Effect of mixed-grass cover and native-soil filter on urban runoff quality.

Abstract

A grass-covered soil filter of native calcareous loam, 200-feet long, 4-feet wide and 5-feet deep, was tested for effectiveness as a water-quality treatment for Tucson urban storm runoff. Water was pumped from Arcadia Wash and applied to the filter in four trials in Fall 1971. Inflow and outflow volumes were measured, sampled and analyzed for important water-quality variables. For grass and grass-soil filtration respectively, the following maximum percent reductions, compared to untreated runoff, occurred: For chemical oxygen demand (COD), 19 and 88; for suspended solids, 34 and 99.6; for volatile suspended solids, 26 and 97; for turbidity, 97 and 98; for total coliforms, 84 and 98; and for fecal coliforms, 50 and 98. Grass-soil filtration, during the four trials, produced water too saline for most uses. After the initial stabilization period, grass-soil filtration was more effective than grass filtration in water-quality improvement. Grass soil filtration had, with time, increasing COD and salt concentrations in the early part of each trial, and decreasing infiltration rates. Grass maturity and soil compaction, following construction of the filter, initially increased efficiency of the water-quality improvement process. Grass filtration upgraded cool-season urban runoff for recreation, irrigation, artificial groundwater recharge, fisheries and wildlife, except in the initial part of some trials and in the grass establishment period when COD, volatile suspended solids and coliforms increased. Chlorination of treated runoff is necessary, particularly for warm-season flows, when recreational water use is anticipated.