Unconventional Type Curves: Useful, or Sirens of Destruction?

Abstract

Abstract Considerable faith is placed in Type Curves; Type Curves that are ultimately responsible for resource estimation, revenue generation, and value determination. We assume that they are valid representations of what will happen, but is this faith warranted? In and of itself, a type curve is a purely deterministic forecast of a multi-dimensional uncertainty. To their credit, explorationists have recognized range-based volumetric assessments as providing significant advantage in both resource determination and downside risk mitigation for nearly two decades. Given that nearly two-thirds of the uncertainty in NPV for an unconventional opportunity is production profile based, it is vital to understand the true range in the production profile. Unfortunately, given the two dimensional aspect (both rate and time are uncertainties) of time series uncertainty and the scale up issue from individual well ranges, this is not as straightforward as it may seem. Attempts at ranged type curves tend to be biased by center anchoring and unreliable P10-P90 estimates. Ranged type curves are often the "expected" curve factored up and down symmetrically (!). This typically does not reflect valid production behavior. Well life is often inappropriately anchored at 40 years, or even 60 years and the curve is back-calculated to one of the most unreliable elements... initial production. The free variable in the calculation becomes decline, yet we receive decline information early in the life of the well. We have no clear reason to anchor production curves to a set time. As such, time should be the calculated variable. Additionally, the expected type curve when replicated across an area of potential will not be a representative case that delivers the expected value of an opportunity. Type curves can be useful, ranged type curves even more so. Stochastic assessment of production potential allows companies to validly assess the true potential of a project area. It enables prioritized learning plans and downside recognition. Individual profile iterations can be easily ranked using discounted present production (DPP), but a value context is greatly preferred. Valid curves enable a company to recognize if a project will be production- or cost-sensitive and efficiently allocate time, people, and capital, thus optimizing the factory phase of the project. Assessments are made to support decisions. Assessments without a decision context are a waste of time.