This paper is lengthy (at 57 pages), and lavishly endnoted (with 177 references in the bibliography). In this paper, Jaworowski, Segalstad and Ono present a listing of supposed problems with ice-core gas measurements.
Here’s the title and abstract:
This paper fails to present any coherent explanation of why ice core measurements in general should be considered unreliable. Instead, Jaworowski and colleagues develop an unconnected list of complaints regarding ice core methodology, without presenting any testable theories as to what the effects of those problems should be. In some cases, e.g. their assertion that the “age assumption” regarding firn consolidation is unproven, they ignore successful research that contradicts their thesis. In other cases, as noted in connection with the history of atmospheric CO2 measurement, they distort the conclusions of other researchers in an apparent attempt to spread a cloud of doubt over the entire enterprise.
Their criticisms of ice core measurements fall into a few broad categories:
(1) Lower CO2 concentrations have been reported by more recent research for similar ice deposits.
(2) Horizontal, impermeable barriers in snow and firn prevent gas diffusion between older and newer layers; hence the gas is the same age as the ice.
(3) Liquid water is present in ice, even at very cold temperatures, allowing CO2 to go into solution under pressure.
(4) Samples are contaminated in a variety of ways, primarily by drilling fluid.
(5) Samples crack in situ, during extraction and handling, permitting gases to diffuse between bubbles.
(6) A host of other physical and chemical processes change gas concentrations in ice, making measurement unreliable.
All of these are perfectly reasonable concerns, but the way Jaworowski and colleagues go about exploring them is not that of well-meaning critics trying in good faith to improve the science. Rather, their arguments at every turn are directed to calling into question the reliability of any ice core measurements, and disputing any conclusions drawn from them.
The paper makes much of the fact that the typical reported values of CO2 concentration for air extracted from ice have changed over the decades. This should not be too surprising: as the discipline evolved and measurement methods improved, researchers have learned how to do a better job. But Jaworowski et al. argue instead that the earlier measurements should cast doubt on the later ones:
Two important observations were made in these early studies. It was found that the CO2 content of the air trapped in pre-industrial and ancient ice is rather high, and has a very wide concentration range of about 100 – 7400 ppm (Table 1). Even more important was the finding that several physical and chemical processes … lead to differentiation of the original atmospheric ratios of N2, O2, Ar and CO2, and to depletion or enrichment of CO2 in the ice.
Here is Table 1:
In Table 1 we see the early history of a rapidly improving science. In many of the papers cited by this one, researchers quite openly discuss the limitations of their methods, and the steps they have taken to improve them. But Jaworowski et al. use the changing nature of the discipline to argue that we shouldn’t trust the whole business. This does not seem like an argument offered in good faith.
Their argument regarding the age of the “close-off” layer is even worse. At the time this paper was written, some very good research had been done showing that there is substantial mixing of air down to the firn-ice transition layer (And quite a bit more has been done since then.) Some of the most convincing evidence for this mixing, and thus for an age difference between air and ice, is the gradient in heavier isotope concentration caused by gravitational enrichment right down to the transition layer, where it levels off dramatically. Jaworowski et al. either ignore this research, or try to turn it to their purposes. They dismiss one 1982 article by Craig and Chou, saying that a 1988 article by Craig et al. had “revoked their earlier assumption” regarding the age difference. They then turn around and cite another 1988 article by Craig and a different set of collaborators, saying that the enrichment “strongly indicates that the air in porous firn is protected from convective and other motions.” So, gravitational enrichment is unreliable as an indication of how long the air has been mixing, but it is reliable as an indication that there is not much mixing?
Their argument regarding liquid water in polar ice is mostly hand-waving. For example. they cite the discovery by Mulvaney et al. of liquid sulfuric acid at grain boundaries as providing “conditions for diffusion of gas from bubbles to or from the intergranular liquid.” To describe this as speculative may be too kind, given the extremely tiny volume of the liquid inclusions, and the absence of any evidence (or even a plausible mechanism) for the proposed effect.
The extensive recitation of supposed problems with sample contamination suffers from the same defects as the other discussions. Jaworowski et al. list a number of cases where samples were believed to have been contaminated, then imply that the problem is endemic and inevitable, and that it makes all gas measurements from ice cores suspect. Nowhere would the reader learn that great care is taken to avoid contaminating ice samples, that techniques are constantly improving, or that drilling fluids are chosen for, among other properties, their lack of interaction with the studies being performed. This section seems to be an attempt to throw everything that comes to hand at the subject, and hope that something will stick.
Sample cracking? See above. Similar to the other discussions, this is an unorganized listing of cracking mechanisms in ice, coupled with vague assertions that this will lead to mixing of gases and unreliable results. In fact, most observations indicate that the bubbles in ice cores are substantially isolated from one another up to the point of crushing in the sample chamber. More to the point, Jaworowski et al. owe us some explicit theory of how their proposed mechanism might affect the results. Even if there was mixing among bubbles in samples, how rapidly would gases diffuse, how could we predict the type and magnitude of the effect, and how might the predicted results be made to agree with observation? We are given no hint.
Lastly, just in case we can still see through the smoke cloud generated by the earlier arguments, Jaworowski et al. give us a list of the processes active in ice cores and invalidating measurements from them:
This is just another laundry-list of supposed problems, masquerading as a summary of detailed criticism. By attempting to make up in volume what it lacks in argument. this table exemplifies everything that is wrong with the paper.
If Jaworowski et al. had taken even one of these processes, developed an explicit and detailed theory of how it might affect measurements of gases in ice cores, proposed an experiment to test that theory, then performed the experiment and reported the results, they would deserve more respect.
If they had acknowledged that they were discussing a vital and rapidly changing area of science, and that improvements in technique were to be expected, their arguments would have been better received.
If they had evinced a willingness to identify specific problems and develop solutions to them in a constructive spirit, their criticisms would have been welcomed by the research community.
Unfortunately, however, this very ambitious paper was shaped from the beginning by its mission: to undermine the credibility of ice core research generally, and thus to weaken any recommendations that might arise from it. Except for one paragraph at the end, this paper offers no proposals to improve the science. Instead, whenever a valid criticism threatens to grow into a constructive suggestion, the mission takes over and turns discussion to another unhelpful blanket attack on the research. Sadly, this ideological bent has poisoned and largely destroyed the value of what might have been a valuable contribution.