Comment by pjkundert | Hacker News Reader

Comment by pjkundert | Hacker News Reader

Parent Full thread

> Those now attempting to figure out the mystery of the hot climate models think one factor might have caused the recent unusual results: clouds. It turns out simulated clouds often cause headaches for climate modelers.

The impact of cloud cover in these models has been incorrect, and has such a large impact. Doesn't this strike anyone as a major problem?

We're burning about 100 billion gallons of aviation fuel per year.

> This means each gallon of jet fuel (6.5lbs) will combine with 23lbs of Oxygen and turn into twenty pounds of CO2, and just over nine pounds of water! (see: https://paullaherty.com/2015/01/10/calculating-aircraft-co2-...).

So that's 100B gallons * 9lbs is about a half a billion metric tons of water vapor going into the dry upper troposphere annually from commercial aviation.

Which, evidence would seem to suggest, has not been modeled properly (at all?) by these climate models?

If I was producing these models, I would be embarassed.

Rough estimate: There are 1.4 × 10^17 liters of water in atmosphere (https://whyfiles.org/2010/how-much-water-is-in-the-atmospher...).

Your number "about a half a billion metric tons" is 0.5 × 10^12 liters of water, which is 5 orders of magnitude lower number. Or, less than 1/10 000 of atmospheric water.

To say in your words, "If I was quoting this number as a significant contribution, I would be embarassed."

The atmosphere at FL300 is super cold and dry. Injecting water there has an enormous impact -- much greater than at lower altitudes, where the bulk of atmospheric water resides.

So, 500 million metric tons sounds like a lot, doesn't it? But there's the fact that it's getting dumped into an already big system: the atmosphere at large.

Let's check it out it percentage terms. According to wikipedia, the mean mass of water vapor in the air is 1.27 * 10^16 kg. To get the units correct, that aviation contribution is 5 * 10^11 kg. Dividing the latter by the former, we find that the annual commercial aviation water vapor represents 0.004% of the mean mass of water vapor.

That's rounding error.

FL300 is very cold and very dry, and bears almost none of the "atmospheric water". Your estimation of impact is not correct. This is not a linear system.

FL300 (30,000 feet) is usually within the troposphere. And that's highly relevant, because the troposphere is well-mixed. Water vapor emissions at 30,000 feet are going to come back down, for the most part.

Cloud cover and insolation (solar heating) is the primary determinant of weather (short-term) temperature. Over the long term, average cloud cover and insolation is determined by ... what? Something not modeled, apparently.

It's in no small part due to "solar weather" (sunspots), and long term levels of high-level water vapor:

> Contrails during day cause cooling because of reflecting of sunlight back into space. During night, they trap infrared heat causing heating. So it is a balance between the two time intervals. We would like to have more CICs (contrail-induced cirrus clouds) during day and none during night.; FAA Scientist: We Want Clouds By Day, None By Night

How did you get from

> It turns out simulated clouds often cause headaches for climate modelers.

to

> Which, evidence would seem to suggest, has not been modeled properly (at all?) by these climate models?

Embarrassment may be premature

Remember, before you make the obvious mistake of "well, aktually, the atmosphere contains 5 orders of magnitude more water than that" -- we're not discussing a linear system.

Almost none of that atmospheric water resides at FL300. That's the point. You wildly adjust atmospheric parameters in the real world, with (apparently) not accurately accounting for them, and then claim your model is predictive?

This is why modelling complex, chaotic systems is hard. I've done it, and that's why I would be embarrassed.

Not because of the mistakes. Because I was so certain that my models were correct. The level of hubris displayed here is mind-boggling.

Reply on news.ycombinator.com