INFILL WELLS CONTRADICT CLAIMS OF NEW GAS IN HUGE HUGOTON FIELD

Thomas F. McCoy, Dave E. Reese, M.J. Fetkovich, Riley B. Needham, Bruce E. Freeman

Phillips Petroleum Co. Bartlesville, Okla.

OPTIONAL INFILL DRILLING IN THE KANSAS HUGOTON 0 GAS FIELD has failed to prove increased gas reserves, contradicting claims that in some infill wells higher pressures and deliverabilities indicate new reserves.

A study by Phillips Petroleum Co. Determined that the higher pressures and deliverabilities result from differing completion efficiencies and/or infill well completions in different layers than in the companion original well. But an improved completion was found not to be unique for an infill well; the same improvements also could be obtained by restimulating the original well.

Statistical analysis of infilled units, on production through September 1992, confirms conclusions that infill wells have not encountered gas not being drained by the original Wells. 1-3 Although deliverability and wellhead shut-in pressures of infill wells differ from the companion original well, this difference is not from contacting untapped gas-in-place.

These higher pressures and deliverabilities were shown to be caused by the pressure gradient in the most permeable layers and by completion efficiency differences between the infill and original wells.

Case histories in this article illustrate some causes for the different performances of the infill and companion original well. These comparisons include:

Oldest infill wells with almost 5 years of performance data
Infill wells encountering the highest initial wellhead shut-in pressures
Infill wells with the largest differences in deliverability.

BACKGROUND

The Hugoton field, the largest gas accumulation in the lower 48 states, covers 6,500 sq miles in three states. About two thirds of the field lies in southwest Kansas.

In September 1992, the Kansas Hugoton had 5,889 producing gas wells, including 1,513 infill wells. Initial gas in place (GIP) was about 30 tscf. About 10 tscf remain.

The Kansas Corporation Commission (KCC), in April 1986, amended its basic proration order 4 to permit drilling a second optional well on all basic proration units of 480 acres or greater in the Kansas Hugoton field. The KCC based its decision, in part, on the premise that these wells would recover an additional 3.5-5.0 tscf of gas that could not be recovered from existing wells.

The gas pay in the Hugoton field is the Chase group that is subdivided primarily into carbonate units and interlayed shaly units." The productive layers, composed of both clastics and carbonates, include from bottom up the Fort Riley, Towanda, Winfield, Krider, Herington, and Hollenberg. Reservoir intervals are separated by sealing shaly units.

The reservoir and shaly units exhibit characteristic log signatures recognizable throughout the field.

Parts 1 and 2 1-3 of the Phillip's study were presented in 1990. Part 3 is discussed in this article.

Part 1 focused on the performance of the first 659 infill wells drilled in the Kansas Hugoton. Part 2 reviewed Mesa Petroleum Co.'s 1977 test program to determine if increased reserves and improved deliverability could be obtained by drilling an additional well on each of five 640-acre units. 7 8

One aspect of the five-replacement-well study was that the original wells were shut in for nearly 10 years, and wellhead pressures were recorded monthly while producing the replacement wells.

The wellhead pressures, combined with pressure and flowmeter data taken on each of the four no-crossflow layers in the productive interval, provide a unique look at the performance relationship between two wells on the same 640-acre unit. No evidence was found that the replacement wells encountered gas not already being drained by the original wells.

Companion papers on the Guymon-Hugoton field 6 9 10 support conclusions reached in Phillips' three-part study on infill wells.

DATA ANALYSIS

The 1,513 infill wells analyzed in Part 3 of the study are listed in the October 1992 Annual Hugoton Field Report. The statistical study is based on 1,431 infill and original well pairs. The other 82 pairs had a missing test for either the infill or original well.

As of October 1992, only 1,513 of the 4,176 infill locations (36%) had been completed. The overall infill drilling activity continues at a significantly slower pace than allowed by the KCC 1986 order.

Infill wells have not appreciably affected total production from the field.

SHUT-IN PRESSURES

Initial shut-in wellhead pressures, Piwhs, encountered by infill wells between January 1987 and September 1992 averaged 149.3 psia. These pressures are significantly lower than the original field pressure of about 450 psia. The average reduction in initial pressure of more than 300 psi is evidence that reservoirs in every new infill well were drained by existing wells.

Fig. 1 updates the cumulative frequency and frequency distribution histogram for the pressure difference, [see equation]. The average difference is 17.4 psi, with an expected value of 13.0 psi.

Part 1 reported a pressure difference that was 3.8 psi lower. The slightly larger difference can be attributed to a larger percentage of the newer wells being drilled on lower-permeability edges of the field. These larger differences, in general, reflect the pressure gradient toward the original well in the more permeable layers and do not reflect new gas.

Only 7.1% of the infill wells encountered pressures that are 50 psi or higher than pressure in existing wells. In these cases, the pressure difference is, in general, caused by differences in completion efficiencies between the infill and original wells.

Case histories, discussed later, illustrate that in all cases examined, completion efficiencies play an important role in infill well pressures.

OFFICIAL DELIVERABILITY

Part 1 demonstrated that the higher official infill well deliverabilities did not reliably indicate new gas.

Fig. 2 updates a cumulative frequency and frequency distribution histogram for the difference in infill and original well official deliverabilities used in determining well allowables. The update, for all practical purposes, is identical to the histogram in Part 1. The higher deliverabilities do not reflect contact with new gas and are caused by a combination of two factors:

For equal completion efficiencies, the average higher initial pressures give infill wells a temporary higher deliverability.
Infill wells with better completion efficiencies can have a higher official deliverability.

Fig. 2 shows that about 30% of the infill wells have lower official deliverabilities than the companion original well. This is because many of these infill wells have "poorer" completions than the original well.

In infill wells with significantly higher official deliverabilities compared to the companion original well, completion efficiency also dominates the well response.

CASE HISTORIES

Infill well histories were examined to determine if the gas found was in communication with the original well or other offset original wells. The data were obtained mainly from publicly available information from KCC, Petroleum Information Corp., and Dwight's Energydata Inc.

These data consist of completion information submitted to the KCC, monthly rate-time data, and 72-hr test data. For each infill and original well pair, these data are assembled into a monthly rate-vs.-time plot, a wellhead depletion back pressure curve, location plat, wellhead shut-in pressure-vs.-cumulative production, and a cross section between the infill and original well.

The key to understanding the performance of an individual well pair is the proper weighing of all available data.

UNIT A

Unit A illustrates a poor infill well completion that reduced the unit allowable.

The original well in Unit A was completed open hole in the Herington, Krider, and Winfield with 13,500 gal of acid on Nov. 6, 1946. The infill well was completed 43 years later Feb. 28, 1989 with 5 1/2-in. casing, cemented through the productive interval. This well was fracture stimulated with 111,825 gal of crosslinked gelled 2% KCl water and 367,800 lb of 12/20 sand.

Fig. 3 shows well performance and completion data for Unit A. As implied by the two shifts in the back pressure curve, in the 1960s, the original well was restimulated twice, resulting in significant deliverability increases.

The infill well's Piwhs was 134.7 psia or much lower than the original well's 419.4 psia. The infill well's 72-hr shut-in pressures fall on the data trend established by the original well. This indicates that the infill well did not encounter gas that was not already in communication with the original well.

The 72-hr test data for the infill well falls far to the left of the back pressure curve for the original well. This indicates that the infill well is suffering from very poor stimulation results when compared to the original well.

This unit has lost allowable because of the infill well. In the November 1992 Hugoton Gas Report, the basic allowable was 7,722 Mscf/month. If the infill well had not been drilled, the basic allowable would be 11,414 Mscf/month, or 48% greater. Because of the poor infill well completion, this unit potentially could lose reserves to surrounding units.

UNIT B

In Unit B, the cause for the infill well's higher pressure is a higher pressure gradient to the original well, and the higher production rates result from a more effective completion.

The original well in Unit B was spudded July 17, 1946, and completed Aug. 21, 1946, with a 5-in. liner set, cemented through the productive interval. The Krider and the Winfield were perforated and acidized with 10,000 gal of acid. Initial potential was 6.47 MMscfd with a Piwhs of 421.0 psia.

The infill well, an unsuccessful Morrow deep test, on Nov. 21, 1989, was recompleted in the Chase Group as an infill well. This well was acidized over the entire perforated interval with 5,000 gal and then fractured with 55,700 gal of fluid.

Fig. 4 plots the available well performance and completion data for Unit B. Unlike many other wells in the Kansas Hugoton field, the depletion back pressure curve indicates that the original well was not restimulated with sand fracturing.

The infill well's Piwhs on Dec. 14, 1989, of 260.5 psia was about 100 psi higher than the original well's 159.0 psi on Oct. 12, 1990. After about 2 years of production, on Sept. 5, 1991, the Pwhs for the infill well was 165.4 psia, or almost 100 psi less.

The latest infill well pressure point falls nearly on the pressure trend established by the original well. The higher Piwhs observed in the infill well does not reflect contact with new gas but probably is caused by the pressure gradient that results from flow in the most permeable layers.

The monthly rate-vs.-time portion shows that the infill well is producing significantly more gas than the original well. The infill well's depletion back pressure curve lies far to the right of the original well's curve and, therefore, indicates a much better completion for the infill well.

Because the original well has a lower production rate at the same drawdown as the infill well, the stimulation in the original well was probably less effective than in the infill well. Therefore, the infill well's higher production rates are a function of completion efficiency and do not indicate contact with gas that is not being drained by the original well.

The better stimulation of the infill well and the subsequent significantly higher production rates have potentially increased this unit's reserves at the expense of the surrounding units.

Under the current field rules, the unit's production rates could be increased by abandoning the original well and only producing gas from the infill well.

This would result in a 65% increase in allowable, from 11,466 to 18,902 Mscf/month.

UNIT D

The infill well in Unit D shows how water in the well bore can lead to measuring higher pressures in the infill well compared to the original well.

The original well in Unit D was spudded July 27, 1963, and completed Aug. 20, 1963, with 4 1/2-in. casing, cemented over the productive interval. The Winfield was perforated and acidized with 250 gal before being squeezed off with 125 sacks of cement, presumably because the interval was wet.

After the Herington and Krider were perforated and acidized with 3,000 gal, open flow was 1.08 MMscfd. Following an additional 4,000 gal acid treatment, open flow was 1.15 MMscfd with 100 bw/d.

The infill well, originally completed Aug. 10, 1977, as an oil well in the Kansas City "B," was recompleted Mar. 29, 1988, as a infill gas well. The 5 1/2-in. casing was perforated in the Herrington and Krider and treated with 11,350 gal of water and 11,350 gal of 15% HCl. The zones tested 219 Mscfd and 63 bw/d. Fig. 5 shows the completion and performance data for Unit D wells.

The first official test of the infill well after pipeline hook-up on Dec. 15, 1988, 240 days after the completion, showed a Piwhs of 302.6 psia, and a test rate of 24 Mscfd at a working wellhead pressure, Pwhs of 270.9 psia. No water production was reported.

Thirteen days later, Dec. 18, 1988, the infill well was tested again. The pressures and gas rate were the same, but 16 bw/d were also reported. The Piwhs for the infill well was more than 140 psi higher than the Pwhs in the original well.

For the infill well, the first three official test points lie far to the left of the depletion back pressure curve for the original well. This indicates that the infill well has significantly less gas production capacity than the original well.

The Krider is generally much more productive than the Herington. It is probable that while the infill well was waiting on a pipeline hookup, the water level in the infill well rose above the Krider. Therefore, the infill Pwhs's measured, the first three data points, represent only the Herington. This is consistent with the higher Pwhs, relatively poorer performance, and historical water production of the infill well compared to the original well.

After the infill well started producing more gas in 1991, the well appears to be lifting some water from the bottom of the well bore and the Krider starts to contribute as evidenced by the official test on Mar. 5, 1992, showing a Pwhs of 226.5 psia, gas rate of 46.62 Mscfd, water of 40 bw/d, and a Pw of 200.6 psia, The Pwhs had dropped more than 90 psi and the 1992 test point is considerably to the right on the depletion back pressure curve.

The initial higher pressures observed in the infill well do not reflect contact with new gas but simply reflect water in the well bore that allows measurement of the pressure and rate only from the Herington layer.

UNIT E

The wells in Unit E show how the inability to lift water is a probable cause for the deterioration of well performance.

The original well in Unit E was spudded Apr. 10, 1949, and completed May 18, 1949, with a 5-in. slotted liner. The well was acidized over the entire productive interval with 15,000 gal. Initial flow potential was 21.968 MMscfd with a Piwhs of 445.5 psia.

The infill well, spudded June 22, 1990, was completed Sept. 14, 1990, with a 5 1/2 in. liner, cemented through the productive interval. The Hollenberg, Herington, Winfield, and Towanda were acidized with 5,000 gal of 15% HCl and then fractured with 90,600 gal of 30-lb crosslinked gel. Fig. 6 shows the completion and performance data for the unit.

The original well's back pressure curve shows scattered data that do not fall on a straight line. This indicates a steady deterioration in the performance of the original well. The official test data indicate water production.

Performance deterioration is also evident in the lower production rates beginning in 1977 and in the slope change on the Pwhs-vs.-Cp plot beginning at about 6 bcf.

The original well produced about 6 bcf in the first 30 years, but only about 1 bcf during the last 16 years. This long-term performance decline is probably because of an inability, to lift produced water.

The infill well's better performance is exhibited by the increased production rates and by the infill well's back pressure curve being to the right of the original well's curve. The infill well's pressures fall on the pressure trend established by the original well.

Compared to the original well, the infill well is producing significantly more gas. These comparatively higher rates are caused by performance deterioration of the original well and not by the infill well encountering gas not drained by the original well.

UNITS G AND H

Infill Unit G and offset Unit H wells show that better completions in infill wells and recompletions of original wells can both increase well production.

The original well in Unit G, spudded Apr. 12, 1947, was completed with 5-in. casing, cemented over the entire productive interval. The Fort Riley was perforated and acidized with 500 gal. A subsequent test flowed 1. 13 MMscfd and 22 bw/d. Presumably because of water production, the operator squeezed this zone with cement.

The Herington, Krider, and Winfield were then perforated and acidized with 28,500 gal. The absolute open flow was 15.9 MMscfd with a Piwhs of 416 psia.

The infill well, spudded June 24, 1987, was drilled to depth of 5,800 ft and was probably an unsuccessful deep test that was plugged back to 2,803 ft and completed as a Kansas Hugoton infill well. The Chase was fractured with 94 bbl of 2% KCl water, acidized with 5,000 gal, and then fractured with 42,000 gal of 25-lb gel. Fig. 7 shows the performance and completion data for Unit G.

The infill well's Piwhs is about 100 psi higher than the current pressure for the original well. The Pwhs-vs.-Gp plot shows that the infill pressure rise corresponds to a shift in the depletion back pressure curve to the right. This is a cleanup effect. The infill depletion back pressure curve falls on the curve for the original well. The rate-vs.-time plot shows that the infill well is producing more gas than the original well.

The infill well is located west of the original well and higher on the structure. The infill well is also perforated in the Krider, while the original well is not. The higher pressures in the infill well are simply a function of completion efficiency and do not reflect contact with new gas. This effect can be explained by the offset well in Unit H, a non-infilled unit located directly to the northwest of Unit G.

The well in Unit H, spudded June 3, 1954, was completed June 30, 1954, with a 5 1/2-in. slotted liner in the Herington, Krider, and Winfield. The well was acidized with 28,000 gal and had an initial potential of 14.1 MMscfd. The performance and completion information for the original well is found in Fig. 8.

The performance data for the Unit H and the Unit G original wells are nearly identical. In 1990, the Unit H well was deepened 40 ft to the Towanda, and a 4 1/2-in. liner was run to TD inside the 5 1/2-in. liner. The well was then restimulated.

After the recompletion to the Towanda, the Pwhs increased about 100 psi with only a small shift in the back pressure curve. The pressures and production rates for Unit H increased substantially without drilling an infill well. Based upon the performance of the Unit H well, the Unit G infill well is only taking advantage of a more-efficient completion and did not encounter gas that was not being drained by other wells in the field.

ACKNOWLEDGMENTS

We thank Phillips Petroleum Co. for permission to publish this article. Also, we would like to acknowledge Lester Wilkonsen for sharing his insight and historical perspective of the Kansas Hugoton. The assistance of C.D. Javine and S.A. Baughman is also gratefully acknowledged, We are also grateful to our colleagues at Mobil Exploration & Producing U.S. Inc. for the stimulating exchange of ideas and their contribution of well performance data.

REFERENCES

McCoy, T.F., et al., "Analysis of Kansas Hugoton Infill Drilling--Part 1: Total Field Results," Paper No. SPE 20756, SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.
Fetkovich, M.J., Needham, R.B., and Mccoy, T.F., "Analysis of Kansas Hugoton Infill Drilling--Part 2: Twelve Year Performance History of Five Replacement Wells," Paper No. SPE 20779, SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.
Mccoy, T.F., et al., "Analysis of the Kansas Hugoton Infill Drilling Program," Transaction SPE 20779, JPT, June 1992, pp. 714-23.
Kansas Corporation Commission, Order Docket No. C-164, Topeka, July 18, 1986.
Clausing, R.G., "Prefiled Testimony," Docket No. C-164, Kansas Corporation Commission, Topeka, 1985, pp. 1-26.
Siemers, W.T., and Ahr, W.H., "Reservoir Facies, Pore Characteristics, and Flow Units-Lower Permian Chase Group, Guymon-Hugoton Field, Oklahoma," Paper No. SPE 20753, SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.
Carnes, L.M. Jr., "Replacement Well Drilling Results--Hugoton Gas Field," Kansas University Heart of America Drilling and Production Institute Meeting, Liberal, Kan., Feb. 6-7, 1979.
Daugherty, M.S., "Report of Data Collected During Mesa's Five Replacement Well Drilling Program," Mesa Petroleum Co., Amarillo, Tex., 1977.
Fetkovich, M.J., Ebbs, D.J., and Voelker, J.J., "Development of a Multiwell, Multilayer Model to Evaluate Infill Drilling Potential in the Guymon-Hugoton Field," Paper No. SPE 20778, 1990 SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.
Ebbs, D.J., Works, A.M., and Fetkovich, M.J., "A Field Case Study of Replacement Well Analysis Guymon-Hugoton Field, Oklahoma," Paper SPE 20755, SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.