Cased-hole density offers new reservoir analysis alternative

Richard Odom, Shawn Bailey
Computalog U.S.A. Inc.
Fort Worth, Tex.

A new cased-hole logging technique can be used for prospecting in old wells, monitoring reservoir dynamics, or as an alternative when open hole logging is impractical or uneconomical.

The new technique being developed determines neutron and density porosities in cased wells with a through-tubing (15/8-in. diameter) pulsed-neutron system.^1-3

Neutron porosity measurements have been available for analyzing cased reservoirs for many years. This hydrogen-based porosity was valid in single-lithology reservoir sequences containing oil or water.

With the addition of a cased-hole density porosity, one can obtain a more accurate porosity determination in mixed-lithology reservoir sequences. This development brings to cased-reservoir analysis the possibility of identifying gas by overlaying density and neutron porosities.

Measurement theory

Pulsed-neutron systems typically create a pulse of fast neutrons and monitor the gamma rays created by the incident radiation pulse.

During the firing pulse of the accelerator, fast neutrons create a dispersed gamma-ray source near the accelerator, and the subsequent transport of these gamma rays is dependent on the density of the medium (Fig. 1 [212,229 bytes]).

In open hole logging, density is measured by focusing gamma rays from a chemical source, in a shielded pad, onto the borehole wall. This technique has a limited investigation depth and is applicable only when the pad is in contact with the formation.

In comparison, the pulsed-neutron density measurement from a dispersed gamma source has larger geometric and energy scales that allow sufficient penetration⁴ to measure density behind well casing.

A qualitative density indicator was commercialized in 1993 for the Computalog PND-S pulsed neutron system. The measured density parameter was normalized with data from offset wells or core data.

To improve the technique, research has recently been directed toward the following:

• Establish an exact theoretical model for measurement dynamics.
• Analyze uncertainties such as cement thickness and quality, well bore fluids, lithology, etc.
• Generate a borehole-compensated density measurement characterized for specific standard boreholes.

The existence of other simultaneous radiation and the dynamic nature of the gamma-ray source complicate the density measurement using the transport of gamma rays from inelastic scattering.

The density measurement for each detector requires a deconvolution process to isolate the gamma rays created by inelastic scattering. Because the gamma-ray source geometry and strength may vary, the gamma-ray transport must be measured in a differential sense. The received signals at both detectors are used to derive the gamma-ray diffusion length (LRHO).

As modeled, LRHO is a solution of the diffusion theory approximation of the Boltzmann equation.⁵

Given the centric geometry, the LRHO parameter is an average value for both borehole environment and formation. The semiquantitative density parameter is then characterized for the borehole environment.

Fig. 2 [44,675 bytes] compares the cased-hole pulsed-neutron (PN) density porosity to the open hole (OH) density porosity for density measurements from 14 wells in offshore Gulf of Mexico, the Gulf Coast, and Canada.

The mean-squared error of the cased-hole porosity compared to the open hole porosity for 30,500 one-half ft samples in 51/2-in. casing was 3.02 porosity units (p.u.).

The uncertainties inherent in cased-hole logging, such as open hole size, cement presence, hardware, etc., may require special consideration or normalization. Cross-correlation to offset wells, core data, or other sections in the well with open hole logs can help to qualify the accuracy of the cased-hole results.

Open hole alternative

In one application, after several unsuccessful attempts to reach the well TD with open hole logging tools, the operator decided to case the well and use the PND system to analyze the deepest target in the reservoir through casing. This target is a Wilcox sandstone gas reservoir that is a prominent oil and gas horizon across the Texas and Louisiana gulf coast.

Fig. 3 [69,549 bytes] shows the open hole and cased-hole results. The left-hand track shows the density and neutron porosities from open hole logging, while the right-hand track is the cased-hole density and neutron porosities extended to TD. Both tracks are scaled 30 to 0 p.u.

Over the section where both density measurements exist, the accuracy of the cased-hole density compared to the open hole density shows an average 1.77 p.u. error, the square root of the mean-squared error compared to open hole density porosity.

Using only the neutron porosity in this type of reservoir would leave the log analyst with an information set that was incomplete. Because the neutron porosity is based on hydrogen content, it cannot distinguish between changes in porosity and changes in gas content.

Field study

Often when a reservoir engineer does a field study in an old field, the first step is to tie together various types of data collected at different times over the reservoir life. Cased-reservoir analysis provides a consistent data set for the field that is current and instantaneous in petrophysical time.

In this field study, the operator planned a water flood for a previously developed San Andres and Grayburg field on the northern shelf of the Delaware basin in New Mexico. These formations are dolomites with interbedded sections of sandstones and siltstones. Reservoir analysis is further complicated by secondary mineralization of anhydrite.

The field has limited open hole logs from its initial development, with less than 10% of the wells having open hole logs or core data.

Typically in complex lithology reservoirs, the density and neutron porosities and the "Pe" (photoelectric absorption) factor are used to determine the lithology and true porosity. The Pe factor measured from modern open hole density systems is useful in determining the rock type because it is sensitive to the abundance of calcium (Ca). The cased-hole, density-neutron porosities and a cased-hole "Pe" factor from neutron-induced spectroscopy³ provide the log analyst a method to run the same types of complex-lithology reservoir models as with open hole logs.

In Fig. 4 [120,730 bytes], the first track is the lithology (percent by volume) determined from the log measurements. The second and third tracks show the correlation of the open hole to the cased-hole measurements, where the porosities are scaled 30 to -10 and the Pe factors are scaled 0 to 10. The current analysis of the bulk volume of oil (black) and water (blue), shown in the fourth track, is derived from carbon/oxygen water-saturation measurements.

The cased-hole model is developed and fine-tuned by correlating a few of the wells that have open hole and core data. Once the parameters for the cased-hole reservoir model have been established, cased-hole logging can provide a consistent reservoir analysis in the many wells included in the field study.

References

1. Odom, R.C., Streeter, R.W., Hogan II, G.P., and Tittle, C.W., "A New 1.625-in. Diameter Pulsed Neutron Capture and Inelastic/Capture Spectral Combination System Provides Answers in Complex Reservoirs," SPWLA Paper O, Society of Professional Well Log Analysts Annual Logging Symposium, Tulsa, June 19-22, 1994.
2. Odom, R.C., Hogan II, G.P., Crosby, B.W., and Archer, M.P., "Applications and Derivation of a New Cased-hole Density Porosity in Shaly Sands," SPE Paper No. 38699, SPE Annual Technical Conference and Exhibition, San Antonio, Oct. 5-8, 1997.
3. Odom, R.C., Hogan II, G.P., Rogers, C.B., Smith, R.L., Sirgo III, M.A., and Bailey, S.M., "Cased-Hole Lithology And Density Measurements," SPWLA Paper ZZ, Society of Professional Well Log Analysts Annual Logging Symposium, Keystone, Colo., May 26-29, 1998.
4. Wilson R.D., "Bulk Density Logging With High-Energy Gammas Produced by Fast Neutron Reactions with Formation Oxygen Atoms," IEEE Nuclear Science Symposium, San Francisco, Oct. 25-28, 1995.
5. Tittman, J., and Wahl, J.S., "The Physical Foundations of Formation Density Logging (Gamma-Gamma)," Vol. 30, No. 2, Geophysics, April 1965.

The Authors