Impact of Storm Water Recharge Practices on Boston Groundwater Elevations

Abstract

Over the past century, the City of Boston has periodically experienced a decline in groundwater elevations and the associated deterioration of untreated wood piles, which support building foundations. To combat declining water tables, Boston enacted a groundwater conservation overlay district enforced by city zoning boards to require storm water recharge practices for any activity that triggers the zoning bylaw. In Boston, recharge to the water table results from the infiltration of rainfall and snowmelt, leakage from water mains, and recharge from artificial systems. Increased mitigation activities to reduce unaccounted-for water have reduced leakage from water mains in the city. Given the high percentage of impervious cover in Boston, the remaining sources of recharge are primarily artificial systems, including pump and infiltrate systems and storm water recharge best management practices (BMPs). The primary objective of this research was to exploit existing information on groundwater elevations and recharge practices to quantify the effect of the required recharge BMPs on the behavior of groundwater elevations in the Back Bay region of Boston. Regional multivariate regression models were developed to determine the potential effects of recharge BMPs on observed groundwater elevations. The literature review revealed several analogous multivariate linear regression studies, none which focused on behavior of stormwater BMPs. The model reveals that the installation of recharge BMPs has a small but highly statistically significant positive effect on groundwater elevations in the Back Bay with the effect being proportional to their capacity and inversely proportional to their distance from the location of interest. The resulting model can be used to predict the effect on average groundwater elevations at a particular location resulting from the installation of a recharge BMP or a set of such BMPs of a particular capacity at a particular distance from the location of interest.