Differential bleaching of quartz and feldspar luminescence signals under high-turbidity conditions

Abstract. Sediment burial dating using optically stimulated luminescence (OSL) is a well-established tool in geochronology. An important but often inapplicable requirement for its successful use is that the OSL signal is sufficiently reset prior to deposition. However, subaqueous bleaching conditions during fluvial transport are vastly understudied; for example the effect of turbidity and sediment mixing on luminescence bleaching rates is only poorly established. The possibility that slow bleaching rates may dominate under certain transport conditions led to the concept that OSL could be used to derive sediment transport histories. The feasibility of this concept is still to be demonstrated, and experimental set-ups are still to be tested. Our contribution to this scientific challenge involves subaquatic bleaching experiments, in which we suspend saturated coastal sand of Miocene age in a circular flume and illuminate it for discrete time intervals with natural light. We record the in situ energy flux density received by the suspended grains in the UV-NIR frequency range by using a broadband spectrometer with a submersible probe. Our analysis includes pre-profiling of each sample following a polymineral multiple signal (PMS) protocol. Using the PMS, the quartz-dominated, blue-stimulated luminescence signal at 125 ∘C (BSL-125) decays slower than the K-feldspar-dominated, infrared-stimulated luminescence signal at 25 ∘C (IR-25) even under subaerial conditions. The BSL-125 from purified quartz shows the opposite behaviour, which renders the PMS unreliable in our case. We find a negative correlation between suspended-sediment concentration and bleaching rate for all the measured signals. For outdoor bleaching experiments we propose to relate the measured luminescence dose to the cumulative received irradiance rather than to the bleaching time. Increases in the sediment concentration lead to a stronger attenuation of the UV–blue compared to the red–NIR wavelength. This attenuation thereby follows an exponential decay that is controlled by the sediment concentration and a wavelength-dependent decay constant, λ. As such λ could potentially be used in numerical models of luminescence signal resetting in turbid suspensions.