An ecological connectivity network maintains genetic diversity of a flagship wildflower, Pulsatilla vulgaris

Ecological connectivity networks have been proposed as an efficient way to reconnect communities in fragmented landscapes. Yet few studies have evaluated if they are successful at enhancing actual functional connectivity (i.e. realized dispersal or gene flow) of focal species, or if this enhanced connectivity is enough to maintain genetic diversity and fitness of plant populations. Here we test the efficacy of an ecological connectivity network implemented in southern Germany since 1989 to reconnect calcareous grassland fragments through rotational shepherding. We genotyped 1449 individuals from 57 populations and measured fitness-related traits in 10 populations of Pulsatilla vulgaris, a flagship species of calcareous grasslands in Europe. We tested if the shepherding network explained functional connectivity in P. vulgaris and if higher connectivity translated to higher genetic diversity and fitness of populations. We found that population-specific Fst was lowest in populations that had high connectivity within the shepherding network, and that well-connected populations within the network had significantly higher genetic diversity than ungrazed and more isolated grazed populations. Moreover, genetic diversity was significantly positively correlated with both seed set and seed mass. Together our results suggest that the implementation of an ecological shepherding network is an effective management measure to maintain functional connectivity and genetic diversity at the landscape scale for a calcareous grassland specialist. Populations with reduced genetic diversity would likely benefit from inclusion, or better integration into the ecological connectivity network. Our study demonstrates the often postulated but rarely tested sequence of positive associations between connectivity, genetic diversity, and fitness at the landscape scale, and provides a framework for testing the efficacy of ecological connectivity networks for focal species using molecular genetic tools.