Global Evaluation of AI and NWP Precipitation Forecasts During Atmospheric River Events

Marina Vicens-Miquel; Taylor Mandelbaum; Amy McGovern; Aaron J. Hill; Daniel Rothenberg

Home/Archive/arXiv

arXiv · 2610.03758

Global Evaluation of AI and NWP Precipitation Forecasts During Atmospheric River Events

Marina Vicens-Miquel·Taylor Mandelbaum·Amy McGovern·Aaron J. Hill·Daniel Rothenberg

View original source ↗Read source document ↗Export metadata ↓

Abstract

Atmospheric rivers (ARs) produce many of the world's most extreme precipitation events and hydrometeorological hazards. Although artificial intelligence weather prediction (AIWP) models have demonstrated skill comparable to or exceeding numerical weather prediction (NWP) systems for large-scale atmospheric variables, their ability to forecast AR-related precipitation remains insufficiently characterized globally. Here, we evaluate 24-hour precipitation forecasts from the Global Forecast System (GFS), Global Ensemble Forecast System (GEFS), GraphCast, and Artificial Intelligence Forecasting System (AIFS) from Day 1 through Day 10 globally and across North America, Europe, and Australia and New Zealand. Using the Extreme Weather Bench framework, forecasts are evaluated against Integrated Multi-satellitE Retrievals for GPM (IMERG) observations using measures of precipitation magnitude, spatial structure, and localization. GraphCast and AIFS exhibit greater spatial skill than GFS and GEFS, particularly for heavy precipitation and at longer lead times, and better preserve the spatial organization of AR-related precipitation through Day 10. However, this improved spatial skill does not translate into accurate precipitation magnitudes. AIWP models tend to overpredict moderate-to-heavy accumulations while underpredicting the heaviest precipitation at longer lead times, whereas NWP systems develop pronounced dry biases. These results reveal distinct strengths and limitations of AIWP for high-impact precipitation forecasting and provide a reproducible benchmark.

Explore related subjects

physics.ao-ph cs.AI

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Marina Vicens-Miquel, Taylor Mandelbaum, Amy McGovern, Aaron J. Hill, Daniel Rothenberg. 2026-09-26. Global Evaluation of AI and NWP Precipitation Forecasts During Atmospheric River Events. https://arxiv.org/abs/2610.03758

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

OTProf: estimating high-resolution profiles of optical turbulence ($C_n^2$) from reanalysis using deep learning

Accurate, high-resolution vertical profiles of optical turbulence ($C_n^2$), which reflect local meteorology and topography, are crucial for ground-based optical astronomy and free-space optical communication. However, measuring these profiles or generating them with numerical weather models requires substantial operational or computational effort. In this work, we present OTProf, a deep-learning method that estimates high-resolution $C_n^2$ profiles from widely available coarse-resolution ERA5 reanalysis data. OTProf is based on the Squeezeformer architecture, modified for super-resolution by introducing grid-agnostic interpolation in the latent space. We train and evaluate OTProf in the Netherlands using high-resolution $C_n^2$ data obtained from the Weather Research and Forecasting (WRF) model. The commonly used Hufnagel-Valley model serves as the baseline. Overall, OTProf reproduces the vertical structure of WRF-based $C_n^2$ more accurately than Hufnagel-Valley and yields more accurate estimates of the Fried parameter $r_0$ and the scintillation index $σ_I^2$. Out-of-domain performance is assessed using a smaller WRF-based $C_n^2$ dataset covering the Canary Islands. Initial results indicate a performance drop because the orography differs markedly from the training domain. We show that output bias correction based on a few local simulations can help OTProf to adapt, but more work for domain adaptation is needed. OTProf is therefore primarily recommended for in-domain applications, e.g., generating $C_n^2$ climatologies from long ERA5 records.

physics.ao-ph↗

arXiv2026

State-resolved multimodal contributions to stratospheric polar vortex predictability

The dynamical basis of stratospheric polar vortex (SPV) predictability remains unclear, particularly the relative roles of persistence, structural variability, and cross-level coupling. Here we combine eigen microstate theory with mesoscopic Granger-causality analysis of ERA5 geopotential height fields to quantify the predictive contributions of eigen microstates. We show that short-term predictability is dominated by the persistence of the leading stratospheric state, whereas higher-order stratospheric structures and tropospheric variability provide additional predictive information at subseasonal leads. Including these additional states increases the all-winter 30-day prediction correlation from 0.24 to 0.33 and reduces forecast error. These contributions exhibit strong lead-time dependence and vary between winters with and without sudden stratospheric warmings. Part of this predictive information overlaps with that contained in preceding 100 hPa eddy heat flux, linking state-resolved predictability to wave-related circulation variability. Our results unify SPV predictability within a multimodal, state-resolved framework and provide a physically interpretable pathway.

physics.ao-ph↗

arXiv2026

Cleaner air strengthens the Pacific trade winds without a detectable change in El Nino

Air quality policy will cut aerosol and ozone precursor emissions, and studies disagree on how the tropical Pacific circulation and the El Nino Southern Oscillation (ENSO) will respond. Here we show, in 12 Earth system models with and without air quality controls, that the cuts strengthen the Pacific trade winds and offset half to two thirds of the weakening projected without them. The response is absent with prescribed sea surface temperatures and detectable only after 2040. It is consistent with the Indian Ocean warming relative to the Pacific, which also ranks the responses to cuts made region by region. ENSO amplitude changes by less than a tenth of its value, the smallest change this ensemble could detect. The change implied by the mean state shift is a sixth of that, and an ENSO model trained without the reduced emission runs predicts none. El Nino's regional impacts still change, because the background climate does.

physics.ao-ph↗