Treelines in the Swiss Alps: Growth Dynamics and Forest Succession in a Changing Climate
The alpine treeline is one of the most prominent ecosystem boundaries in nature. The two ecosystems that are separated by this ecotone – the subalpine forest and the alpine tundra – provide essential ecosystem goods and services (EGS) in densely populated mountain regions like the European Alps. Ong...
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Format: | Doctoral or Postdoctoral Thesis |
| Language: | English |
| Published: |
ETH Zurich
2017
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/20.500.11850/264159 https://doi.org/10.3929/ethz-b-000264159 |
| Summary: | The alpine treeline is one of the most prominent ecosystem boundaries in nature. The two ecosystems that are separated by this ecotone – the subalpine forest and the alpine tundra – provide essential ecosystem goods and services (EGS) in densely populated mountain regions like the European Alps. Ongoing global climate change with its temperature increase affects the major limiting factor of tree growth close to treeline. Moreover, mountain regions have experienced a more pronounced temperature increase than the global average, and this is expected to continue into the future, with far-reaching consequences for forest dynamics at treeline and therefore also for the EGS provided by both ecosystem types. Although the alpine treeline has attracted scientific interest for centuries, understanding the interplay between factors influencing treeline formation and subalpine forest dynamics, especially at the regional to local scale, remains a challenge. To reliably quantify the consequences of global warming for subalpine forests approaching treeline, an in-depth understanding of the large-scale drivers is required, but also of the regional- to local-scale modulators of tree population dynamics. The aim of this PhD thesis is to enhance this understanding by means of empirical analyses of the past and simulations of future treeline dynamics. In particular, I wanted to (i) identify the climatic drivers of short-term tree growth dynamics, reveal the underlying functional form of the growthtemperature relationship, and elucidate the role of inter- and intra-species variability in modulating the growth response of trees; (ii) scrutinize the relative importance of air and soil temperature variables for determining individual tree growth; (iii) assess treeline dynamics at larger spatial and temporal scales, by exploring the causes of population-wide long-term growth trends, and by isolating important regional- to local-scale drivers of tree population dynamics; and (iv) synoptically validate all these findings regarding their importance as drivers of treeline dynamics, which then allows to robustly explore the development of forest dynamics at and directly below treeline under changing climatic conditions. All analyses of my PhD thesis use the three study sites Bosco/Gurin, Hohgant and Zermatt. They experience three distinct climate regimes of the Swiss Alps (Southern, Northern and Central Alps) and are representative for a variety of prevailing growth conditions. In Part I, I analyzed the effects of past climate variability on short- to long-term tree growth dynamics, with a focus on among-tree variability and potential climate change feedback effects. Even though previous research had suggested among-tree variability to be of large significance, it has not yet been systematically considered in studies of tree growth dynamics. Thus, I conducted a dendroecological study of four major tree species (Picea abies, Larix decidua, Pinus cembra, Pinus mugo) and employed linear mixed-effects models (LMMs) to quantify climate influences on short- and long-term growth dynamics. LMMs simultaneously capture among-tree variability and feedback effects due to inter-species competition. These results were complemented with stand-level analyses of size and age structure using a combination of remotely sensed and field-sampled data. The LMMs revealed large amounts of previously unquantified among-tree variability regarding when and to what extent individual trees invest assimilates into growth. Furthermore, the LMMs indicated strongly positive temperature effects on growth during short summer periods across all species, and significant contributions of fall conditions (L. decidua). Furthermore, the influence of snow was implicitly indicated by current year’s spring effects (L. decidua, P. abies). In the longer term, all species showed positive growth trends at the highest elevations, but indifferent patterns with just slightly decreasing elevation. Larix decidua exhibited even negative growth trends just below treeline, whereas P. abies, P. cembra, and P. mugo showed a range of positive trends with decreasing elevation. Clearly, the population-wide positive response at the highest elevations reflected the effects of ameliorated climatic conditions over time, but at the same time it revealed the prominent role of among-tree variability in controlling the response of treeline growth dynamics to climate change. Due to increased competition, first signs of long-suspected negative feedback effects of climate change on stand dynamics at treeline emerged. In Part II, I further scrutinized the role of temperature as the most important determinant of tree growth as the cold edge of a species’ distribution is approached. Both radial and height growth of trees are determined by several interlinked processes, which exhibit different reactions to temperature. I unraveled the relative importance of (i) soil vs. air temperature, (ii) absolute temperatures vs. time-integrated temperatures, and (iii) reserves from the previous growing season for annual radial tree growth, and I attempted to capture the form of the functional relationship between temperature and growth. To this end, I measured long-term, high-resolution (both spatially and temporally) air and soil temperature along elevational transects at each study site and paired them with basal area increment (BAI) data. In addition, I covered the temporal component, i.e. past and present climate variability. Hence, I was able to quantify the growth reactions to temporal as opposed to spatial variability. Close to treeline, BAI of all species depended on integrated rather than absolute temperatures. While BAI was best explained by combining air temperature of the previous and current growing seasons, soil temperature was important only at Bosco/Gurin, a site with extensive snow cover in late winter/spring. When moving down from upper treeline, the temperature-growth relationship was strongly non-linear, showing a rapid decrease of temperature limitation and an increasing importance of other growth-determining factors. For the last 50 years, temperatures have been rising substantially at all sites, with isotherms moving upward 160–260m in elevation. The threshold-dependence of growth to temperature has led to a positive trend of BAI over time, which consequently was consistent throughout the population only at highest elevations. In Part III, I employed the dynamic forest succession model ForClim to assess the ecological relevance of different abiotic and biotic factors driving subalpine forest dynamics, which were partly derived from the first two parts. To this end, I evaluated different model variants based on their capability to reproduce observed individual-tree and stand-scale properties derived from (i) my dendroecological data, (ii) forest size structure as remotely sensed and measured in the field, and (iii) in-situ temperature data from the study sites. Subsequently, using the best model and climate change scenarios, I projected the future development of treeline elevation, growth dynamics of the subalpine forest and its species composition. This mechanistic approach revealed a pivotal role of the precipitation regime in modulating growth conditions at treeline. Precipitation mainly controlled incoming radiation and, via the snow melt date, determined the start of the growing season. Projections of subalpine forest dynamics under future climate suggested substantial changes compared to present-day conditions in all three Alpine regions that I studied. Only after a lag of ca. 100 years, the subalpine forest moved uphill by several 100m even under the low-end climate change scenario. Simultaneously, a strong immigration of montane forest species at the lower end of the present-day subalpine forest became apparent. Overall, I showed that the reaction of treeline forest dynamics to past and future climates is strongly influenced by temperature variability. At the same time, it is significantly modulated by other abiotic forcings such as the precipitation regime and biotic drivers like the substantial among-tree growth variability or competition effects. These drivers act on different spatial scales, from climatic regions down to the individual tree and tree-specific structures such as roots vs. needles. Consequently, it is a misconception to assume that increased temperatures due to climate change generally ameliorate growth conditions at treeline. The highly non-linear relationship between radial tree growth and temperature leads to a pronounced increase in growth variability, just after the temperature limitation is overcome. This behavior is due to other growth-limiting factors setting in, but also partly founded in the previously unquantified among-tree variability. Declining growth trends with decreasing elevation may indicate negative feedback effects of climate change on tree growth due to increased stem density. In addition, the precipitation regime significantly modulates growth conditions at treeline, first by controlling cloud cover, air humidity and therefore the direct radiation energy available to the trees, and second through the amount of snow, thus restricting the period during which trees can actually grow. In summary, the growth of trees at treeline reacts strongly to climate variability but is also decisively modulated by regional- to local-scale drivers. This can go as far as reversing the per se positive implications of increasing temperatures for tree growth close to its cold limit. Consequently, considering the above-mentioned population-internal and -external processes on smaller spatial scales allows for a significantly more reliable assessment of climate impacts on tree growth. |
|---|