Evaluation of three vegetation treatments in bioretention gardens in a semi-arid climate

Bioretention is a well-established tool to reduce nutrient transport from impervious urban landscapes to sensitive riparian habitat in mesic climates. However, the effectiveness of bioretention is less tested in arid and semi-arid climates. Nutrient retention performance was evaluated in three 10 m(2) bioretention cells with different vegetation communities: (I) an irrigated wetland vegetation community, (2) an un-irrigated upland vegetation community, and (3) a no-vegetation control. Synthetic stormwater was added to each cell to simulate the average annual runoff of precipitation from a 220 m(2) impervious surface in Salt Lake City, UT. A significant amount of phosphate (approximate to 50%) was retained by all treatments during the 12-month study. However, total nitrogen (TN) retention was only achieved in the Wetland and Upland treatments (59% and 22%, respectively), and nitrate retention was only achieved in the Wetland treatment (38%). In contrast, the Upland and Control treatments exported 2 and 9 times more nitrate than was added in the simulated rainfall events. Improved nitrogen retention by the Wetland treatment came at the cost of over 12,000I (3200 gal) of irrigation to sustain the vegetation through the hot, dry summer. We hypothesize that plant uptake and soil microbial communities are driving nutrient retention in bioretention systems, and that increasing net primary production will increase nutrient retention. In water-limited climates, this can be sustainably achieved by either: increasing native upland vegetation densities above naturally expected densities, or, by using gray water instead of municipal water sources to irrigate wetland communities through dry summer periods. (C) 2014 Elsevier B.V. All rights reserved.