Large-scale climate variability controls on climate, vegetation coverage, lake and groundwater storage in the Lake Urmia watershed using SSA and wavelet analysis

Abstract

Lake Urmia has shrunk by 88% since 1995 and is an outstanding example of an environmental tragedy in the Middle East, and the lake plays a critical role in the environment, economics, and society in the north-western part of Iran. It has been hypothesized that the drying of Lake Urmia has caused by climate variation and a climate-derived increase in droughts. Therefore, it is necessary to understand the teleconnections between the interannual to multidecadal climate variability and Lake Urmia because of the tangible implications for water resource management and policy decisions in the region. In this study, we use singular spectrum analysis (SSA), wavelet coherence analysis, and lag correlation calculations to analyze and quantify the impacts of the El Niño Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO), and Atlantic Multidecadal Oscillation (AMO) on hydro-climate variables of precipitation, temperature, lake level, groundwater fluctuations, soil moisture, vegetation coverage, and insolation clearness index in the Lake Urmia watershed. Overall, the results indicate that climate oscillations attributed to the Pacific Ocean (i.e., ENSO and PDO) have a more powerful influence than Atlantic Ocean oscillations (NAO and AMO) on the variability in the water level of Lake Urmia as well as on other hydro-climate variables, except for temperature that appears influenced by the Atlantic Ocean oscillations, particularly AMO. PDO is the first dominant mode of variability in all the hydro-climate variables (63.46% on average), except for the temperature. Overall, the wavelet coherence analysis findings indicate relatively greater PDO influence than ENSO on variability in the precipitation, soil moisture, vegetation coverage, and insolation clearness index. Furthermore, hydro-climate variables in the area have a relatively highest statistical correlation with PDO (0.69 on average, ranging from 0.54 to 0.78) compared to ENSO, NAO, and AMO. Moreover, a moderate coherence between PDO and the groundwater levels in most adjacent aquifers has occurred at the >8-year period from ~1980 to 2015. In general, the hydro-climate variables statistically have a weak lag correlation with NAO (0.19 on average, ranging from 0.13 to 0.24). AMO comprises the first mode variability in temperature (71.77%), and its coherence with temperature is moderate (~0.5) at >16-year period for the time earlier than 2000. The lag correlation between AMO and temperature (0.66) is relatively near strong. These findings have important implications for decision-makers and scientists to improve water resources planning and operations in Lake Urmia under future climate uncertainty.

Keywords: Groundwater; Hydro-climate variables; Lake Urmia; Large-scale climate variability; SSA; Wavelet coherence analysis.