Sunday 17 May 2020

Is bird size changing in the Fynbos due to climate change?

In this recently published paper, I show there is no evidence for sustained directional body change attributable to climate for the most commonly ringed birds around the south western Cape: but weather certainly influences how much the birds weigh.

Body mass and condition of a fynbos bird community: investigating impacts of time, weather and raptor abundance from long-term citizen- science datasets

The history behind this paper, published recently in Ostrich: Journal of African Ornithology, goes back a way: and I’m not just talking about the data set which stretches back to 1980 or so.

During my postdoctoral period, I read from several papers that an impact of climate change on birds was that many were shrinking in size due to global warming as smaller birds are able to dissipate heat better than large birds, which conserve heat better. This had been reported for the UK from ringing records (Yom Tov et al ) as well as for Australia by Janet Gardner. She went on to propose that a third law of climate change was that species should respond by getting smaller, and I had the honour of seeing her present at the IOC held in Japan in 2014. I was very interested in doing something similar for the Fynbos birds: since the set of endemic birds seem especially vulnerable to climate change: results I presented at that IOC and published the following year in Bird Conservation International.

 As a bird ringer, I take many measurements, but I have not taken them for long enough to examine patterns from my own dataset alone: I needed more data. I approached Dieter Oschadleus, then head of SAFRING (the organization responsible for managing ringing records and activities in South Africa and further afield in Africa). Ringers submit their biometric information to SAFRING, but Dieter requested that I needed to get permission from the individual ringers in order to use their information. Unfortunately, because of data protection and old records, this turned out to be a lot harder than it seemed. Also: many ringers had passed away during the lost history of the project. Nonetheless I persevered until I had permission to use most of the data available in the Western Cape.

It was a big day for me when the huge data file from SAFRING finally landed in my inbox. Since I’d previously attempted to examine ringing records for the nectarivores, I knew that this was the start of the journey: this data was historically unvetted, and so coming up with a sensible way of data checking was the next challenge. Also, trying to get spatially and temporally good data was also a challenge: there are only intermittent records for most of the Fynbos, so I had to select the region around Cape Town. That was fine, since that is still the only true Mediterranean type climate of the Fynbos (with winter rainfall), which also would reduce the confounder of different sub-climates. I also needed to get South African Weather Service weather station data, which involved signing contracts to ensure the data would not be shared. And then that data was also less than perfect and some serious thinking was required in order to come up with the best way of getting a weather series that matched the ringing data. In the end I opted for a mean approach, based on the 4 stations with the best data. 

Of course, field work was also a distraction: that year was an especially busy one tracking down Hottentot Buttonquails and catching Cape Rockjumpers, so progress with analysis was slow, and before I knew it: it was the end of 2016 and my postdoc. I’d also taken on the role of Editor-in-Chief of Ostrich, and part of my commitment to improving the journal was to publish my own best work there. A few years later I’d learn this is frowned upon, but the commitments had been made by then, and I’ll get onto the independent editorial process in a bit. In the meantime, behind on the analysis I was still analysing data in my hotel room at the Pan African Ornithological Congress in Senegal for the presentation to be given that week to Africa’s select ornithologists. At that stage it did appear that some species were showing significant body mass decrease with time: but that was a rudimentary analysis not taking into account the impact of ringer, season and other variables: and these had a very large effect on the patterns.

At the conference, leading ornithologists Will Cresswell and Colin Beale pointed out that an explanation for body mass change could simply be that birds were adapting to changes in the predator landscape: the predation-starvation hypothesis. Raptors would thus also need to be considered when considering body size change. I missed my chance to submit the article to the PAOC special issue, as analyses became more and more complicated. It would only be 2018 that I would first submit, having performed reams of statistical tests, learnt time-series analysis and other statistical techniques to try and get to grips with this dataset and what it meant.

The peer-review process was independently handled by former editor-in-chief of Ostrich, Lizanne Roxburgh, with Amanda Ridley (whom I have never met) in charge of the decision. Amanda managed to get no less than four reviewers to peer review the article: I’ve never had to deal with that many before (2 is standard). As such, there were many comments and a general mood of skepticism from the reviewers, many of which called for different re-analysis from different modelling stand points. It would take no less than 3 resubmissions, most with additional angles or modelling approaches, before the reviewers were satisfied and relinquished that at least in this case the expected body size change attributable to climate change was mostly not observed. It is important to note that this does not mean that climate is not impacting these birds: clear body mass responses were noted in relation to temperature for several species: but at the short-term scale, with temperature of the previous day and rainfall negatively impacting the weight of many species.

Since this article was submitted and accepted for publication, two other notable studies have come out that have received wide publicity. In South Africa, an analysis of ringing data for Mountain Wagtail suggested directional body size change with the title: “Are animals shrinking due to climate change? Temperaturemediated selection on body mass in mountain wagtails.”

This study found that mass of birds was getting lower, and that survival at different temperature was linked to body mass. First, it is important to note that mass is often a poor proxy of size, especially at the levels measured in this study. Mass is dependent on many things: age, sex, time of capture, predation pressure and the recapture record for this study is very uneven. While they show that mass of ‘replacement’ birds decreased with time, they also found that the mass of recaptured birds was lower: clearly these were not changing size. The magnitude of these mass changes was tiny (0.027 g/year), which could be a measurement artifact (scales measured to 0.1g). In addition: there was no statistical change in temperature for this ringing site. It is thus a stretch to link climate change to body size change, which is the message of the paper. By contrast, we tested not only mass, but also wing length and body condition indices (mass controlled for by size, where we use wing measurement) and found no temporally directional response for most of these birds. An exception was Barn Swallow: the only migrant in our set of 40 species.

The second notable paper that was taken up in the media as evidence for climate change influences on body morphology https://onlinelibrary.wiley.com/doi/full/10.1111/ele.13434. This impressive study showed increasing wing length and decreasing tarsus length for a diverse range of migratory species from north America, with changes explained by increasing summer temperatures.

However, when one examines the literature on body size change in relation to climate change, it is notable that one of the biggest papers in the field found exactly the opposite result in Australia (decreasing wing length, used as a proxy for body size) but also attributed this to climate change.

At this stage it would appear that while attributing change in what we measure to climate is fashionable, there is certainly no evidence one way or the other for a ‘universal law’ at this period in climate time.  

Is this bird smaller than a few years ago? It may seem so from this ringer's perspective since she is now much bigger.


I would like to thank my co-authors on the paper for their valuable contributions: Phoebe Barnard especially for her faith in me over the years and positive attitude, Chris Lennard for insights into climate change, Ben Smit for his intellectual prowess, Mike Fraser (for the European perspective plus ringing hundreds of birds in the Fynbos) and Dieter Oschadleus: who also contributed much data and editorial advice on the paper.

Related Posts Plugin for WordPress, Blogger...