How many panthers?!

Since writing about our experiences in finding DNA evidence for the Panthera genus last summer, there has been a great deal of interest, including a string of newspaper reports in December (The Sun here, Mirror here, Daily Express here). Nice to have such impact! However, one aspect of these reports requires some clarification. Specifically, the claim that there are a 100 panthers out there, which I have been reported 'to believe' . Firstly, at this point, aside from sighting based evidence (which we do not deal with), there is no evidence that supports (or refutes) this claim. I had ended the previous report with the point that we need more evidence, more samples and genomic level information. This would allow us to establish whether there is a coherent (demonstrably related) population, which we think would be the evidence that would be more convincing to the world at large of the presence of panthers in the UK. This leads to the question, which the journalist naturally asked, how big does a population have to be?

This topic is a well studied area of population genetics, and indeed part of what I teach my students about. One of the most useful ways to study populations, and ascertain whether they may need conservation action, is population viability analysis. Basically, it's a simulation model approach. The first problem populations have to overcome to be viable, exist in the long term, is demographic stochasticity. This is the fluctuation in population size caused by the variance in offspring number each generation. While the average brood size may be say 3 or 4 for instance, the actual number of offspring from each mother will be a random number (following a Poisson distribution if you are statistically inclined), of 3 or 4 is the mean. It could be 7, or it could be 0. When populations are very small there is an increasing risk that out of bad luck very few offspring may be produced. Sooner or later this will hit 0, and the population will cease. Statistically, just to be clear of the problems of demographic stochasticity, there need to be well over 50 individuals when we are thinking about mammals of this sort of brood size. Then of course there are other problems one might think about, such as inbreeding which can expose the mutation load (deleterious genetic variation, often hidden in the heterozygous state). Such factors push the required population size up - it is quite complex, and dependent on the previous history of the population. Once you've considered these two aspects, then the next level of problem a population may have to negotiate is environmental variance (such as particularly harsh winters, for instance), which again requires increasingly large populations to be robust enough to survive. As a broad brushstroke, a viable population would have to be something over 100 individuals. Hence the newspaper reports. However, that is IF there IS a viable population, which is something I would emphasize that no one really knows.

So, while I would not say that I 'believe' there are 100 or more panthers out there, it would be more accurate to say, if there is a population that is viable, then it would have to be that big. At least. Of course the alternative is that if there are panthers in the UK, but they are not a viable population, then they must represent recent releases. Hopefully, genetic evidence if it comes will be able to decide between the two possibilities.