Wayward·House

Essays System Signals Library Topics About 🐾 Ask Pepper

Subscribe

Crude Oil Pipelines

Darcy-Weisbach hydraulics, capacity utilization, and the netback price model

The Enbridge Mainline carries more crude oil per day than Colombia produces. This article derives the hydraulic equations governing pipeline capacity, applies them to real Alberta infrastructure, and builds the netback arithmetic that determines what a barrel of bitumen actually earns after it pays to travel from Hardisty to market.

Part of the Alberta Pipeline Geography cluster. This is P1 of five articles. P2 — NGL and Condensate · P3 — Natural Gas Transmission · P4 — Refined Products Distribution · P5 — The Integrated Network

---

Before You Start

You should know Basic algebra. What pressure means physically — force per unit area. That fluids flow from high pressure to low pressure.

You will learn How to calculate how much oil a pipeline can carry (the Darcy-Weisbach equation and what generates friction losses), what determines whether you add a pump station or widen the pipe, and how to build the netback price model that converts a spot price in Chicago into what a bitumen producer in Fort McMurray actually receives.

Why this matters Alberta’s pipeline capacity debate is conducted almost entirely in political language. The hydraulic equations in this article set a physical ceiling on what any pipe of a given diameter can move. The netback model explains why the same barrel of crude earns different amounts depending on which corridor it travels — and why the Trans Mountain Expansion changed Alberta’s pricing geography in a way that volume-only arguments about pipeline capacity rarely capture.

If this gets hard, focus on… The key insight is that friction in a pipe scales with the square of velocity. Push twice as much oil through the same pipe and you don’t double the friction — you quadruple it. Everything else in this article follows from that one fact.

---

Somewhere northeast of Edmonton, a 36-inch steel pipe is moving oil. Not metaphorically — right now, at this moment, at roughly 2 metres per second, approximately 800,000 barrels of diluted bitumen and synthetic crude are transiting Enbridge Line 67 every day toward refineries in the U.S. Great Lakes (Enbridge Pipelines Inc. 2024b). Line 67 is one of five parallel lines in the Enbridge Mainline corridor. Together they carry approximately 3 million barrels per day — more crude oil, every single day, than the entire national production of Colombia (Canada Energy Regulator 2024b).

The physics governing that movement were worked out by Henri Darcy in the 1850s, surveying water supply systems for the city of Dijon, and extended by Julius Weisbach into the equation that now bears their names. The Darcy-Weisbach equation appears in every pipeline engineering textbook. This article derives it from first principles, applies it to real Alberta infrastructure with real numbers, and then uses the same arithmetic to build the netback price model — the calculation that determines what a barrel of Alberta crude actually earns after paying its way from Hardisty to market.

---

1. The Question

How much crude oil can a pipeline move, what does it cost to move it, and what does the producer receive after transport?

These are three separate but linked questions. The first is a fluid mechanics problem. The second is a tariff and operating cost problem. The third — the netback price — connects both to commodity markets and sets the economics of Alberta oil production.

A pipeline moves fluid by maintaining a pressure gradient along its length: high pressure at the inlet (pump station), lower pressure at the outlet (delivery terminal), with friction in the pipe consuming that pressure as the fluid travels. The fundamental constraint is that pump stations can only maintain so much pressure before the pipe itself — rated to a maximum operating pressure — becomes the limit. Within that constraint, the question becomes: how do pipe geometry (diameter, length, roughness) and fluid properties (density, viscosity) determine flow rate?

---

2. The Conceptual Model

A pipeline is a resistance problem. Fluid flows because there is a pressure difference between inlet and outlet. That pressure difference is consumed by friction — the resistance of the pipe wall to the moving fluid. Add more pressure (more pump stations), and you can push more fluid. Increase the pipe diameter, and you reduce friction for a given flow rate. Change the fluid — from light synthetic crude to viscous diluted bitumen — and friction changes too.

Three properties of the system govern everything:

The pipe: inner diameter D (metres), length L (metres), and wall roughness \varepsilon (metres — typically 0.046 mm for commercial steel).

The fluid: density \rho (kg/m³) and dynamic viscosity \mu (Pa·s). Both vary by crude grade and, critically, by temperature.

The flow: mean velocity v (m/s), which determines both throughput and friction losses.

The relationship between these quantities runs through two intermediate steps: first, characterising the regime of the flow — laminar or turbulent — using the Reynolds number; then, using that regime to find the friction factor that appears in the Darcy-Weisbach equation.

---

3. Building the Mathematical Model

3.1 Fluid Properties: Density and Viscosity

Density is straightforward — kilograms per cubic metre:

\rho = \frac{m}{V} \quad [\text{kg/m}^3]

Pipeline specifications constrain density directly. The Enbridge Mainline accepts crude grades with density not exceeding approximately 940 kg/m³ (Enbridge Pipelines Inc. 2024a). Diluted bitumen (dilbit) is blended to stay within specification — typically 880–920 kg/m³ at pipeline temperature, depending on the diluent ratio and the heaviness of the bitumen stream.

Viscosity is less intuitive. Dynamic viscosity \mu measures how strongly a fluid resists being deformed by shear — colloquially, how “thick” it is. Water at 20°C has \mu \approx 0.001 Pa·s. Raw bitumen at ambient Alberta temperatures is effectively a solid — thousands of Pa·s. Diluted bitumen at pipeline operating temperature is blended to achieve viscosity below 350 centistokes, the Enbridge maximum specification (Enbridge Pipelines Inc. 2024a); for dilbit at 900 kg/m³ this corresponds to roughly \mu \approx 0.012–0.015 Pa·s.

Viscosity is strongly temperature-dependent for heavy crudes. A 10°C drop in pipeline temperature can double the viscosity of dilbit — which is why cold winter soil temperatures affect pipeline operations and why pipelines through the oil sands region are often operated at elevated temperatures.

3.2 Flow Regime: The Reynolds Number

Before calculating friction, we need to know what kind of flow we have. Fluids move in two fundamentally different regimes:

• Laminar: Smooth, layered. Fluid molecules travel in parallel paths. Characteristic of slow flow and very viscous fluids.
• Turbulent: Chaotic, mixing. Fluid moves in swirling eddies. Characteristic of fast flow and lower viscosities.

The Reynolds number Re determines which regime applies:

Re = \frac{\rho v D}{\mu}

where \rho is density (kg/m³), v is mean velocity (m/s), D is inner diameter (m), and \mu is dynamic viscosity (Pa·s). Reynolds number is dimensionless.

Regime boundaries:

• Re < 2{,}300: Laminar
• 2{,}300 < Re < 4{,}000: Transitional (unstable)
• Re > 4{,}000: Turbulent

Commercial oil pipelines operate well into the turbulent regime — Reynolds numbers in the range of 50,000 to 500,000 are typical. Laminar flow in a large crude oil pipeline would require either extremely slow velocities or extremely high viscosities — neither of which is economically useful.

Physical meaning: The Reynolds number is a ratio of inertial forces (the momentum of the moving fluid, \rho v D) to viscous forces (the fluid’s resistance to deformation, \mu). When inertia dominates, flow is turbulent. When viscosity dominates, flow is laminar. For crude oil in a metre-wide pipe moving at two metres per second, inertia wins by a wide margin.

3.3 Friction Factor

The friction factor f (the Darcy or Moody friction factor) quantifies how much resistance the pipe wall exerts on the flowing fluid, normalised by the flow’s kinetic energy.

In laminar flow, the solution is exact:

f = \frac{64}{Re} \quad \text{(laminar, } Re < 2{,}300\text{)}

In turbulent flow, friction depends on both Re and wall roughness \varepsilon/D. The classical result is the implicit Colebrook-White equation (Colebrook 1939):

\frac{1}{\sqrt{f}} = -2 \log_{10}\!\left(\frac{\varepsilon}{3.7D} + \frac{2.51}{Re\sqrt{f}}\right)

This cannot be solved algebraically for f — the right-hand side depends on f itself. Pipeline engineers iterate numerically or use explicit approximations.

The most useful explicit form for computational work is the Churchill (1977) equation (Churchill 1977), which is valid across all flow regimes — laminar, transitional, and turbulent — without switching between formulas:

f = 8\left[\left(\frac{8}{Re}\right)^{12} + \left(A + B\right)^{-3/2}\right]^{1/12}

where:

A = \left[-2.457 \ln\!\left(\left(\frac{7}{Re}\right)^{0.9} + \frac{0.27\varepsilon}{D}\right)\right]^{16}

B = \left(\frac{37530}{Re}\right)^{16}

The formula looks intimidating, but the structure is elegant: in laminar flow, the (8/Re)^{12} term dominates and recovers f = 64/Re exactly. In turbulent flow, the A term dominates and reproduces the Colebrook-White result. The Churchill equation is used in the code below.

• Key relationships:

• Higher Re → lower f in turbulent flow (friction factor decreases as flow becomes more turbulent)

• Rougher pipe (\varepsilon/D larger) → higher f in turbulent flow

• In laminar flow, wall roughness is completely irrelevant — only Re matters

The roughness value \varepsilon = 0.046 mm for commercial steel pipe is the standard engineering figure tabulated by Moody (1944) and used in all subsequent pipeline design practice.

3.4 Pressure Drop: The Darcy-Weisbach Equation

With the friction factor established, pressure drop along a pipe section is:

\Delta P = f \cdot \frac{L}{D} \cdot \frac{\rho v^2}{2}

where:

• \Delta P — pressure drop [Pa]
• f — Darcy friction factor [dimensionless]
• L — pipe length [m]
• D — pipe inner diameter [m]
• \rho — fluid density [kg/m³]
• v — mean flow velocity [m/s]

The \rho v^2/2 term is the dynamic pressure — kinetic energy per unit volume of the moving fluid. The f \cdot L/D term is a dimensionless multiplier: the number of “pipe diameters of kinetic energy” lost to friction over length L. Together they give pressure drop in Pascals.

The velocity-squared relationship is the key result. Double the flow rate, double the velocity — but the dynamic pressure (\rho v^2/2) quadruples, and so does the pressure drop per unit length. Pushing twice as much oil through a pipe requires four times as much pump pressure, not twice as much. This is the physical reason why pipeline capacity is not simply doubled by running two pumps: the friction penalty rises faster than the throughput gain.

3.5 Volumetric Flow Rate and Barrels per Day

Velocity and volumetric flow rate are related by the pipe’s cross-sectional area:

Q = v \cdot A = v \cdot \frac{\pi D^2}{4} \quad [\text{m}^3/\text{s}]

Converting to the industry unit of barrels per day (1 barrel = 0.158987 m³):

Q_{\text{bbl/day}} = Q_{\text{m}^3/\text{s}} \times 86{,}400 \;\text{s/day} \times 6.2898 \;\text{bbl/m}^3

Substituting the velocity-to-flow relationship (v = 4Q/\pi D^2) into the Darcy-Weisbach equation gives the flow-rate form:

\boxed{\Delta P = \frac{8 f \rho L Q^2}{\pi^2 D^5}}

This form makes the diameter dependence explicit: pressure drop scales as D^{-5}. Increase pipe diameter by 10% and pressure drop for the same flow rate falls by (1.1)^5 \approx 1.61 — a 38% reduction. This is the dominant design variable for long-distance pipelines, which is why the Trans Mountain Expansion achieved a near-tripling of corridor capacity by adding a single new pipe to an existing right-of-way.

3.6 Pump Stations: Restoring Pressure Along the Route

No single pump station can sustain the pressure gradient over hundreds of kilometres. Stations are placed at intervals to re-pressurise the flow as friction consumes the available head.

If maximum allowable operating pressure is P_{\max} and the minimum inlet pressure at each station (to protect pump seals and avoid cavitation) is P_{\min}, each station can supply a pressure boost of \Delta P_{\text{station}} = P_{\max} - P_{\min} before the next station is needed. Setting the Darcy-Weisbach pressure drop equal to this budget gives the maximum pump station spacing:

L_{\text{station}} = \frac{\Delta P_{\text{station}} \cdot \pi^2 D^5}{8 f \rho Q^2}

Enbridge Mainline pump stations are spaced roughly 80–200 km along the corridor depending on terrain, pipe diameter, and throughput (Canada Energy Regulator 2024b). The Mainline runs approximately 50 stations from Edmonton to Superior, Wisconsin.

---

4. Worked Example by Hand

Scenario: Calculate the pressure drop over a 100 km segment of Enbridge Line 67 (the Alberta Clipper) operating at 750,000 bbl/day. Determine whether a pump station is needed within this segment.

Given parameters:

Parameter	Value
Pipe inner diameter D	0.914 m (36 inches)
Wall roughness \varepsilon	4.6 \times 10^{-5} m (commercial steel)
Fluid density \rho	900 kg/m³ (dilbit blend)
Dynamic viscosity \mu	0.012 Pa·s (~12 cP at 15°C)
Throughput Q	750,000 bbl/day
Segment length L	100,000 m
Max allowable operating pressure P_{\max}	8.0 MPa
Minimum delivery pressure P_{\min}	0.5 MPa

Step 1 — Convert flow rate to SI units

Q = 750{,}000 \;\frac{\text{bbl}}{\text{day}} \times \frac{0.158987 \;\text{m}^3}{\text{bbl}} \times \frac{1 \;\text{day}}{86{,}400 \;\text{s}} = 1.380 \;\text{m}^3/\text{s}

Step 2 — Mean velocity

v = \frac{4Q}{\pi D^2} = \frac{4 \times 1.380}{\pi \times (0.914)^2} = \frac{5.521}{2.624} = 2.104 \;\text{m/s}

Step 3 — Reynolds number

Re = \frac{\rho v D}{\mu} = \frac{900 \times 2.104 \times 0.914}{0.012} = \frac{1{,}731}{0.012} = 144{,}200

Well into turbulent regime (Re \gg 4{,}000).

Step 4 — Relative roughness

\frac{\varepsilon}{D} = \frac{4.6 \times 10^{-5}}{0.914} = 5.03 \times 10^{-5}

Step 5 — Friction factor (Swamee-Jain explicit approximation)

For a quick hand calculation, the Swamee-Jain equation (Swamee and Jain 1976) is explicit and accurate to within 3% of Colebrook-White for 10^{-6} < \varepsilon/D < 10^{-2} and 5{,}000 < Re < 10^8:

f = \frac{0.25}{\left[\log_{10}\!\left(\dfrac{\varepsilon}{3.7D} + \dfrac{5.74}{Re^{0.9}}\right)\right]^2}

Computing the argument:

\frac{\varepsilon}{3.7D} = \frac{5.03 \times 10^{-5}}{3.7} = 1.36 \times 10^{-5}

For Re^{0.9}: take \ln(144{,}200) = 11.878, multiply by 0.9 to get 10.690, exponentiate: e^{10.690} \approx 43{,}900.

\frac{5.74}{43{,}900} = 1.31 \times 10^{-4}

\log_{10}(1.36 \times 10^{-5} + 1.31 \times 10^{-4}) = \log_{10}(1.45 \times 10^{-4}) = -3.840

f = \frac{0.25}{(-3.840)^2} = \frac{0.25}{14.75} = 0.01695 \approx 0.0170

Step 6 — Pressure drop (Darcy-Weisbach)

\Delta P = f \cdot \frac{L}{D} \cdot \frac{\rho v^2}{2} = 0.0170 \times \frac{100{,}000}{0.914} \times \frac{900 \times (2.104)^2}{2}

= 0.0170 \times 109{,}409 \times \frac{900 \times 4.427}{2} = 0.0170 \times 109{,}409 \times 1{,}992

= 0.0170 \times 217{,}967{,}000 = 3{,}705{,}000 \;\text{Pa} \approx \mathbf{3.71 \;\text{MPa}}

Step 7 — Assessment

The 100 km segment consumes 3.71 MPa of pressure. The station budget is P_{\max} - P_{\min} = 8.0 - 0.5 = 7.5 MPa. The maximum reach from a single station is:

L_{\max} = 100 \;\text{km} \times \frac{7.5}{3.71} \approx \mathbf{202 \;\text{km}}

A single pump station can push this throughput approximately 200 km before the next station is required — consistent with Enbridge’s actual spacing of 80–200 km along the Mainline corridor.

---

5. Computational Implementation

import numpy as np
import matplotlib.pyplot as plt

# ── Parameters ─────────────────────────────────────────────────────────────
rho       = 900.0          # density [kg/m³], dilbit
mu        = 0.012          # dynamic viscosity [Pa·s]
eps       = 4.6e-5         # wall roughness [m], commercial steel
D         = 0.914          # inner diameter [m], 36 inches — Enbridge Line 67
L         = 100_000.0      # segment length [m]
P_max     = 8.0e6          # max allowable operating pressure [Pa]
P_min     = 0.5e6          # minimum delivery pressure [Pa]

BBL_PER_M3   = 6.2898
SECS_PER_DAY = 86_400.0

# ── Functions ───────────────────────────────────────────────────────────────

def to_si(Q_bbl):
    """Barrels/day → m³/s."""
    return Q_bbl / BBL_PER_M3 / SECS_PER_DAY

def mean_velocity(Q_si, D):
    return Q_si / (np.pi * D**2 / 4)

def reynolds(rho, v, D, mu):
    return rho * v * D / mu

def churchill_f(Re, eps, D):
    """
    Churchill (1977) Darcy friction factor.
    Valid for all flow regimes: laminar, transitional, turbulent.
    """
    eps_D = eps / D
    A = (-2.457 * np.log((7.0 / Re)**0.9 + 0.27 * eps_D))**16
    B = (37530.0 / Re)**16
    return 8.0 * ((8.0 / Re)**12 + (A + B)**(-1.5))**(1.0 / 12.0)

def darcy_weisbach(f, L, D, rho, v):
    """Pressure drop [Pa]."""
    return f * (L / D) * (rho * v**2 / 2.0)

# ── Sweep over throughput range ─────────────────────────────────────────────
Q_bbl = np.linspace(100_000, 1_050_000, 500)
Q_si  = to_si(Q_bbl)
v     = mean_velocity(Q_si, D)
Re    = reynolds(rho, v, D, mu)
f     = churchill_f(Re, eps, D)
dP    = darcy_weisbach(f, L, D, rho, v)

# Maximum station spacing given MAOP
avail_head    = P_max - P_min                       # [Pa]
L_max_km      = (L * avail_head / dP) / 1000.0     # [km]

# Design point
idx = np.argmin(np.abs(Q_bbl - 750_000))

# ── Figure 1 ────────────────────────────────────────────────────────────────
fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(9, 8), sharex=True)

ax1.plot(Q_bbl / 1e3, dP / 1e6, color='#c0392b', linewidth=2.0)
ax1.axhline(P_max / 1e6, color='#95a5a6', linestyle='--', linewidth=1.2,
            label=f'MAOP ({P_max/1e6:.0f} MPa)')
ax1.axvline(Q_bbl[idx] / 1e3, color='#2980b9', linestyle=':', linewidth=1.5,
            label=f'Design point — {Q_bbl[idx]/1e3:.0f}k bbl/day '
                  f'(ΔP = {dP[idx]/1e6:.2f} MPa)')
ax1.set_ylabel('Pressure drop over 100 km [MPa]', fontsize=11)
ax1.set_ylim(0, 14)
ax1.legend(fontsize=9)
ax1.set_title('Enbridge Line 67 — Hydraulic Performance\n'
              '36-inch pipe, dilbit, ρ = 900 kg/m³, μ = 12 cP', fontsize=12, pad=10)
ax1.grid(alpha=0.3)

ax2.plot(Q_bbl / 1e3, L_max_km, color='#27ae60', linewidth=2.0)
ax2.axvline(Q_bbl[idx] / 1e3, color='#2980b9', linestyle=':', linewidth=1.5,
            label=f'Design point — max spacing ≈ {L_max_km[idx]:.0f} km')
ax2.axhspan(0, 50,  alpha=0.06, color='#e74c3c', label='Impractical — stations too close')
ax2.axhspan(200, ax2.get_ylim()[1] if ax2.get_ylim()[1] > 200 else 600,
            alpha=0.04, color='#2ecc71')
ax2.set_xlabel('Throughput [thousand bbl/day]', fontsize=11)
ax2.set_ylabel('Max pump station spacing [km]', fontsize=11)
ax2.set_ylim(0, 600)
ax2.legend(fontsize=9)
ax2.grid(alpha=0.3)

plt.tight_layout()
plt.show()

Figure 1: Top: Pressure drop over 100 km as a function of throughput for a 36-inch dilbit line. The velocity-squared relationship is visible in the curve’s upward acceleration. Bottom: Maximum pump station spacing — at 750,000 bbl/day a single station covers ~200 km.

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))

# ── Moody diagram ────────────────────────────────────────────────────────────
Re_range  = np.logspace(2.3, 6.2, 600)
f_ch      = churchill_f(Re_range, eps, D)

# Fully rough asymptote: f = [2 log10(3.7 D/ε)]^{-2}
f_rough   = (2.0 * np.log10(3.7 * D / eps))**(-2) * np.ones_like(Re_range)

ax1.loglog(Re_range, f_ch,    color='#c0392b', linewidth=2.0,
           label=f'ε/D = {eps/D:.1e}  (Line 67, steel)')
ax1.loglog(Re_range, f_rough, color='#7f8c8d', linewidth=1.0, linestyle='--',
           label='Fully-rough asymptote')

Re_design = Re[idx]
f_design  = f[idx]
ax1.plot(Re_design, f_design, 'o', color='#2980b9', markersize=9, zorder=5,
         label=f'Design point  Re = {Re_design:,.0f},  f = {f_design:.4f}')

ax1.axvline(2300, color='#bdc3c7', linewidth=0.8, linestyle=':')
ax1.text(2300, 0.085, 'Laminar / turbulent', fontsize=7, color='#7f8c8d',
         ha='center', va='bottom', rotation=90)

ax1.set_xlabel('Reynolds number', fontsize=10)
ax1.set_ylabel('Darcy friction factor  f', fontsize=10)
ax1.set_title('Moody Diagram — Line 67 Operating Region', fontsize=11)
ax1.legend(fontsize=8)
ax1.grid(alpha=0.3, which='both')
ax1.set_xlim(2e2, 1.6e6)
ax1.set_ylim(0.008, 0.10)

# ── Diameter comparison ───────────────────────────────────────────────────────
D_range  = np.linspace(0.40, 1.20, 300)   # 16 to 48 inches
Q_fixed  = to_si(750_000)
dP_D = []
for d in D_range:
    v_  = mean_velocity(Q_fixed, d)
    Re_ = reynolds(rho, v_, d, mu)
    f_  = churchill_f(Re_, eps, d)
    dP_D.append(darcy_weisbach(f_, L, d, rho, v_))
dP_D = np.array(dP_D)

ax2.plot(D_range * 39.37, dP_D / 1e6, color='#8e44ad', linewidth=2.0)
ax2.axvline(D * 39.37, color='#2980b9', linestyle=':', linewidth=1.5,
            label=f'Line 67  ({D*39.37:.0f} in,  ΔP = {dP[idx]/1e6:.2f} MPa)')
ax2.axhline(P_max / 1e6, color='#95a5a6', linestyle='--', linewidth=1.0,
            label=f'MAOP  ({P_max/1e6:.0f} MPa per 100 km)')

ax2.set_xlabel('Pipe inner diameter [inches]', fontsize=10)
ax2.set_ylabel('Pressure drop over 100 km [MPa]', fontsize=10)
ax2.set_title('Why Diameter Matters:  ΔP ∝ D⁻⁵\n(750,000 bbl/day, dilbit)', fontsize=11)
ax2.legend(fontsize=8)
ax2.grid(alpha=0.3)
ax2.set_ylim(0, 35)

plt.tight_layout()
plt.show()

Figure 2: Left: Moody diagram for the Line 67 operating region. The friction factor falls with rising Reynolds number until pipe roughness, not turbulence intensity, limits it. Right: Pressure drop at 750,000 bbl/day as a function of pipe diameter. The D⁻⁵ relationship means a 20% wider pipe cuts friction losses in half.

---

6. The Netback Price Model

The hydraulic model tells us what is physically possible. The netback model tells us what it is worth.

Netback price is what a producer receives per barrel after all transportation, diluent, and quality costs are deducted. It is the number that ultimately determines whether a project is economical — not the spot price at Hardisty, and not the tariff alone, but their combination with the diluent cost that makes bitumen shippable.

For a barrel of diluted bitumen shipped from Hardisty to Chicago on the Enbridge Mainline, the netback per barrel of dilbit is simply:

\text{Netback}_{\text{dilbit}} = P_{\text{WCS}} - T_{\text{pipeline}} - T_{\text{terminal}}

where P_{\text{WCS}} is the Western Canadian Select spot price at Hardisty and T represents tariff and handling costs.

But the product shipped is diluted bitumen. The producer blended condensate (C5+) into the bitumen at diluent ratio d \approx 0.30 — approximately 30 barrels of condensate per 100 barrels of dilbit. To find the netback per barrel of bitumen produced, account for the condensate cost and the dilution arithmetic:

\boxed{\text{Netback}_{\text{bitumen}} = \frac{\left(P_{\text{WCS}} - T\right) - P_{C5+} \cdot d}{1 - d}}

where P_{C5+} is the condensate price paid by the producer and T = T_{\text{pipeline}} + T_{\text{terminal}} is the total transport cost per barrel of dilbit shipped. The (1 - d) denominator converts from a per-dilbit-barrel basis to a per-bitumen-barrel basis.

Worked example — two corridors (Enbridge Pipelines Inc. 2024a; Trans Mountain Pipeline ULC 2024; Canada Energy Regulator 2024a):

Parameter	Enbridge (Chicago)	Trans Mountain (Westridge)
Reference price at destination	USD 57/bbl WCS	USD 63/bbl (Brent-linked heavy)
Pipeline tariff	USD 6.50/bbl	USD 9.25/bbl
Terminal/marine fee	USD 0.50/bbl	USD 1.50/bbl
Total transport T	USD 7.00/bbl	USD 10.75/bbl
Condensate price P_{C5+}	USD 72/bbl	USD 72/bbl
Diluent ratio d	0.30	0.30

Enbridge netback: \text{Netback} = \frac{(57 - 7.00) - 72 \times 0.30}{1 - 0.30} = \frac{50.00 - 21.60}{0.70} = \frac{28.40}{0.70} \approx \textbf{USD 40.57/bbl of bitumen}

Trans Mountain netback: \text{Netback} = \frac{(63 - 10.75) - 72 \times 0.30}{0.70} = \frac{52.25 - 21.60}{0.70} = \frac{30.65}{0.70} \approx \textbf{USD 43.79/bbl of bitumen}

The Trans Mountain corridor costs $3.75/bbl more to use. But it accesses a reference price $6.00/bbl higher, netting the bitumen producer $3.22/bbl more than the Enbridge corridor in this scenario. This is the market access argument in arithmetic form — not that Trans Mountain is cheap to use, but that the destination price differential exceeds the transport cost differential.

The sign can reverse. When Brent-WTI spreads compress, or when Trans Mountain capacity is tight and tolls rise in the next regulatory period, or when condensate prices spike, the corridor advantage shifts. The netback equation makes those sensitivities explicit.

import matplotlib.ticker as mticker

d     = 0.30     # diluent ratio
P_C5  = 72.0    # condensate price [USD/bbl]

WCS_range    = np.linspace(35, 90, 300)
tariff_range = np.linspace(3, 16, 300)
WCS_grid, T_grid = np.meshgrid(WCS_range, tariff_range)

netback = ((WCS_grid - T_grid) - P_C5 * d) / (1.0 - d)

fig, ax = plt.subplots(figsize=(9, 6.5))

cf  = ax.contourf(WCS_grid, T_grid, netback, levels=24, cmap='RdYlGn',
                  vmin=-20, vmax=70)
cs0 = ax.contour(WCS_grid, T_grid, netback, levels=[0],
                 colors='#2c3e50', linewidths=1.8)
cs  = ax.contour(WCS_grid, T_grid, netback,
                 levels=[-10, 10, 20, 30, 40, 50, 60],
                 colors='black', linewidths=0.7, alpha=0.5)
ax.clabel(cs,  fmt='$%d', fontsize=8)
ax.clabel(cs0, fmt='Zero netback', fontsize=8.5, inline=True)

cbar = plt.colorbar(cf, ax=ax, pad=0.02)
cbar.set_label('Bitumen netback [USD/bbl]', fontsize=10)

# Operating points
ax.plot(57, 7.00,  'o', color='#2980b9', markersize=11, zorder=6,
        label='Enbridge Mainline — Chicago  ($40.57/bbl)')
ax.plot(63, 10.75, 's', color='#e67e22', markersize=11, zorder=6,
        label='Trans Mountain — Westridge   ($43.79/bbl)')

# Annotate
ax.annotate('Enbridge', xy=(57, 7.00), xytext=(50, 4.5),
            arrowprops=dict(arrowstyle='->', color='#2980b9', lw=1.2),
            fontsize=9, color='#2980b9')
ax.annotate('Trans Mountain', xy=(63, 10.75), xytext=(67, 8.5),
            arrowprops=dict(arrowstyle='->', color='#e67e22', lw=1.2),
            fontsize=9, color='#e67e22')

ax.set_xlabel('WCS spot price at Hardisty [USD/bbl]', fontsize=11)
ax.set_ylabel('Total pipeline + terminal tariff [USD/bbl of dilbit]', fontsize=11)
ax.set_title(f'Bitumen Netback Price Surface\n'
             f'Condensate = USD {P_C5}/bbl,  diluent ratio = {d}',
             fontsize=12)
ax.legend(fontsize=9, loc='upper left')
ax.grid(alpha=0.15)

plt.tight_layout()
plt.show()

Figure 3: Bitumen netback as a function of WCS spot price and pipeline tariff, at condensate (C5+) = USD 72/bbl and diluent ratio d = 0.30. The Enbridge Mainline and Trans Mountain design points are plotted. The zero-netback contour is the floor below which producers will not accept pipeline nominations.

The green region is commercially viable; the red region is not. The zero-netback contour defines the floor below which no rational producer will accept a pipeline nomination — the point at which transport and diluent costs exactly consume the WCS price. When WCS collapsed toward USD 20/bbl in April 2020 — COVID demand destruction compressing both price and differentials — producers with committed pipeline capacity faced exactly this arithmetic: negative netbacks, paying to move crude that earned less than it cost to ship.

---

7. Interpretation

The Mainline’s corridor geometry

The five parallel lines of the Enbridge Mainline — ranging from 30 to 36 inches in diameter — represent approximately 3 million bbl/day of capacity accumulated over seven decades of incremental pipe-laying (Canada Energy Regulator 2024b). Each additional line added capacity through the same corridor, sharing right-of-way and pump stations, which made the second, third, and fourth lines progressively cheaper in per-barrel terms than the first.

The hydraulic equations explain the toll structure directly. Enbridge’s heavy crude tariff is lower per barrel than Trans Mountain’s partly because the eastern corridor uses large-diameter pipe (high D, low \Delta P per unit flow) and the original capital has been largely amortised. Trans Mountain’s tariff must recover approximately CAD 34 billion in expansion construction costs over a regulatory asset life (Trans Mountain Corporation 2024). That arithmetic flows directly from the pressure drop equation back into the tariff structure.

Capacity utilization and the fixed-cost trap

Pipeline infrastructure is most economical at high utilization. At ~78% of nameplate capacity (Alberta’s approximate position across all export corridors), the system is running within a normal operating range (Canada Energy Regulator 2025). But the cost structure is largely fixed: pump stations run whether the pipe is at 60% or 95%. Tariffs are set to recover costs at a projected throughput level in each regulatory period; if throughput falls short, the tariff rises in the next period. This is the mechanism that protects pipeline companies when production declines — but it also means that producers face rising tolls precisely when commodity markets are weakest.

What Trans Mountain actually changed

Before the Trans Mountain Expansion entered service in 2024, approximately 97% of Alberta’s crude exports moved eastward or southward, all pricing against the WTI benchmark at Cushing, Oklahoma (Canada Energy Regulator 2025). TMX added a meaningful Pacific corridor, allowing a fraction of production to price against Brent and Dubai — benchmarks that have a different structure and can trade at different spreads relative to Alberta heavy crude. The netback surface in Figure 3 shows why this matters: access to a second price benchmark is structurally valuable even when the netback advantage is not always positive. It is the equivalent of a farmer having two grain elevators rather than one.

---

8. What Could Go Wrong

Viscosity temperature dependence

The worked example assumed \mu = 0.012 Pa·s (12 cP) — appropriate for dilbit at approximately 15°C. But dilbit viscosity is highly temperature-sensitive: a 10°C drop roughly doubles \mu, increases \Delta P for the same flow, and potentially forces a throughput reduction to stay within MAOP. In cold Alberta winters, soil temperatures at shallow pipeline depths can fall well below 0°C in sections with thin cover. This is why Enbridge operates some Mainline segments at elevated temperatures and why very cold winters occasionally constrain dilbit nominations in ways that the summer-condition capacity numbers do not capture.

Drag-reducing agents

Real pipeline operations routinely inject drag-reducing agents (DRAs) — long-chain polymer additives — at concentrations of a few parts per million. DRAs interfere with turbulent eddies and reduce the effective friction factor by 20–60% without any change to pipe geometry or pump pressure rating. This allows throughput to be increased beyond what the Moody chart friction factor would suggest.

The Darcy-Weisbach equation as derived here treats f as fully determined by Re and \varepsilon/D. With DRA injection, f is reduced below its Moody value by an empirical drag reduction fraction \phi_{\text{DRA}}:

f_{\text{DRA}} = f_{\text{Moody}} \times (1 - \phi_{\text{DRA}})

DRA degrades as it passes through pump impellers, requiring re-injection at each pump station. The economic optimisation between DRA consumption and additional pump capacity is a standard operational problem on high-throughput lines.

Batch density and viscosity variation

The worked example used a single fluid with \rho = 900 kg/m³ and \mu = 0.012 Pa·s. In practice, the Enbridge Mainline carries multiple grades in sequence — light synthetic crude at 820 kg/m³, heavy dilbit at 910 kg/m³, condensate returning north at 730 kg/m³ — each with different hydraulic characteristics. The pipeline’s pressure profile changes continuously as each batch transits a given point. Pipeline operators use SCADA and model-predictive control to adjust pump station operating points as the density-viscosity profile of the line shifts. The single-fluid model here is the foundation; live operations layer real-time control on top of it.

Terrain and elevation head

The Darcy-Weisbach equation gives friction pressure drop only. Elevation changes add static head:

\Delta P_{\text{total}} = \Delta P_{\text{friction}} + \rho g \,\Delta z

where \Delta z is net elevation gain (positive = uphill). On the Trans Mountain route across the Rockies, elevation changes of several hundred metres over short distances create both significant energy penalties on uphill sections and the risk of excessive line pressure or flow runaway on steep downhill sections if pumps trip. The flat prairie corridors of the Enbridge Mainline and Keystone do not face this complexity at the same scale.

---

9. Extension: Pump Station Power and Operating Cost

The hydraulic work done by a pump station, divided by efficiency, gives the shaft power required:

P_{\text{pump}} = \frac{Q \cdot \Delta P_{\text{station}}}{\eta}

For Line 67 at the design point (Q = 1.380 m³/s, \Delta P_{\text{station}} = 7.5 MPa, \eta = 0.80):

P_{\text{pump}} = \frac{1.380 \times 7.5 \times 10^6}{0.80} = 12.9 \;\text{MW per station}

At CAD 0.08/kWh and continuous operation, one pump station consumes:

12.9 \;\text{MW} \times 8{,}760 \;\text{hr/yr} \times \$0.08/\text{kWh} \approx \$9 \;\text{million/yr}

Across a full pipeline with 40–50 pump stations, annual energy costs run to several hundred million dollars — substantial, but secondary to capital costs and still modest per barrel at this throughput.

The relationship between pump power and throughput is cubic. Because \Delta P \propto Q^2 (from the Darcy-Weisbach flow-rate form), and power is P = Q \cdot \Delta P:

P_{\text{pump}} \propto Q \cdot Q^2 = Q^3

Operating a pipeline at 90% of capacity rather than 75% requires (0.90/0.75)^3 = 1.73\times as much pump power — a 73% energy increase for a 20% throughput gain. Near the capacity ceiling, each marginal barrel becomes progressively more expensive in energy terms and infrastructure stress. This is why pipeline operators publish operating capacity bands rather than single nameplate numbers, and why running at sustained peak throughput accelerates maintenance cycles.

---

10. Math Refresher: Dimensional Analysis and the D^{-5} Result

Why dimensionless numbers?

The Darcy-Weisbach equation mixes metres, seconds, kilograms, and Pascals. Dimensional analysis organises these into groups that are universal — the same physics applies regardless of whether you are moving dilbit in Alberta or water in Dijon in 1857.

Reynolds number: Re = \rho v D / \mu

Checking dimensions:

\frac{[\text{kg/m}^3] \cdot [\text{m/s}] \cdot [\text{m}]}{[\text{kg/(m·s)}]} = \frac{\text{kg/(m·s)}}{\text{kg/(m·s)}} = 1 \quad \checkmark

The Reynolds number is dimensionless. A value of 144,000 predicts turbulent flow whether the fluid is dilbit, water, or liquid nitrogen, whether the pipe is 3 inches or 48 inches.

Friction factor: f = \Delta P / \bigl[({L}/{D})({\rho v^2}/{2})\bigr]

Also dimensionless. The \rho v^2/2 term (dynamic pressure) has units of Pa; multiplied by the dimensionless L/D and divided into \Delta P [Pa], the result is 1. ✓

Deriving the D^{-5} scaling

Start from the Darcy-Weisbach pressure drop equation:

\Delta P = f \cdot \frac{L}{D} \cdot \frac{\rho v^2}{2}

Substitute the velocity-to-flow rate relationship v = 4Q / \pi D^2:

v^2 = \frac{16 Q^2}{\pi^2 D^4}

\Delta P = f \cdot \frac{L}{D} \cdot \frac{\rho}{2} \cdot \frac{16 Q^2}{\pi^2 D^4} = \frac{8 f \rho L Q^2}{\pi^2 D^5}

The D^{-5} in the denominator comes from two sources: D^{-4} from substituting the velocity (v \propto Q/D^2, so v^2 \propto Q^2/D^4) and one additional D^{-1} from the L/D term. Combined: D^{-4} \times D^{-1} = D^{-5}.

The consequence: increase pipe diameter by a factor k and pressure drop for the same flow rate falls by k^5.

Diameter increase	Pressure drop reduction
+5% (k = 1.05)	\times (1.05)^{-5} = 0.784 → 22% reduction
+10% (k = 1.10)	\times (1.10)^{-5} = 0.621 → 38% reduction
+20% (k = 1.20)	\times (1.20)^{-5} = 0.402 → 60% reduction
+41% (20→34 in)	\times (1.41)^{-5} = 0.178 → 82% reduction

This is why the Trans Mountain Expansion could nearly triple corridor capacity by adding a 24-inch expansion line alongside the original 20-inch mainline, using much of the same pump infrastructure. And it is why debates about adding “more pumping capacity” to an existing line hit diminishing returns quickly: you can restore some lost head, but you cannot overcome the D^{-5} penalty of an undersized pipe through pump horsepower alone.

---

Summary

Concept	Key equation	What it governs
Reynolds number	Re = \rho v D / \mu	Flow regime: laminar vs turbulent
Friction factor (laminar)	f = 64/Re	Exact; wall roughness irrelevant
Friction factor (turbulent)	Churchill (1977)	Moody chart, all regimes
Pressure drop	\Delta P = f(L/D)(\rho v^2/2)	Energy consumed per unit length
Flow-rate form	\Delta P = 8f\rho L Q^2/\pi^2 D^5	Capacity vs diameter
Pump station spacing	L = \Delta P_{\text{MAOP}} \cdot \pi^2 D^5 / 8f\rho Q^2	Infrastructure design
Pump power	P = Q \cdot \Delta P / \eta	Operating energy cost
Netback per barrel of bitumen	(P_{\text{WCS}} - T - P_{C5+} \cdot d)/(1-d)	Producer economics

The physical result that matters most: Pressure drop scales with Q^2 and D^{-5}. Double throughput and friction losses quadruple. Increase pipe diameter by 10% and friction falls 38%. Every debate about Alberta pipeline capacity eventually reduces to these two relationships.

The economic result that matters most: The netback per barrel of bitumen is determined not by the WCS spot price alone, not by the tariff alone, but by their combination with the condensate cost and the diluent ratio, compared against the reference price at the destination. Trans Mountain changed Alberta’s netback arithmetic not by being cheap, but by opening access to a second price benchmark — and the value of that access varies with spreads that the hydraulic equations cannot predict.

---

Next: P2 — NGL and Condensate Systems — the fractionation cascade, the diluent supply chain, and mass balance mathematics through a Fort Saskatchewan fractionator.

References

Canada Energy Regulator. 2024a. Market Snapshot: WCS-WTI Price Differential. Canada Energy Regulator. https://www.cer-rec.gc.ca/en/data-analysis/energy-markets/market-snapshots/.

Canada Energy Regulator. 2024b. Pipeline Profiles: Enbridge Mainline and Trans Mountain Pipeline. Canada Energy Regulator. https://www.cer-rec.gc.ca/en/data-analysis/infrastructure/pipeline-profiles/.

Canada Energy Regulator. 2025. Canadian Crude Oil Exports and Markets. Canada Energy Regulator. https://www.cer-rec.gc.ca/en/data-analysis/energy-commodities/crude-oil-petroleum-products/.

Churchill, Stuart W. 1977. “Friction Factor Equation Spans All Fluid-Flow Regimes.” Chemical Engineering 84 (24): 91–92.

Colebrook, Cyril F. 1939. “Turbulent Flow in Pipes, with Particular Reference to the Transition Region Between the Smooth and Rough Pipe Laws.” Journal of the Institution of Civil Engineers 11 (4): 133–56. https://doi.org/10.1680/ijoti.1939.13150.

Enbridge Pipelines Inc. 2024a. Enbridge Mainline Tolls, Terms and Conditions of Service. Enbridge Pipelines Inc. https://www.cer-rec.gc.ca/en/applications-hearings/view-applications-decisions/.

Enbridge Pipelines Inc. 2024b. Line 67 (Alberta Clipper) Facility Information. Enbridge Pipelines Inc. https://www.enbridge.com/energy-transportation/crude-oil/mainline.

Moody, Lewis F. 1944. “Friction Factors for Pipe Flow.” Transactions of the ASME 66 (8): 671–84.

Swamee, Prabhata K., and Akalank K. Jain. 1976. “Explicit Equations for Pipe-Flow Problems.” Journal of the Hydraulics Division 102 (5): 657–64.

Trans Mountain Corporation. 2024. Trans Mountain Expansion Project: In-Service Announcement. Trans Mountain Corporation. https://www.transmountain.com/expansion-project.

Trans Mountain Pipeline ULC. 2024. Trans Mountain Pipeline Toll Principles, Methodology and Tariff. Trans Mountain Pipeline ULC. https://www.cer-rec.gc.ca/en/applications-hearings/view-applications-decisions/.