Decline Model Comparison

Compare multiple decline curve models for the same well. Different models may be more appropriate depending on reservoir type and flow regime. Arps is standard for conventional, Duong for unconventional transient, and PLE for tight/shale.

Models Compared:

• Arps: Classic hyperbolic decline (conventional reservoirs)
• Duong: Power-law decline (unconventional, transient linear flow)
• PLE: Power Law Exponential (tight/shale, early time)

Units: rates are bbl/month and every decline constant is 1/month, so the cumulative block is a true volume (shown as Mstb). All three model families are unit-agnostic — rate and time must share the same time base.

All three parameter sets describe one well. That is what makes the comparison mean anything. The Arps curve is the reference; the Duong and PLE parameters are fitted to it rather than chosen independently — both anchored at month 12, with Duong matched through month 24 and PLE through month 120. Three unrelated parameter sets would produce a spread that says nothing about the models, only about the parameters.

Read the spread column with that in mind. It is tightest (~4%) at month 12 where the fits are anchored, sits in the mid-teens through month 60, and widens to about a third by month 120 — models constrained by the same history separating once the history runs out. The wider figure at month 1 is expected: PLE's qi is a fitted parameter, not the observed initial rate, so its curve starts below the other two by construction.

If you re-fit one model, re-fit the others to the same history or this stops being a comparison.