High resolution OSL and post-IR IRSL dating of the last interglacial–glacial cycle at the Sanbahuo loess site (northeastern China)

Highlights

• Quartz SAR OSL and K-feldspar pIRIR₂₉₀ methods were applied to loess section.

• Feldspar pIRIR₂₉₀ luminescence characteristics are investigated.

• Quartz OSL and pIRIR₂₉₀ ages are compared against each other.

• The pIRIR₂₉₀ ages are used to derive the sedimentation rates of the loess section.

Abstract

Northeastern China is located in the East Asian monsoon region; it is sensitive to both high and low latitude global climate systems. Loess deposits in the region have considerable potential as sensitive archives of past climate changes. However, research into loess deposition and climate change in this region is restricted by the lack of independent age control. In this study, coarse-grained quartz SAR OSL and K-feldspar post-IR infrared (IR) stimulated luminescence (post-IR IRSL; pIRIR₂₉₀) methods have been used to date the Sanbahuo loess site in northeastern China. The quartz OSL characteristics are satisfactory. The measured pIRIR₂₉₀ D_e's do not vary significantly with IR stimulation temperatures between 50 °C and 260 °C; a first IR stimulation temperature of 200 °C was adopted. Dose recovery tests were performed by adding different laboratory doses to both laboratory bleached (300 h SOL2) samples and natural samples; the results are satisfactory up to ∼800 Gy. Resulting quartz OSL and feldspar pIRIR₂₉₀ ages are in good agreement at least back to ∼44 ka; beyond this feldspar pIRIR₂₉₀ ages are older. The feldspar ages are consistent with the expected age of the S1 palaeosol (MIS 5). There appears to have been a period of fast loess deposition at ∼62 ka, perhaps indicative of winter monsoon intensification with a very cold and dry climate that lead to a serious desertification of dunefields in northeastern China.

Introduction

The loess record of northern China is considered to be semi-continuous and is one of the most widely studied terrestrial sediments in world (Liu, 1985, An, 2000, Guo et al., 2002, Lu et al., 2010). These loess-palaeosol sequences contain detailed archives of palaeoenvironmental changes and are highly sensitive to climatic changes, especially to shifts in the Asian summer and winter monsoon and/or Northern Hemisphere westerly circulation (Liu and Ding, 1998). Chronological control of loess sequences provides insights on the timing of past glaciation and dust transportation conditions (e.g. Lai, 2010, Stevens et al., 2008). However, the majority of studies have been carried out on the thickest and most complete loess deposits in the Central Loess Plateau and the northeastern Tibetan Plateau (Liu and Ding, 1998, Guo et al., 2002, Stevens et al., 2008, Lu et al., 2010). There are also extensive loess deposits in the downwind areas of the Horqin and Otindag dunefields in northeastern China; these deposits form a valuable archive to study the history of aridity and palaeoclimatic change in this region (Fig. 1). Until recently, only a limited number of studies have been undertaken concerning the loess stratigraphy and chronology (Zeng et al., 2011, Yi et al., 2012). A detailed numerical chronological framework for this region would contribute to a better correlation of individual sections and provide a timeframe for the regional climatic reconstruction in northeastern China.

Optically stimulated luminescence (OSL) dating, using the single aliquot regeneration (SAR) protocol (Wintle and Murray, 2006) and its derivatives is one of the most robust dating methods for establishing absolute chronologies for loess deposits worldwide (Roberts, 2008). Over the past decade, a considerable amount of research has been undertaken on luminescence dating of Chinese loess–paleosol sequences, mainly based on quartz OSL (Stevens et al., 2006, Stevens et al., 2008, Buylaert et al., 2007, Buylaert et al., 2008, Roberts, 2008, Lai, 2010, Li and Li, 2011). Unfortunately, the quartz OSL signal usually saturates at relatively low doses (∼200 Gy) - for loess deposits with a typical dose rate 3–4 Gy/ka, this restricts the use of quartz OSL to the last 50–70 ka (Buylaert et al., 2007, Roberts, 2008). The infrared stimulated luminescence (IRSL) signal from K-feldspar saturates at much higher doses than does quartz OSL, and offers an alternative approach. However, feldspar IRSL is known to often suffer from athermal (anomalous) fading (Wintle, 1973, Spooner, 1994, Huntley and Lamothe, 2001); this leads to an underestimation of the equivalent dose (D_e) and so the luminescence age of the sample. Recently, it has been suggested that IRSL signals measured at elevated temperatures after an infrared (IR) stimulation are more stable than the standard IRSL signal (Thomsen et al., 2008, Jain and Ankjærgaard, 2011); this has led to the development of so-called post-IR IRSL protocols (Buylaert et al., 2009, Buylaert et al., 2012, Thiel et al., 2011). Since then, several single-aliquot-based pIRIR dating protocols for feldspar have been developed and have been shown to give accurate ages both for very young (<10 ka, Reimann and Tsukamoto, 2012, Fu and Li, 2013) and old (>100 ka, Buylaert et al., 2012, Li and Li, 2011, Li and Li, 2012a, Li and Li, 2012b) samples. Post-IR IRSL dating thus has the potential to extend the loess chronostratigraphy beyond the limits of quartz OSL in northeastern China.

In this work, both standard quartz SAR OSL and K-feldspar post-IR infrared (IR) stimulated luminescence (post-IR IRSL; pIRIR₂₉₀) methods are employed to date the Sanbahuo (SBH) loess site in northeastern China. First, the quartz SAR OSL characteristics are documented. Then we investigate the luminescence characteristics of the elevated temperature IRSL signals in a SAR post-IR IRSL protocol. The quartz OSL and pIRIR₂₉₀ ages are compared with each other and with the loess-palaeosol stratigraphy. Finally, the luminescence ages are used to derive sedimentation rates for the upper part of the SBH loess section.