Comparing the S2S hindcast skills to forecast Iran’s precipitation and capturing climate drivers signals over the Middle East

To enhance the capabilities of the Subseasonal to Seasonal (S2S) database for forecasting Iran’s southwest precipitation from 1 to 4 weeks ahead, we compared observed precipitation and atmospheric variables with the corresponding hindcasts generated by the KMA, UKMO, ECWMF, and Meteo-France (MF) research centers. This analysis involved several deterministic and probabilistic metrics. Our reference datasets included daily precipitation data from 176 rain gauge stations and the NOAA-based atmospheric circulations data for Dec-April 1995–2014. Most hindcasts underestimated wet events in southern and eastern districts but overestimated them in the western and northern regions. Additionally, all hindcasts overforecasted the frequency of wet events across all lead times. The correlation scores were highest in the first week and declined as lead times increased. The ECMWF had the best correlation in all regions, showing superior deterministic and probabilistic forecast skills in western districts. The UKMO hindcasts, whose accuracy has the highest dependancy on precipitation amount, effectively captured signals of the El-Niño Southern Oscillation (ENSO) and Madden Julian Oscillation (MJO) over the study area and the Middle East. They accurately forecasted the 850 hPa moisture transport and 500 hPa vertical velocity features in these regions, particularly during the rainy phases of MJO. These findings offer valuable insights to enhance the accuracy of operational S2S precipitation forecasts for planning and decision-making in Iran and the Middle East.