October 18, 2011

Why on an island?

Every time I tell some friend about my research on the Lundy Island House sparrows, I get asked the same question. They always ask: “Why do you study this on an island? Wouldn’t it be much cheaper and easier to do that in your backyard?”

Fig. 1: Why the effort?

It must indeed seem odd that we study the ubiquitous house sparrow on such a remote location like Lundy Island. Our reasons for doing so go far beyond a simple wish for an exotic field location: island populations minimize the chances of immigration and emigration, and such a “closed” island population is much easier to survey, nearly comparable to a population in the laboratory. Such natural laboratories are extremely well suited to answer questions about evolution and ecology.
Fig. 2: We would not mind to study birds on a more exotic island every now and then, especially in the winter.

But why is that so? It is because only a closed population allow us to estimate fitness precisely. Fitness is an important concept in evolutionary biology, and we must measure it in order to study evolution. Evolution is the frequency change of inherited traits in a population over time, and we know that this happens through selection acting on phenotypes. Selection is the mechanism where genes that are not beneficial will be weeded out, while advantageous genes will persist. How can one determine which bird possess beneficial genes and which does not? Are there birds that contribute more to the future generation than others?

We estimate the success an individual has in passing on its’ genes to the next generation by counting the number of its’ offspring that survive until they breed themselves. This total number, or the life-time reproductive success is often called reproductive fitness by evolutionary biologists.
However, counting all surviving offspring of a bird is not an easy task. First, we must make sure that all chicks of individual birds are individually marked, so that once they have fledged, we can still know which one belongs to which parent. Birds are extremely well suited for this because the method to individually mark them – leg ringing ¬– is extraordinarily well studied. Most rings do not harm birds, or affect their behaviour. On Lundy, we use metal rings from the BTO, those have a unique running number engraved, and, in addition, colour rings. We use four different colour rings per bird, and having ten different colours, that gives us 10^4 possible unique colour ring combinations, enough to individually mark 10000 birds.

Fig. 3: Colour rings

Now comes the difficult part: finding the birds back. In a normal situation, say, on your average farm, sparrows can fly away, maybe to the next farm, and start a brood there. However, most often, scientists do not have the means (financially but also logistically) to monitor all surrounding farms for present sparrows from their own population, and even if they do, they can never be sure that a sparrow did not travel on to the next farm, which may have been considered too far away to be controlled. So it is, under non-island circumstances, extremely difficult to determine whether a bird has emigrated (or, in biologists linga: dispersed), or has died. Counting only those sparrows that breed at the farm where they were born thus very likely underestimates real reproductive fitness. Worse even, if those sparrows that prefer to stay at their home location are somewhat different from those adventurous sparrows that disperse, it may be that the whole dataset is biased towards sparrows who do not like to explore their surroundings, which may be problematic for certain studies.

Lucky for us Lundy sparrow reserachers, we do not have that problem, simply because sparrows do not fly over water. Lundy is an island, and the nearest mainland is 19km away. That means, if we have no dispersal, we can be sure that those birds that cannot be found back in the next year have died.

Fig. 4: Lundy is an island 19 km away from the mainland. The sea is unusually quiet in this picture.

Well, we want to be completely honest here. There is the odd chance that sparrows could potentially fly away from Lundy to the mainland, but because it’s 19 k of open sea and strong winds, chances of any sparrow attempting this also reaching the mainland are pretty low. We can estimate those chances, and we think that it’s less than one bird per year that crosses over from Lundy to the mainland (how we come to that number is a subject of a future blog post). Since this number is really, really low, especially in comparison to mainland populations (who often have 40% or more birds who disperse annually), we can ignore it most of the times.

And therefore, the reproductive fitness estimates of our sparrows on Lundy island are really, really precise, and unbiased. And that is one reason why we study sparrows on an island, in our natural laboratory. It makes our data much more valuable. And that is worth the often very rough boat trip over the Bristol Channel with the Oldenburg.

Fig. 5: The Oldenburg in Bideford.

No comments:

Post a Comment