Microseismic Monitoring of Hydraulic Fractures in Prudhoe Bay

Abstract

Abstract Production stimulation is commonly performed on reservoirs by hydraulically fracturing the formation. Fracture geometry is controlled by the regional stress field, strength and contrast in local rock stress and fracture fluid properties. Computer models predict the length and height of the fracture as a function of injection volume, pressure and rate, and, fluid and rock properties. Until recently, these computations were unable to be verified except for the fracture half length with pressure transient analysis and through log interpretation in close proximity to the well. Direct confirmation of fracture geometry has been shown possible with availability of downhole geophysical monitoring of the seismicity associated with the fracture production. This paper presents results of a test to directly measure the height, length, and azimuth of a fracture generated in the Prudhoe Bay Field, Alaska. Two triaxial geophone tools were deployed during August 1993 to monitor for microseismicity associated with a hydraulic fracture. One tool was deployed in an offsite monitoring well, and the second tool was deployed in the rathole of the fractured well. The objectives of the test were (1) to determine if seismicity existed and if it could be detected from within the fractured well and from the distant offsite monitoring well, (2) to evaluate techniques used to locate the fracture, and (3) provide guidance for the development of seismic recording systems used in fracture monitoring. Introduction Three techniques were used to monitor the hydraulic fracture. Each offered different operational advantages and disadvantages. This experiment attempted to evaluate each of these for future development. They were:offsite monitoring of the hydraulic fracture from a well that was 1450 feet away,monitoring from within the rathole of the fractured well,documenting the change in the vertical-to-horizontal earth noise ratio (H/Z) prior to, and following the hydraulic fracture in a deviated well for fracture height determination. Seismic monitoring of hydraulic fracturing has been successful in locating fracture paths in crystalline rock at the Hot Dry Rock Geothermal sites in New Mexico and Great Britain. It was hoped that these same techniques would work to monitor a hydraulic fracture in reservoir rock from an offsite monitoring well which was approximately 1450 feet away. Additionally, fracture height, length and orientation have been successfully monitored from the rathole of a water injection well in the North Sea. Finally, the Gas Research Institute has published a report and secured a patent outlining a unique method of determining fracture height known as H/Z. The primary goals of this test were to first determine if seismicity existed and if it occurred at rates high enough to be mapped in the Prudhoe Bay reservoir. Secondly, to evaluate and test the three techniques outlined above to determine if they would provide acceptable data for fracture monitoring. Finally, if adequate data was obtained, to map a hydraulic fracture using one or all of the above techniques. Background seismicity and perforation shot data were collected to determine the levels of seismic activity in the formation and document how well the formation propagated seismic energy. These data also allowed calibration of the seismic recording systems, and provided velocity information used in the location algorithms. It should be noted that the offsite DS 17-16 monitoring well had a lower zone fractured 10 days prior to the monitored fracture, and that proppant had been left over the open perforations to provide a "quiet environment for monitoring. The selection of well 17-16 permitted long baseline measurement of 10 day old fracture related seismicity. Any events located in the near wellbore area of well 17- 16 were mostly likely associated with the 10 day old fracture. Data were recorded in both the liner and the tubing of the monitoring well to determine if data could be obtained in the tubing of a completed well. This had obvious operational considerations, since a successful recording would obviate the need to pull tubing and set bridge plugs in future projects. The hydraulic fracture treatment was designed to simultaneously fracture two different zones, Sag river and Zone 4. In an attempt to get the best hydraulic fracture design possible, the fracture program consisted of three major steps. First, a high rate injection profile was performed to determine the distribution of flow between the two zones at rates above fracture pressure. Next, a comprehensive data frac was performed, and finally, the hydraulic fracture with proppant was completed. P. 387^