Incorporating large-scale atmospheric variables in long-term seasonal rainfall forecasting using artificial neural networks: an application to the Ping Basin in Thailand

摘要: Understanding long-term seasonal or annual inter-annual rainfall variability and its relationship with large-scale atmospheric variables (LSAVs) is important for water resource planning management. In this study, forecasting models using the artificial neural network technique were developed to forecast in May–June–July (MJJ), August–September–October (ASO), November–December–January (NDJ), February–March–April (FMA) determine effects of climate change on rainfall. LSAVs, temperature, pressure, wind, precipitable water, relative humidity at different lead times identified as significant predictors. To impacts predictors obtained from two general circulation models, CSIRO Mk3.6 MPI-ESM-MR, used quantile mapping bias correction. Our results show that best performance FMA MJJ seasons are able one month advance these ASO NDJ do so months advance. Under RCP4.5 scenario, a decreasing trend an increasing can be observed 2011 2040. For dry season, while decreases, increases same period time.