Progress Reports: December 2006 June 2006 December 2005
Risk Assessment of Submarine Slope Stability – Hydroplaning
Note: This project is an extension of Submarine Slope Stability Risk
OBJECTIVE: This project is developing a model to predict the movement of submarine slides, with emphasis on slides that travel large distances once they are initiated. In particular the research seeks to develop a numerical model for predicting the initiation of hydroplaning of a slide mass and the subsequent motion of the mass once hydroplaning is initiated.The numerical model developed in this project will be applicable to subaqueous slides of any scale, and will be able to determine if hydroplaning is likely to be initiated as well as the magnitude of movements when hydroplaning is initiated. The sliding process will be simulated based on the initial geometry of a slope failure, the geology of the nearby seafloor, and the mechanical properties of the slide material (including shear strength, stress-deformation properties and unit weight). The model will describe (1) the variation of the velocity of the slide mass in time and space, and (2) the eventual runout distance and geometry of the slide mass. This information is essential in judging the potential risk associated with submarine slides.
INTRODUCTION: Previous research on submarine slides has consisted of numerical and physical modeling and the development of both empirical and numerical models to predict the initiation and movement of slides. This research shows that under certain conditions a moving slide mass can hydroplane on a layer of water that becomes trapped between the moving slide mass and the underlying soil. One of the most important aspects of hydroplaning is the interaction between the moving slide mass and surrounding fluid. Most of the previous work has been based on simplified assumptions regarding the interaction between the sliding mass and fluid. Often the assumptions are based on the interaction between fluid and simple structural elements that often have different geometries and are subjected to quite different motions than a slide mass moving along the ocean bottom. The assumptions have not been verified and most likely introduce errors of a large, but unknown magnitude.
BENEFITS TO MMS & INDUSTRY: Provide a means of including the likelihood and impact of hydroplaning on submarine soil slides to better evaluate the risk of slides impacting a subsea facility or pipeline planned for a specific site in the Gulf of Mexico.
DEPLOYMENT OF RESULTS: Results of this research will be published in Project Reports, theses, and dissertations as well as in conference proceedings and journal articles. Guidance on the likelihood and impact of hydroplaning on slides that could occur in the Gulf of Mexico will be provided, and the model will be available for application.
PROJECT ORGANIZATION & TIMING: During the Phases 1 and 2, research focused on modeling the interaction between a moving slide mass and surrounding fluid. Various numerical models for fluid response, including the commercial software Fluent have been used for this purpose. This modeling has led to an improved understanding of the characteristics of the fluid-slide mass interaction. However, up to this point the focus of the research has been on the fluid, rather than on the motion and deformation of the slide mass. In work done to this point the motion of the slide mass has been stipulated as a boundary condition. In Phase 3 we will incorporate the understanding of the fluid-mass interaction into a numerical model that includes the moving and deforming soil slide mass.
ANTICIPATED NUMBER OF PHASES: 4
PROJECT PLAN FOR PHASE 3 (2005-2006):
Scope and Plan: A numerical model will be developed for describing the initiation, movement and development of hydroplaning of a submarine slide. The work in Phase 3 will consist of the following tasks:
Task 1: Toward description of the progression of slide initiation, movement and eventual hydroplaning, numerical models with the requisite levels of complexity and sophistication will be formulated. The models, initially simple but gradually enhanced and increasingly detailed, will incorporate the nonlinear constitutive properties of the sediment mass as well as the appropriate fluid-sediment interface conditions. Work in the previous Phases 1 and 2 will provide a basis for characterizing the fluid response as well as the interaction between the moving slide mass and underlying sediment. The investigators have extensive prior experience with modeling deformable soil masses for both slope stability and related offshore problems that will be utilized in formulating the numerical model for the soil mass.
For the numerical model, an iterative transient scheme will be adopted in tracing the evolution of the sliding process. In each time step, the flow field around and stress applied to a slide mass with a given geometry and velocity will be computed based on the work completed in Phase 2 of this study. Next, the deformation and acceleration of the slide mass (soil) subjected to the fluid stress field within this increment of time will be computed numerically. This will produce a new geometry for the debris body and a new velocity boundary condition, which will then be substituted into a new analysis of the surrounding flow for the next time step. Iterations between the solutions for the surrounding flow and the deforming slide mass will be performed to represent the process of sediment sliding. Such a decoupled approach is computationally more attractive than a fully coupled procedure because it will enable certain types of nonlinearities to be considered that are otherwise much more difficult to consider in a fully coupled soil-fluid model.
Task 2: In this task we will implement the numerical models formulated in Task 1 by application and adaptation of available computational tools. To the extent possible we will use general-purpose computer codes for nonlinear, dynamic finite-element analysis along with special modules incorporating suitable constitutive descriptions of soil behavior. As necessary we will pursue the development of a novel computational capability, relying on the iterative scheme outlined above, specifically for modeling the initiation and motion of a slide mass where hydroplaning occurs and where there is significant soil-fluid interaction. The work in Task 2 will overlap that of Task 1 - as models are formulated they will be implemented, tested and possibly modified based on results of the initial testing. For example we anticipate that an initial series of tests will be performed using linear soil constitutive models before implementing more sophisticated nonlinear soil models.
Task 3: Perform numerical simulations for potential slides in the Gulf of Mexico and well as other locations where hydroplaning has occurred. This task is intended to accomplish several objectives:
• Determination of sensitivity of the model to various parameters and conditions.
• Identify and implement further refinements and modifications needed in the numerical models
• Determine conditions where hydroplaning may be expected to occur, with special emphasis on the Gulf of Mexico.Task 4: Task 4 is devoted to the planning of the fourth and final phase (Phase 4) of this study. Phase 4 will consist of experimental verification of the numerical model through bench-scale and possibly larger experiments. Appropriate numerical analyses will be performed in Task 4 to identify the needs and appropriate conditions for the subsequent experimental studies.
Task 5: This task is devoted to the preparation of the final report for Phase 3.
Anticipated Milestones & Results: A numerical model will be completed to estimate slide movements and possible hydroplaning will be completed. Illustrative example analyses will be completed to investigate the conditions the cause hydroplaning to occur and model sensitivities. The analyses will be used in developing a plan for Phase 4. Results will be documented in a Phase 3 Report.
PROJECT PLAN FOR PHASE 4 (2006-2007):
Scope of Work: The primary scope of the work in Phase 4 is experimental. In this phase laboratory experiments will be developed, performed and used to verify and calibrate the numerical model developed in Phase 3. The model will also be compared to any available field data. The calibrated model will be applied to hypothetical situations typical of Gulf of Mexico conditions to illustrate situations in which hydroplaning is likely and its impact on the slide behavior.
Anticipated Results: Numerical model that predicts the potential for slide movements including possible hydroplaning. This will enable a more rational assessment of the risk and possible consequences imposed by submarine slides and help to identify conditions where such slides can be anticipated.
PRINCIPAL INVESTIGATORS AND OTHERS INVOLVED IN THE PROJECT:
PI’s: Dr. Stephen G. Wright, Dr John Tassoulas
Others: Hongrui Hu - Ph.D. candidate
Date: December, 2006Project Title: Risk Assessment of Submarine Slope Stability - Hydroplaning
MMS Project: 556 TO Number: 39323
Project PI: Stephen Wright and John Tassoulas
COTR: M. Else
Estimated Completion Date: 1/31/2007
Project Description:
The objective of this project is to study the mechanism of hydroplaning as it relates to submarine slope stability. Hydroplaning is believed to be one of the major reasons why some submarine slides travel large distances. A “block” model has been developed to simulate the process of sliding including hydroplaning. Current research is focused on the verification and calibration of the block model using data from laboratory experiments conducted by Mohrig, et al. (1998, 1999). Once validated the block model will be applied to simulate the movement of actual slides reported in the literature.Progress:
The block model has been formulated and implemented in a computer code. After preliminary studies with the model it was modified to allow the actual conditions of Mohrig’s experiments to be simulated correctly. In the experiments, the soil mass was dumped from a soil tank at the head of a sloping channel. The soil mass remaining in the soil tank pushed the dumped mass down the channel. The original block model only represented the soil mass that is already in motion down a slope or channel. Thus, to simulate the experiments an additional force was added to the end of the block in the block model to simulate the thrust applied by the soil behind the block. Mohrig et. al. (1998, 1999) also reported that the soil mass in their experiments could be highly viscous. Accordingly a strain rate effect on soil resistance has also been added in the block model.Reports and Publications:
Hongrui Hu, Stephen G. Wright, and Spyros A. Kinnas. (2006). “Hydroplaning of Submarine Slides and the Influence of Hydrodynamic Stresses”. Proceedings, 16th International Offshore and Polar Engineering Conference, San Francisco, California, USA. pp. 482-489.
DATE: June 2006
Project Title: Risk Assessment of Submarine Slope Stability – Hydroplaning
MMS Project: 556 TO Number: 39323
PI: Stephen Wright and John Tassoulas
COTR: M. Else
Estimated Completion Date: September 30, 2006
Project Description: This project is developing numerical models to predict the movement of submarine slides, with emphasis on slides that travel large distances once they are initiated. In particular the research seeks to predict the initiation of hydroplaning of a slide mass and the subsequent motion of the mass once hydroplaning is initiated.
The numerical models developed in this project will be applicable to subaqueous slides of any scale, and will be able to determine if hydroplaning is likely to be initiated as well as determine the movements once hydroplaning is initiated. The sliding process will be simulated based on the initial geometry of a slope failure, the geomorphology of the nearby seafloor, and the mechanical properties of the slide material (including shear strength, stress-deformation properties and unit weight). The models will describe (1) the variation of the velocity of the slide mass in time and space, and (2) the eventual runout distance and geometry of the slide mass. This information is essential in judging the potential risk associated with submarine slides.
Progress: A block model is developed to simulate the sliding process of submarine slides. The slide body is represented as a 2-D deformable rectangular block with constant volume. The motion of the block is obtained by solving the dynamic equilibrium equations of the block. The soil body is assumed to remain rectangular although it deforms (elongates/compresses) while moving. The deformation of the block is determined based on the change of average stresses within the block and the stiffness of the soil. The stresses applied on the slide by the surrounding fluid are determined based on the numerical modeling of the flow around the slide mass. The normal stress applied by the underlying ground is simulated by springs between the bottom of the block and the ground. A constant coefficient of friction is assumed along the interface between the block and the ground. Hydroplaning is initiated when the normal force applied by the ground is equal to the kinetic pressure along the bottom of the block. During hydroplaning, the forces applied by the ground are replaced by the kinetic pressure and viscous shear along the bottom of the block. In the block model, the sliding process is simulated by a Newmark finite difference scheme. At the end of each time increment, the 2-D dynamic equilibrium of the block and deformation of the block are computed. The block model is implemented in software programmed using the C programming language. The block model will be used to study the influence of hydrodynamic stresses, initial dimension of the soil mass, stiffness and deformation of the soil mass and the geomorphology of seafloor on the sliding velocity, occurrence of hydroplaning and final run-out distance of the slides.
A more sophisticated FEM model is also being developed to simulate the progression of slide initiation, movement and deposition including possible hydroplaning. The FEM model will be implemented by applying and modifying program FEP++ developed by Murthy N. Guddati, Gonzalo Vasquez and Dilip R. Maniar. The FEM model has been tested for two cases: 1) plane strain consolidation under static loading and 2) dynamic response of a rigid block sliding down a slope.
Publications & Reports:
Hongrui Hu, Stephen G. Wright, Spyros A. Kinnas. (2006). “Hydroplaning of submarine slides and the influence of hydrodynamic stresses.” The 16th International Offshore and Polar Engineering Conference.
DATE: December 2005
Project Title: Risk Assessment of Submarine Slope Stability – Hydroplaning
MMS Project: 556 TO Number: 39323
PI: Stephen Wright and John Tassoulas
COTR: M. Else
Estimated Completion Date: September 2006
Project Description: This project is developing a model to predict the movement of submarine slides, with emphasis on slides that travel large distances once they are initiated. In particular the research seeks to develop a numerical model for predicting the initiation of hydroplaning of a slide mass and the subsequent motion of the mass once hydroplaning is initiated.
The numerical model developed in this project will be applicable to subaqueous slides of any scale, and will be able to determine if hydroplaning is likely to be initiated as well as determine the magnitude of movements once hydroplaning is initiated. The sliding process will be simulated based on the initial geometry of a slope failure, the geology of the nearby seafloor, and the mechanical properties of the slide material (including shear strength, stress-deformation properties and unit weight). The model will describe (1) the variation of the velocity of the slide mass in time and space, and (2) the eventual runout distance and geometry of the slide mass. This information is essential in judging the potential risk associated with submarine slides.
Progress: A block model is being developed to simulate the movement and deformation of a slide mass under the forces applied by the underlying ground and the surrounding fluid. Hydroplaning occurs when the total normal stress is no larger than the excess pore water pressure on the bottom of the block. The excess pore water pressure on the bottom of the block is assumed to be equal to the kinetic pressure along the bottom of the soil mass when hydroplaning happens. The movement of the block in the direction normal to the ground surface determines the thickness of the water film between the slide mass and the ground. The effect of hydroplaning is accounted for by applying hydrodynamic stresses along the bottom surface of the slide mass based on the numerical modeling of the flow around a soil mass. Before hydroplaning is initiated the normal and shear stresses along the bottom of the soil mass are the pressure and friction between the soil and underlying ground.
In the block model, the sliding process including possible hydroplaning is simulated by a Newmark finite difference scheme. During each time increment the acceleration of the block is assumed to be constant and equal to the average value of the accelerations at the beginning and end of the time increment. At the end of each time increment, the 2-D dynamic equilibrium of the block and deformation of the block are computed. The soil body is assumed to remain rectangular although it deforms during movement. The total volume of the soil mass does not change since undrained conditions are assumed. This simple block model can provide insights into the sliding process and the influence of hydroplaning. Computer code is being developed to implement the block model.A more sophisticated numerical model is also being developed to simulate the progression of slide initiation, movement and deposition including possible hydroplaning. In the more sophisticated numerical model, an Euler forward iterative scheme is adopted to trace the evolution of the sliding process. For each time step T, the 2-D dynamic equilibrium of the soil mass is computed using the finite element method. A total stress analysis is conducted and the soil mass is considered linearly elastic. The stresses applied by the surrounding fluid to a slide mass with the geometry and velocity at the time step, T, are computed based on numerical modeling of the water flow around the soil mass. The deformation and acceleration of the slide mass (soil) subjected to the fluid stress field or/and the stresses applied by the underlying ground are computed numerically. During each time increment, dT, the acceleration of the soil mass is assumed to be constant and equal to that at the beginning of the time increment T. This produces a new velocity of the slide mass for the next time step T+dT. A new geometry of the soil body for the next time step is also produced using the geometry and deformation at the previous time step, T.
The onset of hydroplaning and the development of the water film between the slide mass and the ground are modeled by tracing the position of the bottom of the soil mass element by element. The water film exists under the elements where the total normal stress is less than the kinetic pressure along the bottom of the soil mass. The distance between the bottom of the element and the ground surface is taken as the thickness of the water film. The effect of hydroplaning is accounted for by applying hydrodynamic stresses along the bottom surfaces of the elements above the water film based on the numerical modeling of the flow around a soil mass. Before hydroplaning is initiated the normal and shear stresses along the bottom of these elements are the pressure and friction between the soil and underlying ground. A computer program is being developed to implement the more sophisticated numerical model described above.
Publications & Reports:
Hongrui Hu, Stephen G. Wright, Spyros A. Kinnas. (2005). “Hydroplaning of submarine slides and the influence of hydrodynamic stresses.” The 16th International Offshore and Polar Engineering Conference. (submitted and under review – not yet accepted)