Spatial and Temporal Effects on Low Frequency DAS and Microseismic Implications on Hydraulic Fracture Geometry and Well Interactions

Abstract

Monitoring the geometry of hydraulically induced fractures using a variety of frac diagnostic tools is a relatively expensive but necessary step to achieve successful development of unconventional reservoirs. Valuable insights about the hydraulic fracturing process can be determined and cross-validated using an assortment of diagnostic tools. In recent years, the monitoring of pressure, temperature, and strain interactions between wells during stimulation has been providing new and detailed information about the geometry of hydraulic fractures not available before. Low Frequency Distributed Acoustic Sensing (LF-DAS) and Distributed Strain Sensing (DSS) can be used to determine the timing and location of approaching, intercepting and re-opening hydraulically induced fractures in nearby wells instrumented with Fiber Optics (FO). In the Hydraulic Fracture Test Site 2 (HFTS2) the integration of these technologies with traditional fracture geometry diagnostic tools such as Microseismic (MS) are revealing unique information about fracture geometry and the propagation tendencies of fractures generated while stimulating the Wolfcamp in the Permian Delaware Basin. In HFTS2, hydraulic fractures are sub-vertical, occurring in swarms and forming well-defined and separate Frac-Zone-Domains (FZDs). In HFTS2, both the number of microseismic events and strain interactions increase as stimulation progresses within the pad. The observed spatial and temporal dependencies in the MS and LF-DAS datasets strongly suggest there is a lot of re-opening and re-fracturing of pre-existing fractures, not only of the FZDs from the "parent/primary" wells but also from pre-existing fractures of the FZDs from offset "child/infill" wells that are part of the same stimulation vintage. These newly observed spatial and temporal effects must be considered when using MS, LF-DAS, and other fracture driven interaction (FDI) techniques to determine fracture growth velocities and when interpreting frac diagnostics results to calibrate fracture models. The data suggest that fracture growth is impacted by the location of pre-existing hydraulic fractures and their stress state. The findings in HFTS2 highlight the risk not only of misinterpreting microseismic derived Stimulated Rock Volume (SRV) but also, and more importantly the development of possible new completion practices and wells stimulation sequencing. The information from HFTS2 has helped to understand not only the fracture geometry on this Permian Wolfcamp site but also the uses and limitations of different fracture diagnostic tools.