Mayfly mandibles: as seen in the IPCC report

I have been reading through the palaeoclimate chapter in the new IPCC report, in part so I can write a post on transfer functions in the report for Victor Venema’s climate scientists’ reviews. This post is not that review, instead I want to focus on half a sentence in section 5.5.5 on megadroughts and floods.

while lake-sediment proxies from the same region [Scandinavia] suggest wetter winters (Luoto et al., 2013).

The reference is:

Luoto, TP,  Helama, S & Nevalainen, L 2013: Stream flow intensity of the Saavanjoki River, eastern Finland, during the past 1500 years reflected by mayfly and caddisfly mandibles in adjacent lake sediments. Journal of Hydrology, 476, 147-153.

A mayfly nymph.

I know the first author’s work and have commented on it before. I had read this paper before, but had decided not to comment on it even though it fits into my blog’s themes of novel proxies and palaeohydrology, as I thought it would be quietly ignored. But now the IPCC has cited it. I’ve checked the draft versions of the palaeoclimate chapter that I reviewed in 2011 and 2012, this paper is not cited – I didn’t miss it – it was presumably added in response to reviewer comments on the second draft version.

The paper attempts to reconstruct stream-flow, and hence winter precipitation, from the abundance of caddisfly and mayfly mandibles found in the sediment from the middle of a lake. The basic assumption is that when there is increased stream-flow, more mandibles will be washed from the river into the lake, and so their abundance in the core will be higher.

Caddisfly and mayfly are aquatic insects little used as palaeoecological proxies in Quaternary studies. One of the few prior examples is Solem & Birks (2000) who identified caddisfly remains to species level and made inferences about the temperature of Kråkenes, western Norway, during the deglaciation. I didn’t find any literature using mayfly. Given this minimal use of these insect groups, probably in part reflecting their low abundance in the sediment, it should be exciting to see them being used to reconstruct a variable that would normally be difficult to constrain. However, …

The obvious problem is that both caddisfly and mayfly can live in both lakes and rivers. Luoto et al acknowledge this by citing relevant literature, but do not dwell on the consequence. The consequence is that variability in the abundance of caddisfly and mayfly mandibles in the sediment core could be ascribed to either 1) variability in within-lake production or 2) variability in transport of mandibles into the lake (and this could reflect 2a) variability in production in the river or 2b) variability in transport efficiency). Because of these various possible causes of downcore variability in mandible abundance, it is not possible to conclude anything about stream-flow variability.

A caddisfly larva

Species-level identification may have helped to distinguish in-lake production from transported mandibles. Given the enormous number of mandibles being deposited in the lake (lake area * sediment accumulation rate * mandible density), it is doubtful that the river is contributing more than a small proportion of them.

The statistics in Luoto et al are fairly strange (for example, they treat their binary (lake/river) predictor as though it were continuous and somehow persuade their software to fit a quadratic model even though there are only levels in the predictor). These issues are irrelevant to the main problem in the paper as the statistics test whether mandible abundance is the same in the lake as in the river. They do not test where the mandibles in the lake are coming from, nor if precipitation variability can explain downcore variability in mandible abundance.

This is a paper so flawed that it should never have been submitted, let alone published. Part of the reason for the failure of peer review must be that it was submitted to a journal that publishes very few ecological or palaeoecological papers, so it is likely outside the expertise of the editor and the journal’s usual reviewers.

Climate skeptics wishing to denigrate the IPCC should not get excited by the inclusion of  one flawed paper amongst ~900 in the palaeoclimate chapter. The chapter represents the paper correctly, and its inclusion has no impact on any of the chapters conclusions. All that can be concluded is that none of the IPCC chapter 5 authors are experts on mayfly and caddisfly palaeoecology.

Other climate skeptics may wish to follow McIntyre in complaining that the text was altered after the second-order-draft was reviewed. This is such a silly argument. If the draft text was final, what would be the point of reviewing it. So perhaps another exhausting round of reviews to check the changes to the SOD? And then another to check the revisions made after that round. And so on ad infinitum. No thanks, the IPCC process is burdensome enough already. Some text will inevitably change between the last draft and the final version, references may be added and some dropped, all this allows the possibility for some stray problems to enter the text, but these are very unlikely to affect more than details.

Luoto et al is a flawed paper, but its inclusion in the IPCC’s palaeoclimate chapter is no more important to the conclusions of that chapter than the inclusion of a stray comma I found (and subsequently lost).

About these ads

About Richard Telford

Ecologist with interests in quantitative methods and palaeoenvironments
This entry was posted in climate, Novel proxies, Palaeohydrology, Peer reviewed literature and tagged , , , . Bookmark the permalink.

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