Beyşehir Lake is the largest freshwater lake in the Mediterranean region of Turkey that is used for drinking and irrigation purposes. The aim of this paper is to examine the potential for data-driven methods to predict long-term lake levels. The surface water level variability was forecast using conventional machine learning models, including autoregressive moving average (ARMA), autoregressive integrated moving average (ARIMA), and seasonal autoregressive integrated moving average (SARIMA). Based on the monthly water levels of Beyşehir Lake from 1992 to 2016, future water levels were predicted up to 24 months in advance. Water level predictions were obtained using conventional time series stochastic models, including autoregressive moving average, autoregressive integrated moving average, and seasonal autoregressive integrated moving average. Using historical records from the same period, prediction models for precipitation and evaporation were also developed. In order to assess the model’s accuracy, statistical performance metrics were applied. The results indicated that the seasonal autoregressive integrated moving average model outperformed all other models for lake level, precipitation, and evaporation prediction. The obtained results suggested the importance of incorporating the seasonality component for climate predictions in the region. The findings of this study demonstrated that simple stochastic models are effective in predicting the temporal evolution of hydrometeorological variables and fluctuations in lake water levels.

Groundwater quality modelling plays an important role in water resources management decision making processes. Accordingly, models must be developed to account for the uncertainty inherent in the modelling process, from the sample measurement stage through to the data interpretation stages. Artificial intelligence models, particularly fuzzy inference sys-tems (FIS), have been shown to be effective in groundwater quality evaluation for complex aquifers. In the current study, fuzzy set theory is applied to groundwater-quality related decision-making in an agricultural production context; the Mamdani, Sugeno, and Larsen fuzzy logic-based models (MFL, SFL, and LFL, respectively) are used to develop a series of new, generalized, rule-based fuzzy models for water quality evaluation using widely accepted irrigation indices and hydro-logical data from the Sarab Plain, Iran. Rather than drawing upon physiochemical groundwater quality parameters, the pre-sent research employs widely accepted agricultural indices (e.g., irrigation criteria) when developing the MFL, SFL and LFL groundwater quality models. These newly-developed models, generated significantly more consistent results than the United States Soil Laboratory (USSL) diagram, addressed the inherent uncertainty in threshold data, and were effective in assessing groundwater quality for agricultural uses. The SFL model is recommended as it outperforms both MFL and LFL in terms of accuracy when assessing groundwater quality using irrigation indices.