--> Abstract: Estimation of Leakage Potential Through Abandoned Wells from CO2 or Acid-Gas Storage Sites in Deep Saline Aquifers, by Michael A. Celia, Jan M. Nordbotten, Stefan Bachu, Dmitri Kavetski, Sarah E. Gasda, and Helge K. Dahle; #90039 (2005)

Datapages, Inc.Print this page

Estimation of Leakage Potential Through Abandoned Wells from CO2 or Acid-Gas Storage Sites in Deep Saline Aquifers

Michael A. Celia1, Jan M. Nordbotten2, Stefan Bachu3, Dmitri Kavetski1, Sarah E. Gasda1, and Helge K. Dahle2
1 Princeton University, Princeton, NJ
2 University of Bergen, Bergen, Norway
3 Alberta Geological Survey, Edmonton, AB

One major concern associated with large-scale geological storage of CO2 is the potential for leakage into other subsurface formations, and eventually to the atmosphere. In mature basins like many of those in North America, thousands to millions of deep wells have been drilled to date. These wells represent possible leakage pathways for injected CO2, hence any leakage analysis must include a quantitative assessment of these wells. Injection dynamics, plume evolution, interactions and possible leakage of the injected CO2 along existing wells, and the flow of the leaking CO2 into intermediate formations along the length of the well are incorporated into a newly developed mathematical model. Based on well statistics from the Alberta Basin in western Canada, we have simulated leakage over a range of possible conditions. From those simulations we can determine leakage rates and patterns due to leakage along existing wells. Because wells are complex composites of different materials, and these materials degrade with time, effective properties, such as a bulk permeability of the well, are highly uncertain. To accommodate this uncertainty, we have run many simulations, with the well permeabilities assigned from a probability distribution meant to capture uncertainties in well parameters. The results of these simulations allow identification of threshold statistics for well properties, such that leakage will remain below a specified target amount if the statistics of the well properties respect the threshold values. Such analysis can provide regulatory guidance regarding characterization and remediation of existing wells.

AAPG Search and Discovery Article #90039©2005 AAPG Calgary, Alberta, June 16-19, 2005