CEDS time series data finalized

[eleanororourke]

Visuals of the final CEDS CMIP7 time series from Steve and @rhoesly
Uploading Changes since v2024_07_08 release.zip…
Uploading Changes vs 2025_02_09 review data.zip…

Two sets of visuals:

1. Changes vs 2025_02_09 review data (that are posted below)

Most changes are in earlier years, but some changes in BC, OC, CO out to 2023 in some countries.

2. Changes since v2024_07_08 release

Comparisons vs CMIP6 release coming (we have to update comparison scripts to handle the different sectors).

We will be gridding the data reviewing some enhanced gridding diagnostics to do a final check for any issues. Hopefully gridded data will be in ESGF publication que within a couple days.

We will get the aggregate data on first, then the supplemental data (spectated VOCs, solid biofuel, and methane extension back to 1750).

[jkikstra]

Hi @eleanororourke, the links you posted unfortunately don't seem to work (it just links back to this discussion).
And Steve here is @ssmithClimate

[bethdingley]

Hopefully these files work now

1. Changes since v2024_07_08 release.zip
2. Changes vs 2025_02_09 review data.zip

[ssmithClimate]

And some additional global and regional comparisons with other inventories.
Compare_inventory_to_CEDS_BC.pdf
Compare_inventory_to_CEDS_CO.pdf
Compare_inventory_to_CEDS_NH3.pdf
Compare_inventory_to_CEDS_NMVOC.pdf
Compare_inventory_to_CEDS_NOx.pdf
Compare_inventory_to_CEDS_OC.pdf
Compare_inventory_to_CEDS_SO2.pdf

[jkikstra]

Those plots are very helpful!
Adding CO2, CH4, and N2O in the same format would be great too; because they are important for CMIP forcings.

[ssmithClimate]

The GHG emissions are there for this set:
Changes since v2024_07_08 release

but not the others - codes not updated to include GHGs yet in those.

[ssmithClimate]

Here are comparisons with CMIP6. The changes are more dramatic here, but are incremental from the previous couple CEDS data releases. Individual graphs, and as a zipped bundle.

(We are finalizing the CH4 emissions extension, that will be out shortly as well. Unlike CMIP6 ,the CMIP7 emissions extension will be annual, in the exact same format as the modern emissions.)

Comparison_Plots_SO2.pdf
Comparison_Plots_OC.pdf
Comparison_Plots_NOx.pdf
Comparison_Plots_NMVOC.pdf
Comparison_Plots_NH3.pdf
Comparison_Plots_CO2.pdf
Comparison_Plots_CO.pdf
Comparison_Plots_BC.pdf

summary difference with CMIP6.zip

[jkikstra]

CH4 extension; also N2O, or is that going to be different?

[ssmithClimate]

CH4 extension is on its way. We weren't planning on extending N2O, but we could. Current-day estimates for both are done.

[jkikstra]

I think a consistent dataset with all emissions going back to 1750 would be really great. Also then the emulator group doesn't have to infill it.

So yes, if possible, please!

[ssmithClimate]

@jkikstra Ok, we will produce N2O and CH4 extensions. (I just learned a lot about the history of nitric and adipic acid production in the last 24 hours....)

[jkikstra]

That's fantastic!

[ssmithClimate]

And here are the global data. I you see any anomalies or issues please speak up quickly! (For ScenarioMIP folks, the detailed data is on the ScenarioMIP drive.)

v_2025_03_11_global_emissions_by_fuel_Sector.zip

[ssmithClimate]

DOI for this aggregate data: 10.5281/zenodo.15001433
(Gridded data will have a different DOI of course.)

[vnaik60]

Hi @ssmithClimate, nice to see the aggregates, thanks!. Are you going to also release them here https://github.com/JGCRI/CEDS/releases? Also, we are eagerly looking forward to the gridded data, especially since Michaela et al are awaiting them (especially 1850 dataset) for running the models to produce ozone concentration and nitrogen deposition. Any ETA on the gridded files would be very helpful. Thanks!

[ssmithClimate]

Yes! Aggregate data is ready, will take a couple days to post on CEDS, but will get that to the IAM community today.

@jkikstra found a couple of anomalies in the aggregate data that we fixed (turns out they didn't impact the gridded data, but we wanted to make sure nothing else was impacted), and we found one other small issue with SO2 that we fixed. Aggregate data should be ready (HPC willing) for upload to ESGF today. All the supplemental files soon after (including new CH4 and N2O extension). The CH4 extension has been improved from CMIP6 with more sectoral resolution, so the results are a bit different (lower in the early years). I'll post something here on the CH4 and N2O extension.

[vnaik60]

Sounds great, thanks @ssmithClimate!

[ssmithClimate]

FYI: graph of the various CH4 extrapolations.

Our new historical values are lower because we've disaggregated the historical extrapolation more by sector. Which results in lower emissions. For example, the first sanitary landfills were in the 1930s, so solid waste sector emissions before that would have been very small.

RCP_CH4_Extension_Comparison.pdf

(Substantial uncertainties recently of course.)

[jkikstra]

Hi @ssmithClimate thanks for this new release.

I've mapped the new data onto harmonization sectors now, and made some plots comparing to the previous version.
Did not immediately see any big issues.

Here the plots, in case they would be informative to others:

• On the country-level (many plots!): ceds-update_all_percountry.pdf
• Globally aggregated (both per emissions species, and per sector): ceds-update_all_world.pdf
• Globally aggregated (totals): ceds-update-world-totals_all-species.pdf

Some significant updates that I noticed:

• Energy Sector (VOC energy sector quick rise and fall now gone)
• International Shipping (CO2 pre-1950 now higher)
• Waste (CH4 lower)
• RCO (CO2 bit higher; BC and OC rises now less steep; with OC and NH3 pre-industrial now at higher levels)
• CO (increased in russia, )
• CO2 (industrial sector pre-SU-collapse bump now gone)
• N2O (industrial sector bit lower with rebound now almost gone)
• NOx (Agriculture growth now 30% lower, energy sector also bit lower)

[ssmithClimate]

Yes, there are all changes that we noticed and can explain.

[chrisroadmap]

From 1990-2022 the total CO2 emissions have gone up by 4-5% - I wonder if this was a switch to a different dataset used in v2025_03_11 compared to v2024_07_08?

[ssmithClimate]

@chrisroadmap yes, we fixed an error in the natural gas emission factor. There's been a couple corrections over the years since CMIP6, taken together, we're similar in aggregate to where we were then!

[chrisroadmap]

ok cool, thanks!

[ssmithClimate]

Don't use the N2O Extension for final analysis (I don't think this is critical path for anything), just discovered what seems to be s machine-dependent small number bug that results in an artificial bump in the historical curve that I want to fix.

[ssmithClimate]

Looking at this a bit more, its not really an issue at the global level, so we'll leave it.