Avaliação do funcionamento de um dispositivo de biorretenção: monitoramento e modelagem

Abstract

The bioretention is a kind of trap to collect, store, infiltrate and treat urban full water runoff, consisting of a mixture of high hydraulic conduction of sand, soil and organic matter, surrounded by native, generally less permeable soils. It has become popular in many parts of the world, but so far, it has not been implemented as a common practice in Brazil. However, there are some bioretentions installed in certain Brazilian universities, built in real scale that can provide quantitative and qualitative performance data. This work is a study on a bioretention cell constructed at the Federal University of Alagoas. The device is prismatic in shape with approximately 400 cm long, 160 cm wide and 100 cm deep and, divided into 3 layers, an upper part with a mixture of black earth and humus, an intermediate layer satisfies by gravel and last layer contains sand. In this context, initially, a comparison was made on different forms of device design between a cell constructed at the Federal University of Alagoas and others designed for inside Brazil. In summary, as bioretentions do not follow a pattern, both in Brazil and in other countries. The only ones was noticed on all the devices analyzed for a root area layer and a layer of internal water storage. It is concluded that in spite of the little scientific production is not with this object of study, it is possible to introduce the practice in manuals of drainage as the option of handling of rainwater. Then a performance analysis was performed from the update of fully absorbed events without exceeding a maximum storage capacity. For this, it has 18 months of quantitative monitoring where phases of the hydrological cycle are observed, such as precipitation, infiltration and surface runoff. Results showed that the storage layer absorbed 89% of the events and of the 11% found extravasation records. However, a bioretenation was efficient in reducing the flow generated by its contribution area and highly recommended its installation in residential lots. And finally, a simple low-data-demand model is applied to represent a real-scale bioretenation. For this, the Puls Method is used as a quantitative bioretention simulation tool. The results showed good Nash-Sutcliffe and R² coefficients in the calibration phase. However, the parameter infiltration rate presents a great variation between the coefficients. However, the Puls model proved to be efficient when representing a bioretention system, but it was deficient in the simulation of 1 event in the single reservoir mode and 2 events in the double reservoir mode.