More from the first day of the Climate Change and Governance Conference. I should point out that apparently they'll be putting video of the various presentations on the web over the next few weeks, so if you're interested (and have bandwidth) you can watch them yourself.
Dr Kevin Trenberth (US National Center for Atmospheric Research) spoke on "observed changes to the climate and their causes". I ended up missing the first half of this, and came in just as he was talking about hurricanes; there's some evidence that hurricane intensity (and hence damage) has increased along with the rise in average sea temperatures. He also talked about the big picture - retreating glaciers, the decline in sea ice and snow-cover, and the effects of lower snowpack on water supplies and droughts (generally bad). His overall conclusion was that natural forcing does not account for warming after about 1970.
Dr Myles Allen (Oxford University) spoke via videolink on "climate change modelling and projections". His first point was that there is strong agreement among scientists (including those critical of global warming) on the expected level of warming to 2040 or so. Where people disagree is on the sensitivity of the climate (how strongly it will react to extra CO2) and the long-term response. His second point was that the nature of the problem makes the models difficult to validate; CO2 levels this high are unprecedented with the current configuration of continents (and hence global circulation patterns), so we can't check a run of the model (with appropriate inputs) against the geological record. And of course we don't have the luxury of having a control. All we can do is try and be as rigorous as possible, and use ensemble forecasting to try and get a grasp on the uncertainties.
This causes problems for trying to set a "stabilisation limit" for greenhouse gases (e.g. trying to stabilise GHG concentrations at twice the preindustrial level, or half of it, or any other arbitrary level), because we won't know whether we've chosen the right limit until it is too late. Worse, the models all have large "tails" of high warming with reasonably high probabilities. For example, stabilising total GHG concentrations at 40% of 1990 levels still leaves us with a >50% chance of more than 2 degrees of warming. But this isn't the case for a "containment limit", in which we limit the total amount of GHGs we dump into the atmosphere; here, we have far more certainty about the consequences and whether they will be dangerous.
His conclusion was that the idea of a "sustainable per-capita emissions limit" is scientifically indefensible, because we cannot tell if it is in fact sustainable. By contrast, a total emissions cap is sustainable, in that we can predict the effects with a far more certainty. We're emitted about 500 GigaTonnes of GHGs so far, and Dr Allen's suggestion was that if we wanted to limit warming to less than 2 degrees, we should try and emit less than a trillion tons.
(There's a PDF presentation here with the graphs and charts he used, though I think its for the same talk given elsewhere).
Professor Ronald Prinn (MIT Joint Program on the Science and Policy of Global Change) gave an interesting talk on "predicting future climate through the integration of science and economics". Much of this was about the MIT Joint Program's Integrated Global System Model, which incorporates a sophisticated economic model which feeds into (and gets feedback from) a climate change model. This allows the effects of various policies to be tested on a global scale. He likewise pointed out the uncertainties in the models - that long "tail" of extreme warming due to various levels of feedback in the climate system. One key fact to emerge from his modelling was that globally, strong climate change control policies don't have much effect before 2050; instead, the payoff come around 2100, in the form of dramatically reducing those uncertainties and the probability of dangerous levels of warming. So it is very much a long-term project. As for the cost, he used the metaphor of "wheel of climate change", a game-show where a wheel is spun to produce various levels of warming. Policy gives a new wheel, with different (and better) odds. Whether we do it or not is a question of how much we are willing to pay to avoid those disastrous scenarios.
After lunch the conference split into several streams. I listened in to talks by Dr David Whitehead (Landcare Research) on "the role of forests in climate change mitigation" (which was mostly about the prospects for "carbon farming" on marginal East Coast land by returning it to native forest), Dr Mark Howden (CSIRO) on "climate change and Australian agriculture" (their wheat industry will probably be badly affected due to increased temperatures and drought, though there's a small probability that they'll come out ahead if everyone moves their farms to follow the climate), Professor Blair Fitzharris (Otago) on New Zealand's vulnerability to global warming (basically, our farming paterrns will have to shift, and coastal communities will face significant risks), and Dr David Wratt (NIWA) on water and adaptation in NZ (its either too much or too litle, with potentially very costly effects). Finally, there was an interesting talk from Dr Harry Clark (AgResearch) on animal emissions and the prospects for controlling them (difficult given our farming practices, but not impossible).
There was also a talk from Penehuro Lefale (WMO) in another stream on the effects of climate change on small island states which was apparantly very interesting, but which I didn't get to go to. I do however have hardcopy, and I'll post anything interesting later.
The conference dinner had a very interesting and wide-ranging talk from Lord Ron Oxburgh (former chairman of Shell, among other things) on "the climate change challenge". It also had an excellent chocolate mousse pie.