Skip to content
DraconiticResearch
Search Sources API
201,818 studies31 sourcesUpdated 11 min ago
PreprintCC BY-NC-NDarXiv (Cornell University)

AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning

  • Ewan Pinnington1
  • Peter Lean1
  • Mihai Alexe1
  • Eulalie Boucher1
  • Simon T. K. Lang1
  • Patrick Laloyaux1
  • Gert Mertes1
  • Tomáš Kráľ1
  • Patricia de Rosnay1
  • Matthew Chantry1
  • Anthony Mcnally1
  1. 1European Centre for Medium-Range Weather Forecasts (ECMWF
Open at sourcePDFDOICite

Abstract

We introduce the Artificial Intelligence Forecasting System for Direct Observation Prediction (AIFS-DOP). AIFS-DOP is trained on a 40-year harmonized dataset of gridded observations, without using numerical weather prediction (NWP) reanalysis or model data. The resulting model is competitive with ECMWF's Integrated Forecasting System (IFS) when scored on a one year period of forecasts across 2021/2022. This progress on Direct Observation Prediction represents the first time that a data-driven model, trained solely on observations, is competitive with the IFS at medium ranges for several key upper-air and surface headline scores, when verified against observation data.

Related

  1. FastNet: Improving the Physical Consistency of Machine Learning Weather Prediction Models through Loss Function Design
    Tom Dunstan, Oliver Strickson, Thusal Bennett +34
    Artificial Intelligence for the Earth SystemsPeer-reviewedLicence not statedFull text10 citedAtmospheric science & meteorologyOpenAlex
  2. Butterfly Effect and the Kinetic Energy Cascade in Probabilistic Machine Learning Weather Prediction Models
    Jiakai Chen, Joel Oskarsson, Simon Driscoll +1
    arXivPreprintCC BY 4.0Full textAtmospheric science & meteorologyarXiv
  3. Aries: A Proprietary Medium-Range Weather Prediction Model for the Energy Industry
    Lukas Hedegaard Morsing, Arian Bakhtiarnia, Jonas Lynge Olesen +2
    arXivPreprintCC BY-NC-ND 4.0Full textAtmospheric science & meteorologyarXiv
  4. MAUSAM: An Observations‐Focused Assessment of Global AI Weather Prediction Models During the South Asian Monsoon
    Aman Gupta, Aditi Sheshadri, Dhruv Suri
    Journal of Advances in Modeling Earth SystemsPeer-reviewedCC BY5 citedAtmospheric science & meteorologyOpenAlex
  5. Learning joint probabilistic weather forecasts from station observations alone
    Chaeyeon Yi, Yun Am Seo
    arXivPreprintCC BY 4.0Full textAtmospheric science & meteorologyStatisticsarXiv

Cite

BibTeXRIS
BibTeX
@misc{pinnington2026aifsdop,
  title = {{AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning}},
  author = {Ewan Pinnington and Peter Lean and Mihai Alexe and Eulalie Boucher and Simon T. K. Lang and Patrick Laloyaux and Gert Mertes and Tomáš Kráľ and Patricia de Rosnay and Matthew Chantry and Anthony Mcnally},
  howpublished = {arXiv (Cornell University)},
  year = {2026},
  month = jun,
  doi = {10.48550/arxiv.2606.19093},
  url = {https://arxiv.org/abs/2606.19093},
  note = {CC BY-NC-ND},
}
RIS
TY  - UNPB
TI  - AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning
AU  - Ewan Pinnington
AU  - Peter Lean
AU  - Mihai Alexe
AU  - Eulalie Boucher
AU  - Simon T. K. Lang
AU  - Patrick Laloyaux
AU  - Gert Mertes
AU  - Tomáš Kráľ
AU  - Patricia de Rosnay
AU  - Matthew Chantry
AU  - Anthony Mcnally
PY  - 2026
DA  - 2026/06/17
PB  - arXiv (Cornell University)
DO  - 10.48550/arxiv.2606.19093
UR  - https://arxiv.org/abs/2606.19093
AB  - We introduce the Artificial Intelligence Forecasting System for Direct Observation Prediction (AIFS-DOP). AIFS-DOP is trained on a 40-year harmonized dataset of gridded observations, without using numerical weather prediction (NWP) reanalysis or model data. The resulting model is competitive with ECMWF's Integrated Forecasting System (IFS) when scored on a one year period of forecasts across 2021/2022. This progress on Direct Observation Prediction represents the first time that a data-driven model, trained solely on observations, is competitive with the IFS at medium ranges for several key upper-air and surface headline scores, when verified against observation data.
SN  - 2331-8422
ER  -