Economics of Barnett Shale

Organized Concurrent Session at the 31st IAEE/USAEE North American Conference, Austin, Texas

Wednesday, 7 November 2012 - 8:30 AM










Gürcan Gülen

Bureau of Economic Geology, The University of Texas at Austin

Svetlana Ikonnikova

Bureau of Economic Geology, The University of Texas at Austin

Kenneth Medlock

Baker Institute, Rice University (Houston, Texas)

 

In this session, we will discuss the economics of the Barnett shale play based on the results of a major interdisciplinary project at the Bureau of Economic Geology funded by the Sloan Foundation.

This project represents the most detailed look to date at the geology, production and economics of shale gas production in the Barnett formation.  Numerous studies were undertaken in order to understand the nature of Barnett production: the decline analysis via a new approach, Rock Exposure Index, or REI (an index that reflects decline profiles for Barnett wells better than Arps equations), spacing analysis, inferred-refracturing analysis, interference analysis, well economics, panel analysis of production behavior and geologic mapping of reservoir properties. Cumulative production from Barnett is more than 10 tcf as of mid-2012 but our analysis shows that 3-6 times more remains to be produced depending primarily on the price of natural gas, but also a few other factors.

Ken Medlock will present a paper that employs panel data techniques to empirically test the hypothesis that production from Barnett shale gas wells is adequately described by the REI equation.  Specifically, we analyze monthly production data for over 16,000 wells drilled in the Barnett shale during the period 1990-2011. Variables included in the analysis are the well’s production history, the year of first production, operator size, whether or not the well has been refractured, and a set of variables to capture geological characteristics of the shale at the well’s location.  In addition, to reflect market conditions, the average of the 12 month strip of natural gas futures, is also included.  Finally, given the difference in production rates between vertical and horizontal wells, data for the two groups is analyzed separately.

Svetlana Ikonnikova will demonstrate the process of identifying wells that were refractured.  We build a distribution of refractures’ success in terms of incremental production and explore how the distribution changes with time and depending on various parameters, including a well’s original performance, geological characteristics, and operator experience among others.  In parallel, we compare the incremental production from refracturing and new well drilling depending on rock quality.  Next, we look at the economics of refracturing versus new well drilling in an attempt to understand both the pace of activity and how the future may unfold given various uncertainties (e.g., the extent to which refracturing can extend the productive life of wells and thus increase the producible resource of natural gas). The developed well economics framework enables us to take a close look at the trade-off between the refracturing and new well drilling: drilling of a new well is considerably more expensive than refracturing, but the increase in production is much lower in the case of a refracture.  We identify the key parameters for the operator decision making, like costs, technological uncertainty, and the quality of rock available for drilling.

Gürcan Gülen will present the Barnett production outlook model, which utilizes the data from all other studies to group more than 4,000 square-mile blocks into 10 tiers, from best (Tier 1) to worst (Tier 10) performers.  We use historical average horizontal well production profile of each tier as the baseline for our forecasts by tier because horizontal wells replaced vertical and directional wells to a great extent.  We normalize the horizontal wells to 3,000-feet as this length seems to have become the norm.  The model is driven by the pace of drilling activity in our forecast, which is determined based on the remaining areas suitable for drilling and the breakeven economics relative to natural gas price.  The model tracks the remaining drillwell prospect inventory in each tier on untested acreage and throttles drilling pace as each tier becomes fully developed.  Our model uses these historical drilling trends in each tier (bounded by a minimum and a maximum). The model also accounts for well-attrition rates, which have worsened since the introduction of horizontal wells and are worse for poorer tiers.  The model has the flexibility to incorporate the impact of various factors such as technological improvements, reserve depletion, severance tax changes, and new regulations.  Various scenarios on natural gas prices and other key variables are evaluated.

Click to view a printable version of this article.