NO_x climate impacts

Emitted nitrogen oxides (NO_x) have only a minor direct influence on the atmosphere’s radiation balance. However, through chemical reactions, the concentrations of other greenhouse gases are affected. In OpenAirClim, the climate impacts of the species ozone, methane, Primary Mode Ozone (PMO) and Stratospheric Water Vapour (SWV) are calculated. Due to different processes and properties of these species (e.g. different lifetimes), the methods for calculating the radiative forcings are different for each of these species as shown in following flowchart:

        flowchart LR
    inv["NO<sub>x</sub> emissions"]
    inv --> rf_o3["RF(O<sub>3</sub>)"]
    inv --> ch4["$$\mathrm{\Delta CH_4}$$"]
    ch4 --> rf_ch4["RF(CH<sub>4</sub>)"]
    rf_ch4 --> rf_pmo["RF(PMO)"]
    ch4 --> swv["$$\mathrm{\Delta SWV}$$"]
    swv --> rf_swv["RF(SWV)"]

In OpenAirClim, the climate impacts of emitted NO_x can be simulated using one of the two approaches: tagging or perturbation. The underlying response surfaces are different and have been created by different climate-chemistry model setups.

Tagging approach

Important

Our baseline development is built upon the tagging approach for NO_x climate impacts. Currently, NO_x response surfaces using the tagging approach are not yet validated and have a limited resolution. Therefore, these responses should currently not be used for scientific purposes. Meanwhile, the response surfaces using the perturbation approach can be used. Be aware that due to the non-linear NO_x chemistry, different results are expected when comparing both approaches. Particularly, the two approaches should not be mixed between species, e.g. O₃ and CH₄.

Perturbation approach

The worfklows for calculating the responses differ for the individual species. This is due to different species lifetimes and possible interdependencies between species. The workflows and the underlying response surfaces for the perturbation approach have been adpoted from the AirClim framework [11].

Ozone

Ozone has relatively short lifetimes and a direct relation between NO_x emission and resulting O₃ radiative forcing is applied as shown in the flowchart above.

In detail, the RF for a perturbation scenario is calculated by folding the emissions in the input inventory with the pre-calculated radiative forcings of the idealized emission regions. Here, the values of the pre-calculated RF are normalized using the imported NO data for each idealized emission region.

\[\mathrm{RF} = \sum_n E_n \sum_k \epsilon_k \frac{\mathrm{RF}(i_k, j_k)}{\mathrm{N}(i_k, j_k)}\]

where:

\(E_n\) are the NO_x emission masses for each location in the inventory.
\(\epsilon_k\) are the weighting factors of the surrounding neighbours \((k = 1, .., 4)\) on the latitude and altitude dimensions for each emission location.
\(\mathrm{RF}(i_k, j_k)\) are the latitude and altitude dependent radiative forcings of the idealized emission regions.
\(\mathrm{N}(i_k, j_k)\) are the latitude and altitude dependent masses of imported NO of the idealized emission regions.

Methane

The lifetimes of methane are significantly longer than the one-year step size of the OpenAirClim simulations. Therefore, the workflow is different from the calculation procedure of ozone RF. According to the flowchart shown above, first the CH₄ concentration changes are calculated from the NO_x emissions which are used subsequently to evaluate radiative forcing.

The change in methane concentration is derived by regarding the difference of two differential equations [11] which is solved numerically in OpenAirClim:

\[\frac{d}{dt} \Delta C^{\mathrm{CH_4}} = \frac{\delta}{1+\delta} \tau_{\mathrm{CH_4}}^{-1} C_0^{\mathrm{CH_4}} - \frac{1}{1+\delta} \tau_{\mathrm{CH_4}}^{-1} \Delta C^{\mathrm{CH_4}}\]

where:

\(\Delta C^{\mathrm{CH_4}}\) is the change in methane concentration.
\(\delta\) is the relative change in lifetime.
\(\tau_{\mathrm{CH_4}}\) is the methane perturbation lifetime (here: 12 years).
\(C_0^{\mathrm{CH_4}}\) is the time dependent background methane concentration.

The relative lifetimes are evaluated on the basis of inventory years. Similary as for the ozone radiative forcings, relative lifetime changes have been pre-calculated for idealized emission regions, and an average \(\delta\) is computed by folding the emissions in the input inventory with pre-calculated lifetime changes of the idealized emission regions.

Finally, the radiative forcing is calculated from the \(\Delta C^{\mathrm{CH_4}}\) time series by applying the relation defined in Etminan et al. (2016, see their Table 1) [8].

PMO and SWV

The climate impacts of Primary Mode Ozone (PMO) and Stratospheric Water Vapour (SWV) are computed independently from the chosen approach (tagging vs. perturbation). As shown in the flowchart above, both species depend on the previously evaluated methane.

According to Dahlmann et al. (2016, see their appendix A.3) [7], the radiative forcing of PMO is calculated using a linear relationship to the forcing of methane:

\[\mathrm{RF}(\mathrm{PMO}) = 0.29 \cdot \mathrm{RF}(\mathrm{CH_4})\]

The methods for the computation of climate impacts from SWV are based on the work from Harmsen (2026) [12]. For details, refer to the documentation in Stratospheric Water Vapour Module.

NOx climate impacts