The dispersion and migration behavior of hydrocarbon products leaking at low rates (i.e. 1bbl/day and 10 bbl/day) from a pipeline have been studied using a combination of experimental leakage tests and numerical simulations. The focus of this study was to determine the influence of subsurface engineered boundaries associated with the trench walls, and the presence of a water table, upon the leakage behavior of a range of hydrocarbon products. The project numerically modelled three products including diesel, diluted bitumen (dilbit) and gasoline; which were chosen to span a range of fluid types...
The dispersion and migration behavior of hydrocarbon products leaking at low rates (i.e. 1bbl/day and 10 bbl/day) from a pipeline have been studied using a combination of experimental leakage tests and numerical simulations. The focus of this study was to determine the influence of subsurface engineered boundaries associated with the trench walls, and the presence of a water table, upon the leakage behavior of a range of hydrocarbon products. The project numerically modelled three products including diesel, diluted bitumen (dilbit) and gasoline; which were chosen to span a range of fluid types and viscosities. Laboratory simulations of leakage were carried out for the most viscous product (i.e. dilbit) in order to capture plume dispersion in semi-real time, and to allow numerical predictions to be assessed against experimental data. Direct comparisons between observed plume dimensions over time and numerically predicted behavior suggested a good match under low moisture conditions, providing confidence that the numerical simulation was sufficiently reliable to model field-scale applications.
Following a simulated two year initialization period, the leakage of products, their associated gas phase migration, thermal and geomechanical effects were simulated for a period of 365 days. Comparisons between product leakage rate, product type and soil moisture content were made and the spatial impacts of leakage were summarized. Variably compacted backfill within the trench, surrounded by undisturbed and more compacted natural soils, results porosity and permeability differences which control the migration of liquids, gases, thermal effects and surface heave. Dilbit migration is influenced heavily by the trench, and also its increasing viscosity as it cools and degases after leakage. Diesel and gasoline liquid plumes are also affected by the trench structure, but to a lesser extent, resulting in wider and longer plumes in the subsurface. In all cases, the migration of liquids and gases is facilitated by higher permeability zones at the base of the pipe. Volatile Organic Compounds (VOCs) migrate along the trench and break through at the surface within days of the leak. Temperature changes within the trench may increase due liquid migration, however the change in predicted temperature at the surface above the leak is less than 0.5°C above background. For gasoline, the large amount of degassing and diffusion through the soil results in cooling of the soil by up to 1°C. Induced surface displacement was predicted for dilbit and for one case of diesel, but only in the order of 0.2cm above baseline.
Based upon the information gathered, recommendations are provided for the use and placement of generic leak detection sensor types (e.g liquid, gas, thermal, displacement) within the trench and / or above the ground surface. The monitoring locations suggested take into account requirements to detect pipeline leakage as early as possible in order to facilitate notification of the operator and to predict the potential extent of site characterization required during spill response and longer term remediation activities.