Shale Well Frac Stages and Lateral Length. How Many? How Long?

With the shale revolution in full swing, I’d like to highlight how research institutions such as Stanford are playing catch-up to industry activity and how work I’m doing can help operators develop shale resources efficiently. In the past, new technologies in the oil and gas space such as EOR and thermal recovery moved slowly from small science-projects to commercial development. Academics in engineering department s worked in tandem with large companies to evaluate these technologies along the way. The development in shale, however, evolved in a markedly different manner with smaller operators experimenting in the field rather than a research lab. Now that the ramifications of the independents’ success is well understood, there’s been a rush for increased technical understanding of shale production with large funding for research programs at petroleum engineering departments around the country.

[Posted by contributor Kurt Wilson, a petroleum engineering student pursuing a master’s degree at Stanford University.]

When I set out to design my research project, I wanted to capture a central problem facing companies in shale plays: How do I develop this asset efficiently? With a contiguous resource, geological risk is minimal and capital efficiency is paramount. The current practice often utilizes cookie-cutter well designs with the occasional down-spacing trial. Operators sell investors on type curves that reflect the average results of their well designs, but rarely tailor those wells to any knowledge of local geology.  My work involves using simple reservoir models along with some economics to determine the best well designs in shale gas fields. The goal is to help operators move away from the trial-by-error approach to something more technically grounded.

In a simple example, I build a two square mile reservoir model with homogenous geological properties representative of the Barnett shale. I have combined this model with optimization software that allows the systematic testing of thousands of possible well combinations.  In this example, I place five shale gas wells in this field and allow the software to determine the optimal well lengths and number of fracture stages based on a calculation of net present value. Capital costs were estimated from publically available data on drill cost per foot, completion costs per stage, etc. To add variability to the example, I placed a slight pressure gradient across the field, which leads one side of the reservoir to have more gas in place (GIP). If the software is worth its salt, it will place the shale gas wells in this higher quality rock.

This figure shows the initial guess and the optimized case for this example. In the base case, the five wells are equally spaced with 3000 ft. horizontal lengths and eight fracture stages apiece. In the optimized case, the five wells are pushed toward the section with higher initial pressure (red area). Additionally, the software calls for a differing number of stages in the wells, with the best well fractured with 18 stages. Importantly, the economics of this field using a $3.50/mcf gas price were improved substantially, with NPV-10 increasing from $2.5MM in the base case to over $14MM in the optimized scenario. Clearly this is a toy problem, but it represents a method for designing wells in shale plays that is technically grounded and can incorporate knowledge of local geology. Additionally, this method can quickly adapt to volatile commodity and service costs.  If gas prices rise substantially, is it smart to drill longer laterals, place more fracture stages, or both?  Shale resources are an increasingly important part of the oil and gas asset base, and it’s in everyone’s best interest that they are developed in an efficient manner.  

Kurt Wilson is a petroleum engineering student pursuing a master’s degree at Stanford University. He would like to acknowledge his advisor Dr. Louis Durlofsky and the support of the reservoir simulation affiliates. More information can be found at (http://pangea.stanford.edu/researchgroups/supri-b/).

Each business day RBN Energy posts a Blog or Markets entry covering some aspect of energy market behavior. Receive the morning RBN Energy email by simply providing your email address – click here.