Fracture network characterization method

1. A method of representing fractures and computing a petrophysical parameter for a simulation model of a subterranean reservoir, said method comprising:

providing, to a processor, measurements that characterize the subterranean reservoir to derive a statistical description of a full fracture network of the subterranean reservoir;

partitioning the statistical description of the full fracture network based at least in part on a criterion;

based on the partitioning of the statistical description of the full fracture network, creating, and storing to a storage device, a discrete fracture network for a portion of the full fracture network;

computing the petrophysical parameter using the stored discrete fracture network and at least a portion of the statistical description of the full fracture network or separately computing the petrophysical parameter using the stored discrete fracture network and using at least a portion of the statistical description of the full fracture network wherein the computing and separately computing using the at least a portion of the statistical description of the full fracture network comprise calculation of stochastic realizations of one or more random variables representing the petrophysical parameter, in part, by deriving a spatial correlation parameter with respect to a separation distance using, from at least a portion of the statistical description of the full fracture network, fracture density and probability density functions of fracture length and of fracture orientation.

2. The method in accordance with claim 1 , wherein the discrete fracture network comprises fractures of the full fracture network expected to have the greatest impact on the reservoir connectivity or permeability.

3. The method in accordance with claim 1 , wherein the criterion is based on at least one parameter selected from a group consisting of size of the fracture, permeability of the fracture, maximum number of fractures to be included into said discrete fracture network, and maximum proportion of a total fracture density to be included into said discrete fracture network.

4. The method in accordance with claim 1 , wherein the statistical description of the full fracture network comprises values and spatial variations of fracture density, of size distribution of fractures, of distribution of fracture orientations and of distribution of fracture apertures.

5. The method in accordance with claim 1 , wherein the statistical description of the full fracture network comprises values and spatial variations of parameters representative of fracture density, size distribution of the fractures, distribution of fracture orientations and distribution of fracture apertures.

6. The method in accordance to claim 1 wherein the separation distance comprises a grid cell to grid cell separation distance for grid cells of the simulation model.

7. The method in accordance to claim 1 further comprising deriving from the at least a portion of the statistical description of the full fracture network the petrophysical parameter while allowing for an arbitrary orientation distribution of the fractures in the at least a portion of the statistical description of the full fracture network.

8. The method of claim 1 wherein the criterion comprises a spatially variable criterion.

9. The method of claim 8 wherein the spatially variable criterion accounts for well and inter-well space of the simulation model.

10. The method of claim 1 wherein the criterion comprises a criterion that depends at least in part on fracture permeability.

11. A method of representing fractures for computing one or more petrophysical parameters for a simulation model of a subterranean reservoir, said method comprising:

providing, to a processor, measurements that characterize the subterranean reservoir to derive, and store to a storage device, a statistical description of a fracture network;

based at least in part on the stored statistical description of the fracture network, deriving a parameter representative of spatial correlation as caused by size of fractures in the fracture network in relation to size of grid cells of the simulation model wherein the parameter is a function of a variance of a fracture density and a proportion of fractures which intersect grid cells separated by a distance h; and

computing the one or more petrophysical parameters based at least in part on the derived parameter representative of spatial correlation.

12. The method in accordance with claim 11 , wherein the parameter is a semivariogram.

13. The method in accordance with claim 11 , wherein the variance of the fracture density and the proportion of fractures intersecting grid cells separated by the distance h are derived from the statistical description of the fracture network.

14. A method of deriving petrophysical parameters for a simulation model of a subterranean reservoir, the method comprising:

providing, to a processor, measurements that characterize the subterranean reservoir to derive, and store to a storage device, a statistical description of a fracture network that comprises fewer than all fractures of a full fracture network of the subterranean reservoir; and

based at least in part on the stored statistical description of the fracture network, computing the petrophysical parameters while allowing for an arbitrary orientation distribution of the fractures in the fracture network wherein the computed petrophysical parameters comprise stochastic realizations of random variables and wherein the computing comprises, in part, deriving a spatial correlation parameter with respect to a separation distance using, from at least a portion of the stored statistical description of the fracture network, fracture density and probability density functions of fracture length and of fracture orientation.

15. The method in accordance with claim 14 , comprising integrating numerically or analytically over the probability density functions of fracture length and of fracture orientation.