Updated…
One of the more amusing moments in the annals of climate change denial was surely when the billionaire conservative Koch brothers decided to fund the Berkeley Earth project to reassess global temperature change. Presumably they expected their select team (including Ms Curry) to obtain the “right” answer, unlike all those fools before them — especially the wicked US government ones at NOAA and NASA. Unsurprisingly, after studying long and hard on the matter, the Berkeley Earth team came up with pretty much the same answer as everyone else. How disappointing. The denialist trope of grossly exaggerated 20th century warming vanished right about then, to be replaced by the one that denies warming has continued this century (and has actually accelerated). Such are the games played.
Anyway, don’t imagine that the Berkeley Earth project has been a waste of time and Koch money. Berkeley Earth brought to the task a degree of rigor and sophistication that is a step above previous efforts. In particular, they apply Krige interpolation to the whole dataset (not just to the data holes, as Cowtan & Way do). That plus sophisticated homogenisation procedures allow them to incorporate more and disparate data sources, and to compute convincing confidence limits throughout. I expect the Berkeley Earth products to become the default, go to, estimates of global temperature. (That is until a sufficiently convincing reanalysis effort trumps them; smart physics usually beats smart statistics in the end.)
Last month Berkeley Earth finally published their land+ocean series (previously they provided just land temperature estimates). I’ve incorporated that into my plots of global temperature, which now show data from eight different sources. Anyone else want to pretend that all eight are wrong? Details are on the graph pages.
More:
I’m leery enough of wanton extrapolation of fitted functions, but it’s in no way amusing that if you extrapolate the polynomial fit, it neatly bisects the RCP6.0 and RCP8.5 global warming projections for 2050 (which of course derive from the massive international mechanistic modelling effort coordinated by the IPCC, not from some crude extrapolation):
As of now we’re pretty much right on that polynomial trend.
[…] pretty much “business as usual” IPCC greenhouse gas emission scenarios (A1B and A1FI), consistent with observed warming to date. If there were to be substantial emissions reduction over the ensuing 86 years, the snowpack […]