The CATRINE project is proposing a new model inter-comparison protocol that aims to assess the importance of high resolution for modelling of CO2 from fossil fuel anthropogenic emissions.
Monitoring the greenhouse gases carbon dioxide (CO2) and methane (CH4) from space has made great strides over the past decade, in particular in terms of imaging capability (e.g. the NASA's Orbiting Carbon Observatory-2 (OCO-2) and OCO-3, TROPOMI instrument on board Copernicus Sentinel-5P), as well as emerging retrieval capabilities at higher spatial resolution (e.g. Copernicus Sentinel-2 or from Greenhouse Gases Satellite GHGSat) and near-future planned missions such as the Copernicus CO2 Monitoring Mission (CO2M) will extend the imaging capability of CO2 to a level comparable to the existing one for CH4 from TROPOMI. This new wealth of CO2 and CH4 satellite retrievals pushes the application of such satellite data towards the inference of anthropogenic emissions and absorptions, with target spatial scales much finer than the traditional zonal or subcontinental scales of global atmospheric inversions. In this regard, the current generation of global atmospheric transport models that operate primarily at resolutions coarser than one degree, risks obsolescence because most models cannot exploit the higher resolution information in which most of the anthropogenic signals reside.
The protocol proposed by CATRINE aims for tracer transport simulations at high resolution with respect to what is commonly used in atmospheric inversion systems (i.e. resolutions that are at least about 1 degree × 1 degree resolution in the Tropics, or higher). On the vertical, we put a simple lower limit of 50 layers between the surface and the top of the atmosphere, assuming that it implies sufficient refinement of the planetary boundary layer and around the tropopause for constructive intercomparison. A first round of tracer transport simulations is currently being prepared at such “high resolution” under a common protocol and compared together as well as with independent observations. The simulations include carbon dioxide CO2 as the key tracer, sulphur hexafluoride (SF6) for large-scale transport, the radon isotope 222Rn for short-range transport, and an anthropogenic CO2 tracer is used to study the transport of the imposed fossil CO2 emission fields.
In addition, humidity can also be used by online models as an optional tracer given its importance and the availability of radiosonde observations. The target period goes from 1 January 2022 at 00:00 UTC until 31 December 2023 at 24:00 UTC. This period is preceded by one month spin up, starting from initial conditions provided by the protocol.
As well as initial conditions, the protocol also provides surface fluxes for all the tracers. Sensitivity simulations with different horizontal and vertical resolutions using the same model are encouraged. Indeed, the model vertical profiles will be particularly looked at in this exercise.
We expect that the evaluation for several tracers using observed profiles from tall towers, aircraft and AirCore soundings will shed some light into the current discrepancies of atmospheric inversions that assimilate column retrievals or air-sample measurements.
Metrics will be defined to score model performance, considering factors like vertical resolution. Diagnostics will focus on transport errors' impact on the seasonal cycle at different layers and total column, aiding error attribution. The high-resolution simulation database, which will be made publicly available, will have broad applications beyond model evaluation, potentially providing a reference for model uncertainty quantification and AI training.
This initiative, originating from the EU CATRINE project, seeks global participation from other partners to contribute expertise and models.
More details of the protocol can be found in D7.1 (CATRINE-D7.1-V1.pdf)
