Case History of Drainage Mapping and Effective Fracture Length in the Bakken

Abstract

Abstract Understanding effective fracture length and characterizing drainage patterns is critical for optimal development of unconventional resources. This paper documents a comprehensive field experiment in the Bakken formation, where several fracture diagnostic technologies and drainage mapping methods were used in a unique project setup to measure effective fracture length and map drainage. Two vertical wells (V1 and V2) were drilled 1,000 ft away from a Bakken lateral (H1) with 10 years of production (Fig. 1). The two vertical wells were 200 ft apart. V2 was used for microseismic and deformation (downhole tilt) measurements, while V1 was used for pressure measurements and hydraulic fracture characterization. The project consisted of re-pressurizing the existing lateral (parent well), using microseismic monitoring to map drainage (designated MDD, Dohmen et al. 2013, 2014, 2017). A DFIT was performed in the V1 well before the MDD to measure local stress and pore pressure. Following the MDD, a small propped fracture treatment was pumped in the V1. The H1 well was then produced for 4 months and DFITs pumped in the V1 and V2 wells. This comprehensive fracture diagnostic dataset was integrated with detailed core and log measurements, hydraulic fracture modeling, and advanced reservoir simulation to characterize the hydraulic fracture performance. The H1 MDD indicated that the major pressure depletion (drainage) was approximately 500 ft on either side of the lateral. H1 BHP showed local reservoir pressure was 1,000 psi. The initial V1 DFIT showed virgin reservoir pressure, but surprisingly, the 20 bbl DFIT injection was detected on the H1 BHP gauge. Microseismic mapping of the V1 fracture treatment (15 klbs, 600 bbl) showed a planar fracture with a half-length of 1,000 ft. The V1 fracture "hit" the H1, measured by microseismic and confirmed by a 1,650 psi increase in H1 BHP. The microseismic showed a symmetrical fracture, suggesting that the H1 re-pressurization mitigated the detrimental effects of parent well depletion that can cause severe asymmetry. Downhole tilt and microseismic showed fracture height quickly extended downward through the lower Bakken shale into the Three Forks. The V1 was not produced. However, the H1 oil rate doubled after the V1 frac hit, indicating significant stimulation. V1 DFIT #2 showed 1,800 psi depletion, while the V2 DFIT showed approximately 1,000 psi depletion, confirming that the V1 fracture is "flowing" into the H1 lateral 1,000 ft away. The reservoir simulation history match indicated low fracture conductivity, but enough to improve well productivity and drain oil over 1,000-ft from the H1 lateral. This paper details a comprehensive fracture diagnostic dataset gathered in a unique field laboratory where a single hydraulic fracture from a vertical well is used to characterize fracture conductivity and flowing length.