Predicting saffron yield from meteorological data using expert system, Razavi and South Khorasan Provinces

Khorasan Province is one of the most important provinces of Iran، especially as regards agricultural product. The prediction of crop yield with available data has important effects on socio-economic and political decisions at the regional scale. Recently، the application of Artificial Neural Network (ANN) has been developed as a powerful tool which enables to solve accurately the most complicated equations and to perform appropriate numerical analysis. This study shows the ability of Artificial Neural Network (ANN) technology for the prediction of saffron (Corcus sativus) yield، based on the available daily weather and yearly agricultural data. Evapotranspiration، temperature (max، min، and dew temperatures)، precipitation and daily average relative humidity for 20 years at synoptic stations were the weather data used. The potential of ANN and Multi-Layered Preceptron (MLP) methods were examined to predict saffron yield. The MLP models of Artificial Neural Networks and regression using maximum temperature، precipitation، evapotranspiration and relative humidity of autumn and last year yield، as independent variables in predicting the crop yield (R2=0. 8832، RMSE= 0. 689 kg. ha-1، MAE= 0. 560 kg. ha-1)، the most efficiency was achieved.