William Connolley is somewhat bemused by Christopher Monckton’s review of the IPCC’s Summary for Policy Makers. Because the IPCC changed the way sea level rises were reported, critics seem to inevitably misunderstand them and claim that the IPCC has substantially reduced its projections.
Also, the UN, in its 2007 report, has more than halved its high-end best estimate of the rise in sea level by 2100 from 3 feet to just 17 inches.
And after that, the sea level rise? The TAR SPM pic is here. The SLR is 0.88 top-of-range and this is presumably what M is using for 3 feet. But thats all-models all-SRES plus land ice uncertainty (except for the WAIS). The value sans land ice uncertainty (which is what the AR4 uses) is 0.7 in the TAR against 0.59 in the AR4. But hold on, 0.7m is 27″. 17″ is 0.43m. Where has M got that from? Top of the B2 range? Average of the A1F1 range? Why would he choose either of those? Also I’m told, but have not verified, that the TAR range is 95% but the AR4 range is 90% uncertainty. So perhaps its better to look at the mid value for which AR4 sez “For each scenario, the midpoint of the range in Table SPM-2 is within 10% of the TAR model average for 2090-2099.”
Take rising sea levels. In its 2001 report, the U.N.’s best high-end estimate of the rise in sea levels by 2100 was three feet. Lord Monckton notes that the upcoming report’s high-end best estimate is 17 inches, or half the previous prediction.
There’s more, but let me pick out some more good bits from Monckton:
There was almost certainly less Arctic sea-ice in the early 1940s than there is now, and there may have been none in Summer in the middle ages.
Because, according to Monckton a Chinese naval squadron sailed right round the Arctic in 1421 and found none.
The UN’s simulations omitted the important El NiÃ±o ocean oscillation which has been more prominent in recent years, and underestimated urban heat-island effects: thermometers are mostly near towns.