{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiaw23lz5zj5ma2omu6atpay5os6msj6cb5sogye5l6dtbnn4tgrjy",
    "uri": "at://did:plc:3smsdppscbbmgh33ttbch2od/app.bsky.feed.post/3mjhngq3hvrf2"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreihwapye4apjzpskeumzhmts2op4q6hccmg2yp6ew2dt6qzynqaavu"
    },
    "mimeType": "image/png",
    "size": 512974
  },
  "path": "/editor-highlights/machine-learning-can-improve-the-use-of-atmospheric-observations-in-the-tropics",
  "publishedAt": "2026-04-14T12:00:00.000Z",
  "site": "https://eos.org",
  "tags": [
    "Editors' Highlights",
    "climate",
    "everything atmospheric",
    "Journal of Advances in Modeling Earth Systems (JAMES)",
    "machine learning & AI",
    "tropics",
    "Melinc et al. [2026]"
  ],
  "textContent": "Illustration of the effect of the assimilation of a single observation taken at a middle vertical level of the atmosphere. The gold star marks the location of the observation. The color shades indicate the information provided by the observation about the atmospheric state at the different locations (a and b) at the same vertical level and (c) near the earth’s surface. Credit: Melinc et al. [2026], Figure 1",
  "title": "Machine Learning Can Improve the Use of Atmospheric Observations in the Tropics"
}