The City of Richmond is located on the Fraser River Delta which overlies deltaic sediment deposits up to 300m thick. These sediments are composed primarily of sand and silt and make up the Holocene layer. Beneath the Holocene is a layer of Pleistocene and beneath that, the bedrock [1]. Bedrock varies from 200m to 1000m from the surface with an average of 500m [2]. These soil deposits are prone to amplification, and, as a result of a high ground water table, are also prone to liquefaction [1]. On the North Shore, the sediments are much thinner and overlay the rocks of the coast plutonic complex [2]. The composition of the soil layers varies from till to gravel, sand and small amounts of clay. In some areas, the bedrock is exposed. The bulk of this area is not at high risk of liquefaction [5]. 2.2 Measuring Equipment The equipment used in this study is owned by The University of British Columbia‟s Earthquake Engineering Research Facility (EERF). The sensors, Pinocchio WL380s, are geophone-based velocity meters capable of taking long measurements (up to 12hrs) at sampling rates of 100sps and can be seen in Figure 2. Each unit has an internal power supply and GPS antenna, increasing the portability and enabling precise location and time synchronization. Two sets of tri-axial geophones make up each Pinocchio, one for each of high and low amplitude motion. The geophones have a natural frequency of 4.5Hz and 56% damping. Fig. 2 Pinocchio WL380 Sensors in the Field 2.3 Testing Procedure For each test, the units were programmed using a laptop and Secure Digital card (SD). Tests on the North Shore were 20 minutes in duration and tests in Richmond, 30 minutes. According to Ventura et al [3] and SESAME guidelines [12], longer tests are required to identify longer periods. Previous testing in the City of Richmond confirmed this when it failed to identify the longer periods with 5 minute tests [3]. All tests were performed at 100 samples per second. Once the test was complete, the SD card was removed from the sensor and data saved on the laptop. Where possible, sensors were placed on concrete or asphalt for the tests as this appeared to give the most precise results. Where concrete was not available, steel plates with four steel spikes were placed into softer ground and the sensor placed on top of the plate. The spikes secured the sensor in place horizontally. The sensors were oriented with the Y axis in the North direction and levelled. On average, moving to the next location, setting up the sensor, programming the test, and running the test took one hour. A total of 212 tests were performed, with a resolution of approximately 500-800m spacing. 453
RkJQdWJsaXNoZXIy MTMzNzEzMQ==