Subsea pipelines are sometimes buried in rockfill berms for the purpose of buckle mitigation. To estimate the axial resistance of such pipelines requires an estimation of the normal force acting on the pipe. This paper explores the degree to which berm size influences the normal force around a pipe relative to the reference case of uniform cover (i.e. an infinite berm). The goal of this study is to provide normal force estimations for the valuable buried axial test data presented in Eiksund et al. (2013), accounting for the reduction in normal force due to the finite cover used therein. The influence of a shallow boundary (such as is present in test tanks used for buried axial tests) is also considered. The results feed into a companion paper that collates axial rockfill test data from various sources.


Burial of a pipeline in rockfill is a common design approach to control the buckling of subsea pipelines. This can include burial of a pipeline resting at the base of a trench or resting on the seabed. When a pipeline resting on the seabed is buried in rockfill it typically involves formation of an embankment-shaped rockberm around the pipeline. Geotechnical quantification of the axial restraint involves estimation of the normal force acting around the pipe and estimation of an appropriate friction coefficient acting between the pipe and rockfill (as well as between pipe and underlying soil). The normal force is typically taken to be a function of factors such as cover height, lateral earth pressure coefficient, pipe diameter and so forth. The rockfill friction coefficient is usually estimated from the valuable body of test data presented in Eiksund et al. (2013), as recommended by DNVGL (2019) in the design code DNVGL-RP-F114.

The Eiksund et al. (2013) test data was derived using a shallow test tank of length 3m and width 6.5m. The buried tests involved berm-shaped rockfill deposits of side-slope 1(v):2(h) with various cover heights, crest widths, pipe properties and water level. Friction coefficients were interpreted based on the assumption of a stress distribution resulting from uniform cover (i.e. an infinite berm). Two difficulties in the use of this data have to be grappled with. First, to what extent the finite cover offered by a berm results in lower normal force than if the pipe was buried under uniform cover. Second, whether the shallow rigid boundary of the test tank influences the assumed stress distribution (since no such boundary typically exists in the field). Both of these issues would affect the interpreted friction coefficient.

This content is only available via PDF.
You can access this article if you purchase or spend a download.