An Analysis of the Caprock Failure at Joslyn

Abstract

Abstract A significant caprock failure occurred on the Joslyn SAGD property in 2006, which has had a wide impact on the approval process for future SAGD projects. Two reports were released by the Alberta Government: "Total E&P Canada Ltd., Surface Steam Release of May 18, 2006, Joslyn Creek SAGD Thermal Operation, ERCB Staff Review and Analysis, February 11th, 2010" and "Summary of Investigations into the Joslyn May 18th, 2006 Steam Release, Total E&P Canada Ltd.". The latter report is very large. A number of potential mechanisms are postulated without definitive resolution. The most likely failure suggested involved transmission of fluids up a 50,000 mD chimney and a pancake shaped lens of high pressure steam that resulted in a shear failure of the caprock. The author believes that this scenario is unlikely from a geological and heat transfer perspective. There are other anomalies. Peak stresses in most caprock coupled reservoir-geomechanical simulations normally peak between 3 to 7 years after start-up. However, the Joslyn failure occurred immediately after conversion from partial SAGD to full SAGD, in which the producing well was killed and a PCP pump was run. There are other issues that should be considered that include: How hydraulic fractures are initiated and propagated. The literature shows us: computer design programs that predict fracture propagation and shape,analytical solutions that also show expected fracture morphology; and,physical examples which have been dug up or coredIn addition the geological record provides excellent analogue information on the morphology of high pressure intrusions. These various types of data will be compared to the observed morphology derived from the 3D seismic. Earlier work by Edmunds and Good has shown that extremely transient conditions exist in SAGD wells that result in "geysering" or slugging within the producing well. The energy available from water and steam is enormous. The nuclear and power industry previously made these similar discoveries after a series of catastrophic failures of piping systems. Analytical and computer simulation tools from the nuclear industry will be used to show that extremely high pressure transients, many times over fracture initiation pressures, can be expected. This is the same mechanism that resulted in the catastrophic failure of MEG Energy’s Christina Lake main steam distribution line. In summary, there is every reason to believe that when hot steam was injected into a cooled (water) condensate filled well; very large pressure transients can be expected from phase changes within the piping that should result in a frac to surface.