uFigures

uIntroduction

uInitial Drilling

uIsolation Sidetrack

uConclusions

uReferences

uAcknowledgments

uFigures

uIntroduction

uInitial Drilling

uIsolation Sidetrack

uConclusions

uReferences

uAcknowledgments

	Figure 1. Location map of the Wilmington Field, California.
	Figure 2. Structure map used in planning WTU 2369 horizontal section.
	Figure 3. Curtain plot – pre-well plan and structure.
	Figure 4. Distance-to-bed inversion provides mapping of the top of the reservoir.
	Figure 5. Revised curtain plot showing old sidetrack well responsible for water coning.
	Figure 6. Final section with isolation shale zone and distance to bed mapping.

 

Introduction

Horizontal wells are not usually used in the industry to update reservoir structure maps, which results in maps that are missing crucial information about the geometry of the reservoir. Consequently, wells are being planned and drilled with increased uncertainty. Unexpected exits and drilling through water contacts are increasing costs. These unexpected events can be mitigated if appropriate sensor technology and appropriate data is acquired and used during the horizontal drilling of wells. In highly produced clastic reservoirs, such as the Wilmington field (Figure 1), well control forms the backbone of structure mapping. Conventional interpretation uses the abundant vertical wells data in the field to develop a structure map. The asset team uses the structure map to identify and plan horizontal wells to access unproduced pockets of hydrocarbons (Figure 2).

Well #1 of the WTU 2369 horizontal section was planned to place the wellbore 15 to 20 ft below the shale cap. To navigate the geologically complex structure (Figure 3), the InSite ADRTM resistivity tool was selected to geosteer the well (Bittar et al., 2007). This system provides the necessary data and the needed distance to boundary information to enable accurate placement of the well in the target zone.

Initial Drilling

After drilling the casing shoe and entering the reservoir, drilling proceeded as planned. The distance to bed inversion provided mapping of the top of the reservoir until 3,680 ft MD (Figure 4). At this point, an unexpected fluid change was encountered, causing the inversion to collapse. The resistivity in the zone had changed from 40 ohm-m to 6 ohm-m. This change was unexpected from the pre-drill model, and it was apparent that water had displaced the oil in this area.

Revisiting the structure map and reviewing old records of the area showed that an old well that had produced considerable water caused coning in the immediate area around the wellbore (Figure 5).

Immediate remedial action was required. The drilling BHA was tripped and a cement plug was set up into the casing. A revised plan was developed where the new sidetrack was to drill to the north of the original plan, and to intersect the top shale cap of the reservoir. This must be placed precisely to enable the well to crest in the shale before reentering the reservoir, minimizing the lost production section while isolating the water cone from the wellbore. A plan was developed, but the execution was left to the geosteering team to optimize the entry and exit positions. It was realized that, even with good well control, a direct measurement of the boundaries would give tight control and optimize the well path, while minimizing the amount of non-productive zone.

Isolation Sidetrack

Drilling resumed from the casing shoe, with the well being deviated to the north and away from the original hole. The distance to bed inversion algorithm began to pick up the upper boundary of the reservoir at 3,400 ft MD and indicated the boundary position at 20 ft above the well (Figure 6).

This was at the maximum range of the tool and the subsequent inversion was “noisy,” but we mapped the boundaries and found that they compared very well with the revised structural map.

The well was drilled up with a desired exit from the reservoir at 3,700 ft MD (2,660 ft true vertical depth (TVD)). The boundary was mapped and the exit was made at the correct point. The boundary was continually mapped below the tool so that the exit point could be refined, which was done at 2,875 ft MD (2,667 ft TVD).

From the reentry point to the end of the well at 5,905 ft MD, the upper boundary was in almost constant range of the tool. At one point, at 4,850 ft MD, the structural variation took the boundary out of range of the tool. This defined the range of boundary detection of the InSite ADR™ tool in this reservoir to be 22 ft in radius. The boundary was mapped shortly afterward, at 5,010 ft MD, and mapped successfully to TD.

Conclusions

By using the best of existing new technology, well placement has undergone a revolution in its capability to deliver results. This example illustrates real-time mapping of boundaries and the flood zones, in which zonal isolation was accomplished without compromising well design or incurring additional completion costs. The ability of a well-trained and capable team, given the most appropriate tools for the job, to perform real-time well plan modifications to the extent shown here demonstrates how efficient and cost effective such systems can be in the right application.

References

Bittar, M., J. Klein, R. Beste, G. Hu, M. Wu, J. Pitcher, C. Golla, G. Althoff, V. Sitka, V. Minosyan, and P. Paulk, 2007, A new Azimuthal deep-reading resistivity tool for Geosteering and advanced formation evaluation: SPE Annual Technical Conference and Exhibition, 11-14 November 2007, Anaheim, California, Paper SPE 109971. Web accessed 22 September 2009 http://www.onepetro.org/mslib/servlet/onepetropreview?id=SPE-109971-PA&soc=SPE

Acknowledgments

The authors would like to thank the management of Warren E&P and Halliburton Energy Services for permission to publish this work.

Return to top.