Physical Modelling

Model description

Modelling of spatially and temporally resolved CO2 concentrations is realized with the modelling system COSMO-GHG. COSMO stands for Consortium for Small-Scale Modelling and is the operational weather forecast model used by MeteoSwiss. The extension of a greenhouse gas (GHG) module has been developed by Empa and is directly coupled to the latest version of COSMO. The model allows simulating the emissions and transport of CO2 (and other greenhouse gases) for whole Switzerland.


Top left: Background concentration. Top right: Fraction by anthropogenic emissions. Bottom left: Fraction by respiration (RA). Bottom right: Fraction by gross photosynthetic production (GPP)

The model is set up for a domain covering Switzerland at a horizontal resolution of 1 km x 1 km, allowing to resolve detailed concentration patterns resulting from surface fluxes and atmospheric transport. The main strength of the modelling system is that different contributions to the total atmospheric CO2 are separated (anthropogenic sources by type, biogenic uptake and release, CO2 from outside domain). Anthropogenic emissions within Switzerland are taken from the high resolution (500 m) CarboCount-CH inventory, whereas those outside Switzerland are based on the TNO GHGco inventory (6 km). Biogenic CO2 fluxes are calculated online using the Vegetation Photsynthesis and Respiration Model (VPRM) taking air temperature and fraction of absorbed photosynthetically active radiation into account, which are directly available in COSMO. Further inputs are surface reflectance fields from ongoing satellite observations.

The picture on the left shows a snapshot of those CO2 contributions simulated by COSMO-GHG.

Simulations are carried out within a quasi-operational system, where a minimal amount of input data has to be pre-processed.

Example simulation

COSMO-GHG simulation of total CO2 concentration (left panel), biospheric component (top right) and from anthropogenic sources (bottom right). All fields are at the lowest model layer.

Simulation time period is March 2019. After a stormy period in the first half of March, a high pressure system dominated during the second half. This also becomes apparent in the concentration levels of CO2.

Software used for video: Python and ffmpeg.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License