Item

Computer simulation of rockfalls - application to rockfalls at Fox Glacier, West Coast, New Zealand

Rayudu, Durga N.
Date
1997
Type
Thesis
Fields of Research
Abstract
This thesis reviews computer simulation of rockfalls in general and an attempt has been made to analyse and predict rockfalls at Fox Glacier, West Coast, New Zealand, using rockfall simulation programs. A comprehensive comparison of five rockfall simulation programs was carried out to help decide upon which program is the best to use for a detailed analysis of rockfalls. It was found from the comparison that the program Rockfal2 is the best to use for this purpose. Certain differences were noted with Rockfal2 and so it was modified using Visual Basic, based in MSEXCEL™. Additional randomness has been incorporated to variate the starting position and velocity of the boulder, and to generate boulder trajectories and display them in graphical form. The modified program WinRock was used to simulate rockfalls at Fox Glacier. Back analyses were carried out (using the boulder distribution from past rockfalls, as surveyed and recorded in the field), to find the representative coefficients of restitution that are essential to accurately simulate rockfalls. These coefficients were subsequently used to simulate and predict rockfalls in future. Conclusions were drawn that rockfalls at Undercite Creek are relatively stable (with an exception that boulders in excess of 5.5m diameter have more potential to reach the access road) and the Cone Rock rockfalls may increase in due course. An overall assessment of rockfall hazards for all the degrading slopes at Fox Glacier was carried out using the Rockfall Hazard Rating System (1993) and CAN/CSA (1991) guidelines. This assessment identified and "quantified" the hazard that is involved with various slopes. From the hazard analysis it was found that the probability of fatalities are under the proposed and published acceptable limits for major civil engineering projects world-wide. An attempt was made to find out an easy means of obtaining the coefficient of restitution by easy laboratory methods. Investigation of a correlation between Schmidt number and the coefficient of restitution (of a steel ball bouncing on a rock slab clamped to the ground) revealed good results (correlation coefficient = 0.89). This indicates that a good correlation may also exist between Schmidt number and the restitution coefficient when a rock impacts rock.