Stormwater bioretention systems

A bioretention cell is a depression filled with engineered filter media into which surface runoff is directed. The plants are what you see, and they usually get the credit. In fact each pollutant class is removed by its own mechanism, and nearly all of those mechanisms are properties of the media and the hydraulics.

Different pollutants, different routes#

Suspended solids are removed physically, by settling in the ponding zone and straining in the upper media. That matters more than it sounds, because most metals and polycyclic aromatic hydrocarbons in road runoff arrive bound to particles; removing solids removes them too.

Dissolved metals are held by sorption and cation exchange on organic matter and on the surfaces of iron and aluminium oxides and hydroxides. The process is strongly pH-dependent: as pH rises, surface hydroxyl groups deprotonate and metal sorption climbs steeply. Acidic media give metals back.

Phosphorus is where media chemistry decides everything. Sorption capacity is set by amorphous iron and aluminium content, and a media generously amended with compost turns out to be a source rather than a sink: cells routinely export dissolved phosphorus at concentrations above the influent. Modern specifications therefore cap the compost fraction and require the media to be characterised by a phosphorus sorption index rather than by fertility.

Nitrate, and why a saturated zone exists#

An aerobic filter does the opposite of what is expected with nitrogen. Ammonium sorbs and is nitrified, and nitrate is an anion with nothing to hold it on negatively charged media surfaces. Only denitrification removes it, and that requires anoxia plus available organic carbon as an electron donor. Hence the design element that does not appear in photographs: an internal saturated storage zone created by an upturned elbow on the underdrain, usually with a wood-chip carbon source. Its effectiveness is residence time in the saturated layer and the length of the interevent dry period — after a long dry spell the zone re-aerates, and the next storm flushes the nitrate that accumulated.

Hydraulics set the ceiling#

A cell treats only the water that passes through it. Too high an infiltration rate cuts contact time to nothing; too low a rate sends flow to the overflow, and untreated bypass outweighs any efficiency achieved on the treated fraction in an annual load balance. The governing parameters are therefore the ratio of cell area to catchment area and the design storm, not species selection. The asset’s life is hydraulic too: silt accumulates in the top few centimetres, forms a low-conductivity crust, and rehabilitation means removing that layer.

Plants do three things, and they are worth naming precisely. Roots and root turnover maintain macropores and resist clogging. Transpiration recovers pore volume between storms, restoring storage. Direct nitrogen uptake into biomass is real but a small share of the load. None of the three makes species choice decisive.

What a cell does not do#

Chloride from de-icing salt is conservative: there is no retention mechanism for it in such a system. Worse, the accompanying sodium displaces sorbed metals by cation exchange, releasing pulses of previously captured metal, and simultaneously disperses media aggregates, lowering conductivity. In a climate with winter salting this is a limitation, not an operating detail.