SeparatorOilShrinkage
Automating Separator Oil Shrinkage Using the Montney EOS:
Case Study with ARC Resources
Case Study with ARC Resources
01 May 2026
Written by: Phil Carpentier, ARC Resources and Mathias Carlsen, whitson
In Canada’s unconventional plays, thousands of wells produce into complex gathering systems. The challenge is that many wells are only measured at separator conditions, and then quickly routed into extensive pipeline networks where fluids are commingled.
That sounds normal. But it creates a surprisingly painful engineering problem:
How do you consistently convert separator oil rates into stock-tank oil volumes on a well-by-well basis?
Stock-tank rates are often unavailable at the single-well level, yet they are essential for:
And when shrinkage is handled poorly, the error can be huge. Our internal studies show that failing to account for time-varying shrinkage can lead to stock-tank oil rate errors exceeding 20%, especially in near-critical volatile oils and condensate systems.
That’s exactly why ARC Resources and whitson partnered to automate separator oil shrinkage calculations using a Montney-specific equation of state (EOS) model.
The problem: separator oil rates are not stock-tank oil
In tight unconventionals, rates are typically measured daily at separator conditions, but those fluids rarely reach stock-tank conditions on a single well basis.
The industry workaround is usually one of these:
Option A: Ignore shrinkage
Use separator oil volumes directly.
This is common, but it overestimates oil volumes and profitability.
Option B: Apply one constant shrinkage factor
Pick a shrinkage factor for the well or field and use it forever.
The problem: shrinkage is not constant.
Shrinkage factor (SF) depends on:
Option C: Single-point shrinkage calculations
Utilize conventional process modelling software to calculate a shrinkage factor each time a new well-specific fluid sample is taken.
Point in time. Results are used until the next sample is taken (1, 6, 12, 24 months), meaning they are not always representative of current conditions.
Completed one-by-one. Extremely labor intensive and prone to keying errors - manual data entry. Extremely costly if contracted to a third-party.
Why shrinkage changes over time (and why it matters)
Separator oil shrinkage is a moving target because:
The combination can shift shrinkage factors significantly over the life of a well.
In some Montney-style near-critical systems, the spread between low and high shrinkage can be ~20% even without changing facilities, simply due to fluid evolution and operating variability.
So if you are allocating, forecasting, or reporting using stale shrinkage assumptions, your numbers drift quietly and continuously.
ARC + whitson: building an automated shrinkage workflow
ARC and whitson collaborated on a workflow to industrialize separator shrinkage calculations at scale, using a Montney-specific EOS model and automated reporting.
The goal was simple:
Take measured separator data and automatically produce defensible stock-tank volumes for every well.
The solution: Montney EOS shrinkage automation
Together, we implemented a repeatable workflow that looks like this:
Step 1: Pull field measurements from ProTrend
ARC regularly pulls well-level surface measurements from ProTrend, including:
Step 2: Split measured C7+ into C7 to C36+
(Montney-specific gamma model)
Most field measured datasets stop at C7+.
To solve this, measured C7+ is split into C7 through C36+ using a gamma distribution model (C.H. Whitson 1984).
This step is essential because shrinkage is heavily influenced by the heavy-end fraction and how it evolves with time.
Step 3: Bulk-run separator oil shrinkage using the Montney EOS
Once the separator oil composition is split to C36+, it’s run through the relevant surface process model (single or multi-stage) to compute separator oil shrinkage factor (SF).
Step 4: Automated reporting for compliance + audit trail
Finally, ARC and whitson automated reporting outputs so the workflow generates a report with relevant information and metadata. This matters because shrinkage workflows tend to become “spreadsheet folklore” inside organizations.
The value: turning shrinkage into an asset, not a headache
Once this workflow is automated, shrinkage stops being a messy spreadsheet problem and becomes a real operational advantage. ARC and whitson built a physics-based workflow that scales across thousands of wells, produces auditable results, and makes separator shrinkage something you can operationalize instead of something you argue about. And maybe the biggest win: it saves an enormous amount of time and cost, with estimated savings exceeding 2 million USD over the next decade.
In Canada’s unconventional plays, thousands of wells produce into complex gathering systems. The challenge is that many wells are only measured at separator conditions, and then quickly routed into extensive pipeline networks where fluids are commingled.
That sounds normal. But it creates a surprisingly painful engineering problem:
How do you consistently convert separator oil rates into stock-tank oil volumes on a well-by-well basis?
Stock-tank rates are often unavailable at the single-well level, yet they are essential for:
- engineering surveillance
- reporting and compliance
- production allocation
- reserves estimation
- forecasting and economics
And when shrinkage is handled poorly, the error can be huge. Our internal studies show that failing to account for time-varying shrinkage can lead to stock-tank oil rate errors exceeding 20%, especially in near-critical volatile oils and condensate systems.
That’s exactly why ARC Resources and whitson partnered to automate separator oil shrinkage calculations using a Montney-specific equation of state (EOS) model.
The problem: separator oil rates are not stock-tank oil
In tight unconventionals, rates are typically measured daily at separator conditions, but those fluids rarely reach stock-tank conditions on a single well basis.
The industry workaround is usually one of these:
Option A: Ignore shrinkage
Use separator oil volumes directly.
This is common, but it overestimates oil volumes and profitability.
Option B: Apply one constant shrinkage factor
Pick a shrinkage factor for the well or field and use it forever.
The problem: shrinkage is not constant.
Shrinkage factor (SF) depends on:
- surface process configuration (# of stages)
- separator pressure and temperature (psep, Tsep)
- wellstream composition (zi)
Option C: Single-point shrinkage calculations
Utilize conventional process modelling software to calculate a shrinkage factor each time a new well-specific fluid sample is taken.
Point in time. Results are used until the next sample is taken (1, 6, 12, 24 months), meaning they are not always representative of current conditions.
Completed one-by-one. Extremely labor intensive and prone to keying errors - manual data entry. Extremely costly if contracted to a third-party.
Why shrinkage changes over time (and why it matters)
Separator oil shrinkage is a moving target because:
- separator pressures often decline until stabilized
- separator temperatures cycle seasonally
- produced compositions evolve with depletion
- shut-ins cause transient “CGR kicks” and compositional swings
The combination can shift shrinkage factors significantly over the life of a well.
In some Montney-style near-critical systems, the spread between low and high shrinkage can be ~20% even without changing facilities, simply due to fluid evolution and operating variability.
So if you are allocating, forecasting, or reporting using stale shrinkage assumptions, your numbers drift quietly and continuously.
ARC + whitson: building an automated shrinkage workflow
ARC and whitson collaborated on a workflow to industrialize separator shrinkage calculations at scale, using a Montney-specific EOS model and automated reporting.
The goal was simple:
Take measured separator data and automatically produce defensible stock-tank volumes for every well.
The solution: Montney EOS shrinkage automation
Together, we implemented a repeatable workflow that looks like this:
Step 1: Pull field measurements from ProTrend
ARC regularly pulls well-level surface measurements from ProTrend, including:
- separator oil composition (xi)
- separator pressure
- separator temperature
Step 2: Split measured C7+ into C7 to C36+
(Montney-specific gamma model)
Most field measured datasets stop at C7+.
To solve this, measured C7+ is split into C7 through C36+ using a gamma distribution model (C.H. Whitson 1984).
This step is essential because shrinkage is heavily influenced by the heavy-end fraction and how it evolves with time.
Step 3: Bulk-run separator oil shrinkage using the Montney EOS
Once the separator oil composition is split to C36+, it’s run through the relevant surface process model (single or multi-stage) to compute separator oil shrinkage factor (SF).
Step 4: Automated reporting for compliance + audit trail
Finally, ARC and whitson automated reporting outputs so the workflow generates a report with relevant information and metadata. This matters because shrinkage workflows tend to become “spreadsheet folklore” inside organizations.
The value: turning shrinkage into an asset, not a headache
Once this workflow is automated, shrinkage stops being a messy spreadsheet problem and becomes a real operational advantage. ARC and whitson built a physics-based workflow that scales across thousands of wells, produces auditable results, and makes separator shrinkage something you can operationalize instead of something you argue about. And maybe the biggest win: it saves an enormous amount of time and cost, with estimated savings exceeding 2 million USD over the next decade.