The Montney Formation stretches from southwestern Alberta to northeast British Columbia in Canada, and is one of the largest and most prolific shale plays in North America. The Montney Formation is also unique because it has conventional, over-pressured gas and an over-pressured liquids rich fairway. However, since the first multiple fractured horizontal well was drilled in 2005, there has been proposals for optimizing completions using different fracturing fluid systems and completion techniques. The low oil and gas price environment and the ensuing cost control mechanisms coupled with better understanding of what works in the Montney Formation, made the utility of some of the previously proposed optimization designs like "fracture effectiveness" which used energized fracturing fluids less desirable completions method. However, completion optimization methods like "operational effectiveness" which used high-rate slick water with increasing proppant mass per stage become the dominant stimulation method in the Montney Formation. But what has been missing was how to integrate fracture design and optimizations using all available information such as step-rate test, mini-frac, DFIT (diagnostic fracture injection test) analysis, well logs, geo-mechanical data, fracability index, core data, micro-seismic mapping data, and post-fracture analysis to improve fracture design and optimize the well completions.
The objective of this paper is to present a new methodology for building calibrated fracture models from low quality micro-seismic data that has either location uncertainty or signal-to-noise ratio issues, and use it to optimize well completions. The process involves two-steps; first, the hydraulic fracture design was modeled and then calibrated using only micro-seismic mapping data from fracture stages that were closest to the micro-seismic geophones (avoiding location uncertainty or signal-to-noise ratio issues). This allowed us to construct a robust and reliable fracture geometry model. For each of the wells in the study, all fracture stages were then history matched and remodeled using the calibrated fracture model. Secondly, each well was optimized by incorporating fracture cluster sensitivity (2, 3, 4, and 5 clusters per stage), proppant mass sensitivity (50 kg, 75 kg, 100 kg, and 150 kg per stage) and fracture spacing sensitivity (20 m, 25 m, 33 m, 49 m and 98 m per stage).
The result from this study shows that a highly optimized fracture model can be constructed from low quality micro-seismic mapping data that had location uncertainty due to the use of one monitoring well or signal-to-noise ratio issues. Secondly, the result also shows that increasing the number of clusters per stage and proppant mass per stage improves well production and recovery. However, the question is are these improvements short time gains, and what is the balance between well productivity and economics? Thirdly, in this study, we propose using measureable and known metrics to optimize wells such as average "hydraulic" fracture half-length, propped fracture half-length and conductivity for multi-clustered fracture stages. Ideally, well performance should be obtained from lookbacks instead of pounds per lateral length of the horizontal well (i.e. 2,400 lb. /ft.) or fixed volume/proppant for each stage or fixed clusters per stage without any empirical data to support it.
While there are no two shale formations that are alike, most of the findings from this study are transferable and applicable to other unconventional resources. For instance, the paper presents;
A new method for building calibrated fracture models from low quality micro-seismic mapping data that has location uncertainty or signal-to- noise ratio issues.
A new method for optimizing fracture designs using cluster sensitivity analysis with varying proppant mass per fracture stage that can be used for scenarios analysis.
A methodology for optimizing fracture design models by adjusting fracture treatment volumes and proppant mass per stage based on well stage location and available net treatment pressure.