MODWT and wavelet coherence-based analysis of groundwater levels changes detection

摘要

This study used a model to capture trends and dominant periods in groundwater level (GWL), streamflow, and precipitation (GSP) data over the Azarshahr plain, Iran. The trend of GSP was predicted to find dominant time scales affecting the trends observed in the datasets. This study used maximal overlap discrete wavelet transform and Mann-Kendall trend tests to analyze and detect trends in monthly, seasonally based, and annual data from three GWL, one streamflow, and one precipitation gauges in Azarshahr plain during 1968-2015. The present research demonstrated that at the monthly, seasonal, and annual time scales, the trends in stream flow and GW data were significant and downward. Results indicated that this significant negative trend appeared after 1971 in GWL. In addition, the 2-month and 16-month (individual months), 2- and 8-month components were dominant periods at the monthly time scale, and the 6-month, 12-month, and 6- and 12-month components were dominant periods at the seasonal time scale. The 2-year and 4-year components were dominant periods at the annual time scale. Results suggested that climate teleconnections might affect the fluctuations of the GSP process. Wavelet coherency was used to study the correlation and interaction of the variables, where no correlation was detected.