# Open weather/climate data for Deep Learning model training

**URL:** <https://discourse.pangeo.io/t/open-weather-climate-data-for-deep-learning-model-training/3726>\
**Category:** Uncategorized\
**Created:** [October 3, 2023, 2:47pm UTC](https://discourse.pangeo.io/t/open-weather-climate-data-for-deep-learning-model-training/3726 "2023-10-03T14:47:40Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Yves\_Moisan](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.pangeo.io/yves_moisan/32/2070_2.png) [@Yves\_Moisan](https://discourse.pangeo.io/u/Yves_Moisan)\
**Post date:** [October 3, 2023, 2:47pm UTC](https://discourse.pangeo.io/t/open-weather-climate-data-for-deep-learning-model-training/3726/1 "2023-10-03T14:47:40Z")

</div>

Hi All. There is an increasing consensus for considering more than atmospheric variables for weather forecasting, especially in data-driven models : see “[ECMWF Earth System](https://www.ecmwf.int/en/about/media-centre/focus/2021/fact-sheet-earth-system-modelling-ecmwf)” model. To make a long story short : future modelling initiatives will likey need access to traditional “EO” data.

There was [an announcement](https://radiant.earth/blog/2023/05/radiant-earth-announces-new-initiatives-to-accelerate-sharing-of-earth-science-data/) a few months ago : “[Source Cooperative](https://source.coop/) and the [Cloud-Native Geospatial Foundation](https://cloudnativegeo.org/) … designed to make Earth science data more accessible and easier to use, which will help scientists to better understand our planet and the challenges it faces.”

Any idea how those initiatives might be interesting for (or interested in?) holding met-ocean open data ?