Fracture Characterization Methods for Tight Gas Sandstones
By
Stephen Laubach1, Randall Marrett1, Jon Olson1
(1) University of Texas at Austin, Austin, TX
Ongoing research is aimed at enhancing exploration and production for
fractured reservoirs by development and integration of emerging and existing
technologies for the observation, prediction, and fluid-flow modeling of natural
fractures. Macroscopic fractures produce the largest impact on fluid flow
through fractured rock, however they are orders of magnitude
less abundant than
microscopic fractures. Macro-fractures in subsurface reservoirs typically are
poorly represented by data acquired with conventional techniques. Due to the
abundance of microfractures, they can be well studied even in small samples from
the subsurface. We are exploring the hypotheses that micro- and macro-fractures
are different size fractions of the same fracture sets, and that microfractures
can be used to predict the critical characteristics (in terms of fluid flow) of
associated macrofractures.
Previously invisible microfractures are readily observed and characterized when their cathodoluminescence is imaged using scanning electron microscopy. This facilitates determination of the orientations, timing (relative to diagenetic events), and sizes of the numerous microfractures typically present in prospective fractured reservoirs. These observations may be made systematically on a bed-by-bed basis. Orientations and timing of microfractures commonly compare favorably with those of associated conductive macro-fractures.
Microfractures are sufficiently abundant in the numerous fractured units we
have studied that the size distributions can be readily quantified. Under
special circumstances, the sizes (i.e., mechanical apertures and/or lengths) of
both micro- and macro-fractures can be reliably measured in the same fractured
rock volume. The spatial frequency of fractures, as a function of fracture size,
follows power-law distributions over at least 4 to 5 orders of magnitude
in
these cases. This confirms that microfracture sizes can be used to
quantitatively predict spatial frequencies of associated macro-fractures.