Modeling Production from Shale

Mitchell and his team of geologists and engineers began working on the shale challenge in 1981, trying different combinations of processes and technologies before ultimately succeeding in 1997 with the use of a “slick-water” frac that made Barnett Shale economical to develop and in turn changed the future of the US natural gas industry [7]. Continuing on Mitchell’s success progress followed a path that included horizontal wells, multi-cluster, multistage, hydraulic fracturing of horizontal wells and pad drilling, and the rest is history. The success in overcoming the technical difficulties to unlock the huge potentials of oil and gas production from shale is very much tied to an integration of long lateral horizontal drilling, coupled with multistage, multi-cluster hydraulic fracturing that initiates new fractures while activating a system of natural fracture networks in shale. This system of highly permeable conduits introduces the highly pressurized shale formation to a lower pressure, causing a pressure gradient that triggers the flow of oil and natural gas to the surface at very high rates, albeit, with steep decline. Looking at the list of most important technological innovations that have made production from shale possible one can see the challenges that are associated with understanding and modeling and eventually optimizing the production of hydrocarbon from shale. So let us examine what are the realities that we are aware of, and how we go about modeling them in our traditional techniques. First and foremost is the fact that shale is naturally fractured. The natural fractures are mostly sealed by material that have precipitated in them and/or reside in the natural fracture as a result of chemical reactions of the material present in the fractures throughout the geologic time. The only way we can detect presence of the system of natural fractures in the area being developed is by detecting (using logs, video, etc.) the presence of the intersection of these natural fractures with the wellbore. Even when we are successful in detecting the presence of natural fractures, we only have indications about their width (openings) and direction. We