My main research interests focus on how genetic variation is distributed across species ranges, and the effects of past, present and future climate change on levels and patterns of intraspecific diversity.
Populations of species affected by global warming will either have to adapt in situ to new climatic conditions, migrate to track suitable habitat, or become extinct. The potential effects of present and future climate change can be assessed through modelling approaches, but further insights can be gained by examining the effects of previous periods of climate change associated with the glacial cycles of the Quaternary period (ca. 2.6 MA - present). More information on the effects of the ice ages on species distributions can be found in Provan & Bennett (2008) Trends Ecol. Evol. 23, 564-571 and Bennett & Provan (2008) Quatern. Sci. Rev. 27, 2449-2455.
Comparative phylogeography of Orthilia secunda and Monotropa hypopitys
Orthilia secunda (top) and Monotropa hypopitys (bottom) are two members of family Pyrolaceae which share broadly overlapping distributions across North America and Eurasia. In North America, both species exhibit disjunct distributions, being found in the northeast and northwest of the continent, but generally absent from the central grasslands of the Great Plains. Despite this, phylogeographic analysis indicated that whilst M. hypopitys persisted in multiple refugia in both the east and the west throughout the last glacial maximum (LGM; ca. 18 - 21 KA), O. secunda was confined to exclusively western refugia and recolonized eastwards after the retreat of the ice sheets. Thus, despite exhibiting largely similar distributions today, these species have experienced very different histories of response to previous periods of climate change throughout the most recent glacial period. These findings have been published as Beatty & Provan (2010) Mol. Ecol. 19, 5009-5021 and Beatty & Provan (2011) J. Biogeogr. [DOI: 10.1111/j.1365-2699.2011.02513.x]
A comparative study of both species across Europe (Beatty & Provan  BMC Evol. Biol. 11, 29) showed that both species persisted throughout the LGM in southern refugia. O. secunda displayed higher levels of genetic diversity in recolonized regions, however, with populations of M. hypopitys north of putative refugial areas being fixed for one of two haplotypes. As models of projected species distributions under climate change indicate loss of rear-edge populations of both species, the concentration of genetic diversity in southern populations of M. hypopitys means that loss of these populations will have a greater impact on rangewide genetic diversity than in O. secunda.
Unique genetic variation at a species rear edge under threat from global climate change
Studies on a range of seaweeds across the northeastern North Atlantic over the last few decades have documented a northward shift in the southern range limits of many species, including Chondrus crispus (left). Phylogeographic analysis of populations from across the range of this species on both sides of the North Atlantic revealed that it persisted through the LGM in both North American and European (Iberia and the English Channel) refugia. Populations from Iberia contain a high proportion of unique endemic haplotypes, and a statistical phylogeographic analysis indicated that this is because populations from the Iberian refugium did not contribute to the recolonization of Europe after the LGM. Consequently, loss of these populations as a result of global warming will have a disproportionately large impact on the rangewide genetic diversity of the species, a case similar to that of the European populations of M. hypopitys described above. This research has been published in Provan & Maggs (2012) Proc. R. Soc. B 279, 39-47
Origin of the Lusitanian flora
The Lusitanian flora and fauna represent one of the great unsolved puzzles of biogeography. These species, which include plants such as the Strawberry Tree (Arbutus unedo), St Daboec's Heath (Daboecia cantabrica) and St Patrick's Cabbage (Saxifraga spathularis), and animals such as the Kerry Slug (Geomalacus maculosus), are only found in southern and western Ireland and northern Spain and Portugal (and occasionally Brittany), but are largely or completely absent from intervening countries, including England. Debate concerning the origin and distribution of the Lusitanian element has been ongoing since the mid-19th Century. Some naturalists, primarily Edward Forbes, suggested that Irish populations must have survived through the last glaciation in situ, whilst others, including Clement Reid, insisted that very few or no species could have persisted during the maximum advance of the ice sheets, and that recolonisation from Iberia was a much more likely scenario.
In this project, funded by the Leverhulme Trust and being carried out by Gemma Beatty, we are using a phylogeographic approach to determine the origin of Irish populations of several Lusitanian plant species, including Saxifraga spathularis, Pinguicula grandiflora (Large-flowered Butterwort, above left) and Euphorbia hyberna (Irish Spurge, below left). Palaeodistribution modelling of our species' ranges at the LGM has been used to develop hypotheses concerning their possible glacial histories, and statistical phylogeographic analysis of a combination of chloroplast and single-copy nuclear markers will allow us to discriminate between the possible scenarios of in situ persistence through the LGM or postglacial recolonisation from Iberia.
Effect of climate change on genetic diversity in marine invertebrates
The cool water copepod Calanus finmarchicus is a key species in North Atlantic marine ecosystems. Over the last 40 years, however, data from the Continuous Plankton Recorder Survey have highlighted a 70% reduction in C. finmarchicus biomass, coupled with a gradual northward shift in the species’ distribution, which have both been linked with climate change. To determine the potential for C. finmarchicus to track changes in habitat availability and maintain stable effective population sizes, we assessed levels of gene flow and dispersal in current-day populations, as well as using a coalescent approach together with palaeodistribution modelling to elucidate the historical population demography of the species over previous changes in the Earth’s climate. We found high levels of dispersal and a constant effective population size over the last 359,000 – 566,000 YBP, which together suggest that C. finmarchicus may possess the capacity to track changes in available habitat, a feature that may be of crucial importance to the species’ ability to cope with the current period of global climate change. (Provan et al.  Proc. R. Soc. B 276, 301-307).
A study on populations of the Black Katy Chiton (Katharina tunicata) from Vancouver Island, BC, Canada using single nucleotide polymorphisms (SNPs) also revealed high levels of gene flow between populations, as well as evidence of long-term population stability (Doonan et al.  Biol. J. Linnean Soc. 106, 589-597). As with C. finmarchicus, these suggest that habitat tracking could increase the species' chance of survival in the face of global climate change.
Cryptic marine refugia in the red seaweed Palmaria palmata
The red seaweed Palmaria palmata has a similar distribution to that of Chondrus crispus (see above), being found on both sides of the northern North Atlantic. We carried out a phylogeographic analysis of P. palmata across its range to identify potential refugial areas during the LGM. As well as finding evidence for a refugium at the southern edge of the species current-day range, we also found evidence of a cryptic refugium in the English Channel, possibly associated with the Hurd Deep, an enigmatic trench that might have persisted as a marine lake during the LGM. Many subsequent studies on a range of marine species have confirmed the existence of this Channel refugium, first highlighted in Provan et al. (2005) Mol. Ecol. 14, 793-803.
Range edge effects and genetic diversity in peripheral populations
Peripheral populations of species are often characterised by low levels of genetic diversity as a result of genetic drift, restricted gene flow, inbreeding and, in plants, asexual reproduction. Low levels of genetic diversity in populations tend to be associated with reduced adaptive potential. Given that peripheral populations will be most affected by global climate change, both at the leading and the rear edge, knowledge of the genetic composition of such populations at both neutral loci and loci putatively associated with adaptive traits is of great importance in predicting the response of species to global warming.
Levels of clonality in peripheral populations
Populations of both Orthilia secunda and Monotropa hypopitys in Ireland exist at the extreme western limit of their European ranges and are highly fragmented (left). A comparison between populations of O. secunda from the centre of the species range and populations from the species range edge revealed a decrease in levels of genetic diversity corresponding to an increase in the incidence of clonal reproduction. Irish populations were completely monoclonal, with each population comprising a single clone, even in larger populations numbering several hundred individuals. This was not the case, however, for M. hypopitys, where levels of clonal diversity were much higher and towards the upper end of values reported in other plants that exhibit clonal reproduction. Nevertheless, levels of genetic diversity in both species were significantly lower than those in populations elsewhere across both species ranges in Europe. Although the species are found in the same two areas in Northern Ireland, the differences in levels of clonality be be due to the fact that M. hypopitys is a more temperate species than O. secunda, which has a largely boreal distribution. Consequently, global climate change may affect the two species differently. This research has been published as Beatty et al. (2008) Diversity Distrib. 14, 546-555 and Beatty & Provan (2011) Annals Bot. 107, 663-670.
Extinction via hybridization at a species range edge: implications for habitat tracking
Putative hybrid plants between Pyrola minor (top) and Pyrola grandiflora (bottom) have been reported from Qeqertarsuaq, Greenland, where P. minor exists as small, fragmented populations at the northern edge of its range but where P. grandiflora is common. Genetic analysis using nuclear and chloroplast SNPs confirmed the hybrid nature of the morphologically intermediate individuals, but revealed that hybridization was unidirectional, with P. minor plants always pollinated by P. grandiflora. A similar scenario was also found in Churchill, Canada, where P. minor is also at the edge of its range. Cryptic hybrids were also detected in both locations. Such extensive unidirectional hybridization could ultimately lead to the extinction of peripheral populations of P. minor as a result of genetic assimilation, the replacement of the P. minor gene pool as a result of pollen swamping by P. grandiflora. If such population extinction via hybridization can effectively act as a barrier to northward migration of P. minor into areas where P. grandiflora is common, this could compromise the ability of the species to respond to climate change by habitat tracking. These findings have been published as Beatty et al. (2010) Diversity Distrib. 16, 1-9