Here is a list of history's deadliest natural disasters. The world's population is growing and climate change is likely to increase the quantity and severity of these events. Given these realities, how could the death count actually decline? In my 2005 paper, the Death Toll from Natural Disasters, I document the negative correlation between deaths from disasters and national per-capita income. If over the next 40 years, poor nations open up their economies to world trade and grow --- then my estimates suggest that natural disaster deaths could fall sharply. This very nice paper by Kellenberg and Mobarak argues that for quickly urbanizing poor nations that some of my optimism is over-stated. This merits more research.
My new book Climatopolis argues that on top of the direct income effect that new information (generated by climate scientists) about the exacerbated risks that coastal areas face and areas facing heat waves face will provide strong incentives encouraging self interested people to protect themselves. The combination of rising income and increased information about natural disasters will help to protect people all over the world and raises my confidence that the death count from natural disasters could fall over time. This would be direct (observable) evidence that we are adapting to climate change.
Do you smell a new "Simon vs. Ehrlich Bet"? details