|
The problem with this argument is that it is both an argument for and against climate change, and so both for and against taking action on climate change. Humans have a history of (locally) causing massive pollution and other various disasters, like removing topsoil and destroying agriculture. Mining. Extermination of non-tiny animals in a large area. The list goes on. Every time this turned out not to be a problem. Will it now ? If so it will be the first time. But of course, climate change is not based on physical evidence, nor first principles. It is based on statistical trend analysis. Climate change only has basic explanations of mechanisms of change in the athmosphere, they cannot physically model a warming planet from first principles, only by extrapolating past observations into the future can predictions be made. Of course the truth is that both facts are chaotic. In other words, every event in human history is a first, and every change in the athmosphere is a first. For either domain, any particular event is without precedent, and will never happen the same way again. This does not exclude, of course, that one observes patterns. Temperature has been rising, and there are many historically analogous periods, but none that are quite the same. So will the same thing happen ? Unfortunately we know the mathematical theory underpinning this. The answer is sure to disappoint all involved : we don't know, and we don't know the chances either. And no algorithm can ever be used to arrive at a correct answer (note the problem is NOT that we don't know, it's that we can't ever know [1]). It is impossible. It is extremely obvious in human history. When we move past oil (and we will, in the next decade or two), there will be a key ingredient that wasn't there before that completely determines the outcome. It will be a first, that's for sure. Either we find a new energy source that economically blows oil out of the water, or it will be the first long-term sustained drop in global economic activity due to running out of energy. It may also be something different entirely (e.g. WWIII). Maybe thorium reactors start taking the world by storm. Nobody can tell you, with any amount of certainty, what this grand event will be, but it is a surety that it will come. What isn't entirely obvious is that the same is true for climate. Whatever happens, it will be a first, and there will be some event, some factor, that completely determines the outcome, but we can't see it just yet. Firstly, there is of course the potential for large external factors that are obviously so large as to be much more powerful than anything happening on the earth surface itself. Think volcanic activity or a comet, but likely there are 100 such events that could happen we don't even know about. Both events can both raise and drop the temperature by ridiculous amounts, far outstripping what we could ever do with co2. Other things could happen. Suddenly the sea floor starts leaking methane on a large scale, like what happens in any earthquake, but it is sustained. There's certainly enough methane down there to make human solar forcing look like a cow fart. Maybe we're wrong about cloud cover (still a possibility) and the situation simply self-corrects past a certain threshold, something that no statistical model of the past would ever have been able to uncover. Maybe we do get a single world government and actually stand a chance of fixing the situation. Maybe Saudi Arabia and Iran start doing what they've been threatening eachother with for 100 years now and Iran nukes the oil fields there. Maybe it doesn't correct at all, solar forcing happens, and animal life becomes very difficult for a few centuries/millenia. But what is an absolute certainty, as you say, is that the situation will not continue as it's been progressing so far. That is pretty much the only impossibility. Note that the obvious defense of climate theory here, that it results from deterministic actions, and so is not random/chaotic at all, is both correct and completely beside the point. Random systems are not chaotic nor vice versa. Deterministic systems that exhibit 3 properties are chaotic. None of them involves randomness. Neither chaotic nor random systems are predictable [1]. As to whether chaos theory really applies to climate theory the following can be said to summarize things : for 150+ years, the study of chaos theory and the study of climate were one and the same field of study. This has changed, but not for good reasons. There is a longstanding argument that long-term effect dampen the medium term unpredictability, but there is no proof for this, in fact there is plenty of contradicting evidence. And the last defence of climate theory is that climate theory only studies averages. They admit that short-term change might be chaotic, but long-term averages aren't. They certainly don't look it, after all. This flies in the face of the central insight of statistics : the central limit theorem (it basically states that any variable you study must be convergent, that statistics doesn't work on divergent series. And it doesn't, that can be easily tested). This argument is incompatible with statistics on a theoretical level. Therefore it does not matter if this argument is true or not. If this argument is true, then there is no correct way to calculate even an average. If it is false, then you simply can't correctly average certain variables (including a lot of climate variables). The "false" seems to me the sane option, but it doesn't even matter which you pick. Lots of scientific fields of study have these "existential" problems. That doesn't mean they have no value, just that they are fundamentally limited in a few ways. [1] http://en.wikipedia.org/wiki/Chaos_theory |