| Résumé | This paper offers the most comprehensive set of property measurements on multi-year ice to date, in the interest of addressing one of the most significant unknowns for Arctic engineering: multi-year ice strength. Borehole strength results are presented from 56 old ice floes in the gray literature and more than 600 tests conducted on 23 multi-year ice floes over the past decade, including the first-published results over the full thickness of a 12.7. m thick, cold multi-year hummock. The ice borehole strength is obtained by categorizing the pressure vs. time histories for each test into one of four main types of failure behavior. Vertical profiles of the temperature, salinity and borehole strength of multi-year floes in spring and summer demonstrate that the properties of multi-year ice are highly variable in space and time. The mean borehole strength and standard deviation of cold (- 13. °C) multi-year ice is 34.2 ± 9.1 MPa, although strengths as high as 49.2. MPa do occur, making multi-year ice nearly twice as strong as cold first-year ice. The mean borehole strength and standard deviation of warm multi-year ice is 19.6 ± 7.2. MPa (at - 5 °C) and 10.3 ± 5.3 MPa (at 0°C). Ice temperature is shown to be the single largest factor influencing borehole strength: strength increases with decreasing ice temperature, however complex factors such as the ice failure mode and ice consolidation also bear upon the relation. For example, strengths measured in thick, level multi-year ice can be substantially higher than hummocked multi-year ice sampled at the same temperature, time of year and latitude. Similarly, a thoroughly weathered multi-year ice hummock in late summer can have considerably higher strength than a less weathered multi-year hummock in early spring. The study shows that multi-year ice does not deteriorate in the same manner as first-year ice, strength equations based solely on brine volume are not appropriate for multi-year ice and warm multi-year ice should not be assumed deteriorated. The viability of estimating the ice borehole strength from known ice temperatures is explored by fitting linear regressions to strength-temperature data for the two most common failure processes: well-defined yield failures (Type 2) and poorly-defined yield failures (Type 3). The Type 2 failure equation reproduces the measured strength profiles more closely than the Type 3 failure equation, but results are not ideal. A similar comparison was made for the effective borehole strength, i.e. the strength averaged over all test depths in a particular borehole. For the 64 boreholes examined, the Type 2 failure equation produced an upper bound for the effective borehole strength, but only when ice temperatures had been documented over at least half of the total ice thickness. |
|---|
