Or, Pielke versus the world. To put my prejudices up front, my money would be on Pielke. Since I get to write this whilst watching a backup of my laptop (for for some odd reason) I’ll have time to read the sources as I write this.
Background: global warming is happening, and will continue into the future. But how much of a problem is it now, and how much will it be in the future? These are difficult questions. Many organisations and people (the Greenpeace types) appear to automatically assume that All Will Be Ill, and there is no particular need to study this question or even think about it. They are wrong, of course.
Which brings us to the Global Humanitarian Forum and its Human Impact Report “The Anatomy of a Silent Crisis” (WTF? Silent crisis? GW is nothing if it isn’t a noisy crisis, rarely out of the papers). It says of itself that it is specifically and exclusively focused on the adverse impacts of the climate change on human society across the world so there can’t be much doubt what they set out to find; and climate change is already today a major constraint on all human efforts. I has been creeping up on the world for years, doing its deadly work in the dark by aggravating a host of other major problems affection society, such as malnutrition, malaria and poverty. This report aims at breaking the silent suffering of millions. Its findings indicate that the impacts of climate change are each year responsible for hundreds of thousands of deaths with hundreds of millions of lives affected. Climate change is a serious threat to close to three quarters of the world population. Half a billion people are at extreme risk. Worst affected are the worldÂ´s poorest groups, who lack any responsibility for causing climate change. from the summary tells you what they did find. So far so boring; the only surprise is that they have the gall to bill this as the first report of its kind.
Oddly enough, RP Jr didn’t like it and thought it was a “A Methodological Embarassment”. Brian Schmidt thought this was just “Dog bites man, and Roger Pielke Jr. goes about things wrong “. Brian thinks that the bit RP is complaining about is only ~5% of deaths (and some other stuff too, which is less interesting). I’m sure other people are saying much the same as Brian, but I haven’t bothered trawl to find out whom. RP Jr is happy to maintain and extend his critique.
Reading (OK, I admit it, skimming) the reports finds the first interesting bit An estimated 325 million people are seriously affected by climate change every year. This estimate is derived by attributing a 40 percent proportion of the increase in the number of weather-related disasters from 1980 to current to climate change and a 4 percent proportion of the total seriously affected by environmental degradation based on negative health outcomes. So criticising RP for focussing on disasters seems rather unfair – it is the first thing the report mentions. If they are only 5% of the deaths, then I think BS should be criticising the GHF for being misleading!
While I’m there, just below that is their figure 1 – showing in graphical form the relative effects of climate change and other stuff. The LH panel is somewhat meaningless, since the large red bar is “affected by cl ch” and who knows what “affected” means. But it does compare to the 24M affected by traffic accidents. The RH panel is more interesting, as at least we knw what death is, so the 315k dead of cl ch can be compared to the… oh hold on. Why haven’t they put in the obvious comparison against deaths from traffic accidents? Wouldn’t that be the obvious thing to do? Could it possibly be because Global Traffic Deaths Put at Million a Year. OK, that is only the NYT so who cares, how about we try the Transport Research Laboratory which says “estimated that in 1999 between 750,000 and 880,000 people died from road crashes and that the majority of these deaths occurred in developing and transitional nations (85 per cent). Almost half of all estimated deaths occurred in the Asia-Pacific region. This compares with a recent estimate by the World Health Organisation of over a million deaths in 1998”. Gosh, that rather puts things into perspective.
By this point, I’d be pretty well ready to give up on the GHF, who whilst probably Nice People who you’d love to have as neighbours are clearly not up to the realities of the world. Notice how “silent crisis” would be so much more applicable to traffic deaths than GW deaths. But I’ll stop harping on that; it is going to go nowhere.
So anyway, RP leads me to some bloke commenting on dotearth that The very first line of the Executive Summary of the health chapter (CH8 in WG2) reads “Climate change currently contributes to the global burden of disease and premature deaths (very high confidence).” WGII isn’t a patch on WGI, of course, but can’t be completely dismissed. The quoter fails to quote the end of that para, which is At this early stage the effects are small but are projected to progressively increase in all countries and regions. [8.4.1]
So we inquire: what is the source of the very high confidence? What will section 8.4.1 say? It sez The World Health Organization conducted a regional and global comparative risk assessment to quantify the amount of premature morbidity and mortality due to a range of risk factors, including climate change, and to estimate the benefit of interventions to remove or reduce these risk factors. In the year 2000, climate change is estimated to have caused the loss of over 150,000 lives and 5,500,000 DALYs (0.3% of deaths and 0.4% of DALYs, respectively) (Campbell-Lendrum et al., 2003; Ezzati et al., 2004; McMichael, 2004) RP points me to a 2003 McMichael (also including C-L; so I’m going to hope that will do). This says that (forward-fitting, then extrapolating back to get the change from 1990 to 2000) that ClChDeaths in 2000 is 166k. IPCC say >150k, so I suppose that fits. But more interestingly, the results say They should be interpreted with caution as, in contrast to most other risk factors, they relied on modelled rather than directly observed outcomes… The models may also be useful for the secondary purpose of indicating the magnitude of health impacts that might already be caused by climate change, but which may not be detected by direct observation using current surveillance systems. That seems to me to be very much more cautious that the IPCC’s very high confidence of impacts (except in the rather trivial sense that since the study finds a positive connection between T and death, then any increase in T will lead to some increase in death).
Oh dear, I think I’m rambling, But I’ve seen enough now. It is clear that these death estimates are ClCh only, and not very reliable, and ignore other factors (like people getting richer. McM notes that this is a problem; based on the models they use, Europe should have malaria, but of course it doesn’t, because we got rid of it). The largest cause of ClChDeath in McM is malnutrition, and as I’ve said before I don’t think this we’re going to run out of food. So where do McM get their malnutrition deaths from? Well, first of all we have a puzzle: their models end up with an increase in malnutrition deaths from 1990 to 2000. But per capita food consumption has been going up, not down. The answer to this puzzle probably lies in Climate change may affect this through the balance of the (broadly negative) effects of changes in temperature and precipitation, and (broadly positive) effects of higher CO2 levels on yields of major food crops – in other words, they are going to ignore the (large) effects of technological progress (explicitly: no major changes in the political or economic context of world food trade or in food production technology).
So I think it is safe to say that while the McI study may be an interesting abstract modelling study, it has very little relevance to the pattern of deaths in the real world. Which in turn means the IPCC conclusions are dubious, as is the GHF report.
Conclusion: RP Jr is right.