Multi-method (14C, 36Cl, 234U/230Th) age bracketing of the Tschirgant rock avalanche (Eastern Alps): implications for absolute dating of catastrophic mass-wasting
Abstract
Correct and precise age determination of prehistorical catastrophic rock-slope failures prerequisites any hypotheses relating this type of mass wasting to past climatic regimes or palaeo-seismic records. Despite good exposure, easy accessibility and a long tradition of absolute dating, the age of the 230 million m3 carbonate-lithic Tschirgant rock avalanche event of the Eastern Alps (Austria) still is relatively poorly constrained. We herein review the age of mass-wasting based on a total of 17 absolute ages produced with three different methods (14C, 36Cl, 234U/230Th). Chlorine-36 (36Cl) cosmogenic surface exposure dating of five boulders of the rock avalanche deposit indicates a mean event age of 3.06 ± 0.62 ka. Uranium-234/thorium-230 (234U/230Th) dating of soda-straw stalactites formed in microcaves beneath boulders indicate mean precipitation ages of three individual soda straws at 3.20 ± 0.26 ka, 3.04 ± 0.10 ka and 2.81 ± 0.15 ka; notwithstanding potential internal errors, these ages provide an ‘older-than’ (ante quam) proxy for mass-wasting.
Based on radiocarbon ages (nine sites) only, it was previously suggested that the present rock avalanche deposit represents two successive failures (3.75 ± 0.19 ka bp, 3.15 ± 0.19 ka bp). There is, however, no evidence for two events neither in surface outcrops nor in LiDAR derived imagery and drill logs. The temporal distribution of all absolute ages (14C, 36Cl, 234U/230Th) also does not necessarily indicate two successive events but suggest that a single catastrophic mass-wasting took place between 3.4 and 2.4 ka bp. Taking into account the maximum age boundary given by reinterpreted radiocarbon datings and the minimum U/Th-ages of calcite precipitations within the rock avalanche deposits, a most probable event age of 3.01 ± 0.10 ka bp can be proposed. Our results underscore the difficulty to accurately date catastrophic rock slope failures, but also the potential to increase the accuracy of age determination by combining methods. Copyright © 2016 John Wiley & Sons, Ltd.
Introduction
Gravity-driven, extremely fast (‘catastrophic’) rockslides and rock avalanches exceeding 105–109 m3 in volume (Evans et al., 2006) are a major process of orogenic erosion and mass balance and represent a major threat to population and facilities in mountain ranges. With anthropogenic global change now in effect (Marzeion et al., 2014) a key issue in hazard assessment is whether an increased incidence of such events might be expected in future. One approach to this question is to correlate past event ages of catastrophic mass-wasting with reconstructed climatic and seismic records. Such correlations of course hinge on absolute age determination not only of mass-wastings, but also of earthquakes and climate. Letting aside problems of accuracy of past climatic and seismic records, the question herein is how correctly and precisely catastrophic rock-slope failures can be dated with methods presently available.
In the Alps, of at least some 600 post-glacial catastrophic slope failures, ~13% are historical events of more-or-less precisely known age (Ostermann and Sanders, 2012). Of the remaining prehistorical events, absolute ages of highly variable quality exist for a mere ~7%. This implies that some 80% of all post-Last Glacial rockslides of the Alps are still undated (Ostermann and Sanders, 2012). Absolute ages established so far indicate that previous, qualitative age assignments based on ‘glacial till’, ‘glacial overprint’ or other features most commonly were incorrect (Prager et al., 2008). In the literature, in many cases, one or two absolute ages – typically radiocarbon ages – often are cited as ‘the’ event age of catastrophic mass failure. Most radiocarbon ages are derived from soils or organic remnants: (a) originally underneath, (b) within, or (c) above a rockslide deposit (e.g. Deplazes et al., 2007; Patzelt, 2012a). Combining radiocarbon ages of these different types may allow for good age bracketing; unfortunately, this approach often is precluded by lack of suited deposits or exposures, absence of organic remnants or of remnants suited for age-dating, and/or because carbon-14 (14C) ages are biased (e.g. Prager et al., 2009). Only under highly favourable, rare circumstances 14C dating provides a proxy of event age sufficiently accurate and precise to allow straightforward correlation to palaeo-environmental records (climate, seismicity) (cf. Nicolussi et al., 2015). Cosmogenic nuclide surface exposure dating [here chlorine-36 (36Cl)] of detachment scarps of rock avalanches and/or of boulders within deposits is the only ‘direct’ approach to determine the age of a catastrophic slope failure event (Ballantyne et al., 1998; Antinao and Gosse, 2009). Recently, several rockslides and rock avalanches in the Alps (e.g. Ivy-Ochs et al., 2009; Ostermann et al., 2012; Zerathe et al., 2014) and worldwide (e.g. Hermanns et al., 2004; Sturzenegger et al., 2014) have been dated reliably using this method. Uranium-234/thorium-230 (234U/230Th) disequilibrium dating of diagenetic cements, in turn, is subject to the heuristic limit that it can only provide a minimum (ante quam) age of the event (Ostermann et al., 2007; Sanders et al., 2010). This still pioneering approach has been successfully tested at the Fern Pass rock avalanche (Prager et al., 2009) and several other sites provided suitable material for further analysis.
Herein we present a study involving 14C, 36Cl and 234U/230Th dating to constrain the event age of the Tschirgant rock avalanche in the northern Calcareous Alps (NCA) of Austria (Figure 1a). Some of the largest prehistoric rock slope failures in the Alps are concentrated in the Upper Inn Valley–Ötz Valley–Gurgl Valley area (Northern Tyrol, Austria) summarized in the Fern Pass rock avalanche cluster (Figure 1a). The Tschirgant rock avalanche represents a prominent example of a carbonate rock avalanche within this cluster.
We used 17 absolute ages – a figure so far never produced for such a type of deposit – to delimit the timing of rock slope failure (Figure 1b). Our results highlight the difficulty to determine a correct and sufficiently precise event age of rock-slope failure to support any conclusive correlation with other palaeorecords. If possible, catastrophic rock-slope failures should be dated by a multi-methodical approach. The Tschirgant study confirms the applicability of 234U/230Th dating of diagenetic cements as a method of ‘older-than’ or ante quam age determination of mass-wasting.
Geographical and Geological Setting of the Tschirgant Rock Avalanche
The Tschirgant rock avalanche is situated in the densely populated Inn valley, at the confluence of the Ötz River and the Inn River (Figure 2a). Mountain flanks, with a topographic relief of ~1200 to 1600 m, are very steep and rugged. Both the Ötz and the Inn valley were shaped by glacial ice streams, and are floored by glacial and inter-glacial deposits. The detachment area of the rock avalanche is the ‘Weißwand’ cliff more than 1400 m in height, beneath the summit of Mount Tschirgant [2370 m above sea level (a.s.l.)]. The scarp area is part of the southern margin of the NCA, which are separated by a fault, buried beneath the Quaternary deposits of the Inn valley, from the Ötztal-Stubai metamorphic basement complex (ÖSC) (Figure 1a) (Brandner, 1980; Linzer et al., 1997).
The rock avalanche detached from an intensely deformed, subvertically tilted succession of Triassic rocks, mainly dolostones, limestones, cellular dolomites and, subordinately, marls and siltstones (Figure 1a) (Pagliarini, 2008). The rockavalanche debris covers an area of 9.9 km2. The average thickness of the deposit is estimated at ~30 m (Abele, 1974), with a maximum thickness of 65 m (Patzelt and Poscher, 1993). For geographic information system (GIS)-supported volume calculations performed with Arcgis 10.1 and AutoCAD 14, a former valley surface has been interpolated from drilling logs and field observations, and combined with LiDAR (light detection and ranging) derived digital elevation model (DEM) analyses, resulting in a volume of ~230 million m3 of rock debris; this overlaps with calculations of former authors (Abele, 1974: 180–240 million m3; Patzelt, 2012a: 200–250 million m3). The maximum runout is 6.2 km, with a fahrböschung angle of 12.7° (Prager et al., 2008; Patzelt, 2012a).
Methods
Radiocarbon ages
The radiocarbon dataset (nine charcoal respectively wood samples, Table 1) we refer to has been published by Patzelt (2012a) and were obtained by accelerated mass spectrometry at different laboratories in Vienna and Zürich. Radiocarbon laboratory dates are calibrated to calendar years (quoted before present, bp) using the software OxCal Version 4.2.3. (Bronk Ramsey, 2001, 2009; Bronk Ramsey et al., 2001) and its implemented calibration curve IntCal13.
| Dating method | Sample | Coordinates (WGS84 UTM 32 N) | Elevation (m a.s.l.) | Sample material | 14C–Age (yr bp) | cal bc (yr) | cal bp (yr) | Remarks on sample circumstances | Reference | Age type | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Northing (m) | Easting (m) | ||||||||||
| 14C | T1 | 641081 | 5230895 | 750 | charcoal | 3465 ± 45 | 1896–1664 | 3846–3614 | overridden charcoal on fluviatile & debris flow sediments | Patzelt (2012a) | maximum age |
| 14C | T2a | 633686 | 5233088 | 794 | charcoal | 3465 ± 35 | 1885–1691 | 3835–3641 | burned layer in an overridden paleosol | Patzelt (2012a) | maximum age |
| 14C | T2b | 633686 | 5233088 | 794 | charcoal | 3355 ± 30 | 1740–1535 | 3690–3485 | organic remnants in talus material | Patzelt (2012a) | maximum age |
| 14C | T2c | 633686 | 5233088 | 794 | charcoal | 3615 ± 50 | 2138–1785 | 4088–3735 | burned layer in an overridden paleosol | Patzelt (2012a) | maximum age |
| 14C | T3 | 641462 | 5230907 | 812 | charcoal | 3507 ± 39 | 1936–1700 | 3886–3650 | burned layer in an overridden paleosol on lateclacial sediments | Patzelt (2012a) | maximum age |
| 14C | T4 | 639841 | 5230785 | 813 | wood | 2885 ± 20 | 1126–1000 | 3076–2950 | spruce trunk in clay-rich diamicton mobilized by the rockslide | Patzelt (2012a) | maximum age |
| 14C | T5a | 638893 | 5230472 | 910 | charcoal | 3230 ± 90 | 1737–1289 | 3687–3239 | burned layer in an overridden paleosol | Heuberger (1975) | maximum age |
| 14C | T5b | 638893 | 5230472 | 910 | charcoal | 2820 ± 110 | 1282–798 | 3232–2748 | burned layer in an overridden paleosol | Heuberger (1975) | maximum age |
| 14C | T6 | 637902 | 5232644 | 774 | charcoal | 2380 ± 35 | 730–391 | 2680–2341 | burned layer on top of rockslide deposits below an alluvial cone build up after the rockslide event | Patzelt (2012a) | minimum age |
| 234U/230Th | KN | 639802 | 5232394 | 718 | soda-straw stalactites | 3200 ± 260 | underside of boulder (artificial outcrop) | this study | minimum age | ||
| 234U/230Th | RO | 638474 | 5231657 | 732 | soda-straw stalactites | 2810 ± 150 | underside of boulder (artificial outcrop) | this study | minimum age | ||
| 234U/230Th | TSCH_3 | 639806 | 5231754 | 740 | soda-straw stalactites | 3040 ± 100 | underside of boulder (natural outcrop) | this study | minimum age | ||
| 36Cl | TS_1 | 639832 | 5231833 | 737 | rock surface | 3870 ± 500 | huge block on top of the rockslide debris | this study | event age | ||
| 36Cl | TS_2 | 639979 | 5231766 | 739 | rock surface | 2260 ± 500 | huge block on top of the rockslide debris | this study | event age | ||
| 36Cl | TS_3 | 640054 | 5231694 | 750 | rock surface | 2900 ± 500 | huge block on top of the rockslide debris | this study | event age | ||
| 36Cl | TS_4 | 640325 | 5231205 | 758 | rock surface | 7550 ± 500 | huge block on top of the rockslide debris | this study | pre exposure? | ||
| 36Cl | TS_5 | 638495 | 5231577 | 736 | rock surface | 2800 ± 500 | huge block on top of the rockslide debris | this study | event age | ||
- Note: The sample names in the first column go along with the sample locations indicated in Figure 1b.
The interpretation of radiocarbon ages in terms of a minimum age (ante quam age), event age, or maximum age (post quam age) depends on the stratigraphic context of samples and type of sampled material (e. g. paleosol, discrete wood fragments) (Lang et al., 1999). All of the sampling sites (Figure 1b) have been re-visited, and sample position was reconstructed as far as present outcrop allowed. The published dates by Patzelt (2012a) have been re-calculated, re-calibrated and re-interpreted. We used a radial plot (Galbraith, 1990; Vermeesch, 2009) to compare the different dates and to apply proper error propagation when weighted mean values were calculated.
Cosmic ray exposure dating
For cosmic ray exposure dating with 36Cl, the surfaces of five boulders were sampled in summer 2009 (Figure 1b). All sampled boulders were more than 4 m in diameter in exposed size, and were part of the rock avalanche deposit. We carefully selected blocks of Wettersteinkalk lithology that showed a relatively flat upper surface without or with little vegetation cover. The blocks had to be solid without significant internal shattering and they had to be in an exposed position without being prone to tilting or any downslope motions. To avoid spalled surfaces, we had sampled weathered surfaces comprising the top of boulders. The sampled surfaces were located between approximately 1.5 to 3 m above their surroundings. The positions of the boulders in the rock avalanche mass more than approximately 2 km off the toe of the detachment scar also precludes that they were derived by post-rock avalanche rockfalls. Shielding by snow has been taken into account as far as possible: at the nearby meteorological station Haiming (695 m a.s.l.), the mean annual number of days with a snow cover of ≥20 cm is 13.8 in the period 1971–2000 (see www.zamg.ac.at/fix/klima). The 36Cl was extracted from limestone and dolomite samples using standard procedures (Ivy-Ochs et al., 2004, 2009). The amounts of total-rock chlorine and 36Cl were determined by accelerator mass spectrometry (AMS) with the 6MV tandem at the AMS facility of ETH Zurich (Synal et al., 1997). Surface exposure ages and production rates were calculated according to Alfimov and Ivy-Ochs (2009). To calculate exposure ages, a production rate of 54 ± 3.5 36Cl atoms g Ca-1 yr-1 was scaled to the latitude and altitude of sampling sites according to Stone (2000), with a muon contribution of 9.6% (Stone et al., 1996, 1998). The uncertainties of the 36Cl exposure ages include both the uncertainties in the AMS measurements (listed with the atoms per gram of rock in Table 1) and the uncertainties for the individual production rates (all given in Alfimov and Ivy-Ochs, 2009). The uncertainties are added in quadrature as described in Balco et al. (2008).
234U/230Th Dating
For 234U/230Th dating, crusts of flowstone cement and soda-straw stalactites up to 5 cm in length were sampled at three different locations (Figure 1b). The cements and soda-straws had precipitated along the underside of large boulders, by precipitation in a meteoric-vadose diagenetic setting [see Ostermann et al. (2007) and Sanders et al. (2010) for diagenetic settings and model]. The monophase calcite formation is driven by meteoric dissolution–re-precipitation of micritic abrasive rock powder generated by dynamic disintegration during the rock avalanche event. For U/Th-dating, the fringes of calcite cement were sampled with a microdrill under the microscope. Organic material was removed physically as far as possible. Cleaned samples approximately 0.25 g in mass were spiked with a mixed 236U + 229Th spike and dissolved in HNO [see Ostermann et al. (2006, 2007) for detailed descriptions of methods]. Thorium and U analysis were done with an multi collector inductively coupled plasma (MC-ICP) mass spectrometer of the University of Bern. The U content of the samples ranged from 20 to 120 ppb.
Common problems in 234U/230Th disequilibrium dating of meteoric carbonates are open-system conditions and initial detrital contamination (e. g. Ludwig and Titterington, 1994; Debaene, 2003; Geyh, 2001, 2005). Closed-system conditions since mineral precipitation are tested for by generating a 234U/238U–230Th/238U activity ratio plot. This plot is interpreted with the confidence that subsamples subject to a re-opening of the diagenetic system plot distinctly off the cluster of most values (cf. Geyh, 2001). For terrestrial carbonates, a detrital Th contamination which is difficult to remove completely by physico-chemical treatment, the total sample dissolution (TSD)-’isochron’ approach is appropriate (Ku and Liang, 1984; Luo and Ku, 1991; Kaufman, 1993; Geyh, 2001, 2005) and has been applied here.
Standard deviations were calculated according to the maximum likelihood algorithm of Ludwig and Titterington (1994). For each subsample, supplementary single U/Th disequilibrium-age calculations were made with an unpublished Visual Basic program by Jan Kramers (retired leader of the research group on isotope geology, University of Bern, Switzerland) based on the equation of Kaufman and Broecker (1965). All U/Th-ages have been referred to the year ad 1950, the reference point for radiocarbon ages.
The Tschirgant Rock Avalanche Deposits
Dufresne et al. (2016) recently published a detailed morpho-lithological study of the rock avalanche deposit that links sediment characteristics with processes of emplacement; therefore, only the most salient features will be outlined here. From proximal to distal along the rock avalanche deposit, the relative arrangement of Triassic stratigraphic units as present in the detachment scarp (Pagliarini, 2008) is roughly preserved. The proximal part of the rock avalanche deposit thus consists of Wetterstein Dolomite, whereas the medial and distal parts also include other stratigraphic units and lithologies, respectively (different types of limestones, dolostones, cellular dolomites) (Prager et al., 2008; Patzelt, 2012a, Dufresne et al., 2016). Overall preservation of structural-stratigraphic arrangement is a typical feature of deposits of catastrophic mass movements (e.g. Strom, 2006). In the considered deposit, however, this was complicated by overthrusting and other types of deformation (folding, normal faults, tear faults; Dufresne et al., 2016) within the rock avalanche deposit, including the shear-off and incorporation of overridden fluvial deposits of Inn and Ötz valley derivation. Previous interpretations of incorporation of these fluvial sediments into the rock avalanche stressed injection upon water escape as a relevant process, but remained indefinite as to when injection took place relative to rock avalanching (Abele, 1974; Erismann and Abele, 2001). Recent interpretations stress synkinematic emplacement, that is, shear-off and entrainment of fluvial deposits – combined with water escape – during avalanche movement (Prager et al., 2008, 2012; Dufresne et al., 2014, 2016). Over most of its area, the rock avalanche deposit is littered with boulders of less than 1 m3 up to ~100 m3 in volume. The larger boulders are most common in the central part of the accumulation area (see Dufresne et al., 2016, for details). In addition, the deposit is sculpted with hummocks and transversal ridges that probably originated during propagation and deposition of the rock avalanche (e.g. Abele, 1997; Erismann and Abele, 2001; Patzelt, 2012a; Prager et al., 2012; Dufresne et al., 2014, 2016).
Results of New Dating Approaches, Previous Radiocarbon Ages and Difficulties in Dating
234U/230Th-Dating
At three locations in the rock avalanche deposit, diagenetic cements suited for 234U/230Th disequilibrium dating were found (Figure 1b). Thin sections showed that the growth of soda straws selected for dating was not interrupted by dissolution truncation of crystals. Closed-system conditions were indicated by 234U/238U–230Th/238U activity ratio plots. Several subsamples treated mathematically with the errorchron method (see earlier) indicated three mean precipitation ages of 3.20 ± 0.26 ka bp, 2.81 ± 0.15 ka bp, and 3.04 ± 0.10 ka bp (Table 1). These ages, of course, can only provide ‘older than’ (ante quam) half-brackets of the rock avalanche event.
Cosmic ray exposure dating Tschirgant rock avalanche
The 36Cl surface exposure dating of five rock surfaces resulted in ages between 2.3 ± 0.5 and 7.5 ± 0.5 ka (Table 1). The block which gave the 7.5 ± 0.5 ka age (TS_4) seems to represent an inherited nuclide concentration. Neglecting the 7.5 ± 0.5 ka ‘outlier’ age the other four 36Cl exposure ages range from 2.3 ± 0.5 to 3.9 ± 0.5 ka with a mean value of 3.06 ± 0.62 ka for the catastrophic rock avalanche event at Tschirgant.
Previous radiocarbon ages
Radiocarbon ages T5a and T5b (Heuberger, 1975) first indicated that the rock avalanche took place during the Holocene. Nevertheless, Heuberger (1975) discounted the results and postulated a glacier related formation of some rock avalanche morphologies. Although the exact sampling circumstances and the precise stratigraphic positions are not totally clear, we involved these dates in our calculations as two maximum ages for the event.
Based on seven radiocarbon samples from five different sites, Patzelt (2012a) distinguished two separate rock avalanche events separated in time by approximately 600 years. Rock avalanche deposits accumulated by successive dated mass-wastings are documented from several locations in the Alps (e.g. Lenhardt, 2007; Patzelt, 2012b). For the Tschirgant deposit, however, our own field investigations, interpretation of high-resolution LiDAR derived hill shade images, and numerous drill logs provided no conclusive evidence for two successive events. Based on radiocarbon, the lower age limit is bracketed by T6 (older than 2680–2341 cal bp); the upper age limit is given by T4 (3076–2950 cal bp) and by T5b (3232–2748 cal bp) (see Table 1). All other radiocarbon samples (T1, T2a, T2b, T2c, T3, T5a) in turn are older and mark maximum ages at about 3740 cal bp of that single event.
Because of the position of the samples within the rock debris accumulations (Figure 1b) and the site specific situation (Figure 3), we cannot support the assumptions of Patzelt (2012a) that T4 represents a second event age. Instead, we consider T4 as a minimum age and not T2b as the first event age rather a maximum age because the sampling circumstances are not to assume an event age.
Difficulties in Dating Rock Avalanches
All the evidence available to us, from field to numerical ages, to us indicates that a single catastrophic rockslide that turned into a rock avalanche happened between about 2.9 and 3.1 ka ago at Tschirgant in the Upper Inn Valley (Austria). Precise and accurate age estimates are not easy to obtain. The choice of the most appropriate dating method is not a simple matter and material suitable for dating is often scarce (Lang et al., 1999) with exception for cosmogenic nuclide dating. At Tschirgant, the extraordinary outcrop situation and a long research history provides an ideal place to perform a case study applying and comparing three different radiometric dating methods.
Our data highlight the difficulty to precisely date prehistorical catastrophic rock slope failures. We provide an age range from ~2.4 to ~3.4 ka bp for mass-wasting (Figure 4). A weighted mean from U/Th-dating and surface exposure dating results in 3014 ± 154 yr bp; taking into account all dates (published and new ones), their individual relationship to the catastrophic event and considering the overlap of the uncertainties a refined most probable period for the failure can be assumed at 3.01 ± 0.10 ka.
These results imply that catastrophic rock slope failure events dated by, say, a single or a few absolute ages by whatever method must be considered with reservation. As mentioned, precise dating is prerequisite to correlate magnitude/frequency of rockslides with past climate and seismicity. Even in the Alps, with their unparalleled data density, a mere 7% of at least 600 post-glacial rockslides deposits may be sufficiently well-dated to be potentially useful in comparison with palaeoclimatic records (Ostermann and Sanders, 2012).
This is our second study indicating that 234U/230Th disequilibrium dating of diagenetic products in rock avalanche deposits can yield ‘older-than’ proxy-event ages that are not inferior to other absolute dating methods with respect to scatter and error range. This supports our previous hypothesis that, in many cases at least, carbonate cements and/or speleothems suited for U/Th dating tend to precipitate shortly after mass-wasting (Ostermann et al., 2007). The U/Th method as a new tool in age constrainment of rock avalanches, and of their subsequent geomorphic changes (Sanders et al., 2010), is best applied in a frame of multi-method dating. Relative to other approaches, the main advantage of U/Th dating is easy and rapid sampling combined with small sample size. In addition, U/Th disequilibrium should be applicable up to about 490–500 ka, i.e. beyond the range of radiocarbon dating (cf. Geyh, 2005). Recognition of glacially-overprinted remnants of pre-Last Glacial Maximum mass-wastings (Gruber et al., 2009), and age-dating of such remnants, is an imminent challenge in rockslide research.
Conclusions
- A catastrophic rockslide that turned into a rock avalanche happened between about 2.9 and 3.1 ka ago at Tschirgant in the Upper Inn Valley (Austria) as a single event. About 230 million m3 of mainly carbonatic material spread over an area of at least 9.9 km2.
- The exceptionally good outcrop situation and data based on a long research history there allowed applying surface exposure dating as well as U/Th-dating successfully. The results agree with existing radiocarbon dates in a remarkable high degree.
- The 36Cl surface exposure dating of four boulders within the rock avalanche accumulation area resulted in an age of 3.06 ± 0.62 ka for the catastrophic event. Especially for rock-slope failures surface exposure dating provides reliable dates and is the only method that directly dates the event.
- The U/Th-dating of calcite precipitations of cements and small stalactites taken from megapores on boulder undersides that were newly formed shortly after the rock debris accumulation resulted in a mean age of 3.00 ± 0.16 ka. Lithification by cement precipitation within carbonatic rock avalanche accumulations is a common phenomenon and can be used to proxy-date the rock avalanche event and/or subsequent geomorphic changes of the rock avalanche mass.
- The re-calibration and re-interpretation of nine published radiocarbon dates taken from two authors resulted in a maximum age of 3.53 ± 0.26 ka bp and a minimum age of 2.51 ± 0.17 ka bp. Depending on the sampling circumstances and the sample material each sample has its own temporal relationship to the rock avalanche event in terms of minimal age and maximal age.
- Considering all dates from Tschirgant it is possible to estimate a most probable event age of 3.01 ± 0.10 ka bp resulting from the temporal overlap between two U/Th-ages (minimum ages; KN: 3200 ± 260 yr bp; TSCH_3: 3040 ± 100 yr bp) and a 14C–age (maximum age; T4: 3013 ± 63 cal bp).
- Our study shows that three independent radiometric dating methods provide comparable results within a time span of about 700 years and underlines the difficulty to precisely date a catastrophic slope failure. Keeping in mind that for most other rock avalanche accumulations except Tschirgant not all described dating methods are applicable it is shown that even a single method approach can result in a consistent event age-proxy.
Acknowledgements
This research was supported by the Austrian Science Fund FWF (P-20890-N10, to M. O.). Jan Kramers (University of Bern, Switzerland) is gratefully acknowledged for his encouragement, and his support in U/Th sample preparation and measurements. Rainer Brandner and Lucas Pagliarini (both University of Innsbruck, Austria) provided discussions on the structure of the detachment scarp.
