A NEW HYDROLOGICAL APPROACH TO SOURCE EVALUATION FOR PUBLIC PRIVATE PARTICIPATION PROJECTS FOR URBAN DRINKING WATER

Vasudev Shirahatti

Abstract: Application of 2- site model for generating synthetic stream flows is discussed. Historical simulations based on insufficient field data may often be not reliable due to small sample bias. In the case of public Private Participation (P-P-P) projects rigorous testing of reservoir yields is critically important. The 2-site model applied to a field situation wherein main river having 29-years of daily flow data is treated as key river and its main tributary having only 14 -years concurrent data is treated as follower. Leveraging the lag-zero cross correlation that must exist between two adjacent catchments falling under the same hydrometeorological conditions, record of follower river was extended through application of innovative 2-site model. Study basin lies in tropical wet-dry climate of peninsular India as such high skewness and serial lag properties characterise historical time series. These higher model moments have great economic significance in reservoir sizing for sustaining commercial yields and this calls for extreme care in modelling. In the model high sample skewness was treated by Wilson-Hilfertytransformation in the generating equations for generating longer synthetic time series of 100 items as may be necessary for reservoir simulation purposes. Preserving all historical statistics including higher order moments like skew coefficient in generated samples is of priority in commercial applications. Paper discusses novel evaluation technique devised for evaluating model generated samples

Keywords: Historical simulations, Public Private Participation Projects, synthetic simulations, commercial yield, reservoir simulation, 2 -site model, peninsular Indian rivers, skewness coefficient, serial correlation coefficient, cross correlation coefficient, model evaluation criteria

DOI: https://doi.org/10.15623/ijret.2018.0704010