basically everyone agrees that the climate is changing, has always changed, and will always change. there is debate over how much of climate change is anthropogenic, but this is irrelevant to the point that predictive climate models are functionally useless, due to the complexity of the thing it's modeling.
[0] https://www.scientificamerican.com/article/clouds-may-be-the...
downvoters: I would appreciate an explanation as to how my understanding of this situation is inaccurate. my views are never set in stone, I'm always open to learn new things. I just don't see how, short of a complete holistic Matrix-style simulation of the planet, one can obtain meaningful predictions from a data-driven climate model. I'm wrong about a lot of stuff all the time and I'd like to know where my understanding of this situation is incomplete or inaccurate.
That is false. Although you’re correct that climate is very complex and it is impossible to model every detail. The notion that you cannot make valid conclusions from modeling because a set of observations do not match models is fallacious because it discounts the correct conclusions that climate models have achieved. For example, the first climate models from the 1960s predicted that the mesosphere will contract and cool as a result of increasing co2 concentrations, and this conclusion has been backed by empirical observations. Furthermore, modeling from the 1970s by scientists at Exxon Mobile found that by 2020, net warming would reach an average of 1 degree centigrade, which we’ve since exceeded.
Conclusion
Climate models published since 1973 have generally been quite skillful in projecting future warming. While some were too low and some too high, they all show outcomes reasonably close to what has actually occurred, especially when discrepancies between predicted and actual CO2 concentrations and other climate forcings are taken into account.
A clock that is stuck is a poor model of time because it does not model the progression of time, even though it is still correct twice a day.
We can make a very simplistic climate model that would be valid. We have a sphere of radius 6,357km, which is illuminated by electro-magnetic radiation centered at wavelengths of 0.6um. When the light hits the surface, it is absorbed and is re-emitted as black body radiation with a wavelength centered at 10um. It has an atmosphere, which contains a gas, which blocks the escape of wavelengths at 4um and 15um. This reflects some of the black body radiation back to the sphere, causing it to warm. In 2009 for the earth, this lead to a net accumulation of 0.6 W/m^2 of energy absorption [1]. If accumulated over the entire area of the sphere and accumulated over time, this would lead to an increase in temperature. Although the model does not take into account land formations, ocean currents, air currents, seasons, etc, it still models the underlying process of the earth’s energy balance and can answer the question of whether the planet will warm and can provide estimates at how quickly this will occur.
I think you’re hinting at systems being so complex that they can vary based on minute changes in initial conditions, you’d be correct to say that climate is chaotic. Not all chaotic systems are the same though. A double pendulum will swing in eccentric motions, but if you look at something like the stock market, there’s a lot of short term variability with long term certainty. I cannot tell you what my stocks will do next week. But markets generally converge on a doubling of money in 10 years. In fact that’s reliable enough that I’m willing to have a 401k. With the pendulum, I can tell you with a lot of certainty that if it’s damped, it will settle to its lowest energy state and stay there permanently.
A climate model will not predict the weather. But in the same way that I can tell you that winter will be cold, I can also tell you that Arctic amplification and Hadley cell expansion will lead to very different weather patterns that civilization has not seen before, such as much more extreme swings in temperature and weather.
[1] https://en.m.wikipedia.org/wiki/Earth's_energy_budget#/media...
At large scale climate change isn’t that complicated - sun hits earth, heats it up, some energy escapes. We can get a pretty good picture from models and the details just add to that.
no—I, knowing basically nothing about climate science and having been taught to more or less blindly trust climate models as accurate predictors of the future, read an article that said we have no idea how to model clouds with anything resembling accuracy, because clouds by themselves are crazy complex things. this led me to realize that this should have been intuitive knowledge without even reading the article, which led me to truly consider the scope of the climate system w.r.t. how it could possibly be modeled by computers, which led me to realize that the idea that it's possible for a data-driven computer model to predict anything about the future with any kind of reliability is pretty silly to begin with.
A couple counterpoints from former climate skeptics:
- Richard Mueller is a prominent physicist who founded an institute to debunk anthropogenic climate change, only to change his mind after thorough examination of the data. https://youtu.be/Sme8WQ4Wb5w
- Jerry Taylor is a libertarian and former climate debunker for the Cato Institute; a noteworthy pivot moment was speaking to a risk analyst for a financial institution, whose job it is to make bets on complex systemic risks in the context of uncertainty. He claimed that from a financial lens, AGW was a slam-dunk case for mitigation, not because of the certainty of any one model, but of the full probability distribution in aggregate. https://www.reckonings.show/episodes/17
But as good Bayesians, we also have to consider the opposite case: crises and "black swan" events do happen. How do we filter signal from noise? We probably could have benefitted from more "enlightened alarmism" re: COVID in Jan/Feb 2020, when billions in preventative measures could have saved trillions to world GDP (and millions of lives).
It's a double-bind politically: given a non-trivial long-term risk, how does one frame its urgency? A calm and measured tone spurs little action (the unsqueaky wheel doesn't get the grease); a tone of urgency and panic bypasses badly-needed rational faculties, and/or flips the "bullshit detectors" based on the history of proverbial boys crying wolf.
While I'm not opposed to FDR-ish green infrastructure plans, the answer from economists is undramatic and uncontroversial [0]: tax the externality to establish a price signal, and return the dividend back to citizens, whose consumption behaviors then change in response to those signals [1].
Of course, we've had to contend with decades of FUD from vested O&G interests, no different than the tobacco cancer debacle [2], despite having had credible evidence for AGW since the late 70's. I'm happy to concede the potential for scientific groupthink and perverse Big Government incentives on the green side, so long as we're also acknowledging the enormous financial incentives for muddying the waters (uncertainty = inaction = status quo = profit).
[0] https://www.wsj.com/articles/economists-statement-on-carbon-...
I am under the impression that the worst-case scenarios in most models have pretty much all come to pass, often earlier than expected. Seems to me that they’ve been very predictive. It would have been nice had the media more accurately delivered those predictions; perhaps it would have prompted some early action.
Our current CO2 levels, ocean temperature & acidity, glacier melt, and global temperature are in excess of all “good case” predictions. We failed to rein in CO2 production and now we’re pretty much living the worst-case scenarios… as predicted.
If the model predicts that a glacier disappears in 2050 and it disappears in 2070 instead?
Or if the lack-of-model makes a lack-of-prediction about glacier disappearance, and then it disappears in 2070 by surprise?
Error accumulation makes sense to worry about if you're planning a voyage and worried that dead-reckoning navigation will leave you kilometres off course. But if you're on the voyage already, out on the open ocean, you can argue with the navigator's math but you still need to navigate -- you won't do better by just blindly insisting the boat never left the harbour.
Think of climate modeling as saying that for a given species of trees, half of them will be dead at a given latitude in 50-100 years, not pointing at a specific tree and predicting that it will die on July 7th, 2047.
I also think we need to seperate out how humans are driving species extinct, and climate change. These two can be very different things, yet are always conflated in the media for hype.
Most species are going extinct because their habitats are actively destroyed by humans, or are fished to extinction, not because the planet is warming.