Past forest cover explains current genetic differentiation in the Carpathian newt (Lissotriton montandoni), but not in the smooth newt (L. vulgaris)

Aim: Current genetic variation and differentiation are expected to reflect the effects of past rather than present landscapes due to time lags, that is, the time necessary for genetic diversity to reach equilibrium and reflect demography. Time lags can affect our ability to infer landscape use and model connectivity and also obscure the genetic consequences of recent landscape changes. In this work, we test whether past forest cover better explains contemporary patterns of genetic differentiation in two closely related but ecologically distinct newt species-Lissotriton montandoni and L. vulgaris.Location: Carpathian Mountains and foothills.Methods: Genetic differentiation between populations was related to landscape resistance optimized with tools from landscape genetics, for multiple timeframes, using forest-cover data from 1963 to 2015. Analyses were conducted for pairs of populations at distances from 1 to 50 km.Results: We find evidence for a time lag in L. montandoni, with forest cover from 40 years ago (ca. 10 newt generations) better explaining current genetic differentiation. In L. vulgaris, current genetic differentiation was better predicted by present land-cover models with lower resistance given to open forests. This result may reflect the generalist ecology of L. vulgaris, its lower effective population sizes and exposure to habitat destruction and fragmentation.Main Conclusions: Our study provides evidence for time lags in L. montandoni, showing that the genetic consequences of landscape change for some species are not yet evident. Our findings highlight the interspecific variation in time lag prevalence and demonstrate that current patterns of genetic differentiation should be interpreted in the context of historical landscape changes.