A system comprised by a UASB (Upflow Anaerobic Sludge Blanket) reactor followed by two horizontal subsurface-flow constructed wetlands in parallel was evaluated for the treatment of the wastewater generated in the city of Belo Horizonte, Brazil (50 inhabitants each unit). One unit was planted (Typha latifolia) and the other was unplanted. Influent and effluent samples were collected for a period of seven months. The systems were able to produce final effluents with low concentrations of organic matter and suspended solids, but showed not to be efficient in the removal of nutrients. Mean effluent concentrations for the planted and unplanted units were, respectively: BOD: 15 and 19 mg/L; COD: 42 and 64 mg/L; TSS: 3 and 5 mg/L; TN: 27 and 33 mg/L; N-NH(3): 25 and 29 mg/L; P Total: 1.2 and 1.5 mg/L. The planted wetland presented effluent concentrations and removal efficiencies significantly (Wilcoxon matched-pairs test, 5% significance level) better than the unplanted unit for most constituents. The study shows that horizontal subsurface-flow constructed wetlands can be effectively used as a post-treatment option for the effluent from UASB reactors.

Ammonia and phosphorus removal were evaluated over four years monitoring of an experimental wastewater treatment plant in Brazil: initially a UASB reactor (real scale) + three shallow polishing ponds (pilot scale), afterwards a UASB + submerged aerated biofilter (BF) (real scale) + the same pond series. Shallow ponds (h < 0.90 m) provided considerable ammonia removal, but phosphorus removal was somewhat poorer. An equation was derived for the estimation of ammonia pond effluent, based on the same assumptions of the well known Pano and Middlebrooks model. According to the experimental data, an effluent suitable for aquaculture should be achieved with retention time of 25 days.

This paper presents the results of a comparison of the performance of two treatment systems operating in parallel, with the same influent wastewater. The investigated systems are (i) UASB + three polishing ponds in series + coarse filter (200 population equivalents) and (ii) UASB + subsurface flow constructed wetlands (50 population equivalents). Two wetland units, operating in parallel, were analysed, being one planted (Typha latifolia) and the other unplanted. The systems were located in Belo Horizonte, Brazil. The wetland systems showed to be more efficient in the removal of organic matter and suspended solids, leading to good effluent BOD and COD concentrations and excellent SS concentrations. The planted wetland performed better than the unplanted unit, but the latter was also able to provide a good effluent quality. The polishing pond system was more efficient in the removal of nitrogen (ammonia) and coliforms (E. coli). Land requirements and cost considerations are presented.

Resumo Apresenta-se uma síntese de trabalhos de pesquisa conduzidos ao longo de quatro anos em tratamento de esgotos sanitários, em um sistema UASB + BF + lagoas de polimento e utilização de efluentes em fertirrigação e produção animal. A irrigação de forrageiras (Brachiaria e milho hidropônico) com efluente do sistema UASB +BF e seu fornecimento para alimentação de caprinos e bovinos não resultou em riscos reais à saúde animal nem potenciais à saúde humana. O sistema de lagoas de polimento mostrou-se eficiente na produção de efluentes, de acordo com as recomendações da OMS, para piscicultura e irrigação, logrando produção de hortaliças e peixes (tilápia do Nilo) de qualidade apta ao consumo. O potencial fertilizante e nutricional dos efluentes tratados demonstrou a possibilidade de sua aplicação em atividades produtivas com economia de insumos (adubo e ração).