Global Treeline Range Expansion Experiment (G-TREE) Davos

Project lead

Esther Frei


Christian Rixen

Project staff

Peter Bebi

Project duration

2013 - 2025

We investigate seedling emergence and survival of Norway spruce and European larch in an experiment at Stillberg near Davos. This project is part of the research initiative G-TREE that studies the influence of biotic and abiotic factors on the emergence and survival of tree seedlings at Arctic and alpine locations worldwide, allowing for a better understanding of treeline shifts in the face of climate change.

The Global Treeline Range Expansion Experiment (G-TREE) is a globally distributed collaborative project involving researchers from Europe, North America, Australia and New Zealand. At about 15 different treeline sites worldwide the mechanisms determining the distribution of tree populations at alpine and Arctic treelines are studied using a standardized experimental approach. It is expected that treelines will shift under climate change because growth conditions for trees at treeline will become more favourable. However, climate change-induced range expansion of treeline populations depends on successful recruitment, which requires dispersal of viable seeds followed by successful establishment of individual propagules. In the G-TREE project we disentangle various biotic (e.g. seed source, competition) and abiotic (e.g. elevation, soil substrate, weather) effects on early tree seedling recruitment at treeline, which will provide better insights into where, and under what circumstances, treeline expansion can occur.

The experiment

In 2013, we established a G-TREE experiment at Stillberg near Davos, with three experimental sites located along an elevational gradient across an alpine treeline ecotone. The lowest study site was at 1,930 m a.s.l., about 150 metres below the current treeline, the mid-elevation site was at the current treeline elevation (2,100 m a.s.l.), and the uppermost site was at 2,410 m a.s.l., i.e. approx. 300 metres above the current treeline.

At each of the three experimental sites, 20 plots were set up and assigned to different treatment combinations. At the beginning of the experiment, we removed the natural vegetation from part of the experimental plots. This treatment allowed us to study whether existing ground vegetation facilitates or inhibits emergence and growth of tree seedlings. Subsequently, seeds with a low- or a high-elevation origin (provenance) of the two subalpine tree species Norway spruce (Picea abies) and European larch (Larix decidua) were sown on half of the experimental plots, whereas no seeds were added to the remaining control plots. Additionally, every second experimental plot was covered by a metal cage immediately after seed addition to exclude herbivores.

Since 2013, seedling recruitment has been assessed by counting the seedlings at the beginning and the end of each growing season. Furthermore, seedling heights are measured in autumn. These measurements allow us to compare not only seedling emergence but also winter survival and growth of seedlings.

First results

The seedling inventories from the first two years demonstrate that tree seedlings can emerge well above the current treeline. Surprisingly, the proportion of germinating spruce and larch seedlings was even highest at the uppermost site, but only when seeds were added. The lack of emergence on control plots implies a lack of natural recruitment.

At treeline, dense vegetation further constrained emergence. Both spruce and larch germinated better on plots where natural vegetation had been removed. This shows that competition by dwarf shrubs and other ground covering vegetation reduced emergence. At the uppermost site, however, the vegetation cover did not constrain tree seedling emergence, suggesting that the scarce alpine vegetation did not negatively affect seedling emergence above treeline. Other factors such as seed source and herbivory also influenced early recruitment success.

Despite successful emergence, our findings suggest that longer-term tree seedling survival above treeline requires favourable climate conditions. After two years, the proportion of surviving spruce and larch seedlings was considerably smaller above treeline than at treeline, suggesting that harsh winter conditions are an important bottleneck for early seedling recruitment.


We will combine the findings of our experimental site with those from the other G-TREE sites. The synthesis of the findings will allow us to better understand the importance of different abiotic and biotic drivers of tree seedling recruitment at alpine and Arctic treelines, thus leading to a better understanding of treeline shifts worldwide.

Furthermore, we will continue to assess the survival and growth of seedlings at our experimental sites at Stillberg in the coming years. So far, continued observations indicate that early seedling recruitment is affected differently by environmental factors than post-emergence seedling establishment. Ultimately, treelines will only expand if emerged seedlings are able to survive in the longer term.