Open Collections

Slope stability prediction techniques for forest management purposes : a case study from the Queen Charlotte Islands, British Columbia

University of British Columbia

This study examines the predictive ability of three slope stability assessment techniques used in British Columbia. Two of the methods, environmentally sensitive area mapping (ESA) and slope hazard ratings generated from terrain mapping (SRS), are routinely applied in forest management; the third, still under development and so far utilized for research purposes only, develops a failure probability rating from terrain mapping and clearcut slope failure frequencies (SGA). The specific objectives are to (1) appraise the predictive ability of these stability prediction methods in logged and natural terrain, (2) appraise the comparative performance of the methods, and (3) determine the method most useful to British Columbia land managers. This study focuses on testing/comparing each method's predictions. To be successful, a stability assessment method must predict where the greatest increase in failures will occur if the hillslopes are logged. Similarly, a method is considered successful if, regardless of treatment, the stability class failure frequencies are ranked in tandem with the predicted likelihood of failure (i.e., the least stable class will have the largest failure frequency). Two study regions on the Queen Charlotte Islands were subjected to each prediction method using 1:20,000 scale pre-logging (1977) aerial photographs and failure inventories completed in each region. Failure frequencies by stability rating (as determined for each approach), for both logged and natural terrain, were determined from the recent (1988) airphotos. Failure frequency per unit area was the analytical unit utilized for statistical comparisons of predictive success. Two non-parametric statistical techniques, Mann-Whitney U Test and Spearman's Rank Correlation were employed in the analysis. Both regions had the majority of new failures happen in logged terrain. The overall failure frequency was 1.9 per km2. McClinton Bay's unlogged and logged area failure frequencies were 1.4 and 3.1 per km2, respectively. Louise Island's unlogged and logged area failure frequencies were 0.9 and 4.8 per km2. SRS successfully predicted 94% of all failures (52% of land designated medium-high hazard), ESA predicted 73% of all failures (23% of land medium-high hazard), and SGA predicted 52% of all failures (21% of land medium-high hazard). Which method is better? If economics are not considered then SRS is without qualification the most accurate. ES A is the most cost-effective. As typical with many applied geomorphology questions, the final analysis displays the tension between scientific understanding and hands-on management. In seeking to bring the greatest understanding of the complex factors influencing surficial terrain failure the scientist is often at odds with the land manager who wishes to avoid complex classification. Thus, if understanding is the prime consideration then the SRS method is recommended; otherwise, from an economic and management stance, the ESA method appears to hold the greatest promise. The importance of this applied geomorphology thesis lies in the development of a methodological approach to critically assess slope stability prediction methods, the failure inventory, the use of non-parametric statistics, the discussion of tension between 'science and management,' and, of course, the results.

Thesis/Dissertation

application/pdf

2008-12-19

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/3228

Geography

University of British Columbia

UBCV

DSpace