Phytolith Research As An Indicator Of Paleo-climate In The Late-Holoncene Peat Deposits Of Northeast China

With the development of PAGES, the paleo-climatic and paleo-environmentalreconstruction has become one of the most productive research projects. In this project, it isimportant to choose and study some geological records and climatic indicators. In this thesis,primary researches from different points of view are carried on the application potential ofphytolith in paleo-climate reconstruction of peat deposits, using the Holocene peat depositsas information records and phytolith as a new climatic indicator.The peat deposits of Northeast China are mainly developed in Songliao basin area. Thegeological environment of long term depression and humid climate provide favorableconditions for peat developing in Sanjiang Plain, Songnen Plain, Dahinganling &Xiaohinganling Moutains, Changbai Mountains, etc. Many researches of pollen, clayminerals and plant relicts etc. have been done to reconstruct paleoclimate of the peatdeposits in this area. All the past achievements show that the peat deposits of NortheastChina are palying a more and more important role in Holocene paleo-climatic reconstructionof Quaternary environmental research.According to the surface annual and monthly climatic data from Chinese internationalexchange stations, some isograms were drawn (including annual temperature, annualprecipitation, annual relative humidity, annual wind speed in 2 seconds, annual sunlighthours, annual air pressure in the period of 1971~2002yr and monthly temperature, monthlyprecipitation, monthly relative humidity, monthly wind speed in 2 seconds, monthly sunlighthours, monthly air pressure from 1971~2002yr). On the basis of these isograms, we collectpeat surface soils from 41 locations of Manzhouli, Yakeshi, Mangui, Mohe, Tahe, Jiagedaqi,Fuyu, Hailun, Tieli, Hegang, Tangwanghe, Fujin, Yushu, Fangzheng, Qitaihe, Muling,Mishan, Mudanjiang, Wangqing, Antu, Tumen, Dunhua, Huadian, Jinchuan, Fusong,Northern slope of Changbai Mountain, Baishan, Xinbin, Kuandian, Huanren, Dandong,Zhuanghe, Gaizhou, Panjin, Anshan, Fengku, Xibaihua, Kailu, Zhangwu, Qian'an and Da'an.So that the author can extract phytoliths from the samples and analyse the relationshipbetween phytolith assemblages and climatic parameters. Most of these sites are dispersedvertically to the climatic parameter isolines in the climate parameter isograms of the studiedarea.Phytoliths are extracted from the peat through heavy liquid seperation after wiping offthe humic and organic. According the morphological classification, together with manypointviews and the actual conditions of peat bogs, the phytoliths in this thesis are dividedinto 11 types by the author: Dumbbell-shaped, fan-shaped, square-shaped,rectangular-shaped, trapezium-shaped, saddled, crenate-shaped, cap-shaped, point-shaped,elongate, and other shapes on the basis of Wang and Lv's classifacion and the ICPN1.0.Besides the single tpes of square-shaped, rectangualar-shaped, trapezium-shaped andsaddle-shaped phytolith and some unidentified shapes, dumbbell-shaped phytolith includes 2kinds, fan-shaped, crenate-shaped and elongate-shaped phytolith all include 5 kinds,cap-shaped and point-shaped phytolith both includes 3 kinds.The distribution maps of the 11-types phytoliths are made based on the phytolithidentification. The maps show that the latitude, topography, temperature and heat conditionsare controlling factors to the distribution of phytolith assemblages in the peat surface soils.There is certain corresponding relationships between the phytolith distribution and meantemperature isolines, mean precipitation isolines and mean relative humidity isolines. Withthe latitude rising, the contents of cap-shaped, point-shaped, crenate-shaped andelongate-shaped phytoliths increase and the contents of dumbbell-shaped, saddle-shaped,fan-shaped, square-shaped, rectangular shaped, and trapezium-shaped phytoliths decrease. Ina similar latitude area, the contents of cap-shaped, cap-shaped, point-shaped, crenate-shapedand elongate-shaped phytoliths are lower in plain area than those in mountain area, whereasthe contents of dumbbell-shaped, saddle-shaped, fan-shaped, square-shaped, rectangularshaped, and trapezium-shaped phytoliths are higher in plain area than those in mountainarea.To qualify the relationship between the phytolith assemblages and climate changing,canonical correlation analysis and clustering analysis were adopted to study the correlativitybetween the phytolith assemblages and climatic parameters. The results shows that,Fan-shaped phytolith is corresponding to warm and wet climate;crenate-shaped,elongate-shaped, and point-shaped phytolith are corresponding to cooler and drier climate.Relationships between the monthly climate parameters and the phytolith assemblages arenarrated as the following: there is no evident relationship between the phytolith assemblagesand climate in winter;there is obvious correlation between the mean temperature andcrenate-shaped, elongate-shaped and point-shaped phytolith assemblages in spring;there isevident relationship between the mean temperature and crenate-shaped and elongate-shapedphytolith in summer, and in autumn the relation between the mean temperature andfan-shaped phytolith is obvious. Synthesizing the above results, it is concluded that the meantemperature and mean relative humidity influences phytolith assemblages obviously, meanprecipitation and mean wind speed take the second place.R factor analysis on the phytolith assemblages from both peat surface and peat profileis conducted to extract the dominating factors in the explanation of relativity betweenoriginal variables. 7 factors (F1, F2, F3, F4, F5, F6, F7) are extracted after calculation, with thededicative ratio of 93.2%. With the annual mean temperature, annual mean precipitation,annual mean relative humidity and annual mean wind speed;phytolith-climatic transfermodel is accomplished after stepwise regression:Y1= 22.741+8.254X1+8.647X2+9.507X6;Y2=5252.048+635.991X1-326.642X6;Y3=65.694-2.336 X6;Y4=29.458-2.834 X3。Y1---annual mean temperature, Y2---annual mean precipitation, Y3---anuual mean relativehumidity, Y4---annual mean wind speed in 2 seconds;X1,X2,X3 and X6 are representativefactors selected after stepwise regression.Thus, the validity of the model was tested during the late Holocene paleo-climaticreconstruction in the peat profiles of Jixiang and Yushu in Dunhua Basin. At the same time,Morlet time-frequencey characteristics of the paleo-climate reflected by phytolith-climatetransfer model are analyzed and shows its potential advantages in the time-frequency studyof paleo-climate. When scale a≥30, Paleo-climate changes in profiles of Jixiang and Yushuare consistent, when scale a≤20, there is a little difference between different climaticparameters in the same profile, but the time-frequency of same parameters in two profilesare the same. The result also shows that higher precision and differentiation of paleo-climatein peat deposits can be achieved through phytolith ananlysis. Connect with the Morlettime-frequency analysis and historical recordation, the deductive results of the modeldivided the Late Holocene paleo-climate process into 6 stages: warm stage---coolstage---temperate and wet stage---cooling stage---warm and wet stage---cooling stage and14 sub-stages.At the same time, to testify the precision and differentiation of the model, total organiccarbon (TOC), geochemical elements and plant relicts in Jixiang peat profile are also studiedwith the phytolith analysis. Some conclusions are made as: ① Plaeo-climate precisiondeduced from the phytolith-climate transfer model is higher than that of TOC. Thetransform of organic is controlled by many factors including temperature and humidity, sothat its precision isn't as high as the model which only shows temperature or humidity. ②Result from the analysis of phytolith-climate transfer model is more credible than that ofelements, because the geochemistry study to deduce paleo-climate and paleo-environment isjust in its initial stage and geochemical behaviors of different elements are not clear. ③Compared with the plant relicts analysis, phytolith ananlysis will deduce the paleo-climateand paleo-environment more precisely. Because of the quantitative limits, it is difficult toreconstruct the plant panorama of the peat profile. While the stable physical geochemicalcharacteristics of phytoliths can be used in paleo-environment reconstruction. All the aboveanalysis not only tests the accuracy of the model, but also confirms the important role ofphytolith analysis in paleo-environment reconstruction in peat deposits.Finally, the results of the phytolith analysis are compared with C-O isotopes, pollen andclay mineral ananlysis. At the same time the influence factors of paleo-climate reflected bythe phytolith-climate transfer model were are analyzed. It is shown that, C-O shows highprecision in paleo-climate reconstruction except the deficiency in the vegetationreconstruction. Pollen ananlysis shows its advantage in the paleo-climate reconstruction of alarge area, while the phytolith analysis shows its adavantage in a smaller area. Clay mineralsare sensitive to environment changes except the disturbing of some autogenetic andhysterogenic minerals. On the mean temperature charts of paleo-climate deduced fromphytolith-climate transfer model, there are obvious responses of 8 solar activities and 3volcanic activities in the past 3000yrs. All these analysis show the great potential ofphytolith analysis in the paleo-climate research.