Estimation of fracture density by coupling regression and hypothesis testing.

@inproceedings{ViseurGM2014,
 abstract = { Fracture density is an important parameter for any characterization of fractured reservoirs. Many stochastic simulation algorithms that generate fracture networks rely on the determination of a fracture density to populate the reservoir zones with individual fracture surfaces. Many approaches exist to obtain fracture density from well data or outcrop analogues. A particular case of fracture study corresponds to fracture corridors. In this kind of studies, a difficulty is to objectively separate fracture corridors characterized by a high fracture density from zones where fractures are present but more sparsely.
In this paper, an approach is proposed to distinguish different zones characterized by different fracture densities. This means to be able to compute local fracture densities while distinguishing which densities could be considered as different. The proposed approach relies on 1D acquisitions of fracture locations, which then allows this protocol to be applied on outcrops but also on well data. It uses plots between fracture number and distances. The key point is to couple regression and hypothesis testing. The regression aims at computing local average fracture density and the hypothesis testing at individualizing areas that are characterized by statistically different densities. An application on a dataset collected on a calcareous quarry (Calvisson, France) is presented to illustrate the interest of this approach. },
 author = { Viseur, Sophie AND Châtelée, Sebastien AND Lamarche, Juliette },
 booktitle = { Proc. 34th Gocad Meeting, Nancy },
 title = { Estimation of fracture density by coupling regression and hypothesis testing. },
 year = { 2014 }
}