Evidence for niche similarities in the allopatric sister species Lepus castroviejoi and Lepus corsicanus. (23rd January 2014)

• Record Type: Journal Article
• Authors: Acevedo, Pelayo; Melo‐Ferreira, José; Real, Raimundo; Alves, Paulo C.; Silman, Miles
• Is Part Of: Journal of biogeography. Volume 41:Number 5(2014:May)
• Page Start: 977
• Page End: 986
• Publication Date: 2014-01-23
• Subjects: Biogeography -- Periodicals; 578.09
• DOI: 10.1111/jbi.12270
• Languages: English
• ISSNs: 0305-0270

Abstract

<abstract abstract-type="main" id="jbi12270-abs-0001"> <title>Abstract</title> <sec id="jbi12270-sec-0001" sec-type="section"> <title>Aim</title> <p> <italic>Lepus castroviejoi</italic> and <italic>Lepus corsicanus</italic> are sister species with allopatric distributions that share extensive phenotypic and genetic variation. Under the framework of niche conservatism, we assessed whether these species have similar ecological niches, which could provide insights into their mode of divergence, conservation, and taxonomic status.</p> </sec> <sec id="jbi12270-sec-0002" sec-type="section"> <title>Location</title> <p>The distribution range of <italic>L. castroviejoi</italic> in the northern Iberian Peninsula, and that of <italic>L. corsicanus</italic> in mainland Italy and Sicily.</p> </sec> <sec id="jbi12270-sec-0003" sec-type="section"> <title>Methods</title> <p>We developed spatially explicit ecological models to characterize the niches of the two species by modelling them separately and together. Individual models were transferred to the territory of the sister species to explore their niche relationships. Predictions were assessed for discrimination and calibration in a cross‐assessment procedure.</p> </sec> <sec id="jbi12270-sec-0004" sec-type="section"> <title>Results</title> <p>The model trained with <italic>L. castroviejoi</italic> was not able to predict the range of <italic>L. corsicanus</italic>, whereas the model trained with <italic>L. corsicanus</italic> was able to discriminate the <italic>L. castroviejoi</italic> distribution better than by chance alone (AUC = 0.814), although the reliability of the predictions was limited. The model trained with <italic>L. corsicanus</italic> in Italy's mainland (excluding the range in Sicily), however, discriminated <italic>L. castroviejoi</italic> presences/absences (AUC = 0.788) and accurately predicted its probability of occurrence. Furthermore, a well‐calibrated model, which was able to discriminate the species distributions (<italic>L. castroviejoi</italic>, AUC = 0.828; <italic>L. corsicanus</italic>, AUC = 0.956), was obtained when the species were considered together.</p> </sec> <sec id="jbi12270-sec-0005" sec-type="section"> <title>Main conclusions</title> <p>Our results suggest that <italic>L. castroviejoi</italic> and <italic>L. corsicanus</italic> share extensive niche properties, which reinforces their possible conspecific status. The ecological niche of their ancestor may have resembled the present occupied niche of <italic>L. corsicanus</italic> in mainland Italy, given that this model was able to accurately predict the distribution range of both species. Finally, ecological evidence suggests that niche conservatism may explain the fragmentation in the distribution range of their ancestor, which may have been the driver of the initial stages of divergence.</p> </sec> </abstract>