The recent failure of a coastal buried pipeline along the Italian coast is presented. Extensive initial investigations to assess failure causes are conducted and several shreds of evidence are identified leading to wave-induced liquefaction as the main source of failure. This thesis is subsequently confirmed through the application of two known analytical models, i.e. Ishihara and Yamazaki 1984 and Sumer et al., 2012. The results show that despite the rather inexpensive computational cost of the models, they can be successfully used to identify the area affected by the wave-induced liquefaction with good agreement with field observation.


Wave-induced instability of seabed soils around buried pipelines is an increasingly important research subject concerning the stability of pipelines transporting hydrocarbons from wells to processing facilities (Pisanò et al., 2022), of pipelines for outflow discharge and of cable connections for marine renewable energy. When directly laid on the seabed, pipelines are often exposed to harsh hydrodynamic loads or collisions with hard bodies that may negatively impact their structural performance. A typical stabilization option is to lay pipelines in trenches backfilled with rocks or sand. Since sand backfills are often loose (uncompacted) and shallow (i.e. subject to low effective compression), Pipelines buried in sandy backfill may suffer from the consequences of soil liquefaction because they can move in liquefied sands (Pisanò et al., 2020). Liquefaction can be triggered by a number of factors, including structural vibrations, ocean waves, and earthquakes (Sumer, 2006). The occurrence of liquefaction is generally associated with low values of relative density of the backfill in combination with low effective stresses at shallow soil depth, however, influence of the previous stress/strain history can also play a role (Nelson and Okamura, 2019). Liquefied soils are characterized by low strength and stiffness inducing segments of buried pipelines to experience excessive displacement, for instance in the form of vertical flotation or sinking. In the presence of either relatively light or heavy pipelines, the difference between the pipeline and the liquefied sand weight is the main trigger of the flotation or sinking process. The process of buried pipelines flotation is well known since 1966 with the study conducted within the Pipeline Floatation Research Council framework in US (Pipeline Flotation Research Council, 1966), while later additional studies and documented pipeline failures were reported by Christian et al. (1974), Herbich et al. (1984) and Damgaard et al. (2006). EU wise, considerable research effort was invested on the wave induced liquefaction problem through the LIMAS project ended in 2004, (Damgaard et al., 2006; Sumer, 2006; Sumer, 2014; Sumer et al., 2010; Sumer et al., 2006; Sumer, 2012) while more recently the ongoing NuLIMAS project is developing open source numerical tools to model the liquefaction around marine structures, (Shanmugasundaram et al., 2022; Sumer and Kirca, 2022). Some relevant outcomes of the previous research efforts are reflected within some industry standards, however, it is the opinion of this paper's authors that no specific methodologies are explicitly mentioned to assess the risk of wave induced liquefaction. Indeed, industry practice relies on the designer experience who rather often assumes liquefaction around buried pipelines as a given fact, whose negative effects can be mitigated by designing heavier pipelines and rock protection. The twofold goal of this work is to present a recent case of coastal buried pipeline failure due to wave induced liquefaction and to highlight the predictive capability of simple and computationally inexpensive models such as Ishihara and Sumer's models (Ishihara and Yamazaki, 1984; Sumer, 2012). The effectiveness of the models will be quantified by contrasting their prediction in terms of liquefied trench area and the observed pipelines failure.

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