Optimization of Infill Drilling in Naturally Fractured Tight Gas Sandstone Reservoirs in the San Juan Basin

Based on a workshop co-sponsored by PTTC's Southwest Region on May 11, 2000, in Farmington, New Mexico.

BOTTOM LINE

Integrated, multi-disciplinary reservoir characterization and simulation studies of tight gas sandstone reservoirs in the San Juan Basin are required to determine the optimum location and expected recovery from infill wells. Core and outcrop analysis, seismic data, production data analysis techniques and reservoir simulation are tools used in the analysis. A spreadsheet-based infill drilling locator program has been developed to assist producers in analyzing infill drilling potential in these and other similarly fractured reservoirs.

PROBLEM ADDRESSED

Production from tight gas sandstone reservoirs in the San Juan Basin is highly dependent on natural fractures. Although enhancing overall permeability, the fractures create significant permeability anisotropy, which causes drainage areas around wells to be elliptical. Elongated drainage areas increase the potential for large sections of the reservoir being undrained. To optimally site infill wells, and know that there are sufficient reserves to justify drilling, one must understand reservoir geology, the natural fracture system, drainage patterns and the relative stage of depletion. Further data analysis and numerical modeling are often warranted to define upside potential.

KEY WORDS:

Naturally Fractured Reservoirs, Optimized Infill Drilling, Permeability Anisotropy, San Juan Basin, Tight Gas Sandstones

SPEAKERS

Overview
Lawrence Teufel, New Mexico Tech

Outcrop and Core Fracture Study
Scott Cooper & John Lorenz, Sandia National Lab

Seismic Analysis of Dakota Formation
Bruce Hart and Robin Pearson, New Mexico Institute of Mining and Technology, New Mexico Bureau of Mines and Mineral Resources

Reservoir Simulation of Mesaverde Pilot Areas
Hamoud Al-Hadrami, New Mexico Tech

Analyzing Tight Gas Production Data
Her-Yuan Chen, New Mexico Tech

Producing Characteristics and Drainage Volume
of Dakota Wells
Arild Sunde , New Mexico Tech

Infill Well Location Calculator
Mike Kelly, New Mexico Tech

TECHNOLOGY OVERVIEW

The US Department of Energy is sponsoring an integrated, multi-disciplinary study of infill drilling potentially in the naturally fractured reservoirs of the San Juan Basin. The investigation involves describing the natural fracture systems and reservoir heterogeneities, geological and geophysical modeling, definition of anisotropies and resulting elliptical drainage areas, and determining the optimal location for infill wells and estimating their incremental recovery.

Natural fractures in outcrops along the northern edge of the San Juan Basin indicate that the dominant fracture set within both the Mesaverde Group and Dakota Sandstone is composed of bed-normal extension fractures that strike between N5°W to N25°E when bedding is returned to horizontal. Two additional pairs of conjugate fractures, a pair with a bed-parallel axis of intersection and a pair with a bed-normal axis of intersection, are limited to siliceous sandstones of the Dakota formation. These fracture sets and orientations are also observed within core and are compatible with a tectonic reconstruction in which the fractures formed during overall N-S to NNE-SSW direction, Laramide-age shortening.

Local variability of fracture strike around the basin is common and is due to local changes in the maximum shortening direction and to mechanical stratigraphy. Differences in the direction of shortening are attributed to structures bounding the basin. These structures include a series of basement-involved thrusts on the eastern, northern and western margins of the basin. Mechanical stratigraphy is a primary control on fracturing and hydrocarbon production in the basin. Changes in mechanical stratigraphy that control fracture characteristics are attributed to compositional, diagenetic and depositional variability across the basin. Stratigraphic control on mechanical rock properties and by extension, on fracture characteristics, is evident in the differences between fracture patterns found in the Dakota and Mesaverde sandstones.

Three 3-D seismic surveys covering 90 square miles were analyzed to establish structural and stratigraphic frameworks within the basin and to evaluate means of identifying fractures. Complicating factors for seismic analysis were artifacts related to geometries of the original surveys, low signal-to-noise ratios, coherent noise and seismic thin beds. In analyzing the data, the seismic data were separated into three subsets corresponding to the original surveys. Logs from 150 wells, production data and outcrop observations were considered in the analysis.

For the Dakota interval in the eastern study area, a combination of three seismic attributes was able to predict net sand thickness of the Two Wells (an upper Dakota sand member) with reasonable confidence. The depositional environment there is dominated by offset, stacked shore parallel sand bodies. Individual sand bodies within these members were not seismically resolvable. In the west study area where Dakota sands are thickest and the best production (with inferred higher fracturing) occurs, seismic attributes did not correlate with sand thickness.

In the Mesaverde, lithostratigraphy makes formations mappable, but flow units cross boundaries and depositional histories are obscured. High dip areas with higher fracture density were found to be where the best producers were located. Seismic horizon/curvature attributes were able to detect fracture swarms.

Clean sandstones are more brittle and have a higher fracture density than do dirty, shaly sands, an observation confirmed in outcrops and cores. Stratigraphic complexity is apparent in both the Dakota and Mesaverde Sandstones. Both stratigraphy and lithology, which influence fracturing, are important controls on production. Different locations/orientations of fracture swarms and faults exist at Mesaverde-Dakota-Basement levels. Overall, there is a higher fracture density at the Dakota level.

A reservoir model was developed for the Mesaverde Group and well performance numerically simulated. The Mesaverde Group contains three basic units—the continuous Cliff House Sandstone, the Menefee which is a heterogeneous shaly interval, and the continuous Point Lookout Sandstone. Initially the area was developed on 320-acre spacing, infilled to 160-acre spacing beginning in the late 1970s, followed by further downspacing to 80 acres beginning in 1998. Two models were developed: a small-scale model with three layers (one for each interval) and a large-scale model with 10 layers (two in the Cliff House, six in the Menefee, and two in the Point Lookout). The small-scale model was made to evaluate anisotropy, which turned out to be 10:1 in a N-S direction. Production through 1989 was history-matched using the large-scale model, then predictions from 1989 through 1999 were used to verify the model. Later, predictions for another 30 years history were made. In the two pilot areas simulated, ultimate recovery with properly located 80-acre infill wells was 26 to 44% higher than observed with 160-acre spacing, confirming that downspacing to 80 acres is justified.

Transient flow, that is the period of time before stabilized flow is achieved, is the most important parameter for tight gas production. Pressure transient analysis and material balance methods do not work well in tight gas reservoirs due to the extremely long shut-in times, with associated lost production, required. But several methods are available for analyzing production data during the transient flow period. Rate vs. time log-log plots best evaluate stimulation effectiveness. Cumulative production vs. time log-log plots make it easy to determine recovery efficiency at a given cut-off time. Rate vs. cumulative production log-log plots give good resolution during transient flow and are very good for removing the effect of variable flowing pressures. Dimensionless radius (reD) equals equivalent drainage radius (re) divided by the apparent well radius (rwa). Stimulation, or negative skin, makes rwa behave larger, making reD smaller. Lower values of reD correlate to higher flow rates during the transient period, higher recovery at a given time, and higher flow rates at a given economic recovery. Type curve matching allows one to calculate these critical parameters from production data. These parameters can then be used in numerical simulation efforts.

New Mexico Tech developed a spreadsheet-based infill well location calculator program to assist producers in planning infill wells in tight gas reservoirs. The model uses data normally known for the well (porosity, net thickness, gas properties, saturations, etc.), combined with data on the direction and degree of permeability anisotropy, to calculate the optimum location for an infill well. The model takes data from up to four surrounding wells and determines optimum location for an infill well based on the total production of all the wells.

LESSONS LEARNED

When analyzing tight gas sandstone reservoirs, producers must develop a thorough understanding of the natural fracture systems which typically control reservoir productivity. Complementary tools of geological and geophysical modeling, production data analysis and reservoir simulation are recommended to explain/understand past reservoir performance and predict performance with different infill well scenarios. First and foremost, the degree and direction of permeability anisotropy must be determined. Then considering the stage of reservoir depletion, reserves and production from optimally located infill wells can be determined.

CONNECTIONS:

Lawrence Teufel
New Mexico Tech
801 Leroy Place
Socorro, NM 87801
Phone: 505-835-5483, e-mail: teufel@griffy.nmt.edu

Scott Cooper
Sandia National Lab
P.O. Box 5800
MS 0750
Albuquerque, NM 87185-5800
Phone: 505-844-3977, e-mail: spcoope@sandia.gov

Bruce Hart
McGill University
3450 University Street
Montreal, Quebec H3A 2A7
Phone: 514-398-3677, e-mail: hart@eps.mcgill.ca

Her-Yuan Chen
New Mexico Tech
801 Leroy Place
Socorro, NM 87801
Phone: 505-835-5743, e-mail: her@nmt.edu

Mike Kelly
Keltic Services
P.O. Box 116
Roswell, NM 88202
Phone: 505-622-2550, e-mail: mkelly@baervan.nmt.edu

For information on PTTC’s Southwest Region and its activities contact:
Robert Lee, Director, Petroleum Recovery Research Center
801 Leroy Place, Socorro, NM 87801
ph 505-835-5938, fax 505-835-6031, e-mail prrc@prrc.nmt.edu

Disclaimer: No specific application of products or services is endorsed by PTTC. Reasonable steps are taken to ensure the reliability of sources for information that PTTC disseminates; individuals and institutions are solely responsible for the consequences of its use.

The not-for-profit Petroleum Technology Transfer Council is funded primarily by the US Department of Energy’s Office of Fossil Energy, with additional funding from universities, state geological surveys, several state governments, and industry donations.

Petroleum Technology Transfer Council, 2916 West T. C. Jester, Suite 103, Houston, TX 77018
Toll-free 1-888-THE-PTTC; Fax 713-688-0935; E-mail hq@pttc.org; web www.pttc.org