Grantee Research Project Results
2010 Progress Report: A Hierarchical Modeling Framework for Geological Storage of Carbon Dioxide
EPA Grant Number: R834385Title: A Hierarchical Modeling Framework for Geological Storage of Carbon Dioxide
Investigators: Celia, Michael A.
Institution: Princeton University
EPA Project Officer: Aja, Hayley
Project Period: December 1, 2009 through November 30, 2013
Project Period Covered by this Report: September 1, 2009 through August 31,2010
Project Amount: $870,009
RFA: Integrated Design, Modeling, and Monitoring of Geologic Sequestration of Anthropogenic Carbon Dioxide to Safeguard Sources of Drinking Water (2009) RFA Text | Recipients Lists
Research Category: Drinking Water , Water
Objective:
The project has four general goals and objectives to: (1) develop and enhance a set of analytical and numerical computational tools for simulation of CO2 injection, migration, and possible leakage, (2) develop a "hierarchical modeling framework" within which different computational tools can be combined to produce effective hybrid models, (3) provide guidance on level of model complexity for different scenarios, and (4) provide simple-to-use web-based interfaces for different versions of the codes we develop.Progress Summary:
During the first year of the project, we have made significant progress on all four of these objectives. A brief overview is provided herein.
In general, we have strengthened our overall modeling and analysis approach through a systematic analysis of length and time scales, and a thorough discussion of how these scales guide the kinds of models that are appropriate for a specific situation. This leads to a general concept of multi-scale modeling, with the scales at which the model is resolved being related directly to the practical questions being asked. An example is provided below in Figure 1, which is taken from Celia and Nordbotten (2010). Of particular significance is the time scale required for vertical segregation of the fluids (assumed to be CO2 and brine) due to buoyancy. For times large compared to this time scale, we can assume vertical segregation, and the associated vertical equilibrium of pressures. This allows for vertical integration of the governing equations, which greatly simplifies the equations to be solved. The multi-scale nature of the approximations we have developed, in particular the treatment of saturation profiles in the vertical direction, is discussed in Celia and Nordbotten (2010). This also gives rise to the concept of capillary fringe and the ability to incorporate local-scale capillary effects into the larger scale models. A thick capillary fringe can have very significant consequences on the flows, a topic that we are currently exploring in detail and for which we expect to complete a manuscript shortly.
A brief summary of progress on all four of the objectives is given as follows:
Objective 1: We have expanded both analytical and numerical solutions in significant ways. In terms of numerical solutions, we have included diffuse leakage across caprock formations, and we have relaxed our initial assumption of a sharp interface between the two fluids through the inclusion of a capillary fringe. The diffuse leakage is now working, the capillary fringe is still being tested. In addition, we have almost completed work on analytical solutions for leakage along a conductive fault zone. This analytical solution will serve as a sub-scale correction within coarse-scale numerical models.
Objective 2: We are in the process of adding two local-scale "hybrid" features to our existing models and are working on a much more general third one (note that in this case we use "hybrid" to mean an overall model that uses both numerical and analytical solutions). The two local-scale corrections are (1) a correction for interface location and pressure in grid cells that contain injection wells, and (2) the local-scale correction for leakage along a fault as discussed earlier. More generally, we continue to work on possible algorithms that will allow domains of interest to be broken into regions in which numerical solutions are calculated, and regions where analytical solutions are used, with a seamless boundary between them. We hope to have a working prototype for such a code with a restriction to only one type of solution used in any formation (geological layer).
Objective 3: As noted earlier, we have put together an analysis of space and time scales that allows multi-scale modeling ideas to be developed transparently. Into this analysis, we have embedded explicitly the questions that are to be answered. In this way, we are trying to match the model choice to the questions being asked. The following figure shows our initial estimates of length and time scales relevant for the CO2 injection problem.
Figure 1 - Proceses and Spatial Scales (from Celiana dn Nordbotten (2010)).
Figure 2 - Processes and Temporal Scales (from Celia and Nordbotten (2010)).
In addition to this scaling analysis, we are also using the models we have developed to determine optimal placement of monitoring wells to detect leakage. We have a paper that is under revision (see Nogues et al. (2011) in publication list), wherein we show how the kinds of very efficient models we develop can allow many tens of thousands of simulations to be performed in a Monte Carlo sense. This can then be used directly to determine monitoring strategies. We are initiating a similar application study based on brine extraction and the more general idea of water management in conjunction with carbon management. The papers by Court et al. present these ideas.
Objective 4: We continue to work on a web-based interface that would allow any user to have access to simple versions of the software we have developed. We currently have a live site where the CO2 plume and the pressure increases can be calculated and visualized. We also have a second option to analyze the behavior of a system with one injection well, one leaky well, and two aquifers separated by a caprock. The web interface can be found at http://monty.princeton.edu/CO2interface/.
2. Key Personnel and Status Change
We have not had any changes in key personnel. One Masters student, Adam Janzen, has completed his degree. His MSE thesis is listed in our publications list.
3. Expenditures to Date:
Expenditures are within the expected range, and we do not anticipate any changes to the budget.
4. Data Quality Assurance:
Preliminary data acquisition has started for real-world GS project sites. Data we have obtained undergoes consistency checks, wherein: geometric information about formations is self-consistent; that well data is both self-consistent (wells are at discrete non-overlapping locations) and consistent with other information (depths consistent with depths of formations); and that measured temperatures are consistent with reported geothermal gradients. We have yet to define tolerances, determined from published standards and literature on empirical value-ranges, for when data are considered unreliable.
5. Significance of Results to Date:
A summary of results has been provided in Section 1 of this report. Here, we highlight the most significant accomplishments to date: (1) development of a multi-scale framework that serves as a guide to the kinds of models that are most appropriate for a given question or problem that is being posed. This relies on scaling arguments and the associated mathematics of multi-scale modeling. (2) A live web interface that can provide insights into both plume behavior and pressure build-up for injection of CO2 into a deep saline aquifer. We now also have an option to simulate a relatively simple leakage problem involving one injection well, one leaky well, and two aquifers. (3) Continued development of hybrid models where local-scale behavior is represented by analytical solutions which are then embedded into coarser-scale numerical simulators. The numerical simulators themselves are simplified by using vertical integration of the governing equations, coupled with a sub-scale (in this case vertical) reconstruction of the details of the saturation profile along z.
Future Activities:
We will continue to work on all four objectives, with a focus on the hybrid multi-scale models and on applications to specific field locations, including continued work in the Alberta Basin and new applications in the Illinois Basin. We will also continue to improve our web interface, and will make a significant effort to interact more strongly with our EPA colleagues, especially in Athens.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 25 publications | 18 publications in selected types | All 18 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Celia MA, Nordbotten JM, Court B, Dobossy M, Bachu S. Field-scale application of a semi-analytical model for estimation of CO2 and brine leakage along old wells. International Journal of Greenhouse Gas Control 2011;5(2):257-269. |
R834385 (2010) R834385 (2011) R834385 (2012) R834385 (Final) |
Exit Exit Exit |
|
Gasda SE, Nordbotten JM, Celia MA. Vertically-averaged approaches for CO2 migration with solubility trapping. Water Resources Research 2011;47(5):W05528. |
R834385 (2010) R834385 (2011) R834385 (2012) R834385 (Final) |
Exit Exit Exit Exit |
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.