This study aims to assess the short-term response of groundwater to the main hydro-meteorological variables of drought in a coastal unconfined aquifer. For this purpose, a multiple fuzzy linear regression-based methodology is implemented in order to relate rainfall, streamflow and the potential evapotranspiration to groundwater. Fuzzy regression analysis is recommended when there is a lack of data. The uncertainty of the system is incorporated into the regression coefficients which, in this study, are considered to be fuzzy symmetrical triangular numbers. Two objective functions are used producing a fuzzy band in which all the observed data must be included. The first objective function, based on Tanaka’s model, minimizes the total width of the produced fuzzy band. The second one includes the first while additionally minimizing the distance between the central value of the fuzzy output of the model and the observed value. Validity of the model is checked through suitability measures. The present methodology is applied at the east part of the Nestos River Delta in the Prefecture of Xanthi (Greece), where the observed values of the depth of groundwater level of four wells are examined.

中文翻译：

---

通过模糊线性回归将无水层中水文气象变量与地下水位联系起来

这项研究旨在评估沿海无限制含水层中地下水对干旱的主要水文气象变量的短期响应。为此，为了将降雨，水流和潜在的蒸散量与地下水联系起来，实施了一种基于多重模糊线性回归的方法。当缺乏数据时，建议进行模糊回归分析。系统的不确定性被纳入回归系数中，在本研究中，回归系数被认为是模糊对称的三角数。使用两个目标函数产生一个模糊带，其中必须包含所有观察到的数据。基于田中模型的第一个目标函数使产生的模糊带的总宽度最小。第二个包括第一个，同时另外最小化了模型的模糊输出的中心值和观测值之间的距离。该模型的有效性通过适用性度量进行检查。本方法在Xanthi州（希腊）的Nestos河三角洲的东部应用，在该处检查了四口井的地下水位观测值。