Material Balance vs Volumetrics: Whether to Order Waterflood Pumps Now
A faulted anticline has produced for 38 months from 12 wells. Initial pressure was 322 bar at 2,950 m TVD; current pressure is 268 bar, and the last three static surveys show decline slowing to 0.6 bar/month. Cumulative production is 8.4 MMstb, with GOR rising from 180 to 310 scf/stb. Volumetrics from 3D seismic and nine wells give a P50 OOIP of 72 MMstb, with a range of 60-95 MMstb. A consultant's material balance assuming a small aquifer and no gas cap returns 38 MMstb, which he calls 'the real OOIP'. Bubble point is 240 bar, so the reservoir is still undersaturated. The field development plan assumed 14,000 rb/d offtake and water injection at 18,000 bw/d starting in year four. Partners must decide within 60 days whether to order injection pumps with a 16-week lead time, or continue depletion and monitor. Only one dedicated observation well exists, 1.1 km from the crest.
1. Reconcile the 38 MMstb material-balance result with the 60-95 MMstb volumetric range. What is the most likely explanation, and what would you measure before revising the OOIP used for waterflood design?
Read the worked answer
The two estimates are not measuring the same thing. Material balance with an assumed small aquifer and no gas cap is a non-unique history match: if pressure support exists, the calculation forces a small OOIP to explain the observed pressure. The evidence points to support rather than a 38 MMstb tank. Decline has slowed to 0.6 bar/month while cumulative production reached 8.4 MMstb; average voidage is roughly 8.4 MMstb x 1.25 / 38 months, or about 9,200 rb/d, so 54 bar of depletion over that voidage implies unusually strong rock/fluid compressibility for a undersaturated system unless there is influx. Rising GOR from 180 to 310 scf/stb could be solution-gas release, but pressure remains 28 bar above bubble point, so it more likely reflects lift or coning behaviour and should be checked, not used as proof of a gas cap.
Before changing the OOIP, acquire: (1) a pressure gradient and repeat formation tester in the observation well to locate contacts and confirm continuity; (2) an interference or pulse test between the observation well and a producer to quantify aquifer support; (3) production logging to identify water entry and gas breakthrough; (4) a material-balance sensitivity with aquifer strength as a variable, reported as a range, not a single number. Carry 65-75 MMstb as the working range and design injection for voidage replacement rather than for a fixed OOIP.
2. The pumps must be ordered now or the decision deferred nine months. Recommend a decision with explicit conditions, and quantify the cost of deferral as far as the data allow.
Read the worked answer
Order the pumps, but attach conditions that protect the downside. At the current 0.6 bar/month decline, nine months of deferral costs about 5.4 bar of pressure, moving the reservoir from 268 bar toward 263 bar and reducing the buffer to bubble point from 28 bar to about 23 bar. More important than the pressure itself is lost voidage replacement: at 14,000 rb/d offtake for 270 days, deferral leaves roughly 3.8 million rb of voidage unreplaced. If that accelerates decline and shaves even 2% from a 70 MMstb OOIP, the deferred recovery is about 1.4 MMstb, worth far more than the pump order.
The decision is not 'is the OOIP 38 or 72?' but 'what is the cost of waiting versus the cost of idle equipment?' Conditions: place the order with a cancellation or deferral clause and staged payments; confirm injection water quality and injector location using the observation well data; set a decision gate at 90 days with pressure and GOR thresholds; and do not commit the full 18,000 bw/d capacity until interference testing confirms connectivity. If the observation well shows no support and pressure falls below 250 bar, convert the programme to gas-lift or gas injection planning rather than force a waterflood.