A HOLOCENE CLIMATIC RECORD DENOTED BY GEOCHEMICAL INDICATORS FROM BARKOL LAKE IN THE NORTHEASTERN XINJIANG,
NW CHINA
© 2010 Zhong Weia b*, Xue Jibina, Li Xiaodongc, Xu Huajunb, Ouyang Juna
aSchool of Geography Science, South China Normal University, Guangzhou, 510631, China bKey Laboratory of Oasis Ecology (Ministry of Education), Xinjiang University, Urumqi 830046, China c School of Tourism, Xinjiang University, Urumqi 830046, China * E-mail: DL06@scnu.edu.en Received 02.09.2008
The Barkol Lake, as a closed inland lake, is located at the northeast Xinjiang in northwest China. A combination of geochemical indicators including S18O and 813C of carbonate, TOC, carbonate contents, as well as grain size proxies and magnetic susceptibility of sediments obtained from a newly recovered section at this lake, provides a high-resolution history of climatic change in the past 9400 years. Multi-indicators reflect that Holocene climatic change in the study region generally follows the Westerly Wind pattern of Holocene, and three climatic periods can be identified. Between 9400 and 7500 cal a B.P., climate was characterized by relatively drier and colder condition. From 7500 to 5800 cal a B.P., a relatively warmer and moister climate prevailed, but between 5800 and 3500 cal a B.P., climate shifted towards warmer and drier conditions. A relatively colder and wetter climate prevailed during 3500~1000 cal a B.P., then it changed towards cold and dry between 1000 and 500 cal a B.P.; after 500 cal a B.P., climate changed towards warm and dry conditions again. This study reflects that during the Middle Holocene (from ca 7000 to 3500 cal a B.P.), variations of carbonate S18O of sediments from several lakes in the northern Xinjiang were synchronous with that of Qinghai Lake, where was strongly influenced by the South Asian monsoon; however, after 3500 cal a B.R this consistency was interrupted, possibly resulting from a re-domination of the Westerly Wind and the retreat of South Asian monsoon in the northern Xinjiang.
INTRODUCTION
In recent years, the important significance of the Westerly Wind has attracted great interests of scientists because the Westerly Wind as a link between the climates of the North Atlantic and the East Asian (EA) monsoon, in some sense, controls the northern bound of the EA monsoon and plays an important role in variation of hydrothermal characteristics in Asian inland terrestrial climate [1]. The Westerly Wind interacting with the Tibetan Plateau profoundly influences Asian monsoon climate and global climate. In China, the Westerly Wind affected areas mainly concentrate in the northwest arid and semiarid regions (mainly in NW China). Though considerable researches concentrating on climatic changes in the areas have been carried out since the end of 1980's [2—6], however, discrepancies on the details of climatic changes and patterns of hydro-thermal combination still exist.
In the past two decades, several paleoclimatic studies have been performed on Barkol Lake, where was strongly influenced by the Westerly Wind [5, 7—9], among which the most detailed study focusing on the Holocene climate was contributed by Han's core ZKOOA [5]. However, we do not think this core was able to provide a highresolution record of Holocene climatic changes because its sampling intervals (generally at 10~15 cm intervals) could not satisfy high-resolution study on paleoclimatic
changes; on the other hand, the radiocarbon 14C ages of those samples were not calibrated and the carbon "reservoir" effect was not taken into consideration. In this paper, we try to present a high-resolution record of Ho-locene climatic changes based on a section newly recovered from Barkol Lake and to explore the possible cause responsible for climate changes in the Holocene.
MATERIALS AND METHODS
Barkol Lake (43°36' ~ 43°43' N, 92°43' ~ 92°51' E), a closed inland salty lake at the altitude of1580 m above the sea level, was developed in the Barkol basin which is surrounded by Barkol Mountains (in the eastern part of the Tian-shan Mountains) and Moxinwula Mountains (in the eastern part of the Altai Mountains) and is located about 14 km west of the county town of Barkol Hasakustan Autonomous County (Fig. l). This lake is mainly supplied by melting water from the surrounding mountains and has a catchment of 4514 km2. At present, this lake has shrunk to about 90 km2, much of ancient Barkol Lake bed was covered with glauberite and saline pan deposits ofhalite with minor amounts ofgypsum and exposed mudflat.
Present climate of Barkol Lake region reflects a typical temperate arid climate. Variation of modern precipitation shows evidently negative correlation with temper-
(A)
(B)
70°
—r~
90°
110°
,130° E
Siberian* \
Mongolian High .
-■ai\O......^esfÎ
IJ
_ -40
n
md
- (Tibetan plateau
1-30°
;.................ö................
20° Mk:-"
0_300 km S-"'"
.....
P.R. China
t
\
\ tfv
Fig. 1. The locations of the Barkol Lake (including the sites of cores and section at Barkol Lake) and other lakes mentioned in this paper (A); the climatic background of the study region (B) (Modified after [9] and [10]).
J240
§200
o
&160
e
>120 io tati 80
f 40
0 -
(a)
a Temperature • Precipitaion 0 Evaporation -- Surface runoff
123456789 10
11 12 Month
20
10
O
V
P t
re o
--0
-10
-20
1960
1970
1980
1990 Year
Fig. 2. Variations of monthly meteorologic factors, surface runoff (a) and changes of temperature and precipitation since 1950's (b) in the Barkol Lake region (after [8]).
ature (Fig. 2). Mean annual evaporation and mean annual precipitation are 1638 mm and 202.3 mm, respectively. Precipitation mainly concentrates in June-November with a peak value of about 100 mm in July (Fig. 2). Mean annual temperature is 1.1°, the average temperatures of January and July are —18.6 and 16.9°, respectively. Generally, this lake was covered with ice from November to March.
In July 2004, we carried out an excavation and retrieved a 250-cm-deep sediment section (named BLK-1 section, 43°42' N , 92°50' E ) at the center of Barkol Lake (Fig. 1). Totally 236 samples were taken at 1-cm (from surface to 230 cm depth) and 3-cm intervals (mainly in the bottom 20 cm of the section), respectively. Samples were sealed on the spot and taken to laboratory, where those were completely air-dried. Seven organic enriched bulk samples were taken for the conventional radiocarbon 14C measurements. Radiocarbon ages were calculated with a half-life of 5568 years and were calibrated to the calendar ages using Calib 4.0 and the INTCAL98 data set [11].
Only the <4 ^m size fraction of each sample was analyzed for carbonate oxygen and carbon isotopes using a Finnigan MAT-252 Mass Spectrometer. Prior to the analysis, the sample was heated for 1 hour under vacuum at about 475 and then treated with 100% H3PO4. The isotopic analysis of carbonate was performed on CO2 prepared from the carbonate of the sediment. All isotope values were reported in standard delta notation relative to the PDB standard. Precision of analysis was ±0.02%.
Treated with 4 N HC1, carbonate content of sample was measured using Bernard Calcimeter with a relative
Radiocarbon dating results of section BLK-1 of the Barkol Lake
Sample No. Lab. No. Depth/cm 14C /a B.P. Calib(2a) /cal a B.P. Material dated
BLK—1—243—246 LZU05—44—1 4-7 907 ± 63 698-930 TOC
BLK-2-J231-234 LZU 05-43 16-19 1590 ±65 1344-1686 TOC
BLK—3—198—200 LZU 05-42-1 50-52 2245 ± 58 2125-2350 TOC
BLK—4—170—173 LZU 05-41-2 77-80 3422 ± 60 3486-3839 TOC
BLK—5—142—145 LZU 05-40 105-108 4340 ± 60 4823-5266 TOC
BLK—6—110—113 LZU05-39 137-140 5166±65 5742-6176 TOC
BLK—8—040—03 5 LZU 05-37 -1 210-215 8111±72 8774-9280 TOC
error less than 3%. Total organic matter (TOC) was determined by weight loss on ignition (LOI) at 550° for 2 h. Particle size of the sample was determined using a Malvern Mastersizer 2000 Laser Analyzer with a measurement range of 0.02—2000 ^m. Samples were pre-treated with 10—20 ml of 30% H2O2 to remove organic matter and then with 10 ml of 10% HC1 to remove carbonates. About 2000 ml of deionized water was added and the sample solution was kept for 24 h to rinse the acidic ions. The sample residue was finally treated with 10 ml of 0.05 M (NaPO3)6 on an ultrasonic vibrator for 10 minutes to facilitate dispersion before grain-size analysis. The Mastersizer 2000 automatically yields the median diameter and the percentages of the related size fractions of a sample with a relative error of less than 1%. Mass magnetic susceptibility (SI) of sediments was measured using a Bartington-MS 2B Magnetic Susceptibility Meter.
RESULTS AND INTERPRETATION
Variation oflithology
Sediments of BLK-1 section are of different colors and are mainly composed of clay or silty clay. No sedimentary hiatus was identified. The overall sedimentary characteristics of BLK-1 section were shown in Fig. 3. Litho-units of this section are in good accord with that of core ZK00A and section RG reported by Han et al. [5] (Fig. 3).
Chronology
Radiocarbon dating results and the relationship between age and depth were shown in Table and Fig. 3, respectively. In saline lake sediments, the "reservoir" effect is frequently the most important factor prohibiting accu-
rate radiocarbon ages. Several methods, both indirect and direct, can be utilized to determine the reservoir effect in saline lake [12, 13]. Based on a previous study the "reservoir" effect of Barkol Lake was determined to be 750 years [14]. Therefore, after subtracting 750 years from all radiocarbon ages, the chronological sequence of section BLK-1 was reconstructed based on linear interpolation for every sample, its bottom age was determined to be ca 9400 cal a B.P. and the average resolution of each sample was ca 38a.
Carbon and oxygen isotopes of carbonate
In the arid area, 818O of authigenic carbonate (818Ocarb) in the closed lakes was primarily controlled by both the isot
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.