Clouds and global warming

[StreetScooby 5]

Two very interesting articles. Current grid size for simulations is around 60 square kilometers. They need to get that down to 1 kilometer to properly address cloud formation. Basically, climate models != science, at present, but I'm hoping others will share their own conclusions after reading these papers.

Heavy weather: why we need supercomputers to teach us how clouds and climate change work

Remote clouds responsible for climate models’ glitch in tropical rainfall

We are all engines of karma

[lawrocket 3]

They're improving it. Just a few years ago grid boxes were in the hundreds of km^2. Thus if a cloud had 100% humidity, the mean humidity of a large gridbox will be much less. That means that the modelers have to parametize the large scale means into smaller scale distributions. That’s where the climate modelers make assumptions. Each climate model makes different assumptions, meaning each will have different results, meaning there is a margin of error in it for cloud albedo (albedo is not the only climate effect of clouds – at night clouds increase the greenhouse effect). And different clouds act differently (cirrus clouds let in more energy and reflect more energy back to earth, while stratus are very good at bouncing energy back to space). Each has to be parametized.

Of course, adding these margins of error up all up increase the margin of error. Considering the thousands of relationships that must also be parametized, it’s where we see issues. For example, a couple of years ago it was reported that Antarctica had warmed. Problem: that warming was well within the margin of error. Meaning that Antarctica could have cooled. Margins of error add up, and with climate models it means more uncertainty.

My wife is hotter than your wife.

[StreetScooby 5]

Quote

They're improving it

Yes, they are. And the guys who are trying to do the real science (sans politics) inspire a lot of confidence in me.

We are all engines of karma

[billvon 3,120]

It's a thorny problem for a lot of reasons.

One, as the articles point out, it is very, very difficult to model cloud formation well - although we're getting much better at it.

Second, overall temperature is very sensitive to the timing of cloud formation. More clouds during the day? (i.e. afternoon thunderstorms) Overall, we see cooling since clouds are good at reflecting IR. More clouds at night? (low clouds and fog) This increases warming since it traps reflected IR. Thus it's not sufficient to get an accurate trend of total cloud formation; we need to figure out when it's occurring as well.

[StreetScooby 5]

I understand the issues here, and that's why I notice these articles. In no way should long term, capital intensive, freedom restricting decisions be made by the government based upon current global warming computer models, IMO.

We are all engines of karma