Réf. Krautblatter & Moser 2009 - A

Référence bibliographique complète

KRAUTBLATTER, M. MOSER, M. 2009. A nonlinear model coupling rockfall and rainfall intensity based on a four year measurement in a high Alpine rock wall (Reintal, German Alps). Natural Hazards and Earth System Sciences, 9: 1425–1432.

Abstract: A total of more than 140 000 kg of small magnitude rockfall deposits was measured in eight rockfall collectors of altogether 940m² in size between 1999–2003 below a 400–600m high rock face in the Reintal, German Alps. Measurements were conducted with a temporal resolution up to single days to attribute rockfall intensity to observed triggering events. Precipitation was assessed by a rain gauge and high-resolution precipitation radar. Intense rainstorms triggered previously unreported rockfall intensities of up to 300 000 g/(m²h) that [the authors] term “secondary rockfall event.” In comparison to dry periods without frost (10−2g/(m²h)), rockfall deposition increased by 2–218 times during wet freeze-thaw cycles and by 56-thousand to 40-million times during secondary rockfall events. [The authors] obtained three nonlinear logistic growth models that relate rockfall intensity [g/(m²h)] to rainfall intensity [mm/h]. The models account for different rock wall intermediate storage volumes, triggering thresholds and storage depletion. They apply to all rockfall collector positions with correlations from R²=0.89 to 0.99. Thus, the timing of more than 90% of the encountered rockfall is explained by the triggering factor rainfall intensity. A combination of rockfall response models with radar-supported storm cell forecast could be used to anticipate hazardous rockfall events, and help to reduce the exposure of individuals and mobile structures (e.g. cable cars) to the hazard. According to meteorological recordings, the frequency of these intense rockfall events is likely to increase in response to global warming.

Mots-clés
 

Organismes / Contact

• Institute of Geography, University of Bonn, Bonn, Germany: michael.krautblatter@giub.uni-bonn.de
• Department of Applied Geology, University of Erlangen-Nuremberg, Erlangen-Nuremberg, Germany


(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
      Rockfalls

Pays / Zone
Massif / Secteur
Site(s) d'étude
Exposition
Altitude
Période(s) d'observation
Germany Bavarian Alps Reintal North Face (400–600m high) N   1999–2003

(1) - Modifications des paramètres atmosphériques
Reconstitutions
 
Observations

The record of the nearby meteorological station “Hohenpeissenberg” indicates that the moderate warming tendency between 1879 and 1999 coincided with a more than a two-fold increase in the frequency of intense summer rainstorms (DWD, 2001). Further evidence for increasing frequency and magnitude of convective rainstorms has recently been provided by a number of meteorological stations in the Alps and is closely linked to changes in the probability of certain circulation systems that provide air masses rich in water vapour (Fricke and Kaminski, 2002).

Modélisations
 
Hypothèses
 

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

(2) - Effets du changement climatique sur le milieu naturel
Reconstitutions
 
Observations
 
Modélisations
 
Hypothèses
 

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

 


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

The combination of high temporal resolution and direct observation enabled to link rockfall intensity to certain triggering events such as rainstorms or intense freeze-thaw conditions. [...] The increased rockfall response to rainstorms is strictly confined to a time window of less than one hour of surface runoff (see e.g.Wetzel, 1994) in the rock face. [...]

Two types of triggers exert a significant influence on rockfall activity: intense rainstorms and moisture-saturated freeze-thaw conditions, especially when combined with direct solar radiation during the day. For the latter [the authors] measured an increase in rockfall intensity [in g/(m²h)] of 2 to 218 times in comparison to dry conditions without frost. During rainstorms exceeding a threshold of 9–13mm/h, the rockfall intensity reached values of 0.68 kg/(m²h) to 300 kg/(m²h) in all rockfall collectors. The rainstorms which triggered intense rockfall occurred on 1 August 2002 and on 14 June 2003 lasted less than an hour and appeared on precipitation radar images as one or two central two-kilometre pixels with rainfall intensities of 9 to 38mm per 15 min surrounded by squares with 2 to 8mm per 15 min. In comparison to values of rockfall situations without meteorological triggers, secondary rockfall events yielded a 56-thousand to 40-million fold increase in hourly rockfall intensity and thus the deposition exceeded the usual rockfall amount of a whole year without rainstorm by 0.5 to 50 times.

As reported previously, the combination of high moisture supply and deep freezing leads to a moderate increase in rockfall intensity compared to dry, no-frost conditions [...]. In contrast to previous studies, this study related rockfall activity to hourly and 30-min rainfall intensity. The general mode of rockfall response to short-term rainfall intensity appeared to be similar on all nets. Rainfall intensities up to 10–14mm per hour or respectively 9–13mm in 30 min led to no or only a moderate increase in rockfall activity. Rainfall intensities exceeding this range of threshold values triggered secondary rockfall events and thus initiated a sudden increase in rockfall deposition. In response to intense rainstorms a previously undocumented type of rockfall activity were observed, that [the authors] term “secondary rockfall event”. The term secondary refers to the fact that the rockfall material derives from intermediate storage areas such as wedges and couloirs in the rock face. Such events were observed at rainfall intensities that are capable of creating surface runoff and triggered the onset of fluvial processes, hyperconcentrated flows and debris flows on intermediate storage areas in the rock face. [...] The mobilized material moved down the several hundred meters high rock cliff as a more or less freefalling particle cloud (Erismann and Abele, 2001). A secondary rockfall event can thus be defined as a short-term mass deposition of fine-grained rockfall material that originates from intermediate storages in the rock wall and is released by fluvial processes and debris-saturated flows active in the rock face; the short-term intensity of rockfall deposition typically exceeds the deposition during dry periods without frost by a factor of at least 104.

Nonlinear logistic growth functions of rockfall response in relation to hourly rainfall intensity produce significant high correlations from R²=0.93 and 1.00 for all of the eight rockfall nets. The range of correlations in response to 30-min rainfall intensities indicates values between R²=0.93 and 1.00, as well. [...]

Modélisations
 
Hypothèses

If the warming tendency continues or accelerates in future, the intensity and frequency of rainstorms is also likely to increase and will trigger an enhanced activity of secondary rockfall events (Krautblatter and Moser, 2006).


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

A number of direct rockfall measurements have been conducted in the past [see references in the study]. Whereas these studies provided qualitative evidence for a general rainfall – rockfall relation, no quantitative models were developed, which are crucial for the prediction of the specific hazard. Several authors have promoted the idea that the high discrepancy of rock wall retreat rates encountered in various studies is caused by the fact that different rockfall generating processes were assessed. [...] Luckman (1976) distinguished geological factors that initially influence type, spatial distribution and intensity of rockfall activity and climatic factors that finally control rockfall trigger mechanisms. A similar concept was stated by Dorren (2003) who divided up rockfall promoters and factors that cause the actual start of the rockfall movement. [...]

Between 1999–2003 rockfall was continuously measured on twelve robust construction nets installed at eight scree slope positions on the foot of the rock face. As [the present] study aimed to investigate the impact of certain triggers on rockfall activity, we decided to run measurements on large nets with 11 to 234m² to provide a reliable informative basis even for small deposition intensities [g/m²]. The nets were equipped with 60 cm high steel fences to prevent rocks from rolling out of the net area. Rock particles that fell onto the nets were weighed and removed in all 133 measurements.

Between 1 August and 10 October 2002, the deposition of rockfall particles in the collectors was assessed with a high temporal resolution of one day to two weeks. Measurements were continued on a one to two-monthly basis between 23 May and 16 October 2003. As no extreme rainstorms occurred within the study area between 1999 and 2001, this period was taken as a reference period for average rockfall deposition excluding secondary rockfall events. From 1999 to 2001, nets were measured every few months and yielded low rockfall deposition. Dry frost-free conditions were monitored in summer 2002. The combination of high temporal resolution and direct observation enabled to link rockfall intensity to certain triggering events such as rainstorms or intense freeze-thaw conditions. [...] The increased rockfall response to rainstorms is strictly confined to a time window of less than one hour of surface runoff (see e.g.Wetzel, 1994) in the rock face. Therefore, g/(m²h) are the most appropriate unit to quantify and compare rockfall intensity coincident to the duration of triggering conditions. [...] The volumetric importance of intermediate storage was evaluated using air photos and high-resolution photos that were taken horizontally from opposite slope with a telephoto lens.

Precipitation data was obtained from hourly rain gauge measurements at a distance of one to three kilometres from the rockfall nets. Air temperatures were monitored automatically at the same position. The German Seismic Data Analysis Center (SDAC) recorded no earthquakes exceeding a magnitude of ML=4.0 during the examination period which is regarded as the minimum earthquake magnitude for the activation of rock slides and rockfalls according to Keefer (1984, 2002). For a detailed analysis of the rainstorms of 1 August 2002 and 14 June 2003 precipitation radar images [...] with a spatial resolution of two kilometres and a temporal resolution of 15 min were used to calculate average 30 and 60 min intensities for the examination area.

For relating maximum rainfall intensity and rockfall intensity, logistic growth functions were applied [...] Three groups of typical rockfall conditions characterized by the influence of intermediate storage and rockfall exposure were combined to produce three different predictive rockfall response models. [...]


(4) - Remarques générales

The traditional concept of primary rockfall, which explains rockfall as freshly detached particles from the rock face, is incapable of explaining storage effects and the enormous temporal variability of rockfall deposition. Matznetter (1956) introduced the concept of secondary rockfall that explains the time lag between back-weathering and the actual start of the rockfall deposition (Rapp, 1960; Whalley, 1984). He described the fact that rockfall particles, which are totally detached from the rock face by weathering, can still rest for a long time in intermediate storage or in situ in the rock face until they are removed as a secondary rockfall. This explains why rockfall deposition rates at the foot of the rock wall are (in the short term) often inconsistent with rates of back-weathering. Krautblatter and Dikau (2007) tried to reference the type of output of existing rockfall studies in terms of primary and secondary rockfalls. They developed conceptual mathematical models on basis of empiric measurements. These describe rockfall in respect to promoting pre-weathering and weathering conditions as well as internal and external triggers. They postulate that (i) promoting (pre-weathering and weathering) conditions delimit the speed of back-weathering and (ii) triggering conditions determine rockfall supply while (iii) both are linked by the time-dependent intermediate storage in the structured rock face. [...]

[In the present study], nonlinear logistic growth functions of rockfall response in relation to hourly rainfall intensity produce significant high correlations from R²=0.93 and 1.00 for all of the eight rockfall nets. The range of correlations in response to 30-min rainfall intensities indicates values between R²=0.93 and 1.00, as well. [...] [The authors] expect that [their] models will apply well in large alpine rock wall systems with significant intermediate storage that are, therefore, dominated by secondary rockfalls (Krautblatter and Dikau, 2007; Sass and Krautblatter, 2007). Even if certain adjustments of [this] models in respect to hydrology, geology and climatic characteristics can be necessary in other mountain environments, the principles of the initiation of secondary rockfall events by debris-saturated flows within the rock face remain the same.

These findings influence hazard assessment and risk mitigation strategies. 90% of rockfall deposition is confined to a small window of time associated with the activity of rainstorms. Hazardous rates of rockfall deposition that endanger human lives and structures correspond with the spatial and temporal occurrence of high short-term rainfall intensities evident on precipitation radar images. Due to the modern achievements of radar technology and software developments such as RADVOR-OP (Radar Supported Near Real-Time Precipitation Forecast for Operation Use) of the German Weather Service (DWD) or similar products in other countries, spatial movements and intensity changes of storm cells can be forecast over short spans of time (Hering et al., 2004; Kober and Tafferner, 2009). A combination of the logistic rockfall response functions and thresholds presented in this paper with radar supported precipitation forecast systems could be used to predict intensity, spatial and temporal occurrence of hazardous rockfall events early enough to activate a warning system. Due to the fact that precipitation radar systems are already in use in most mountain ranges in developed countries, such a warning system would have a wide spatial range of applicability. [...]


(5) - Syntèses et préconisations

Conclusions:
Small-magnitude rockfalls are difficult to quantify, as their size distribution is not fully covered by laser-scanning and their deposits can often not be referenced accurately to a certain time span of deposition (as is the case for large rockfall deposits). At the same time, small magnitude rockfalls may cause the highest number of (rockfall) casualties in many mountain environments and have an overwhelming importance for sediment budgets.

During four years of rockfall measurements, [the authors] found out that (i) rockfall intensity is only coupled to rainfall intensity above a certain threshold (here 9–13 mm/30 min) and (ii) that the rockfall response to rainfall intensity above the threshold is highly nonlinear. The rockfall deposition during two rainstorms reached previously unreported intensities of up to 300 kg/(m²h). [The authors] termed these events “secondary rockfall events” as their predominantly fine grain size distribution matches exactly the size composition of intermediate storage in the rock face. A secondary rockfall event can be defined as a short-term mass deposition of fine-grained rockfall material that originates from intermediate storage in the rock wall and is released by fluvial processes and debris-saturated flows active in the rock face; the rockfall deposition occurs during minutes in free falling particle clouds. Secondary rockfall events exceeded the usual annual rockfall deposition by 0.5 to 50 times. Historical descriptions show that secondary rockfall events are likely to account for a double digit number of casualties in the Reintal alone in the last century.

[The authors] established three nonlinear logistic growth functions for different topographic settings that model rockfall response to shortterm rainfall intensity with correlations from R²= 0.89 to 0.99. The models account for (i) no rainfall-rockfall coupling at low rainfall intensity, (ii) coupling due to the onset of particle flows in the rock face at a certain rainfall intensity threshold, (iii) secondary rockfall events and (iv) decoupling due to storage depletion at very high rainfall intensities. Nonlinear rock fall response models could be used in combination with radar-supported storm cell forecasts to predict hazardous secondary rockfall events.

Data indicate that intermediate storage is presently depleted 2–4 times faster by the activity of secondary rockfall events than refilled by back-weathering. Nevertheless, the capacity of intermediate storage in the study area appears to be sufficient to allow an elevated level of events for at least decades.

Références citées :

Dorren, L. K. A.: A review of rockfall mechanics and modelling approaches, Progress in Physical Geography, 26, 69–87, 2003.

DWD: Der langj¨ahrige Niederschlagstrend am Hohenpeißenberg: Die Bedeutung von Extremwerten, Global Atmosphere Watch Brief des Deutschen Wetterdienstes, 5, 1–2, 2001.

Erismann, T. H. and Abele, G.: Dynamics of Rockslides and Rockfalls, Springer, Heidelberg, 316 pp., 2001.

Fricke, W. and Kaminski, U.: Ist die Zunahme von Starkniederschlägen auf veränderte Wetterlagen zurückzuführen?, GAW-Brief des DWD, 12, 1–2, 2002.

Hering, A. M., Morel, C., Galli, G., Sénési, P., Ambrosetti, P., and Boscacci, M.: Nowcasting thunderstorms in the Alpine region using a radar based adaptive thresholding scheme, Third European Conference on Radar Meteorology (ERAD), Visby, Sweden, 206–211, 2004.

Keefer, D. K.: Landslides Caused by Earthquakes, Geol. Soc. Am. Bull., 95, 406–421, 1984.

Keefer, D. K.: Investigating Landslides caused by Earthquakes – historical review, Surv. Geophys., 23, 473–510, 2002.

Kober, K. and Tafferner, A.: Tracking and nowcasting of convective cells using remote sensing data from radar and satellite, Meteorol. Z., 18, 75–84, 2009.

Krautblatter, M. and Moser, M.: Will we face an increase in hazardous secondary rockfall events in response to global warming in the foreseeable future?, in: Global Change in Mountain Regions, edited by: Price, M. F., Sapiens Publishing, Duncow, 2006.

Krautblatter, M. and Dikau, R.: Towards a uniform concept for the comparison and extrapolation of rockwall retreat and rockfall supply, Geogr. Ann. A, 89, 21–40, 2007.

Luckman, B. H.: Rockfalls and rockfall inventory data; Some observations from the Surprise Valley, Jasper National Park, Canada., Earth Surf. Proc. Land., 1, 287–298, 1976.

Matznetter, K.: Der Vorgang der Massenbewegungen an Beispielen des Klostertales in Vorarlberg, Geogr. Jahresber. O¨ sterreich, 25, 108 pp., 1956.

Rapp, A.: Recent developement in the Mountain Slopes in Kärkevagge and Surroundings, Northern Scandinavia, Geogr. Ann., 42, 1–158, 1960.

Sass, O. and Krautblatter, M.: Debris flow-dominated and rockfall-dominated talus slopes: Genetic models derived from GPR measurements, Geomorphology, 86, 176–192, 2007.

Whalley, W. B.: Rockfalls, in: Slope Instability, edited by: Brundsden, D., and Prior, D. B.,Wiley & Sons, London, 217–256, 1984.