Ecology, biogeography and evolutionary history of the alpine long-eared bat plecotus macrobullaris

Resumen

6 Ecology, biogeography and evolutionary history of P. macrobullarisSummaryThe alpine long-eared bat Plecotus macrobullaris is a recently described species with a limitedknowledge and controversial understanding of its ecology and evolutionary history. Although itwas initially described as an alpine species, such alleged alpine nature was almost rejected bythe scientific community during the following years when it was reported at low elevations andeven at sea level. Additionally, the species has been captured in multiple types of habitats, anddifferent studies have reported contrasting conclusions about its ecological preferences. Thepresent dissertation aims to give insight into the ecological preferences, the biogeography andthe evolutionary history of the species.As a first step, I gathered information about the distribution of P. macrobullaris by revisitingpublished information and enquiring bat researchers. I obtained 311 presence points, 113 ofthem unpublished. That initial step unveiled possible geographical and altitudinal samplingbiases, since more than 50% of the presence records came from the Alps, and besides, alpineenvironments (areas above the upper forest treeline) had been almost completely overlooked inbat surveys. Thus, aiming to diminish the possible effect of sampling bias in the biogeographicperception of P. macrobullaris, I performed several field samplings in geographic areas andaltitudinal belts that had not been thoroughly studied before. The samplings carried out in thePyrenees, the Pindos Mountains, the Caucasus, and alpine areas in the Alps resulted in 45 newpresence points, 41 of them located above the treeline. Additionally, the nettings in the alpinebelt unveiled an unknown community of bats comprised by ten species and leaded by thealpine long-eared bat (70% of the captured bats belonged to this species).The high number of P. macrobullaris captured in alpine environments suggested that its affinityto such habitats could be greater than previously thought. To test whether those bats wereactually foraging at supraforestal habitats and identify their dietary preferences, I analysedthe diet of multiple specimens captured at alpine environments using molecular tools. Themethodology employed for the diet analysis was based on PCR+cloning and traditionalsequencing, which resulted to be a useful approach for isolating DNA sequences andidentifying prey species at species level, even though it has been rendered obsolete in thelast few years by Next Generation Sequencing technology. I identified 29 taxa consumed byP. macrobullaris at the species level, all but one of them being moths. Some of them are onlypresent in alpine meadows. Besides, the results did not support the idea of the bat foraging inthe forest as forest dwelling moths were lacking in the diet. Additionally, 96% of the identifiedprey species were tympanate moths and no evidence of gleaning (capturing insects from theground or the vegetation) behaviour was revealed, suggesting that P. macrobullaris foragesin alpine environments probably by aerial hawking using faint echolocation pulses to avoiddetection and evasive responses by hearing moths.After discovering that bats were using alpine environments for foraging, I studied their roostingbehaviour by radio-tracking 37 bats captured in alpine environments. I found that the PyreneanP. macrobullaris also roost and breed in the alpine belt, up to 2450 m above the sea level, andabout 1000 m above the treeline. Bats used rock crevices (30 roosts), scree deposits (21) andbuildings (3) for roosting, and most lactating and pregnant females relied on crevices. Batsselected areas with high meadow availability near the roost, while avoiding densely forestedareas. Foraging areas and roosting sites were located at the same elevation, indicating thatalpine long-eared bats use alpine areas for both roosting and foraging in the Pyrenees.However, breeding females roosted at lower elevations than nulliparous females and males,though they remained above the treeline.7The revision of the distribution of P. macrobullaris brought out a peculiar distribution pattern, since the species was found in a large geographic area extending from the Pyrenees (Western Europe) to the Zagros Mountains (Middle East), but always restricted to mountain environments. The comparison of the geographic distribution of P. macrobullaris with the distributions of other 503 European vertebrates showed that such a distribution is unique among bats, but is shared with other five vertebrates, namely the white-winged snowfinch Montifringilla nivalis, the alpine chough Pyrrhocorax graculus, the wallcreeper Tichodroma muraria, the alpine accentor Prunella collaris and the snow vole Chionomys nivalis. In order to understand the factors shaping the distribution of these animals with wide geographic distribution restricted to mountain environments, I modelled the distribution and the ecological niche of P. macrobullaris at two different scales. The broad scale modelling covered the whole Western Palearctic and was generated using presence points acquired in the previous step after applying several filters. The fine scale modelling was restricted to the Pyrenees mountain range, and based on high-quality roosting locations obtained by radio-tracking. The models fitted the observed data, even at low-elevation areas where the species had been reported. The broad scale modelling showed that topographic variables outperformed climatic predictors, and the abruptness of the landscape was identified as the most important variable shaping the distribution of P. macrobullaris. This variable likely acts as a surrogate of several ecological factors, which drive this species to be stacked to mountain areas. One of them is high rock-availability, which was confirmed in the fine-scale model as the variable with the best predictive ability. In fact, all the aforementioned species share the quality of foraging in open-space areas and sheltering in natural rock structures such as fissures or talus slopes, including the alpine long-eared bat. The best explanatory climatic variable was the mean summer temperature, which showed that P. macrobullaris is able to cope with temperatures spanning from 6 to 22 ºC, suggesting that it is not a cold-adapted, but rather a cold-tolerant eurythermic organism. Therefore, this study led me to conclude that the distribution of P. macrobullaris, and probably the rest of the mentioned species, is not primarily shaped by climatic determinants, but by topographic factors that provide rock-abundant and open-space habitats.Regarding the evolutionary history of the species, I used an integrative approach combining complete mitochondrial genomes with analytical techniques drawn from the fields of population genetics, phylogenetics, demographic history inference, and Ecological Niche Modelling (ENM) to understand how Pleistocene climatic oscillations affected the demographic history of different populations. I used the complete mitogenome sequences of 57 specimens collected all over the distribution of P. macrobullaris. I identified two deep lineages with very different demographic histories: the Western, restricted to the Pyrenees and the Alps, and the Eastern, which expanded across the mountain ranges east of the Dinaric Alps. ENM projections of past conditions predicted that suitability drops during cold stages in the areas inhabited by the Western lineage, while the opposite trend was observed in the mountains inhabited by the Eastern lineage. Accordingly, genetic diversity and Bayesian phylogenies showed a different molecular structuring in both lineages. The paleodemographic scenario that best fitted the data indicated that the Western lineage likely shrunk due to the repeated extensive glaciation events that the Alps and the Pyrenees experienced during the Pleistocene, while the Eastern lineage maintained a constant population size owing to more limited glaciation in the mountains of South-eastern Europe and the Middle East. The main conclusion obtained was that the evolutionary history of P. macrobullaris was shaped by the differential effect of glaciation events in different8 Ecology, biogeography and evolutionary history of P. macrobullarismountain systems, yielding complex population dynamics, with contrasting patterns ofexpansion and retreat depending on the geographical location. This study showed that thedemographic responses of species to events of climate variation were not homogeneous,but complex processes that can be successfully studied when integrating multiple newlydeveloped techniques.All the performed studies led me to conclude that the initial perception of P. macrobullarisas an alpine species, as well as its common name ¿alpine long-eared bat¿, do reflect itsactual ecological preferences and evolutionary history, which are unique among Europeanbats, and thus a special effort should be done to assure its conservation.