Spatial dynamics of intraspecific genetic variation in European beech (Fagus sylvatica L.)

Fagus sylvatica is one of the most important tree species in Europe. Due to its extensive distribution and dominance in its range, the performance of European forests in the contexts of perceived climate change will largely depend on the reaction of beech forests. Adequate information about the genetic diversity and structure of the species is indispensable for the conservation and sustainable utilization of the forests. Forest management systems currently in use in Germany are considered sustainable. However, their impact on the genetic diversity and spatial genetic structure of F. sylvatica is largely unknown. In these contexts, the main objectives of this dissertation were a) to assess the genetic diversity and differentiation of F. sylvatica at different spatial scales, b) to determine the level and magnitude of fine-scale genetic structure of the beech stands, c) to investigate the effect of forest management activities on the genetic diversity and fine-scale genetic structure of the beech stands independent of the marker used and d) to compare microsatellite and AFLP markers to estimate genetic diversity and structure of the stands, and their applicability to detect management impacts on genetic structure of the stand. Exhaustive sampling was performed in most of the plots to harvest the leaf samples from 3620 beech trees from 30 different stands located at three different regions (exploratories) covering major differences in the population histories and natural geographical distribution of beech forests in Germany. Ten beech stands, including managed and unmanaged, were sampled at each exploratory. Within population and among population genetic variability and fine-scale genetic structure was assessed by using two types of markers, codominant and biparentally inherited microsatellite (six SSRs and three EST-SSRs) for all individuals and dominant AFLP markers (126 loci) for a subset of 100 individuals from each plot across six beech stands. The differences in genetic diversity, differentiation and SGS parameters between managed and unmanaged beech stands were analyzed and their significance levels were estimated for all stands together and separately at exploratory level. As expected, European beech retains a high level of genetic diversity at all studied beech stands across the three exploratories. Despite the low difference between extreme values, the mean genetic diversity (AR and HS) of the beech stands across exploratories was significantly different based on microsatellite data (p<0.05). These significant differences were not detected by using neutral SSRs. However, genic SSR (three EST-SSR loci) revealed highly significant (p<0.001) differences in these diversity parameters among exploratories, which could be linked to some sort of adaptive selection. The mean genetic diversity among beech stands was increasing from Southern to Central and North-East Germany. Mantel tests revealed a positive and significant relation between genetic and geographic distances for all beech stands across the three exploratories. However, the relationships were non-significant among beech stands at each exploratory level. Most of the genetic variation was retained within stands (98%) followed by among stands within exploratories (1.17%) and among exploratories (0.82%); all of these variation components were highly significant. Moreover, about 90% of the population pairs based on microsatellite data were significantly differentiated from each other. Beech stands at S. Alb have the strongest genetic differentiation from other exploratories. Fagus sylvatica at Hainich and S. Chorin were genetically more similar (mean pairwise FST=0.0126) than S. Alb and Hainich (pairwise FST=0.0176) despite similar geographical separation between them (~300 km). The strongest differentiation was measured between the most widely separated exploratories S. Chorin and S. Alb (~600 km, mean pairwise FST=0.0271). These two exploratories vary in population histories, recolonization period, geographical elevations, soil types and climatic variables. The strength of fine-scale spatial genetic structure was quantified by the Sp statistics. Due to the high variability in the stand structures, geographic and climatic patterns, very heterogeneous levels of SGS were estimated across studied beech stands. A very weak to moderate levels of SGS were estimated with microsatellite (Sp=0.00004 to 0.3208) and AFLP data (Sp=0.00583 to 0.04661). The mean kinship coefficients among pairs of individuals at first distance class (0-10 m) were also small suggesting a weak family structure. Spatial autocorrelograms between the mean kinship coefficients over logarithms of distance classes revealed the maximum extension of significant SGS up to the distance of 30 m by both SSR and AFLP markers. Twenty-six out of 29 beech stands studied with SSR markers and all six beech stands studied with AFLP markers showed a significant fine-scale genetic structure, which is primarily due to restricted seed dispersals. In general, both microsatellite and AFLP markers showed that forest management has no impact on genetic diversity but it has significantly reduced fine-scale genetic structure of the beech stands. There was no significant impact at any of the diversity parameters due to management activities in Hainich and S. Chorin exploratories. However, management has significantly decreased the observed heterozygosity (HO) and genetic diversity (HS) in S. Alb. Moreover, the management activities have significantly reduced inbreeding coefficient (FIS) and genetic differentiation (FST) in Hainich and S. Alb, respectively. Additionally, AFLP markers did not reveal any significant difference in genetic diversity and differentiation parameters between managed and unmanaged beech stands. Seedlings at the close proximity of the seed tree are genetically more related due to the restricted seed dispersals in beech. Removal of individuals through various cleaning, thinning and harvesting activities in managed forests eliminate the continuity of related genotypes, consequently reduces the spatial genetic structure of the stand. Reduced level of SGS was ascertained in the managed compared to the unmanaged beech stands by all three exploratories. Selections of genetic markers have direct effects on the estimation of genetic diversity and finescale genetic structure of the stands due to their inherent properties. Six SSR markers confirmed the highest levels of genetic diversity parameters compared to 3 EST-SSRs and 126 AFLP loci. However, AFLP markers exhibited the strongest genetic differentiation among all three used marker types. Further, AFLP markers showed a higher level of fine-scale genetic structure compared to SSR markers. However, both of these markers were significantly and positively correlated to estimate the SGS (r=0.9, p<0.05). Both SSR and AFLP markers demonstrated similar trend of kinship coefficients over the natural logarithm of distances across beech stands. Efficient gene flow, long life-span and large effective population sizes of F. sylvatica contribute to the maintenance of genetic diversity. High level of intraspecific genetic variation within naturally regenerated beech stand may have possibly buffer the management disturbances on the genetic diversity. This information about the management impacts on the genetic diversity and their spatial distribution within populations can be utilized more effectively for the sustainable utilization of the beech forests and conservation of genetic resources in the study area.