Proceedings of the International Workshop on Seismic Design and Reassessment of Offshore Structures

Abstract

This Executive Summary presents the major conclusions and recommendations of the International Workshop on Seismic Design and Reassessment of Offshore Structures. Full versions of the text of all invited lecturers and complete working group reports are contained elsewhere in these Proceedings. Site Seismic Hazard and Ground Motion Continuous improvements have been made in the state-of-the-art of estimation of seismic ground motion for design and reassessment of offshore platforms. C. B. Crouse of Dames & Moore provides an overview of methods used to determine seismic design parameters for offshore structures. A site﷓specific analysis is generally performed to develop seismic design parameters for a particular site. The first step is to define the seismic sources in the site region, and develop a model of the attenuation of ground motion that is appropriate for the region. Next, probabilistic and/or deterministic analyses are performed to estimate the ground motion parameters. For new designs, both approaches are commonly used. For reassessment and requalification, a probabilistic approach is generally preferred. There are three basic inputs needed for a probabilistic approach; 1) identification of earthquake sources in the site region, 2) definition of the earthquake recurrence for each source, and 3) definition of the attenuation of earthquake ground motion, usually expressed as a function of distance from the source and earthquake magnitude. An example is given of a typical offshore platform located in a subduction zone environment. This example is used to illustrate some of the more important issues in the determination of seismic parameters. These issues include: extrapolation from available data to larger design level events, inclusion of basin effects on soil amplification, the relationship between vertical and horizontal earthquake motions, and the effect of the water column on vertical acceleration. It is concluded that estimation of ground motion for offshore structural design and reassessment is a challenging process involving several technical disciplines including seismology, geology, geotechnical and structural engineering, as well as political considerations. Unfortunately, many uncertainties still exist in the specification of seismic design parameters based on site-specific seismic hazard analyses. The Working Group on Site Seismic Hazard and Ground Motion, co﷓chaired by Allin Cornell of Stanford University and Paul Somerville of Woodward Clyde Consultants, makes the following major conclusions and recommendations: 1. There is a critical need for more recorded earthquake seafloor motion data in regions where offshore facilities are proposed or currently constructed. 2. There is a need for the application of the latest state-of-the-art techniques in the prediction of ground motion, especially in the period range of 1-4 seconds. 3. Further research should be undertaken related to defining the subsurface input reference point for strong ground motion. 4. Deep geophysical data is needed to better characterize offshore seismic sources. Design, Reassessment And Requalification Robert Bea of the University of California observes that the primary objectives of a seismic design methodology are to assure that a new offshore structure will have sufficient strength and ductility to satisfy its intended purposes without undue expense or risk. He argues that the primary focus of the platform designer should be on the response that develops after first significant yielding occurs in the structural elements and components that comprise the structural system of the platform, and the damage states that can lead to collapse of the platform. In the design criteria, member resistance and factors should be chosen to provide acceptable reliability against significant damage or collapse. A strength level earthquake is not anticipated to induce significant damage or inelastic response in the structural elements. Static push-over analyses have been used extensively to demonstrate the capacity and ductility of offshore platforms. The margin of safety beyond the elastic performance requirement is often expressed by a platform Reserve Strength Ratio (RSR) which is the ratio of the maximum lateral load capacity of the platform to a reference load induced by the strength level earthquake. Bea has found that platforms designed according to current American Petroleum Institute (API) guidelines generally have an RSR 2. In contrast to a seismic design methodology, a seismic requalification methodology should provide a set of processes and parameters to insure that an existing platform will be able to develop acceptable performance during its proposed service period. Bea summarizes his "comprehensive" seismic requalification methodology. In this methodology, the judgment of fitness for purpose is based on the total probability of failure of the platform system, including topside equipment and operations, the potential costs, and the consequences associated with failure. If a platform is judged not fit-for-purpose, guidance is provided as to how to improve its performance characteristics. Two different approaches may be used to define the fitness-for-purpose criteria in this methodology: "utility optimization" or "standard of practice." The utility optimization approach is based on an evaluation of the positive and negative utilities associated with different risks. Utility may be measured in a variety of ways. Costs and benefits are commonly measured in monetary terms. The objective is usually expressed in terms of minimization of total expected cost. The standard-of-practice approach is based on current design and requalification decisions which have been made for other platforms. The underlying premise of this approach is that, over time, the profession, industry, and regulatory agencies have defined acceptable combinations of probabilities of failure and consequences associated with failure. In the standard-of-practice approach, the decision on whether a structure is fit-for-purpose is based upon its probability of failure and total cost of failure as compared to other new and existing platforms. Straight line "acceptable" and "marginal" boundaries have been proposed in a log-log total probability vs. total cost space. Bea expresses the opinion that none of the available seismic requalification methodologies are as highly developed as seismic design methodologies. It is argued that the industry has the necessary technology to develop definitive engineering guidelines from the available requalification methodologies and that high priority should be given to developing such guidelines. It is further asserted that generally accepted fitness-for-purpose criteria that are dependent on consequences and likelihoods of failure need to be developed for both new and existing offshore structures. David Wisch of Texaco Corporation presents the results of a 1992 API requalification project. A panel, consisting of Wilfred Iwan (Chairman), Allin Cornell, George Housner and Charles Thiel, was constituted by the API to recommend an approach to seismic requalification of offshore structures. The API Panel's strategy is based on the establishment of separate acceptable levels of risk for life safety and environmental hazards. The Panel reasoned that life safety and environment hazards are essentially different in character and need not, indeed cannot, be measured by a common scale such as dollars. The Panel's acceptable risk for life safety is based on the concept that the risk to a worker on an offshore platform should be comparable to that for a worker in a similar facility onshore. This was believed to be a publicly acceptable level of risk. Acceptable environmental risk is based upon the risk of spills that occur from other sources in the same region which are deemed publicly acceptable. The Panel uncovered no safety or environmental issues associated with the seismic safety reassessment process that indicated platforms should be subject to risk criteria more restrictive than those for land-based industrial facilities. The Panel argued that the focus of seismic safety reassessment should be on limiting, to an acceptable level, the risk due to catastrophic impacts of earthquakes, where a catastrophic impact is defined as one that has unacceptable large life and/or environmental safety consequences. It was further argued that offshore facilities should have more rigorous site hazard and engineering behavior analysis than onshore facilities in order to achieve these goals, even though they have comparable quantitative risk limits. The Panel recommend that decisions related to cost and economic return be left to platform owners and not be included in reassessment or requalification requirements imposed by regulatory agencies. The Panel concluded that use of the design and analysis guidelines presented in the API recommended practice document, RP 2A, 19th edition, will produce a structure with life safety comparable to that of well-engineered structures onshore provided that: 1) the hazard study and structural analyses are peer reviewed, 2) the seismic hazards are determined in accordance with strength and ductility level earthquakes with 200 and 1,000 year return periods, respectively, 3) a ductility level earthquake analysis is always performed, and 4) proper allowance is made for life﷓safety risks associated with platform appurtenances. For structures which do not meet the guidelines of RP 2A, the Panel determined that an objective of an annual probability of failure (collapse) of less than 1 x 10 -3 would be consistent with providing a level of life safety comparable to that of onshore structures. The Panel also concluded that median value results from hazard analyses should be used in any probabilistic analyses to verify satisfaction of this objective. For Southern California waters, the Panel recommended an environmental performance objective of a release of no more than 2,000 barrels from any possible source including wells, pipelines, and onboard storage. Wisch notes that the API Panel study is only one of a series of steps being undertaken by the offshore industry. Some of the questions that are being and will be answered by these studies include: 1) whether platforms can be categorized as to their expected response based on type or age, 2) whether an effective screening procedure can be developed to avoid detailed case-by-case assessment of offshore structures, 3) how independent review and peer review should best be incorporated into the reassessment process, 4) the different issues that must be addressed in reassessment outside the U.S., and 5) how acceptability criteria should relate to recommended industry practice, codes, regulatory requirements, and economic considerations. The Working Group on Design, Reassessment and Requalification, co-chaired by Kris Digre of Shell Oil Co. and William Ibbs of U.C. Berkeley, arrives at the following conclusions and recommendations: 1. The strength level and ductility level earthquakes employed in current API recommended practice need to be more precisely specified; perhaps tied to some other criteria. 2. A ductility level analysis should always be required. 3. Accelerographs should be installed in all offshore structures for the purpose of acquiring data on structural performance during actual earthquake excitation. 4. Technology should be further developed and shared regarding the following: joint and member capacities (including in-frames), mass coefficients for wave loading, effects of marine growth on structural response, and the relation between static and dynamic load analysis procedures. 5. A majority of the working group felt that there should be a separation of life safety, environmental consequences, and economic decisions for requalification but a consensus on this matter could not be reached. 6. A full probability-of-failure risk analysis should be performed rather than a mere determination of the survival of a structure for a given return period earthquake. 7. A strength level earthquake analysis need not be performed for requalification as long as a ductility level analysis is performed. 8. A careful peer review should be conducted of both the structural and seismic hazard elements of any design, reassessment, or requalification process. More work is needed to develop an appropriate peer review process and guidelines for reviewer qualification. 9. Research is needed to more precisely define manned and unmanned operations, and catastrophic consequences of environmental pollution. 10. There is a need to build greater consensus on the appropriate criteria for safety goals for design and requalification. Further research is needed on such issues as: specification of life safety and environmental safety objectives, the use engineering judgment, and how to properly incorporate the consequences of failure into any performance objective.

Files

Additional details