S.M. Papalexiou, and P. Kossieris, Theoretical documentation of model for synthetic hyetograph generation, DEUCALION – Assessment of flood flows in Greece under conditions of hydroclimatic variability: Development of physically-established conceptual-probabilistic framework and computational tools, Contractors: ETME: Peppas & Collaborators, Grafeio Mahera, Department of Water Resources and Environmental Engineering – National Technical University of Athens, National Observatory of Athens, 97 pages, May 2014.
The simulation of flood events necessitates the simulation of the rainfall over small times scales (e.g., smaller than the monthly scale). Nevertheless, rainfall modelling at small time scales is not simple as rainfall at these scales is an intermittent process and exhibits large variability in its statistical-stochastic characteristics. In this context, a flexible multivariate framework of stochastic simulation of rainfall was developed that can be applied to a large range of times scales. The proposed methodology is based on the cyclostationary multivariate autoregressive model of order 1 (PAR1), while the intermittency characteristics were reproduced using a novel transformation structure. The methodology was verified in the basin of Boeotikos Kephisos and it was verified that the model preserves satisfactorily the basic statistical characteristics of daily rainfall, including the probability dry, as well as the autocorrelation and the cross correlation structures. As an alternative for the generation of synthetic hyetographs the stochastic model known as the rectangular pulse Bartlett-Lewis model is presented. This model is widely accepted for the single-variate simulation of rainfall at fine time scales and in continuous time. The implementation was done in R programming environment and is available through the computer package HyetosR.
Full text (3599 KB)
Related project: DEUCALION – Assessment of flood flows in Greece under conditions of hydroclimatic variability: Development of physically-established conceptual-probabilistic framework and computational tools