Genetic diversity, gene flow and inbreeding in pedunculate oak (Quercus robur L.) in fragmented forest stands

Deforestation and fragmentation of once large and continuous forests are probably one of the most important and widespread human-induced changes that have been made to forests worldwide. Small forest fragments will contain small tree populations, which are more prone to extinction and to losses of genetic variation through increased random genetic drift, selfing and mating among closely related individuals. A good understanding of the processes that shape genetic diversity across generations within tree populations is therefore indispensable for the evaluation and conservation of the evolutionary potential of trees in the long term and for increasing individual tree performance in the short term. Nevertheless, despite the field of forest landscape genetics has been an attractive science domain throughout the past decades, our knowledge of the genetic consequences of anthropogenic disturbances for tree populations remains relatively limited. Furthermore the effects of forest fragmentation and forest management practices on the genetic diversity of tree populations are highly debated in the scientific literature, resulting in what is known as the paradox of forest fragmentation genetics . This thesis aimed at gaining better insights into the effects of human land use and forest management activities on the genetic diversity, gene flow and inbreeding of small and fragmented pedunculate oak populations (Quercus robur L.). To do so, a multidisciplinary approach was adopted, in which we first conducted a meta-analysis of the available literature on forest landscape genetics. Subsequently we examined the maintenance of genetic diversity, contemporary gene flow and individual tree performance in small and fragmented pedunculate oak stands in Northern Belgium. Based on our meta-analysis, we could reject the traditional idea that woody plant species are relatively resistant to genetic losses following habitat fragmentation, as they were as prone to genetic erosion as herbaceous species studied in earlier meta-analyses. Although wind-pollinated species are presumed to have extensive pollen flow, we found evidence for pollen limitation and genetic losses, which were similar to those observed in insect-pollinated species. When we compared the results of this meta-analysis with what was found in our in-depth study on pedunculate oak, no clear losses of genetic diversity were observed across generations in the studied pedunculate oak stands. This suggest that the effective population sizes (Ne) were large enough to avoid the negative impacts of genetic drift and inbreeding. Since high out-of-plot pollen immigration rates were observed in all study plots, extensive pollen flow may have counteracted the negative genetic consequence associated with small tree populations. Nevertheless, even under high out-of-plot pollen immigration rates, we found that an important fraction of mating events occurred at short distances between neighbouring trees. Also in our study plots, less diverse local pollen pools were obtained insmall and low-density forest stands, which suggested that pollen limitation due to a restricted number of local mating partners was not totally compensated by the observed high out-of-plot pollen immigration rates. Ultimately, this may increase the likelihood of inbreeding in subsequent generations. The heterozygosity level of tree individuals should however be maximized in natural populations, as we detected a weak but significant relationship between heterozygosity and tree performance. Moreover our study was one of the first that showed an effect of high drought stress levels on the strength of the observed heterozygosity-fitness correlations. This suggests that in view of climate change the genetic consequences of forest fragmentation could be further exacerbated.To mitigate the potential negative genetic consequences through increased inbreeding and to maintain rare, potentially adaptive alleles in small pedunculate oak stands, Ne. should be maximized in the seedling cohort. In this respect conservation resources should be directed to establish genetic connectivity across fragmented oak populations. However this is not always realistic in the highly fragmented forest landscapes of Northern Belgium, through which more customized management practices are needed to maintain or increase Ne. Supplementation of natural regeneration by composite provenancing may be a good option to maintain and increase genetic diversity in small and fragmented oak stands and to enlarge the evolutionary potential to current and future environmental changes.status: publishe