Is the Well Depleted or Damaged? A Productivity Index That Fell 44%
Well X-12 produced 1,200 BOPD at a flowing bottomhole pressure of 2,800 psia when the reservoir pressure was 3,200 psia. Six months later it makes 700 BOPD at a flowing bottomhole pressure of 2,300 psia. A static gradient survey run 6 hours after shut-in gives 2,720 psia. Water cut has risen from 5% to 35%, the bubble point is 2,400 psia, and the well is on continuous gas lift at 1.2 MMscf/d. The offset well 2,000 ft away in the same sand still calculates 3,200 psia from its own surveys. Management has a $1.2 million workover ready to approve and wants a yes or no this week.
1. Compute the productivity index now and before. How much of the rate loss is explained by depletion and water, and how much is not?
Read the worked answer
Productivity index
- Initial: PI = 1,200 / (3,200 - 2,800) = 3.0 BOPD/psi
- Now, on oil: PI = 700 / (2,720 - 2,300) = 700 / 420 = 1.67 BOPD/psi
- Now, on total liquid: 700 / 0.65 = 1,077 BFPD; PI = 1,077 / 420 = 2.56 bbl/psi
What depletion alone would give. If the PI had stayed at 3.0 and there were no water, the well would make 3.0 x 420 = 1,260 BOPD. It makes 700.
Applying water cut. At 35% water cut, the liquids that a 3.0 bbl/psi PI would deliver are 1,260 BFPD, of which oil = 1,260 x 0.65 = 819 BOPD. That is the rate the well should be making on depletion plus water cut alone, assuming the liquid PI is unchanged.
The gap. Actual 700 BOPD versus 819 expected is a shortfall of about 15% of liquid productivity (2.56 versus 3.0 bbl/psi). That part is not explained by depletion or water cut.
Caveats that matter here
- The 2,720 psia figure came from only 6 hours of shut-in. A damaged or low-permeability well can still be building up after 6 hours, in which case the true static pressure is higher than 2,720 and the PI is even lower than calculated above.
- Flowing BHP of 2,300 psia is below the 2,400 psia bubble point, so free gas is evolving in the near-wellbore, which suppresses the oil PI independent of any damage.
So the honest answer is: depletion and water cut explain most of the drop, but roughly 15% is unexplained and may be recoverable.
2. What mechanisms could account for the unexplained loss, and what data would separate them?
Read the worked answer
Candidate mechanisms
- Skin damage - scale (water breakthrough came with it, so scaling is plausible), migrating fines, asphaltene/paraffin deposition, or a water block. Shows up as positive skin with unchanged kh.
- Water coning or channelling reducing effective oil permeability. Water cut jumped 5% to 35% in six months; if that water is coming from a high-permeability streak it may be short-circuiting the oil column.
- Compartmentalisation. The offset well at 2,000 ft still sees 3,200 psia while X-12 sees 2,720 psia. Either X-12 is in a separate compartment, or the 6-hour survey is still building up. These have opposite implications.
- Lift inefficiency. Rising water cut and a fixed 1.2 MMscf/d of lift gas will load the well up; the recorded 2,300 psia flowing BHP may not be what the well can actually produce against.
- Gas evolution below the bubble point in the near-wellbore, as noted above.
Data that discriminates
- A proper pressure build-up (72 hours minimum, with a downhole shut-in if possible): gives skin, kh, wellbore storage and the distance to any boundary. This one test answers items 1, 3 and largely 5.
- A production log while flowing: identifies which perforations are making water and which are making oil. Answers item 2.
- A flowing gradient survey: confirms the true flowing BHP and the lift performance, rather than relying on a surface-derived number. Answers item 4.
- Water chemistry comparison between produced water and any suspected source (aquifer versus injection water) to identify where the water is coming from.
3. Do you approve the $1.2 million workover? Give your recommendation and the decision logic behind it.
Read the worked answer
No - not this week, and not for $1.2 million on current information.
Why refusing is the technically defensible answer
- The analysis above shows that most of the decline is depletion and water cut, which a workover cannot fix. At best you are chasing a 15% liquid-productivity shortfall plus whatever lift inefficiency exists.
- The single most important number - the true static pressure - is based on a 6-hour shut-in that is probably still building up. A $1.2M rig workover justified by a number that might be wrong is exactly the decision this industry gets wrong.
- A workover is a rig job. Several of the likely fixes do not need one.
What to approve instead (roughly $150-250k)
- 72-hour build-up with downhole shut-in: skin, kh, boundary.
- Production log: water entry by interval.
- Flowing gradient survey: actual Pwf and lift performance.
Then decide on the results
- High skin, non-damaging water from the oil zone: a coiled-tubing scale squeeze or acid job at a fraction of $1.2M, not a rig workover.
- Log shows water from one discrete interval: water shut-off (cement plug, polymer, or mechanical isolation) is the fix, and it is cheap relative to a workover.
- Build-up shows a boundary at 2,000 ft or less: the well has a small drainage volume, the offset is in a different compartment, and neither a workover nor a stimulation will recover much - the money belongs in an infill location evaluation instead.
- Flowing BHP is materially worse than assumed: this is a gas-lift optimisation problem, not a wellbore problem. Fix the lift first.
Approve the diagnostic programme now and hold the workover approval until the build-up is interpreted.