Roger Pielke Jr seems to receive quite a lot of criticism from some quarters for what he chooses to say about climate change, in particular the role of climate change in exacerbating (or not) damage and losses due to extreme weather events. He received extensive cricism for his first 538 article claiming that disasters cost more, but not because of climate change, which he defended in an interview with Keith Kloor.
Now, from what I’ve read and done, what Roger Pielke Jr says about disaster trends is probably broadly correct, at least in the sense that we can’t detect a trend associated with climate change. We’re talking here about the impact on us (humans) of events that, by themselves, are reasonably rare. Although there is some evidence for trends in some extreme events, it’s not surprising that we can’t detect a climate change related trend in damage/losses due to these events.
Yesterday I noticed a tweet from Roger that linked to a blog post of his about a new paper on disaster losses and climate change. The paper is titled On the relation between weather-related disaster impacts, vulnerability and climate change by Hans Visser, Arthur C. Petersen, and Willem Ligtvoet.
In his blogpost, Roger says
Some have argued that our methodological inability to fully account for possible changes in vulnerability to losses over time may mask a climate change signal in the data.
I suspect he’s right; some do indeed argue that this may be the case. Roger then quotes various passages from the paper above and concludes with (my emphasis)
The bottom line? Once again, we see further reinforcement for the conclusion that there is no detectable evidence of a role for human-caused climate change in increasing disaster losses. In plain English: Disaster losses have been increasing, but it is not due to climate change.
Now, I haven’t worked through the paper in detail, but from the abstract
Despite these variations, our overall conclusion is that the increasing exposure of people and economic assets is the major cause of increasing trends in disaster impacts.
Okay, that seems broadly consistent with Roger’s interpretation. However, the abstract then says
in all three cases, the role played by climate change cannot be excluded.
Hmmm, that doesn’t quite seem the same as but it is not due to climate change. This seems to be saying precisely what those who Roger is criticising have been saying : we can’t rule out that there might be a climate change signal in the data. Of course, Roger might simply mean that the influence of climate change is not dominant/important at this stage, but that would – again – seem consistent with those he’s criticising.
To be clear, I haven’t read the paper, so maybe the abstract is actually saying something different to what the paper says (or there is some subtlety that I’ve missed). I’m also not really that interested in this type of thing, since it seems perfectly reasonable that we can’t actually find a climate change signal in the data. Looking for something that we are almost certainly not going to find and then arguing that not finding it has some significance, just seems a little silly. If we want to understand what might happen in the future, we should – in my view – be considering the physical climatology, not whether or not we’re able to detect – today – a trend in past economic data. However, it does appear as though Roger has used a paper that says we can’t exclude that there is a climate change signal in the data to argue that those who suggest that there may be a masked climate change signal in the data are wrong. Again, maybe I’ve misunderstood something, but if this is what Roger has done, it’s not surprising that some give him a rather hard time.