TREELIM – Population processes and evolution

B3 Evolutionary adaptation (reciprocal common gardens)

Nursery grown tree seedlings performed very well at ALL transplant gardens. In fact, they did so well and grew so fast, that we had considered terminating the trial in mid 2011, but then felt the benefit of completing the 2011 season was greater than the drawback of losing root information. There was a clear reduction of growth with elevation, a shift in phenology (paper submitted by Vitasse et al.), and a dominance of local environmental conditions over genetic properties (small provenance effects), suggesting only minor ecotypic differentiation across the elevational gradient of sampled seed mother trees. The transplant gardens that we established on recommendation by the ERC peers above the current tree limit are still intact, and will be further observed to obtain longer performance data.

By reciprocally growing tree seedlings from low and high elevation provenances in common gardens at different elevations, we are aiming to reveal evolutionary adaptation of different species to cold growth conditions and disentangle genetic from phenotypic plasticity in these trees. For this experiment we used seedlings of 7 common broad-leaved, deciduous tree species raised from seeds that were collected at lower elevations and at the upper edge of their respective distribution ranges in two regions of the Swiss Alps (near Martigny, i.e. ‘western provenances’ and in the area of Chur, i.e. ‘eastern provenances’). The gardens with over 4500 seedlings were established in spring 2010 at four elevations in the western Alps and eastern Alps, respectively, and covered an elevation gradient form 430 m asl. to 1700 m asl. In each garden seedlings from high and low populations from both regions were grown for two consecutive seasons (Fig. 6). To avoid site-specific effects apart from temperature, all seedlings were grown in pots with standardized soil and watered throughout summer.

In 2010 and 2011, larger field campaigns were accomplished to measure growth parameters (seedling height and stem diameter, health status) and collect standardized leaf-discs for the measurements of specific leaf area and chemical analyses. Additional leaf material for genetic analyses was sampled in cooperation with Felix Gugerli (research unit ecological genetics and evolution, Swiss federal research institute WSL). Starting in March 2011, the phenological stages (bud break, leaf unfolding and bud set) of all seedlings were individually assessed in all gardens. The final harvest of all seedlings took place in autumn 2011. The data collected over the last two years are currently under scrutiny.

The assessment of bud break and leaf unfolding in 2011 revealed genetic adaptation of the leaf unfolding date between low and high elevation provenances of six out of seven tree species. Except of Fagus sylvatica, all species showed later leaf unfolding of high elevation populations compared to low elevation population in most gardens. However, this ecotypic adaptation was largely overridden by a high degree of phenotypic plasticity in all species, with leaf unfolding occurring later in the higher the gardens (the lower the temperatures, Fig. 7). Thus, in spite of their genetic differentiation, populations are shifting their phenology to a similar extent in response to temperature variation, which suggests that populations from the leading upslope edge of the species distribution are likely to respond similarly than populations from the middle of the species distribution range to ongoing climate warming. A manuscript presenting the results of this study is currently in preparation.

The measurements for growth and biomass production in the reciprocal common gardens were completed in December 2011. The first preliminary analyses indicated ecotypic adaptation in most species and a very strong temperature effect among the gardens. Irrespective of species, seedlings from low elevation provenances exhibited better growth than their high elevation counterparts (as depict exemplarily for Acer in Fig. 8). In contrast, the environmental and genetic influences on SLA appear to be small, but some of the species indicate genetic variation in this leaf trait (Fig. 8).