# Missing FieldMetaData in hyb-3dvar\_gfs\_aero.yaml

**URL:** <https://forums.jcsda.org/t/missing-fieldmetadata-in-hyb-3dvar-gfs-aero-yaml/72>\
**Category:** JEDI\
**Created:** [February 11, 2021, 8:04pm UTC](https://forums.jcsda.org/t/missing-fieldmetadata-in-hyb-3dvar-gfs-aero-yaml/72 "2021-02-11T20:04:14Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![andytangborn](https://yyz1.discourse-cdn.com/flex031/user_avatar/forums.jcsda.org/andytangborn/32/48_2.png) [@andytangborn](https://forums.jcsda.org/u/andytangborn)\
**Post date:** [February 11, 2021, 8:04pm UTC](https://forums.jcsda.org/t/missing-fieldmetadata-in-hyb-3dvar-gfs-aero-yaml/72/1 "2021-02-11T20:04:14Z")

</div>

I’m running fv3jedi-var.x with the hyb-3dvar\_gfs\_aero.yaml. I’m getting an error message:

FieldMetadata::getField: Field mass\_fraction\_of\_dust001\_in\_air not found in FieldsMetadata

Is this caused by an inconsistency in the yaml file? Or is it in the forward operator (AodCRTM). I haven’t seen this in earlier hyb-3dvar runs. The difference here is that I am using the calculation of stddev when I generate the bumpparameter files.

---

<div class="post-metadata">

**Author:** ![andytangborn](https://yyz1.discourse-cdn.com/flex031/user_avatar/forums.jcsda.org/andytangborn/32/48_2.png) [@andytangborn](https://forums.jcsda.org/u/andytangborn)\
**Post date:** [February 14, 2021, 2:12am UTC](https://forums.jcsda.org/t/missing-fieldmetadata-in-hyb-3dvar-gfs-aero-yaml/72/2 "2021-02-14T02:12:27Z")

</div>

I found the problem. The experiment directory linked the wrong fieldsets directory, so it was not using the updated version. This is now resolved.

---

<div class="post-metadata">

**Author:** ![mmiesch](https://yyz1.discourse-cdn.com/flex031/user_avatar/forums.jcsda.org/mmiesch/32/39_2.png) [@mmiesch](https://forums.jcsda.org/u/mmiesch)\
**Post date:** [February 16, 2021, 3:33pm UTC](https://forums.jcsda.org/t/missing-fieldmetadata-in-hyb-3dvar-gfs-aero-yaml/72/3 "2021-02-16T15:33:59Z")

</div>

Thanks @andytangborn for letting us know - glad you found the solution!