Flow Interference Between Frac Clusters (Part 2): Field Example From the Midland Basin (Wolfcamp Formation, Spraberry Trend Field) With Implications for Hydraulic Fracture Design

Weijermars, Ruud (Texas A&M University) | van Harmelen, Arnaud (Texas A&M University) | Zuo, Lihua (Texas A&M University)



This study expands the high-resolution visualization of the drainage around horizontal wells using a newly developed analytical streamline simulator. A first case of single synthetic fractured well highlights the stagnation points occurring between the frac clusters, where drainage is ineffectively slow. A second synthetic case of two parallel wellbores with multiple fracture hits may enlarge the stagnant zones, with negligible drainage, occurring between the frac clusters. Finally, our flow tracking method is used to history match the total production with fracture allocated production for a multi-stage fracked well in the Wolfcamp formation, Midland basin. Progressive drainage of the SRV is visualized using the actual production data. The recovery factor after 5 years is 4% but rapidly slows and reaches only 6% after 40 years. When natural and/or hydraulically induced micro-fractures are assumed absent, only the near-frac regions in the SRV are drained even for tight frac spacing. Pressure near those fracs is rapidly lost and the productivity index of unconventional wells remains very low. Unrecovered oil remains trapped in so-called dead zones occurring in the matrix domains of extremely slow flow between the principal frac zones. Our simulation suggests that refracs between the original frac clusters can tap the oil in the dead zones after the first few years of production and may increase recovery factors and thus improve well economics. This conclusion is contingent on the assumption that the matrix domain between the fracs remains unfractured, When the matrix is micro-fractured, deeper domains would be drained (future work), leaving only residual, un-recovered hydrocarbons entrapped in the matrix blocks engulfed by the microfractures.


Contrary to the main trend of solving reservoir flows using numerical solvers for discrete elements, our reservoir modeling method is based on linear differential equations and closed-form solutions (Weijermars et al. 2017, Part 1, URTeC 2670073A). A major advantage of solutions based on the analytical description is infinite resolution due to the closed-form algorithms used. Reservoir drainage area, drained volume (using time-of-flight contours for successive fluid batches to reach the frac zones) and pressure fields can all be accurately visualized with our new method. The analytical results perfectly match those of the commercial, numerical simulator augmented with a streamline tracing algorithm (Part 1, URTeC 2670073A). The continuous solution allows us to visualize the flow around individual fractures in greater detail and with accuracy limited only by the uncertainty in input parameters and not by any grid sizing.