Population genetics of south European Atlantic salmon under global change AMERICA GARCIA VALIENTE *, EDWARD BEALL w andEVAGARCIA-VAZQUEZ * *Departamento de Biologia Funcional, Universidad de Oviedo. C/Julian Claveria s/n, 33006-Oviedo, Spain, wUMR ECOBIOP, INRA Station d’Hydrobiologie, 64310 Saint Pe ´e sur Nivelle, France Abstract Populations at the edge of species distributions are especially vulnerable to climate change. Genetic changes as well as modification of their population structure are expected as reactions to global warming. Atlantic salmon (Salmo salar) inhabiting south France has been chosen as a model for studying the effect of global warming in marginal populations during the last 15 years. Increased gene flow between neighboring popula- tions and dichotomy of maturation age between sexes have been identified as two main population changes significantly associated with high values of the North Atlantic Oscillation index, a global climate indicator. Although occurrence of isolated popula- tions in each river (or even tributary) is a paradigm for this species, at least in northern areas, increased gene flow between rivers is forecasted as long as climate warming increases, favoring metapopulations at regional level. Keywords: between-cohort mating, gene flow, genetic variation, global warming, life history, mature male parr, NAO index, population structure, Salmo salar , straying Received 23 December 2008 and accepted 28 February 2009 Introduction Fish population variability and fisheries activities are closely related to weather and climate dynamics (Leho- dey et al., 2006). There are many examples of relation- ships between climate variability and fish populations. Catches of many species are declining associated with intense global climate change during the last decades (Parsons & Lear, 2001; Beaugrand et al., 2003; Chavez et al., 2003; Condron et al., 2005; Astthorsson et al., 2007; Brander, 2007; Greene & Pershing, 2007; Hannesson, 2007; Stenevik & Sundby, 2007). Among them, Atlantic salmon Salmo salar L. can be chosen as a particularly valuable model because of its anadromous condition (it is vulnerable to changes in both freshwater and marine habitats), very long migration (more than 6000 km for south European salmon) and because it is subjected to both marine and freshwater fisheries. For Atlantic salmon, and other anadromous fish, it has been predicted that changes in patterns of develop- ment (maturation and smoltification) and of return timing may be the principal consequences of climate change (Mangel, 1994). In the last decade many studies have focused on the effects of climate change on Atlan- tic salmon population dynamics and demography. Climate change can be responsible for long-term changes in sea age composition of salmon populations (Gudjonsson et al., 1995). Survival of salmon post-smolt at sea is highly influenced by ocean conditions in the marine nursery areas, suggesting that declines in Atlan- tic salmon are in part due to global climate change (Friedland et al., 2003). Unfavorable climatic conditions have been claimed to be the ultimate reason for the poor salmon production (Jonsson & Jonsson, 2004a). Changes in life histories, abundance and return rate of Atlantic salmon have been described in association with in- creased sea temperatures (Niemela et al., 2006). Under conditions of global climate change, it is expected that the populations at the edge of a species’ geographical distribution are the most vulnerable (Walther et al., 2002). Although northward range expan- sion is not necessarily associated with southern range retraction and species may simply expand their geogra- phical distribution (Parmesan et al., 1999; Thomas & Lennon, 1999), in some cases the southernmost areas become unsuitable for northern species. For example, the capelin (Mallotus villosus) and the halibut (Hipoglos- sus hipoglossus) have been displaced northwards, aban- doning their southernmost distribution areas (Schrank, Correspondence: Eva Garcia-Vazquez, fax 1 34985103534, e-mail: egv@uniovi.es Global Change Biology (2010) 16, 36–47, doi: 10.1111/j.1365-2486.2009.01922.x 36 r 2009 Blackwell Publishing Ltd