Premium
A non‐Gaussian multicomponent model for river flow
Author(s) -
Vandewiele G. L.,
Dom A.
Publication year - 1989
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr025i003p00397
Subject(s) - base flow , series (stratigraphy) , surface runoff , gaussian , flow (mathematics) , fourier series , mathematics , statistical physics , geology , mathematical analysis , physics , geometry , drainage basin , geography , paleontology , ecology , cartography , quantum mechanics , biology
In order to model satisfactorily sharp rises and slow decreases appearing in any pure runoff time series with a small enough time base, exponentially decaying recession and base flow terms are first subtracted from total runoff. The remaining part is deseasonalized by a truncated Fourier series multiplicator and then modeled as a second‐order Markov stationary series. The use of non‐Gaussian distributions allows the modeling of zero contributions, which correspond to dry spells. An example has been optimized and simulated and the observed and simulated series have been compared, even on hydrologically interesting properties such as return periods, which were not modeled explicitly. The results were satisfactory.