Questions for Yan et al (2015), Willie Soon’s new paper

Willie Soon and Bob Carter are coauthors on a new paper in Nature Geosciences by Yan et al. It looks like an interesting paper exploring the movements of the intertropical convergence zone between China and Australia over the last millennium using palaeohydrology records and relating this to solar activity. Unlike many papers looking for evidence of the impact of solar variability on the Earth’s climate in palaeo-data, it is not immediately obvious that this paper is flawed. So I’m going to ask some critical questions about this paper.

Coauthorship

Papers published by journal in the Nature group all include author contribution statements.

H.Y. designed the study and wrote the manuscript. W.W. contributed to the section discussing climate model results. W.S. contributed significantly to improvements in the manuscript. Z.A., W.Z. and Z.L. contributed to discussion of the results and manuscript refinement. Y.W. and R.M.C. contributed to improving the English.

Contributing to the “discussion of the results and management refinement” could certainly merit coauthorship, whereas contribution “to improving the English”does not obviously merit coauthorship. Nature’s policies on coauthorship are not prescriptive

The author list should include all appropriate researchers and no others. Authorship provides credit for a researcher’s contributions to a study and carries accountability. The Nature journals do not prescribe the kinds of contributions that warrant authorship but encourage transparency by publishing author contributions statements.

so no foul there. But the Australian Research Council’s Code for responsible conduct of research has well defined criteria for authorship (Carter is Australian).

To be named as an author, a researcher must have made a substantial scholarly contribution to the work and be able to take responsibility for at least that part of the work they contributed.

Attribution of authorship depends to some extent on the discipline, but in all cases, authorship must be based on substantial contributions in a combination of:

  • conception and design of the project
  •  analysis and interpretation of research data
  •  drafting significant parts of the work or critically revising it so as to contribute to the interpretation.

Substantial intellectual involvement is required.

How do the authors reconcile their inclusion of Carter as a coauthor for “improving the English” with this code of conduct?

Conflicts of interest

Yan et al (2015) make the following statement about conflicts of interest

The authors declare no competing financial interests.

Willie Soon has made this statement before. It turned out not to be entirely true.

Nature’s guidance to authors is clear.

… competing interests are defined as those of a financial nature that, through their potential influence on behaviour or content or from perception of such potential influences, could undermine the objectivity, integrity or perceived value of a publication.

They can include any of the following:

Funding: Research support (including salaries, equipment, supplies, reimbursement for attending symposia, and other expenses) by organizations that may gain or lose financially through this publication.

It is difficult to specify a threshold at which a financial interest becomes significant, … , so we offer as one possible practical alternative guideline: “Any undeclared competing financial interests that could embarrass you were they to become publicly known after your work was published.”

Since Willie Soon appears to be impossible to embarrass, perhaps he feels that the $1.2 million funding from the fossil fuel industry and speakers fees for attending the Heartland and Doctors for Disaster Preparedness meetings are below the necessary threshold. Would his coathors agree?

Volcanic eruptions

And so onto the science. The authors write

The Little Ice Age (defined as AD 1400–1850) was associated with low solar irradiance and high atmospheric aerosol concentrations as a result of several large volcanic eruptions.

and

strong volcanic eruptions have been detected during the LIA (that is, coinciding with the Maunder Minimum and Dalton Minimum)

Yet somehow the paper focuses only on the solar variability. Why?

Solar variability

How sensitive are the results to the choice of the amplitude of total solar irradiance variability? Yan et al set this to 0.25% rather than the smaller amplitude preferred by the IPCC.

Site selection

Yan et al show differences in reconstructed palaeohydrology from a set of sites across a region stretching from Australia to China. They show that the northern and southern sites become drier during the Little Ice Age, whereas the equatorial sites become wetter. Site selection is a crucial component of any analysis of this type. What does Yan et al have to say about site selection criteria? Nothing. Nothing at all. Nothing about how good the chronologies are or about the confidence to which proxy variability can be attributed to hydrological change. The paper should not have been published without this information.

Schematic map depicting the current extent of the monsoon/ITCZ in the  East Asia-Australia area, and the hypothesized contraction of the monsoon/ITCZ zone  during the LIA. The palaeo-hydrologic records are marked. Pink lines indicate the northern and southern limits of modern EASM and ASM. Red shaded areas indicate the areas with possibly lower precipitation during the LIA than during the MCA. Blue shaded area indicate the converse.

Schematic map depicting the current extent of the monsoon/ITCZ in the East Asia-Australia area, and the hypothesized contraction of the monsoon/ITCZ zone during the LIA. The palaeo-hydrologic records are marked. Pink lines indicate the northern and southern limits of modern EASM and ASM. Red shaded areas indicate the areas with possibly lower precipitation during the LIA than during the MCA. Blue shaded area indicate the converse.

It might be fruitful to compare the geographic patterns in hydrological change that Yan et al find over the last thousand years with those reported by Tian et al (2013).

Tian et al. Comparison of effective moisture from Lake Khuisiin with other records from Asia, using a five–point moisture scale. (a) The spatial pattern of index of moisture variation. (b) The coloured profiles on the left indicate the proportion of records available, with the following moisture codes: blue line >0, green shading = 1, blue shading = 2, red line <0, yellow shading = −1, and red shading = −2; the black profile on the right indicates the index of moisture variations through time.

Tian et al. Comparison of effective moisture from Lake Khuisiin with other records from Asia, using a five–point moisture scale. (a) The spatial pattern of index of moisture variation. (b) The coloured profiles on the left indicate the proportion of records available, with the following moisture codes: blue line >0, green shading = 1, blue shading = 2, red line <0, yellow shading = −1, and red shading = −2; the black profile on the right indicates the index of moisture variations through time.

The latitudinal patterns reported by Yan et al do not seem so obvious in Tian et al’s compilation. So what criteria were used?

Advertisements

About richard telford

Ecologist with interests in quantitative methods and palaeoenvironments
This entry was posted in Peer reviewed literature, solar variability and tagged , . Bookmark the permalink.

11 Responses to Questions for Yan et al (2015), Willie Soon’s new paper

  1. Kaustubh says:

    Nice post – I was wondering when you would get to this. Yan et al. also neglect recent rainfall reconstructions in the Western Pacific that don’t really show any correspondence with solar forcing nor have a “well-defined” Little Ice Age. For example, Partin et al., 2013 and Maupin et al., 2014. I was also surprised that there was no mention of the South Pacific Convergence Zone, which is an integral component of precipitation variability in the Western Pacific Warm Pool.

    • Thanks. I only found this paper last night – my RSS feeder is slow for Nature Geosciences.

      Any idea how reliable the coral d18O records are as hydrological rather than temperature records?

      • Kaustubh says:

        Coral d18O is first and foremost a function of the temperature and d18O of the seawater that the coral grew in. Depending on the location and time scale of interest (annual, interannual, decadal etc.), coral d18O can be dominated by d18O seawater variability or temperature variability, such that it can be inferred as one or the other. This has to be demonstrated for that location of course, prior to interpretation. The d18O of seawater is a proxy for salinity. Though salinity can be indicative of hydrology, many parameters can influence salinity in a location, and precipitation is only one of those variables apart from evaporation and advection.

        In the Western Pacific, one has to be careful dealing with coral d18O due to advection of fresher/saltier waters on seasonal, interannual, and multidecadal timescales, apart from salinity changes due to precipitation directly. As a further example of slack/selective proxy picking in Yan et al., they choose a Sr/Ca and d18O record from New Caledonia (D6, D7 in their paper). Both of these records, in context of Yan et al.’s time period of interest, and according to the articles themselves, should be considered as sea-surface temperature proxies and not proxies of rainfall…and I haven’t even gone through the other proxies in detail.

      • Shades of Soon and Baliunas (2003) perchance? If Yan et al are mixing carefully selected hydrological & temperature proxies together, this might get interesting.
        I think the Pages2k database of palaeohydrogical records is available – it will be interesting to see which they have chosen, and how any extras can be justified.

      • I’ve had a look at the other proxies Yan et al use. The New Caledonian record in the only one that the original authors believe to be a temperature record rather than a moisture record. However Burrows et al (D10) is not convinced that their record of peat humidification represents regional climate rather than local factors as two nearby bogs have different timing of wet and dry phases.
        For unknown reasons, D11 is not shown in the supplementary material. This and D12 are based on an unpublished conference paper (page 15) which does not have enough information to evaluate the reconstruction properly, e.g., the age depth model is not shown.

      • Kaustubh says:

        Interesting! I don’t think it is anywhere close to being a repeat of Soon & Baliunas and sadly, mixing proxies is not unique to this high-profile paper. For example, Neukom et al. 2014 include proxies of rainfall that correlate highly with temperature (stalagmite d18O for example). At least they included their selection metrics in their text…

      • It can be easy to justify including precipitation proxies in a temperature analysis or vice versa. For example, rainfall at a site in Peru might reflect El Nino and a its large temperature anomaly. But it does need to be justified.

  2. gregladen says:

    Frankly, I think “improving the English” is a potentially author-worthy part of a paper. I’ve done a LOT of that sort of work myself, and it is often not a matter of simply making grammar better, but actually, in extracting meaning and making meaning meaningful, as it were. There are things that are said in papers, and cited, that only exist in the literature because somebody “improved the english” and there are plenty of things …. such as those things the professor tells the young grad student to find by “giving this a very careful read” that are not so obvious and that should have been better said.

    Having said that I quickly add that I’ve never actually been listed as a co-author for “improving english” if that is all I did on a paper. But, I would not fault it. It is a judgement call and I’d rather see a bit of encouragement (by being an author) on papers for this so that the damn papers are better.

    So I’m calling that particular issue not important.

    As for the science in the paper, I only read the paper through once and I didn’t get the impression there as anything glaring, but like you, I had questions about methods, data selection, etc. Thanks very much for your commentary on this, very interesting.

    • If Y.W. and R.M.C. had been listed as contributing “to discussion of the results and manuscript refinement.” in addition to “improving the English”, I would not have commented. But I don’t think that improving the English by itself merits more than an acknowledgement. If it deserved coauthorship, my partner, who gives ecology and geology manuscripts a language-wash (rates available on request), would have many more papers that I do.

      But yes, this is not the most important aspect of this paper.

      • gregladen says:

        As I said, I’ve never “improved english” and then gotten co-authorship for that, but I can see it being such an essential and time consuming part of a paper, that one could. Indeed, I did this as part of a translation process (from a non-Enlgish language) once, and was paid to do it. The reason I was hired was because I was one of the few people on Earth who really understood the very narrow subfield of science being discussed AND was not already on one side or another of a big fight over what to say about it or otherwise an interested party. So my job was to objectively interpret meaning across two very different languages and cultures. That was “improving English” and little else, but totally deserved authorship.

        But it was a thesis so instead of co-authorship I got a pile of money. I really think it is better left open and not made into a wedge, though I think we are generally in agreement on it.

        One more small point: I suspect (an educated suspicion, tho) that the attempts to list what you SHOULD be co-author for are in part an answer to the problem of people being co-author because they either a) got a grant that funded the research or b) own or control a piece of equipment or a lab in which the research was done, and otherwise had absolutely nothing to do with it.

    • Eli Rabett says:

      That’s really a call for the editors and as you say not very important except for construction of h numbers.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s