There’s a press release that reports on a drastic shift in sex ratio of offspring of water voles in animals with radio tracking collars.
Radio-tracking associated with ‘dramatic shift’ in water vole sex ratio
Wildlife researchers are being warned that radio-tracking could be affecting the animals they are studying. According to new research published today in the British Ecological Society’s Journal of Applied Ecology, fitting radio-collars to water voles was associated with a “dramatic shift” in the sex ratio of the animals’ offspring, casting doubt on the assumption that radio-tracking does not fundamentally affect the biology of radio-collared water voles.
The water vole (Arvicola terrestris) is an endangered species in the UK, and ecologists Dr Tom Moorhouse and Professor David Macdonald of Oxford University’s Wildlife Conservation Research Unit had been monitoring the size of two populations – one in Norfolk and one Wiltshire – when they noticed a 48% decline in the number of females born at the Norfolk site. At both sites vole numbers had been monitored using small traps baited with apple and carrot during 2002 and 2003, voles being released after counting. However, Moorhouse and Macdonald had also fitted radio-collars to 38 of the voles caught at the Norfolk site for in-depth study of the voles’ movements.
According to Moorhouse and Macdonald: “Our analysis revealed that the most likely cause for the female decline was a shift in the sex ratio of young raised by radio-collared females. This result has implications for conservation research, especially for monitoring water vole populations.”
Skewed sex ratios have been reported in stressed and malnourished females of various species, including water voles. The ecologists suggest this could be explained in terms of the local resource competition hypothesis, which predicts that mothers with access to poor resources will produce offspring of the sex most likely to disperse and therefore reduce local competition for resources.
“Radio-collars clearly have the potential to cause some stress to water voles, and it is possible that this might stimulate sex-ratio adjustment as part of an evolutionary mechanism mitigating impacts of suboptimal habitats, similar to the sex-ratio bias and stress response in food deprived water voles,” they say.
Researchers have long been aware that the techniques they use had the potential to cause unexpected effects, and there have been many studies into the effects of radio-collars. However, this is the first study to show an association between radio-collars and sex ratio, although further work is needed to establish a causal link. According to Moorhouse and Macdonald: “We would expect any such effect to be species-specific, but our results will alert those studying other small mammals to look for similar associations. Our findings are a reminder that the assumption that the use of radio-collars does not fundamentally affect the biology of the subjects always requires careful checking. This study emphasises that the effects of commonplace wildlife marking and tracking techniques may be difficult to detect and yet both important and revealing. Clearly, it is both scientifically and ethically important to be aware of, and to strive to minimise, any such effects.”
One simplified view of Heisenberg’s uncertainty principle is that observation changes the system being observed. While Heisenberg’s principle applies to the very small end of things, biologists have to come to grips with the issue of whether their techniques actually change what they are trying to study. This has tripped people up in many cases. For example, ornithologists discovered that different colors of leg bands used to identify individuals made a difference in mating success.
There was a lecture that Diane and I attended in 1993 that described physiologists trying to establish whether a certain species of penguin was bioenergetically at its limits. The person giving the lecture was a krill biologist who was talking about the work of some of her colleagues. The species had rookeries in the Antarctic several miles inland from the edge of the ice. This cuts down wonderfully on the predators taking advantage of not-very-mobile adults and vulnerable young. But it also means that the parents switch off care of the young to go to the ocean and stuff themselves with krill. If a penguin that is foraging is nabbed by a seal or killer whale its mate will eventually have to abandon the egg or young bird. So along come the physiologists with their nets, scooping up penguins fresh from the ocean and headed back to the rookery. To get a measure of energy resources, one up-ends the penguin to get the stomach contents, and then does bomb calorimetry or a similar analytical technique to figure out energy input. Coupled with known data on penguin metabolism, the result obtained by the physiologists was that these penguins were operating at the very edge of the bioenergetic resources they had.
After the lecture, we approached the lecturer with a question. “Do you know,” we asked, “what the physiologists put back into the penguins they analyzed as a replacement diet?” The lecturer gave us a stricken look. She didn’t know, in fact, what they had put in, or indeed if they had put in anything. The outcome of not replacing the penguin’s repast was easily predictable, in light of the research results themselves. If they had simply dumped a now-empty penguin back on the ice edge, it would have no choice but to go back into the ocean for another round of foraging. And back at the rookery, its mate would have no choice but to abandon its offspring so that it, operating at its bioenergetic limit, would have enough energy to make it back to the ocean to forage for itself. We don’t know that the penguins in the study went hungry, but we don’t know for certain that they didn’t, either. But it does illustrate another way in which the act of observing can have an effect, even if it doesn’t show up in the specific results of the research.
As biologists, we owe it to our research subjects to think very carefully about the effects that our interaction with them may produce. There is no pat answer. “Model it on a computer!” works when we already have a lot of information to base a model upon, not when we are mostly ignorant concerning some aspect of biology. I say this as someone who routinely has applied computational techniques to biological research, including a variety of models. Most of the time we still need to make observations of real organisms. The reasons to question our observational techniques are multiple: to assure that the results are not unduly skewed by the mode of observation, to avoid introducing extraneous factors into the study population, and to minimize any negative effects our observations may have for the individuals involved.