Réf. Keller & al. 2000 - A

Référence bibliographique complète
KELLER F., KIENAST F., BENISTON M. Evidence of response of vegetation to climatic change on high elevation sites in the Swiss Alps. Regional and Environmental Change, 2000, 1, p. 70-77.

Abstract: Climate change has in the past led to shifts in vegetation patterns; in a future, warmer climate due to enhanced greenhouse-gas concentrations, vegetation is also likely to be highly responsive to such warming. Mountain regions are considered to be particularly sensitive to such changes. In this paper the authors present an approach to assess the impact of climate change on long-term vegetation plots at the high elevation site of the Schynige Patte, 2000 m above sea level, in the Bernese Alps (Switzerland). Records of vegetation spanning the period 1928 up till today at two different sites, each with several plots, were considered. The observed change in the species composition was then related to changes in land use and climate. Daily values of temperature, snow and precipitation from several high elevation weather stations were used to conduct these analyses. The correlation between climate and vegetation patterns revealed that species which prefer low thermal conditions move out of the plots, i.e., their frequency of occurrence is negatively correlated with the average number of degree-days over the last six decades. On the other hand, species with higher thermal demands are seen to be invading the plots, i.e., their frequency of occurrence is positively correlated to the average number of degree-days. Nutrient changes - though independent from climate - also seem to play an important role for the observed shifts in species.

Mountain vegetation, climate change, Schynige Platte, Switzerland.

Organismes / Contact
Department of Geography, University of Fribourg, Pérolles, 1700 Fribourg, Switzerland. Tel: +41 26 300 90 26 ; franziska.keller@unifr.ch.
Swiss Federal Institute for Forest, Snow and Landscape Research (WSL), Birmensdorf, Switzerland.

(1) - Paramètre(s) atmosphérique(s) modifié(s)
(2) - Elément(s) du milieu impacté(s)
(3) - Type(s) d'aléa impacté(s)
(3) - Sous-type(s) d'aléa
Temperature, Precipitation Vegetation, Snow cover    

Pays / Zone
Massif / Secteur
Site(s) d'étude
Période(s) d'observation
Switzerland Bernese Alps Schynige Platte (vegetation)
Säntis (climate)
  2000 m (Schynige Platte)
2500 m (Säntis)
Since 1930s (vegetation)
20th century (climate)

(1) - Modifications des paramètres atmosphériques
At the Säntis climatological observing site, mean monthly temperatures have risen from about -2.5°C at the beginning of the 20th Century to around -0.5°C in the 1990s. The increase has been particularly remarkable since the early 1980s; at the beginning of the 1990s, the highest average temperatures of the record were reached. High values were also measured at the end of the 1940s and at the beginning of the 1950s. Only the minimum temperatures show an increasing trend of about 2°C since the beginning of the century, while the temperature maxima have remained at a more or less constant level.

The linear trend of the degree-days exhibits a systematic increase. The mean temperatures of the degree-days were high in the 1930s, although the curve of the degree-days shows a decreasing tendency. Temperatures for days above 3°C were quite high, while the contrary seems to be the case at the end of the 1970s and from the 1980s until today, where the average temperatures of the days above 3°C are lower but the number of degree-days are high. Days with temperatures just above 3°C are more frequent but extreme maximum temperatures occur less often. The trend towards a warmer climate is significant.

Precipitation records of the weather stations in the Bernese Alps do not show any clear trend and every climatological site has its own particular evolution. Snow depth and duration have decreased since the beginning of the century. Years without any snow or only small snow depth have not yet been recorded at elevations as high as 2000 m above sea level, although this has been observed at low to medium elevations (< 1,200 m) in the late 1980s and early 1990s (Beniston, 1997).

Informations complémentaires (données utilisées, méthode, scénarios, etc.)
Meteorological data from the Säntis were used (eastern Switzerland at 2,500 m asl), with daily records from 1901 up till today. Although the Säntis is located some distance away from the Schynige Platte, it is nevertheless representative of temperature and precipitation in the Alps. Exhaustive analyses were undertaken to ensure that the synchronism of annual and internannual fluctuations observed at Säntis correspond to those observed in the available parts of the records taken from the vicinity of the Schynige Platte. The correlation was excellent.

To generate a more physiologically-meaningful surrogate for thermal energy input, the number of days with mean temperatures exceeding 3°C was calculated as well as the average temperature of these days. An adiabatic temperature correction of 6.5°C/km was applied to account for the altitudinal difference between Säntis and the Schynige Platte.

(2) - Effets du changement climatique sur le milieu naturel
The ordination results of the Alpengarten and Lüdiwiese sites clearly show a shifting species composition between the 1930s and the 1990s. For the Alpengarten site, it is seen that the 6 vegetation plots have changed considerably their species composition over time. The difference in species composition seems to be greater for the first records than for the more recent ones. The Lüdiwiese site also exhibits a clear shift in species composition since 1934.

For the Alpengarten site, analysis of the probability of occurrence of warmth-demanding species over time shows an increase over the last sixty years. On the other hand, probabilities of occurrence of species that tolerate low thermal energy impact decrease over the observed period. The probabilities of the indicator values tested with the Kruskal-Wallis test do not always prove significant. Analyses for the continentality indicator value support the fact that the probability of occurrence of the species that prefer lower temperature differences increases over time, although the trend is not always significant. Looking at the nutrient indicator values, the probabilities show a constant increase up to the 1970s, with most species prefering a medium-rich soil. Subsequently, a strong change occurs and the species growing on nutrient-rich soils exhibit a higher occurrence. The reason for this sudden change is not clear, however.

At the Lüdiwiese site, essentially identical conclusions can be drawn to those for the Alpengarten site. The probability of finding a species that prefers warmer climatic conditions has doubled since the 1930s, and a tendency for a shift from alpine to subalpine species can be observed. Concerning continentality the study also indicates an increase of species that require milder temperatures with smaller temperature amplitudes.

The observed shift in species composition is as much the result of modified temperature conditions as to changed nutrient availability. As a result, two independent signals of two independent impacts can be detected. The change in species composition of the experimental plots is most probably due to an influence of changed nutrient amounts and soil acidity.

Degree-days and probability of plants with high or low thermal requirements, respectively, show clear dependencies. A negative correlation between the probability of occurrence of species with a lower thermal demand, and the average number of the degree-days of the seven years prior to the vegetation surveys was observed. On the contrary, a positive correlation between the probability of occurrence of species with a higher thermic demand and the respective of degree-days can be infered.

There is thus a clear relationship between the decreasing probability of occurrence of species preferring cooler conditions on the Schynige Platte, the increasing frequency of species which respond positively to warmer average temperatures and the number of degree-days, and hence the evolution of temperatures this century. These processes can be considered to be a directed change in species composition due to changes in environmental conditions; these changes affect primarily species with either low or high thermal requirements.

From the analyses of the evolution of temperature, it can be concluded that on the Schynige Platte, while average temperatures are slightly higher (essentially due to an increase in minimum temperatures), the amplitude of daily temperature range has become smaller since the 1930s (Rebetez and Beniston 1998) and therefore the trend is towards a milder climatic regime. This is indirectly reflected in the changed species composition observed on the Schynige Platte vegetation plots.

Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
Temperature (number of degree-days and amplitude of daily temperature)
The present study is restricted to the lowermost level of the functional hierarchy of ecological systems, i.e., individual plants with a short life-span rather than long-lived plant-communities or biomes. This prerequisite follows the reasoning that climatic change will first be visible in the species or subspecies composition or in minor changes of the functional pathways. Only greater levels of climatic change would lead to detectable signals in the higher hierarchies of ecological systems.

The study area is the Schynige Platte at 2000 m a.s.l., in the Bernese Alps (Switzerland), which has a long vegetation history that reaches back to the 1930s. The assessment was performed on experimental plots which have not been treated with any kind of fertiliser. The plots are grouped into the area of the Alpengarten with the phytosociological association of the Seslerio-Caricetum and the Lüdiwiese with a Geo montani-Nardetum .

Multivariate analysis of the different time steps was used to detect possible changes in the species composition. The MULVA-5 program (Wildi and Orlòci 1996), specifically designed to analyse composition properties of phytosociological relevés as a function of site, climate and other physical properties (Wildi 1993) has been used here. The species scores were transformed to presence-absence and the relevés were arranged according to the year of sampling.

To analyse the resemblence pattern of the relevés, a resemblence matrix was calculated using van der Maarel's similarity ratio (van der Maarel 1979). The cover-abundance scale of Braun-Blanquet (Braun-Blanquet 1964) had to be transformed to quantitative rank scores according to the suggestion of van der Maarel (1979). In a last step a principal coordinate analysis (PCOA) was calculated. It is a simple numerical technique for multidimensional scaling based on principal component analysis (PCA) methods, but more general.

The indicator value concept (Landolt 1977) was employed to check whether a postulated change in composition is either undirected or directed. The concept of the indicator values defines ecological profiles. For this study, the indicator value for temperature, continentality, nutrients and pH-reaction were chosen for every species

A probability of presence was calculated for each species and site summarizing all the plots of one site. Consequently, a species that was present in 1933 on all plots of the Lüdiwiese is assigned a value of 100%, while if it is present on only 30% of all plots in 1997, the probability of presence is 30%.

For both sites, groups were formed for all the species having identical indicator values. The average probability of presence was calculated such that 100% was reached if all species of a group with the same indicator value were present in every record on all the plots at one site.

The average number of degree-days of the seven years prior to every vegetation record (i. e. 1929, 1933, 1939, 1943, 1954, 1972, 1990, 1997) was calculated and compacted with the probabilities of occurrence of any species group formed on the basis of temperature traits (indicator values).

(3) - Effets du changement climatique sur l'aléa

Paramètre de l'aléa
Sensibilité du paramètres de l'aléa à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)

(4) - Remarques générales

(5) - Syntèses et préconisations

Références citées :

Beniston M (1997) Variations of snow depth and duration in the Swiss Alps over the last 50 years: links to changes in large-scale climatic forcings. Climatic Change 36: 281 300.

Braun-Blanquet J (1964) Pflanzensoziologie, Grundzüge der Vegetationskunde. 3. Aufl. Springer Berlin, Wien, New York.

Landolt E (1977) Ökologische Zeigerwerte zur Schweizer Flora. Veröff. des geobot. Instituts der ETH, Stiftung Rübel, Zürich, Heft 64.

Rebetez M and Beniston M (1998) Changes in sunshine duration are correlated with changes in daily temperature range this century. An analysis of Swiss climatological data. Geophys. Res. Letters 25: 3611-3613

van der Maarel E (1979) Transformation of cover-abundance values in phytosociology and its effects on community similarity. Vegetatio 39: 97 - 114.

Wildi O, Orlòci L (1996) Numerical exploration of community patterns. A guide to the use of MULVA-5. 2nd Ed., SPB Academic Publishing bv, The Hague, 171p.