| Title | Landslides on Haida Gwaii during and after the Earthquake |
| |
| Author | Millard, T; Geertsema, M; Clague, J J; Bobrowsky, P ; Hasler, A; Sakals, M |
| Source | Risky Ground, newsletter of the Centre for natural hazards research, Simon Fraser University Winter, 2012 p. 12-13 |
| Image |  |
| Year | 2012 |
| Alt Series | Earth Sciences Sector, Contribution Series 20120393 |
| Document | serial |
| Lang. | English |
| Media | paper |
| File format | pdf |
| Province | British Columbia |
| NTS | 103B; 103C; 103F; 103G/13; 103J/04; 103K/01; 103K/02; 103K/03 |
| Area | Haida Gwaii |
| Lat/Long WENS | -134.0000 -130.2500 54.2500 51.5000 |
| Subjects | geophysics; engineering geology; landslide deposits; landslides; earthquakes; earthquake studies; earthquake magnitudes; earthquake mechanisms; slope deposits; slope failures; slope stability |
| Illustrations | location maps; photographs |
| Program | Public Safety Geoscience National-Scale Geohazard Assessments |
| Abstract | Haida Gwaii, located off the northern mainland coast of British Columbia, has some of the most landslide-prone terrain in Canada. The earthquake of October 27, 2012, and subsequent aftershocks caused
numerous, although fewer than expected, landslides on Haida Gwaii. We observed fresh landslides on the west coast of Graham Island, across Moresby Island, and on adjacent smaller islands during reconnaissance helicopter (Millard) and fixed-wing
(Geertsema) flights (Figure 1) shortly after the earthquake. A strong wind and rain storm on November 4, immediately before our first flight, may have triggered some of the landslides, confounding our analysis.
We identified nearly 100 fresh
landslides during our reconnaissance flights (Figure 1). One-third of the landslides occurred on Graham Island (the northern island) in the Rennell Sound area. Two-thirds of the landslides happened on Moresby Island, to the south, with the highest
concentration (one-third of the total) directly south of Cowgala Bay (Figure 1 inset). About 90% of the landslides were shallow debris slides (Figure 2), typical of those in coastal British Columbia. The remainder were rock falls, rock slides and
debris flows.
This is the second known report of earthquake-triggered landslides in British Columbia. A large (Mw 7.3) earthquake occurred beneath central Vancouver Island in June 1946. Mathews (1979) concluded that this earthquake triggered
hundreds of small landslides (rock falls and debris avalanches) similar to those that occurred on Haida Gwaii in 2012. The number of landslides attributed to the Vancouver Island earthquake is greater than the number that occurred during the 2012
Haida Gwaii earthquake. Mathews, however, based his estimate on comparison of pre-earthquake aerial photographs and airphotos taken 11 years after the earthquake, which may have inflated the numbers.
Although some of the 2012 debris slides may
have been associated with windthrow from the November 4 storm, they are few and not plotted. It is unclear whether the debris slides occurred immediately after the earthquake or after the November 4 storm, but the rock-slope failures were likely
coseismic. The high concentration of debris slides coincides roughly with the epicentre of a Mw 4.3 aftershock (Figure 1, see also Figure 2 in Cassidy et al. in this newsletter), but a Mw 5.2 aftershock epicentre farther south occurred in an area
with few observed landslides. Although Keefer (1984) has shown that a Mw 4 earthquake is sufficiently powerful to trigger landslides, a lack of dry antecedent soil moisture conditions in the autumn of 2012 may have been a factor in the relatively
low number of earthquake-triggered landslides. Nevertheless, the mainshock and the swarm of aftershocks may have reduced the threshold for future landslide initiation, as observed by Lin et al. (2003) following the Mw 7.3 Chi-Chi earthquake in
Taiwan. Time will tell. As a precaution, government officials now require forest workers to shut down operations and leave landslide-prone terrain when the water input into soils reaches 60% of previous thresholds for operational shut-down.
We are
acquiring satellite imagery of the area to better catalogue landslides associated with the earthquake. Repeated acquisition of imagery will assist in evaluating the stability of slopes in the months and years to come. |
| GEOSCAN ID | 292217 |
|
|