Annual precipitation forecast for west, southwest, and south provinces of Iran using artificial neural networks

Abstract

Rainfed agriculture plays an important role in the agricultural production of the southern and western provinces of Iran. In rainfed agriculture, the adequacy of annual precipitation is considered as an important factor for dryland field and supplemental irrigation management. Different methods can be used for predicting the annual precipitation based on climatic and non-climatic inputs. Among which artificial neural networks (ANN) is one of these methods. The purpose of this research was to predict the annual precipitation amount (millimeters) in the west, southwest, and south of Islamic Republic of Iran with the total area of 394,259 km 2 , by applying non-climatic inputs according to the long-time average precipitation in each station (millimeters), 47.5 mm precipitation since the first of autumn (day), t 47.5 , and other effective parameters like coordinate and altitude of the stations, by using the artificial neural networks. In order to intelligently estimate the annual amount of precipitation in the study regions (ten provinces), feedforward backpropagation artificial neural network model has been used (method I).To predict the annual precipitation amount more accurately, the region under study was divided into three sub-regions, according to the precipitation mapping, and for each subregion, the neural networks were developed using t 47.5 and long-time average annual precipitation in each station (method II). It is concluded that neural networks did not significantly increase the prediction accuracy in the study area compared with multiple regression model proposed by other investigators. However, in case of ANN, it is better to use a structure of 2-6-6-10-1 and Levenberg-Marquardt learning algorithm and sigmoid logistic activation function for prediction of annual precipitation.

An ensemble forecast of semi‐arid rainfall using large‐scale climate predictors

“…These results are in agreement with those of Fallah-Ghalhary et al (2010), who used synoptic patterns including the sea-level pressure, temperature difference, sea-level temperature, relative humidity and sea-level pressure difference for predicting spring rainfall. These results are further compatible with those of El-Shafie et al (2011), Farokhnia et al (2011, Azadi and Sepaskhah (2012), Sanikhani and Kisi (2012)…”

“…(), Farokhnia et al . (), Azadi and Sepaskhah (), Sanikhani and Kisi () and Choubin et al . () who reported the superiority of the ANFIS for rainfall prediction.…”

Multiple linear regression, multi-layer perceptron network and adaptive neuro-fuzzy inference system for forecasting precipitation based on large-scale climate signals

“…So, MLR is not capable enough to predict the wet years and droughts. Azadi and Sepaskhah (2012) suggested that the neural networks did not significantly increase the prediction accuracy compared with the MLR. Dastorani et al (2010) concluded that potential of ANN in dry land precipitation prediction is almost same with the ANFIS model.…”

“…Dahamsheh and Aksoy (2009) suggested that the ANNs were slightly better than the multiple linear regression (MLR) in forecasting the monthly total precipitation of arid regions. Azadi and Sepaskhah (2012) concluded that the neural networks did not significantly increase the prediction accuracy compared with the multiple regression model. Dastorani et al (2010) indicated that the potential of ANN in dryland precipitation prediction is almost same with the ANFIS model.…”

Application of several data-driven techniques to predict a standardized precipitation index

“…Today, more non-linear models are applied to prediction. In previous studies, Dahamsheh and Aksoy (2009), Azadi andSepaskhah (2012), andRezaeian-Zadeh et al (2012) used artificial neural networks (ANNs), and El-Shafie et al (2011), Sanikhani and Kisi (2012), Jeong et al (2012), and Choubin et al (2014a) successfully applied the adaptive neuro-fuzzy inference system (ANFIS) to predict precipitation. In eastern Australia, Deo and Sahin (2015) investigated the application of the ANN model for the prediction of monthly SPIs using hydrometeorological parameters and climate indices.…”