Several developments have taken place in regions characterized by seasonal ice cover, including the US Beaufort, North Caspian, and Sakhalin Island. These projects incorporated pipeline transportation systems that are a cost-effective, safe, and reliable mode of hydrocarbon transport to shore. Ice gouging is one of the key design issues that affect engineering considerations with respect to strain-based design, target burial depth requirements, cost, and safety. Other challenges that must be considered include strudel and hydrodynamic scour, thaw settlement of permafrost and frost heave, and upheaval buckling. These considerations also may influence design requirements.

Ice gouging

It is generally accepted that Arctic pipelines would need to be trenched to some depth below the mudline to protect the pipeline from ice keels. Ice gouging of the seafloor (sometimes referred to as ice scouring) is a near-shore feature for most of the northern continents. Sea ice is driven by wind and current forces and tends to pile up, creating pressure ridges when the ice moves. Pressure ridges have keels extending below the water surface, which move with the ice sheet. Occasionally, these ice keels intrude into water depths less than the ice keel draft and form a gouge in the seafloor.

As an ice keel passes over any point in the seabed, vertical and lateral stresses are applied to the soil at the keel base, resulting in some distribution of vertical and lateral soil displacements with depth beneath the ice keel. The movement of the soil also loads or moves the pipelines in the trench. The configuration of the pipeline after gouging, and hence the strain in the pipelines, depends on the pipeline properties, the soil characteristics, and the depth of the pipeline as well as the soil displacements. The soil displacements induced at the pipeline depth due to ice gouging and resulting strains in the pipeline must be calculated. The burial depth of the pipeline must be sufficient so that the maximum predicted ice gouge will result in pipeline strains within acceptable levels.

Strudel scour and hydrodynamic scour

Nearshore Arctic zones typically develop a bottomfast ice sheet during the winter season. If an onshore river flow encounters such an area during the spring breakup, the river water will overflow the bottomfast ice sheet in the nearshore zone. This overflow water will spread offshore and drain through tidal and thermal cracks or seal breathing holes in the ice sheet. If the drainage rate is high, hydrodynamics (high-velocity currents) of the draining water at the seafloor can scour seabed sediment (leaving a circular or linear scour in the seabed), which can potentially expose and impose current loads on a pipeline. These are known as strudel scours, and they usually occur in 2 m to 8 m (6 ft to 26 ft) depth offshore from river deltas. It is unlikely that every drain hole in the sheet ice produces a scour in the seafloor. The deepest scours are found in shallow water – 2 m to 3 m (10 ft) deep – where the strudel flow is sufficiently powerful to excavate the seafloor sediments immediately below the ice.

If a strudel scour happens on top of a pipeline, there is the possibility that the scour could result in an unacceptable pipeline span. In extreme conditions, the pipeline span could possibly experience hydrodynamic loads from the water flow, and vortex-induced vibration effects may need to be checked.

Thaw settlement and frost heave

If permafrost was continuous and uniform in terms of soil and ice conditions along the pipeline route, and if the pipeline was operated at a constant temperature along the route, then differential thaw settlement and its effects would not be an issue as settlement would be uniform along the pipeline. These ideal conditions do not exist in practice. Irregular, discontinuous ice-bonded permafrost soil conditions are more common, and pipeline temperature can vary along its length.

When the pipeline becomes operational, the temperature of the pipeline will typically increase, thereby warming surrounding soil and creating a permafrost thaw bulb. The extent of the thaw bulb, the soil type, ice and moisture content, and the stratigraphic profile are the primary factors that determine the potential for differential thaw settlement along the pipeline alignment. If the thaw settlement area is adjacent to an area that is thaw-stable, the differential settlement can induce considerable strain in the pipeline and must be accounted for in design, which can be analyzed through geothermal analyses and finite element modeling.

Upheaval buckling

A buried steel pipeline will try to expand longitudinally when operated at a temperature and pressure higher than that experienced during installation. A long buried pipeline is not free to expand due to the restraint provided by the surrounding soil and thus will develop a locked-in axial compressive force. If the buried pipeline has some residual vertical curvature, possibly due to trench bottom irregularities during installation, the tendency of the axial force near the high points of these trench irregularities will be to buckle the pipeline upward. If the upward force exceeds the downward force due to the combination of the resistance of the soil cover, the pipeline stiffness, and the pipeline self-weight, then the pipeline will move upward and may become exposed above the seabed. This phenomenon is known as upheaval buckling and has been frequently documented for offshore pipelines.

The immediate effect for Arctic offshore pipelines is that the pipeline could have less burial depth or even become exposed at the seabed, which increases the risk of interaction with ice keels. Problems associated with upheaval buckling may include high bending stresses and loss of protective soil cover but may not directly cause a leak or exceed other limit states. However, upheaval buckling is an undesirable condition that must be considered in design. Upheaval buckling analysis must be carried out to determine the minimum backfill thickness over the pipeline for the selected design parameters and maximum allowable vertical variance (prop or imperfection) of the installed pipeline profile.

Pipeline design, construction implications

The design of offshore pipelines in Arctic and northern ice environments must evaluate environmental and geotechnical load effects for potential large deformation ground movement events that may affect pipeline mechanical integrity. Evaluation of the system demand and system capacity influences engineering design considerations that may impact the construction and operational phases.

The first two design issues listed above happen over very limited distances of the nearshore portion of the pipeline and have been addressed in existing pipeline design. Upheaval buckling can happen anywhere in the pipeline (and also in non-Arctic pipelines) if conditions are conducive to buckling. Installation procedures have been implemented on past projects – such as Northstar – to minimize the risk of upheaval buckling. Thaw settlement or upheaval buckling may govern nearshore burial depths as well as trench/backfill design, whereas ice gouging would govern burial depth in deeper waters. It is generally felt that these considerations can be designed for in future projects.

Pipelines have been designed, constructed, and are operational in Arctic (e.g. US Beaufort) and subarctic regions (e.g. Sakhalin Island). The pipelines operating in the true Arctic are in relatively shallow water depths and fairly close to shore. Developments further offshore in deeper water will require additional consideration to aspects related to pipeline design, in particular with respect to burial for protection against ice gouging. Pipeline burial for protection in water depths from approximately 20 m to 40 m (66 ft to 131 ft) will be a challenge given the more severe gouging in these water depths. Modern gouges up to 3 m deep in the US Beaufort and 5 m (16 ft) deep in the Canadian Beaufort have been found. Conceptual development of specialized trenching systems to permit burial to the required depths has been evolving, and advanced technology may need to be developed to allow projects to proceed.

Acknowledgment
This article is based on OTC Paper 24607, which was presented at the Arctic Technology Conference in Houston in February 2014.

References available on request.