Z. Moreno,
A. Paster,
T. Kamai
Drywell infiltration is a common approach to recharge groundwater and reduce load from drainage systems. In order to properly design a drywell, it is critical to predict its infiltration capacity, that is, its response to anticipated precipitation/stormwater/flood events. This is commonly conducted using models that solve the unsaturated flow in the subsurface using complex and costly numerical schemes. This work proposes a different approach, based on a solution for a sharp interface wetting front. The proposed model predicts the water level in the well and the subsurface wetting front location during and after an infiltration event. The model was tested and compared with numerical simulations of Richards' equation and with data from a field experiment, and proved to be sufficiently accurate. The typical run times of the model are smaller than 1 s and about three orders of magnitude shorter compared to the numerical model of Richards' equation. For illustrating possible applications, we use field data: the model is used to estimate the hydraulic properties via parameter optimization. Finally, a sensitivity analysis of the drywell response was conducted, demonstrating some practical applications for analysis, which may be used for properly determining site-specific drywell design.
Z. Moreno,
A. Paster,
T. Kamai
Drywell infiltration is a common approach to recharge groundwater and reduce load from drainage systems. In order to properly design a drywell, it is critical to predict its infiltration capacity, that is, its response to anticipated precipitation/stormwater/flood events. This is commonly conducted using models that solve the unsaturated flow in the subsurface using complex and costly numerical schemes. This work proposes a different approach, based on a solution for a sharp interface wetting front. The proposed model predicts the water level in the well and the subsurface wetting front location during and after an infiltration event. The model was tested and compared with numerical simulations of Richards' equation and with data from a field experiment, and proved to be sufficiently accurate. The typical run times of the model are smaller than 1 s and about three orders of magnitude shorter compared to the numerical model of Richards' equation. For illustrating possible applications, we use field data: the model is used to estimate the hydraulic properties via parameter optimization. Finally, a sensitivity analysis of the drywell response was conducted, demonstrating some practical applications for analysis, which may be used for properly determining site-specific drywell design.