A test of rock surface luminescence dating using glaciofluvial boulders from the Chinese Pamir

Four gneiss boulders on Holocene glaciofluvial fans offset by the Kongur normal fault in the Chinese Pamir Plateau were sampled to test the reliability and applicability of rock surface luminescence dating. The 10Be ages of the boulders ranged from 2.1 to 7.5 ka. Depth profiles of luminescence intensity were measured using a modified multi-elevated-temperature post-infrared infrared stimulated luminescence (MET-pIRIR) protocol, in which two consecutive stimulations with different power, 15 and 131 mW/cm2, were performed for each temperature. The decay rate of the trapped charge at the surface σφ0¯ were calibrated using the 10Be age for each boulder sample. The resulting values for the parameter σφ0¯ were utilized to estimate the exposure time of other samples. The shielding effect of rock varnish was examined in relation to the attenuation coefficient μ and the wavelength of the sunlight that penetrates a rock surface. Uncertainties in μ derived from fitting each MET-pIRIR signal depth profile, corresponding the parameter σφ0t¯, and calculated ages were too large to allow precise comparisons to be made between vanished and unvarnished rock surfaces. The μ was assumed to be the same for all MET-pIRIR signals, and so was fitted simultaneously to all profiles derived from a single sample; this significantly reduces the fitting uncertainties in the parameters μ and σφ0t¯. The long-term relative bleaching rate of the IR50, MET-pIRIR110, MET-pIRIR170 and MET-pIRIR225 signals were evaluated by comparing the ratios of σφ0t¯ derived from fitting of luminescence-depth profile. The IR50 signal, MET-pIRIR110 and MET-pIRIR170 bleached tens, ∼5, and 2 times faster than the MET-pIRIR225 signals, respectively. These results highlight the potential and challenges in using rock surface luminescence methods as a viable dating tool for exposure ages.