Réf. Stoffel & al 2005b - A

Référence bibliographique complète
STOFFEL M., SCHNEUWLY D., BOLLSCHWEILER M., LIÈVRE I., DELALOYE R., MYINT M., MONBARON M. Analyzing rockfall activity (1600–2002) in a protection forest - a case study using dendrogeomorphology. Geomorphology, 2005, vol. 68, 224–241.

Abstract: Throughout the last four centuries, rockfall has caused growth disturbances GD to the trees sampled on the Täschgufer slope, most frequently in the form of low magnitude–high frequency events. In addition, the auhors identified one high magnitude–low frequency event in 1720. To analyze past rockfall activity, they introduce a "rate" defined as the number of impacts per meter width of all tree surfaces sampled per decade. Results clearly demonstrate that this rockfall "rate" continually decreased in both sectors after the large 1720 rockfall event. Significantly low rockfall "rates" can be observed during the 1850s, 1960s and 1970s in the northern and during the 1820s in the southern sector. In contrast, high rockfall "rates" were identified during the 1870s and 1990s in the northern, and during the 1770s in the southern sector. Reconstructed data further show that the forest recolonizing the southern sector after the 1720 event gradually improved its protective function, reducing "rates" by a factor of 13 between the 1740s and the 1990s. In the well-established forest of the northern sector, the efficacy of the protective forest was temporarily reduced by the rockfalls in 1720, resulting in increased rockfall "rates". Since then, the protective function of the forest stand has increased again.

Dendrogeomorphology, rockfall, growth disturbances, frequency, magnitude, protection forest, Swiss Alps

Organismes / Contact

Groupe de Recherches en Géomorphologie (GReG), Department of Geosciences, Geography, chemin du Musée4, University of Fribourg, 1700 Fribourg, Switzerland
Remote Sensing and GIS Unit, Department of Geosciences, Geography, University of Fribourg, 1700 Fribourg, 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
    Mass movements Rock falls

Pays / Zone
Massif / Secteur
Site(s) d'étude
Période(s) d'observation
Swiss Alps Valais Täschgufer slope West ~1500-3200m 1600–2002

(1) - Modifications des paramètres atmosphériques

Informations complémentaires (données utilisées, méthode, scénarios, etc.)

(2) - Effets du changement climatique sur le milieu naturel

Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)


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

Reconstructed rockfall activity varied greatly across the slope. The spatial distribution of interpolated recurrence intervals show that GD abundantly occurred in the trees located above 1700 m a.s.l., where surfaces gradually become more sparsely forested. Not surprisingly, the spatial pattern of recurrence intervals largely coincides with the distribution of the age structure and oldest trees are commonly found in areas with relatively low numbers of GD. In contrast, youngest trees are largely concentrated in areas where rockfall repeatedly caused GD.

Throughout the last four centuries, rockfall fragments have continuously caused GD to the trees sampled for analysis. There seems to have been no period since AD 1600 without rockfall, and activity most commonly consisted of low magnitude–high frequency events. In addition, tree-ring and age structure analyses also allowed identification of one high magnitude–low frequency event, which (almost) completely destroyed the forest stand in the southern sector of the slope in 1720. As an indirect consequence of the high-magnitude event in 1720, trees abundantly (re)colonized the slope.

Decadal variations in rockfall activity:

Northern sector (1600–1999):
The number of GD was reduced to 400 events derived from 78 trees. Before 1720, rockfall "rates" indicate considerable fluctuations in reconstructed numbers of GD. Since, data indicate that rockfall activity continually decreased after the 1720 event, interrupted by decades with significantly low (1850s, 1960s, 1970s) and significantly high rockfall "rates" (1870s, 1990s). During the 19th century, rockfall "rates" indicate rather low values during the 1850s and, to a minor extent, the 1810s. Thereafter, rockfall caused higher "rates" during the late 19th and the early 20th century. Within the second half of the 20th century, rockfall "rates" indicate a period with low activity lasting from the 1950s until the 1970s. This period was followed by an increase in the rockfall "rate" during the 1980s and, even more, the 1990s. The importance of this increase as well as the influence of dam construction works on the decadal ratios of the 1980s and 1990s will be analyzed below.

Southern sector (1740–1999):
In total, 341 GD were dated in the 57 trees sampled. Similarly to the northern sector, reconstructed rockfall "rates" continually decreased in the southern sector after the high magnitude rockfalls of 1720. After a slight decrease in the 1750s and 1760s, the "rate" indicates a considerable increase in rockfall activity in the 1770s, lasted until the 1780s or even 1790s. In contrast to the high magnitude rockfalls of 1720s, the considerable number of GD identified for the last three decades of the 18th century were not caused by one high-magnitude event, but rather by a series of years with high rockfall activity. Between the 1870s and the 1940s, comparably low numbers of GD were found in the trees sampled, resulting in a considerable decrease in rockfall "rates". In contrast to the northern sector, the rockfall "rate" started to increase again in the 1940s. Again, the influence of the dam construction works on the decadal ratios of the 1980s and 1990s will be analyzed further down.

Yearly fluctuations in rockfall activity (1950–2002):
In total, the authors identified 172 GD, 105 in the southern (61%) and 67 in the northern sector (39%).
The number of reconstructed GD remained on a comparably low level in the northern sector during the first three decades of investigation. After 1980, rockfall repeatedly caused more GD to the trees sampled. Reconstructed data neatly reflect the period of increased rockfall activity in the 1980s and the early 1990s.
In the southern sector, yearly resolved "rates" indicate that during the first three decades, (short) periods with increased numbers of GD (e.g., 1959–1964) were followed by several years with (almost) no GD. Similar to the northern sector, considerably increased rockfall "rates" can be discerned in 1980 and 1992.

The construction of Dams between 1988-89 only slightly influenced the rockfall "rates". In contrast, reconstructed rockfall "rates" prove to be influenced by the construction works of the two protection dams in the late 1990s. The rockfall "rates" of the 1990s were clearly influenced by anthropogenic intervention on the slope, making it difficult to estimate the undisturbed rockfall frequency over this period. As a further consequence, anthropogenic intervention on the slope hindered rockfall fragments to cause GD to the trees sampled after 1998. The recently built dams efficiently stopped rockfall in the recent past and so further influenced the rockfall "rates" of the 1990s.

According to Lauber (1995), rockfall activity at Täschgufer apparently increased in the 1980s and again after 1993. These observations have recently been confirmed with dendrochronological analysis (Stoffel et al., in press), which further indicated that almost 90% of the intra-annual rockfall activity at Täschgufer occurs in April and May.




Paramètres de l'aléa
Sensibilité du paramètre de l'aléa à des paramètres climatiques et du milieu
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
Rock falls activity

The area investigated was the west-facing Täschgufer slope. Rockfall frequently occurs on the slope, originating from the heavily disintegrated gneissic rockwalls below the Leiterspitzen summit. The main rockfall source areas on the slope are located between 2300 and 2600 m a.s.l. (Rockfall Source Area 2) and above 2700 m a.s.l. (Rockfall Source Area 1), where bedrock is highly fractured with many joints.

The authors sampled severely affected trees with obvious signs of growth disturbances (GD) from both the rockslide deposits (northern sector) and the southern sector of the slope. In this investigation, at least 4 cores were extracted per tree using increment borers. In addition, undisturbed reference trees have been sampled. In total, 135 L. decidua Mill. trees (564 cores) from the rockfall slope and 17 trees (34 cores) from the undisturbed reference site were sampled in 2002. The analysis of the 564 cores allowed identification of 786 GD attributed to rockfall activity on the slope. In some cases, impacts caused GD in more than one core of the same tree, reducing the number of different rockfall events to 761.

Samples were analyzed and data processed following the standard procedures described in Bräker (2002). Growth curves of the disturbed samples were then crossdated with the reference chronology (1596–2002) constructed from 17 undisturbed trees. This procedure allowed differentiation of climatically driven fluctuations in tree growth within the study area from GD caused by rockfall activity.

The authors therefore use a rockfall "rate" expressed as number of rockfalls (GD) per meter width of all tree surfaces sampled per decade instead of analyzing absolute values. In a further step, the rockfall "rates" were converted from real into logarithmic values, a power regression trend line calculated and residuals from the regression model determined. Finally, the authors used yearly rockfall "rates" (real values) to analyze short-term fluctuations in rockfall activity for the period 1950–2002. This use of yearly resolved data on GD also allowed estimation of the influence of anthropogenic rockfall triggering during the periods of dam construction work in 1988-89 and 1996-1998 as well as the distribution of GD thereafter.

Data were investigated with the ArcGIS Geostatistical Analyst software in order to examine spatial relationships between all sample points. Within this study, interpolations were performed to visualize the age structure of the forest stand and the spatial distribution of recurrence intervals.

(4) - Remarques générales


(5) - Syntèses et préconisations

The approach outlined in this study proved to be a useful tool for analyzing fluctuations in inter-annual and inter-decadal rockfall activity on a forested slope. Although the techniques used in this investigation need further refinement and minor modifications in the sampling strategy, the results presented above clearly show that dendrogeomorphic investigations have the potential to produce results on yearly fluctuations and decadal ratios of rockfall activity over several centuries. Moreover, the authors have been able to determine spatial variations in rockfall activity, considerable differences in recurrence intervals and the changes in the efficacy of the protective function of the two forest stands at Täschgufer.

Références citées :

Berger, F., Quetel, C., Dorren, L.K.A., 2002. Forest: a natural protection mean against rockfall, but with which efficiency? The objectives and methodology of the ROCKFOR project. Proc. Int. Congress Interpraevent 2002 in the Pacific Rim, Matsumoto, Japan, pp. 815 – 826.

Braam, R.R., Weiss, E.E.J, Burrough, P.A., 1987. Spatial and temporal analysis of mass movement using dendrochronology. Catena 14, 573 – 584.

Shroder, J.F., 1980. Dendrogeomorphology: review and new techniques of tree-ring dating. Prog. Phys. Geogr. 4, 161 – 188.

Solomina, O., 2002. Dendrogeomorphology: research requirements. Dendrochronologia 20, 233 – 245.

Stoffel, M., Lièvre, I., Conus, D., Grichting, M., Raetzo, H., Gärtner, H., Monbaron, M., 2005a. 400 years of debris flow activity and triggering weather conditions: Ritigraben VS, Switzerland. Arct. Antarct. Alp. Res. 37 (1) in press. [Fiche Biblio]

Stoffel, M., Lièvre, I., Monbaron, M., Perret, S. Seasonal timing of rockfall activity on a forested slope at Täschgufer (Swiss Alps)—a dendrochronological approach. Z. Geomorphol. 49, in press.