Past, present and future climate variability and extremes: The impact on forests.
Summary of the Wengen-97 International Workshop on Global Change Research

At a scientific workshop in Wengen held from 22-25 September, scientists from 13 different countries examined the problems that changes in climate pose for forests. Particular emphasis was given to changes in extreme events, such as storms, droughts and changes in rainfall patterns.

The workshop identified a number of outstanding problems. Many resource managers are being asked to ensure that changes to forests lie within the "range of natural variability". This presents major problems, as there is often insufficient information to indicate what this range is. The difficulties were explained by Prof. Malcolm Hughes (University of Arizona, USA), who used 8000 year tree-ring series to show that the frequency of droughts in California has changed over the last 2000 years.

Although the climatic changes that have been forecasted will be rapid, even greater changes have occurred in the past. Prof. Brigitte Ammann (University of Bern, Switzerland) has studied the effects of climatic changes at the end of the last glacial period. She has used pollen analysis to show that the responses of animals and vegetation to such changes may be very rapid if no migration is involved.

Predicting the effects of climate change on forests may be very difficult, particularly in mountain regions. This is because each tree may respond in a slightly different fashion, depending on a range of different factors. For example, at high-altitude sites, temperature generally limits growth. In southern France, Dr. Lucien Tessier and Dr. Christelle Belingard (University of Marseille, France) showed how moisture limitations may determine how fast trees grow as well as the signal of direct anthropogenic interference on recolonization. As a result, the same change in climate may have different effects in different forests, a point also made by Prof. Brian Luckman (University of Western Ontario, Canada)

New methods for analysing the responses of forests to climate are being developed. Isotope analysis offers considerable potential, according to Dr. Matthias Saurer (Paul Scherrer Institute, Switzerland), as well as the monitoring of atmospheric pollutant species at high alpine sites. These are generally far removed from local perturbations (urban effects, etc.) and therefore provide useful information on long-distance pollution which can affect tree growth and mortality (Margit Schwikowski, Paul-Scherrer Institute, Switzerland).

Drought plays an important role in forest dynamics. It has been responsible for pulses of tree mortality in the Argentinian Andes, according to Dr. Riccardo Villalba (Mendoza, Argentina). In the Canadian Rockies, Prof. Brian Luckman identified periods of extreme cold as a cause of tree death. Conversely, in the Czech Republic, wind, snow and frost are the main causes of forest damage (Prof. Rudolf Brazdil, Masaryk University, Czech Republic).

A much better understanding of climate variability in Europe is developing as a result of recent progress in the recognition of the North Atlantic Oscillation. During positive phases of this Oscillation, winds bring greater quantities of moist winds to Europe, according to Dr James Hurrell (NCAR, USA). Lack of snow in the Alps in the late 1980s and early 1990s, and its late arrival during the 1995/1996 season can be to a large degree attributed to the high positive values of the North Atlantic Oscillation Index during these periods.

In Switzerland, climate change is already evident in the form of temperature increases that are well above the global average (Martine Rebetez, WSL, Switzerland). The greatest increases have been in the minimum temperatures, with maximum temperatures being less affected. In particular, winter minimum temperatures have tended to increase. There has also been an increase in extreme precipitation events, particularly during autumn, although this is not reflected in average rainfall statistics for the period. Increases in the intensity of autumn rainfall events are likely to continue, and both the frequency and intensity of winter precipitation is likely to increase (Christoph Frei, ETHZ, Switzerland).

The responses of forests to these changes are very uncertain. A variety of different climatically-induced stresses are important for forests (John Innes, WSL, Switzerland), and these are likely to have an impact on future changes in forest composition.. Whether or not this will be reflected in changes in the tree-line remains unclear as it is often difficult to determine the most important factors affecting the upper growth limit of trees (Christian Körner, University of Basel, Switzerland). Temperature may be important at some locations but at others, soil moisture may be critical (Tommaso Anfodillo, University of Padua, Italy). Alternatively, the length of snowcover may be the limiting factor. Changes in the tree-line may be particularly important if these occur at the expense of alpine meadows, and there is much evidence to suggest that such changes are occurring (David Peterson, University of Washington, USA).

Trees themselves may help to provide information on extreme events. For example, severe frosts are sometimes recorded as frost cracks in tree rings (Paulo Cherubini, WSL, Switzerland). However, when trees are lost to such extremes, the forest industry may be in a poor position to react, as demonstrated by events following the Vivian storm, which hit Western and Central Europe in february 1990 (Jean Combe, WSL, Switzerland).

Much of the information that we have today about future responses of forests to climate change has been derived by modelling. Recent developments in modelling have demonstrated the importance of extreme in events in determining forest development (Harald Bugmann, Potsdam Institute for Climate Impact Research, Germany). Considerable progress has also been made in refining early models that dealt only with mean monthly temperatures. Including for example soil moisture based on daily values results in radically different results (Petra Lasch, Potsdam Institute for Climate Impact Research, Germany). Similarly models including moisture values based on precipitation and subsequent water storage also produce different results (Norbert Kräuchi, WSL, Switzerland). Models are also being developed that enable the effects bark beetles to be included (Manfred Lexer, University of Agricultural Sciences, Austria).

Although many changes in forests may occur, these are most likely to occur near the upper tree-line. Other forests may be fairly resistant to change, such as the Norway spruce plantations of the Swiss Plateau (Felix Kienast, WSL, Birmensdorf). Such changes need to be looked at on a landscape scale, as the interactions between patches may be particularly important (Martin Sykes, Lund University, Sweden).

The Workshop was organised jointly by Professor Martin Beniston of the University of Fribourg and Dr. John Innes of the Swiss Federal Institute for Forest, Snow and Landscape Research (WSL). Financial support for the Workshop was provided by BUWAL, the Swiss National Science Foundation, University of Fribourg (Institute of Geography), and by the WSL.


The most important points to come out of this meeting were:

1. Tree ring records reveal that mean temperatures are important for tree growth, but that tree rings contain much more information. This information reveals the importance of extreme events in determining the growth of individual trees.

2. The factors affecting tree growth vary between species and between sites. Consequently, a knowledge of the ecology of a particular species is needed before predictions are possible about how it will respond to climate change. Site factors need to be taken into account, as they will influence how a tree responds to a specific climatic change.

3. The main climatic events that are likely to impact on forests are wind storms, droughts, frosts, and the duration of snow-cover. These can have both direct (e.g. freezing injury) and indirect (e.g. by influencing bark beetle populations) effects.

4. The changes in climate which are projected by General Circulation Models on the basis of scenarios of the Intergovernmental Panel on Climate Change are not unique: similar (but not identical) very rapid changes have occurred in the past. Forests responded to these changes differently, making prediction about the impacts of the current changes difficult.

5. The understanding of climatic variability is becoming increasingly clear. The Southern Oscillation in the Pacific Ocean has been recognised for some time, but the implications of the presence of a North Atlantic Oscillation are now being recognised.

6. Analysis of climatological records from Switzerland and elsewhere indicate changes in the patterns of extreme events. These include less extreme cold periods in winter and more intensive rainfall events in autumn.

7. The impacts of these changes on forests are currently being evaluated. Changes in temperature may affect the winter chilling requirements of some species. Changes in precipitation patterns will affect soil moisture conditions, with major implications for forest development.

8. Models of forest development are increasingly incorporating information on extreme conditions. A new generation of forest models is using information on daily values for, for example, soil moisture availability. The results generated by these models are very different from those derived from older, simpler models.

9. Simple models of changes in the distribution of trees based on mean temperatures are now seen as being too simplistic. Changes in the distribution of forests will be affected by the availability of suitable sites for tree growth. For example, the upward movement of the treeline may be restricted by the absence of soils suitable for tree growth.

For more information, please contact either:

Prof. Martin Beniston Institute of Geography University of Fribourg Perolles CH ­ 1700 Fribourg Switzerland E-Mail: Martin.Beniston@Unifr.CH
Dr. John Innes Swiss Federal Institute of Forest, Snow and Landscape Research Zürcherstrasse 111 CH ­ 8903 Birmensdorf Switzerland E-Mail: john.innes@wsl.ch

    Home | General Information |Teaching | Research | Publications | Events | Links 

    Franziska Keller, last update, 05.01