skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Discrete Fracture Network Models for Risk Assessment of Carbon Sequestration in Coal

Technical Report ·
DOI:https://doi.org/10.2172/941131· OSTI ID:941131

A software package called DFNModeler has been developed to assess the potential risks associated with carbon sequestration in coal. Natural fractures provide the principal conduits for fluid flow in coal-bearing strata, and these fractures present the most tangible risks for the leakage of injected carbon dioxide. The objectives of this study were to develop discrete fracture network (DFN) modeling tools for risk assessment and to use these tools to assess risks in the Black Warrior Basin of Alabama, where coal-bearing strata have high potential for carbon sequestration and enhanced coalbed methane recovery. DFNModeler provides a user-friendly interface for the construction, visualization, and analysis of DFN models. DFNModeler employs an OpenGL graphics engine that enables real-time manipulation of DFN models. Analytical capabilities in DFNModeler include display of structural and hydrologic parameters, compartmentalization analysis, and fluid pathways analysis. DFN models can be exported to third-party software packages for flow modeling. DFN models were constructed to simulate fracturing in coal-bearing strata of the upper Pottsville Formation in the Black Warrior Basin. Outcrops and wireline cores were used to characterize fracture systems, which include joint systems, cleat systems, and fault-related shear fractures. DFN models were constructed to simulate jointing, cleating, faulting, and hydraulic fracturing. Analysis of DFN models indicates that strata-bound jointing compartmentalizes the Pottsville hydrologic system and helps protect shallow aquifers from injection operations at reservoir depth. Analysis of fault zones, however, suggests that faulting can facilitate cross-formational flow. For this reason, faults should be avoided when siting injection wells. DFN-based flow models constructed in TOUGH2 indicate that fracture aperture and connectivity are critical variables affecting the leakage of injected CO{sub 2} from coal. Highly transmissive joints near an injection well have potential to divert a large percentage of an injected CO{sub 2} stream away from a target coal seam. However, the strata-bound nature of Pottsville fracture systems is a natural factor that mitigates the risk of long-range leakage and surface seepage. Flow models indicate that cross-formational flow in strata-bound joint networks is low and is dissipated by about an order of magnitude at each successive bedding contact. These models help confirm that strata-bound joint networks are self-compartmentalizing and that the thick successions of interbedded shale and sandstone separating the Pottsville coal zones are confining units that protect shallow aquifers from injection operations at reservoir depth. DFN models are powerful tools for the simulation and analysis of fracture networks and can play an important role in the assessment of risks associated with carbon sequestration and enhanced coalbed methane recovery. Importantly, the stochastic nature DFN models dictates that they cannot be used to precisely reproduce reservoir conditions in a specific field area. Rather, these models are most useful for simulating the fundamental geometric and statistical properties of fracture networks. Because the specifics of fracture architecture in a given area can be uncertain, multiple realizations of DFN models and DFN-based flow models can help define variability that may be encountered during field operations. Using this type of approach, modelers can inform the risk assessment process by characterizing the types and variability of fracture architecture that may exist in geologic carbon sinks containing natural fractures.

Research Organization:
Geological Survey Of Alabama
Sponsoring Organization:
USDOE
DOE Contract Number:
FC26-05NT42435
OSTI ID:
941131
Country of Publication:
United States
Language:
English