More Like this

1. A Comparison of AI Weather Prediction and Numerical Weather Prediction Models for 1–7-Day Precipitation Forecasts

   https://doi.org/10.1175/WAF-D-24-0081.1  

   Radford, Jacob T; Ebert-Uphoff, Imme; Stewart, Jebb Q (April 2025, Weather and Forecasting)

   Abstract Pure artificial intelligence (AI)-based weather prediction (AIWP) models have made waves within the scientific community and the media, claiming superior performance to numerical weather prediction (NWP) models. However, these models often lack impactful output variables such as precipitation. One exception is Google DeepMind’s GraphCast model, which became the first mainstream AIWP model to predict precipitation, but performed only limited verification. We present an analysis of the ECMWF’s Integrated Forecasting System (IFS)-initialized (GRAP_IFS) and the NCEP’s Global Forecast System (GFS)-initialized (GRAP_GFS) GraphCast precipitation forecasts over the contiguous United States and compare to results from the GFS and IFS models using 1) grid-based, 2) neighborhood, and 3) object-oriented metrics verified against the fifth major global reanalysis produced by ECMWF (ERA5) and the NCEP/Environmental Modeling Center (EMC) stage IV precipitation analysis datasets. We affirmed that GRAP_GFSand GRAP_IFSperform better than the GFS and IFS in terms of root-mean-square error and stable equitable errors in probability space, but the GFS and IFS precipitation distributions more closely align with the ERA5 and stage IV distributions. Equitable threat score also generally favored GraphCast, particularly for lower accumulation thresholds. Fractions skill score for increasing neighborhood sizes shows greater gains for the GFS and IFS than GraphCast, suggesting the NWP models may have a better handle on intensity but struggle with the location. Object-based verification for GraphCast found positive area biases at low accumulation thresholds and large negative biases at high accumulation thresholds. GRAP_GFSsaw similar performance gains to GRAP_IFSwhen compared to their NWP counterparts, but initializing with the less familiar GFS conditions appeared to lead to an increase in light precipitation. Significance StatementPure artificial intelligence (AI)-based weather prediction (AIWP) has exploded in popularity with promises of better performance and faster run times than numerical weather prediction (NWP) models. However, less attention has been paid to their capability to predict impactful, sensible weather like precipitation, precipitation type, or specific meteorological features. We seek to address this gap by comparing precipitation forecast performance by an AI model called GraphCast to the Global Forecast System (GFS) and the Integrated Forecasting System (IFS) NWP models. While GraphCast does perform better on many verification metrics, it has some limitations for intense precipitation forecasts. In particular, it less frequently predicts intense precipitation events than the GFS or IFS. Overall, this article emphasizes the promise of AIWP while at the same time stresses the need for robust verification by domain experts. 

   more » « less
2. Improving Short-Term QPF Using Geostationary Satellite All-Sky Infrared Radiances: Real-Time Ensemble Data Assimilation and Forecast during the PRECIP 2020 and 2021 Experiments

   https://doi.org/10.1175/WAF-D-22-0156.1  

   Zhang, Yunji; Chen, Xingchao; Bell, Michael M. (April 2023, Weather and Forecasting)

   Abstract The Prediction of Rainfall Extremes Campaign In the Pacific (PRECIP) aims to improve our understanding of extreme rainfall processes in the East Asian summer monsoon. A convection-permitting ensemble-based data assimilation and forecast system (the PSU WRF-EnKF system) was run in real time in the summers of 2020–21 in advance of the 2022 field campaign, assimilating all-sky infrared (IR) radiances from the geostationary Himawari-8 and GOES-16 satellites, and providing 48-h ensemble forecasts every day for weather briefings and discussions. This is the first time that all-sky IR data assimilation has been performed in a real-time forecast system at a convection-permitting resolution for several seasons. Compared with retrospective forecasts that exclude all-sky IR radiances, rainfall predictions are statistically significantly improved out to at least 4–6 h for the real-time forecasts, which is comparable to the time scale of improvements gained from assimilating observations from the dense ground-based Doppler weather radars. The assimilation of all-sky IR radiances also reduced the forecast errors of large-scale environments and helped to maintain a more reasonable ensemble spread compared with the counterpart experiments that did not assimilate all-sky IR radiances. The results indicate strong potential for improving routine short-term quantitative precipitation forecasts using these high-spatiotemporal-resolution satellite observations in the future. Significance Statement During the summers of 2020/21, the PSU WRF-EnKF data assimilation and forecast system was run in real time in advance of the 2022 Prediction of Rainfall Extremes Campaign In the Pacific (PRECIP), assimilating all-sky (clear-sky and cloudy) infrared radiances from geostationary satellites into a numerical weather prediction model and providing ensemble forecasts. This study presents the first-of-its-kind systematic evaluation of the impacts of assimilating all-sky infrared radiances on short-term qualitative precipitation forecasts using multiyear, multiregion, real-time ensemble forecasts. Results suggest that rainfall forecasts are improved out to at least 4–6 h with the assimilation of all-sky infrared radiances, comparable to the influence of assimilating radar observations, with benefits in forecasting large-scale environments and representing atmospheric uncertainties as well. 

   more » « less
3. Impact study of increased radio occultation observations during the ROMEX period using JEDI and the GFS atmospheric model

   https://doi.org/10.5194/egusphere-2025-3235  

   Zhang, Hailing; Shao, Hui; Ruston, Benjamin; Braun, John (July 2025, EGUsphere)

   Abstract. The international collaborative Radio Occultation Modeling EXperiment (ROMEX) project marks the first time using a large volume of real data to assess the impact of increased Global Navigation Satellite System (GNSS) radio occultation (RO) observations beyond current operational levels, moving past previous theoretical simulation-based studies. The ROMEX project enabled the use of approximately 35,000 RO profiles– nearly triple the number typically available to operational centers, which is about 8,000 to 12,000 per day. This study investigates the impact of increased RO profiles on numerical weather prediction (NWP) with the Joint Effort for Data assimilation Integration (JEDI) and the global forecast system (GFS), as part of the ROMEX effort. A series of experiments were conducted assimilating varying amounts of RO data along with a common set of other key observations. The results confirm that assimilating additional RO data further improves forecasts across all major meteorological fields, including temperature, humidity, geopotential height, and wind speed, for most of vertical levels. These improvements are significantly evident in verification against both critical observations and the European Center for Medium-Range Weather Forecasts (ECMWF) analyses, with beneficial impacts lasting up to five days. Conversely, withholding RO data resulted in forecast degradations. The results also suggest that forecast improvements scale approximately logarithmically with the number of assimilated profiles, and no evidence of saturation was observed. Biases in the forecast of temperature and geopotential height over the lower stratosphere are discussed, and they are consistent with findings from other studies in the ROMEX community. 

   more » « less
4. Can Convection‐Permitting Modeling Provide Decent Precipitation for Offline High‐Resolution Snowpack Simulations Over Mountains?

   https://doi.org/10.1029/2019JD030823  

   He, Cenlin; Chen, Fei; Barlage, Michael; Liu, Changhai; Newman, Andrew; Tang, Wenfu; Ikeda, Kyoko; Rasmussen, Roy (December 2019, Journal of Geophysical Research: Atmospheres)

   Abstract Accurate precipitation estimates are critical to simulating seasonal snowpack evolution. We conduct and evaluate high‐resolution (4‐km) snowpack simulations over the western United States (WUS) mountains in Water Year 2013 using the Noah with multi‐parameterization (Noah‐MP) land surface model driven by precipitation forcing from convection‐permitting (4‐km) Weather Research and Forecasting (WRF) modeling and four widely used high‐resolution datasets that are derived from statistical interpolation based on in situ measurements. Substantial differences in the precipitation amount among these five datasets, particularly over the western and northern portions of WUS mountains, significantly affect simulated snow water equivalent (SWE) and snow depth (SD) but have relatively limited effects on snow cover fraction (SCF) and surface albedo. WRF generally captures observed precipitation patterns and results in an overall best‐performed SWE and SD in the western and northern portions of WUS mountains, where the statistically interpolated datasets lead to underpredicted precipitation, SWE, and SD. Over the interior WUS mountains, all the datasets consistently underestimate precipitation, causing significant negative biases in SWE and SD, among which the results driven by the WRF precipitation show an average performance. Further analysis reveals systematic positive biases in SCF and surface albedo across the WUS mountains, with similar bias patterns and magnitudes for simulations driven by different precipitation datasets, suggesting an urgent need to improve the Noah‐MP snowpack physics. This study highlights that convection‐permitting modeling with proper configurations can have added values in providing decent precipitation for high‐resolution snowpack simulations over the WUS mountains in a typical ENSO‐neutral year, particularly over observation‐scarce regions. 

   more » « less
5. Performance of the Pangu‐Weather Deep Learning Model in Forecasting Tornadic Environments

   https://doi.org/10.1029/2024GL109611  

   Hua, Zhanxiang; Hakim, Gregory; Anderson‐Frey, Alexandra (April 2025, Geophysical Research Letters)

   Abstract The development of deep learning (DL) weather forecasting models has made rapid progress and achieved comparable or better skill than traditional Numerical Weather prediction (NWP) models, which are generally computationally intensive. However, applications of these DL models have yet to be fully explored, including for severe convective events. We evaluate the DL model Pangu‐Weather in forecasting tornadic environments with one‐day lead times using convective available potential energy (CAPE), 0–6 bulk wind difference (BWD6), and 0–3 km storm‐relative helicity (SRH3). We also compare its performance to the National Centers for Environmental Prediction (NCEP)'s Global Forecast System (GFS), a traditional NWP model. Pangu‐Weather generally outperforms GFS in predicting BWD6 and SRH3 at the closest grid point and hour of the storm report. However, Pangu‐Weather tends to underpredict the maximum values of all convective parameters in the 1–2 hr before the storm across the surrounding grid points compared to the GFS. 

   more » « less