Fix mom6 get close state

[hkershaw-brown]

Description:

fix index passed to below sea floor
fix: dist calculation only after call to convert_vertical_state

Cannot assume convert_all_state_verticals_first is set to true assim_tools_mod
and even if you are doing horizonal distance only for localization, still need
to convert the state location to m to check against the depth of the ocean floor
in m.

Fixes issue

fixes #973

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

Documentation changes needed?

• My change requires a change to the documentation.
  • I have updated the documentation accordingly.

Tests

Please describe any tests you ran to verify your changes.

Ran filter with 1 obs and 1 for the increment (⚠️ not this is not commited to the branch) to make spheres (with vertical localization on) or cylinders (for horizontal localization only). Using z* case. The plots are (output_to_filter.nc - input_to_filter.nc

diff --git a/assimilation_code/modules/assimilation/assim_tools_mod.f90 b/assimilation_code/modules/assimilation/assim_tools_mod.f90
index 0c595170b..9f57f91f3 100644
--- a/assimilation_code/modules/assimilation/assim_tools_mod.f90
+++ b/assimilation_code/modules/assimilation/assim_tools_mod.f90
@@ -2057,6 +2057,8 @@ endif

! Get the updated ensemble
if(.not. inflate_only) ens = ens + final_factor * increment
+ens = ens + final_factor * 1.0_r8
+

end subroutine obs_updates_ens

Checklist for merging

• Updated changelog entry
• Documentation updated
• Update conf.py

Checklist for release

• Merge into main
• Create release from the main branch with appropriate tag
• Delete feature-branch

Testing Datasets

• Dataset needed for testing available upon request
• Dataset download instructions included
• No dataset needed

[mgharamti]

I'm trying to understand the motivation for not computing the distances here. Is it faster to just get the indices and then compute the distances for the valid ones below?

[hkershaw-brown]

Yes correct, the get_dist is the expensive bit of the calculation. This is the pattern a lot of our "bigger" models use.

[mgharamti]

Tested the changes by running filter with a single ob. It fixes the localization issue.
It's a clean change and Nicely done @hkershaw-brown!