Fracturing Creates Hundreds of Fractures - Only a Few Matter

Abstract

It would be exceedingly hard to find anyone with more experience extracting data from a fractured reservoir than Kevin Raterman. After nearly a decade as co-leader of ConocoPhillip’s one-a-kind test site in the Eagle Ford, Raterman, a reservoir engineering advisor at the company, has had the opportunity to use just about every sort of diagnostic test. He was the lead writer on the latest technical paper (URTeC 263) about the test site that studied “core, image logs, proppant tracer, distributed temperature sensing, distributed acoustic sensing, and pressure, which shows that not all hydraulic fractures are created equal.” Among those, the measurements from an array of downhole pressure gauges loomed largest. Raterman’s advice to those analyzing fractured reservoirs is to install more downhole pressure gauges to identify where fractures are draining the rock and where they are not. Those carefully placed sensors observed large pressure differences at locations as close as 45 ft apart. “The more spatial pressure data you have, the better off you are,” he said. Pressure gauges are one of the oldest tools for reservoir analysis, and one of the more affordable. The test site plan to gather spatial data, though, significantly upped the ante. It required installing gauges in the reservoir at carefully chosen locations in what would normally be inaccessible rock. ConocoPhillips installed 14 pressure gauges in monitoring wells around the fractured producing well, where the 15th gauge was located. Based on other diagnostic tests, the locations showed how a small number of large fractures can play a dominant role in draining the reservoir. It was part of ConocoPhillips’ exhaustive reservoir test in the Eagle Ford that was launched in the heady days when $100/bbl oil seemed like a given. The new technical paper examined 12 pressure gauges installed in a monitoring well originally drilled for testing, which had focused on imaging thousands of feet of rock and collecting one of the largest samples of fractured core•ever (Fig. 1). The core samples were the stars of the first test site paper in 2017 (URTeC• 2670034). That attracted a lot of attention because the samples contained a complex mix of large and small fractures that contradicted widespread assumptions about fracturing. It was widely assumed that the millions of pounds of sand pumped to prop open fractures extended to a large area around the well. But at those far field locations, from 60–400 ft from the well, “evidence for well propped fractures is sparse.” Those samples were also interesting because hydraulically fractured core is rarely collected. Coring fractured rock is costly and difficult. The ConocoPhillips team had to develop a method to collect the fragile, cracked samples. There are limits to what can be learned from the cracks in a cylinder of rock a few inches wide. Those fracture segments are a static piece of a dynamic system.