Stormwater biofiltration systems can alleviate eutrophication by reducing nitrogen loadings from stormwater runoff. These systems can be modified to further enhance nitrogen removal by incorporating a submerged layer with a carbon source to promote denitrification. The objectives of this research were to compare the nitrate (NO À 3 ) removal performance of various media mixtures under varying environmental conditions and to estimate the longevity of carbon media in denitrifying biofilter microcosms. Media types evaluated consisted of mixtures containing sand, gravel, and/or eucalyptus wood chips. Negligible NO À 3 removal and high concentrations of total Kjeldahl nitrogen (TKN), phosphate (PO 3 À 4 ), and dissolved organic carbon (DOC) were observed from carbon-containing media that were left unsaturated before analysis. High removal of NO À 3 (85-100%) and low concentrations of TKN and DOC were observed when carbon-containing media were acclimated under saturated conditions. A mixture of gravel and eucalyptus-wood media exhibited high NO À 3 removal rates, low final DOC concentrations, and was selected for further analysis. Results indicated that carbon-containing biofilters should include gravel instead of sand media and be designed to maintain saturation to increase NO À 3 removal rates and decrease TKN, PO 3 À 4 , and DOC export. Results from the gravel-wood media were then imported into a novel model developed to estimate media longevity. Based on observed DOC dissolution rate, estimated longevity of the gravel-wood media in denitrifying biofilters was greater than a decade. Results indicate that permanently saturated carbon-containing media will supply a sustainable source of carbon over the functional lifespan of stormwater biofilters.

No abstract

No abstract

Denitrifying stormwater biofiltration systems that include submerged zones containing organic carbon sources can alleviate eutrophication by reducing nitrogen loadings from stormwater runoff. The objective of this research was to evaluate performance of denitrifying biofilters containing wood chip medium under varying dynamic loading rates and antecedent dry conditions (ADCs). Results showed that during ADCs, dissolved organic carbon (DOC) concentrations increased in pore water due to dissolution of the wood chip medium. During a storm event, nitrate (NO 3 -) removal was enhanced due to high pore water DOC concentrations and mixing of NO 3 -from the influent with pore water; however, as DOC was gradually flushed out of the biofilter, higher effluent NO 3 -concentrations were observed over time. At lower hydraulic loading rates, effluent NO 3 -concentrations were low due to sufficient hydraulic residence time for denitrification. For all storm events tested, an overall NO 3 -mass removal efficiency of 85% was achieved. Sulfate reduction was observed at long ADCs (up to 30 days). Phosphate release was observed when NO 3 -was absent from the effluent, possibly due to dissolution of iron oxyhydroxides under low redox conditions. Findings from this study are being used to develop a mathematical model of nitrogen removal in denitrifying stormwater biofilters.