The Extraction Of Fracture Surfaces Based On Terrestrial Laser Scanning And The Clustering Analysis Of Tectonic Paleo Shear Strain Tensors

Stresses in the Earth's crust are affected by tectonic loading processes;while crustal stresses are also the immediate driving forces of tectonic deformations.The brittle tectonic deformations,to a large extent,are in the form of developing fracture/fault systems in the upper crust.Regional studies of the development of fracture/fault systems and their corresponding spatial and tempo?ral variation of crustal tectonic strain can therefore provide insight into the complex interactions between fracture/fault systems,tectonic paleo-strian and crustal stresses.Previous studies of fractures at laboratory rock sample scale examined the relationships be?tween many factors and the development of fractures,and established constitutive models based on statistical or physical processes.However,it's difficult to generalize them in fracture-strain problems at larger scale with complex geological background.On the other hand,many enlight-ening results have been obtained from the study of fractures at outcrop scale by reasoning based on observations,geological background and geological knowledge.However,compared with the complex geological background,the data used for analysis may be relatively limited to be statis-tically representative or too qualitative,which may leads to the weak foundation of the reasoning results.In order to overcome the above drawbacks,this dissertation presents innovative methods for fracture system and tectonic strain research based on high-precision and high-resolution outcrop laser scanning data.The methods can be sequentially divided into the following three topics.The primary problem of doing fracture system research on outcrop point cloud data is to build the digital fracture model,i.e.,the automatic extraction of fracture surfaces from the point cloud data.In this dissertation,we propose an algorithm using a region-growing approach for the au-tomatic extraction of the full extent of individual fractures from outcrop terrestrial laser scanning data.Compared with manually acquired field survey data,our method obtained substantially greater amount of and better-quality fracture data.In the digital fracture model,the shear deformations on each rock fracture surface is the key part of understanding the development of fracture systems.In this dissertation,we propose a quan-titative method to derive historical shear deformations of rock fractures from fracture surface data based on the analysis of effects of indicating structures(quasi fault striations and quasi fault steps)on the she.ar strength parameter of the fracture surface.The validity of the proposed method was proved by testing it on a constructed fracture surface with striations and an outcrop fracture surface with clear fault steps.The application of this method on an example outcrop shows an intuitive idea of how the rock mass was deformed and that the distribution,occurrence and mode of new frac-tures are strictly controlled by preexisting fractures,i.e.,the historical slip data on fracture surfaces obtained by this method has broad prospects of applications.Finally,we propose a clustering analysis based method that utilizes large population of frac-ture historical slip data to do the tectonic strain analysis.Firstly,the Lagrangian strain tensors corresponding to historical shear slip on each fracture surface are calculated,and then the principal shortening directions of these strain tensors are clustered to obtain the tectonic shortening direc-tions experienced by the outcrop.The results of the applications of the method on 21 outcrops in the study area of this dissertation show two main tectonic shortening events.From this,the devel-opment pattern of fracture systems in these two tectonic shortening events can be analyzed,and the spatial variation of the degree of shear deformations on fracture surfaces in shortening events provides an important basis for the reconstruction of tectonic paleo strain.