Abstract
In northern regions, ice forces, or actions, must be considered in the design of structures such as light piers, bridge piers, and offshore platforms. Estimates of ice forces in Canadian waters are usually obtained by consulting design standards such as those developed by the International Organization for Standardization (ISO) and the Canadian Standards Association (CSA). These design standards draw on available analytical formulae. Field measurements are available from several sources that suggest reasonable agreement with analytical results for simple cases involving wide structures.
One of the remaining uncertainties in estimating design loads, however, is the contribution of force imposed below the waterline due to unconsolidated keels of ice ridges. Only cursory guidance is provided by the standards associations and their analytical design equations. Close inspection of those formulae show that force estimates can become excessive in situations where the expected keel depth is great compared to the designed structure width. Such scenarios would be expected in offshore oil and gas operations where drilling risers, jack-up legs, and even jacket structures may be exposed to ice ridges.
The present work examines available approaches for evaluating ridge keel forces, including passive pressure calculations. The processes of ice rubble failure and the corresponding stress distributions are considered in the context of classical soil mechanics applied in geotechnical engineering. Design standards are also used to calculate ice forces for a range of ridge keel properties, keel geometries, and structure design widths. Field measurements from the Norströmsgrund lighthouse and the offshore Molikpag caisson are then examined and compared to the forces obtained using these approaches.
The authors conclude that the shape factor adopted in ISO 19906 plays an important role in calculations considering narrow structures and deep keels. It is also shown that the sensitivity of ridge keel load calculation to geometric factors varies considerably with structure width. Furthermore, an absence of real world data from ridge keel interactions with very narrow structures precludes validation of present models in these situations and should be the focus of data collection and model refinement.
