Record-Breaking 24-Hour Rainfall Exceeds 1,000mm in Vietnam: A Review of Our Precipitation Forecast Accuracy

Central Vietnam experienced significant weather impacts starting October 21, 2025, when moist air began flowing into the region due to the approach of Typhoon No. 24 and strengthening northeast monsoon conditions. Even after Typhoon No. 24 passed through, small-scale atmospheric disturbances persisted between Da Nang and Hue, continuing the convergence of warm, humid air. This resulted in approximately one week of continuous rainfall lasting through October 27.

The sustained heavy rainfall triggered widespread flooding, tragically resulting in 13 deaths and missing persons.¹ The damage to infrastructure and property was extensive: 64 homes collapsed or sustained major damage, over 100,000 houses experienced flooding, and more than 300 road sections—including national highways—were affected by landslides. The agricultural and livestock industries also suffered significant losses. ²

Evaluating Rainfall Forecast Accuracy

To assess how well we anticipated this extreme weather event, Weathernews analyzed the performance of seven global weather models developed by national meteorological agencies worldwide.

According to the Vietnam Meteorological Agency, the heaviest rainfall was recorded in Nam Dong, Hue City, where an extraordinary 1,138.6 mm of precipitation fell between 07:00 UTC on October 26 and 07:00 UTC on October 27. To put this in perspective, Hue City received more than an entire month's worth of rain in just 24 hours.³

Looking back at the forecasts issued at 12:00 UTC on October 25, several models did indicate the potential for heavy rainfall exceeding 100 mm. The most aggressive prediction came from the U.S. National Centers for Environmental Prediction (NCEP) model, which forecasted 378 mm of rainfall. However, even this highest prediction fell dramatically short—the actual observed rainfall was nearly three times greater than what any model anticipated (Figure).

ECMWF: European Centre for Medium-Range Weather Forecasts NCEP: U.S. National Centers for Environmental Prediction JMA: Japan Meteorological Agency BOM: Australian Bureau of Meteorology CMC: Canadian Meteorological Centre DWD: Deutscher Wetterdienst

At Weathernews, we strive to deliver highly accurate forecasts by utilizing our proprietary prediction models alongside ensemble forecasting that incorporates models from other agencies. Through continuous accuracy evaluation, we work to steadily improve our forecasting capabilities. Moving forward, we will continue conducting rigorous accuracy verification across various prediction models, including AI-based weather forecasting systems. By deepening our understanding of each model's characteristics through this ongoing analysis, we remain committed to providing you with increasingly precise weather information