22 January 2012

Future Tropical Cyclone Damage

Kerry Emanuel of MIT (WCAS in PDF) and Robert Medelsohn of Yale University, Emanuel and colleagues (Nature Climate Change), have new papers out on future hurricane damage. The findings of both papers reinforce existing literature on the very long time necessary to detect a signal of human caused climate change in the disaster record under recent projections and the relative role of the importance of development over human-caused climate change in future losses from tropical cyclones.

Emanuel (2011, PDF)  implemented an alternative methodology to Crompton et al. (2011) to assess under various scenarios when the signal of human-caused climate change would be detectable in the damage record of Atlantic hurricanes. He looked at four different models, and three of them showed increasing losses and one a small decrease. Of the three models that showed increasing losses the time until detection is 40, 113 and 170 years.

This time to detection is shorter than that which we found in Crompton et al. (2011). Why is that? Emanuel use an older set of model runs (we used Bender et al. 2010 -- I wonder why they used something different) and that probably accounts for the difference. It would have been nice if Emanuel had used the Bender runs, as that would have allowed an apples to apples comparison. I'd speculate that our numbers would be quite similar apples-to-apples.

Regardless, the two papers are in agreement that the time to detection of a signal of human-caused climate climate change, assuming that recent projections are correct, is a long, long time. Like, not in our lifetimes and certainly not now.

Mendelson et al. (2012) examine a range of scenarios for how tropical cyclone damage will increase to 2100. That paper concludes that tropical cyclone damage will decrease as a proportion of global GDP from 0.04% today to 0.01% in 2100 assuming human-caused climate change, using the same four models as used in Emanuel (2011, PDF). That is right, decrease as a portion of GDP. (Apparently, this result was not sexy enough for Nature Climate Change which headlined their homepage announcement of the paper rather misleadingly as, "Tropical Cyclone Damage Set to Double" referring to the expected increase in aggregate damages to 2100.) The paper also explores how damages might increase in regions around the world, though it is important to recognize that both the climate and socio-economic assumptions of the paper are highly speculative.

Mendelsohn et al. (2012) explain that their findings are consistent with existing work and as such, adds to our growing understanding that under a very wide range of scenarios for how climate might change and how society might develop, socio-economic factors will dominate the future damage record (see this paper in PDF for an unrealistically broad range of explored scenarios) independent of a wide range of assumptions and uncertainties.

Anyone claiming that they can see a human-caused climate signal in the hurricane damage record (or even the hurricane record itself) is facing a growing mountain of peer-reviewed research to overcome.

H/T to Revkin and Kloor