In November 2005, participants at the Workshop on Geological Storage of CO2 at Princeton University agreed on the need for a common test problem to assess various models to simulate the fate of CO2 injected into the subsurface. Alberta Geological Survey offered to make available the data for the Wabamun Lake area in Alberta, Canada, which were assembled to develop a comprehensive model for studying CO2 geological storage.
The Wabamun Lake area, southwest of Edmonton in central Alberta, was selected as the test area because a variety of favourable conditions identified it as a potential site for future, large-scale CO2 injection. Several large, industrial CO2 point sources are in the area, resulting in short transportation distances of the captured gas. Various deep saline formations with sufficient capacity to accept and store large volumes of CO2 in supercritical phase exist at the appropriate depth and are overlain by thick confining shale units. Most importantly, a wealth of data exist (i.e., stratigraphy, rock properties, mineralogy, fluid composition, formation pressure, information about well completions, etc.), collected by the petroleum industry and submitted to the Alberta Energy and Utilities Board. For these reasons, the Wabamun Lake area is an ideal location to characterize a CO2 storage site and analyze the potential risks.
The transient pressure measurements taken at different times during drillstem tests (DSTs) were extrapolated to the formation pressure by sources specialized in DST interpretation (Canadian Institute for Formation Evaluation - CIFE, HydroFax) using the Horner method. Based on the built-up pressure, the industry qualifies the quality of a DST using flags in quality-descending order, such as A, B, C, D, E and F, which signify, respectively, best quality, nearing stabilization, possible plugging, questionable, low permeability and low pressure, and low permeability and high pressure. The accuracy as specified by various manufacturers ranges between 0.025% and 0.25% of the full scale for newer pressure gauges (1970-2000) and is as low as 0.5% for gauges from the 1950s and 1960s. This translates to a maximum error of approximately F200 kPa (accepted pressure difference between first and second shut-in measurement for a C-quality DST) or F20 m of hydraulic head. On the other hand, errors introduced by the mechanical procedure of the test are much larger, and empirically determined errors in pressure measurements generally increase with depth and range between 170 and 650 kPa.
To minimize inaccuracies due to methodology error, only DSTs of A, B and C quality and Horner-extrapolated pressures are in this dataset for the Wabamun Lake area. In addition, DSTs were culled that were suspected to have been influenced by production from nearby wells, hence not reflecting the original "virgin" formation pressure.