The characteristics of first-year and multi-year sea ice and icebergs in different offshore regions in the world vary significantly, resulting in varied ice load distributions depending on the specific ice types, environment, and platform characteristics. As part of the development and validation of action factors for design ice loads for the new International Standard ISO 19906 "Petroleum and natural gas industries-Arctic offshore structures", ice load distributions were derived for a broad range of arctic and subarctic conditions and structure types. The results provided necessary input for the calibration of ISO 19906. A methodology for probabilistic analysis of ice loads for various regions has been developed. Important results on the hazard curves were obtained, and these results point to areas where further research is needed in arctic structure design. In the subsequent calibration of action factors, generally results consistent with accepted practice were found, resulting in action factors of 1.35 for Extreme-Level Ice Events (ELIE), and 1.0 for Abnormal-Level Ice Events (ALIE). It was found that the Wang and Croasdale models for multi-year ridge interactions with sloped and conical structures required action factors higher than 1.35. On further review, it was confirmed that these models do result in load distributions with relatively fat tails, requiring higher action factors. It is believed that these results occur because the effect of scaling of flexural strength and other factors related to the scale effect are not taken into account, and that a higher action factor is not necessarily warranted. An overview of the scenarios considered and the methods used to determine load distributions is provided in this paper. The calibration procedure and resulting action factors are described in related papers.
T he design procedures in the ISO 19906 International Standard are based on estimated ice loads associated with Extreme-Level Ice Events (ELIE) and Abnormal-Level Ice Events (ALIE). ELIE corresponds to an annual probability of exceedance of 10–2 and ALIE corresponds to an annual probability of exceedance of 10–4 (Thomas, 2011). Action and resistance factors are calibrated to achieve target reliability levels (Maes, 2011). Good calibration requires that the slope of the tails of annual maximum load distributions be well defined. For example, if the design load is the ELIE load multiplied by an action factor, then to ensure reasonably consistent reliability levels, the probability of exceeding loads larger than the ELIE load must be reasonably well established. The process is illustrated in Figure 1. Given load and resistance distributions, a convolution integration can be performed to determine the probability that the load is greater than the resistance. Given characteristic values, such as the load and resistance at 10–2 and 10–1 annual probabilities of exceedance respectively, the action and resistance factors are used to control the spacing between the distributions and hence the probability of failure. If the tails of distributions behave consistently then it is possible to calibrate to achieve consistent action factors.
