Simulating the effect of rainwater harvesting on flood mitigation: the case of Asunción, Paraguay

This work uses computer simulations to assess the effect of rainwater harvesting on the mitigation of flash floods in Asunción, Paraguay, where local rainfalls frequently cause damages, disruptions, and personal harm. Minute-by-minute rainfall data available for the year 2017 were used to select the storms for this study. Terrain, hydrographic, land cover, buildings information, and street maps were retrieved from various freely available sources. One of the hydrographic basins in the municipality of Asunción was chosen for analysis. Terrain and buildings data were merged so that the simulations predict how the built structures divert water flow. An effective rainfall equation was derived to account for the collective effect of thousands of rainwater harvesting systems (RWHS), which were assumed to be installed under different scenarios. The HEC-RAS software (Hydrologic Engineering Center River Analysis System, U.S. Army Corps of Engineers) was used. Three specific locations within the simulated basin were selected for a detailed study. Two peak flow rates were observed in most scenarios: the first of them is the immediate outcome of the rainfall and the second results from the downhill water flow. According to the simulations, the RWHS delay the largest peak flow rate by a few minutes (from 3 to 21 min, depending on the case), which might guide the possible deployment of early warning systems. In a scenario of RWHS with large tanks, installed on every roof – taken as the upper limit of what could be accomplished – reductions of up to 52.2% and 73.3% in the first and second peak flow rate, respectively, were calculated under a heavy rainfall. Under a moderate rainfall, these values were 45.5% and 78.1%. If it is considered that only 50% of the roof area is connected to RWHS, with smaller storage tanks, the peak flow rate reductions are of up to 17.7% (first peak) and 17.8% (second peak) under a heavy rainfall. Under a moderate rainfall, the corresponding values were 23.1% and 40.3%. The overall conclusion is that the effect of RWHS on flood mitigation, as measured by the peak flow rate, is highly dependent on the rain intensity, location, and RWHS features. A limitation of this study is its focus on a single water basin within Asunción’s metropolitan region. However, applying the established methodology to all water basins in a region of interest should provide information to support the decision-making process of policy makers at local and national levels.