Rankine Lateral Earth Pressure Analysis for Retaining Wall Design: A Senior Geotechnical Engineer's Guide

Engineering Guide

What Is This Calculation and Why It Matters

Lateral earth pressure calculation is a foundational geotechnical analysis underpinning the safe, economical, and code-compliant design of retaining structures—ranging from cantilevered concrete walls and mechanically stabilized earth (MSE) systems to basement walls and bridge abutments. Unlike vertical stress (which arises directly from overburden), lateral earth pressure is a horizontal stress state induced by soil’s self-weight, cohesion, internal friction, and boundary conditions. Its magnitude and distribution govern critical structural demands: bending moments in wall stems, shear forces at the base, foundation bearing pressures, and global stability against sliding and overturning.

The Rankine theory—developed by William John Macquorn Rankine in 1857—is one of the most widely applied classical methods for estimating lateral earth pressure under active and passive conditions. It assumes a homogeneous, isotropic, cohesionless (or cohesive) soil mass bounded by a smooth, vertical, rigid retaining wall with no wall-soil friction (δ = 0°). While simplified, Rankine remains indispensable in preliminary design, code-based load modeling (e.g., ASCE 7-16), and educational frameworks due to its analytical transparency, computational efficiency, and strong correlation with field behavior for well-drained, granular backfills.

Neglecting or misapplying this calculation risks catastrophic failure modes: excessive wall deflection leading to cracking and water infiltration; inadequate reinforcement causing flexural collapse; or insufficient base width resulting in sliding or rotational instability. Conversely, overestimating pressure leads to unnecessarily thick, costly, and resource-intensive walls—violating sustainability and value-engineering principles. Thus, mastery of Rankine analysis is not merely academic—it is an ethical and professional obligation for structural and geotechnical engineers.

Theory and Formula Walkthrough

Rankine theory derives from Mohr–Coulomb failure criteria applied to a soil element in a state of plastic equilibrium. It identifies two limiting stress states:

Active state: Occurs when the retaining wall moves away from the soil, allowing the soil to expand laterally and reach failure. Lateral pressure is minimized.
Passive state: Occurs when the wall is pushed into the soil, compressing it laterally to failure. Lateral pressure is maximized.

For a cohesionless soil (c = 0), the Rankine coefficients depend solely on the effective angle of internal friction, φ′:

Active Earth Pressure Coefficient (Kₐ)

$$ K_a = \tan^2\left(45^\circ - \frac{\phi'}{2}\right) $$

φ′ (phi prime): Effective angle of internal friction (degrees), representing the soil’s resistance to shear under drained conditions. It must be determined via consolidated-drained (CD) triaxial tests or direct shear tests on representative, properly sampled specimens. In practice, φ′ is often taken as the peak friction angle for dense sands or the critical-state angle for loose or heavily vibrated soils.
The term $45^\circ - \phi'/2$ defines the inclination of the major principal stress plane relative to the horizontal in the active wedge. As φ′ increases, Kₐ decreases—reflecting greater soil strength and reduced lateral thrust.

Passive Earth Pressure Coefficient (Kₚ)

$$ K_p = \tan^2\left(45^\circ + \frac{\phi'}{2}\right) $$

Identical derivation but with sign reversal: the major principal stress rotates toward the wall. Note that $K_p = 1/K_a$ only for cohesionless soils with δ = 0° and vertical backface—this reciprocal relationship underscores the asymmetry between mobilization capacity (active) and resistance capacity (passive).

Lateral Earth Pressure at Depth (σₕ)

For a level, surcharge-free, dry backfill:

$$ \sigma_h = K \cdot \sigma_v = K \cdot \gamma \cdot z $$

γ: Unit weight of soil (kN/m³). For partially saturated or submerged soils, use the effective unit weight γ′ = γₛₐₜ − γ_w where γ_w = 9.81 kN/m³. Ignoring buoyancy above the water table is a frequent source of error.
z: Vertical depth below ground surface (m), measured orthogonally to the ground surface.
K: Either Kₐ (for active design loads) or Kₚ (for passive resistance checks, e.g., at the toe or in front of embedded walls).
σₕ: Total lateral pressure (kN/m²) acting horizontally per unit area at depth z. It varies linearly with depth—forming a triangular distribution with zero at the top and maximum at the base.

When cohesion (c′) is present (e.g., silty clays), Rankine modifies to:

$$ \sigma_h = K \cdot \sigma_v \mp 2c'\sqrt{K} $$

where the minus sign applies to active pressure (reducing thrust) and the plus sign to passive (increasing resistance). However, the tool specification assumes c′ = 0, consistent with typical granular backfill design assumptions.

Standard Requirements

Both ASCE 7-16 and EN 1997-1 mandate rigorous treatment of lateral earth pressures—but with distinct philosophical approaches.

ASCE 7-16 (Chapter 6: Soil and Hydrostatic Pressures)

Section 6.3.1 requires lateral earth pressure to be calculated using “accepted engineering principles” — explicitly endorsing Rankine, Coulomb, and log-spiral methods depending on wall geometry and soil conditions.
Section 6.3.2.1 specifies that active pressures shall be used for design of retaining walls unless passive resistance is reliably mobilized (e.g., via structural connection to foundations or sufficient embedment depth). Passive resistance may only be included if justified by analysis and verified by site-specific conditions.
Section 6.3.3 mandates accounting for hydrostatic pressure when groundwater is present. The effective unit weight γ′ must be used below the water table, and pore water pressure (u = γ_w·z_submerged) must be added as a separate triangular load.
Load factors are applied per Section 2.3: active earth pressure is treated as a live load with a factor of 1.6 in LRFD combinations (e.g., 1.2D + 1.6H), while passive resistance is factored at 0.7 (i.e., reduced by 30%) to reflect uncertainty in mobilization.

Eurocode 7 (EN 1997-1:2004+A1:2013, Section 3.4: Actions on Structures)

Clause 3.4.1(3) defines earth pressure as an imposed action, requiring classification as either favorable (passive resistance) or unfavorable (active thrust) depending on limit state.
Clause 3.4.2(2) requires partial factors: γ_G = 1.35 for permanent actions (soil weight), γ_Q = 1.5 for variable actions (surcharge), and γ_E = 1.3 for earth pressure itself in DA1 Combination 1 (STR/GEO ultimate limit states).
Critically, Annex A.3.4 notes that Rankine theory is appropriate only when wall movement is sufficient to develop full active/passive states—and recommends reduction factors (e.g., 0.8–0.9 on Kₐ) for restrained walls (e.g., basement walls with floor slabs) where mobilization is limited.

Both standards emphasize that theoretical coefficients must be calibrated with site investigation data—not textbook defaults.

Common Mistakes and How to Avoid Them

Using total unit weight instead of effective unit weight below the water table
Consequence: Overestimation of lateral load by up to 40–50% in saturated sands.
Fix: Always delineate the phreatic surface from borehole logs or piezometer data. Compute γ′ = γ_sat − 9.81 kN/m³ for submerged zones.
Applying Rankine to cohesive soils without verifying mobilization assumptions
Consequence: Underestimating active pressure in stiff clays (where undrained conditions dominate) or ignoring time-dependent consolidation effects.
Fix: For short-term (undrained) design, use total stress parameters (φ_u ≈ 0°, c_u > 0) and consider alternative methods like Terzaghi’s or numerical modeling. Rankine is strictly valid for drained conditions.
Ignoring wall batter or backfill slope
Consequence: Significant error—e.g., a 10° backfill slope increases Kₐ by ~15% vs. level backfill.
Fix: Use modified Rankine equations or switch to Coulomb theory when β (backfill slope) ≠ 0° or α (wall batter) ≠ 90°.
Assuming full passive resistance without verifying mobilization distance
Consequence: Unconservative design—passive resistance requires ~0.01–0.05H wall movement (H = embedment depth); restrained walls rarely achieve this.
Fix: Per Eurocode 7 Annex A.3.4, apply a mobilization factor (e.g., 0.5–0.7) to Kₚ unless instrumentation confirms displacement. Never rely on passive pressure for global stability without explicit verification.
Omitting surcharge loads
Consequence: Under-designed stem reinforcement and increased long-term creep deformation.
Fix: Model uniform surcharge (q) as an equivalent additional height of soil: $\Delta\sigma_h = K_a \cdot q$. Include live loads (e.g., 10–20 kPa for roadways) and dead loads (e.g., adjacent structures) per ASCE 7 Table 6.1.

Worked Example with Realistic Numbers

Scenario: A freestanding cantilever retaining wall retains a well-graded sand backfill. Site investigation yields:

φ′ = 32° (from CD triaxial tests on dense sand)
γ = 18.5 kN/m³ (dry unit weight above water table)
Groundwater table at 2.0 m depth → γ′ = 18.5 − 9.81 = 8.69 kN/m³ below
Design depth z = 5.0 m (wall height)

Step 1: Compute Rankine coefficients

$$ K_a = \tan^2\left(45^\circ - \frac{32^\circ}{2}\right) = \tan^2(29^\circ) = (0.554)^2 = 0.307 \approx 0.31 $$

$$ K_p = \tan^2\left(45^\circ + \frac{32^\circ}{2}\right) = \tan^2(61^\circ) = (1.804)^2 = 3.256 \approx 3.26 $$

Step 2: Compute lateral pressure at z = 5.0 m

Because groundwater exists, split calculation:

Above water table (z = 0 to 2.0 m): σᵥ = 18.5 × z → σₕ,active = 0.307 × 18.5 × z
Below water table (z = 2.0 to 5.0 m): σᵥ = 18.5×2.0 + 8.69×(z−2.0) → σₕ,active = 0.307 × [37.0 + 8.69(z−2.0)]

At z = 5.0 m:

σᵥ = 37.0 + 8.69 × 3.0 = 37.0 + 26.07 = 63.07 kPa
σₕ = 0.307 × 63.07 = 19.36 kN/m²

Add hydrostatic pressure: u = 9.81 × 3.0 = 29.43 kPa → total lateral pressure = 19.36 + 29.43 = 48.79 kN/m²

Step 3: Interpretation & Design Integration

Active coefficient Kₐ = 0.31 indicates moderate lateral thrust—typical for dense sand. Compare with typical values: φ′ = 25° → Kₐ = 0.406; φ′ = 40° → Kₐ = 0.217.
Passive coefficient Kₚ = 3.26 reflects high resistance capacity—however, per ASCE 7-16 §6.3.2.1 and Eurocode 7 §3.4.1, passive resistance at the toe should be reduced by at least 30% (factored Kₚ = 0.7 × 3.26 = 2.28) and only included if embedment ≥ 0.7 m and soil is competent.
The computed 48.79 kN/m² at 5 m depth anchors the triangular active pressure diagram (0 at top, 48.79 at base), yielding resultant force Hₐ = ½ × 48.79 × 5 = 122.0 kN/m acting at 5/3 = 1.67 m above base.

This result feeds directly into moment equilibrium checks: overturning moment = Hₐ × (height to resultant) = 122.0 × 1.67 = 203.7 kN·m/m; resisting moment depends on wall weight and passive pressure. Stability ratios (e.g., overturning = 3.0+, sliding = 1.5+) must then satisfy ASCE 7-16 §6.4 and EN 1997-1 §11.6.

In summary, Rankine analysis delivers actionable, code-aligned inputs—but only when grounded in rigorous site characterization, correct parameter selection, and disciplined application of standard-mandated safety factors. Treat it not as a black-box calculator, but as a diagnostic lens into soil–structure interaction physics.

← Back to Lateral Earth Pressure Calculator

📜 Applicable Standards

ASCE7-16 (Chapter 6) EUROCODE7 (Section 3.4)

💬 Frequently Asked Questions

What is the Rankine active earth pressure coefficient formula, and how does it relate to the angle of internal friction? ▼

The Rankine active earth pressure coefficient is calculated as $K_a = \tan^2(45^\circ - \phi'/2)$, where $\phi'$ is the effective angle of internal friction (in degrees). This expression assumes a cohesionless, homogeneous, isotropic soil with a horizontal ground surface and a smooth, vertical retaining wall—key assumptions of classical Rankine theory. For $\phi' = 30^\circ$, $K_a = 0.33$; increasing $\phi'$ reduces $K_a$, reflecting greater soil stability. Note that Rankine theory neglects wall friction and soil cohesion—so for cohesive soils or rough walls, Coulomb or more advanced methods (e.g., log-spiral or numerical analysis per Eurocode 7 Annex D) are preferred. ASCE 7-16 Section 3.2.2 permits Rankine for preliminary design but mandates verification with site-specific shear strength parameters.

How does groundwater affect lateral earth pressure calculations in the Rankine model? ▼

Groundwater significantly increases lateral pressure by reducing effective stress and introducing pore water pressure. Below the water table, use the effective unit weight ($\gamma' = \gamma_{sat} - \gamma_w$) for computing the effective lateral pressure, then add hydrostatic pressure ($u = \gamma_w \cdot h_{water}$) linearly. Total lateral pressure becomes $\sigma_h = K_a \cdot \sigma'_v + u$. Ignoring this leads to under-designed walls—common cause of failure. Per Eurocode 7 (EN 1997-1 §9.6.2), partial factors must be applied separately to $\gamma'$ and $u$. ASCE 7-16 requires saturated conditions to be modeled explicitly in load combinations (e.g., Load Case 6: $D + H + F + W$). Always confirm phreatic level via piezometer data—not assumed.

Can I use the Rankine calculator for clayey soils with cohesion? ▼

No—Rankine theory assumes zero cohesion ($c' = 0$) and relies solely on $\phi'$ for lateral pressure prediction. Applying it directly to cohesive soils (e.g., clays) will underestimate active pressure near the wall top and overestimate it at depth, risking instability. For $c' > 0$, use the extended Rankine formulation: $\sigma_{a} = K_a \sigma_v' - 2c'\sqrt{K_a}$, but only if tension cracks are accounted for (depth $z_c = 2c'/\gamma\sqrt{K_a}$). Better practice: adopt Coulomb with cohesion, or—per Eurocode 7 §C.3.2—use undrained analysis ($\phi_u = 0$, $c_u$) for short-term clay conditions. Field vane tests or consolidated-undrained triaxials are essential to characterize $c'$ and $\phi'$ reliably.

What’s the difference between active and passive earth pressure coefficients—and why does passive require higher wall movement? ▼

Active pressure ($K_a$) occurs when the wall moves away from soil, mobilizing full shear resistance; passive pressure ($K_p = \tan^2(45^\circ + \phi'/2)$) develops when the wall pushes into soil, requiring ~5–10× greater displacement (typically 0.01–0.05H for granular soils per FHWA NHI-16-007). $K_p$ is always > $K_a$ (e.g., $\phi'=30^\circ$: $K_a=0.33$, $K_p=3.00$), but passive resistance is rarely fully mobilized due to practical displacement limits. ASCE 7-16 Section 3.2.3 cautions against relying on passive pressure unless movement is assured and soil is well-compacted. Eurocode 7 Annex C recommends reducing $K_p$ by 20–30% for design conservatism unless monitored movement confirms mobilization.

How do surcharge loads influence lateral earth pressure—and how should they be added in Rankine analysis? ▼

Surcharge loads (e.g., roadways, buildings) induce additional uniform vertical stress ($q$) that amplifies lateral pressure via $\Delta\sigma_h = K_a \cdot q$. This adds a constant lateral component across depth—unlike soil self-weight, which increases linearly. In layered soils or non-uniform surcharges, use influence charts or Boussinesq-based distribution. ASCE 7-16 Section 3.2.2.2 requires surcharge to be included in load combinations (e.g., $D + H + F + (0.75L + 0.75S + 0.75R)$), with $K_a$ applied to the effective surcharge. Eurocode 7 §9.6.2 mandates separate partial factors for surcharge ($\gamma_Q = 1.5$) and earth pressure ($\gamma_G = 1.35$). Always verify surcharge magnitude and extent via geotechnical report—not estimated.

Is Rankine theory suitable for cantilever retaining walls per modern design standards? ▼

Yes—for preliminary sizing and low-risk applications—but not for final design without verification. Rankine provides conservative $K_a$ estimates for vertical, smooth walls with level backfill, satisfying ASCE 7-16’s ‘simplified method’ criteria (§3.2.2). However, cantilever walls experience base rotation and wall-soil interaction not captured by Rankine. Eurocode 7 §9.6.2 requires global stability analysis (sliding, overturning, bearing) using characteristic actions and partial factors—Rankine alone is insufficient. Modern practice combines Rankine for initial $\sigma_h$ estimation, then refines with limit equilibrium (e.g., using software like Slide or RSPile) or finite-element modeling (per ASTM D6027) to capture nonlinear behavior and wall flexibility.

Why does the calculator output lateral pressure in kN/m² instead of total force—and how do I convert it for structural design? ▼

The calculator outputs pressure ($\sigma_h$) in kN/m² (equivalent to kPa) because Rankine theory gives stress at a specific depth—not integrated force. To obtain total lateral force per meter width ($P_a$), integrate $\sigma_h(z)$ over height $H$: $P_a = \frac{1}{2} K_a \gamma H^2$ (for dry, level backfill). For design, this triangular pressure distribution governs bending moment ($M = P_a \cdot H/3$) and shear at the base. ASCE 7-16 Section 3.2.2.1 specifies that $P_a$ must be combined with other loads (e.g., seismic $F_h = 0.35P_a$ per §12.13.2) using appropriate load factors. Always apply pressure at the centroid of its diagram—critical for overturning checks per Eurocode 7 §11.6.2.

📈 Case Studies

Urban Retaining Wall for Slope Stabilization in Seattle

Case Study 1: Urban Retaining Wall for Slope Stabilization in Seattle

Scenario A mixed-use development in Seattle’s Capitol Hill neighborhood required a 6-m-tall cantilever retaining wall to stabilize a steep, cut slope adjacent to an existing historic building. Site constraints included limited right-of-way (max 1.2 m excavation setback), strict settlement limits (<5 mm), and presence of glacial till overlain by weathered silt—requiring conservative soil parameters. Geotechnical investigation confirmed drained conditions (no groundwater at design depth), but surcharge from adjacent pedestrian walkway (15 kN/m²) was accounted separately per ASCE 7-16.

Given Data

Angle of internal friction: 32°
Unit weight of soil: 19.2 kN/m³
Depth below ground level: 6.0 m

Calculation Using Rankine’s theory for cohesionless, level backfill (no wall friction or inclination):

Active earth pressure coefficient: ( K_a = \tan^2\left(45^\circ - \frac{\phi'}{2}\right) = \tan^2\left(45 - \frac{32}{2}\right) = \tan^2(29^\circ) \approx 0.306 \rightarrow \textbf{0.31} ) (rounded to 2 decimals)
Passive earth pressure coefficient: ( K_p = \tan^2\left(45^\circ + \frac{\phi'}{2}\right) = \tan^2\left(45 + 16\right) = \tan^2(61^\circ) \approx 3.27 \rightarrow \textbf{3.27} )
Vertical effective stress at 6.0 m: ( \sigma'_v = \gamma \cdot z = 19.2 \times 6.0 = 115.2 , \text{kN/m}^2 )
Lateral earth pressure at depth: ( \sigma'_a = K_a \cdot \sigma'_v = 0.306 \times 115.2 \approx 35.25 , \text{kN/m}^2 \rightarrow \textbf{35.25 kN/m²} )

Note: Total active lateral force per meter (for wall design) integrates this triangular distribution: ( P_a = \frac{1}{2} K_a \gamma H^2 = 105.8 , \text{kN/m} ); however, the tool reports point pressure at depth.

Result and Decision The calculated lateral pressure of 35.25 kN/m² at base level governed stem thickness and heel length. Combined with surcharge-increased moment, the design adopted a 0.45-m-thick reinforced concrete stem with 2.1-m-long toe and 1.8-m-long heel. A granular filter and weep holes were specified to maintain drained conditions—critical given Seattle’s high rainfall.

Lesson Even in drained, cohesionless soils, small increases in φ′ (e.g., 32° vs. default 30°) reduce Ka significantly—here by ~7%—directly lowering design loads; always use site-specific φ′ from consolidated-drained triaxial tests, not textbook defaults.

Highway Embankment Support on Coastal Clay in Louisiana

Case Study 2: Highway Embankment Support on Coastal Clay in Louisiana

Scenario LA DOTD upgraded US-90 near Houma, requiring a 4.5-m-high MSE (mechanically stabilized earth) wall to retain a new embankment over soft, normally consolidated Gulf Coast clay. Due to rapid construction schedule and limited access, the design team used a temporary soldier pile–lagging system during construction, later transitioning to permanent MSE. Soil data from CPT and vane shear tests indicated low friction angle due to high water content; undrained analysis was inappropriate because the wall would be built in stages allowing partial consolidation. Thus, effective-stress Rankine analysis applied using consolidated-undrained (CU) triaxial φ′ values.

Given Data

Angle of internal friction: 18° (low due to high plasticity index PI = 52)
Unit weight of soil: 15.8 kN/m³ (saturated unit weight corrected for buoyancy not needed—groundwater table >10 m below surface)
Depth below ground level: 4.5 m

Calculation

Active earth pressure coefficient: ( K_a = \tan^2\left(45^\circ - \frac{18}{2}\right) = \tan^2(36^\circ) \approx 0.529 \rightarrow \textbf{0.53} )
Passive earth pressure coefficient: ( K_p = \tan^2\left(45^\circ + \frac{18}{2}\right) = \tan^2(54^\circ) \approx 1.89 \rightarrow \textbf{1.89} )
Vertical effective stress at 4.5 m: ( \sigma'_v = 15.8 \times 4.5 = 71.1 , \text{kN/m}^2 )
Lateral earth pressure at depth: ( \sigma'_a = 0.529 \times 71.1 \approx 37.61 , \text{kN/m}^2 \rightarrow \textbf{37.61 kN/m²} )

Note: This unusually high Ka (vs. typical 0.3–0.35 for sands) reflects the low φ′—driving higher lateral loads and necessitating tighter reinforcement spacing in the MSE design.

Result and Decision The elevated lateral pressure (37.61 kN/m²) triggered redesign of reinforcement vertical spacing from 0.8 m to 0.6 m in the lower third of the wall and increased geogrid tensile strength from 25 kN/m to 42 kN/m. Settlement analysis confirmed acceptable long-term performance under combined embankment + traffic loading, validated via 6-month post-construction monitoring.

Lesson Low φ′ soils (e.g., plastic clays) produce disproportionately high Ka values—misapplying sand-based assumptions risks underdesign; always cross-check Ka against empirical correlations (e.g., from CPT-derived φ′) and validate with limit equilibrium software when Ka > 0.45.