The purpose of this research project was to develop a procedure for terrain
stability assessment by applying case-control sampling and multiple logistic
regression analysis, widely used statistical techniques in biomedical research and in
epidemiology. The idea of applying statistical methods used in epidemiology to
terrain stability assessment was based on the observation that landslides, like some
diseases, are rare phenomena. The implementation of a terrain stability assessment
based on these statistical techniques was expected to help understand the causeeffect
relationships between landsliding and various terrain attributes. In contrast to
the currently used approaches, the study procedure provided a quantitative tool to
assess the risk of landsliding and to define the most important terrain attributes that
contribute to soil mass movements.
A case-control study of 20x20 m grid cells with average slope greater than 10
degrees was conducted on the Jamieson-Orchid-Elbow Creeks subdrainage of the
Seymour River Basin, British Columbia. All of the 101 landslide cases were
compared with 264 control grid cells. Multi-way cross classification tables were
constructed to study the relationship between landsliding and several terrain
attributes. A possible interaction between slope angle and the drainage condition of
the soil was detected. A logistic regression analysis was then performed within a
Geographic Information System (GIS) environment to develop a landslide risk
model for the Jamieson-Orchid-Elbow Creeks study area. A landslide risk matrix
was then constructed based on the landslide risk model. It was found that sites
located in the transient snow zone, with slope angle greater than 55 degrees, on
bedrock outcrop surficial material type and on shallow soil have the greatest risk of
experiencing rapid, shallow soil mass movements. It was also found that holding all
the other variables constant, slope angle had the greatest effect on the magnitude of
landslide risk. Based on the data, sites with very steep slopes (over 55 degrees)
have, on the average, five times the chance of experiencing a landslide event
relative to sites with gentle slopes (10-25 degrees).
The landslide risk matrix was used to create landslide risk categories. The
spatial distribution of landslide risk, categorized as very low, low, moderate, high
and very high, is portrayed within 20-m square grid cells on the landslide risk map.
The major advantage of using the landslide risk assessment of this study is that
it provides the terrain mapper with quantitative information about the relative risk of
landsliding. This information can be used as a tool in planning watershed
management activities and in an overall risk assessment for a given geographic area.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:BVAU.2429/3807 |
| Date | 11 1900 |
| Creators | Gulyás, Gyula |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Relation | UBC Retrospective Theses Digitization Project [http://www.library.ubc.ca/archives/retro_theses/] |
Page generated in 0.006 seconds