Stormwater biofilter treatment model (MPiRe) for selected micro-pollutants

• Developed quality model is checked against field data at two biofilters.
• Triazines are modelled well for all events on both biofilter cells.
• Glyphosate and chloroform are modelled well in good-practice-design biofilter.
• Calibrated model parameter values agree with literature reported data.
• Long antecedent dry period posed a problem for quality modelling in low flow events.

Biofiltration systems, also known as bioretentions or rain-gardens, are widely used for treatment of stormwater. In order to design them well, it is important to improve models that can predict their performance. This paper presents a rare model that can simulate removal of a wide range of micro-pollutants from stormwater by biofilters. The model is based on (1) a bucket approach for water flow simulation, and (2) advection/dispersion transport equations for pollutant transport and fate. The latter includes chemical non-equilibrium two-site model of sorption, first-order decay, and volatilization, thus is a compromise between the limited availability of data (on stormwater micro-pollutants) and the required complexity to accurately describe the nature of the phenomenon. The model was calibrated and independently validated on two field data series collected for different organic micro-pollutants at two biofilters of different design. This included data on triazines (atrazine, prometryn, and simazine), glyphosate, and chloroform during six simulated stormwater events. The data included variable and challenging biofilter operational conditions; e.g. variable inflow volumes, dry and wet period dynamics, and inflow pollutant concentrations. The model was able to simulate water flow well, with slight discrepancies being observed only during long dry periods when, presumably, soil cracking occurred. In general, the agreement between simulated and measured pollutographs was good. As with flows, the long dry periods posed a problem for water quality simulation (e.g. simazine and prometryn were difficult to model in low inflow events that followed prolonged dry periods). However, it was encouraging that pollutant transport and fate parameters estimated by the model calibration were in agreement with available literature data. This suggests that the model could probably be adopted for assessment of biofilter performance of other stormwater micro-pollutants (PAHs, phenols, phthalates, etc.). The model, therefore, could be applied in practice for sizing of biofilter systems and their validation monitoring, when used for stormwater harvesting.

Figure optionsDownload high-quality image (298 K)Download as PowerPoint slide