Revisiting the Dimensional Analysis of Pipeline Embedment During Installation

In deep water, pipelines are usually laid directly onto the seabed, in which they embed becoming partially buried. The depth a pipe will embed during installation depends on its weight and on the soil strength, but also on many other features of the lay process. The embedment along the pipeline length will govern many aspects of its behavior, for example the axial and lateral stiffness in thermo-mechanical expansion process. Despite its importance, foreseeing this penetration in design stage is still a difficult task, for which many models have been proposed over the years. Verley & Sotberg [1] and Verley & Lund [2] have presented a thorough dimensional analysis of the pipe-soil interaction, as part of a broad study on the lateral stability of pipelines. Their equations for the initial embedment (in particular the last for embedment in undrained conditions) have been often used in design, but usually receiving additional correction factors to account for aspects not originally considered by the authors. Unlike the original dimensional analysis, the calibration of these correction factors has been done in trial and error basis, often leading to unreasonable results. This paper revisits the dimensional analysis of the pipeline embedment during installation, aiming at widening the set of variables involved. The objective is to put together a dimensionally consistent framework in which additional aspects of the embedment process can be included in a rational way. The paper is limited to the desktop study of selecting a new set of dimensionless groups. This might be later used for deriving a new, more accurate, model for foreseeing pipeline embedment levels in design stage. This would though require extensive testing. Which is beyond the scope of the current work.