EFFECT OF MONITORING TIME INTERVAL ON THE RESULTS OF STOCHASTIC ANALYSIS OF KARST WATER RESOURCES

Abstract

Data monitoring is important in the study and analysis of hydrological behaviors of karsticsystems. Monitoring shows the various regimes of groundwater flow (laminar or turbulent), depletion,filling, filtration and karstification degree of karst aquifers. In recent years, analyses of these data hadconsiderable advances using stochastic time series analyses. An important problem related to theseanalyses is the constraint of financial and technical human support. If we can increase the time intervals,the costs are lowered and development of new monitoring networks will be possible. In this research, theeffect of monitoring time intervals on the results of time series analysis and calculated hydrologic delaytimes are studied, using bivariate analyses in spectral domain. The first (shorter) and the second (longer)delay times are related to the flow of groundwater through the larger and smaller fractures of the karsticsystem, respectively. Daily groundwater level data of three piezometers and the discharge of a spring inthe Maharlu karst basin in Iran (52􀁄12′ to 53􀁄28′E and 29􀁄1′ to 30􀁄6′N ) were acquired and then sixdifferent time series having different time intervals (1, 3, 7, 10, 14 and 30 days) were extracted andanalyzed. The results show that the variation of the computed stochastic parameters with increasing timeintervals is unpredictable, but the variation of time (period) corresponding to the frequency at the peak ofamplitude function (curve) shows a linear and predictable relationship, which is unique for every kind ofkarstic aquifer having different degrees of karstification.