Utilização de terras úmidas no tratamento de águas superficiais poluídas: influência do tipo de leito e de macrófita.

Abstract

This work, carried out from January 2000 to September 2002 at the Federal University of Campina Grande, Campina Grande City (7o 13' 11" South, 35° 52' 31" West, 550 m above m.s.l), Paraíba State, northeast Brazil, aimed the analysis of a natural wetland polluted with sewage and the performance of two constructed wetland systems, in pilot-scale, fed with the efíluent of the natural wetland. The natural ecosystem significantly reduced organic matter (between 74 and 87% BOD and from 21 to 57% COD) and fecal indicator microorganisms (thermotolerant coliforms between 92.63 and 96.78 percent, fecal streptococci between 92.91 and 97.74 percent, coliphage between 91.95 and 96.47 percent and bacteriophage between 66.83 and 88.61 percent) but the quality of its efíluent did not comply with the standards for both discharging into receiving bodies and reusing in irrigation, being necessary to be treated. Firstly, constructed wetlands applied for such an objective, comprised 15 experimental units using gravei as substrate media having each set of 5 a hydraulic retention time of 5, 7 or 10 days, respectively, being 4 vegetated with Typha sp. and 1 without vegetation. Controls and vegetated units did not show significant differences in removing organic matter but performed differently in terms of nutrient (total phosphorus between 74 and 87 percent in vegetated units and between 1 and 37 percent in control units, soluble orthophosphate between 86 and 100 percent in vegetated units and less than 31 percent in controls, ammonia within the ranges 98- 100 percent in vegetated units and 71-98 percent in controls) and fecal contamination indicators (thermotolerant coliforms between 98.06 and 99.97 percent in vegetated units and between 95.71 and 98.61 percent in control units, fecal streptococci from 94.47 to 99.93 percent in units with Typha and from 85.56 to 96.09 percent in controls, coliphage within the ranges 95.69-99.92 and 86.62-99.07 percent, respectively, and bacteriophage between 96.76 and 99.88 percent in vegetated units and from 87.71 to 97.69 percent in controls) removais. The second constructed wetland system was made up of 36 units being 18 operated under a hydraulic retention time of 5 days and the other ones with 10 days. For each hydraulic retention time 9 experimental units were filled with gravei and 9 with sand; in each group 3 units were vegetated with Typha sp., 3 vegetated with rice (Oriza sativa L.) and 3 were kept as controls. The association of gravei and Typha sp. promoted the highest removais for total phosphorus (65-87%), soluble orthophosphate (71-97%), ammonia (82-97%), BOD (80-95%) and turbidity (92-97%) but was responsible for the increase of electrical conductivity (40- 102%), total alkalinity (24-75%), bicarbonate (29-124%), sodium (96-248%), hardness (30- 96%), calcium (33-132%), magnesium (34-120) and chloride (62-233%). The association of sand with both types of vegetation caused the highest removais of fecal contamination indicators reaching 99.22-99.86 percent for thermotolerant coliforms and 98.02-99.43 percent for fecal streptococci. In both experiments on constructed wetlands the hydraulic retention time of 10 days showed a better performance than the other ones and in ali cases vegetated units produced effluents with quality for discharges into surface water bodies classified according the brazilian standards classification as adequate for water supply after a conventional treatment, irrigation of trees, fodder and cereal and animal drinking.