Add a pre-industrial nitrogen example

[penguian]

The example notebook reads in the CMIP7 nitrogen forcing datasets as Iris cubes, adds the cubes, and then creates an ancillary from the resulting cube, except that is does not save the ancil file.

The notebook cannot use the /g/data/access/ngm/modules/ants modules because they use separate Singularity containers for Python.

[penguian]

Tested with https://code.metoffice.gov.uk/trac/roses-u/browser/d/q/8/1/9/nitrogen-ozone?rev=337274
This produced a file at gadi:/g/data/tm70/pcl851/CMIP7/esm16_ancil/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2025.11.13/Ndep_1850_cmip7.anc

[MartinDix]

This file still seems to be using the input4MIPS units of kg m^-2 s^-1 rather than ESM 1.6 g m^-2 day^-1

[penguian]

OK. I will change the units to g m^-2 day^-1

[penguian]

Looking at the published STASH codes for NITROGEN DEPOSITION, there is a code m01s00i447 corresponding to kg m^-2 s^-1 but no code that corresponds to g m^-2 day^-1. See also https://code.metoffice.gov.uk/trac/um/wiki/STASH/ReservedCodes which also does not include a nitrogen deposition STASH code with these units.

Examining the file /g/data/vk83/configurations/inputs/access-esm1p5/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2020.05.19/Ndep_1850_ESM1.anc reveals that lbuser4=884, corresponding to a(n unpublished?) STASH code of m01s00i884.

Examining https://forum.access-hive.org.au/t/proposed-changes-to-variable-names-in-um-netcdf-conversion/4599 reveals

s03i884	_	_	NITROGEN DEPOSITION

This makes sense, since STASH section 3 refers to "boundary layer and surface".

I will change the STASH code to m01s03i884.

[penguian]

Tested with https://code.metoffice.gov.uk/trac/roses-u/browser/d/q/8/1/9/nitrogen-ozone?rev=337274
See /g/data/tm70/pcl851/CMIP7/esm16_ancil/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2025.11.14/Ndep_1850_cmip7.anc

[MartinDix]

The units used by CABLE are different to those used by MOSES/JULES so we have to ignore what the STASHmaster file says. The STASH code in the ancillary file has to be m01s00i447, using the code for the section 3 diagnostic field won't work.

Fix for um2nc ACCESS-NRI/um2nc-standalone#208

[penguian]

I spoke to Spencer. See https://github.com/ACCESS-NRI/access-esm1.6-configs/blob/dev-preindustrial%2Bconcentrations/atmosphere/prefix.PRESM_A lines 711-715:

1|    1 |    0 |  884 |NITROGEN DEPOSITION                 |
2|    2 |    0 |    1 |   21 |    5 |   -1 |   -1 |    0 |    0 |    0 |    0 |
3| 000000000000000000000000000000 | 00000000000000000001 |    3 |
4|    1 |  122 | -99  -99  -99  -99  -99  -99  -99  -99  -99  -99 |
5|    0 | 1510 |    0 |  129 |    0 |    0 |    0 | 9999 |    0 |

This agrees with the output of mule-pumf for /g/data/vk83/configurations/inputs/access-esm1p5/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2020.05.19/Ndep_1850_ESM1.anc, so the correct SASh code is m01s00i884, and the units for STASH code m01s00i447 do not need to be changed.

[penguian]

Output is in /g/data/tm70/pcl851/CMIP7/esm16_ancil/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2025.11.18/Ndep_1850_cmip7.anc

$ mule-cumf --stashmaster 7.3 /g/data/vk83/configurations/inputs/access-esm1p5/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2020.05.19/Ndep_1850_ESM1.anc /g/data/tm70/pcl851/CMIP7/esm16_ancil/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2025.11.18/Ndep_1850_cmip7.anc|more
/g/data/xp65/public/apps/med_conda/envs/analysis3-25.09/lib/python3.11/site-packages/mule/validators.py:198: UserWarning: 
File: /g/data/vk83/configurations/inputs/access-esm1p5/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2020.05.19/Ndep_1850_ESM1.anc
Ancillary file contains header components other than the row/column dependent constants - these should be set to "None" for Ancillary files
  warnings.warn(msg)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
* (CUMF-II) Module Information *
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

mule       : /g/data/xp65/public/apps/med_conda/envs/analysis3-25.09/lib/python3.11/site-packages/mule/__init__.py (version 2024.11.1)
um_utils   : /g/data/xp65/public/apps/med_conda/envs/analysis3-25.09/lib/python3.11/site-packages/um_utils/__init__.py (version 2024.11.1)
um_packing : /g/data/xp65/public/apps/med_conda/envs/analysis3-25.09/lib/python3.11/site-packages/um_packing/__init__.py (version 2024.11.1) (packing lib from SHUMlib: 2024111)


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
* CUMF-II Comparison Report *
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

File 1: /g/data/vk83/configurations/inputs/access-esm1p5/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2020.05.19/Ndep_1850_ESM1.anc
File 2: /g/data/tm70/pcl851/CMIP7/esm16_ancil/modern/pre-industrial/atmosphere/land/biogeochemistry/global.N96/2025.11.18/Ndep_1850_cmip7.anc
Files DO NOT compare
  * 11 differences in fixed_length_header (with 7 ignored indices)
  * 1 differences in integer_constants (with 0 ignored indices)
  * 0 differences in level_dependent_constants (with 0 ignored indices)

Compared 0/24 fields, with 0 matches
12 fields found in file 1 were not in file 2
12 fields found in file 2 were not in file 1

%%%%%%%%%%%%%%%%%%%%%%%
* fixed_length_header *
%%%%%%%%%%%%%%%%%%%%%%%
Components DO NOT compare (compared 249/256 values)
Ignored indices:
  Index 35 (t3_year)
  Index 36 (t3_month)
  Index 37 (t3_day)
  Index 38 (t3_hour)
  Index 39 (t3_minute)
  Index 40 (t3_second)
  Index 41 (t3_year_day_number)
Component differences:
  Index 21 (t1_year)                          differs - file_1:      0  file_2:   1850
  Index 24 (t1_hour)                          differs - file_1:      0  file_2:     12
  Index 27 (t1_year_day_number)               differs - file_1:      0  file_2: -32768
  Index 28 (t2_year)                          differs - file_1:      0  file_2:   1850
  Index 31 (t2_hour)                          differs - file_1: -32768  file_2:     12
  Index 34 (t2_year_day_number)               differs - file_1:      0  file_2: -32768
  Index 110 (level_dependent_constants_start) differs - file_1:    278  file_2: -32768
  Index 111 (level_dependent_constants_dim1)  differs - file_1:      1  file_2: -32768
  Index 112 (level_dependent_constants_dim2)  differs - file_1:      4  file_2: -32768
  Index 150 (lookup_start)                    differs - file_1:    282  file_2:    278
  Index 161 (data_dim1)                       differs - file_1: 172032  file_2: 337920
...

[MartinDix]

Looks good now. See #93 (comment)

I was thrown for a while by the 1850 differences between 1-1 and 1-2.

[atteggiani]

Thanks Paul. Looks good.
I left a couple small comments.