Studying the Changes in the Hydro‐Meteorological Components of Water Budget in Lake Urmia

Abrupt changes in the Lake Urmia water level have been addressed in many studies, and yet the link between the water level decline and hydro‐meteorological variables in the basin is a major topic for debate between researchers. In this study, a set of data‐driven techniques is used to investigate the components of the water budget in Lake Urmia. Then, the rate of monthly depth differences (DD), precipitation (P), evaporation (E), potential groundwater head (G), and streamflow (Q) time series between 1974 and 2014 are used in the analysis. Several scenarios and strategies are developed by considering the major changes in the year‐2000, which is believed to be the initiation of the hydrological encroachment in the basin. Simple water budget (WB), dynamic regression (DR), and symbolic regression (SR) techniques are used to simulate the DD with consideration to P, E, G, and Q. Alternatively, the effect of the year 1997 as the potential base‐line for the initiation of significant meteorological trends in the basin is investigated. Conducted analysis showed that the DR models of an autoregressive moving average together with multiple exogenous inputs provide an approximate R2: 0.7 as the best alternative among the selected models. It is shown that the Q and G depict abrupt changes compared to the P and E, while either the year 1997 (climate effect) or the year 2000 (encroachment effect) is considered as the baseline in the study. Plain Language Summary We have investigated the main scenarios regarding the Lake Urmia water level decline. For this, a conceptual water budget approach together with sophisticated data mining techniques and mathematical models are used. Then based on the best fit model results and prior details obtained in data analysis we have questioned the credibility of different hydrological scenarios. Key Points Lake water budget Mathematical model Hydrological data analysis