September 2025 Climate Forecast Discussion for the period Oct-Dec 2025 through Jan-Mar 2026
**Note: ** The IRI seasonal forecasts of precipitation and temperature issued this month are based on an objective calibration procedure that combines four models NCEP-CFSv2, CanSIPS-IC4, COLA-RSMAS-CESM1, and NASA-GEOSS2S models. The climatological base period for normal is 1991-2020. The SST anomalies cited below refer to the OISST.v2.1 SST data set. Starting from June 2024, the observed climatology used to compute the dry mask was updated to 1991-2020, and the criterion modified to mask points where 10% of the training sample (i.e. 3 or more years) received less than or equal to 1 mm over a 3-month season.
During August 2025, sea surface temperature (SST) anomalies remained in the ENSO-neutral range over the central-eastern equatorial Pacific Ocean, with the monthly NINO3.4 anomaly index value at -0.33 °C. Weekly-averaged SST anomalies in the NINO3.4 region were -0.5 °C for the week centered on September 10, 2025. This indicates the gradual development of a cold ENSO state, which is expected to emerge during fall 2025. In conjunction, the early September CPC ENSO forecast, maintained a “La Nina Watch”. As per the statement, a transition from ENSO-neutral to La Niña is likely during Oct-Dec 2025 (71% chance). Thereafter, La Niña is favored but chances decrease to 54% in Dec-Feb 2025/2026. The Indian Ocean Dipole (IOD) index is negative and expected to remain so for the next couple of months, while the North Atlantic Ocean exhibits slightly warm SST anomalies that have decreased substantially. Moreover, the Pacific Decadal Oscillation Index in August 2025 recorded a value of -3.23, following its record-breaking value of -4.12 in July 2025.
The IRI Multi-Model Probability Forecasts for October 2025 through March 2026 highlight regions with an elevated probability of below-normal precipitation (brown shading), especially across the southern United States and northern Mexico during OND and NDJ, as well as parts of the eastern Mediterranean and the Middle East—persisting through all forecast seasons and peaking in DJF and JFM. Central Southwest Asia also shows an increased probability of below-normal precipitation, particularly from NDJ to JFM. Additional areas with a higher probability of below-normal precipitation include northeast Brazil (mainly in DJF and JFM), parts of southern Africa, and the eastern Horn of Africa during OND and NDJ. These regions face increased risk of drought and water scarcity as the dry signal persists through the boreal winter months. In contrast, regions such as Australia and parts of maritime Southeast Asia—including Indonesia and Papua New Guinea—feature a high probability of above-normal (green shading) precipitation, especially in OND and NDJ. Further areas with increased probability of above-normal rainfall include the southern Amazon (prominent in JFM), southeast South America (Paraguay, southern Brazil, northern Argentina in DJF), and parts of the western Indian Ocean region (Madagascar and adjacent islands in OND). These forecasts underscore the probabilistic nature of seasonal precipitation outlooks and the influence of major climate drivers across different regions and time periods, potentially linked to the development of La Niña conditions. However, the signal remains weak and could transition back to ENSO-neutral later in the winter and spring of 2026.
The temperature forecasts show that above-normal temperatures (red shading) are highly probable in nearly all seasons across much of North America, northern and central Eurasia, the Middle East, and large portions of Africa and South America. There are regions with gray shading across portions of South America (notably the Amazon basin and southern South America), Central Africa, and southeastern Australia. These areas indicate where the probability of the normal temperature category is dominant (i.e., close to the climatological median).
