alexandra_thorn: 2009, taken by Underwatercolor (Default)
alexandra_thorn ([personal profile] alexandra_thorn) wrote2015-02-09 02:15 pm

Global warming: a word on the uncertainty of computer models

Earlier today a climate change skeptic I know in Facebook land made a post grumbling about being expected to believe in anthropogenic climate change just because of computer models. He was happy enough to acknowledge that there's clear experimental evidence that human activities have already heated the planet, but objected to his perception that we're just expected to just believe computer models when it comes to future climate change.

I think this statement comes from a fundamental misunderstanding of what climate models are for. Here's what I wrote in response:

As a computer modeller and an ecologist who works closely with climate modellers, I want to speak to the claim that evidence for global warming is based on computer models. This statement is incorrect but contains a grain of truth. We know from laboratory experiments that certain gasses absorb solar radiation and turn it into heat. And we know from the both recent and paleo atmospheric records that the levels of these gasses in the atmosphere change the climate. We also know from both experimental and field observations that human activities are increasing the concentrations of these gasses, and that some of these activities are also setting off natural processes that are producing more of these gasses such that we wouldn't be able to slam on the brakes even if we instantly stopped everything we're doing right now.

So, we know from experimental and observational data that human activities are changing the composition of the atmosphere in a direction that will cause it to absorb more heat, and that since we are not doing much to change our activities that these changes will continue to accelerate. What we *don't* know is exactly, quantitatively, what the outcomes of these changes will be, and when. This makes it harder to plan for the future that we know we are stuck with. And that's where the computer models come in. Computer models give us best guesses as to how much ocean levels will rise over what time scale, and what the long term effects will be on climate systems. This helps planners figure out reasonable contingencies for actions that might help ameliorate the negative effects of climate change. Should we put up sea walls? How high do they need to be? Which populations are most likely to need evacuation plans? What types of crops are most likely to be robust to the types of changing climate patterns we expect in specific parts of the world?

There are *huge* limitations to the existing models. The fact of the matter is that we have no clue of the details of what's going to happen, other than some basic stuff that comes down to common sense: if we don't somehow kick ourselves into an ice age, the ice caps are going to melt and the ocean levels will eventually increase by a couple hundred feet (we don't know exactly when, but we'll be stuck with it once it happens); and ocean acidification (which happens even if we do get an ice age) from atmospheric CO2 will dissolve the shells of many sea creatures kicking off a mass extinction that we won't be able to do anything about. The fact that our computer models are so uncertain shouldn't reassure anybody. It should make the situation a lot scarier. We know big changes are happening, but we don't know exactly what they will be or when. This makes planning difficult and the whole situation a lot more dangerous.