Two is Better Than One: How Measuring Twice Substantiates Surprising Results. Three Cases Studies from North American Unconventionals

Garrett Fowler; Karen Olson; Maggie Albrecht; Jackson Heffener; Jake Li; Rohan Irvin; Mark McClure
Paper presented at the SPE International Hydraulic Fracturing Technology Conference and Exhibition, Muscat, Sultanate of Oman, September 2023.
Subsurface realizations are uncertain. Uncertain realizations are refined with additional diagnostics and physics-based modeling establishes physical-plausibility of a realization. We examine three case studies that demonstrate the utility of replicating multiple diagnostics in a physics-based model to confirm surprising subsurface realizations.Three case studies are examined, where diagnostic observations are replicated in simulation models, and confirmed with secondary diagnostics.

  1. HFTS 1 Phase 3 (Eagle Ford) – Three different stage designs with varying stage length were tested. Volume to first response (VFR) was measured between the treating well and observation well 691 feet away. Replicating the VFR responses in a model suggested increasing heel bias and decreased cluster efficiency as stage length increased. When compared with fiber optic strain (initially not available), the strain response confirmed the heel bias and low cluster efficiency.
  2. HFTS 2 (Delaware Basin) – A vertical fiber optic strain gauge installed in the middle of a well pad exhibited strain more than 1800 feet above the fractured wells, suggesting fracture height growth in excess of 1800 feet. Simulation models replicated the height growth when stress and pore pressure profiles were input into the model. During production, downhole pressure gauges in the vertical well confirmed depletion from reservoir was consistent with extremely tall fractures.
  3. Stress and toughness calibration (Midland Basin) – Water soluble tracers and production geochemistry provide measured connectivity and zonal allocation of production fluids, respectively, within a well pad. Matching the interpreted total hydraulic and productive geometries required minor adjustments to geomechanical inputs in the model, underscoring the importance of multiple diagnostics.

 

The multiple diagnostics acquired in each of the three projects helped to confirm initially surprising results, and modeling confirmed the physical validity of each and possible mechanisms. By understanding the underlying mechanisms, methods to exploit or mitigate the causational mechanism may be engineered. In each of the case studies presented the substantiated observations and proposed mechanisms yielded design changes to improve returns on future wells:

 

  1. Eagle Ford study – established economic justification for shorter stages with higher designed perforation friction for up to 42% higher productivity per foot.
  2. Delaware study – led to landing zone modifications yielding up to 60% improvement in NPV/section.
  3. Midland study – Calibration to multiple diagnostics reduced zonal allocation error to less than 10% while honoring pad production, providing forecasting confidence.

 

The pace of drilling and experimentation in shale reservoirs allows for rapid design iteration. Selection of diagnostics is critical to understanding what design changes are impacting the reservoir and physics-based modeling establishes plausible mechanisms to explain how and why those changes occur. Validation of mechanisms allows for predictive models to be employed in the optimization of future development.

Search