The Disaggregation Model via Non-Parametric Approach

Abstract

Flow disaggregation models, which are one of the stochastic generation techniques, play a crucial role in the planning, design, and operation of water resource management systems and related projects. One distinguishing feature of these models is their ability to address the issue of missing observed data and compensate for it. They also enable the rescaling of data from a higher temporal level to a lower temporal scale. Data at lower temporal scales are typically required to address hydraulic and operational design problems in water resource projects. There are two main approaches to disaggregation flow data: the parametric approach and the non-parametric approach.One of the advantages of the disaggregation model is its ability to distribute flow data values from a key station to several sub-stations, both temporally and/or spatially, while preserving the basic statistical properties of the time series obtained from the model (mean, standard deviation, minimum, maximum, and correlation coefficient) for the observed data.In the current study, a non-parametric approach was used for the purpose of disaggregation approach. It is assumed that there is aggregated discharge data at a key station, and this data will be disaggregated into a corresponding series of discharges temporally and spatially at sub-stations that are statistically similar, using the SAMS 2010 platform program (Stochastic Analysis, Modeling, and Simulation). Annual and monthly discharge data for five stations measuring discharges on the Tigris River System in Iraq were used, including the Mosul Dam station on the Tigris River, the Asmawah station on the Khazir River, the Askiklik station on the Upper Zab, the Dibs Dam station on the Lower Zab, and the Baiji station on the Tigris River, covering a time span of twenty-three years. The statistical results of the disaggregation approach were compared with their observed counterparts and showed good agreement in most years and months and for all stations. Based on this, the method is recommended disaggregation of the data when decisions required water management strategies in these regions.