This investigation of different multi-criteria evaluations (MCE) showed that by applying a deterministic model that described the effects of slope and soil type on landslide hazard instead of simply including slope and soil type in the MCE gives one more precise results of either extreme (high or low potential). The deterministic MCE also more accurately depicts the relationship between soil type/slope and landslide potential. The slope seems to have a very high correlation with failure. However, for both models the urban populations around Nanaimo and Victoria seem to at low risk of landslides.

Limitations

Data – The data for Vancouver Island is not necessarily of the highest quality and as for all analyses, garbage in is garbage out. The geotechnical parameters for the different geological units were taken from a literature review and not from actual measurements taken in the field. This is not necessarily a bad estimate, but could be more accurate if more measurements were taken. The precipitation had coarse resolution and a better resolution would increase the accuracy of our results.

Deterministic Analysis – Assumes an infinite slope (the depth of soil to the slip surface is far shorter then the length of the area or slope that is failing. The equation we used is for failure in soils and not in rock, this is why we assumed weathered rock, however, this has been used previously to quite accurately predict rockslides.

Weightings for MCE – The weightings we used were somewhat arbitrary as there does not seem to be literature that specifically gives weightings to the factors contributing to slope failure.

Further Research

More research into specific weightings for factors that are related to landslide potential would greatly improve this analysis. A landslide inventory could be used to test the validity of the two different models.

Acknowledgements

We would like to thank Dr. Brian Klinkenberg and Joel Trubilowicz for their guidance in the development of this project.