Question 3 of 8: Rail-Tanker Crude Oil Derailment — Fate in the Vadose Zone

Question A-3: Rail-Tanker Crude Oil Derailment — Fate in the Vadose Zone (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Given. Crude oil is a light non-aqueous phase liquid (LNAPL, specific gravity ≈ 0.85–0.90, lighter than water); the punctured tanker releases the full 10,000 L over 3 h onto silty loam. Soil properties and depth to the unconfined aquifer are tabulated below.

Find. Describe the fate/transport of the released crude oil, and estimate the vadose zone's retention capacity for a given spill footprint to judge whether the water table is threatened.

Approach. Crude oil released at the surface infiltrates under gravity and capillary forces through the unsaturated (vadose) zone. As it moves down, a fraction is trapped by capillarity as residual saturation on soil grains (the same mechanism that leaves “field capacity” water behind), while the remainder continues to advance. Whether the mobile front reaches the water table depends on the available pore space in the flow path versus the volume released, so the calculation below estimates the vadose zone's oil-retention capacity per unit spill footprint and compares it to the tanker volume.

1. Volumetric water content already occupying pore space. The soil's 15% gravimetric water content converts to a volumetric content using the bulk density: $\theta_w = w\,\rho_b/\rho_w = 0.15 \times 1375/1000 = 0.206$ (20.6% of the soil volume is already water-filled).
2. Air-filled (available) porosity for oil. Subtracting from total porosity gives the pore space available to be invaded by oil as it displaces air: $\theta_a = n - \theta_w = 0.51 - 0.206 = \boxed{0.304}$ (30.4%).
3. Retention capacity per unit footprint. Treating the 2.0 m vadose-zone thickness as fully available for oil retention (a conservative upper bound, since some of $\theta_a$ will remain air-filled around residually trapped oil), the volume of oil the column can hold before the front could reach the water table, per square metre of spill footprint, is $R = \theta_a \times z_{wt} = 0.304 \times 2.0 = \boxed{0.6075\ \text{m}^3/\text{m}^2}$ (607.5 L/m²).
4. Minimum footprint to fully retain the spill above the water table. Dividing the spilled volume by this areal capacity gives the smallest spill footprint that keeps the whole 10,000 L release within the vadose zone: $A_{min} = V/R = 10\ \text{m}^3 / 0.6075\ \text{m}^3/\text{m}^2 = \boxed{16.5\ \text{m}^2}$.

A footprint smaller than about 16.5 m² (e.g., oil pooling in a confined rail-bed ditch or ballast trench) would drive the front to the water table; a footprint larger than this — more likely along an open rail right-of-way where the oil can spread laterally across ballast and adjacent ground before infiltrating — keeps the release within the unsaturated zone, at least until rainfall redistributes the retained oil further. In practice the spill will do both: an initial pool near the puncture point (small footprint, high risk of reaching groundwater) surrounded by a wider, thinner smear as the 3-hour drain continues and oil spreads along the rail bed. Once any oil reaches the water table it will float and spread laterally as a pancake-shaped LNAPL body on the capillary fringe (its density is below that of water), smearing a “bathtub ring” interval as the water table fluctuates seasonally, and will continue to dissolve BTEX and other soluble fractions into the underlying groundwater for as long as free product persists. The house 500 m away draws from the deeper confined aquifer (30 m bgs), which is protected from this shallow LNAPL by the intervening aquitard — the principal exposure pathway to that receptor is not the drinking-water well itself (unless the well is poorly constructed and short-circuits the confining layer) but vapour intrusion of light-end hydrocarbons if the plume migrates toward the house, and surface/near-surface contact via the septic field if lateral spreading reaches that far. Because both the ground surface and the shallow water table generally slope toward the wetland described in Question A-4, the more immediate ecological receptor at risk is that wetland, not the house.