Field experience with well to well tracer tests

Interwell tracer tests are widely used. This article reviews some of the studies reported in open literature. The selection introduces different problems that have been addressed, but the original papers should be studied to obtain a more detailed description of the programs.

Tracers in WAG programs

The Snorre field is a giant oil reservoir (sandstone) in the Norwegian sector of the North Sea. Injection water and gas were monitored with tracers, 18 and the resulting tracer measurements are discussed in this page.

The same tracers used in the Snorre field have been injected in the Gullfaks field in the North Sea. The tracers identified unexpected communication paths between layers. The results contributed to methods for improving the WAG recovery performance.

Water and gas tracer injection in fractured reservoir

A tracer test was carried out in the Spraberry trend in west Texas. In this field, the oil was produced primarily from fractures and not from the matrix porous sandstone. To carry out the most efficient waterflooding strategy, the knowledge of fracture direction was imperative. Instead of a costly water-injection pilot project, natural gas with tracer was injected into a central well for 16 weeks to define the fracture pattern. The ⁸⁵Kr tracer applied was injected continuously with a concentration 28 times minimum detectability, which was 13,000 Bq/m³ (10^-8 Ci/ft³) gas. The detector system applied was a thin-wall, beta-sensitive Geiger-Muller tube that records the tracer amount in the effluent gas continuously.

At the beginning of the thirteenth week, a 2-Ci slug was injected, and at the beginning of 14th week, a 1-Ci slug was injected. Radioactive gas was recorded only in two wells. No radioactivity increases were detected at any of the other 10 wells monitored, either continuously or intermittently, during the tracer survey. The cyclic nature of tracer appearance was attributed to the necessity for overcoming the varying hydrostatic head of oil in the well and to the appreciable difference in effective permeability of the principal and cross-fracture systems. Apparently, neither of the two wells in which tracer was recorded intersected the same fracture plane as the injection well. The tracer survey confirmed predictions about the general line of fracture orientation in the lower formation of the Sprabarry trend.

Gas and water tracers in the el furrial field

Viela et al. reported the application of both gas and water tracers in the El Furrial field. The radioactive water tracers applied were isopropyl alcohol, HTO, ²²Na, and thiocyanate. Breakthrough time recorded was between 1 and 3 years. In the same field, gas tracers also were added. The tracers applied were PMCP, PDMCB, and SF₆. The conclusion was that the tracer survey was important to verify expected flow behavior and to identify unexpected communication paths.

Tracers for gas injection

Welge described one of the pioneering works of tracer application to follow injected gas. The communication between the wells in a small part of the Cromwell pool in central Oklahoma was investigated. The well spacing in the area was approximately 65 m. Three radioactive tracers were used: HT (only tritiated hydrogen is mentioned which may be either HT or T₂), CH₃T, and ⁸⁵Kr. One Ci of the tritiated compounds and 280 mCi of ⁸⁵Kr were injected into the same well. In the nearest well, the tracers were produced after 7 days, which indicates an average flow rate of 10 m/day or more than 1 ft/hr. The tracers were not injected simultaneously; therefore, a comparison of the response curves should be done with care. The curves show small variation with the methane peak produced a day or two later than tritium and ⁸⁵Kr. It also seems that the ⁸⁵Kr data are a bit more spread out than the other data. The test results were used to improve the knowledge about the sweep area and the volumetric flow in the different directions.

Calhoun reported the use of several gas tracers in the Fairway field. The program consisted of three phases of injection. In the first phase, 10 Ci of tritiated hydrogen and ⁸⁵Kr were injected. Four months later, 10 Ci of tritiated methane, tritiated hydrogen, and ⁸⁵Kr were injected in a new set of injectors. In the third phase, 10 Ci of tritiated methane and ⁸⁵Kr were injected again in several wells. The extensive tracer program showed the source of gas breakthrough of 25 production wells. Continuous sampling of these wells after gas breakthrough has indicated a change in front configurations and fluid-migration pattern. Tracer responses proved the need for controlled injection and withdrawal to even out sweep configurations. Furthermore, tracers have indicated those areas in which injection rates and gas/water cycles should be changed to reduce fingering of the injected gas. This knowledge has been useful in alternating injection cycles and rates to control GORs at a reasonable level. Radioactive-tracer results also indicate that high-pressure gas injection at the Fairway is yielding additional oil recovery, either through swelling of the residual oil or by partial miscibility.

Tinker reported the use of tritiated hydrogen and ⁸⁵Kr as tracers under methane injection in the Trembler zone II reservoir of the East Coalinga field, California. The tracers revealed that desaturated intervals were often continuous over wide areas of the field. Those intervals could greatly influence an injection project by acting as a thief zone for injection water. The tracer study suggested generally better reservoir-sand continuity than could be inferred from a related outcrop study.

Tracers in the enriched-gas injection

In the Prudhoe Bay field, ⁸⁵Kr, tritiated methane, ethane, and propane have been used as tracers for an enriched-gas flooding. In the South Swan Hills unit, Alberta, Canada, tracers were applied to follow enriched gas in a water/solvent cyclic-injection program. This is a limestone reservoir, and the tracers applied were tritiated hydrogen gas, tritiated ethane, and ⁸⁵Kr.

To gain information about the interwell reservoir characteristics, relative fluid velocities, and volumetric sweep efficiency in the early life of this project, it was desirable to trace the interwell movement of both the injected solvent and water. Tracers were added in as many as 14 solvent and 14 water injectors. The tracer responses in the wells were applied to redesign the injection program to achieve better sweep efficiency. The results proved to be of value as a qualitative indication of sweep, showing water and solvent flowing together to the majority of the offset wells. In some cases, unexpected flow paths were identified. No quantitative interpretation and information on project performance, however, have been derived from these data.

High-pressure N₂ miscible injection

In the Jay/Little Escambia Creek field, five radioactive tracers and one chemical tracer were used to tag injected N₂. The radioactive tracers used were ⁸⁵Kr, tritiated hydrogen, methane, ethane, and propane. Sulfur hexafluoride was the lone chemical tracer. The N₂ was injected at a pressure of up to 7,600 psig for a period of 1 to 2 weeks before the well was switched over to water injection. The tracers produced are ⁸⁵Kr and tritiated propane. The main purpose of tracer injection was to determine the source of N₂ breakthrough. This knowledge enabled adjustment of injection rates and volumes to improve area convergence on the production wells.

Also in the Fordyce field, ⁸⁵Kr, HT, CH₃T, and C₂H₅T were applied in a high-pressure ( > 7,000 psig) miscible injection program. The gas injected was dry natural gas with an N₂ content of approximately 5%. The tracer-production data were used in a gas-sweep model to predict gas movement and to localize unswept areas.

Labeling of CO₂ injection

Craig reported the use of halogen compounds to trace injected CO₂. The compounds applied were freon-11, freon-12, freon-113, and sulfur hexafluoride. The halocarbons were reported to be detectable in concentrations down to 0.5 ppb in laboratory tests on produced fluids. The detection was carried out by separation through a GC column and registration by an electron-capture detector. These four tracers were injected in nine injection wells, and registration of the tracer content in the production was done from 23 production wells.

Residual-oil saturation in a Leduc miscible pilot

Because of the large remaining oil in place, Leduc Woodbend D-2A had the potential to be an ideal miscible-flood candidate. Before the miscible injection, it was important to quantify the remaining oil. An injector/producer pair on 64-m spacing was chosen as a pilot for the tracer test. The partitioning tracer applied was tritiated butanol. Fig. 4 shows the production curves of the tracers. On the basis of the retention of the butanol and methanol and the partition coefficient of these two tracers at the actual conditions, the residual oil between the two wells was measured. The tritiated methanol was regarded as a nonpartitioning tracer. The residual oil, S_or, was calculated on the basis of different arrival times of the two tracers, as shown in the production curve. The "peak," "half-peak height," and "breakthrough" gave the S_or at 34, 35, and 38%, respectively. The results were compared with the oil saturation obtained from sponge coring and single-well push-and-pull tracer tests. The sponge core gave an S or at 33%, while the single-well tracer test gave a result from 35 to 40%, depending on the porosity model applied.

• 

  Fig. 4 – Production curve of tritiated methanol and tritiated butanol in the Leduc field.

Noteworthy papers in OnePetro

Use this section to list papers in OnePetro that a reader who wants to learn more should definitely read

See also

Well to well tracer tests

Interpreting data from well to well tracer tests

Tracer flow in porous reservoir rock

PEH:Well-To-Well_Tracer_Tests