Abstract
Unconventional resource plays, herein referred to as source rock plays, have been able to significantly increase the supply of hydrocarbons to the world. However, majority of the companies developing these resource plays have struggled to generate consistent positive cash flows, even during periods of stable commodity prices and after successfully reducing the development costs. The fundamental reasons for poor financial performance can be attributed to various reasons, such as; rush to lease acreage and drill wells to hold acreage, delayed mapping of sweet spots, slow acknowledgement of high geological variability, spending significant capital in trial and errors to narrow down optimal combinations of well spacing and stimulation designs. The objective of this paper is to present a systematic integrated multidisciplinary analysis of several unconventional plays worldwide which, if used consistently, can lead to significantly improved economics.
We present an analysis of several unconventional plays in the US and Argentina with fluid systems ranging from dry gas to black oil. We utilize the publicly available datasets of well stimulation and production data along with laboratory measured core data to evaluate the sweet spots, the measure of well productivity, and the variability in well productivity. We investigate the design parameters which show the strongest correlation to well productivity. This step allows us to normalize the well productivity in such a way that the underlying well productivity variability due to geology is extracted. We can thus identify the number of wells which should be drilled to establish geology driven productivity variability. Finally, we investigate the impact of well spacing on well productivity.
The data indicates that, for any well, first year cumulative production is a robust measure of ultimate well productivity. The injected slurry volume shows the best correlation to the well productivity and "completion normalized" well productivity can be defined as first year cumulative production per barrel of injected slurry volume. However, if well spacing is smaller than the created hydraulic fracture network, the potential gain of well productivity is negated leading to poor economics. Normalized well productivity is log-normally distributed in any play due to log-normal distribution of permeability and the sweet spots will generally be defined by most permeable portions of the play. Normalized well productivity is shown to be independent of areal scale of any play.
We show that in every play analyzed, typically 20-50 wells (with successful stimulation and production) are sufficient to extract the log-normal productivity distribution depending on play size and target intervals. We demonstrate that once the log-normal behavior is anticipated, creation of production profiles with p10-p50-p90 values is quite straightforward. The way the data analysis is presented can be easily replicated and utilized by any operator worldwide which can be useful in evaluation of unconventional resource play opportunities.
A Reappraisal of the Prebisch-Singer Hypothesis Using Wavelets Analysis
