data preprocessing

Accurate monitoring of groundwater levels and estimating their fluctuations in the future is of importance, especially in arid and semi-arid areas. regarding the high capabilities of AI-based models in the modeling of hydrologic phenomena, this research used the MP5 decision tree, in combination with the wavelet transform, to predict groundwater level fluctuations of the Kerman-Bagheyn plane. To develop the wavelet-decision tree (W-M5p) hybrid model, the wavelet transformation outputs were exported to the MP5 as inputs. Several statistical criteria, including coefficient of correlation (R), agreement index (Ia), and scattering index (SI), were used to evaluate the performance of the hybrid model compared to the single model. The results indicated that, even when the inputs of the hybrid model includes only the meteorological data from a synoptic station (the water level of previous periods were not used in the analysis), that the performance of the WM5P was superior to the single model in the prediction of groundwater fluctuations. The WM5P model with three months of forecast horizon with the Coif4 wavelet and decomposition level of 6 reduced the SI value from 0.6394 to 0.0181 and, at the same time, increased the Ia from 0.6898 to 0.9998. Consequently, the Coiflet4 with decomposition levels of 5 and 6 was the most efficient wavelet in the hybrid model for reliable estimation of the Kerman-Bagheyn plane groundwater level.

Precipitation forecasting is essential in maintaining, managing, allocating, and distributing water resources, determining the volume of water facilities, supplying the crops' water requirement, and determining the amount of erosion and sedimentation. This research aimed to investigate the performance of artificial neural networks in predicting monthly rainfall based on climatic and geographic information such as minimum and maximum temperature, minimum and maximum humidity, precipitation, latitude and longitude, and altitude above sea level in 23 stations of the Fars province. The results on levels 5, 10, and 18 of input data showed that the model accuracy in estimating the monthly rainfall increases with the increment in the number of inputs. The most accurate neural network model is in the rank normalization method with one hidden layer, and also, the best network structures are 5-25-1, 10-30-1, and 18-20-1, respectively. The results indicated that the neural network with 18 inputs has the smallest MSE=0.032 and the highest R=0.62. The best normalization method is the ranking method with an optimal neural network of one layer and 5-25-1 structure, the Levenberg-Marquardt training algorithm, and the sigmoid tangent stimulus function. Therefore, by using an artificial neural network (with 18 introduced inputs), it is possible to predict the amount and distribution of monthly rainfall in a wide area with acceptable accuracy. This issue plays a very decisive role in the management and planning of drinking and agricultural water resources; by taking into account these forecasts, future policies can be planned to optimize costs and maximum productivity facilities.

Solar radiation arriving to the land surface is one of the major variables that is used in projects and hydrological, agricultural, meteorological and climatic models. In this study, the functionality of the principal component analysis (PCA) and the entropy theory (EN) for determination  of inputs to multilevel perceptron artificial neural network (MLP), artificial neural network, radial basis function (RBF), support vector machine (SVM)and genetic programming (GEP), was investigated for estimation of solar radiation at two stations (Kerman and Mashhad) during 1984-2005 and 1980-2004 periods, respectively. The average temperature, mean water deficit pressure, minimum temperature, maximum temperature, sunshine, relative humidity, dew point temperature, hourly vapor pressure, horizontal visibility and water content were selected as inputs of pre-processing methods. The obtained results  in Kerman station showed that the ENT-MLP model with RMSE=38.36 (Mj /m2) and R2 = 0.93 have had the best performance. Also in Mashhad station, PCA-MLP model with RMSE=79.75 (Mj / m2) and R2 = 0.77 had the best performance. In general, the both pre-processing  principal component analysis and entropy theory were recognized  as the proper methods for determination  of estimating models input to estimate  solar radiation.