Quaternary Science Reviews, Vol. 7, pp. 287-293, 1988. 0277-3791/88 $0.00 + .50 Printed in Great Britain. All rights reserved. Copyright© 1988 PergamonPress plc THE POTENTIAL OF USING THERMOLUMINESCENCE TO DATE BURIED SOILS DEVELOPED ON COLLUVIAL AND FLUVIAL SEDIMENTS FROM UTAH AND COLORADO, U.S.A.: PRELIMINARY RESULTS S.L. Forman, M.E. Jackson, J. McCalpin* and P. Maat Center for Geochronological Research, INSTAAR, University of Colorado, Boulder, Colorado 80309-0450, U.S.A. *Department of Geology, Utah State University, Logan, Utah 84322, U.S.A. The natural TL intensity for surface and buried Holocene and Pleistocene A horizons developed on flood-plain silts, near Denver, Colorado exponentially decreases with time. This signal is approaching saturation by ca. 130 ka. The A horizon of the modem flood-plain soil is not fully light bleached. The TL properties and age estimates are presented for radiocarbon dated, eolian-enriched buried-A horizons developed on fault-derived colluvium from the American Fork segment of the Wasatch fault zone, Utah. Dating of these buried soils provide a close age estimate on paleoearthquake events. Mean TL age estimates by regeneration and total bleach techniques for buried A horizons are 0.5 + 0.1 ka and 2.7 + 0.4 ka which are in agreement with corresponding radiocarbon dates of 980 + 70 years BP and 2620 + 70 years BP. A surface sag pond mud formed within an antithetic grabben is well light bleached and yielded a TL age estimate by the total bleach method of 240 + 60 years BP, in agreement with its known age of <300 years BP. This study indicates that relatively brief periods of pedogenesis are not sufficient to light-bleach sediment and that eolian additions enhance the reduction of TL in soils. INTRODUCTION The techniques to date sediment by thermolumin- escence (TL) have mostly been developed for and applied to well light-bleached eolian deposits. Re- cently, buried A horizons have been identified with a residual TL level (I0) similar to that of loess, and thus many of the TL techniques to date loess may be extended to buried soils (Wintle and Catt, 1985). However, the mechanism of light bleaching in soils is not well understood. Vares (1982) attributed the decrease in TL signal in the upper horizons of podzolic soils developed in glacial-fluvial sands from Estonia to chemical weathering. Huntley et al. (1983) hypothesize that the TL of buried A horizons from British Columbia were reduced by light bleaching or possibly by chemical weathering. Wintle and Catt (1985) re- ported the reduction of A horizon TL from light exposure enhanced by pedoturbation for buried soils developed in loess and colluvium, near Kent, England. They also considered the effects of decalcification, gleying and translocation on the stability of the TL signal. Questions remain regarding the duration of soil development necessary to reduce the TL of unbleached or partially bleached parent material. Also, there is a need to assess the effect of eolian additions, a common constituent of soils in the arid western U.S.A. (Birke- land, 1985), to the TL properties of surface horizons. We present the TL properties of incipient modern and buried Holocene and late Pleistocene soils devel- oped on flood-plain silts near Denver, Colorado. Also reported are the TL properties and age estimates for radiocarbon dated eolian-enriched buried soils and sag- pond muds developed on fault-derived colluvium from the American Fork segment of the Wasatch fault zone, Utah. This study of independently dated material tests the accuracy of TL to date buried soils developed on fluvial and colluvial deposits in the western U.S.A. and provides key information for the extension of the technique to date older sediments. LABORATORY PROCEDURES Measurements were made on the fine (4-11 I~m) polymineral sediment fraction after removal of calcium carbonate. The TL was measured by a 'Daybreak' reader using an EMI 9635QB photomultiplier tube fitted with a Chance-Pilkington HA-3 heat absorbing filter and a Schott UG-11 filter. The UG-11 filter is reported to enhance the feldspar signal relative to that of quartz (Debenham and Walton, 1983) although this has been disputed by Berger (1985). Samples were heated at a rate of 5°C/sec in an argon environment after initial evacuation. Dated samples, ITL16, ITL18 and ITL23 were preheated at 150°C for 16 hr to remove an unstable TL component (cf. Wintle, 1985; Forman et al., 1988). A 125 mCi 9°Sr/9°Y beta source was used for irradiation. In order to compare the amounts and sensitivities of TL contributing minerals the discs were normalized to each other by the following procedu':e: (1) after initial glowing all discs received a second beta dose of 21.9 Gy; (2) glowing of discs was delayed 2 hr and; (3) the area between 150-300°C of the second dose glow- curve was used to normalize the TL signal of the initial glow curve. Thick-source alpha counting provided uranium and thorium estimates (Huntley and Wintle, 1981). 287