Workflow development and sensitivity investigation of offset well-to-well interference through 3D fracture modeling and reservoir simulation in the Denver-Julesburg Basin

Abstract

The objective of this research was to construct a calibrated multi-well 3D fracture model to evaluate well-to-well interference, also known as well “bashing”, in the Niobrara and Codell formations of the Denver-Julesburg (DJ) Basin. Hydraulic fracturing is a vital component to produce in unconventional resource plays economically. In the early stages of development in all unconventional resource plays, including the DJ Basin, single well pads were common for either exploratory drilling or lease retention. Initially many shale plays across the world were not fully developed and in order to capture the resources originally left behind, infill drilling is taking place to access additional resources to create a greater effective stimulated rock volume (SRV) and achieve better recovery. Unfortunately, many previously drilled, producing “parent” wells and newly drilled “child” wells are actually yielding a reduction in recovery rates in short and long-term cases due to interference. Research has shown that one of the main contributors to the variability in production is the presence of pressure depletion from production around the parent wells and the impacts it has on subsequent child well completions. As a means of examining well-to-well interference, the focal point of this study was a nine well pad consisting of six wells in the Niobrara and three wells in the Codell. This pad showed evidence of interference measured using bottom hole gauges installed in the two “parent” wells, one in each formation, during the stimulation of the child wells. Oil and water flowback tracers utilized during the treatment of the child wells also supported the presence of communication. A multi-well, multi-stage (total of 500 stages) 3D hydraulic fracture model was developed using offset vertical well logs and geologic interpretation of mud logs to define the variation in rock properties that best represent the reservoir in which these wells were drilled. Calibration of the model consisted of using diagnostic measurements, diagnostic fracture injection tests (DFIT), step-down tests, and pressure matching measured to simulated treatment data. Using the calibrated model, the “parent” wells were simulated to generate individual stage fracture geometries. Results indicated uncontained hydraulic height growth throughout the Niobrara and from the Codell into the Niobrara. Using the results from the hydraulic fracture model, including conductivity and effective half-length, a pressure depletion profile due to production drawdown was created for the two parent wells. History matching the parent wells production (oil, gas, and water), using this numerical reservoir model, resulted in a subsequent pressure gradient radiating from the wellbore out into the reservoir. The history matched reservoir depletion model was then incorporated in the fracture modeling to simulate the treatments of “child” wells more accurately in a depleted reservoir. The parent wells’ hydraulic fracture and reservoir models culminated in the construction of a hydraulic fracture depletion model, which was used as the primary tool to examine well-to-well interference between “parent” wells and “child” wells for this setting. A successful history match of the “parent” well fracture treatments was not fully achieved. However, the reservoir model provides a reasonably accurate representation of how the pressure transient moves away from the wellbore. The fracture geometry generated in the calibrated model was directly utilized in generating the pressure depletion profile simulated in the reservoir model – the key component in understanding the effects of pressure sinks on production. The imported depletion model allows for the simulation of the child well treatments and the impact of pressure sinks on fracture generation and the interaction between “child” wells and “parent” wells. The model also serves as a tool to conduct sensitivities on potential methodologies to mitigate the effects of pressure depletion. A comprehensive workflow of well-to-well interference between “parent” and “child” wells, using hydraulic fracture and reservoir modeling, and subsequent analysis of methodologies to mitigate the effects of pressure depletion, was successfully developed during the course of this study. However, several communication and compatibility issues were identified interacting between the two software packages. The most important being the integration of fracture geometries hydraulic fracture model into the reservoir model, which is challenging given the number of nodes that are generated and correctly characterizing the associated inputs. It is important to take into account these considerations when attempting this workflow until these issues are addressed to improve accuracy. Protection frac simulations indicate that they could be either beneficial or detrimental depending on the amount of repressurization that is achieved and the distance that the pressure transient travels into the reservoir. In the given situation, repressurizing the reservoir surrounding the “parent” wells by 1,000 psi resulted in a reduction of well interference due to the increased stress state near the wellbore. However, repressurizing by only 500 psi resulted in increased well interference between the parent wells. Several wells communicated with both the parent wells as a result of the repressurization not being great enough to offset the depletion. Shutting-in the “parent” wells prior to the “child” well completions, as a method to mitigate well-to-well interference by allowing the reservoir to increase pressure naturally, resulted in an ineffective solution. The modeling showed that shutting-in the well for three months resulted in an increase of only ~250 psi, which is an insufficient pressure increase to offset the effects of depletion.