> We're lucky to get 10Ma uncertainty for 100Ma old formations. Maybe a few decades ago. Ironically, much of my background is in U-Pb CA-ID-TIMS geochronology.

> We're lucky to get 10Ma uncertainty for 100Ma old formations. Maybe a few decades ago. Ironically, much of my background is in U-Pb CA-ID-TIMS geochronology. The current state of the art in my community is between 0.05 and 0.01 % two-sigma relative, i.e. 10 to 50 ka at 100 Ma. To give one example, the chronology of the P-T extinction (the big one) is now constrained to the decamillenial level even though it we're talking about an extinction 252 million years ago [1]. Of course, 50,000 years is still quite long compared to modern anthropogenic effects, but the timescale bias of the geologic record is neither poorly known nor taboo (there are well cited papers about this exact problem [2]). I should probably also reiterate that many records of interest to climate science are (1) much younger, so absolute precision is much higher than at 252 Ma and (2) may come with a built-in relative chronometer with annual resolution, as in ice cores, tree rings, varves, and some speleothems. Imputing incompetence or bad faith to the entire field of climate science is not productive. While I'm not a climate scientist myself, I know many of them, and find them to be both honest and competent. [1] http://www.pnas.org/content/111/9/3316.short http://www.pnas.org/content/111/9/3316.short [2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5227093/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5227093/