Premium
Estimating parameters of groundwater recharge model in frequency domain: Karst springs Jadro and Žrnovnica
Author(s) -
Jukić Damir,
DenićJukić Vesna
Publication year - 2008
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.7057
Subject(s) - groundwater recharge , karst , autocorrelation , exponential function , surface runoff , hydrology (agriculture) , groundwater , transfer function , scale factor (cosmology) , scale (ratio) , time domain , environmental science , parametric statistics , frequency domain , mathematics , aquifer , geology , statistics , geotechnical engineering , computer science , mathematical analysis , geography , physics , ecology , metric expansion of space , dark energy , engineering , biology , paleontology , cosmology , quantum mechanics , computer vision , electrical engineering , cartography
The spectral method is based on the assumption of exponential forms of autocorrelation functions of the input and output time series. The results of practical applications of the spectral method on several karst springs in Croatia show that this assumption cannot always be reliably applied to the time series of rainfall and spring discharge, especially if short time periods are analysed. For this reason, the method is modified. The scale factor between the transfer function of total rainfall and the transfer function of groundwater recharge is considered as an additional model parameter which is determined in the optimization procedure. In addition to the theoretical background, this paper also presents the results of application of the modified spectral method on the springs Jadro and Žrnovnica located in the Dinaric karst area in Croatia. The parameters of groundwater recharge model are determined separately for both the springs and the obtained values are discussed and compared with the results of previous investigations. The values of the scale factors determined in the optimization procedure are compared to the values calculated using the theoretical equation. The comparison shows that the theoretical equation underestimates the optimal values of the scale factor. The underestimations are especially evident for periods of one year. Finally, the characteristics of rainfall–runoff relations of two springs are compared using the identified transfer functions and the results of simulations of the periodogram of observed discharge by the parametric periodogram. Copyright © 2008 John Wiley & Sons, Ltd.