Markov chain-incorporated and synthetic data-supported conditional artificial neural network models for forecasting monthly precipitation in arid regions

• Published in: Journal of hydrology (Amsterdam) Vol. 562; pp. 758 - 779
• Main Authors: Aksoy, Hafzullah, Dahamsheh, Ahmad
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2018
• Subjects: algorithms; Arid region; arid zones; Artificial neural networks; atmospheric precipitation; case studies; hydrology; Intermittent precipitation; Jordan; Markov chain; neural networks; Synthetic data; Thomas-Fiering model; time series analysis; Jordan; Markov chain; Intermittent precipitation; Arid region; Thomas-Fiering model; Artificial neural networks; Synthetic data
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2018.05.030

•Arid region precipitation was studied for one-month ahead forecasting.•ANNs are not powerful in forecasting intermittent precipitation of arid regions.•ANN models were developed and improved by additional tools.•Markov chain and synthetic data improved the model performance slightly.•A conditional statement was considered for substantial improvement in ANN models. For forecasting monthly precipitation in an arid region, the feed forward back-propagation, radial basis function and generalized regression artificial neural networks (ANNs) are used in this study. The ANN models are improved after incorporation of a Markov chain-based algorithm (MC-ANNs) with which the percentage of dry months is forecasted perfectly, thus generation of any non-physical negative precipitation is eliminated. Due to the fact that recorded precipitation time series are usually shorter than the length needed for a proper calibration of ANN models, synthetic monthly precipitation data are generated by Thomas-Fiering model to further improve the performance of forecasting. For case studies from Jordan, it is seen that only a slightly better performance is achieved with the use of MC and synthetic data. A conditional statement is, therefore, established and imbedded into the ANN models after the incorporation of MC and support of synthetic data, to substantially improve the ability of the models for forecasting monthly precipitation in arid regions.