Efeito dos atributos físicos e hidráulicos do solo sobre o comportamento de práticas de desenvolvimento de baixo impacto em sistema de drenagem

Abstract

Low Impact Development (LID) practices are presented as an alternative for seeking a potential balance in the hydrological cycleprocesses in urban environments, as they promote water infiltration. These structures can be modelled and their hydrological performance simulated. Rossman and Huber (2016a) provide a recommended values range for the physical and hydraulic properties of their soil layer, however, there is uncertainty regarding these values, as they are soil-specific. Therefore, the objective of this study was to model, simulate, and assess the performance the LID practices bioretention cell (BC) and rain garden (RG), both including a soil layer. These practices were spatially distributed in relation to the main drainage system conduits, occupying 10% of the impervious area of the drainage system sub-catchments of the Universidade Federal de Lavras (UFLA).For modelling, the soil layer characteristics of these LID practices were defined according to the recommendations of Rossman and Huber (2016a), as well as based on two local soil types, Oxisol (LV) and Inceptisol (CX). The physical and hydraulic attributes of these soils were determined both in laboratory and by the pedotransfer function (PTF) Splintex 2.0. The parameters required for modelling the soil layer included: soil-saturated hydraulic conductivity, field capacity, permanent wilting point, porosity, percolation decay constant, and wetting front suction head.All scenarios were evaluated by continuous temporal simulationas well as design storm events with return periods of 10, 20, and 50 years. Under the conditions analyzed; it was found that LID practices were ineffective in controlling flow for design storms with return periods greater than 10 years.Overall, the performance of the LID practices varied depending on the soil type. For RG, simulations using the recommended values by Rossman and Huber (2016a) or those obtained via PTF are suitable for LV. However, for CX soils, the use of PTF is not recommended. For BC, simulations using recommended valuesby Rossman and Huber (2016a) or those obtained from PTF are only appropriate when LV is the soil in use. For CX soils, simulations should be based only on values obtained from laboratory measurements.