Seasonal Precipitation Skill and Sources of Predictability in the Middle East With a Focus on the Arabian Peninsula: Insights From APEC Climate Center Multi‐Model Ensemble Prediction

Abstract This study comprehensively assesses the seasonal prediction skill of precipitation over the Middle East (ME), with a particular focus on the Arabian Peninsula (AP), using the APEC Climate Center (APCC) multi‐model ensemble (MME) system. Analysis of one‐month lead hindcasts (1993–2016) shows that the MME consistently outperforms individual models and shows higher skill during boreal spring and autumn. These seasons exhibit strong teleconnections with tropical SST‐based climate drivers, such as El Niño‐Southern Oscillation (ENSO), El Niño Modoki, and the Indian Ocean Dipole (IOD), while forecast skill remains lower in boreal summer and winter. Teleconnection analyses reveal that the MME's skill is closely linked to its ability to capture SST‐driven precipitation variability. Forecast skill significantly declines when regression‐based precipitation signals from climate drivers are removed, highlighting the importance of these teleconnections. Although the MME exhibits overall superior performance, challenges remain in simulating winter precipitation over the AP, where the influence of climate drivers is relatively weak and models struggle to accurately reproduce teleconnection patterns. These findings underscore the value of improving the model representation of teleconnections to enhance seasonal forecast reliability. The results provide actionable insights for enhancing dynamical prediction systems and inform future model development tailored to this climate‐sensitive region.