Laserfiche WebLink
Table 1: Summary of Scanline Survey Data <br />Scanline <br />(length in ft) <br />Joint <br />Set <br />Number of <br />Joints <br />Apparent Spacing <br />(ft) <br />Alpha <br />(degrees) <br />Terzaghi <br />Corrected <br />Spacing (ft) <br />1 (600) <br />S l <br />51 <br />11.8 <br />36 <br />9.5 <br />S3 <br />28 <br />21.4 <br />76 <br />5.2 <br />2 (560) <br />S l <br />19 <br />29.5 <br />77 <br />6.6 <br />S3 <br />54 <br />10.4 <br />20 <br />9.7 <br />S4 <br />42 <br />13.3 <br />15 <br />12.9 <br />3 (440) <br />S 1 <br />14 <br />31.4 <br />75 <br />8.1 <br />S3 <br />59 <br />7.5 <br />20 <br />7.0 <br />S4 <br />52 <br />8.5 <br />15 <br />8.2 <br />4 (420) <br />Sl <br />18 <br />23.3 <br />75 <br />67 <br />S3 <br />49 <br />8.6 <br />20 <br />8.1 <br />S4 <br />39 <br />10.8 <br />15 <br />10.4 <br />Using true spacing, all three -set block- forming <br />combinations of joint sets S1, S2, S3 and S4 were <br />analyzed to obtain the frequency distribution of <br />block sizes. Experience has shown that many <br />discontinuity sets have a spacing distribution that <br />approximates reasonably to a negative exponential <br />form (Priest and Hudson, 1976, Priest and Hudson, <br />1981), and this was assumed to apply in this study. <br />The negative exponential distribution is a single - <br />parameter distribution that is completely defined by <br />its mean, so provided the mean spacing is known, <br />the full range of probabilities can be estimated. <br />Analysis of the bedding plane data indicated a <br />lognormal spacing distribution. Using these distribu- <br />tions, the potential discontinuity combinations and <br />resulting block sizes were analyzed using <br />probabilistic modeling software. <br />The results of this analysis, in the form of a <br />cumulative probability distribution, are presented in <br />Figure 2, together with the cumulative distribution of <br />block sizes from the boulder catalog. The results are <br />in good agreement, and converge well in terms of <br />the larger block sizes. Based on this analysis, a worst <br />case block size of about 6000 cubic feet was <br />considered appropriate for Phase I rockfall hazard <br />assessment. An 'equivalent' disk, with a diameter of <br />26 ft, a thickness of 11 ft, and a volume of 5840 cu. <br />ft. was used in the CRSP3 runs. <br />Prior to the main hazard assessment, a number of <br />trial runs were carried out to calibrate the CRSP3 <br />model, and select appropriate site - specific slope <br />surface parameters. These calibration runs were <br />carried out on sections of the slope where previous <br />rockfalls provided a basis for comparing observed <br />and predicted behavior. <br />Typical block sizes were determined from the <br />boulder catalog for each calibration section. These <br />block sizes were converted to equivalent disks or <br />cylinders for input to the CRSP3 model, by setting <br />the diameter equal to the average of the two most <br />similar dimensions, and the thickness or lengths <br />equal to the remaining dimension. <br />2.3 Crsp3 Parameters <br />CRSP3 requires that the surface of the modeled <br />profile be described in terms of tangential and <br />normal (restitution) coefficients. Tables 2 and 3 <br />show the relationship between the surface descrip- <br />tions presented by Pfeiffer and Bowen (1989) for <br />assigning tangential and normal (restitution) <br />coefficients, and inferred equivalent site conditions. <br />Table 4 indicates the ranges for these parameters <br />considered appropriate at this site. <br />