Abstract: Did Checkerboard Steamflooding Work at Kern River?

HARRIS, DAN, Chevron Corp.

The steamflood in the Kern River field was vertically expanded to the K and G sands in the Monte Cristo No. 2 lease in 1991. The sands were heated in a checkerboard style (the K sand in one pattern, the G sand in the next pattern, the K sand in the next pattern, and so on) as suggested by a simulation study. Now that eight years have passed, we evaluated the lease to see how the steamflood fared and to look for any new developments. Technically we found that checkerboard steamflooding was mostly unsuccessful. Several patterns had unswept oil, so we are steaming previously unsteamed zones to recover it. On the bright side, however, the steamflood was a financial success in that more oil was produced with less steam than originally predicted. Time-lapse temperature-observation-well (TOW) data were assembled on a commercial software package for amazingly easy analysis of steam chest growth. TDT logs were run on producing wells that were not near TOW's. The logs were very helpful as we found even more areas of unswept oil. Also, we found more oil from downward hotplate heating and are producing it today. Five injector workovers used new dual-steam-cup packers instead of the conventional dual packer/magnum configuration, and this has saved us a lot of money. We already have plans to use dual-cup packers on other Chevron-operated fields. This case study is intended to serve as a learning experience to those who are thinking about checkerboard-steamflooding a lease. We learned a lot and wish to share our findings and new technologies with the industry.

AAPG Search and Discovery Article #90911©2000 AAPG Pacific Section and Western Region Society of Petroleum Engineers, Long Beach, California