Date of Award

Spring 2018

Project Type


Program or Major

Civil Engineering

Degree Name

Master of Science

First Advisor

Thomas P Ballestero

Second Advisor

James Houle

Third Advisor

Alison Watts


Bioretention filters are a common Best Management Practice used to treat pollutants and water volumes of stormwater runoff from urbanized watersheds. The current static sizing criterium for new design applies Darcy’s law to the Water Quality Volume (WQV) as it filters through the bioretention soil media. In retrofit applications, the WQV is instantaneously stored in the system pore space, including the ponded surface. While these designs are simple to implement, the systems areas are oversized because no infiltration or other outflow is considered in the design.

The retrofit bioretention study site for this research located in Dover, NH (HSBS) treats a 21.9 ac. suburban watershed with 38% impervious cover. The static design treatment rainfall for the HSBS was 0.16 inches. Monitored data yielded a mean peak discharge reduction and mean volume reduction of 62% and 35% with standard deviations of 25% and 37%, respectively. Of 45 observed storms with one-minute logging intervals, 67% of the events were fully treated and ponding did not exceed the designed elevation; 95% of these events were less than 1.27 inches. The static design rainfall was in the 5.4th percentile, almost 800% less than the largest observed treated storm with no bypass. The volume balance model developed in this study used a Green and Ampt approach for the vertical bottom and lateral sidewall infiltrations integrated over the duration of the event. The model requires basic watershed and climatic properties, design dimensions, and anisotropic native soil characteristics. Comparing estimates of bottomonly infiltration and the developed model to observed infiltrated volumes yielded mean errors of -79% and -8%, with RMSE of 3210 cu. ft. and 2650 cu. ft., respectively. On average, 39% of the total infiltrated volume occurred through the sidewalls. The retrofit treatment rainfall for the model was 0.52 inches. This was 225% larger than the current static design rainfall and was in the 49th percentile of the observed, treated events for HSBS, which was an improvement of 44% over the current static design.