An Improved Model for the Prediction of Pipeline Embedment on the Basis of Assessment of Field Data

ences both axial and lateral resistance, and these PSI responses are usually the most-significant uncertainty in the design of pipelines laid on the seabed. The case history presented in this paper shows that this approach is invaluable and provides a significant contribution to good design practice. Two existing export pipelines of significantly different overall pipeline diameter and submerged weight were laid along the same route, in the same soils. These pipelines might have been expected to reach quite different levels of embedment following installation and flooding, and current models for predicting embedment (described in the following) confirmed this; yet, the final levels of embedment were relatively similar and deeper than those that would be predicted with current practice. This is clearly a concern because higher levels of embedment generally lead to higher levels of resistance from the soil, which is often the most-challenging design case in the assessment of lateral buckling (Bruton et al. 2007). This finding has therefore provided an excellent opportunity to modify and calibrate embedment models for use in defining PSI responses on current projects. An assessment of the embedment mechanisms during installation and post-installation flooding has led to a modified methodology supported by geotechnical principles that provide a much-improved correlation between predicted and measured embedment levels for these pipelines. This new approach is recommended for prediction of pipeline embedment levels on current projects. This paper addresses some important revisions to current embedment models: • Improved modeling of penetration resistance because of buoyancy, heave mounds, and bearing capacity at embedment levels greater than one-half diameter, which is a concern in weaker soils. • Improved modeling of the likely operative shear strength at the time of pipeline flooding, to account for the level of strength regained because of reconsolidation of the soil under the weight of the empty pipe. In this assessment, one can assume that sufficient time (2 to 4 months) has passed to achieve a relatively high level of reconsolidation. Further work is required to quantify the likely increase in operative strength with time because the duration between installation and flooding is potentially an important input to the final pipeline embedment. Indeed, this methodology confirms that insufficient time between installation and flooding can result in excessively deep pipeline embedment.