论文部分内容阅读
【Abstract】A high-resolution pollen-based bioclimatic reconstruction from Valikhanov section in Kazakhstan， supported by 6 AMS dates reveals vegetation and climate change history during last glaciations and Holocene. The vegetation probably was Artemisia arid steppe during ～28，000 - ～18，000 14C yr BP.， then turned to be Artemisia steppe during ～18，000 - ～10，000 14C yr BP， an Atermisia-Chenopodiaceae desert steppe during ～10，000 - ～5，000 14C yr BP and a Chenopodiaceae-Artemisia steppe after ～5，000 14C yr BP in this region. Climate is cold and slightly wet during 28，000-10，000 14C yr BP.， dry and mild during 10，000-5，000 14C yr BP.， wet after ～5，000 14C yr BP in this region. Climate in the Central Asian domain was constantly wetter than that in the East Asian domain during MIS2. The Central Asian domain experienced a wetter climate in late Holocene whereas the East Asian domain expereinced a wetter climate in early Holocene.
【Key words】pollen； MIS2； Holocene； Central Asia； climate
Ⅰ. Introduction
MIS2 and Holocene is an important timeframe for understanding climate change and climate variability because， for the most part， the climatic boundary conditions are similar to those experienced now and possibly in the near-future. However， the temporal and spatial patterns of the climatic variability are poorly constrained regionally. In particular， we lack well-dated and high-resolution records from continental interiors. Central Asia is one of such areas， having been shown to display extreme sensitivity to past large-scale force-driven changes. It is a climatic transitional zone between the westerly-dominated “European” system to the west and the monsoon-dominated “East Asian” system to the east. During the Holocene and the last glacial/interglacial cycle， the “European” climate has primarily alternated between a cool/wet mode and a warm/dry mode， whereas the “East Asian” climate has alternated between a warm/wet mode and a cold/dry mode. Pollen data have been frequently used in vegetation and climate reconstructions， as they tend to reflect vegetation and climate changes at a regional scale. However， there is no high-resolution loess pollen record in central Asia. Here， we present the first high-resolution loess pollen record based on 50 pollen samples from a 7.5 m-thick loess section from south-central Kazakhstan to better understand environmental variations and associated vegetation changes from the late Pleistocene to the Holocene in this area. Ⅱ. Physiographic settings
Valikhanov section （43.17o N， 69.33o E， 1000 m elevation） is situated in Valikhanov Valley near City of Zhanatas in south-central Kazakhstan （Fig. 1） and was archaeologically excavated in 2007 by a research team from Kazakhstan National University. The Valikhanov Valley， only ～50 km away from the Moyynkuml Deserts to the north， is presently steppe vegetation dominated by Artemisia， Chenopodiceae， and Gramineae. The landscape in the area is nearly completely mantled with loess deposits. The mean annual precipitation is ～400 mm/yr in this region， while the mean annual January temperature is ～-10℃and the mean annual July temperature is ～22℃.
Ⅲ. Method
Fossil pollen samples were collected at intervals of ～25 cm and totally 50 samples were analyzed. Pollen extraction was carried out according to standard palynological procedures， as follows. Pollen samples were treated with potassium hydroxide， hydrochloric acid， and hydrofluoric acid to remove organic matter， carbonates， and silicates， respectively. One Lycopodium tablet （Lund University， Batch No. 938934） was added before extraction for calculation of pollen concentrations. Pollen types were identified with the aid of pollen identification keys.
Ⅳ. Stratigraphy and chronology
The section is on the second terrace of the Valikhanov River and can be divided into five pedostratigraphic units （Fig. 2）. Unit 1 （0-100 cm） is an accretionary mollisol. Typical mollisols form in semi-arid to semi-humid steppes and have crumbly-structured and organic-enriched A horizon with granularly-structured Bc or/and Bt horizon. An accretionary mollisol is characterized by an over-thickened A horizon without the development of B horizon. It develops when parent material （loess） is being synchronously deposited. Unit 2 （100-200 cm） is a massive loess unit and has no observable marks of biological activities. Unit 3 （200-450 cm） is an accretionary entisol. A typical entisol has a non-structural A horizon and is basically a pedogenically-unaltered parent-material layer. An accretionary entisol has a barely identifiable A horizon with abundant and readily-observable root canals that are filled with organic-enriched sediments or carbonate-enriched sediments. An eolian accretionary entisol develops under conditions of rapid addition of parent material （loess）. In addition， the observed slight gleization suggests that Unit 3 was formed under relatively poorly-drained conditions. Unit 4 （450-600 cm） is a weakly-gleyed （bluish strips and spots） massive loess and it was deposited under poorly drained conditions. Unit 5 （600-750 cm） is a fluvially altered loess unit containing several slightly-laminated and sand-rich layers. Only six AMS dates （Table 1） were obtained and the linear relationship between the dates and the depths has a high correlation coefficient （R2 = 0.933）. Both field-identified pedostratigraphic divisions and laboratory-analyzed proxy data show that the site was under fluvial influences approximately between 28，000 and 22，000 14C yr BP （Unit 5） and under poorly drained eolian deposition conditions between ～22，000 to ～18，000 14C yr BP （Unit 4）. The accretionary entisol （i.e.， Unit 3） seems representing the last deglaciation between ～18，000 and ～10，000 yr 14C yr BP and the observed gleization in this entisol suggests that this entisol was formed under relatively poorly-drained conditions. Unit 2 （a yellowish loess unit） seemed to have deposited during the early Holocene （～5，000 to ～10，000 14C yr BP） under well-drained conditions， and the top soil， an accretionary mollisol （Unit 1）， was formed during the late Holocene （the past ～5，000 years）. The accretionary mollisol （Unit 1） is characterized by relatively high susceptibility values and high organic matter contents. The carbonate leaching toward the bottom of the soil profile is also documented by carbonate concentration at the depth of ～80 cm. It should be noted that the constantly low values in low-frequency magnetic susceptibility and the observed gleization between 250 and 750 cm do suggest that the eolian deposition had occurred primarily under poorly-drained conditions between ～28，000 and ～10，000 14C yr BP.
Ⅴ. Results
We make 3 groups for all pollen. Trees pollen includes Pinus， Picea， Quercus， Cupressus， Ulmus， Betula， Julandaceae and others. Steppe component herbs pollen includes Artemisia， Gramineae， Compositate， Rosaceae， Polygonaceae， Leguminosae， Elaeagnaceae， Urticaceae， Liliaceae， and others. Desert component herbs pollen includes Chenopodiaceae， Ephedra， Nitraria， Zygophyllaceae， and others. Four pollen assemblage zones are recognized from the entire section based on visual inspection and CONISS results （Fig. 3）.
Pollen Zone 1 （750-450 cm； ～28000- ～18000 14C yr BP）， which corresponds with fluvially altered loess and weakly-gleyed loess， is characterized by high pollen percentages of Atermisia. Trees pollen is low with average value ～5%. Atermisia is stable and high （～60%）. Labiatae is 3-10%. Steppe component pollen percentage is 70-90%. Chenopodiaceae is around 5-10%. Desert component pollen is about 15-20%. A/C fluctuates in this zone but remains relatively high. Pollen concentration is low with value 20-30 grains/g. Charcoal concentration is 1000-3000 grains/g. Pollen Zone 2 （450-200 cm； ～18000- ～10000 14C yr BP）， which corresponds with accretionary Entisol， is characterized by stable and diverse pollen assemblage. Trees pollen varies under 10%. Atermisia is stable and high （～60%）. Labiatae is 3-10%. Steppe component pollen percentage is about 80%. Chenopodiaceae is around 6-12%. Desert component pollen is about 20%. A/C is relatively low. Pollen concentration is low with average value ～20 grains/g. Charcoal concentration is 1000-3000 grains/g.
Pollen Zone 3 （200-100 cm； ～10000- ～5000 14C yr BP）， which corresponds with yellowish Loess， is characterized by dramatically variety of Atermisia and Chenopodiaceae. Trees pollen is less than 5%. Atermisia decrease from ～60% to ～30%. Labiatae is 5-10%. Rosaceae is 2-5%. Steppe component pollen percentage increases slightly. Chenopodiaceae fluctuates within 10-30%. Ephdra fluctuate within 2-20%. Desert component pollen is 4-35%. A/C fluctuates dramatically in this zone. Pollen concentration is 25-1200 grains/g. Charcoal concentration is 2000-3000 grains/g.
Pollen Zone 4 （100-0 cm； after ～5000 14C yr BP）， which corresponds with accretionary mollisol， is characterized by high pollen percentages of Chenopodiaceae. Trees pollen increase but still is less than 20%. Atermisia is decrease from ～50% to ～20%. Labiatae is 3-7%. Steppe component pollen percentage decrease sharply. Chenopodiaceae reaches the maximum level of 40%. Ephdra increases sharply from 2% to 40%. Desert component pollen increases from 20% to 60%. A/C fluctuates and reaches the minimum level. Pollen concentration is 40-400grains/g. Charcoal concentration is 3000-7000 grains/g.
Ⅵ. Discussion
28000-18000 14C yr BP （Pollen Zone 1）， low trees pollen means there is almost no tree. Low pollen concentration indicates sparse vegetation， Atermisia pollen dominant this pollen assemblage while others herb pollen are few， it suggests an Atermisia arid steppe vegetation. When loess deposition rate was higher than soil-forming rate under wet conditions， a gleyed loess resulted. When loess deposition rate was lower than soil-forming rate under wet conditions， a gleyed entisol resulted. The fluvially altered loess and weakly-gleyed loess suggests that climate was not dry. Considering of summer insolation in the Northern Hemisphere is relatively low at this period， this vegetation maybe demonstrates a cold and slight wet climate prevailed in this region during 28000-18000 14C yr BP.
18000-10000 14C yr BP （Pollen Zone 2）， low trees pollen means there is almost no tree. Atermisia pollen still dominant the pollen assemblage while others herb pollen are few. Pollen and charcoal concentrations are low. However， the accretionary Entisol suggests that climate was a little wetter. We propose this vegetation is Atermisia steppe. Considering of massive ice sheet exist in the Northern Hemisphere at this period， the vegetation demonstrates a cold and slight wet climate prevailed in this region during 18000-10000 14C yr BP. 10000-5000 14C yr BP （Pollen Zone 3）， lower trees pollen indicates there is no tree at that time. Atermisia pollen decrease while Chenopodiaceae and Ephdra pollen increases in this pollen assemblage， it suggests Atermisia-Chenopodiaceae desert steppe vegetation. Dramatically variety of pollen assemblage indicates climate fluctuates rapidly. Considering of peak summer insolation has melt much ice sheet in the Northern Hemisphere， this vegetation demonstrates a dry and mild climate prevailed in this region during middle Holocene.
After 5000 14C yr BP （Pollen Zone 4）， few trees pollen indicates there are rare trees at that time. Atermisia pollen continues to decrease while Chenopodiaceae and Ephdra pollen increase in this pollen assemblage. Pollen concentration increase much. It suggests Chenopodiaceae-Atermisia steppe vegetation. High charcoal concentrations in this zone are also consistent with dense biomass， which provided more fuels for fires to burn. This vegetation demonstrates a wet climate prevailed in this region during late Holocene.
Ⅶ. Conclusion
The vegetation probably was Artemisia arid steppe during ～28，000 - ～18，000 14C yr BP.， then turned to be Artemisia steppe during ～18，000 - ～10，000 14C yr BP， an Atermisia-Chenopodiaceae desert steppe during ～10，000 - ～5，000 14C yr BP and a Chenopodiaceae-Artemisia steppe after ～5，000 14C yr BP in this region. Climate is cold and slightly wet during 28，000-10，000 14C yr BP.， dry and mild during 10，000-5，000 14C yr BP.， wet after ～5，000 14C yr BP in this region. Climate in the Central Asian domain was constantly wetter than that in the East Asian domain during MIS2. The Central Asian domain experienced a wetter climate in late Holocene whereas the East Asian domain expereinced a wetter climate in early Holocene.
Acknowledgement
The author thank： Chinese Nature Science Foundation for the grant （No. 40671190）； Guizhou provincial science and Technology for the grant （2013-2273）； Thanks to Prof. Kambiu Liu and Prof. Zhaodong Feng！
References：
[1]Bush，A.B.，2005.“CO2/H2O and orbiotally driven climate variability over central Asia through the Holocene”.Quaternary International，136，15-23.
【Key words】pollen； MIS2； Holocene； Central Asia； climate
Ⅰ. Introduction
MIS2 and Holocene is an important timeframe for understanding climate change and climate variability because， for the most part， the climatic boundary conditions are similar to those experienced now and possibly in the near-future. However， the temporal and spatial patterns of the climatic variability are poorly constrained regionally. In particular， we lack well-dated and high-resolution records from continental interiors. Central Asia is one of such areas， having been shown to display extreme sensitivity to past large-scale force-driven changes. It is a climatic transitional zone between the westerly-dominated “European” system to the west and the monsoon-dominated “East Asian” system to the east. During the Holocene and the last glacial/interglacial cycle， the “European” climate has primarily alternated between a cool/wet mode and a warm/dry mode， whereas the “East Asian” climate has alternated between a warm/wet mode and a cold/dry mode. Pollen data have been frequently used in vegetation and climate reconstructions， as they tend to reflect vegetation and climate changes at a regional scale. However， there is no high-resolution loess pollen record in central Asia. Here， we present the first high-resolution loess pollen record based on 50 pollen samples from a 7.5 m-thick loess section from south-central Kazakhstan to better understand environmental variations and associated vegetation changes from the late Pleistocene to the Holocene in this area. Ⅱ. Physiographic settings
Valikhanov section （43.17o N， 69.33o E， 1000 m elevation） is situated in Valikhanov Valley near City of Zhanatas in south-central Kazakhstan （Fig. 1） and was archaeologically excavated in 2007 by a research team from Kazakhstan National University. The Valikhanov Valley， only ～50 km away from the Moyynkuml Deserts to the north， is presently steppe vegetation dominated by Artemisia， Chenopodiceae， and Gramineae. The landscape in the area is nearly completely mantled with loess deposits. The mean annual precipitation is ～400 mm/yr in this region， while the mean annual January temperature is ～-10℃and the mean annual July temperature is ～22℃.
Ⅲ. Method
Fossil pollen samples were collected at intervals of ～25 cm and totally 50 samples were analyzed. Pollen extraction was carried out according to standard palynological procedures， as follows. Pollen samples were treated with potassium hydroxide， hydrochloric acid， and hydrofluoric acid to remove organic matter， carbonates， and silicates， respectively. One Lycopodium tablet （Lund University， Batch No. 938934） was added before extraction for calculation of pollen concentrations. Pollen types were identified with the aid of pollen identification keys.
Ⅳ. Stratigraphy and chronology
The section is on the second terrace of the Valikhanov River and can be divided into five pedostratigraphic units （Fig. 2）. Unit 1 （0-100 cm） is an accretionary mollisol. Typical mollisols form in semi-arid to semi-humid steppes and have crumbly-structured and organic-enriched A horizon with granularly-structured Bc or/and Bt horizon. An accretionary mollisol is characterized by an over-thickened A horizon without the development of B horizon. It develops when parent material （loess） is being synchronously deposited. Unit 2 （100-200 cm） is a massive loess unit and has no observable marks of biological activities. Unit 3 （200-450 cm） is an accretionary entisol. A typical entisol has a non-structural A horizon and is basically a pedogenically-unaltered parent-material layer. An accretionary entisol has a barely identifiable A horizon with abundant and readily-observable root canals that are filled with organic-enriched sediments or carbonate-enriched sediments. An eolian accretionary entisol develops under conditions of rapid addition of parent material （loess）. In addition， the observed slight gleization suggests that Unit 3 was formed under relatively poorly-drained conditions. Unit 4 （450-600 cm） is a weakly-gleyed （bluish strips and spots） massive loess and it was deposited under poorly drained conditions. Unit 5 （600-750 cm） is a fluvially altered loess unit containing several slightly-laminated and sand-rich layers. Only six AMS dates （Table 1） were obtained and the linear relationship between the dates and the depths has a high correlation coefficient （R2 = 0.933）. Both field-identified pedostratigraphic divisions and laboratory-analyzed proxy data show that the site was under fluvial influences approximately between 28，000 and 22，000 14C yr BP （Unit 5） and under poorly drained eolian deposition conditions between ～22，000 to ～18，000 14C yr BP （Unit 4）. The accretionary entisol （i.e.， Unit 3） seems representing the last deglaciation between ～18，000 and ～10，000 yr 14C yr BP and the observed gleization in this entisol suggests that this entisol was formed under relatively poorly-drained conditions. Unit 2 （a yellowish loess unit） seemed to have deposited during the early Holocene （～5，000 to ～10，000 14C yr BP） under well-drained conditions， and the top soil， an accretionary mollisol （Unit 1）， was formed during the late Holocene （the past ～5，000 years）. The accretionary mollisol （Unit 1） is characterized by relatively high susceptibility values and high organic matter contents. The carbonate leaching toward the bottom of the soil profile is also documented by carbonate concentration at the depth of ～80 cm. It should be noted that the constantly low values in low-frequency magnetic susceptibility and the observed gleization between 250 and 750 cm do suggest that the eolian deposition had occurred primarily under poorly-drained conditions between ～28，000 and ～10，000 14C yr BP.
Ⅴ. Results
We make 3 groups for all pollen. Trees pollen includes Pinus， Picea， Quercus， Cupressus， Ulmus， Betula， Julandaceae and others. Steppe component herbs pollen includes Artemisia， Gramineae， Compositate， Rosaceae， Polygonaceae， Leguminosae， Elaeagnaceae， Urticaceae， Liliaceae， and others. Desert component herbs pollen includes Chenopodiaceae， Ephedra， Nitraria， Zygophyllaceae， and others. Four pollen assemblage zones are recognized from the entire section based on visual inspection and CONISS results （Fig. 3）.
Pollen Zone 1 （750-450 cm； ～28000- ～18000 14C yr BP）， which corresponds with fluvially altered loess and weakly-gleyed loess， is characterized by high pollen percentages of Atermisia. Trees pollen is low with average value ～5%. Atermisia is stable and high （～60%）. Labiatae is 3-10%. Steppe component pollen percentage is 70-90%. Chenopodiaceae is around 5-10%. Desert component pollen is about 15-20%. A/C fluctuates in this zone but remains relatively high. Pollen concentration is low with value 20-30 grains/g. Charcoal concentration is 1000-3000 grains/g. Pollen Zone 2 （450-200 cm； ～18000- ～10000 14C yr BP）， which corresponds with accretionary Entisol， is characterized by stable and diverse pollen assemblage. Trees pollen varies under 10%. Atermisia is stable and high （～60%）. Labiatae is 3-10%. Steppe component pollen percentage is about 80%. Chenopodiaceae is around 6-12%. Desert component pollen is about 20%. A/C is relatively low. Pollen concentration is low with average value ～20 grains/g. Charcoal concentration is 1000-3000 grains/g.
Pollen Zone 3 （200-100 cm； ～10000- ～5000 14C yr BP）， which corresponds with yellowish Loess， is characterized by dramatically variety of Atermisia and Chenopodiaceae. Trees pollen is less than 5%. Atermisia decrease from ～60% to ～30%. Labiatae is 5-10%. Rosaceae is 2-5%. Steppe component pollen percentage increases slightly. Chenopodiaceae fluctuates within 10-30%. Ephdra fluctuate within 2-20%. Desert component pollen is 4-35%. A/C fluctuates dramatically in this zone. Pollen concentration is 25-1200 grains/g. Charcoal concentration is 2000-3000 grains/g.
Pollen Zone 4 （100-0 cm； after ～5000 14C yr BP）， which corresponds with accretionary mollisol， is characterized by high pollen percentages of Chenopodiaceae. Trees pollen increase but still is less than 20%. Atermisia is decrease from ～50% to ～20%. Labiatae is 3-7%. Steppe component pollen percentage decrease sharply. Chenopodiaceae reaches the maximum level of 40%. Ephdra increases sharply from 2% to 40%. Desert component pollen increases from 20% to 60%. A/C fluctuates and reaches the minimum level. Pollen concentration is 40-400grains/g. Charcoal concentration is 3000-7000 grains/g.
Ⅵ. Discussion
28000-18000 14C yr BP （Pollen Zone 1）， low trees pollen means there is almost no tree. Low pollen concentration indicates sparse vegetation， Atermisia pollen dominant this pollen assemblage while others herb pollen are few， it suggests an Atermisia arid steppe vegetation. When loess deposition rate was higher than soil-forming rate under wet conditions， a gleyed loess resulted. When loess deposition rate was lower than soil-forming rate under wet conditions， a gleyed entisol resulted. The fluvially altered loess and weakly-gleyed loess suggests that climate was not dry. Considering of summer insolation in the Northern Hemisphere is relatively low at this period， this vegetation maybe demonstrates a cold and slight wet climate prevailed in this region during 28000-18000 14C yr BP.
18000-10000 14C yr BP （Pollen Zone 2）， low trees pollen means there is almost no tree. Atermisia pollen still dominant the pollen assemblage while others herb pollen are few. Pollen and charcoal concentrations are low. However， the accretionary Entisol suggests that climate was a little wetter. We propose this vegetation is Atermisia steppe. Considering of massive ice sheet exist in the Northern Hemisphere at this period， the vegetation demonstrates a cold and slight wet climate prevailed in this region during 18000-10000 14C yr BP. 10000-5000 14C yr BP （Pollen Zone 3）， lower trees pollen indicates there is no tree at that time. Atermisia pollen decrease while Chenopodiaceae and Ephdra pollen increases in this pollen assemblage， it suggests Atermisia-Chenopodiaceae desert steppe vegetation. Dramatically variety of pollen assemblage indicates climate fluctuates rapidly. Considering of peak summer insolation has melt much ice sheet in the Northern Hemisphere， this vegetation demonstrates a dry and mild climate prevailed in this region during middle Holocene.
After 5000 14C yr BP （Pollen Zone 4）， few trees pollen indicates there are rare trees at that time. Atermisia pollen continues to decrease while Chenopodiaceae and Ephdra pollen increase in this pollen assemblage. Pollen concentration increase much. It suggests Chenopodiaceae-Atermisia steppe vegetation. High charcoal concentrations in this zone are also consistent with dense biomass， which provided more fuels for fires to burn. This vegetation demonstrates a wet climate prevailed in this region during late Holocene.
Ⅶ. Conclusion
The vegetation probably was Artemisia arid steppe during ～28，000 - ～18，000 14C yr BP.， then turned to be Artemisia steppe during ～18，000 - ～10，000 14C yr BP， an Atermisia-Chenopodiaceae desert steppe during ～10，000 - ～5，000 14C yr BP and a Chenopodiaceae-Artemisia steppe after ～5，000 14C yr BP in this region. Climate is cold and slightly wet during 28，000-10，000 14C yr BP.， dry and mild during 10，000-5，000 14C yr BP.， wet after ～5，000 14C yr BP in this region. Climate in the Central Asian domain was constantly wetter than that in the East Asian domain during MIS2. The Central Asian domain experienced a wetter climate in late Holocene whereas the East Asian domain expereinced a wetter climate in early Holocene.
Acknowledgement
The author thank： Chinese Nature Science Foundation for the grant （No. 40671190）； Guizhou provincial science and Technology for the grant （2013-2273）； Thanks to Prof. Kambiu Liu and Prof. Zhaodong Feng！
References：
[1]Bush，A.B.，2005.“CO2/H2O and orbiotally driven climate variability over central Asia through the Holocene”.Quaternary International，136，15-23.