The Tibetan Plateau (TP), known as the “Water Tower of Asia” and “the Third Pole”, is highly relevant to Asian monsoon systems (East Asian monsoon and Indian monsoon), and global climate change due to its massive extension at high altitudes (Krause
Several studies have been conducted to document the paleoenvironmental change on the TP by dating relict beach ridges and lacustrine sediments with optically stimulated luminescence (OSL), cosmogenic radionuclides and 14C methods (e.g., Kong
In this paper, we carried out a combined stratigraphic and chronological analysis on a lacustrine outcrop from Cuoe Lake on the central TP. Both luminescence and radiocarbon dating methods were used to reconstruct the evolution of the lake level during the Holocene.
Selin Co Basin (
The CE1 site (31° 30′ 29″ N, 88° 48′ 20″ E) is about ∼18 m above present lake level (a.p.l.l.), and the ∼2 m thick section exhibits significant variations in sedimentary composition and structure. It can roughly be divided into three stratigraphic units (U1 from the bottom, U2 in the middle and U3 from the top) by apparent unconformities (
All luminescence samples’ preparation and measurement were performed at the Luminescence Dating Laboratory of NIGLAS (Nanjing, China). In subdued red light, the daylight-exposed outer layer of the samples was scraped off for water content and dose rate measurements. The remaining (un-exposed) material was prepared for the equivalent dose (De) measurements.
These six samples were wet sieved to extract the coarse grain size fraction (63–90 μm or 100–200 μm). Samples were firstly treated with 30% H2O2 and 10% HCl to remove organic material and carbonates, respectively. Subsequently, the K-rich feldspar fraction (<2.58 g/cm3), and the quartz-rich fraction (2.62–2.70 g/cm3) were separated with a heavy liquid solution (sodium polytungstate). The quartz grains were etched with 40% hydrofluoric (HF) acid for one hour to remove the outer alpha-irradiated layer of the quartz and to eliminate feldspar contamination, and then the etched fraction was rinsed with 10% HCl to remove fluoride. Considering that HF etching on feldspar tends to attack the cleavage planes of the mineral grains, instead of removing the outer rim uniformly (Duller, 1994), we did not etch the K-feldspar grains with HF. Instead, we took the external alpha-irradiated dose rate of the feldspar grains into account for total dose rate calculation. The pure quartz and K-feldspar grains were mounted as 6 mm and 2 mm diameter monolayers respectively on 9.8 mm diameter stainless steel discs using silicone oil as an adhesive for measurements.
De values were measured with an automated luminescence reader (Risø TL/OSL DA-20) equipped with a 90Sr/90Y beta source. For quartz OSL, blue LEDs (470 nm) were used for stimulating the aliquots and a 7.5-mm Hoya U-340 filter was used as the signal detection filter in front of the photomultiplier tube. For the infrared (IR) stimulation of K-feldspar, infrared LEDs (870 nm) were used and the infrared-stimulated luminescence (IRSL) signal was detected through a combination of Schott BG-39 and BG-3 filters in the blue-violet spectrum between 320 and 450 nm.
Approximately 60 g of sediments from each luminescence sample were dried for 4 days at 50°C until constant weight to determine the present-day water content. The dried material was homogenised, and a subsample of 5 g was ground to powder for the measurement of dose rate related elements,
The De of CG quartz was determined using a routine single-aliquot regenerative-dose (SAR) protocol (Murray and Wintle, 2000) with an additional step of IR depletion to monitor feldspar contamination (Duller, 2003) in the SAR sequence. A previous study pointed out anomalous fading on quartz samples from an adjacent region (Rades
Thomsen
Protocol used to measure pIRIR150.
1 | Dose | |
2 | Preheat at 180°C for 60 s | |
3 | IRSL measurement at 50°C for 100 s | Lx1 |
4 | IRSL measurement at 150°C for 200 s | Lx2 |
5 | Test dose | |
6 | Preheat at 180°C for 60 s | |
7 | IRSL measurement at 50°C for 100 s | Tx1 |
8 | IRSL measurement at 150°C for 200 s | Lx2 |
9 | Return to step 1 |
Radiocarbon dating was determined at Beta Analytic. The samples were processed using a conventional acid-alkali-acid treatment to remove carbonate and other contaminants. The alkali-insoluble organics were obtained and then prepared as graphite for AMS radiocarbon measurements. The measured 14C ages were calibrated to calendar years using the CALIB program with the IntCal 13 dataset after 13C/12C adjustment (Reimer
Summary of the AMS 14C ages.
Beta-471695 | 162 | shell | 13060±30 | 15.6±0.1 |
Beta-471698 | 0 | shell | 1320±30 | 1.3±0.04 |
In order to confirm the suitability of the used measurement protocol, it is commonly accepted that three routine tests should be carried out for any new SAR protocol,
Furthermore, dose recovery tests were undertaken for all six samples (three aliquots for each). First, the aliquots were bleached with a solar simulator (Hönle SOL2) lamp for 4 h (e.g., Reimann and Tsukamoto, 2012; Buylaert
It is well known that most feldspar samples suffer from anomalous fading. Huntley and Lamothe (2001) showed that the luminescence signal decayed linearly with the logarithm of time and attributed this loss to anomalous fading which is usually quantified by the
In order to investigate bleaching behaviour for the pIRIR150, residual doses were first measured using three aliquots for each sample after exposure to Hönle2 for 4 h (Zhang
For each sample, 12–24 aliquots were used for De measurement. All Des estimates derived from the pIRIR150 signal for the six samples are listed in
Luminescence dating results for six samples from CE1 profile.
CE1 | NL-1456 | 40 | 2.63±0.10 | 9.81±0.28 | 1.59±0.05 | 4.16±10 | 0.42±0.04 | 3.92±0.25 | 0.45±0.04 | 0.1±0.01 |
NL-1457 | 90 | 2.1±0.09 | 7.02±0.22 | 1.00±0.04 | 10.70±10 | 0.39±0.04 | 3.72±0.26 | 29.70±0.2 | 7.9±0.5 | |
NL-1458 | 150 | 3.89±0.14 | 9.58±0.28 | 1.44±0.05 | 10.86±10 | 0.36±0.04 | 4.08±0.26 | 33.23±0.4 | 8.0±0.5 | |
NL-1459 | 170 | 5.87±0.21 | 6.85±0.23 | 1.03±0.04 | 30.74±10 | 0.35±0.04 | 3.56±0.25 | 27.49±0.6 | 7. 6±0.6 | |
NL-1460 | 210 | 6.67±0.21 | 6.2±0.20 | 0.85±0.04 | 29.88±10 | 0.33±0.03 | 3.52±0.26 | 25.13±0.2 | 7.1±0.5 | |
NL-1461 | 245 | 2.71±0.11 | 9.45±0.27 | 1.61±0.05 | 2.48±10 | 0.32±0.03 | 3.68±0.24 | 34.69±0.3 | 9.4±0.6 |
It is essential to compare the data generated using luminescence with an independent age control (e.g., 14C dating). Terrestrial plant macrofossils are well known to be most suitable for radiocarbon dating of lake sediment, as they are not affected by old carbon reservoir effects (Long
Based on the luminescence dating and stratigraphic analysis, we constructed the chronostratigraphy of the CE1 section (
This study reconstructed the sedimentary history of Cuoe Lake, mainly based on luminescence dating of outcrop CE1, and then lake level fluctuations were inferred. Luminescence characteristic tests show that quartz OSL signal is too dim to date and even suffers from anomalous fading, which limits the application of quartz mineral. Alternatively, the suitability of the pIRIR protocol measured at 150°C (pIRIR150) for K-feldspar fractions is confirmed by a set of luminescence tests. Compared with luminescence age, 14C dating of a shell sample recovered from the lake sediments yields much older age by ∼7 ka, which might be a result of hardwater reservoir effect. Combining the luminescence-based chronology and sedimentary analysis, we conclude that the Cuoe Lake has experienced peak water-level during the early Holocene (∼9.4–7.1 ka), and lake-level drop afterwards; this lake-level variation pattern is probably related to the changes of the dominant summer monsoon system.