The Geomorphology Of The Typical Rivers In The Eastern Tibet And Their Active Tectonic Implications

Under the control from the collision of Indian plate with Eurasian plate, the eASTERnedge of Tibetan Plateau rises significantly. The overall trend of the tectonicgeomorphology in this region is the steep dipping of altitude from the inner part ofplateau to Sichuan Basin which is believed to be caused by long-term erosion andstacking. It is the research of retrieving regional structural activities from thestructural and geometric features of tectonic geomorphology that is considered as amajor direction of relative study. Among these factors, fluvial geomorphology isattained great attention since its close relation with tectonic activity during its creationand evolution process. Tectonic geomorphologic indexes are instrumental inquantifying the accumulated influence of tectonic activities on the fluvial areas. Theincreasing spatial resolution of digital elevation model promotes the quantificationand semi-quantification in tectonic geomorphology, which adds new thoughts intotectonic geomorphology research.EASTERn edge of the Tibetan Plateau is selected asthe study area of this dissertation. Using the high-resolution DEM derived withremote sensing images, a new algorithm for retrieving river terrace based on fuzzyclustering and spatial auto-correlation is designed and implemented. This techniquecan be used to calculate the geographic index of river terrace which is followed byaccuracy evaluation. A combination of geomorphological index extracted with thismethod and data collected in field study is applied in the estimation of rising process,during which some prior published ago-dating is also taken as reference. Then, somediscussions on the indicative significance of these indexes are performed in thisdissertation. Some key results of this dissertation are listed as below:(1) The tectonic geomorphologic indexes of the eASTERn edge of the TibetanPlatueau and the corresponding indicating significance.The different fluvial geomorphology in three drainage areas of Minjiang, Fujiang andBailongjiang suggests the distinct tectonic activities in the study area. Based on ASTER DEM, three geomorphologic indexes are retrieved. The first one isHypsometric Integral which is commonly used to indicate the stage ofgeomorphological erosion cycle; the second is SL index which is always used todescribe the local variance of gradient; the third is Hack profile, which is oftenapplied in the depict of rising process; the forth is fluvial profile which also serves asa tool to characterize the rising process. Minjiang area and Bailongjiang area are onthe relative earlier stage of geomorphological erosion cycle, between whichMingjiang area is younger. On the other hand, Fujiang area is on the relatively laterstage of this cycle. The higher HI of upstream area of Minjiang area indicates that thisbranch area experiences more tense tectonic activities leading to its higher risingspeed. SL index represent the variance of the local gradient of fluvial profile. Theratio of Minjiang having a relatively high SL is higher than the other two branches asshown in the experiments. This phenomenon can be attributed to various reasons indifferent fragments. For example, in the upstream are of Mingjiang area, theHongqiaoguan to Jiaochanghe fragment, the corresponding higher SL is caused by thestrong tectonic activities of Songpinggou fault.(2) The deformation and its causes of fluvial terrace in the upstream area of MinjiangTo further analyse the tectonic activity in Hongqiaoguan-Jiaochang fragment,field study extent was extended in both upstream and downstream directions. Afterinvestigating the dispersion of terrace, altitude relative to river and the sedimentarystructure in Zhanglacun-Maoxian fragment, a four-grade-terrace phenomenon is foundin this area. The terraces with larger extent and higher continuity are used for theanalysis in following steps. Taking prior published age dating data as reference, thisdissertation discuss the transformation of fluvial profile since Quaternary.T4terrace,180~430m-height relative to river surface,experienced rising movementin early-middle Pleistocene. Since late Pleistocene, influenced by the strong tectonicactivities, the formation of T2and T3had begun. T2,20m to40mheightrelative toriver surface, was well developed along the river above Shawan village. At thatmoment, TheStrong fault activitiesof Songpingou Fault led to T4’s deformation, theheight in Tuanjie village raised to430m.Also, faulting caused knickpoint.Leap watererosion made mostly by knickpoint stopped the formation of T2and T3in the hanging wall, from Tuanjie village to Manaoding village. All of the phenomenon could beascribed to the Songpinggou Fault, which location may right be consistent with spurtT4and Slightly down to the knickpoint.（3）A Clustering Approach for Incorporationg Spatial Dependence into theAutomatic Fluvial Terrace Extraction from Digital Elevation Model: A Case Studyfrom the Zhangla Basin along the Upstream of the Minjiang RiverThe terraces along the upper reaches of the Minjiang River record rapid uplift aroundthe Tibetan Plateau since the Quaternary. However, the common method to obtain theterraces elevation data always relies on single point or line to represent the wholelandform. The available results and further analyses are usually not continuous andsystematic. All of these, therefore, restricts the development of further study. A largenumber of qualitative information and the digital terrace models are relied on theimprovement of DEM processing technology. Our present paper applies fuzzy cmeans algorithm to a bunch of the Minjiang River terrace cross sections in differentdimensions to get the integrated distribution characteristics of the terraces in theZhangla basin. SPOT5image was adopted when generating the DEM. Then weconsider both the spatial correlation and distribution characteristics of terrace crosssections into analysis and we consequently find a successful way to extract differentlevels of terraces based on both whole indicator and internal correlation. The precisionevaluation suggests that result of terrace extraction is highly consistent with the fieldsurvey data.