Hansen et al refer to Rignot et al 2013
Rignot being one of the co-authors here, perhaps the two 2015 papers from him and Joughin (not a co-author) on imminent and irreversible Antarctic ice shelf collapse came too late for this draft.
I am still choking on the overly broad sweep of the Hansen article -- two of the building blocks have crumbled (see #36) since this paper was put together (6-8 weeks ago).
Below I dug into a very recent fiasco/scandal concerning the seeming slam-dunk issue of synchronizing Greenland and Antarctic ice cores over just the last 2,500 years, assigning stratospheric volcanoes to dates and modeling their cooling effects on climate through sulfate aerosols.
Mind you, the Sigl paper (first link below) is really excellent but frankly I do not see here an 'offset' or 'seven year bias' in ice core dates, but instead seven years of dating blunders
by a cryosphere community that mistakenly deprecated dendrochronology, followed by years of rubbish and denial from climate aerosol modelers who write and tediously defend papers based on the misinterpreted record (notably ME Mann).
How much confidence can we have in a 115ka Greenland ice core chronology with the first 2.5ka proving such a struggle and extension to the Holocene seemingly a pipe dream? Very little, based on wholesale revisions of radar stratigraphy and Eemian nitrogen isotope dates. Yet a whole lot of paleo and forecast climatology is contingent on this. And this does not get to a tenth of the factual assumptions underlying the Hansen paper. Crawl before walking, walk before flying and all that.
Naturally the Sigl article shows very little curiosity about sources of ice dating mistakes, other than Vesuvius attribution which was nonsense from the get-go and corrected earlier by Baillie. (Indeed, the paper might well have gone point by point through every issue raised by the Baillie papers.) Systemic errors in ice core analytics — such as flat-melt CFA and non-utilization of split beam optics to detect layer distortion — are likely to have even larger cumulative effects in deeper deformed ice. Indeed, after so many siting errors such as NEEM, the new Renland core was drilled precisely where parallel layers could be obtained according to the sled radar grid.
It seems the seven extra years in the ice core chronology were inserted earlier than the 1257 Samalas event but later than 994 Be10
anomalies, an interval of 263 years. Is this better understood yet, that is, how much is still wrong (ie earlier than 1257 but not correctly dated initially in the ice chronology)?
The Sigl paper states that revised ice-core timescales are now accurate to <5 years, with revised age models pointing to previous 11 year errors in Greenland and 14 years in Antarctica. I didn’t quite follow why residual errors are still so high — dendrochronology has zero annual error on this scale, indeed I recall the Inyo bristlecone alone provide a continuous 9 ka record of the Holocene. Is the 5 year error attributable to deference to the Hidden Markov Matrix algorithm now being used on ice cores?
I am very familiar with scientific applications of HMM and 'maximal likelihood' statistics. The first thing to say here is HMM is just an algorithm within a particular non-Bayesian school of parametric statistics; despite the name, it does not necessarily find a more likely outcome than ordinary common sense and does not purport be objective (as the Sigl paper asserts): it merely carries out someone's code written to various buried subjective assumptions not necessarily met by the data feed. I'm ok with HMM on ice cores as a convenient first step that suggests an initial neutral (or at least uniform) baseline for multiple cores. The real reason people use HMM
is because it provides ‘error' estimates, leaving peer-reviewers with less to complain about. The issue getting lost here is formal error within the HMM framework vs real error in the real world.
I can’t envision HMM ever outperforming an expert human annotator in ice corp interpretation. Case in point: right here. The more holistic Baillie approach already has a better outcome than 5 year error. The misunderstanding about subjectivity may have arisen from the checkered history with different sunspot observers where indeed each observer had to be assigned an individual scaling factor.
There’s a fascinating backstory on what actually caused the C14 and B10 excursion in the tree rings and ice. I’ve included the better articles on those as well. How was it even possible for the C14
extravaganza in 775 to be missed by so many for so long? How many historic dates are based on what we now see as erroneous calibration?
Timing and climate forcing of volcanic eruptions for the past 2,500 years
M. Sigl and 24 others 2015http://www.nature.com/nature/journal/vaop/ncurrent/full/nature14565.html
Tree ring effects and ice core acidities clarify the volcanic record of the first millennium
MGL Baillie and J McAneney 2015http://www.clim-past.net/11/105/2015/cp-11-105-2015.html
The mystery of the offset chronologies: Tree rings and the volcanic record of the 1st millennium
J McAneney 2015http://www.realclimate.org/index.php/archives/2015/02/the-mystery-of-the-offset-chronologies-tree-rings-and-the-volcanic-record-of-the-1st-millennium/
Volcanoes, ice-cores and tree-rings: one story or two?
MGL Bailli 2010http://dx.doi.org/10.1017/S0003598X00099877
Proposed re‐dating of the European ice core chronology by seven years prior to the 7th century AD
MGL Baillie 2008http://onlinelibrary.wiley.com/doi/10.1029/2008GL034755/full
Tree rings and volcanic cooling
KJ Anchukaitis 2012http://dx.doi.org/10.1038/ngeo1645
Dendrochronology raises questions about the nature of the AD 536 dust-veil event
MGL Baillie The Holocene, 4, 212–217, 1994http://www.realclimate.org/index.php/archives/2012/11/responses-to-volcanoes-in-tree-rings-and-models/
Mann, M., Fuentes, J. & Rutherford, S. Nature Geosci. 5, 202–205 (2012).
Mann, M., Fuentes, J. & Rutherford, S. Nature Geosci. 5, 837–838 (2012).
Mann, M. E. et al. J. Geophys. Res. A. 118, 7617–7627 (2013)
Cosmic ray event of AD 774–775 shown in quasi‐annual 10Be data from the Antarctic Dome Fuji ice core
F Miyake 2015http://onlinelibrary.wiley.com/doi/10.1002/2014GL062218/abstract
On a solar origin for the cosmogenic nuclide event of 775 AD
EW Cliver 2014http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20150007938.pdf
A solar super-flare as cause for the 14C variation in AD 774/5?
R. Neuhäuser 2014http://onlinelibrary.wiley.com/doi/10.1002/asna.201412071/abstract
Solar activity around AD 775 from aurorae and radiocarbon
R Neuhäuser 2015http://arxiv.org/abs/1503.01581
The AD775 cosmic event revisited: the Sun is to blame
IG Usoskin 2013http://arxiv.org/abs/1302.6897