Brian L. Ehgartner

Geotechnical studies associated with decommisioning the strategic petroleum reserve facility at Weeks Island, Louisiana: A case history

The first sinkhole at the Weeks Island Strategic Petroleum Reserve (SPR) site was initially observed in May 1992. Concurrent with the increasing dissolution of salt over the mined oil storage area below, it has gradually enlarged and deepened. Beginning in 1994 and continuing to the present, the injection of saturated brine directly into the sinkhole throat some 76 m beneath the ground surface essentially arrested further dissolution, providing time to make adequate preparation for the safe and orderly transfer of crude oil to other storage facilities. This mitigation measure marked the first time that such a control procedure has been used in salt mining; previously all control has been achieved by either in-mine or from-surface grouting. A second and much smaller sinkhole was noticed in early 1995 on an opposite edge of the SPR mine, but with a very similar geological and mine mechanics setting. Both sinkholes occur where the edges of upper 152 m and lower 213 m mined storage levels are nearly vertically aligned. Such coincidence maximizes the tensional stress development, leading to fracturing in the salt. This cracking takes 20 or more years to develop. The cracks then become flow paths for brine incursion, which after time progress into the mined openings. Undersaturated ground water gradually enlarges the cracks in salt through dissolution, leading to eventual collapse of the overlying sand to form sinkholes. Other geologic conditions may also be secondary factors in controlling both mining extent and sinkhole location.

Gas releases from salt

The occurrence of gas in salt mines and caverns has presented some serious problems to facility operators. Salt mines have long experienced sudden, usually unexpected expulsions of gas and salt from a production face, commonly known as outbursts. Outbursts can release over one million cubic feet of methane and fractured salt, and are responsible for the lives of numerous miners and explosions. Equipment, production time, and even entire mines have been lost due to outbursts. An outburst creates a cornucopian shaped hole that can reach heights of several hundred feet. The potential occurrence of outbursts must be factored into mine design and mining methods. In caverns, the occurrence of outbursts and steady infiltration of gas into stored product can effect the quality of the product, particularly over the long-term, and in some cases renders the product unusable as is or difficult to transport. Gas has also been known to collect in the roof traps of caverns resulting in safety and operational concerns. The intent of this paper is to summarize the existing knowledge on gas releases from salt. The compiled information can provide a better understanding of the phenomena and gain insight into the causative mechanisms that, once established, can help mitigate the variety of problems associated with gas releases from salt. Outbursts, as documented in mines, are discussed first. This is followed by a discussion of the relatively slow gas infiltration into stored crude oil, as observed and modeled in the caverns of the US Strategic Petroleum Reserve. A model that predicts outburst pressure kicks in caverns is also discussed.

Allowable pillar to diameter ratio for strategic petroleum reserve caverns.

This report compiles 3-D finite element analyses performed to evaluate the stability of Strategic Petroleum Reserve (SPR) caverns over multiple leach cycles. When oil is withdrawn from a cavern in salt using freshwater, the cavern enlarges. As a result, the pillar separating caverns in the SPR fields is reduced over time due to usage of the reserve. The enlarged cavern diameters and smaller pillars reduce underground stability. Advances in geomechanics modeling enable the allowable pillar to diameter ratio (P/D) to be defined. Prior to such modeling capabilities, the allowable P/D was established as 1.78 based on some very limited experience in other cavern fields. While appropriate for 1980, the ratio conservatively limits the allowable number of oil drawdowns and hence limits the overall utility and life of the SPR cavern field. Analyses from all four cavern fields are evaluated along with operating experience gained over the past 30 years to define a new P/D for the reserve. A new ratio of 1.0 is recommended. This ratio is applicable only to existing SPR caverns.

Analysis of Cavern Shapes for the Strategic Petroleum Reserve

This report presents computational analyses to determine the structural integrity of different salt cavern shapes. Three characteristic shapes for increasing cavern volumes are evaluated and compared to the baseline shape of a cylindrical cavern. Caverns with enlarged tops, bottoms, and mid-sections are modeled. The results address pillar to diameter ratios of some existing caverns in the system and will represent the final shape of other caverns if they are repeatedly drawn down. This deliverable is performed in support of the U.S. Strategic Petroleum Reserve. Several three-dimensional models using a close-packed arrangement of 19 caverns have been built and analyzed using a simplified symmetry involving a 30-degree wedge portion of the model. This approach has been used previously for West Hackberry (Ehgartner and Sobolik, 2002) and Big Hill (Park et al., 2005) analyses. A stratigraphy based on the Big Hill site has been incorporated into the model. The caverns are modeled without wells and casing to simplify the calculations. These calculations have been made using the power law creep model. The four cavern shapes were evaluated at several different cavern radii against four design factors. These factors included the dilatant damage safety factor in salt, the cavern volume closure, axial well strain in the caprock, and surface subsidence. The relative performance of each of the cavern shapes varies for the different design factors, although it is apparent that the enlarged bottom design provides the worst overall performance. The results of the calculations are put in the context of the history of cavern analyses assuming cylindrical caverns, and how these results affect previous understanding of cavern behavior in a salt dome.

Sensitivity of storage field performance to geologic and cavern design parameters in salt domes.

A sensitivity study was performed utilizing a three dimensional finite element model to assess allowable cavern field sizes in strategic petroleum reserve salt domes. A potential exists for tensile fracturing and dilatancy damage to salt that can compromise the integrity of a cavern field in situations where high extraction ratios exist. The effects of salt creep rate, depth of salt dome top, dome size, caprock thickness, elastic moduli of caprock and surrounding rock, lateral stress ratio of surrounding rock, cavern size, depth of cavern, and number of caverns are examined numerically. As a result, a correlation table between the parameters and the impact on the performance of a storage field was established. In general, slower salt creep rates, deeper depth of salt dome top, larger elastic moduli of caprock and surrounding rock, and a smaller radius of cavern are better for structural performance of the salt dome.

Three dimensional simulation for Big Hill Strategic Petroleum Reserve (SPR).

3-D finite element analyses were performed to evaluate the structural integrity of caverns located at the Strategic Petroleum Reserves Big Hill site. State-of-art analyses simulated the current site configuration and considered additional caverns. The addition of 5 caverns to account for a full site and a full dome containing 31 caverns were modeled. Operations including both normal and cavern workover pressures and cavern enlargement due to leaching were modeled to account for as many as 5 future oil drawdowns. Under the modeled conditions, caverns were placed very close to the edge of the salt dome. The web of salt separating the caverns and the web of salt between the caverns and edge of the salt dome were reduced due to leaching. The impacts on cavern stability, underground creep closure, surface subsidence and infrastructure, and well integrity were quantified. The analyses included recently derived damage criterion obtained from testing of Big Hill salt cores. The results show that from a structural view point, many additional caverns can be safely added to Big Hill.

3-D CAVERN ENLARGEMENT ANALYSES

Three-dimensional finite element analyses simulate the mechanical response of enlarging existing caverns at the Strategic Petroleum Reserve (SPR). The caverns are located in Gulf Coast salt domes and are enlarged by leaching during oil drawdowns as fresh water is injected to displace the crude oil from the caverns. The current criteria adopted by the SPR limits cavern usage to 5 drawdowns (leaches). As a base case, 5 leaches were modeled over a 25 year period to roughly double the volume of a 19 cavern field. Thirteen additional leaches where then simulated until caverns approached coalescence. The cavern field approximated the geometries and geologic properties found at the West Hackberry site. This enabled comparisons are data collected over nearly 20 years to analysis predictions. The analyses closely predicted the measured surface subsidence and cavern closure rates as inferred from historic well head pressures. This provided the necessary assurance that the model displacements, strains, and stresses are accurate. However, the cavern field has not yet experienced the large scale drawdowns being simulated. Should they occur in the future, code predictions should be validated with actual field behavior at that time. The simulations were performed using JAS3D, a three dimensional finite element analysis code for nonlinear quasi-static solids. The results examine the impacts of leaching and cavern workovers, where internal cavern pressures are reduced, on surface subsidence, well integrity, and cavern stability. The results suggest that the current limit of 5 oil drawdowns may be extended with some mitigative action required on the wells and later on to surface structure due to subsidence strains. The predicted stress state in the salt shows damage to start occurring after 15 drawdowns with significant failure occurring at the 16th drawdown, well beyond the current limit of 5 drawdowns.

Laboratory evaluation of damage criteria and permeability of Big Hill salt.

To establish strength criteria of Big Hill salt, a series of quasi-static triaxial compression tests have been completed. This report summarizes the test methods, set-up, relevant observations, and results. The triaxial compression tests established dilatant damage criteria for Big Hill salt in terms of stress invariants (I{sub 1} and J{sub 2}) and principal stresses ({sigma}{sub a,d} and {sigma}{sub 3}), respectively: {radical}J{sub 2}(psi) = 1746-1320.5 exp{sup -0.00034I{sub 1}(psi)}; {sigma}{sub a,d}(psi) = 2248 + 1.25 {sigma}{sub 3} (psi). For the confining pressure of 1,000 psi, the dilatant damage strength of Big Hill salt is identical to the typical salt strength ({radical}J{sub 2} = 0.27 I{sub 1}). However, for higher confining pressure, the typical strength criterion overestimates the damage strength of Big Hill salt.

Three dimensional simulation for bayou choctaw strategic petroleum reserve (SPR).

Three dimensional finite element analyses were performed to evaluate the structural integrity of the caverns located at the Bayou Choctaw (BC) site which is considered a candidate for expansion. Fifteen active and nine abandoned caverns exist at BC, with a total cavern volume of some 164 MMB. A 3D model allowing control of each cavern individually was constructed because the location and depth of caverns and the date of excavation are irregular. The total cavern volume has practical interest, as this void space affects total creep closure in the BC salt mass. Operations including both cavern workover, where wellhead pressures are temporarily reduced to atmospheric, and cavern enlargement due to leaching during oil drawdowns that use water to displace the oil from the caverns, were modeled to account for as many as the five future oil drawdowns in the six SPR caverns. The impacts on cavern stability, underground creep closure, surface subsidence, infrastructure, and well integrity were quantified.

Analysis of salt and casing fracture mechanisms during cavern integrity testing for SPR salt caverns.

This report presents computational analyses to evaluate the risk of Cavern Certification Testing on casing damage and salt fracturing. This deliverable is performed in support of the U.S. Strategic Petroleum Reserve. Several models have been built utilizing either a 1-cavern or 19-cavern, 30-degree wedge of symmetry as has been used on previous West Hackberry (Ehgartner and Sobolik, 2002) and Big Hill (Park et al., 2005) analyses. Various stages of complexity have been added to each model: mesh generation for both 1-cavern and 19-cavern fields; the inclusion of the small-diameter well going from the surface to the top of the caverns; and the addition of steel casing and cement liners to the well surfaces. A stratigraphy based on the Bayou Choctaw site has been chosen for modeling, and the caverns are modeled as equivalent cylinders to simplify the calculations. Salt dilation, as defined by two separate dilation criteria, has been generated during simulated workover routines and cavern integrity tests, but only for the models that include the steel casing and cement liners. The addition of these features generates the tensile and shear stress conditions in the model necessary to cause stress states that exceed dilation criteria. These predictions have been made using the power law creep model. The results show that the salt can be damaged during high pressure change conditions in the wells and caverns.