Enabling Improved Reservoir Simulation and Production Forecasting

By Kash Kashikar
Vice President, Completions Evaluation, MicroSeismic, Inc.

Last month I attended the SPE Gulf Coast Section Reservoir Engineering forum. The meeting was very well attended and boasted the largest attendance in the history of the Reservoir Engineering Forum. This highlighted the interest and need for improving our ability to accurately forecast production from individual wells and fields in unconventional reservoirs. The morning session focused on 3 presentations related to shale oil and gas. The presenters covered topics ranging from workflows to utilizing stochastic history matching methods, to the use of complex full physics simulators that have the ability to model geological, fluid flow and fracture network complexity.

While significant progress has been made in our ability to model fluid flow in very complex fracture systems that result from hydraulics fracturing, it was clear that there are still many unanswered questions. Some key questions included the following: What is the nature and complexity of the fracture network? What are the dimensions of the fracture network in play over the life of the well? What is the areal and volumetric extent of the stimulated zone and thus the drainage volume?

Microseismic is the singular measurement that extends beyond the wellbore into the reservoir and the intra-well space. However, today there is very limited microseismic data available to guide any large scale use for proper and rigorous reservoir and fracture network description. To alleviate this lack of measurement beyond the wellbore, stochastic modeling of fracture networks is often used for simulating long term production. The problem is compounded by the fact that production data is only available for a short time period of time – a few months to a few years, while the forecast for ultimate recovery is performed over a time period that is several times larger (10 to 30 years). This leaves us with a very large set of possible models that provide a good history match to the short production history, but significant variation in the potential ultimate recovery.

At Microseismic Inc. we have developed a unique methodology to use the measured microseismic activity to deterministically define a discrete fracture network.  The process relies on basic physics and uses the energy released during fracturing to quantify the size, and orientation of individual fractures.  This description of the Discrete Fracture Network captures the inherent complexity of the fractures and accurately defines the areal and volumetric extent of the fracture network.  The resulting Stimulated Rock Volume thus provides an accurate and rigorous description of the reservoir volume required for the simulation.  Analyzing the size, orientation, and density of the discrete fracture network within the stimulated rock volume provides a mechanism to further constrain critical reservoir properties such as permeability, porosity and stress directions required for long term production forecasts.  As the body of our work at MicroSeismic increases over time, we are able to pull out correlations between microseismic results and production, allowing us to begin to rely more on the deterministic results and less on fuzzy models.

At Microseismic Inc., we are focused on helping customers design, monitor and optimize stimulation treatment and better understand the interaction between the reservoir, the operation, and its impact on field and reservoir economics. We are working closely with our customers across North America and abroad to deliver a better frac and improved production forecasting, each and every day.