Temporal scale effect analysis for water supply systems monitoring based on a microcomponent stochastic demand model

Water demands are the main random factor that conditions flow variability within drinking water supply systems. The importance of using high-resolution demands in distribution mains is already well-known, but there is little knowledge of how the temporal scale (i.e. sampling frequency) affects the ability of a metering or monitoring system to explain network performance. The aim of this paper is to analyse the variability (i.e. information) that is lost because of not using a more frequent sampling rate to characterize water demands. For such purpose, a novel analytical approach based on a conceptualization of the microcomponent-based SIMDEUM model (SIMulation of water Demand, an End-Use Model) is presented. This methodology provides the statistical properties of water demands over different sampling frequencies. It is here applied to Benthuizen case study to further explore the effect of temporal and spatial scaling laws under realistic conditions. Results are of major importance for monitoring design, as they highlight the need for properly combining measurements with different levels of resolution. Moreover, they enable to assess the impact of the sampling selection on the potential characterization level of monitored demands within urban water modelling applications