th 1E&G / Vol. 65 / No. 2 / 2016 / th 1–2 / DOI 10.3285/eg.65.2.th1 / © Authors / Creative Commons Attribution License
 E&G Quaternary Science Journal Volume 65 / Number 2 / 2016 / th 1–2 / DOI 10.3285/eg.65.2.th1www.quaternary-science.net GEOZON SCIENCE MEDIAISSN 0424-7116
 THESIS ABSTRACT
 Constraining the timing of deglaciation of the High Alps and 
rates of subglacial erosion with cosmogenic nuclides
 Christian Wirsig
 E-Mail:  wirsig@phys.ethz.ch
 University:  Laboratory of Ion Beam Physics, ETH Zürich, Switzerland
 Supervisors:  Susan Ivy-Ochs, Hans-Arno Synal, Christian Schlüchter, Naki Akçar, Urs H. Fischer, Sean Wille
 Dissertation online:  hp://e-collection.library.ethz.ch/view/eth:48357
 We use in situ produced cosmogenic nuclides to study two 
central aspects of the impact of glaciers on the evolution of 
U-shaped troughs, cirques and overdeepenings in the Alps. 
One is the aspect of available time – how long did large 
glacier systems occupy the Alps in the past? e other is 
the pace – how fast do glaciers erode their beds? 
We report chronological constraints on the onset of de-
 glaciation in three dierent study areas in the High Alps; 
Mont Blanc in the Western Alps, Haslital in the Central Alps 
and Zillertal in the Eastern Alps (W¡ et al. 2016b). In 
each study area we mapped glacial erosional marks (striae, 
crescentic gouges) and trimlines for a local reconstruction 
of the Last Glacial Maximum (LGM) ice surface and ©ow 
paerns. We select spots close to the trimline on exposed 
ridges of truncated spurs as preferred targets for dating the 
onset of ice surface lowering. Furthermore, we constrain the 
ice surface and terminal positions of local Lateglacial glacier 
extents by an analysis of mapping results and numerical 
glacier models. In the Oberhasli region (Fig. 1) we observe 
the ªrst sign of ice surface lowering of the Aare Glacier at 
Fig. 1: Longitudinal prole from Unteraargletser to Guannen showing reconstructions of LGM, Gsnitz and Egesen ice surfaces (W ­ et al. 2016c) 
based on ice-ow direction indicators and surface exposure ages. Two shear stress values, 100 kPa and 150 kPa, were used to model the Gsnitz scenarios. 
e Egesen ice surface (100 kPa) is constrained to be just slightly higher than Sto and lower than the directional ange of striae at Bälital (red line). 
Inset map shows location of Haslital (le) and Goldbergkees (right) study sites. Figure modied from W ­ et al. 2016c.
th 4 E&G / Vol. 65 / No. 2 / 2016 / th 1–2 / DOI 10.3285/eg.65.2.th1 / © Authors / Creative Commons Attribution License
 23.0 ± 0.8 ka (W¡ et al. 2016c). Published chronologies 
in the Alpine forelands suggest a similar date for the end of 
the ªrst LGM advance phase of the piedmont lobes. An only 
slightly lower ice surface of the Aare Glacier was therea¬er 
aained then abandoned at 17.7 ± 0.8 ka. We further deter-
 mine identical ages for the onset of ice surface lowering of 
18.5 ± 1.1 ka in the Val Ferret on the southern side of the 
Mont Blanc massif (Western Alps) and of 18.4 ± 1.4 ka in 
Zillertal in the Eastern Alps (W¡ et al. 2016b). e ages 
suggest synchronous decay of the LGM glaciers in the ac-
 cumulation zones of the Western, Central and Eastern Alps. 
To quantify the fundamental process of subglacial ero-
 sion we present an approach of erosion rate and burial time 
determination based on comparison of apparent exposure 
ages derived from same-sample analysis of 10Be, 14C and 36Cl 
(W¡ et al. 2016a). Due to the more complex production 
systematics, the production rate proªle beneath the rock 
surface of 36Cl has a dierent shape than the ones of 10Be 
and 14C. Erosion therefore results in systematically too old 
apparent 36Cl ages compared to 10Be and 14C. In addition, be-
 cause of the short 14C half-life, decay during burial causes 
too young apparent 14C ages compared to 10Be and 36Cl. At 
Goldbergkees (Eastern Alps), we apply this approach using 
10Be and 36Cl to quantify subglacial abrasion rates in the re-
 cently ice-free (< 100 years) glacier foreªeld. Both nuclides 
show that several of the analyzed surfaces were eroded by 
> 300 cm during the late Holocene (W¡ et al. in press). 
Calculated subglacial abrasion rates are > 5 mm / a. e 
deeply eroded surfaces are from sites more than 1 km 
upvalley of the LIA glacier terminus. On the other hand, 
several sites were hardly erdoded at all as shown by the 
presence of inherited nuclides. We observe severe erosion 
along the lower valley ©anks and both up- and downvalley 
of a cross-valley bedrock riegel (Fig. 2).
 References
 W¡, C., I²³´Oµ¶, S., A·¸¹, N., Lº»·¼, M., H»»¼, K., W¹µ·¼, L., 
V¾µ·¼¿¶ºÀ¼, C. Á Sµ¶ÂÃµ¶Ä¼, C. (2016a): Combined cosmogenic 
10Be, in situ 14C and 36Cl concentrations in proglacial bedrock constrain 
Holocene history and erosion depth of Grueben glacier, Switzerland. 
– Swiss Journal of Geosciences. DOI: 10.1007/s00015-016-0227-2
 W¡, C., I²³´Oµ¶, S., R¼Ä¿¼, J.M., C¶ÄÂ, M., V¾µ·¼¿¶ºÀ¼, C., 
Bµ¶Â¼, M. Á R¼¿ÇÂ, M. (?): Subglacial abrasion rates at Goldberg-
 kees, Hohe Tauern, Austria, determined from cosmogenic 10Be and 36Cl 
concentrations. – Earth Surface Processes and Landforms (in press).
 W¡, C., Z¹¹Ç¿, J., C¶ÄÂ, M., A·¸¹, N. Á I²³´Oµ¶, S. (2016b): 
Dating the onset of LGM ice surface lowering in the High Alps. – 
Èaternary Science Reviews 143, 37–50. hp://dx.doi.org/10.1016/j.
 quascirev.2016.05.001
 W¡, C., Z¹¹Ç¿, J., I²³´Oµ¶, S., C¶ÄÂ, M., K¾À¼, F. Á Sµ¶ÂÃµ¶´
 Ä¼, C. (2016c): A deglaciation model of the Oberhasli, Switzerland. – 
Journal of Èaternary Science 31, 46–59. DOI: 10.1002/jqs.2831
   
Fig. 2: Erosion depths into bedro along 
A) transverse and B) longitudinal valley 
proles at Goldbergkees (Austria) calculated 
with measured nuclide concentrations and 
known time of post-Lile Ice Age exposure 
of < 100 yr (W ­ et al. in press). Erosion 
is deepest part way up the valley sides 
(Gold4,9,10) and at the stoss (Gold13) and lee 
(Gold15) sides of the bedro riegel. e boxes 
showing the erosion depths are not to scale. 
Note that the scale anges from A to B but 
vertical exaggeration (~ 5x) stays the same. 
Figure modied from W ­ et al. in press.