The performance of rainwater harvesting systems in the context of deep uncertainties

Rainwater harvesting systems (RHS) has been a relevant alternative of water supply in urban areas facing increasing water demand associated to limited water availability. However, previous work has showed that the performance of these systems is highly affected by climate data such as precipitation (well characterized uncertainties). The present study aims to assess the impact of system parameters (water demand, tariffs, storage volume, collect area) named here as deep uncertainties factors influence the feasibility of RHS. So, performance criteria were defined such as Percentage of Satisfied Demand - PSD, Reliability - REL, Percentage of Rainwater Harvesting - PRH, Net Present Value - NPV, Net Present Value Volume - NPV and Benefit Cost Rate - BCR for different scenarios that incorporate uncertainties in precipitation regime, water tariff, discount rate and increase of operating costs rate. The simulation of the RHS performance considered eight categories of residential buildings according to representative water consumption (ranging from 4.748 to 44.673 m³/month) and two characteristic catchment areas for each of the four group of demands (ranging from 60 to 400 m²) in the city of Rio Verde located in the central of Brazil. An ensemble of 1000 scenarios was defined using the Latin Hipercube Sampling (LHS) method and booststrapping resampling (in the case of precipitation). Then, it was evaluated how different scenarios affected each indicator and if uncertainties from some of the parameters have a greater impact on the performance criteria. Results showed low influence of precipitation scenarios on the performance criteria, maybe due to the sampling method that did not generate significant variability. For the elements with deep uncertainty, the relationship among the water tariff and the discount rate readjustments with NPV was confirmed. Thus, the importance of evaluating these elements carefully to achieve rainwater harvesting systems projected performance was confirmed.