Comparative genetic structure of two forest tree species resulting from historic human land-use

Background/Question/Methods

Human land-use can have profound genetic impacts on natural tree populations via population size reductions and population fragmentation. These genetic effects, however, can differ across species depending on a range of factors such as life history traits and the particular landscape contexts. This study focuses on the effects of landscape-level forest clearance that have occurred since the 1600s in western/central Massachusetts on the genetic structure of two late-successional tree species, American beech (Fagus grandifolia) and Eastern hemlock (Tsuga canadensis). As disturbance-sensitive species, both showed population declines due to land-use but certain differences in their life history traits may result in different extent of genetic responses. We asked whether this forest clearance and subsequent forest recolonization had the same magnitude of impact on the spatial patterns of genetic variation in both species. We collected hemlock and beech individuals co-occurring in 12 forest stands and genotyped them across five and eight microsatellite loci, respectively.  For each species, we classified and grouped populations based on their forest history: primary (logged but never cleared forests) and secondary (forests that recolonized abandoned agriculture lands) forests. Genetic diversity, population differentiation and gene flow were then calculated and compared between groups for each species.

Results/Conclusions

We observed a significant increase in genetic divergence among secondary compared to primary forests only in hemlock populations (F_st secondary =0.078; F_st primary=0.035, P=0.05) and not in beech (F_st secondary = 0.057, F_st primary =0.037, P=0.66). Genetic bottlenecks were also evident in four hemlock populations while none was observed among beech populations. We examined the overall population genetic structure of both species and found that hemlock populations were partitioned into three genetic clusters. In contrast, beech populations were grouped into a single population, indicating higher levels of gene flow among populations. These results suggest that hemlock and beech populations may be affected differently by human land-use.