{
  "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.",
  "abstract_shown": true,
  "affiliations": [
    "European Centre for Medium-Range Weather Forecasts (ECMWF"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 11,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Ewan Pinnington"
    },
    {
      "affs": [
        0
      ],
      "name": "Peter Lean",
      "orcid": "0000-0002-3662-5382"
    },
    {
      "affs": [
        0
      ],
      "name": "Mihai Alexe"
    },
    {
      "affs": [
        0
      ],
      "name": "Eulalie Boucher",
      "orcid": "0000-0002-6070-2544"
    },
    {
      "affs": [
        0
      ],
      "name": "Simon T. K. Lang",
      "orcid": "0000-0003-3952-586X"
    },
    {
      "affs": [
        0
      ],
      "name": "Patrick Laloyaux",
      "orcid": "0000-0003-2808-0463"
    },
    {
      "affs": [
        0
      ],
      "name": "Gert Mertes"
    },
    {
      "affs": [
        0
      ],
      "name": "Tomáš Kráľ"
    },
    {
      "affs": [
        0
      ],
      "name": "Patricia de Rosnay",
      "orcid": "0000-0002-7374-3820"
    },
    {
      "affs": [
        0
      ],
      "name": "Matthew Chantry",
      "orcid": "0000-0002-1132-0961"
    },
    {
      "affs": [
        0
      ],
      "name": "Anthony Mcnally"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5vjphkcah6nky.bib",
  "cited_by": 8,
  "doi": "10.48550/arxiv.2606.19093",
  "fulltext_url": null,
  "id": "5vjphkcah6nky",
  "issns": [
    "2331-8422"
  ],
  "landing_url": "https://arxiv.org/abs/2606.19093",
  "language": "en",
  "licence": "CC-BY-NC-ND",
  "licence_label": "CC BY-NC-ND",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "1902|Meteorological Phenomena and Simulations"
    ]
  ],
  "openalex": "W7165147083",
  "pdf_url": "https://arxiv.org/pdf/2606.19093",
  "pmcid": null,
  "pmid": null,
  "published": "2026-06-17",
  "publisher": "Cornell University",
  "ris": "https://research.draconitic.com/paper/5vjphkcah6nky.ris",
  "sources": [
    {
      "first_seen": "2026-11-08T13:20:53Z",
      "native_id": "W7165147083",
      "source": "openalex",
      "url": "https://arxiv.org/abs/2606.19093"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "openalex:1902|Meteorological Phenomena and Simulations",
      "subfield": "atmospheric"
    }
  ],
  "title": "AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning",
  "title_text": "AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning",
  "type": "preprint",
  "updated_at": "2026-11-09T13:21:19Z",
  "url": "https://research.draconitic.com/paper/5vjphkcah6nky",
  "venue": "arXiv (Cornell University)",
  "versions": []
}