Real-time monitoring of a multiple retrogressive landslide
In this thesis the data collected from the field monitoring of slope inclinometers, in place inclinometers (IPIs) and vibrating wire piezometers (VWPs) installed in a multiple retrogressive landslide are evaluated and re-interpreted. The goal of this research was to evaluate the IPIs in particular, as part of a real-time slope stability monitoring system. The real-time monitoring system was installed by the Saskatchewan Ministry of Highways and Infrastructure (SMHI) to monitor a multiple retrogressive landslide affecting Highway No. 302 near Prince Albert, Saskatchewan. The instrumentation was connected to an automatic data collection and communication system so that data was available remotely via the ARGUS monitoring website. The instrumentation system was implemented after manual displacement monitoring confirmed high rates of movement. The automated landslide monitoring system utilized IPIs and VWPs. The instruments were installed next to Highway No. 302 at the crest of the slope and next to the North Saskatchewan River at the toe of the slope. Data was collected between September 18, 2005 and July 2, 2006. The near real-time data provided from the instrumentation was pore water pressure, river level, and lateral slope displacement at an hourly interval. Site investigation and slope inclinometer monitoring confirmed a major slip surface located in the glaciolacustrine clay deposit just above the clay/till contact. Slope inclinometer monitoring showed that the toe of the slope was moving at greater rates and magnitudes than the crest of slope. This observation of movement was consistent in the literature regarding other multiple retrogressive landslides using manual methods of monitoring. The installation of the automated monitoring system allowed instrument data to be collected at hourly intervals. Consistent periodic reading intervals captured data from the IPIs that showed the crest of the slope was moving more quickly than the toe. This was a significant observation that may not have been captured with manual readings with a traversing slope inclinometer probe. Observations were also made by correlating displacement data from IPIs at the toe of the slope with river level. It was shown that movement began after a rise and fall in river level. This suggests that displacement of the landslide was triggered by river bank erosion due to higher seasonal flows or alternately a rapid drawdown mechanism. On October 13, 2005, 25 days after installation of the IPIs, a major disturbance in the readings at both the crest and toe of the slope occurred. After this disturbance the crest was moving at greater rates than the toe. The initial disturbance was attributed to a settling of the head assembly of the IPIs triggered by global failure of the slope. Additional disturbances were noted in the IPIs at the toe of the slope and attributed to the casing contacting the IPIs. A geometric analysis was developed to verify that the unexpected movement could have been caused by the casing contacting the IPIs. The geometric analysis included observation of the shear induced deformation profile of the slope inclinometer casing, IPIs diameter, IPIs length and the annular space between the casing and the IPIs. The geometric analysis verified that the IPIs at the toe of the slope were likely affected by the IPIs coming into contact with the casing. Casing contact could affect the IPIs orientation and depth position, resulting in unexpected data. An additional set of IPIs data from a landslide near Aylesbury, Saskatchewan was used to compare to the IPIs data from Highway No. 302. Data comparison showed that there were erratic readings from the IPIs at the Highway No. 302 site while the IPIs at the Aylesbury site generally remained stable and consistent. This data also confirmed that the IPIs at Highway No. 302 were affected by casing contact. The near real-time monitoring system was an excellent risk management tool for SMHI. Due to the rate and magnitude of movement the instrumentation was destroyed sooner than expected. The data set produced allowed for evaluation of the landslide and performance of the IPIs. In conjunction with the river data the IPIs helped identify potential failure mechanisms. Additional work is required to ensure that the data from IPIs can be used to evaluate landslide kinematics. In particular ensuring that the IPIs are not used beyond their capability and providing realistic data.
DegreeMaster of Science (M.Sc.)
DepartmentCivil and Geological Engineering
SupervisorSharma, Jitendrapal; Barbour, S L.
CommitteeHawkes, Christopher D.; Kelly, Allen J.; Antunes, J P.; Kells, James A.; Panesar, Harpreet S.
Copyright DateApril 2012
multiple retrogressive landslide