In this paper, artificial neural network (ANN) was used for downscaling the outputs of general circulation models (GCMs) to evaluate changes in precipitation and mean temperature for a future period in Urmia at the north-west of Iran. MIROC-ESM-CHEM from IPCC AR5 was selected as an acceptable model based on correlation coefficient (CC) values, which is calculated between precipitation of GCM models and precipitation data prepared by Urmia Meteorological Organization for 1951-2000. As a first step, the most important parameters of the MIROC-ESM-CHEM was selected before the downscaling process by ANN in the base period (1951-2000). Afterward, the future projections of precipitation and mean temperature during 2020– 2060 were applied using ANN-based simulation according to the CC method. By comparing the results, the MIROC-ESM-CHEM showed a 2.01% increase under RCP4.5 and a 0.16% decrease under RCP8.5 in annual precipitation. Also, the temperature projection outputs showed the annual mean temperature would increase in the future period in this area, and it is likely to get warmer.
ANN- Correlation coefficient-Climate change-Downscaling-GCM-Urmia
Cite this paper
Nazak Rouzegari, Vahid Nourani, Amir Molajou. (2019) Application of Artificial Neural Network and Predictor Screening Method for Downscaling Climatic Parameters. International Journal of Environmental Science, 4, 112-119
Copyright © 2019 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0