Add recipe for seasonal cycle and time series of Arctic/Antarctic sea ice area (REF) #3891

axel-lauer · 2025-01-29T08:58:17Z

Description

This PR add the recipe ref/recipe_ref_sea_ice_area_basic.yml to calculate the seasonal cycle of Arctic and Antarctic sea ice area as well as time series of the Arctic September and Antarctic February sea ice area for REF.

The new recipe requires

the updated reference dataset OSI-450 (Update CMORizer for OSI-450 aka OSISAF/CCI (sea ice concentration) #3963)
Closes Arctic/Antarctic sea ice extent (seasonal cycle) REF #3890

Checklist

It is the responsibility of the author to make sure the pull request is ready to review. The icons indicate whether the item will be subject to the 🛠 Technical or 🧪 Scientific review.

🛠 This pull request has a descriptive title
🛠 Code is written according to the code quality guidelines
🛠 Documentation is available
🛠 Tests run successfully
🛠 The list of authors is up to date
🛠 Any changed dependencies have been added or removed correctly
🛠 All checks below this pull request were successful

New or updated recipe/diagnostic

🧪 Recipe runs successfully
🧪 Recipe is well documented
🧪 Figure(s) and data look as expected from literature
🛠 Provenance information has been added

TomLav · 2025-03-13T12:20:44Z

Hello, I am the PI for the ESACCI-SEAICE dataset. I am also the PI for another dataset (OSISAF/CCI) that is longer and more up-to-date (although coarser resolution).

I would suggest that the OSISAF/CCI dataset is used as a reference dataset by REF. Can I help getting the OSISAF/CCI dataset as reference for the two REF sea-ice metrics in ESMValTool?

The file format of ESACCI-SEAICE and OSISAF/CCI are identical, except for the filename convention, and the grid spacing (12.5km for ESACCI-SEAICE, and 25km for OSISAF/CCI).

axel-lauer · 2025-03-21T12:22:11Z

I am adding sea ice area to the recipes. I am also running an updated version of the OSI-450 CMORizer to create the OSISAF/CCI v3.0 dataset for this recipe.

TomLav · 2025-03-24T09:24:37Z

Hello @axel-lauer,

Thank you for your work bringing our data into ESMValTool

I do not know ESMValTool enough to find the information myself in the code, but I am wondering:

does this recipe start from daily SIC files, then computes daily sea-ice area, before averaging these to monthly SIA?
is the full range of sea-ice concentration considered when computing SIA or only SIC>=15% (it is better to have the full range)
is the recipe limited to 1995-2014? is it a REF or CMIP standard? the OSISAF/CCI dataset covers 1979-present
could the resulting SIA be referred / labelled with the "OSISAF/CCI" name?

axel-lauer · 2025-03-24T12:24:25Z

Thank you for your work bringing our data into ESMValTool

Thanks for pointing us to the new dataset. I just opened a pull request to update the OSISAF/CCI (aka OSI-450) from version 2 to version 3 (#3963).

I do not know ESMValTool enough to find the information myself in the code, but I am wondering:

does this recipe start from daily SIC files, then computes daily sea-ice area, before averaging these to monthly SIA?

The REF recipe uses the monthly means directly.

is the full range of sea-ice concentration considered when computing SIA or only SIC>=15% (it is better to have the full range)

At first, we used sea ice extent (SIC >= 15%) but we now changed this to sea ice area (calculated as the sum of the sea ice fraction in each grid cell multiplied by the grid cell area).

is the recipe limited to 1995-2014? is it a REF or CMIP standard? the OSISAF/CCI dataset covers 1979-present

The time period 1995-2014 is currently used for calculation of the average seasonal cycle and matches the last 20 years of data from the historical CMIP6 simulations. This could be adjusted once the first CMIP7 data are available. The newly added time series of Arctic September and Antarctic March sea ice area cover the full period of the OSISAF/CCI dataset (1979-2020).

could the resulting SIA be referred / labelled with the "OSISAF/CCI" name?

Sure, I changed this in the plots produced by basic sea ice diagnostics for REF (recipe_ref_sea_ice_area_basic.yml) (#3891).

TomLav · 2025-03-24T12:41:12Z

The newly added time series of Arctic September and Antarctic March sea ice area cover the full period of the OSISAF/CCI dataset (1979-2020).

FYI: the OSI-450 CDR covers 1979-2020. It is extended from 01.01.2021 to present with our Interim CDR. The Interim CDR has codename OSI-430, its sea-ice area can be appended to OSI-450 under the label "OSISAF/CCI". The OSI-430 files have exactly the same format as OSI-450, but are found in a separate location:

https://thredds.met.no/thredds/catalog/osisaf/met.no/reprocessed/ice/conc_cra_files/monthly/catalog.html
ftp://osisaf.met.no/reprocessed/ice/conc-cont-reproc/v3p0/monthly

This in case you want to append more recent years to the SIA timeseries.

The newly added time series of Arctic September and Antarctic March

We usually consider that February is the month with minimum yearly sea-ice area in the Antarctic (but March might be a decision by CMIP or REF).

TomLav · 2025-03-24T12:54:12Z

does this recipe start from daily SIC files, then computes daily sea-ice area, before averaging these to monthly SIA?

The REF recipe uses the monthly means directly.

The original monthly SIC data is on equal area 25x25km grid, so the area of all grid cells is 625km^2. Also, the grid cells with a valid (non-NaN) SIC have no fractional land, so the whole (625km^2 * SIC) contributes to SIC (no need to weight with Ofx).

Sorry if the above is obvious, I just want to double-check the computation that goes to REF.

axel-lauer · 2025-03-24T13:10:51Z

FYI: the OSI-450 CDR covers 1979-2020. It is extended from 01.01.2021 to present with our Interim CDR. The Interim CDR has codename OSI-430, its sea-ice area can be appended to OSI-450 under the label "OSISAF/CCI". The OSI-430 files have exactly the same format as OSI-450, but are found in a separate location:

So far, we did not use ICDR values as we were afraid it would be quite hard to keep track of which actual data has been used in case the ICDR record gets updated.

The newly added time series of Arctic September and Antarctic March

We usually consider that February is the month with minimum yearly sea-ice area in the Antarctic (but March might be a decision by CMIP or REF).

Thanks for spotting this. I just changed the time series of Antarctic sea ice area from March to February (5d26b08).

axel-lauer · 2025-03-24T13:20:29Z

does this recipe start from daily SIC files, then computes daily sea-ice area, before averaging these to monthly SIA?

The REF recipe uses the monthly means directly.

The original monthly SIC data is on equal area 25x25km grid, so the area of all grid cells is 625km^2. Also, the grid cells with a valid (non-NaN) SIC have no fractional land, so the whole (625km^2 * SIC) contributes to SIC (no need to weight with Ofx).

The data remain on their native 25x25 km grid. For irregular grids (i.e. 2-dim latitude and longitude values), ESMValTool usually simply uses the grid cell areas (here "areacello") provided for the particular grid for calculations instead of trying to estimate the grid cell areas. Even though in case of OSISAF/CCI, all values of the field areacello are constant (625 km^2), having such a "fx" file makes running ESMValTool easier.

TomLav · 2025-03-24T13:42:40Z

Thank you for answering my questions @axel-lauer .

In case you want to check the results of the present recipe, we distribute a monthly sea-ice area product from OSI-450/OSI-430:

Others than that I have nothing to add :)

esmvaltool/recipes/ref/recipe_ref_sea_ice_area_basic.yml

hb326 · 2025-03-27T12:08:03Z

esmvaltool/recipes/ref/recipe_ref_sea_ice_area_basic.yml

+            pyplot_kwargs:
+              title: 'Sea ice area (million km2)'
+            plot_kwargs:
+              'MPI-ESM1-2-LR':


Do we want this to be hard-coded? It probably would need to be a little more flexible when implemented in the REF so that it just refers to the model that is used.

Good point. Changed here: 87b2ad5

hb326

Thanks @hb326 for the review. I addressed your three comments in this commit: 87b2ad5

hb326

@axel-lauer: thanks for the changes!
Looks good to me now.
👍

schlunma

Thanks Axel, recipe looks good and works perfectly fine 🚀

Could you please solve the merge conflict?

axel-lauer · 2025-03-28T08:54:25Z

Could you please solve the merge conflict?

Done!

TomLav · 2025-03-28T09:09:39Z

Just a comment to say that I am positively impressed by the process of developing this metric, and the fact that it involved an assigned scientific reviewer (and a scientific review period).

This is probably not the right place to make such a suggestion, but I would still put down that (when possible) the scientific review should also involve the PI of the observation dataset. Both because we can help accessing and using the data, but also because we are then aware of the use in ESMValTool and can inform when changes or new version are made.

As a matter of fact, I can inform you that a new (bug-fix) release of the OSISAF/CCI dataset is underway that will (marginally) change the sea-ice areas computed with this recipe. This will be v3.1 (OSI-450-a1) that should be released in the next month. The sea-ice area computed by the current recipe (from v3.0) are perfectly ok for scientific exploitation by REF and other users. But the new ones (from v3.1) will be marginally better.

It is fine for me if you want to close this PR and I open an issue later when the OSISAF/CCI v3.1 is released.

axel-lauer · 2025-03-28T10:22:27Z

This is probably not the right place to make such a suggestion, but I would still put down that (when possible) the scientific review should also involve the PI of the observation dataset. Both because we can help accessing and using the data, but also because we are then aware of the use in ESMValTool and can inform when changes or new version are made.

That's a very good idea. In most cases, though, dataset PIs are unfortunately not available for this task. So thanks for sharing your expertise with us here!!

As a matter of fact, I can inform you that a new (bug-fix) release of the OSISAF/CCI dataset is underway that will (marginally) change the sea-ice areas computed with this recipe. This will be v3.1 (OSI-450-a1) that should be released in the next month. The sea-ice area computed by the current recipe (from v3.0) are perfectly ok for scientific exploitation by REF and other users. But the new ones (from v3.1) will be marginally better.

Thanks for letting us know. From the ESMValTool side, it would be very easy to switch from v3.0 to v3.1 once available. But I am not sure the tight REF schedule allows for waiting for v3.1. We will check how it would work to update recipes or reference datasets at a later stage.

It is fine for me if you want to close this PR and I open an issue later when the OSISAF/CCI v3.1 is released.

That would be great and we would be very happy to hear about v3.1 once the dataset is released.

schlunma

Great, thanks @axel-lauer and everyone for contributing. This can be merged 🚀

* main: (43 commits) use python-only environment file for Circle build documentation (#3981) Added option to plot horizontal lines in monitoring line plots (#3977) [Condalock] Update Linux condalock file (#3974) Run esmvalbot test PR twice a month (#3976) unpin upper limit `r-base` (#3971) [Condalock] Update Linux condalock file (#3968) Update CMORizer for OSI-450 aka OSISAF/CCI (sea ice concentration) (#3963) Add recipe for seasonal cycle and time series of Arctic/Antarctic sea ice area (REF) (#3891) use Python-only environment for ReadTheDocs builds (#3955) [Condalock] Update Linux condalock file (#3961) Create scatterplots for the REF (#3923) [Condalock] Update Linux condalock file (#3957) Unsafe and manual fixes Automatic fixes Replace flake8, yapf, and isort with ruff Add recipe and diagnostic for ZEC (Zero Emissions Commitment) (#3897) Update ESACCI-OZONE CMORizer(formatter and downloader) for REF (#3899) Update citation and changelog for version 2.12 (#3945) use mamba>=2 for Circle CI upstream development test (#3949) Force install mamba2 in CircleCI and add ipython and pin it to <9.0 (#3947) ...

axel-lauer added 2 commits November 27, 2024 14:25

added REF recipe for seasonal cycle of NH sea ice extent

83d5b62

cleaned up recipe_ref_siextent_seasonal.yml

2da670b

This was referenced Jan 29, 2025

Arctic/Antarctic sea ice extent (seasonal cycle) REF #3890

Closed

Fix derivation for sea ice extent (siextent) ESMValGroup/ESMValCore#2648

Open

hb326 added REF Important for the CMIP Rapid Evaluation Framework (REF) in scientific review labels Feb 3, 2025

axel-lauer added 2 commits February 26, 2025 12:13

snapshot 2025-02-26

0c0ffb9

Merge branch 'main' into ref_sea_ice_extent

0eb375c

added docu for seaice extent recipe (REF)

16a9cc5

axel-lauer changed the title ~~Add recipe for seasonal cycle of Arctic/Antarctic sea ice extent (REF)~~ Add recipe for seasonal cycle of Arctic/Antarctic sea ice area & extent (REF) Mar 21, 2025

axel-lauer marked this pull request as draft March 21, 2025 12:27

axel-lauer added 3 commits March 21, 2025 13:56

added sea ice area

b67057a

update recipe_ref_sea_ice_seasonal.yml

f070ea2

removed sea ice extent from recipe

7077b56

axel-lauer added 2 commits March 24, 2025 12:28

added time series of minimum sea ice area

a20b84f

removed trailing whitespaces in recipe_ref.rst

a442093

axel-lauer changed the title ~~Add recipe for seasonal cycle of Arctic/Antarctic sea ice area & extent (REF)~~ Add recipe for seasonal cycle and time series of Arctic/Antarctic sea ice area (REF) Mar 24, 2025

axel-lauer mentioned this pull request Mar 24, 2025

Update CMORizer for OSI-450 aka OSISAF/CCI (sea ice concentration) #3963

Merged

11 tasks

axel-lauer marked this pull request as ready for review March 24, 2025 12:24

changed from March to February for Antartic sea ice area

5d26b08

axel-lauer requested a review from hb326 March 24, 2025 13:22

updated reference file osi-450.bibtex

9bce723

bouweandela mentioned this pull request Mar 25, 2025

Antartic/Artic sea ice area seasonal cycle Climate-REF/climate-ref#199

Closed

TomLav mentioned this pull request Mar 25, 2025

Prioritized set of observations for inclusion in obs4MIPs until the end of March Climate-REF/climate-ref#181

Closed

9 tasks

hb326 reviewed Mar 27, 2025

View reviewed changes

esmvaltool/recipes/ref/recipe_ref_sea_ice_area_basic.yml Show resolved Hide resolved

hb326 reviewed Mar 27, 2025

View reviewed changes

esmvaltool/recipes/ref/recipe_ref_sea_ice_area_basic.yml Show resolved Hide resolved

hb326 reviewed Mar 27, 2025

View reviewed changes

hb326 requested changes Mar 27, 2025

View reviewed changes

update recipe_ref_sea_ice_area_basic.yml

87b2ad5

hb326 approved these changes Mar 27, 2025

View reviewed changes

hb326 added approved by scientific reviewer and removed in scientific review labels Mar 27, 2025

schlunma self-requested a review March 27, 2025 16:46

schlunma added this to the v2.13.0 milestone Mar 28, 2025

schlunma reviewed Mar 28, 2025

View reviewed changes

Merge branch 'main' into ref_sea_ice_extent

f8ee658

pre-commit

ae91e68

schlunma approved these changes Mar 28, 2025

View reviewed changes

schlunma merged commit eb2972a into main Mar 28, 2025
8 checks passed

schlunma deleted the ref_sea_ice_extent branch March 28, 2025 10:38

schlunma added the approved by technical reviewer label Mar 28, 2025

Uh oh!

Add recipe for seasonal cycle and time series of Arctic/Antarctic sea ice area (REF) #3891

Add recipe for seasonal cycle and time series of Arctic/Antarctic sea ice area (REF) #3891

Uh oh!

Conversation

axel-lauer commented Jan 29, 2025 • edited by schlunma Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist

New or updated recipe/diagnostic

Uh oh!

TomLav commented Mar 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

axel-lauer commented Mar 21, 2025

Uh oh!

TomLav commented Mar 24, 2025

Uh oh!

axel-lauer commented Mar 24, 2025

Uh oh!

TomLav commented Mar 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TomLav commented Mar 24, 2025

Uh oh!

axel-lauer commented Mar 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

axel-lauer commented Mar 24, 2025

Uh oh!

TomLav commented Mar 24, 2025

Uh oh!

Uh oh!

Uh oh!

hb326 Mar 27, 2025

Choose a reason for hiding this comment

Uh oh!

axel-lauer Mar 27, 2025

Choose a reason for hiding this comment

Uh oh!

hb326 left a comment • edited by axel-lauer Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

hb326 left a comment

Choose a reason for hiding this comment

Uh oh!

schlunma left a comment

Choose a reason for hiding this comment

Uh oh!

axel-lauer commented Mar 28, 2025

Uh oh!

TomLav commented Mar 28, 2025

Uh oh!

axel-lauer commented Mar 28, 2025

Uh oh!

schlunma left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

axel-lauer commented Jan 29, 2025 •

edited by schlunma

Loading

TomLav commented Mar 13, 2025 •

edited

Loading

TomLav commented Mar 24, 2025 •

edited

Loading

axel-lauer commented Mar 24, 2025 •

edited

Loading

hb326 left a comment •

edited by axel-lauer

Loading