far too much faith and far too little knowledge in your beliefs.
I think it can be attempted, and more or less well done. Life is sub-optimal.
However, when someone says they know the sign of the “cost” of mitigating ACO2 emissions (let alone the size of the “cost”), I have to wonder how they balanced out the positive and negative externalities to reach that conclusion. I see this frequently among people who say they know the “costs” will be huge, and alongside reaching that conclusion figure in all the many positive externalities resulting from access energy (such as, say, educating women in developing countries and thus reaping all the ancillary benefits).
> The general guide to the relative benefits of any activity is cost. On the whole it’s likely that most negative externalities are offset by positives (recognized and accounted for or not) – which is why so much behavior with supposed negative externalties continues unabated – including the production and use of fossil fuels.
I know that you take that dynamic to be an article of faith, and while I don’t even if I do accept that for the sake of argument there’s still a problem because the of how the negatives being offset by the positives might be distributed. For example, supplying fossil fuel energy at a relatively low price (as distinguished from cost) might prop up a higher GDP which looks great when aggregated, but when broken down shows that poorer people living in poor countries or neighborhoods disproportionately absorb costs such as higher asthma caused by breathing lower quality air. Or they might absorb higher costs from having more proportionally more people dying in wars fought to keep oil flowing. Or they might absorb the costs from a lack of freedom in countries where despotic autocrats rake in huge sums of money from selling oil.
> But even that might have waned by now had environmentalists not blocked nukes. Energy is not an option for modern life, and the more it costs the less of modern life will exist.
Lol. Any and all road leads to owning the libz with you. At any rate, sure, big bad environmentalists (who obviously dominate the US political landscape in your world view but not quite so much in mine) have played a role in depressing nuclear energy in this country. But so have pie-in-the-sky libertarians who pay lip service to nuclear energy but systematically build roadblocks against the kinds of centralized energy policy formation, and federal spending, that accompany nuclear power in all the countries that have significantly more of it than we do (with the possible except of “socialist” Finland). And that’s in part because the long time horizon for payoff for the massive investment with high liability required for nuclear development isn’t merely not particularly attractive to people with that kind of money because of regulations (there’s a complicated clause for ya’) but also just because those are the conditions of the context. Why invest so much money in nuclear when you can get higher returns on a shorter time horizon so many ways. That’s why federal funding is needed and I demand evidence of your unrelenting lobbying your representatives to get it done!
> Life expectancy in developed countries is more than double what it was in 1870 and still rising. That is happening **because** of, not **in spite of** the wealth our economy creates.
Wow. That’s one hell of a slippery slope. Mt. Everest-sized.
> Question for you: how many life years of how much relative quality are gained by the trade that generates the particulate matter?
I don’t have a stock answer other than “It’s complicated.” And that’s where I began here. You seem to have a clear answer to that question. I’ve never said that I do and I’m skeptical of people who think they do. So that’s why I’m asking you for your answer to that question as you comments imply that you’ve done the calculations., Or maybe you’re just assuming an answer as a matter of faith?
Chipmunk –
If my response was too dense and poorly written to get through, just read Dale’s.
Yikes.
The longer response seems to have disappeared. Just for the hell of it, if it doesn’t reappear I’ll try reposting.
Trying longer response again:
Chipmunk –
> If you believe that can be done, you’re mistaken. You’ve invested far too much faith and far too little knowledge in your beliefs.
I think it can be attempted, and more or less well done. Life is sub-optimal.
However, when someone says they know the sign of the “cost” of mitigating ACO2 emissions (let alone the size of the “cost”), I have to wonder how they balanced out the positive and negative externalities to reach that conclusion. I see this frequently among people who say they know the “costs” will be huge, and alongside reaching that conclusion figure in all the many positive externalities resulting from access energy (such as, say, educating women in developing countries and thus reaping all the ancillary benefits).
> The general guide to the relative benefits of any activity is cost. On the whole it’s likely that most negative externalities are offset by positives (recognized and accounted for or not) – which is why so much behavior with supposed negative externalties continues unabated – including the production and use of fossil fuels.
I know that you take that dynamic to be an article of faith, and while I don’t even if I do accept that for the sake of argument there’s still a problem because the of how the negatives being offset by the positives might be distributed. For example, supplying fossil fuel energy at a relatively low price (as distinguished from cost) might prop up a higher GDP which looks great when aggregated, but when broken down shows that poorer people living in poor countries or neighborhoods disproportionately absorb costs such as higher asthma caused by breathing lower quality air. Or they might absorb higher costs from having more proportionally more people dying in wars fought to keep oil flowing. Or they might absorb the costs from a lack of freedom in countries where despotic autocrats rake in huge sums of money from selling oil.
> But even that might have waned by now had environmentalists not blocked nukes. Energy is not an option for modern life, and the more it costs the less of modern life will exist.
Lol. Any and all road leads to owning the libz with you. At any rate, sure, big bad environmentalists (who obviously dominate the US political landscape in your world view but not quite so much in mine) have played a role in depressing nuclear energy in this country. But so have pie-in-the-sky libertarians who pay lip service to nuclear energy but systematically build roadblocks against the kinds of centralized energy policy formation, and federal spending, that accompany nuclear power in all the countries that have significantly more of it than we do (with the possible except of “socialist” Finland). And that’s in part because the long time horizon for payoff for the massive investment with high liability required for nuclear development isn’t merely not particularly attractive to people with that kind of money because of regulations (there’s a complicated clause for ya’) but also just because those are the conditions of the context. Why invest so much money in nuclear when you can get higher returns on a shorter time horizon so many ways. That’s why federal funding is needed and I demand evidence of your unrelenting lobbying your representatives to get it done!
> Life expectancy in developed countries is more than double what it was in 1870 and still rising. That is happening **because** of, not **in spite of** the wealth our economy creates.
Wow. That’s one hell of a slippery slope. Mt. Everest-sized.
> Question for you: how many life years of how much relative quality are gained by the trade that generates the particulate matter?
I don’t have a stock answer other than “It’s complicated.” And that’s where I began here. You seem to have a clear answer to that question. I’ve never said that I do and I’m skeptical of people who think they do. So that’s why I’m asking you for your answer to that question as you comments imply that you’ve done the calculations., Or maybe you’re just assuming an answer as a matter of faith?
John –
> If ever something showed that people are not the rational actors that economics assumes, this is it.
I think the definition of “rational actor” is complicated outside as well as inside of economics. Outside of what Chris mentions above, the a critical component is that what is or isn’t a rational act, particularly in this context, is influenced by cognitive biases such as cultural cognition and motivated cognition.
chipmunk
As usual, you espouse standard economic tripe. You are absolutely correct about the reason that environmental policies (including climate change policies) are difficult to enact and run into so much political opposition. However, your statements about the relative values of environmental resources (fish, etc.) and more marketed resources (e.g. minerals) simply reflects existing market values and do not permit any other sorts of valuation. Some issues are matters of value in the larger sense of the word, and involve competing beliefs. Markets are not always the best ways to resolve such conflicts. Political systems exist for reasons, one of which is to express, argue, and resolve differing belief systems. I probably share your skepticism about how these political systems function, but ignoring them and replacing them with markets is not a solution. Let’s not debate whether there is such a thing as the right of a species to exist: I don’t believe I will change your mind and I know you won’t change mine. And whether you or I are in the majority on such things does not interest me.
Although doing it explicitly would be nice, I think this graph here is more or less that. There is certainly no “global average temperature meter” we can consult to get the right answer, so the numbers on this graph come from a large variety of sources and are integrated together through some kind of means.
Would be happy if anyone who is more familiar with the data availability could point to a single source big old curated dataset of measurements and metadata? It’d be nice to have some baseline credible archive. I see on github about 100 pages of results for search on “climate data” for example. A lot of them are packages to access some APIs etc.
Try https://berkeleyearth.org/data/
The project is overseen by a prominent UCB physicist who didn’t trust the global temperature analyses and wanted to work it all out independently.
Nice, thanks for that. It does seem to be pretty much the thing.
Sure, but the information is easy to track down. For example, a 5 minute perusal of the NASA Giss website (for example) will allow you to discover that there are “error bars” (95% confidence intervals) associated with this data, that the plots aren’t “average global surface temperature” but are “temperature anomalies”; i.e. the difference in averaged temperature relative to some chosen period, and one can also easily find the latitude-dependence and more detailed spatial distribution of warming. You just need to look.
The use of temperature anomalies allows for average global temperature changes to be determined pretty reliably even tho, as you say, the notion of an “average global temperature” is rather meaningless. You can then dig just a little deeper to determine the extent that temperature changes are spatially correlated (pretty high), the contemporary and historical spatial coverage of temperature stations, the contribution of satellite measurements to surface temperature changes (hint: very little contribution in the 19th century) and so on.
Sure the data is collated and used by climate scientists as research effort and research tool as well as a resource to inform policy. The latter is no doubt why so much effort is made to attempt to rubbish climate science! But just like biological researchers study the natural world in order to understand it, provide data to test models and make predictions (and inform policy), so climate scientists study the natural world in order to understand it, provide data to test models and make predictions (and inform policy).
>For example, a 5 minute perusal of the NASA Giss website (for example) will allow you to discover that there are “error bars” (95% confidence intervals) associated with this data
>You just need to look.
But that was one of the main points of my comment. Why? Why should I have to look? When I present a plot of my model fit to data or the inferences made or any plot of unobserved parameters, then I make dang sure that I plot the uncertainty in the estimates (at least as best I am able). I don’t really see an excuse for distilling the information so far down that apparently unobserved parameters become what looks like observed data points in a plot. Why not just use the nicer plot with error bars in the first place? The links that people gave above show some nice plots by some people who I’m sure have done a lot of hard work. Why not use those? I guess in light of the comments above, I really don’t see a point (pun intended) to a plot like this.
Very few graphs of data intended for public consumption have error bars. It’s not just climate data. What are the error bars on immigration numbers? What about on GDP or other economic measures? Poll numbers are one of the very few exceptions, and even then the error bars are stated verbally in the fine print and not put on a graph.
If those “data” points are actually parameters from some model, then I think they should attempt to include the uncertainty in the plot.
I just realized (1) I once read an entire book on how this is modeled globally, and (2) I can’t identify the book or (3) really remember any of the details. Oy. Good book, though, I think!
I read this as a political science post using evolution of attitudes about climate change as the real world example, but not about climate change per se, unlike many of the commenters. I find the insight about directional versus proximity voting especially helpful because I always have trouble understanding why the center does not hold in so many important situations. Ironically the center could collapse again in 2024 due to efforts of those claiming to want to shore up the center who ignore the essentially binary nature of our system.
I agree that the actual point of this post is fascinating. I had never heard of directional versus proximity voting. Proximity voting seems “natural” in terms of what people would do — vote for candidates that reflect their own wishes. Is there any evidence that people actually vote directionally, making a conscious decision to “overshoot” their target?
I also feel like there’s a neat animation that could be made of this — maybe people balancing a seesaw either by shifting positions around a center or adding weights to the end, which works until it doesn’t…
I haven’t voted directionally myself, but I have appreciated when certain points of view, which I don’t endorse myself, nonetheless move the “Overton window,” so that issues that have been ignored get discussed and views I do endorse are nearer the center of the window, and thus more likely to be adopted.
Chris:
Directional and proximity voting are familiar ideas within political science. It was interesting to think about how this applies to pundits. It strikes me that edgelord behavior can often be seen as a form of directional advocacy, and I wonder how the Freakonomics authors think about their edgelord-like promotion of climate-change denial in an environment where climate-change denial is being pushed by political extremists. My guess is that they’d prefer for that old post to be forgotten, just as they might wish they hadn’t promoted the noise-mining of that beauty-and-sex-ratio researcher, and as usual I’m frustrated that they just ignore their past mistakes rather than confronting them. We can learn soooo much by looking carefully about our past mistakes, and the Freakonomics authors are in an excellent place to do so, if they’d only try.
” I’m frustrated that they just ignore their past mistakes”
Could be your mistake, not theirs. Seems likely you’re mistakenly assuming that their objective is to uncover some truth and communicate that truth to the public. But that’s not their objective at all. Their objective is to make money, either directly via advertising or by generating publicity for their brand. They succeeded! Here you are still discussing it how many years later. Where’s the mistake?
Virtually every major news outlet does the same thing every day. The purpose of media is to make money. Whether it provides accurate information or not is a secondary concern to it’s participants – one that concerns them only insofar as a) inaccuracies undermine their ability to make money in the future; or b) earn them general social opprobrium that makes them feel bad. The former almost never happens and the latter is not common. So the hype and misinformation goes on.
Rational discussion of has ended in American media.
> But that’s not their objective at all. Their objective is to make money,
BInary thinking strikes again!
So instead of 100x more warming under democrats, p-hacking can get it to only ~5x. And the rankings are still dominated by democrats at the high end.
More importantly, what does this polynomial predict for the rest of Trump + Biden presidencies? Does it fit with what we observed?
I’m not sure what makes you think it’s 5x, looks like basically the same to me from Nixon to beginning of Obama. The big difference if any is going to come during the Obama presidency. But in general we have a mechanism for recent warming to be faster than past warming (ie. we know more emissions recently from China, India, etc).
You can’t use this kind of polynomial fit to outside the data. There are nice Bayesian techniques to deal with that, but they basically require you to specify a likely behavior, and maybe you’d need some covariates like CO2 levels and ENSO events etc.
I just added up the smoothed values in the top chart since Carter. For three democrats its ~0.7 C, while three republicans (ignoring one year of Trump) is ~0.2 C.
Including the Trump year would subtract 0.145 from the republican number. You can also eyeball the viopoints chart to see its about the same or bigger difference if all the data is used.
I don’t know why I would want to smooth the data in this way anyway. If the model can’t extrapolate to future data then it doesn’t mean anything.
You smooth this data to discover a better estimate of *rates* of change of an underlying process from noisy data. Rates of change are change in y divided by change in x. Presidents don’t all have the same tenure.
https://i.ibb.co/YbyJN3p/temp-By-Pres.png
If I understand your table, I get the following estimated rates of change since Nixon’s election. The final number is in degC/yr
Nixon: .053/5 = .0106
Ford: .088/3 = .0293
Carter: .107/4 = .0266
Reagan: .092/8 = .0115
BushHW: .081/4 = .0203
Clinton: .167/8 = .0209
BushW: .081/8 = .0101
Obama: .409/8 = .0511
So Nixon was a little low, but Ford was a little high compared to Carter, Reagan was lower, then BushHW was basically the same as Clinton, BushW was a lot less than Obama.
So on overall if there is any effect it’s that Reagan had lower rate of change, and Obama had more and this makes the GOP presidents preside over lower rates of change and the Dems preside over higher rates of change.
But they were 30 years apart, and of course we expect warming to be going a lot faster in 2010 or so because China and India and other developing countries started massive Coal burning well after Reagan.
So basically it’s nothing to see here that wouldn’t be predicted by just “later times have higher CO2 and higher rates of change” plus some other variations like ENSO or Mt Pinatubo, or that Icelandic volcano or whatever else idiosyncratic there was through time.
You don’t use this kind of thing to extrapolate because it’s entire purpose is to interpolate derivatives with lower noise.
Say you are a detective who gets an image of a license plate. Would you rather see the top (original) or bottom (smoothed) picture?
https://i.ibb.co/PCn1PGg/platereader.png
If anything I’d think to check at monthly resolution, and this data is *already* heavily smoothed (and otherwise processed) over both time and space.
It is a very interesting choice though. Use complicated methods to blur out any patterns in the data because you assume they are “noise”. Like ENSO and volcano activity gets treated as noise rather than part of the climate system.
Anon:
Enough! I don’t know if you’re trolling or just genuinely confused, but in any case please restrict yourself to no more than two comments per post in the future. Thank you.
Take some function, like sin(2*pi*x) evaluate it at the points 0:0.05:1 and add measurement noise to it like normal(0,0.2). Then fit sin(2*pi*x) to that data using least squares, and evaluate the derivative of the function at a point like say 0.2
Also evaluate the difference (data at 0.25 – data at 0.15) and divide by 0.1 as an alternative estimate of the derivative at 0.2
Do that a bunch of times, and evaluate the distribution of the two techniques, which gives a distribution that is tighter spread around the true value of the derivative of the function at 0.2 (which is 2*pi*cos(2*pi*0.2) = 1.9416)
We are trying to evaluate decades long trends, using data that is heavily averaged in *space* to get a “global average” but is *not* smoothed in time. The idiosyncratic noise in space is meaningless for the decades long trend but appears in each data point as “noise” when it comes to evaluating derivatives. Without an appropriate level of smoothing, the derivative operator *enhances* that noise. Think about splitting a timeseries into a fourier type series:
sin(k*x) + sin(2*k*x) + sin(3*k*x) + sin(4*k*x) … etc
if we take the derivative we get:
k*cos(k*x) + 2*k*cos(2*k*x) + 3*k*cos(3*k*x) + 4*k*cos(4*k*x)… etc
In other words the high frequency, short-timescale variation is amplified by an amount proportional to its frequency. If on the other hand we are interested in estimating stuff that takes about time 8/k then all the stuff that wiggles up and down 24 times in that big of a time interval is not informative, it’s just throwing our estimate of how much things changed in 8/k years off because it’s amplified by a factor of 4 over what we’re interested in.
It’s not that volcanos and such aren’t a part of the climate, it’s that their effect on a given 1 to 2 year timescale is substantially larger than the net effect on a 8 or 30 year timescale.
The data you’re looking at is heavily averaged over *space* in any given year, but does not take into account as far as I know the fact that global temperatures wiggle back and forth around a function that changes relatively slowly in time. Applying that smoothing in time produces better estimates of rates of change of that underlying trend.
Andrew, I was writing my response while you posted yours. I’ll leave it at that. Hopefully the data processing discussion is at least somewhat relevant to someone else with a timeseries question.