Project Introduction
A new approach to age-dating,
and implications for diagenetic changes in sturzstrom deposits.

In populated mountainous areas such as the Alps, rockslides are a substantial threat to humans and facilities (cf. Heim, 1932; Abele, 1974; Hewitt, 2006), and even small-sized rockslides can demand high toll of lives (Elm, Switzerland – Heim, 1882; Vaiont, Italy – Müller, 1964). Apart from the effects of human interference such as exessive quarrying (as was the case for Elm), it is inferred that the main triggers of rockslides are climatic changes, earthquakes and post-glacial stress relaxation (Erismann & Abele, 2001; Ballantyne, 2002). Warming and humidification of climate, such as the actual global climatic warming due to burning of fossil fuels, lead to melting of both glaciers and permafrost which, in turn, may increase the likelihood of large rockfalls and of rockslides (Harris et al., 2003). During the late-Glacial to Holocene interval of time repeated and marked changes between climatic warming and cooling, respectively, took place in the Alps, (Patzelt, 1980, 1995; Furrer et al., 1987; Burga & Perret, 1998). If it is correct that rockslides are associated with climatic changes, then a positive correlation between documented late-Glacial to Holocene phases of climatic warming/humidification and rockslide frequency must be postulated. Earthquakes as triggers of rocksliding may result from tectonic plate interaction, but smaller earthquakes for instance from local stress fields or by post-glacial rebound might also act as triggers. The catastrophic Huascaràn rockslide of the Andes was unambigously triggered by an earthquake, indeed the strongest earthquake documented for historical times in South America (Erismann & Abele, 2001, and refs therein). For nearly all other rockslides that descended in unpopulated areas and/or prior to historical records of seismicity, however, their relation to earthquake activity to date is difficult to document. Following glacial intervals, while deglaciation is perhaps followed by a phase of increased rockslide frequency, slow post-glacial relaxation of formerly loaded rocks may however also trigger large sturzstrom events thousands of years after deglaciation (Ballantyne, 2002). For larger rockfalls and, perhaps, also for small rockslides, post-glacial relaxation of high rock cliffs is probably a relevant trigger mechanism. It is not clear, however, whether large rockslides with a low 'angle of descent' (=‘Fahrböschung’, see e. g. Erismann & Abele, 2001) such as Almtal, Fern Pass, and Flims are directly related to deglaciation.

With respect to each of the (partly interrelated) hypotheses of rockslide triggers mentioned above – climatic changes, earthquakes, and post-glacial relaxation – a crucial question is the age of the numerous documented past rockslide events in the Alps. For assessment of the relation of rockslides to climate and earthquake activity, and of the likelihood of future sturzstroms, thus, age data on past events are crucial. In the last decades, sturzstrom deposits were dated by 14C age determination of wood fragments that are preserved (a) in lacustrine or fluvial successions underneath the rockslide mass, (b) within the rockslide mass, or (c) in newly-formed lakes on top of the sturzstrom (Patzelt & Poscher, 1993; Wassmer et al., 2004; Weidinger, 2004). In each case, the 14C age provides a different constraint for the age of the rockslide event, that is, in case (a) the 14C age represents a maximum age limit for the event, in case (b), which is very rare, the 14C age is a good proxy age of the event, and in case (c) the 14C age provides a minimum age limit. Unfortunately, the 14C approach to age-dating often cannot be applied because of absence of suited deposits or exposures thereof, lack of organic remnants or of remnants suited for age-dating, and/or because the resulting 14C age is fraught with marked imprecision (e. g. Prager et al., in press). In addition, the 14C method is limited in scope to the time interval back to about 50 ka b.p., such that potentially older rockslide events cannot be considered. In the past decade, an increasing number of sturzstroms have been dated by 36Cl exposure dating of detachment scars and/or of surfaces of boulders within the sturzstrom deposit (Ivy-Ochs et al., 1998, Reuther et al., 2006). Exposure dating has the advantage to ‘directly’ yield the age of a rockslide event, yet exposure ages are still characterized by an error range (2 σ standard errors) of more than a thousand years to thousands of years (see Prager et al. 2007, their fig. 4). Relative to the ups-and-downs of late-Glacial to Holocene Alpine climate (Patzelt, 1995), the error range of most exposure ages thus extends over a large part or more of a climatic fluctuation. To verify results and to improve the method of exposure dating cross-checking with other, independent age-dating methods is important (Gosse & Phillips, 2001; Geyh, 2005). At the present state, thus, an increasing number of the numerous Alpine rockslides that hitherto could not be determined by the 14C method will foreseeably be dated by exposure dating, but a cross-check with another method of age determination is desirable in each case.

Diagenesis of rockslide deposits