In this week’s QAECO Reading Group we talked about whether and how protected areas (PAs) may facilitate species’ range expansions – a topic that is becoming increasingly important as the global climate changes.
In their recent PNAS paper, Thomas et al (2012) measured colonization events at the expanding edge of species ranges in the UK. They compared the proportion of colonization events that have occurred inside PAs in comparison to colonization events outside PAs. They showed that in recent decades approximately 40% of all colonization events of 7 focal butterfly and bird species, for which detailed observation records were available, took place inside PAs even though protected areas comprise only 8.4% of UK’s terrestrial area. The species specific metrics suggest that colonization inside PAs is 4.2 times more frequent than expected based on the availability of PAs. However, the results were significant only for 5 out of the 7 focal species (app. 71%). Therefore, to support their conclusions, they performed the same analysis with a data of 256 less-intensively surveyed invertebrate species, and found that in 98% of the cases species were disproportionately colonizing PAs than non-PAs. Consequently, they conclude that protected areas do facilitate species range expansions and therefore play an important role in conservation, particularly in the era of global warming.
We thought that this paper addressed a very important research question and that the take home message was well formulated. However, several members of our group questioned technical aspects of this work, specifically the methods used to account for observation biases, and the interaction between land use, habitat type and PAs. Most of the data came from voluntary organisations such as Biological Records Centre. Voluntary observers typically go places where species of interest are likely to be found (such as PAs), and we felt that the analytical framework did not exclude the possibility that observation bias was driving some of the documented patterns.
We also believe the paper would have benefitted from separating the effects of land use and habitat/vegetation types from tenure (i.e. PA vs. non-PA). Biodiversity-friendly land uses and land cover are likely to be disproportionally represented in PAs. Therefore comparing colonisations within PAs with the rest of the landscape (including highly disturbed areas), without exploring whether other areas of semi-natural vegetation were also performing such a role, makes an unfair comparison. This is particularly the case within a landscape that is quite modified, even within PAs (compared with an Australian’s concept of a protected area, at least). Finally, we pondered how much the results would differ if PAs other than Sites of Special Scientific Interest were included in the analysis (such as private PAs).
To take the ideas of this paper further, an interesting next step would be to explore what happens at the contracting edge of species range. For example, do PAs help to reduce the rate of contraction? It would also be useful to model the dispersal dynamics of species at their expanding range edge, and factors affecting changes in habitat suitability, and therefore account for the processes that drive expansion.
Check out what our colleagues in the Metapopulation Research Group said about this paper.
Interested in protected areas and conservation planning?
Check out these QAECO ecologists:
By Heini, Luke, James, Emily and team
Thomas et al. (2012) Protected areas facilitate species’ range expansions. PNAS, 109(35): 14063-14068.