Module 4 · Month 4 of 6

Reservoir Engineering

The reservoir engineer answers the questions that decide a project's worth: how fast will it flow, how much will we get, and what will it do over twenty years? This is where porosity and saturation become a production forecast, a reserves number, and a recovery strategy — and where shale's brutal decline gets explained.

Chapter 00·~5 min

How to read this #

Reservoir engineering is the quantitative heart of the discipline. Module 3 told you what is in the rock; this module is about how it moves and how much you can recover. It is the most equation-heavy module — work the math by hand.

We build from the fluids (how oil and gas behave under pressure) to flow (Darcy's law and the radial-flow equation that sets a well's rate) to the reservoir as a whole (drive mechanisms, material balance, well testing). Then the forecasting tools — decline curves and reserves — then the big machinery: simulation and enhanced oil recovery. We end where the money is now: the strange physics of shale reservoirs and why a lateral's production falls off a cliff.

Two questions, always

Every reservoir engineering tool answers one of two questions: how much is there (volumetrics, material balance, reserves) and how fast does it come out (Darcy, well testing, decline, simulation). Keep asking which question a method is answering and the field organizes itself.

Chapter 01·Fluids & flow

Reservoir fluids #

What you produce — and how you produce it — depends on the fluid type, which is set by composition and by where the reservoir sits on a pressure-temperature phase diagram. The same molecules behave as liquid, gas, or both depending on conditions.

The five fluid types

Type	Behavior	GOR (scf/STB)
Black oil	Heavy, low shrinkage; the classic oil reservoir	< 2,000
Volatile oil	Light oil, high shrinkage, lots of dissolved gas	2,000–3,300
Gas condensate	Gas in reservoir; drops liquid as pressure falls (retrograde)	3,300–50,000
Wet gas	Gas in reservoir; liquids only at surface	> 50,000
Dry gas	Gas everywhere; no liquids	—

The phase envelope. A reservoir's position relative to the critical point and cricondentherm sets its fluid type. Gas condensates are the tricky ones — drop below the dew point and valuable liquid drops out in the reservoir, where it may be lost.

Chapter 02·Fluids & flow

PVT: how fluids change with pressure #

Oil and gas are compressible and dissolve in each other, so a barrel in the reservoir is not a barrel at surface. PVT (pressure-volume-temperature) properties quantify the difference, and they are essential to every volume and flow calculation.

Formation volume factor (Bo) — reservoir barrels per stock-tank barrel. Always >1 because oil shrinks and loses gas coming up. This is the divisor in STOIIP.
Solution gas-oil ratio (Rs) — scf of gas dissolved per STB of oil at reservoir conditions.
Bubble point (Pb) — the pressure at which the first gas bubble comes out of solution. Above it the oil is undersaturated; below it free gas appears in the reservoir.
Gas FVF (Bg) and the z-factor — for gas, governed by the real-gas law PV = znRT.

Above the bubble point, dropping pressure expands the oil (Bo rises) while all gas stays dissolved (Rs flat). Below it, gas escapes solution — Rs falls and the now gas-depleted oil shrinks (Bo falls). This kink governs solution-gas-drive behavior.

Chapter 03·Fluids & flow

Darcy's law & radial flow #

This is the equation that sets how fast a well flows. Henry Darcy found in 1856 that flow rate through porous rock is proportional to the pressure gradient and permeability, and inversely to viscosity:

q = − (k · A / µ) · (dP/dx) // Darcy's law, linear flow

From linear to radial

A well doesn't drain a straight tube — fluid converges on it from all directions, so flow is radial. Integrating Darcy's law in cylindrical coordinates from the wellbore radius (rw) out to the drainage radius (re) gives the steady-state radial inflow equation in field units:

q = (k·h·(P_e−P_wf)) / (141.2 · B · µ · [ln(re/rw) + s]) // STB/d

Read it carefully — it is the most important equation in production. Rate scales with permeability-thickness (kh, the flow capacity) and with drawdown (Pe−Pwf, how hard you pull). It scales inversely with the log of drainage radius — so doubling the drainage area barely changes rate, a profound and useful fact. And s is the skin factor: near-wellbore damage (positive skin, from drilling) chokes the well; stimulation like a frac (negative skin) boosts it.

Flow converges radially on the well, and pressure drops fastest right at the wellbore — the cone of depression. Most of the drawdown is spent in the last few feet, which is exactly why near-wellbore skin (damage or stimulation) matters so much.

Chapter 04·Fluids & flow

Reservoir drive mechanisms #

What pushes the oil to the well? The drive mechanism is the source of energy, and it sets both the production profile and the ultimate recovery factor. Most fields run on one dominant drive, sometimes several.

Solution-gas (depletion) drive — as pressure drops below bubble point, gas comes out of solution and expands, pushing oil. Cheap but weak: recovery typically only 5–25%.
Gas-cap drive — an overlying free-gas cap expands downward as oil is produced. Better: 20–40%.
Water drive — an aquifer pushes up into the reservoir, maintaining pressure. The strongest natural drive: 35–60%+. The Middle East giants run on it.
Gravity drainage & compaction — gravity segregates fluids; rock compaction squeezes oil out. Often supplementary.

The three main natural drives, weakest to strongest. The recovery factor — what fraction of the oil in place you actually get — is set largely by which drive you have, which is why identifying it early (from material balance and pressure) is so valuable.

Chapter 05·Measuring the tank

Well testing & pressure transient analysis #

A well test is a controlled experiment on the reservoir: change the rate, watch the pressure respond, and infer properties you cannot measure any other way — average permeability, skin, reservoir pressure, and the distance to boundaries.

Drawdown and buildup

In a drawdown test you produce at constant rate and watch pressure fall; in the cleaner buildup test you shut the well in and watch pressure recover. The pressure transient propagates outward through the reservoir over time, so early data sees the near-wellbore (skin), middle data sees the formation (kh), and late data sees the boundaries. The classic Horner plot (buildup) gives kh from its slope and skin from its intercept; the modern log-log pressure-derivative plot is the diagnostic that fingerprints flow regimes and boundaries.

The pressure derivative is the interpreter's fingerprint. A flat middle section is radial flow, from which kh and skin come; the early hump is wellbore storage; a late upturn means the transient hit a boundary — a fault or the edge of the drainage area.

Chapter 06·Measuring the tank

Material balance #

Material balance is conservation of mass applied to a whole reservoir treated as a single tank: what you produced plus what's left equals what you started with, all corrected for how fluids expand as pressure drops. It is the reservoir engineer's most elegant tool — it uses production and pressure history to back out the oil in place and the drive mechanism without any geological model.

The general form (Havlena-Odeh) reads underground withdrawal (F) as the sum of expansion terms:

F = N·(E_o + m·E_g + E_f,w) + (W_e + injection) // N = oil in place

where Eo is oil/dissolved-gas expansion, Eg the gas-cap expansion, Ef,w the rock-and-water expansion, and We the aquifer influx. Plotting it the right way gives a straight line whose slope is N — and the relative size of the terms tells you the drive. For gas, the same logic collapses to the famous straight line:

P/z = (P_i/z_i) · (1 − G_p/G) // gas material balance

For a depletion gas reservoir, P/z falls linearly with cumulative production. Extrapolate the line to P/z = 0 and you read the original gas in place directly off the x-axis — one of the cleanest results in the discipline.

Chapter 07·Measuring the tank

Decline curve analysis #

The most-used forecasting method in the industry, because it needs only production history. As a well depletes, its rate declines in a characteristic way; fit that decline and extrapolate to get the estimated ultimate recovery (EUR). J.J. Arps formalized it in 1945 and the equations still run the business.

The Arps equations

One general form with a single shape parameter b spans three behaviors:

q(t) = q_i / (1 + b·D_i·t)^(1/b)

b = 0 — exponential: constant percentage decline; the conservative classic for conventional wells.
0 < b < 1 — hyperbolic: decline rate itself decreases over time; fits most real wells.
b = 1 — harmonic: the slowest decline; optimistic.

The Arps family from one initial rate. Conventional wells decline gently; shale wells (sienna) lose 60–80% of their rate in the first year, which is why their economics live and die on the first 12–18 months of cash flow.

Arps breaks in shale

Arps assumes boundary-dominated flow, which shale wells don't reach for years. Early shale decline is so steep that fitting Arps with a high b overestimates EUR badly. Modern methods — Duong, stretched-exponential (SEPD), and rate-transient analysis — were built specifically for transient linear flow in fractured shale. Using plain Arps on a shale well is a classic, expensive rookie error.

Chapter 08·Measuring the tank

Reserves & SEC rules #

Reserves are the commercially recoverable, remaining quantities — and because they back company valuations and loans, they are defined and audited with legal rigor. The distinction that trips up newcomers: resources are what's there; reserves are what you can profitably produce under today's conditions.

The categories (SPE-PRMS)

1P — Proved (P90): ≥90% probability the actual recovery exceeds this. Split into PDP (proved developed producing), PDNP, and PUD (proved undeveloped).
2P — Proved + Probable (P50): the median, best-estimate case.
3P — Proved + Probable + Possible (P10): the optimistic case.

For US public companies the SEC sets strict rules on what counts as proved — including the use of a trailing 12-month average price and a five-year limit on booking undeveloped locations. The methods to estimate reserves run from analogy (early) to volumetric (Module 1's STOIIP × recovery factor) to material balance and decline (with production history) to full simulation. Reserves estimation is where reservoir engineering meets the law and the balance sheet — covered further in the Upstream Field Manual.

Chapter 09·Models & recovery

Reservoir simulation #

When the reservoir is too complex for tank models — heterogeneous rock, many wells, multiple fluids, injection — you build a numerical simulation: divide the reservoir into a 3D grid of cells and solve the flow equations (Darcy plus mass conservation) for each cell at each timestep. It is the most detailed forecasting tool and the most data-hungry.

The reservoir is gridded into cells, each with its own rock properties; the simulator marches fluids from injector to producer. The model is trusted only after history matching — tuning it until it reproduces the field's real production — then it forecasts the future.

The honest caveat: a simulation is only as good as its inputs and its history match, and a model that matches the past can still mispredict the future (non-uniqueness). Good reservoir engineers treat simulation as one input among several — cross-checked against material balance and decline — not as an oracle.

Chapter 10·Models & recovery

Enhanced oil recovery #

Primary recovery (natural drive) leaves most of the oil behind — often two-thirds of it. Recovery comes in three stages, and the later stages are how mature fields keep producing.

Primary — natural reservoir energy (the drives of Chapter 4).
Secondary — inject water or gas to maintain pressure and sweep oil toward producers. Waterflooding is by far the most common and is the workhorse of the North Sea and Middle East.
Tertiary (EOR) — change the physics to mobilize the oil water can't move:
- Miscible gas (CO₂, hydrocarbon) — gas that mixes with oil, erasing the interfacial tension. CO₂ flooding doubles as carbon storage.
- Chemical — polymers (improve sweep), surfactants (lower interfacial tension), alkali.
- Thermal — steam injection or in-situ combustion for heavy oil, cutting its viscosity. The basis of Canadian oil sands and California heavy oil.

Waterflooding: injected water pushes oil to the producer, but it fingers through high-permeability paths and bypasses oil in tighter rock. The fraction actually contacted is the sweep efficiency — improving it is most of what EOR is about.

Chapter 11·Models & recovery · LATERAL SPINE

Shale reservoir engineering #

Shale rewrites the rules of this whole module. Matrix permeability is nanodarcy — a billion times tighter than a good sandstone — so without the induced fracture network from a multi-stage frac, nothing flows at all. The reservoir engineer's mental model shifts from "fluid flows radially to a well" to "fluid bleeds from tight matrix blocks into a fracture network and then to the lateral."

The stimulated rock volume

The producing unit is not the well — it is the stimulated rock volume (SRV), the region around the lateral that the fracs have shattered into flow paths. Oil reaches the well by two slow steps: from the ultra-tight matrix into the nearest fracture (the rate-limiting step), then up the fracture network to the wellbore. This gives shale its signature flow regime — long-lived transient linear flow into the fractures — and its brutal decline, because once the rock immediately around the fractures is drained, the matrix can't refill it fast enough.

Production comes from the SRV: the tight matrix bleeds into the fracture planes, which carry fluid to the lateral. Recovery factors are low (5–10%) and decline is steep because the matrix can't keep the near-fracture rock supplied. This is why spacing and frac design (Module 5) are the reservoir engineering of shale.

This reshapes practice. Reserves come from rate-transient analysis and modern decline (not Arps); well spacing is optimized against parent-child interference (a later "child" well draining a depleted "parent's" rock, or a frac hit damaging it); and the asset's value is dominated by the first 18 months. The reservoir engineer, the completion engineer, and the geosteerer now work as one team — which is exactly why this curriculum threads the lateral through every module.

Connects

The fractures drawn in sienna here are designed in Module 5. The SRV depends on the brittleness graded in Module 3 and the stress state from Module 1. Shale reservoir engineering is the integration point of the whole study.

Chapter 12·Reference

Equations & glossary #

The equations of this module

q = (k·h·ΔP) / (141.2·B·µ·[ln(re/rw) + s]) // radial inflow, STB/d

J = q / (P̄_r − P_wf) // productivity index

F = N·(E_o + mE_g + E_f,w) + W_e // material balance

P/z = (P_i/z_i)(1 − G_p/G) // gas material balance

q(t) = q_i / (1 + b D_i t)^1/b // Arps decline

Glossary

Term	Meaning
Bo / Rs	Oil formation volume factor / solution gas-oil ratio (PVT).
Bubble point	Pressure where gas first leaves solution in oil.
Decline (Arps)	Empirical rate-vs-time decline; b sets exponential/hyperbolic/harmonic.
Drive mechanism	Source of energy moving oil to the well; sets recovery factor.
EUR	Estimated ultimate recovery from a well or field.
EOR	Enhanced oil recovery — tertiary methods (gas, chemical, thermal).
kh	Permeability-thickness — a well's flow capacity, from well tests.
Material balance	Mass conservation on the whole reservoir to find OOIP and drive.
Reserves (1P/2P/3P)	Commercially recoverable volumes at P90/P50/P10 confidence.
Skin (s)	Near-wellbore pressure-drop factor; + damage, − stimulation.
SRV	Stimulated rock volume — the fractured region that produces in shale.
Waterflood	Secondary recovery by water injection to sweep oil.

End of Module 4

You can now classify a reservoir fluid, use Darcy's radial equation, identify a drive mechanism, run material balance and decline, understand reserves categories, and explain why shale produces and declines the way it does. Next: Module 5, Completions & Production — building the well to produce, and the hydraulic fracture that makes the lateral pay.