Performance Assessment of Three Types of Rainwater Detention Structures for an Urban Development in Wilsonville, Oregon, USA

The performance of three types of low-impact development (LID) rainwater detention structures were tested using simulated storms and ambient rainfall for water quantity and quality parameters in various seasons. The structures included stormwater planter boxes (n=12), extensive ecoroofs (n=2) and a 93 m2 section of porous pavement. The planter box types varied with respect to soil depths (22-26 cm and 38-42 cm), soil mixes (n=2) and with respect to the use of a fabric filter, resulting in 6 unique planter box types measured, each of which had 2 replicates. For the planter boxes and porous pavement, measurement of water retention and time lag between input and output flow was conducted for 6 artificial storms, with inputs approximating a storm intensity of 1.3 cm/24 hr. For the ecoroof panels, ambient rainfall and resulting ecoroof runoff were collected over a 6 month period. Results showed that water retention varied from 0% to 59% for the stormwater planter boxes over 72 separate storm simulations, with a median stormwater retention of 20%. Lag time between the centroids of input and output flow varied from 1.7 min to 29.9 min, with a mean stormwater lag time of 12.8 min for all planter boxes. Using repeated measures ANOVA, with a significance level of p<0.05, differences were found among the planter box types for water retention during dry antecendent storms and for time lag for all storms. In general, as shown by multiple comparison tests, planter box configurations that used a finer, less porous soil mix and had no fabric filter produced the largest stormwater retention and the longest delay in transmission of stormwater through the planter box. For water quality for 5 simulated storms for all boxes, mean specific conductivity increased by 27 µS/cm from input to output flow, and mean water temperature decreased by 1.9°C. Ecoroof water retention varied from 0% to 100% depending on storm duration, with a negative exponential relationship found between water retention % and storm duration (adj R2 = 0.41, p<.005). Porous pavement storm simulations showed >99% retention for 3 storms, while not performing as well during 3 other storm simulations due to clogging that occurred on site. Overall this study found that all three structures, stormwater boxes, ecoroofs and porous pavement, hold promise for a significant reduction in stormwater in urban and suburban neighborhoods.