How Climate Scenarios Lost Touch with Reality
issues.org
issues.org
And before anyone says so, no it has nothing to do with the supposed lessened attention span smartphones have caused. I'm not wasting 15 minutes of my life reading about social media recommendation algorithms [1] when not even a single paragraph is dedicated to at least some rudimentary technical explanation behind them.
Same goes for 30-minute YouTube videos attempting to maximize monetization for subjects whose explanation in less than 10 minutes for an average intelligence person is adequate.
Here's also a recent example of ideal, imo, technical writing that is sadly getting more scarce:
Recipes tell you how their great grandmother moved to the Americas and made the first apple pie ever. And then detail the entire history of apples and how to make the entire recipe. And then, 70% of the way down the page, right before the comment section, is what you wanted at first glance: what’s in this pie, and how long will it take me to make?
It’s everywhere.
I agree, so I used ChatGPT to summarize this for us:
In the article "Climate Change Scenarios Have Lost Touch with Reality," written by Roger Pielke Jr. and Justin Ritchie and published in Issues in Science and Technology, the authors argue that the climate change scenarios used by policymakers and scientists have become disconnected from empirical reality. They suggest that this has led to an overreliance on untested and unrealistic assumptions, which in turn undermines the credibility of climate change projections and makes it more difficult to develop effective policies.
The authors note that climate change projections are typically based on a set of assumptions about future greenhouse gas emissions, which are then used to generate a range of potential climate outcomes. However, these assumptions often fail to account for a range of important factors, such as economic growth, technological innovation, and changes in land use patterns, which can have a significant impact on emissions and climate outcomes.
Pielke and Ritchie argue that policymakers and scientists need to take a more pragmatic and evidence-based approach to climate change projections, one that is grounded in real-world data and incorporates a wider range of factors. They suggest that this could involve developing more diverse and flexible scenarios that are better able to capture the complexity and uncertainty of the real world.
Overall, the authors stress the importance of ensuring that climate change policies are based on realistic and evidence-based projections, rather than relying on overly simplistic or unrealistic assumptions. They argue that this will require a fundamental rethinking of the way that climate change scenarios are developed and used, and a greater emphasis on collaboration and dialogue between policymakers, scientists, and other stakeholders.
edit: it's not due to any antipathy to AI. I love using ChatGPT myself.
Why?
What point are you actually trying to make? It seems like you’re trying to equate one with the other, especially given your novel use of the word “denialism”. I don’t want to put words in your mouth, but I’m still unclear on what exactly you are implying.
Roger A. Pielke Jr. is an American political scientist and professor, and was the director of the Sports Governance Center within the Department of Athletics at the Center for Science and Technology Policy Research at the University of Colorado Boulder.
He got that from his father who was also an AGW denialist (a climate change believer, but a denier that humans play any meaningful role), albeit one with more relevant credentuals:
Roger A. Pielke Sr. is an American meteorologist with interests in climate variability and climate change, environmental vulnerability, numerical modeling, atmospheric dynamics, land/ocean – atmosphere interactions, and large eddy/turbulent boundary layer modeling.
- Atmospheric CO2, 1960-now.[1]
- Ocean temperature, 1880-now. [2]
- Annual arctic sea ice minimum area, 1978-now. [3]
[1] https://gml.noaa.gov/ccgg/trends/
[2] https://climate.nasa.gov/vital-signs/global-temperature/
Well, because of that feedback loop. Coal is still a cheap way of making energy, as it is in the business-as-usual economics of the scenarios. If we say "coal is not a problem", the pressure shifts away from coal and its usage may linger longer than it should.
The scenarios are about the forces impacting the trends, not the trends themselves.
It's true that we reduced the usage of coal in last few years. It's a big assumption that it happened because of a worst case scenario popularized by IPCC.
From this article[1], IPCC's 8.5 was based on the assumption that between 2017-2100 we will use 5x as much coal as we had done in our entire lifetime. That's a like a "coal boom". The world was already moving away from coal, even before the report, and at best the report and faulty scenario created more panic.
> Everyone knew about the decades of coal-powered economic growth in China, but those working closely on the future of energy had already grown somewhat skeptical that the same model would be deployed across the developing world — and even more skeptical that the rich nations of the world would ever return to coal in a sustained way.
> The most conspicuous example was an emissions pathway called RCP8.5, which required at least a fivefold growth of coal use over the course of the 21st century. Because it was the darkest available do-nothing path, RCP8.5 was reflexively called, in the scientific literature and by journalists covering it, “business as usual.”
Feedback loop as you call it probably slowed down the pace turning the whole thing into a culture war, creating mass hysteria, and getting people to fight each other when they were likely already moving away from coal. We don't have a control group in this case, and climate action is needed, though I am not sure if creating panic was justified at any time.
[1]: https://www.nytimes.com/interactive/2022/10/26/magazine/clim...
What made it a culture war was the fossil fuel companies pumping billions of dollars into politics for decades. The Republicans didn’t used to be the anti-science party – that took years of investment building think tanks, the media bubble, etc. which turned that into a litmus test for candidates and making any sort of consensus off limits (a Republican lobbyist I know socially used to represent the fishing & hunting interests. Around the Obama era, those old coalitions fell apart as politicians would tell him that he was right but there was no way they could support environmental protections and survive the primary media cycle).
It’s very easy to look at coal declining and say it was inevitable but that ignores how it was hard won at each step. As you had countries like China booming it’s really not hard to see how deniers having a little more clout could have lead to locking in decades of elevated emissions.
https://xenetwork.org/ets/episodes/episode-176-climate-scena...
To see all episodes related:
Another less important example is helium : helium is necessary for high end electronics, and electronics are necessary for modern cars. However car manufacturers are generally completely oblivious of their very real dependency upon helium, which came mostly from the US and Russia.
I've just watched an interesting conference by a researcher who's working on new, complex models to build scenarios taking all these entanglements into account, managing thousands of different resources across the whole globe. At the moment there's like 5 of them working on these. Unsurprisingly, these complex, stochastic models treated with lot of Monte-Carlo analyses and now even AI are far from being stable and precise enough for projects such as IPCC. So that's where we are for now.
Classical economics are wildly out of touch with reality here (Nordhaus even got the Nobel in economics with a completely bonkers linear extrapolation of climate effect on global GDP), and these sorts of interdependencies are everywhere.
So OK, I've got it, the scenarios are too simple. But the IPCC working group II is composed of mainstream economists, and mainstream economists are wildly out of their league (in my opinion I'd go as far as saying that macro economy is a pseudo-science at this point anyway), so they're doing what's possible in a mainstream environment. Too bad.
https://www.pnas.org/doi/10.1073/pnas.2007117117
Relevant image: https://www.pnas.org/cms/10.1073/pnas.2007117117/asset/56610...
> It associated the RCP scenarios with not just plausibility but also likelihoods when it labeled the scenario leading to the greatest amount of climate change, called RCP8.5 (indicating a radiative forcing of 8.5 W/m2 in 2100), as the single business-as-usual scenario of the set.
> For instance, RCP8.5 projects to 2100 a six-fold growth in global coal consumption per capita, while the International Energy Agency and other energy forecasting groups collectively agree that coal consumption has already or will soon peak. Also, RCP8.5 foresees carbon dioxide emissions growing rapidly to at least the year 2300 when Earth reaches more than 2,000 ppm of atmospheric carbon dioxide concentrations. But again, according to the IEA and other groups, fossil energy emissions have likely plateaued, and it is plausible to achieve net-zero emissions before the end of the century, if not much sooner. Today, projections that carbon dioxide emissions from fossil fuels will increase dramatically for the next 50, 100, or 300 years are simply implausible.
It’s an eminently reasonable request, one that should be celebrated. I don’t know why it’s necessary to write thousands of words impugning the scientific integrity of climate science.
What is much scarier to me are possible effects of GHGs that could make any emissions path much worse. Methane levels seem to be increasing, which could make current projections obsolete very quickly.
Put differently, are the models precise but inaccurate or both? If inaccurate only, are they still instructive?
Yes, you have to look at the whole picture. But you also have to look from where you are. Not where someone else thought you could be. Right?
But also: if we take drastic action “wrongly” and the end result is a healthier plant? I’m good with that.
I'm reminded of folks that were against LED lightbulbs because folks could just turn off their lights a few hours earlier. Not only would that have been a drastic change in behavior, it is comically incomparable in impact. To get the same impact, you would have to turn off your lights basically all day to get close to the same change. Such that sometimes we do have much better levers than previous drastic changes could have even dreamed of achieving.
As such, it really depends on the drastic action. For the most part, I don't see how aggressive goals for electric vehicles is harming folks. So if that is what we are looking at, fully back on board with you.
With a major caveat of not liking fear mongering. Honesty of where we are is important for its own reasons. Trust in future plans, at the least.
That is, are the models bad, or the scenarios? The models don't control the inputs, after all. If it is just the inputs are different from actuals, that is very different from the outputs are wrong
Scientifically climate change has been a growing concern since at least 1972. In 1990 there was zero support politically for the sort of action that is now starting.
The IPCC's work still matters, the political changes are not bedded in ( see the GOP in USA that is still clinging to its old ideas, but not for much longer). But the urgent, political work it was established for is almost done.
They can take some credit if we save the world
Droughts, wildfires, hurricanes/typhoons seem as real to me as "scientists" continue to squabble. Critical viewpoints are integral to the "integrity of science", but without application/action where is the actual contribution?
Maybe the article authors are suggesting: just set at RCP2.6 or favor the IEA baseline range and call it a day? I would think they would lose funding, speaker fees, followers, etc...
One missing variable in the socioeconomic assumption input to RCP assessment is the proliferation of dis/mis information or lack of open data access.
As the west looks on, some global partners are displaying excellence in cooperative data sharing and full disclosure on consumption patterns to help inform policy development: https://youtu.be/gfIoS9NB0Wg
Stick that in RCP input.
Are you serious?
Two tourists with anecdotes
The topic is science and data not tourist's reckons
IPPC reports are too conservative because insufficient attention has been paid to the importance of tipping points, feedback loops, and outlier events forecasts. The nature of consensus building has tended to marginalize more extreme scenarios.
Humanity already has a non-zero probability that wars and general political trends will reverse technological progress and waste the resources needed to adapt to climate change and transform the economy. Then the RPC8.5 scenario might be too optimistic, not too pessimistic.
Consider that if highly accurate models were really possible someone could make one that would accurately predict NFL game outcomes. When that happens we will know it because Sports Books would no longer take bets on NFL games. Sports Books profit because they get a slight statistical advantage which pays off due to bettor volume.
So we know they can't make an accurately predicting model for football. How accurate can they possibly get with climate change modeling? It's orders of magnitude more difficult. Football is formally rule bounded and tons of data on a long history of games is readily available. In contrast, the vast majority of climate data is based on theoretical methods. This isn't bad, it's just far less precise because the variance of all those theory based estimators accumulates across the model, widening the confidence interval of its outputs to a point where it may not be useful.
Finally, climate scientists are people. They have careers, families, financial responsibilities. What happens if they say the climate is fine? Some other crisis gets the funding and their careers are destroyed. They can't just switch to a different science, they've invested a lot of time and money becoming a specific thing.
I found this article very wordy and skimmed parts, but it looks to me like a goalpost moving exercise to explain why their aggressive predictions for the present time didn't come true.
The book does not "stop taking bets" on the entire event in question. The book simply has to limit the "sharp" individually. Of course then we get into the whole cat and mouse game of chasing sharp action through straw bettors and playing games of information theory (i.e. how much money can a sharp win while still plausibly not outing themselves as an obvious sharp).
The book only would have a serious problem, if the entire world became sharp and every bettor was betting sharply. But alas there you have a paradox: anyone who is currently a sharp, has a strong self interest to preserve the secrecy of their models and methods, because knowing something that nobody else knows is precisely what creates their edge. And so therefore, as long as sharps have this self-interest, then there is no problem for books to keep operating, and the problem is reduced to one of identifying and limiting sharps occasionally (basically a "cost of doing business" that books acknowledge).
The only way for the scenario you describe to happen, is for disinterested actors with no thought of profit for themselves, to not only "solve" a particular market, but also release it to the entire world and make it a commodity. I'm not saying it's impossible, but I'm saying that normal economic incentives work against this outcome.
The sharp works on inside knowledge, not advanced mathematic modeling. The sharp is someone with inside information, e.g. star player twisted his ankle recently and no one knows.
Bit if any moron could pull up a perfect model of the game outcome they'd all do it and the book would have no line to play. Books are happy to take action on anything where they have the advantage and are always wary of sharps ( won't take their money if known).
I really could have picked any prediction model though so we shouldn't get hung up on sports Books.
For example, why are banks still lending big money on expensive sea side property? Shouldn't their internal modeling be showing that these places will soon be flooded? Oceanside property should be a bargain right now. It isn't though, because the people on charge don't actually believe in the climate change models they push.
At least as accurate as weather reports, which can be pretty reliable days in advance. The outcome of a football match often can't be predicted accurately during the match. This should suggest that the atmosphere and a sport are qualitatively different things.
> What happens if they say the climate is fine?
They go into some other statistical modelling field, probably to do with meteorology, and we start to get accurate weather reports two weeks in advance.
It's not like high-level sports players, who have specialised into that sport, and are unlikely to ever be high-level players of any other sport. (If anything, there are incentives for high-level sports players to throw games to keep them more exciting! Maybe that's why we can't predict the games? It's not, but what if it were?)
The models created by the oil companies back in the 70s seem to be “breathtakingly accurate” [1]
Second, the problem with your analogy is that with football you expect the prediction to be X team will win, and similarly expect the predictions to be X weather event at $place will occur. The problem with your expectations of predictions regarding the latter is the sample space.
The sample space is so large that extreme weather events will occur $somewhere. Predicting exactly where and their trajectories is difficult due to chaos and the complexity of the system. There is a bound on our capacity to predict exact events long ahead of time. However, what we can do is give a probability of the event occurring over a large enough region. When you scale said region to the earth, we can ignore very small details and focus on the bigger picture, which is that the variance in weather in creases.
The problem with an increasing variance is that extreme events occur more frequently in all directions. Meaning high and low temperatures, droughts and floods and so on.
To observe this in action, a year or so ago europe had severe floods that killed many people and simultaneously over the next summer experienced historic droughts.
The sample space strikes again and we saw historic droughts in China too.
[1] https://www.theguardian.com/business/2023/jan/12/exxon-clima...
[2] Firmament, by Dr Simon Clark
2. Scientists, especially young ones, have an incentive to challenge dominating scientific views because that is how you make yourself known in the scientific community. Being known can be transformed to getting the next academic position.
Variance accumulates across models. The less actual data you have the wider the variance is on your estimators. When your data is mostly projected the confidence interval on your outcome makes it meaningless, e.g. average temperatures will change by plus or minus 20 degrees.