HORIZONTAL WELL WILL BE EMPLOYED IN HYDRAULIC FRACTURING RESEARCH

A new 10-well research site is planned to enable more controlled experiments for better definition of hydraulic fracturing.

One of the 10 wells will be a near-horizontal well that will monitor microseismic events along its length.

The Gas Research Institute (GRI) has begun evaluating a low-permeability, gas-bearing sandstone as the target stratum for experiments to be conducted at its hydraulic fracture test site (HFTS). During a 4-year period, GRI will use the HFTS as a field laboratory to conduct multidisciplinary research projects to assess the mechanics of hydraulic fracturing.

As a result of a screening process, the Davis sandstone in the Ft. Worth basin has emerged as the tight gas sand which best fits the selected criteria established by GRI and its contractors, GRI says. The Ft. Worth basin is located approximately 50 miles northwest of Ft. Worth.

GRI is planning a research well to fully characterize the Davis prior to making a final decision on the location of the HFTS. If data from the research well indicate the Davis sand does not adequately meet selection criteria, other candidates identified in the screening process will be investigated.

EARLIER PROJECTS

GRI's tight gas sands program has sponsored research projects on hydraulic fracture treating since 1983. The main objective of the research has been to learn how to predict the shape and extent of an hydraulic fracture in real time.

To do so, comprehensive field data have been collected before, during, and after fracture treatments on a number of wells. These data have been analyzed by a team of geologists, geophysicists, and engineers to understand how a fracture propagates and to estimate the three-dimensional properties of natural gas reservoirs and hydraulic fractures created within them.

GRI has conducted field experiments mainly in the Travis Peak and Cotton Valley sands in East Texas, the Canyon sands in West-central Texas, and the Frontier sands in Wyoming. These experiments have led to new theories and methodologies for evaluating the shape and extent of an hydraulic fracture.

GRI and others have concluded that hydraulic fractures grow taller, are wider, and have a shorter length. This conclusion is contrary to conventional model predictions.

Experiments at the HFTS will attempt to validate and refine fracture modeling theory.

Originally, when GRI's tight gas sands project was organized, researchers believed they could develop fracture diagnostic methods to remotely sense the direction and the dimensions of hydraulic fractures. Specifically, they believed that fracture height, azimuth, and length could be determined with reasonable accuracy using fracture diagnostic systems, such as geophones.

Although GRI has developed diagnostic systems to determine fracture azimuth and to estimate fracture height at the wellbore, no system has yet been developed to confidently determine fracture length.

Input from fracture diagnostics is not detailed enough to allow validation of fracture propagation models. GRI's HFTS will be used to measure in situ fracture dimensions and properties to validate a number of different fracturing diagnostic techniques.

The HFTS will provide a well-characterized environment, facilities, and material resources. The experiments will help validate methods, concepts, hypotheses, and models to improve applications of hydraulic fracturing and expedite technology transfer for more effective natural gas production.

OBJECTIVES

Research at the site will cover three main areas: fracture diagnostics, fracture modeling, and fracture fluid characterization. The focus of research on fracture diagnostics will be to develop a technique for measuring fracture length and to enhance fracture azimuth and height diagnostic techniques for more efficient use.

Fracture modeling research will be directed toward verifying fracture simulator theory on pressure distribution within a fracture, including the pressure at the fracture tip, or leading edge. Research will also be conducted to:

Verify fracture closure concepts and develop techniques to better estimate fracture closure from pressure-decline analyses
Develop techniques for measuring and predicting subsurface formation stress and quantify characteristics of effective stress barriers for containment of hydraulic fractures
Investigate proppant transport, fracture proppant conductivity, and the role fracture-fluid viscosity has on created fracture dimensions
Investigate and verify concepts for the growth of hydraulic fractures, including fracture wing asymmetry, multiple fracture patterns, near well bore tortuosity, and fracture closure.

RESEARCH TARGETS

As mentioned previously, fracture-diagnostic equipment will be used in experiments designed to measure the actual shape and pressure distribution of hydraulic fractures. Tiltmeters, inclinometers, and geophones will create an image of the fracture from remote sensing locations.

Instruments installed in intersecting wells will measure the pressure distribution in the fracture as well as the height, width, and length. Data collected will provide insight into unresolved issues regarding hydraulic fracture propagation.

LEADING-EDGE MODELING

Pressure distribution at the fracture tip has a significant impact on fracture geometry, yet remains a major unresolved issue in fracturing mechanics.

Fracture propagation models differ greatly in their assumptions about modeling this leading edge.

Some researchers suggest that pressure increases rapidly within the fracture near the fracture tip. However, other researchers have reported that pressure required to propagate a crack ahead of the fracture fluids is normally negligible.

HFTS experiments will seek to resolve this controversy.

PRESSURE-DROP CALCULATION

Another important unknown parameter is the pressure drop along the length of the fracture. Most models currently use equations for fluid flow, under laminar conditions, in a smooth wall elliptical tube (or parallel plates). However, GRI research and other published literature indicate that a more rigorous equation should be used which deals with wall roughness and pressure gradients associated with fractures that do not have smooth parallel walls.

Experiments at the HFTS will be conducted to measure the pressure gradient down an actual hydraulic fracture for various conditions of viscosity and shear rate to improve the understanding of fluid flow in fractures.

FRACTURE FLUID VISCOSITY

Fracture models vary significantly in their treatment of fracture fluid viscosity. Results of experiments at the HFTS will be used to identify how viscosity effects net pressure and created fracture dimensions.

FRACTURE PROPERTIES

The natural gas industry frequently uses radioactive and temperature-logging methods to estimate fracture height. However, these techniques only attempt to measure fracture height very near the well bore while true effective fracture height may be significantly different away from the well bore.

GRI has funded research to estimate fracture height away from the well bore and azimuth using microseismic information recorded in the treatment well.

HFTS fracturing experiments will provide data to verify and enhance fracture height and azimuth measurements using this technique. Tiltmeters will be deployed to collect independent azimuth data.

The unique microseismic data to be collected at the HFTS will also be very important for the development of a fracture-length diagnostic technique.

IN SITU STRESS

In situ stress is an important parameter in all fracture modeling. Although GRI has made significant progress in stress profiling using cores, logs, and in situ stress tests, problems remain in analyzing stress data and developing a more generalized correlation between logs and measured values.

Thorough characterization of stresses in all rock layers is essential to interpreting fracture diagnostic measurements and computing fracture dimensions.

Extensive core, geophysical logging, and in situ stress test data will be gathered at the HFTS in an attempt to fully characterize each layer of rock in detail.

Research at the HFTS will attempt to resolve problems with current stress determination techniques and to develop new techniques. For example, inclinometer data will be correlated with fracture closure pressure to enhance in situ stress test data interpretation.

PRESSURE DROP

Near-well bore phenomena often cause excessive pressure drops during a fracture treatment. Sometimes perforations are restricted, and other times multiple fractures at the well bore create tortuous paths from the well bore to the main fracture.

Severe well bore restrictions have been responsible, at least in part, for treatment screenouts at some GRI experimental wells. These kinds of potential problems must be considered when designing a fracture treatment and especially when analyzing fracture pressures.

Experiments will be conducted at the HFTS to evaluate near well bore pressure drop and its effect on fracture analysis and treatment.

FRACTURE WIDTH

Besides fracture height and length, engineers need to know the width of an hydraulic fracture. Although various equations are used to calculate fracture width among different models, no single equation stands out as the best.

When the formation has multiple layers with different stresses and mechanical properties, width calculation becomes more complex. Experiments to attempt to determine width will be performed. These properties will be fully characterized in the rock layers at the HFTS and research results will provide information to improve width calculations in models.

GEOLOGIC CONSIDERATIONS

Geologic characteristics of rocks that most affect strength, the effect of natural fractures on hydraulic fracture propagation, and the role of tectonic and burial history on current stress distribution will be investigated at the HFTS.

PROPPANT AND CONDUCTIVITY

Some research has been conducted on proppant transport, but only on small-scale laboratory models. Most theories assume piston-like displacement of the proppant-laden fluid in the fracture, not considering such factors as gravity segregation, viscous fingering, or convection.

Also, published values of fracture conductivity from laboratory tests may not adequately predict the values of fracture conductivity at in situ conditions. Later experiments at the HFTS will address these issues.

SITE SPECIFICATIONS

The site ultimately selected for GRI's HFTS will be fully characterized to better ensure a controlled experimental environment. The formation selected for the site will have adequate fracture barriers (both above and below the target interval), be relatively shallow (between 2,000 and 5,000 ft) to contain R&D costs, and be relatively thick (75-100 ft) and homogeneous.

Facilities will be constructed on a 300-acre site in a region with a moderate climate near a service company infrastructure. The facilities will provide a central workplace for researchers and a repository for hardware and instruments needed to measure, record, and analyze data.

Complete computing capability will be available for quality control and on-site processing of experimental data.

Fig. 1 shows the current proposed data acquisition and processing system to be used at the HFTS.

HFTS PLAN

Fig. 2 presents a conceptual layout of the HFTS. Each well has been strategically placed to accomplish specific functions such as data monitoring and fracture description. Each well will be drilled according to a phased schedule intended to reduce risk while allowing research to progress.

Data Well No. 1 (D1) will be drilled and cased to collect data on the formation properties of the site. Well D, will be designed to completely describe the formations in the vertical sequence.
Other wells will help define the formations in the lateral direction so that a three-dimensional description of the test site can be developed over time.
Stresses, fracture azimuth, and net fracture pressures will be determined from open hole and cased hole stress testing, over coring, and minifracture tests. Eventually, D, may be used as a source/receiver well bore for diagnostic experiments.
Monitor Well No. 1 (M1) will be drilled approximately 300-400 ft from D1. A similar open hole data-acquisition program and detailed formation description will be performed on this well to evaluate the changes in the vertical sequence, as well as the horizontal direction (from D1).
Well M1 will have a series of geophones and inclinometers attached to the outside of the casing and cemented in place. This instrument string will be about 1,000 ft long, extending several hundred feet above and below any fractures to be created in the treatment well.
Treatment Well No. 1 (T1) will be the main hydraulic fracturing experiment well and will be drilled between D1 and M1. The well will be located so that the created hydraulic fracture will be perpendicular to subsequent intersecting wells.
A series of geophones will be attached to the outside of the casing and placed above the producing interval. Tubing run in T, will allow treatments to be pumped down the casing/tubing annulus and direct measurement of bottom hole pressures.
In Monitor Well No. 2 (M2), the fracture azimuth will be determined by various methods on three separate wells. Once this is accomplished, M2 will be drilled as a near-horizontal well running parallel to the expected hydraulic fracture for a distance of about 1,200 ft.
The horizontal section of M2 will be about 100 ft from the fracture created in T1. M2 will have a series of geophones running the length of the horizontal section, cemented on the outside of the casing.
The primary purpose of this well is to provide additional data necessary to monitor and determine created fracture length and height with microseismic diagnostic techniques.
Intersecting Well No. 1 (I1) will be a deviated well intersecting an initial fracture created in T1. The intersection point will be about 100200 ft from T, and will deviate at approximately 60. Adequate core will be cut to recover this initial fracture.
Well I1 will then be completed open hole through this fractured interval so that fracture pressure measurements and fracture width measurements can be made during future fracture experiments. Well I1 will verify fracture azimuth as estimated by seismic and open hole techniques.
Intersecting Wells No. 25 (I2-I5) will be additional intersecting wells that will then be drilled in the direction of fracture propagation. Three of these wells (I2-I4) will be drilled on the same side of T1 as the first intersecting well.
The fifth intersecting well will be drilled on the opposite side of T1 to evaluate fracture wing symmetry.

Experiments will then be conducted to propagate a fracture through the intersecting wells to determine fracture propagation rate and fracture pressure.

ANTICIPATED BENEFITS

One of the main problems in building and proving the accuracy of fracture propagation models is that there has yet to be a definitive view of an actual fracture in an actual formation. Laboratory tests have provided some data, and field experiments have quantified portions of a created fracture, but no single experimental design has included the kind of extensive testing being planned for the HFTS.

Experimental results generated at the HFTS will be used to attempt to resolve a number of issues pertaining to fracture propagation modeling and fracture diagnostic techniques. Actual measurements of pressure distribution at the fracture tip, as well as determination of fracture length, height, and width, will help enhance hydraulic-fracture propagation theory.

Other experimental results will provide information about the effects of viscosity and viscosity distribution on created fracture dimensions. With a seismic source resolution of 5 ft, fracture geometry will be highly detailed.

Tiltmeter data will also be correlated with microseismic data to improve the accuracy of both diagnostic methods. New diagnostic methods could be developed based on results of hydraulic fracturing experiments at the HFTS.

Cecil Parker, senior staff engineer for Conoco PER, has been one of several experts who worked closely with GRI to advise it on specific steps in the development plan for the site.

According to Parker, "Data generated from experiments at the hydraulic fracture test site can help the gas industry standardize and calibrate fracture models. I don't expect that the industry will come away from this project with a complete understanding of the hydraulic fracturing process.

"There are just too many variables in any given fracture treatment," says Parker, "to guarantee that level of success. However, the project should help us better define the important parameters of fracture treatments, which in turn can help us better characterize the sort of fracture we've made."

GRI believes that results of experiments conducted at the HFTS will be very useful to the natural gas industry for planning the development of low-permeability gas reservoirs. Geologists, geophysicists, and engineers involved with exploration and production will be afforded specific insights into optimum stimulation and development techniques for these reservoirs.

The industry will then be better able to provide a more stable and secure supply of natural gas for the consumer at competitive costs.

SITE MANAGEMENT

Proposed research at GRI's hydraulic fracture test site will, admittedly, be costly, complex, and technically challenging. However, it has the potential to yield significant benefits to the natural gas industry.

To ensure that all relevant research issues are identified and adequately addressed, GRI has set up a comprehensive management and organization plan.

The organization plan calls for GRI to provide overall project management and guidance for the site. Providing input to GRI will be a technical review committee comprised of representatives from gas producer and service companies who will review development plans, specific experiments, and results from experiments.

Current committee membership includes Ralph Veatch (Amoco), Malcolm Strubhar (Mobil), Carl Montgomery (ARCO), Jintendra Avasthi (Chevron), Ken Nolte (Dowell Schlumberger), Bob Hannah (BP Exploration), Jacob Schlyapoberski (Shell), Lew Lacy (Exxon), Mohamed Soliman (Halliburton), Cecil Parker (Conoco), and Wayne Pittman (Texaco).

In addition, a representative from an independent gas production company in the area will be solicited once the site has been selected.

Leading oil and gas industry research and consulting organizations will perform the technical tasks necessary for developing and implementing the HFTS program. S.A. Holditch & Associates Inc., CER Corp., Teledyne Geotech, Resources Engineering Systems, the University of Texas Bureau of Economic Geology, NSI Technologies, and Sandia National Laboratory will be primary contributors.

"We designed the management plan to ensure that the oil and gas industry is represented and consulted on all proposed research at the site," explains Allen Shook, GRI's project manager for the HFTS.

Proposed research has and will continue to undergo technical review.

All instrumentation, experiment designs, and experiment executions are subject to quality control review as well as scrutiny from an independent risk analysis consulting firm.

A phased schedule identifies specific milestones for GRI to make explicit go/no go decisions.

According to Shook, "if research at the hydraulic fracture test site is not successful, we want to be sure that it's because the research can't be done; not because the project lacked superior planning and management."

Cecil Parker of Conoco comments that "A research project on hydraulic fracturing this comprehensive probably wouldn't be done outside GRI sponsorship. As members of the technical review committee, we have the opportunity to act as stewards of industry research funding.

"Through direct participation," says Parker, "we've got the ability to change course direction or modify specific experiments, and we can learn as we go along."