Comprehensive Investigations On Joint Roughness Quantification,Rock Mass Statistical Homogeneity,and Rock Slope Stability

In this dissertation,comprehensive investigations have been conducted on joint roughness quantification,rock mass statistical homogeneity,and rock slope stability.First,several 2-D and 3-D slope and/or amplitude statistical roughness metrics as well as fractal roughness metrics computed using the 2-D variogram,the 2-D roughness length,and the 3-D modified divider methods are used to investigate some important factors that could influence the accurate quantification of rock joint roughness.Also,the capability of a new proposed directional 3-D roughness metric,(?)_A+(the mean positive apparent dip in the shear direction),to accurately quantify roughness is evaluated.Second,modified versions of three different methods(polar equal area pole plot comparison,linear correlation coefficient analysis,and contingency table analysis)are used to investigate rock mass statistical homogeneity using over 5800 orientation data;this highly unique set of data was obtained from the Co To-Than Lan islands,and along a 15 km stretch of the 3B highway in North East Vietnam.Third,the stability of the rock slopes along the 3B Highway in North East Vietnam are investigated using kinematic analysis.The results of the joint roughness investigations show that,in the exception of the computed amplitude parameter,non-stationarity resulting from a global linear trend does not have an influence on all the investigated roughness parameters.Also,the results show that the suggested 3-D roughness parameter,(?)_A+,is capable of accurately quantifying the roughness of a joint surface.The results of the joint size and heterogeneity effect investigations show that depending on the roughness heterogeneity,it is possible to have either the negative or positive scale effect.In addition,a significant sampling interval effect was found on all the statistical roughness parameters used for the research.It therefore follows that heterogeneity and sampling interval effects are the main reasons for the conflicting findings on scale effects that have appeared in the literature over the past 40years.As such,there should not be any rule for scale effects on joint roughness as suggested in the literature.The studied rock joint surface also exhibited significant anisotropy on the computed roughness parameters.The statistical homogeneity investigation results show that no strongly reliable objective method is available at present in the engineering geology or rock mechanics literature to firmly demarcate between the statistical homogeneity and non-homogeneity of rock masses.Each of the available methods can only be used to rank rock masses with respect to the strength of statistical homogeneity.This finding clearly indicates that decision making on statistical homogeneity of rock masses is a difficult task using results coming from just one methodology.For both the rock masses on the Co To and Thanh Lan islands,and along the3B Highway,findings from this dissertation indicate that comparison of results coming from several methods allow one to make better judgement on rock mass statistical homogeneity.These investigations presented unique case studies that have illustrated how to make decisions under significant uncertainty using the results coming from three different methods on real world rock mass statistical homogeneity.The rock slope stability investigation results show that significant risks of plane,wedge and toppling failures exist along the 3B Highway.Considering all basic failure modes(including toppling),the dominant mode of failure at 15 survey stations is toppling failure.12 stations have wedge sliding as the dominant failure mode while only one station has plane sliding as the dominant mode of failure.Toppling has a significant effect on the final Maximum Safe Slope Angle(MSSA)values for the slopes along the 3B Highway.Considering all survey stations along the highway,MSSA values range from 25.2°-57.7°(including toppling),and from 25.2°-70°(excluding toppling).For some of the survey stations,MSSA values are considerably higher when toppling is excluded compared to when toppling is included.This occurrence can be attributed to the existence of high angle discontinuities at the various survey stations along the 3B Highway.It is important to note that no comprehensive kinematic analysis such as the one conducted in this dissertation exists for the investigated rock slopes along the 3B Highway.The results of this detailed investigation therefore provide a solid engineering guide for mitigating rock slope failures along the 3B Highway.