Generalized Storage-Reliability-Yield Equations for Rainwater Harvesting Systems

Abstract

Although rainwater harvesting (RWH) is gaining popularity as a sustainable water supply source in urban as well as rural areas, estimating required storage remains an important design challenge. This paper develops a robust, yet computationally simple equation for calculating required storage capacity for a RWH system, which is generally applicable in the United States (U.S.). A simulation model with a daily time step and a yield-after-spill algorithm is used to generate empirical Storage — Reliability — Yield (SRY) relationships for RWH systems at 232 U.S., first-order precipitation gaging stations with long daily precipitation records. A regional regression modeling approach is used to combine system parameters (daily yield, collection area, reliability) with climatic variables (e.g. standard deviation of daily rainfall) to predict required storage capacity. Nationwide regression models for fixed reliability cases (80, 90, 95, 98%) demonstrate good fits (R2> 0.95) between model predictions and simulated storage capacities. The fits improve (R2>0.97) when the nation is broken down into smaller, more climatically homogeneous regions.