Introduction
Extinction and local extirpation with consequent species turnover are expected to occur in cold environments because of reduced habitat suitability for resident species and improved conditions for low elevation or low latitude migrants (Gottfried et al. 2012, Pauli et al. 2012, Porro et al. 2019). Hence, the distribution of ‘cold-adapted’ species is projected to undergo severe range loss in the next future (Steinbauer et al. 2020). Arctic species are predicted to experience range contraction at their southern trailing-range locations while the distribution range of northern, leading-edge populations may even expand beyond their historical cold border by migrating poleward (Hickling et al. 2006). For alpine species, the distribution range of lower-elevation trailing-edge populations is expected to shrink, unless these peripheral plants possess full or partial adaptations to expected climate warming scenarios (Dullinger et al. 2012, Angert et al. 2020). Cold-adapted alpine species are also predicted to shift their ranges towards colder locations, i.e., towards higher elevations (Lenoir et al. 2008) or in micro topographic niches (Körner and Hiltbrunner 2021). Distribution shifts triggered by climate change are projected using correlational bioclimate envelope models, which can overestimate species losses because key aspects such as demographic dynamics, plant plasticity, etc. are ignored by models (Randin et al. 2009, Niskanen et al. 2017, Casazza et al. 2021). Phenotypic plasticity, i.e. the responsiveness of plants to changing environment, consists in the capacity of a genotype to express different phenotypes under diverse environmental conditions (Kawecki and Ebert 2004, Garland and Kelly 2006, Leimu and Fischer 2008). If the performance of populations is correlated with environmental conditions at the provenance site this can be interpreted as a test of local adaptation, although local adaptation may be marginal in certain species and environments (Macel et al. 2007, Beierkuhnlein et al. 2011). Peripheral trailing-edge plant populations often host unique genetic traits when compared to core populations (Eckert et al. 2008, Mathiasen et al. 2021). Local adaptation may be particularly important at range limits where the selective pressure of climatic conditions on a species’ population is usually stronger than in its range centre and where genetic recombination may be limited due to geographic isolation (Choler et al. 2004, Kawecki 2008, Paul et al. 2011). It is less clear whether some of the unique genetic traits of peripheral populations may be useful to cope with climate change. Abeli et al. (2018) hypothesized that peripheral plant populations of cold-adapted species may already be adapted to local warm conditions, especially in southern refugia. This seems to be supported by long-term stability of several trailing-edge populations (Abeli et al. 2012, Mathiasen et al. 2021) and eventually by models in which demographic and genetic simulations are included (Cotto et al. 2017).
Results of common garden experiments suggest that genotypes from warmer sites in many cases have an advantage in terms of demographic traits compared to genotypes from colder sites when exposed to warming (Peterson et al. 2016, Bontrager and Angert 2019). For example, individuals of Eucalyptus grandis experiencing natural long-term exposure to heat waves have higher capacity to express protective proteins than individuals from populations with lower exposure to extreme weather events (Maher et al. 2019). Dickman et al. (2019) recorded adaptation to drought in seeds from peripheral low-elevation populations of Mimulus laciniatus. On the other hand, there is increasing empirical evidence that plants can react to climate warming with mechanisms involving phenotypic plasticity (Schneider 2022). Plasticity allows plants to grow under variable conditions and to cope with environmental changes by immediate (plastic) responses in the absence of genetic (evolutionary) variation. There is strong evidence that phenotypic plasticity is directly conditioned by genotypic diversity (Eller et al. 2017). However, phenotypic plasticity can also derive from changes in gene expression controlled by epigenetic response, which confers plants the ability to respond in the short term to a rapidly changing environment (Nicotra et al. 2010, Pagliarani and Gambino 2019, Ashapkin et al. 2020). Furthermore, phenotypic plasticity can be caused by the intrinsic ability of plants to acclimate to changing ecological conditions (Valdés et al. 2019). On the other hand, when environmental conditions exert strong selective pressures on organisms, like in ecologically/geographically peripheral populations, useful trait selection and stabilisation may result in loss of plasticity (Anstett et al. 2021; Usui et al. 2023). Whatever the mechanisms involved, phenotypic plasticity plays an important role in explaining adaptation of cold-climate plants to warming climate. For example, high level of phenotypic variation buffers the alpine cliff plant Heterotheca brandegeei from the negative impacts of warming and drought in southern North America (Winkler et al. 2019). For arctic-alpine plant species it is more problematic to define where core areas and marginal areas are located within their highly fragmented distribution ranges. Therefore, it is conceptually difficult to evaluate the geographic location of trailing-edge vs. leading-edge populations of arctic-alpine species. In fact, arctic-alpine species underwent multiple re-colonization events during the climatic vicissitudes that occurred during the Pleistocene. So, arctic-alpine species with similar present-day distributions may have had different glacial refugia and their distribution range may thus have differed strongly during the Pleistocene (Brochmann and Brysting 2008). This has profound implication on the putative genetic differentiation among populations. Marginal populations of many plant and animal species often show lower genetic diversity and high differentiation of molecular markers compared to conspecific core populations (Van Rossum et al. 2004, Tsumura et al. 2007, Eckert et al. 2008). Based on their complex colonization history, populations of arctic-alpine species may reflect differing pathways of genetic differentiation, possibly untethered by latitude. Furthermore, populations located at the southern margin of the present-day distribution area do not represent the warmest sites within the species’ range (Abeli et al. 2015). This may constitute a mismatch between geographic and ecological marginality (Soule 1973). It is generally assumed that the distribution range of southern populations of arctic-alpine species will shrink unless these peripheral plants possess full or partial adaptations to expected climate warming scenarios (Abeli et al. 2018, Dullinger et al. 2012). The distribution range of northern populations may even expand beyond their historical cold border. However, northward expansion will be precluded if warming negatively impacts growth of plants from northern populations at cooler temperatures compared to plants from southern populations (DeMarche et al. 2018). Interestingly, a recent study predicted stronger decline of arctic-alpine species at the northern range margins than at southern locations. (Niskanen et al. 2019).
In this study, we used the Alpine catchfly [Viscaria alpina (L.) G. Don] for assessing the potential of northern and southern populations of an arctic-alpine species to cope with a warming climate. V. alpina represents an excellent model to examine multi-faceted responses of different populations to global warming. In fact, V. alpina has a broad distribution range that consists in a vast high-latitude area extending from North America to eastern Eurasia and a set of smaller mid-latitude mountain areas prevalently located in southern Europe, from the Pyrenées and the Cantabrian mountains to the Alps and the Northern Apennines (Nagy 2013). V. alpina exhibits relatively high genetic diversity among populations whose locations differ strongly in terms of climatic, altitudinal and edaphic features (Haraldsen and Wesenberg 1993). Most important, both the southern and the northern populations of V. alpina span a wide thermal range with differences of up to 7-9 °C in average mean annual temperature (Nagy 2013). To assess the ability of V. alpina to cope with a warmer climate, we selected four European populations, two from a northern area and two from a southern area. In both areas one of the two populations stemmed from a colder site and the other from a warmer site with mean annual temperature difference of about 5 °C between the two sites. We cultivated plants originated from seeds collected from the four populations at warm temperature and assessed their responses in terms of morphological and ecophysiological traits based on phenotypic short-term response to high temperature. Furthermore, we cultivated a further set of plants under more extreme thermal conditions that simulated a short-term heat spell mimicking heat wave conditions. Indeed, the frequency of heat waves is increasing in recent years which results in compound effects of heat and drought (Mukherjee and Mishra 2021) besides the single effects of higher temperature. Experiments were conducted on seedlings, as this early life stage is highly sensitive to variation in climate and represents a major bottleneck to species recruitment, playing a key role in the distribution and dynamics of plant populations (Vázquez-Ramírez and Venn 2021).