Advancing Multi‐Phase LNAPL Extraction Models by Incorporating a Wellbore Module

Abstract Current mathematical models are unable to accurately predict the long‐term extraction rate of subsurface LNAPL contaminants due to an overestimation of the mass transfer rate between wellbore and soil. This shortcoming leads to misjudgments regarding the suggested implementation period and decisions on whether the technology should be adopted. Therefore, we developed a multi‐phase transport model that incorporates a wellbore module, which for the first time enables accurate characterization of the pressure difference at the well screen by correlating well configuration, in‐well LNAPL/water layer thicknesses, and LNAPL saturation in the soil. The proposed model reduces the mean absolute percentage error against field measured extraction rate of LNAPL from 112.32% to 17.41% compared to previous numerical models such as TMVOC. Our model is able to quantitatively characterize the inevitable decay of LNAPL extraction rate due to the attenuation of pressure difference and hydraulic conductivity of the LNAPL around the well screen. Our simulation results reveal that rising groundwater tables displace the main body of LNAPL in the soil upwards and away from well screen. This means that the measures to prevent volatile organic compounds from being transmitted to the surface through the wellbore, such as sealing with cement‐bentonite grout, will significantly reduce the extraction efficiency. In conclusion, compared with previous models, the advanced model including a new module that accounts for wellbore activity can better reflect the combined effects of process parameters (e.g., pump position), well configuration, and groundwater depth on LNAPL extraction efficiency.