MIT's "pretty good" climate modeling part 2. The rarest of the rare: something from a national outlet that did their homework on Polis and actually got it right. Last minute add on the Trump ruling.
MIT's "pretty good" climate modeling part 2.
Let me remind you of the question from part 1. Can we predict the future of our planet and its climate?
And, the followup: if you think we can, how well can we predict?
This is the first of 2 in a series looking at predictions and modeling, using an MIT global model as an example.
As with post 1 from last Monday, I am working from a video shared by a reader (linked first below) and the relevant part starts around the 36:57 mark if you want to follow along.
In this post, let's talk about the validity of a model but from a different angle: what are the pieces that get used to build a model, its inner workings, and have they been used correctly?
There are a thousand ways to make models and all of them operate on the rules that you tell them to. They don't make decisions on their own. They don't think. They don't make judgments. They take the inputs you feed them and then produce an output depending on what "guts" you put in them.
So, care is needed in telling the model the rules to play by. Care in particular because if you use a result from physics or from mathematics in a model, the theory you use needs to be valid for the situation in which you use it. Using the theorem or whatever in a way in which it was not defined or in a physical regime in which it's valid will at best make your results (to varying degrees) approximate.
Let me give an analogy to something most of us are likely familiar with. Take a look at screenshot 1 and travel back in time to algebra class. The top equation is called a linear equation because the x is raised only to the power of 1. The middle one is a quadratic equation, because it has an x raised to the power of 2 (in addition to an x to the power of 1 but we classify by the highest exponent).
Good so far? Okay, the third item is the much-maligned quadratic formula. Used correctly, it would give you all the values of x that make the second equation true. If you tried to use the quadratic formula to solve the first equation, the linear one, you'd have a real problem. You would, in fact, be trying to divide by zero (a mathematical no-no).** So what happened?
You're trying to use a mathematical result OUTSIDE of where it is valid. The quadratic formula was invented specifically for the case where "a" doesn't equal zero and you're trying to use it where "a" = 0. You have a logical fault, a mismatch.
The MIT model that you see in the Youtube video has a similar fault in it. I attached a picture from the video as screenshot 2 for convenience's sake.
I'll explain, but some context is needed first.
In physics a blackbody is an idealized piece of matter that is capable of absorbing (and emitting--one curiosity from physics is that good absorbers are also good emitters!) any and all electromagnetic radiation that hits it, whether that's radio, light, infrared, x-rays, whatever.
This object is imagined as being in thermal equilibrium with its surroundings. That is, it's happily exchanging heat energy as radiation (again, the radio, etc.) with the environment it sits in and its temperature is not changing with time.
If we somehow heated this object up, we'd tilt the balance and it would start emitting more energy than it absorbed until it hit a new (higher temperature) equilibrium with its surroundings. The reverse would happen if we cooled it.
If you absorbed the last paragraph you probably get the connection to global warming modeling already. In order to model global warming, you have to have some handle on how our atmosphere absorbs and rejects heat to the ground below and space without, and the primary method by which both the former and the latter happen is by radiation. This would be an obvious, necessary, and fundamental piece of the puzzle for any modeler.
By what I can tell from the video, the MIT people modeled our atmosphere as a blackbody (either in whole or in layers of blackbodies).
This is invalid. It is exactly the same kind of error as I mentioned above about the quadratic formula because, put simply, our earth's atmosphere is NOT a blackbody. It doesn't accept heat and reject it in the way a blackbody does.
Technically, nothing is a PERFECT blackbody, but there are objects that come close and so it's valid to consider them as blackbodies. Your body is pretty decent (97% a blackbody) and the sun is also pretty good too. In order to give you a sense of how close the sun is, take a look at screenshot 3. It's a spectrum (a breakdown of how much light our sun gives off at different wavelengths) of our sun's radiation. A perfect blackbody would be the black curve in the graph. As I mentioned in part 1, the actual and predicted are decently close without a lot of gaps.
But if we can treat the sun (or you) as a blackbody, we cannot treat our atmosphere that way because our atmosphere is way, way too thin. The atoms in our atmosphere, as a dilute gas, are not close enough to feel or be influenced by each other in the way they are in the sun (or you).
I'm reminded here of the old joke about someone coming upon a man bent over looking for something in the beam of a streetlamp. Another man comes along and asks, "what did you lose?" to which the first man replies, "I lost my keys somewhere over there." motioning to a direction off in the darkness.
The first man then asks, "why are you looking here then?" and the second answers with the punchline, "because it's the only place I can see to look."
I have mentioned the idea of a spherical cow in science and op eds before (see the second link below if you want more context). This is shorthand to a scientist for a situation where you have to simplify nature (sometimes to an absurd degree) in order to make progress in understanding it. This is an okay thing to do as long as you acknowledge the limitations brought on in your theory or model by making your cows perfectly spherical.
I have a feeling, though I am not certain, that the MIT people were pretending our atmosphere was a blackbody because, like the gentleman in the streetlamp, it's the only place they could see to look. That is, had they tried to include the actual, real, complexity of nature in the model, it would have been to complex to be tractable. They needed a spherical cow to get a result using current technology.
The problem is that the MIT people (and I suspect others who do climate modeling) are not being forthright in just exactly how many spherical cows are hiding in their results and just what they do to the validity of their results. I KNOW for sure that the media are not digging into this.
That's a shame because that should be part of the conversation. I hope you at least, going forward, have a little better understanding of things to look for in modeling now.
Watch out for "pretty close" models and how "pretty close" they are.
Watch out for how big the errors are relative to the results the modelers are claiming.
And watch out for situations where the modelers are using an invalid result from physics or mathematics relative to what they're modeling.
**I tried to highlight this in red. To use the quadratic formula with a linear equation, you'd set a = 0 and you can see how the zero would end up in the denominator.
https://www.google.com/url?q=https://www.youtube.com/watch?v%3DPpcISHuSCpo%26t%3D2
Care to see a national article that gets it right on Polis?
I don't know about you, but it is a near-endless frustration for me to live in Colorado and read articles from news outlets (across the political spectrum--FoxNews I am wagging my finger at you in disappointment) that paint Gov Jared Polis as something he's not.
I'm tired of "news" and opinion by these outlets that don't seem to bother to take a minute and do some due diligence on who Polis is and what his actual policy has been here.
The essay below, then, was a breath of fresh air. Thought I'd share so we could all have a chance to see what it looks like when a national outlet (for once) puts Polis' claims to the test.
Do you suppose this guy could offer lessons (both to mainstream reporters here and from national outlets)?
https://www.forbes.com/sites/patrickgleason/2023/12/15/polis-pitches-tax-cuts-but-his-democrat-run-statehouse-isnt-on-board/?sh=542c2b306b3e
Our uber-partisan State Supreme Court ruled as expected. There’s a reason they tried this here and not, say, Wyoming after all.
If you disagree with the Colorado State Supreme Court's ruling (as I do for reasons I won't list here), I urge you to remember this and follow my lead.
Vote to NOT retain any judge in Colorado above the district level.
Above that level they are truly dyed-in-the-wool leftists and rule based on what they think the law should be and not what it is.
They've been doing so for years now.
https://coloradosun.com/2023/12/19/donald-trump-colorado-ballot-decision-supreme-court/