LAUSR

Abstract The reliability of quartz electron spin resonance (ESR) dating using the Ti centre was tested through a comparison of the ESR ages of loess from the Luochuan site, Chinese loess plateau, with reference ages. Instead of the commonly used multiple-aliquot additive dose (MAAD) protocol, the single-aliquot regenerative dose (SAR) protocol was used to calculate the equivalent dose (De) values of five samples, with expected ages ranging from 28.5 ± 2.9 ka (unit L1) to 646 ± 65 ka (unit L6). The apparent ages, ranging from 59.4 ± 3.8 to 517 ± 42 ka, were then corrected for thermal signal loss, using the lifetime of the Ti centre, ∼1.7 × 106 years, previously calculated for a sample from the same site. The corrected ages ranged from 60.4 ± 3.8 ka to 617 ± 50 ka. For the youngest two samples both apparent and corrected ages overestimated the expected ages, which was presumably caused by incomplete signal resetting before burial. The apparent age of the oldest sample showed an age underestimation of ∼20%, but after the thermal signal loss from the Ti centre was corrected for, the age was consistent with the reference age. Both apparent and corrected ages of L3 and L4 are in good agreement with expected ages. We show, that with careful consideration of the thermal stability and bleachability of samples, reliable age calculation using the Ti centre is possible.