-
Notifications
You must be signed in to change notification settings - Fork 0
Home
Chris Smith edited this page May 23, 2024
·
10 revisions
- If adding a new constraint set to an existing calibration mechanism and nothing else changes, this is a micro version (e.g. 1.0 to 1.0.1). There is no need to create new folders under this version; everything that applies to v1.0 would be valid for v1.0.1, v1.0.2 etc. Most updated will likely fall into this category.
- If an update or tweak to an existing calibration mechanism would change previously submitted results, this is a minor version. We would create a new folder for these inputs and results. v1.1 would differ from v1.0 for the same constraints.
- If a new calibration strategy is completely different and would break backward compatibility, this is a major version (e.g. 2.0).
Document major and minor versions here.
TO DO
- Fix for the BC on snow distribution which was wrong in v1.2.
- Volcanic forcing uses IGCC until 2022
- Temperature uses IGCC until 2022
- Slight updates to forcing scaling for asymmetric distributions to make them skewnormal rather than halfnormal
- Quantile mapping from 4xCO2 ERF derived from models to the CO2 ERF scaling factor (rather than trial and error estimate)
- Issue fixed with calculation of TCRE
- Output calibration field names have been updated to make it clear which part of the model it belongs to
As v1.0 with following fixes and improvements:
- Aerosol cloud interactions recalibrated to 13 (up from 11) CMIP6 models, and calibration code based on APRP fixed (part of the climateforcing package).
- NOx emissions from RCMIP have been corrected, where biomass burning emissions were in units of NO and agricultural and fossil emissions were in units of NO2, but no conversion was made in RCMIP. This affects historical calibrations for aerosol direct forcing, ozone forcing, and methane lifetime.
- Methane lifetime now reports the correct base lifetime for 1750, which is used in all projections.
- Ozone forcing calibration brought inside the code, rather than using AR6 results.
- 1.5 million prior ensemble
- Climate response calibrated on 49 abrupt-4xCO2 experiments from CMIP6 and sampled using correlated kernel density estimates
- Methane lifetime calibrated on 4 AerChemMIP experiments for 1850 and 2014 (Thornhill et al. 2021a, 2021b). Unlike other variables which are sampled around some prior uncertainty, only the best estimate historical calibration is used.
- Carbon cycle uses the parameters from Leach et al. 2021 calibrated for FaIR 2.0.0 using 11 C4MIP models.
- Aerosol cloud interactions depend on SO2, BC and OC, using calibrations from Smith et al. 2021a (developed for AR6) using 11 RFMIP and AerChemMIP models, with a prior of -2 to 0 W/m2.
- Aerosol radiation interactions use prior values from AR6 Ch6, with a factor of two uncertainty for each species and a prior in the range of -0.6 to 0.0.
- Ozone uses the same coefficeints as AR6 (Smith et al. 2021b).
- Effective radaitive forcing uncertainty follows the distributions in AR6.
Document micro versions here.
- GCP CO2 (to 2023), CEDS now to 2022, all other emissions as in all-2022
- harmonization done to 2022 for non-CO2 species, 2023 for CO2
- GMST constraint to 2023 and recent warming period to 2004-2023
- CO2 concentrations constraint in 2023
- GCP CO2 (to 2022), PRIMAP-hist v2.5.1 other GHGs (to 2022), GFED+CEDS SLCFs (2022, except CEDS which is 2019 and COVID-extended), inverted concentrations for Montreal gases
- no aviation ERF effect (for consistency with WG3)
- IGCC 2022 climate constraints based on ssp245
- RCMIP historical emissions (with corrected NOx)
- no aviation ERF effect (for consistency with WG3)
- IGCC 2022 climate constraints based on ssp245
Same as v1.2.0, implementing the BC on snow distribution fix.
Version suitable for AR6 database assessment. Same constraints/emissions as v1.2.0, except:
- contrails are removed from past and future as in AR6 NOx emissions from aviation are not available
- the correction for NOx being the wrong units has been reversed, to allow calibration/contraint on AR6 database emissions where biogenic NOx emissions are not available
Same constraints/emissions as v1.0, except:
- RMSE constraint is now 0.17°C (up from 0.16°C).
- RMSE constraint compares the observed GMST (on midyears) to the modelled temperature (at the year ends). The rationale is that the forcing has had chance to act on modelled temperatures at year n+1.
- CO2 constraint is updated to 2022 (from IGCC), and uses the average of the 2022 and 2023 timebounds.
- emissions-driven SSP2-4.5 constraint has been dropped.
- Ocean heat content constraint updated to 2020 (from 2018), using IGCC
- Temperature constraint uses 2003-2022 from IGCC (updated from 1995-2014 from AR6)
Same constraints/emissions as v1.0.2.
Same constraints/emissions as v1.0.
Same as v1.0.1 except:
- land use forcing from cumulative CO2 emissions was not consistent with the AR6 assessment because of the change in CO2 emissions dataset and has been fixed.
Same as v1.0, except:
- CO2 FFI and AFOLU emissions are from Global Carbon Project (Friedlingstein et al. 2022), up to and including 2022 which is an estimate.
- SSP CO2 emissions are harmonized, with 2021 as the harmonization year. So 2022 is the first year in which scenarios may differ.
- 1001-member posterior (deliberately chosen).
- Emissions and concentrations from RCMIP (i.e. based on CMIP6)
- Temperature from AR6 WG1 (1850-2020, mean of 4 datasets), constrained using ssp245 projections beyond 2014.
- ssp245 projections for 2081-2100.
- Ocean heat content from AR6 WG1 (1971-2018), linear.
- two step constraining procedure used: first RMSE of less than 0.16K, then 6-variable distribution fitting.
- Aerosol ERF, ERFari and ERFaci as in AR6 WG1
- Friedlingstein et al. 2022: https://doi.org/10.5194/essd-14-4811-2022
- Leach et al. 2021: https://doi.org/10.5194/gmd-14-3007-2021
- Smith et al. 2021a: https://doi.org/10.1029/2020JD033622
- Smith et al. 2021b: https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_FGD_Chapter07_SM.pdf
- Thornhill et al. 2021a: https://doi.org/10.5194/acp-21-853-2021
- Thornhill et al. 2021b: https://doi.org/10.5194/acp-21-1105-2021