Page 44 - 2025年第56卷第10期

impact extreme events［J］ Artificial Intelligence for the Earth Systems，2025，4（1）：1-24.
.
［ 30］ 朱恩达，王亚强，赵妍，等 . 东亚区域人工智能气象大模型预报技巧评估［J］ 应用气象学报，2024，35（6）：
.
641-653.
［ 31］ XU H， ZHAO Y， ZHAO D， et al. Improvement of disastrous extreme precipitation forecasting in North China by
Pangu-weather AI-driven regional WRF model［J］ Environmental Research Letters，2024，19（5）：054051.
.
［ 32］ 杜勇，陈学秋 . 珠江“22·6”特大洪水预报实践［J］ 中国水利，2022（22）：33-35.
.
［ 33］ BECK H E，WOOD E F，PAN M，et al. MSWEP V2 Global 3-hourly 0. 1° precipitation： methodology and quanti‐
tative assessment［J］ Bulletin of the American Meteorological Society，2019，100（3）：473-500.
.
［ 34］ BALSAMO G，RABIER F，BALMASEDA M，et al. Recent progress and outlook for the ECMWF Integrated Fore‐
/
casting System［C］/EGU General Assembly Conference Abstracts. 2023.
.
［ 35］ 李艳忠，星寅聪，庄稼成，等 . 典型遥感降水产品的水文模拟性能评估［J］ 遥感学报，2024，28（2）：
398-413.
.
［ 36］ 苏翔，刘梅，康志明，等 . 2020 年江苏主汛期短期暴雨预报检验［J］ 气象，2022，48（3）：357-371.
［ 37］ CHENG Y， WU H， LI Z L， et al. Evolution and prediction of the extreme rainstorm event in July 2021 in Henan
.
province，China［J］ Atmospheric Science Letters，2023，24（6）：e1156.
.
［ 38］ HERSBACH H，BELL B，BERRISFORD P，et al. The ERA5 global reanalysis［J］ Quarterly Journal of the Royal
Meteorological Society，2020，146（730）：1999-2049.
［ 39］ HUANG Z，ZHAO T. Predictive performance of ensemble hydroclimatic forecasts： Verification metrics，diagnostic
.
plots and forecast attributes［J］ WIREs Water，2022，9（2）：e1580.

Applicability evaluation of large weather models for heavy rainfall events

2
1
1
2
LI Qiang ，ZHAO Tongtiegang ，XU Bo ，LI Yu ，TIAN Yu 3

（1. Center of Water Resources and Environment，Sun Yat-sen University，Guangzhou 510275，China；

2. College of Hydraulic Engineering，Dalian University of Technology，Dalian 116024，China；

3. China Institute of Water Resources and Hydropower Research，Beijing 100038，China）

Abstract：Heavy rainfall and floods are among the most severe natural disasters in China. As large weather models
（LWMs）show great potential in terms of accuracy，timeliness and efficiency in weather forecasting，this paper evalu‐

ates the applicability of different LWMs for heavy rainfall forecasting to provide more accurate，longer lead-time and

more accessible weather forecasting data for flood forecasting. Focusing on two extreme floods in the Beijiang River
Basin of the Pearl River Basin in June 2022 （"2022.6"）and April 2024 （"2024.4"），the predictive performance of

three LWMs，i.e.，the GraphCast，FuXi，and AIFS，in forecasting heavy rainfall under different lead times has been
assessed and compared with the high-resolution forecasts （HRES） from the European Centre for Medium-range

Weather Forecast （ECMWF） The results show that the predictive performance of GraphCast，FuXi，AIFS and
.

HRES is negatively correlated with the intensity of precipitation. Compared with HRES，the performance of Graph‐

Cast，FuXi，and AIFS is more accurate in forecasting the morphology and location of precipitation，but they underes‐
timate the areas and amounts of the two heavy rainfall events. For the "2022.6" event，when the forecast initialization

time is 1 day ahead，the forecasted maximum 6-hour and cumulative precipitation are respectively 20.1 mm and

271.7 mm for the GraphCast，21.8 mm and 222.5 mm for the FuXi and 24.8 mm and 303.0 mm for the AIFS，all of
which underestimate the observed precipitation of 110.9 mm and 416.3 mm. As the lead time increases，the perfor‐

mance of GraphCast，FuXi，and AIFS deteriorates，but they remain more accurate than HRES in forecasting the spa‐

tial location and occurrence time of heavy rainfall events. For the "2024.4" event，when the forecast initialization time

is 3 days ahead，the errors in forecasting the occurrence time of the maximum 6-hour precipitation are respectively

24，-6，72，and 96 hours for GraphCast，FuXi，AIFS，and HRES，respectively. These LWMs demonstrate great

potential in forecasting heavy rainfall and can provide valuable weather forecasts for flood forecasting，serving the

practical needs of disaster prevention，mitigation，and emergency management.

Keywords：extreme precipitation；global climate model；precipitation forecast；artificial intelligence；forecast veri‐

fication
（责任编辑：耿庆斋）
— 1291 —