Direct dating of fault gouge using feldspar luminescence: a note of caution

Luminescence thermochronology using feldspar has opened up an exciting new possibility to constrain Quaternary landscape evolution, as no other method is capable of (e.g. King et al., 2016; Biswas et al., 2018). The method has also the potential to estimate the past activity of faults, directly from fault gouge. In this study, we test the applicability of feldspar luminescence dating on fault gouge samples from the Simplon Fault, one of the major normal faults in the central European Alps, using three samples from an outcrop in the Rhone Valley. The samples were gently crushed and sieved, and then density separated to extract Na-rich feldspar (2.58<d<2.62) and K-rich feldspar (d<2.58) grains. The post-IR infrared-stimulated luminescence signal, measured at 225°C from the Na-rich feldspar fraction, showed an extremely small characteristic saturation dose (D0) of below 100 Gy, and the natural signal was indistinguishable from saturation. However, multigrain aliquots of the K-rich feldspar fraction yielded diverse equivalent dose values from as low as ~20 Gy to saturation. Thin-section analysis of the samples suggests that some feldspar grains have been highly retrograded to other minerals e.g. muscovite. However, the age of such neo-crystallised muscovite was estimated to ~10-14 Ma using the 40Ar/39Ar method (Campani et al., 2010). Our results suggest the existence of recrystallised feldspar in the fault gouge, and therefore a careful thin section analysis will be required to interpret the feldspar luminescence ages.