CH4 underestimation in CMAQv5.3.2 using ECH4 emissions and IC/BC

CMAQ Chemistry and Aerosols
1

Hi all,

I am attempting to simulate CH4 using CMAQv5.3.2.
I have prepared CH4 emission inputs and properly configured the initial and boundary conditions (IC/BC), but the resulting ECH4 concentrations in the model are significantly lower than observed.

For example, when the observed CH4 concentration is about 2 ppm, the model output is only around 50 ppb.

From the mech_cb6r3_ae6_aq.def file, it seems that the CH4 species is associated with a constant background concentration via the parameter ATM_CH4 = 1.85 ppm, which likely serves as the background CH4 level used as an OH sink.

On the other hand, ECH4 appears to be the species that is actually calculated in the model using CH4 emissions from the input files.

As a result, the ECH4 variable in the CONC output is much lower than expected, and does not accumulate over time as one might anticipate for a long-lived species like CH4.
Compared with other pollutants such as CO, ECH4 appears to decay too quickly.

I have two questions:

  1. The only chemical reaction defined for ECH4 in the mech_cb6r3_ae6_aq.def file is as follows:
    <R216> ECH4 + OH = MEO2 + RO2 # 1.85E-12@1690;
    Could this reaction be the main reason ECH4 is being depleted too rapidly in the model?

  2. Is there a way to treat CH4 in CMAQ like other pollutants (e.g., CO, NO2, PM10) — that is, to model its spatial and temporal distribution using both emissions and a defined background concentration, instead of separating CH4 (fixed background) and ECH4 (emitted)?

Thank you in advance for your insights.

Best regards,
Jeonghyeok

2

Hi Jeonghyeok,

Below are the general steps I would suggest to model methane explicitly in CMAQ. Some of these I guess you may have already done. Essentially, you could treat the ECH4 species as total methane while removing any influence of the fixed global background methane (ATM_CH4=1.85 ppm) on OH.

  1. Prepare initial and boundary conditions for your domain containing the ECH4 variable to represent background methane concentrations. You could use an assumption of a constant global background, though you might get better results if you obtain methane concentrations for your IC/BC from a global model such as GEOS-Chem.
  2. In CCTM/src/MECHS/cb6r3_ae6_aq/mech_cb6r3_ae6_aq.def delete the reaction <R124> CH4 + OH = MEO2 + RO2 # 1.85E-12@1690; to prevent the fixed global background methane from removing OH. You may also want to remove the reaction <CL10> CL + CH4 = HCL + MEO2 + RO2 # 6.6E-12 @ 1240; or maybe change the CH4 to ECH4 in this reaction if you want to account for losses of ECH4 by chlorine. After modifying this file, you will need to generate updated RXNS_DATA_MODULE.F90 and RXNS_FUNC_MODULE.F90 files along with updated ebi solver files (files in CMAQ/CCTM/src/gas/ebi_cb6r3_ae6_aq) by running the chemmech and create_ebi utilities for the changes to take effect. There are some options in more recent versions of CMAQ to have the bldit_cctm.csh script set up and run these utilities, but I do not believe this capability was in CMAQv5.3.2.
  3. Make sure you have accounted for all methane sources in the emissions you have prepared, including methane from sources such as wetlands, livestock, and landfills which may not be considered in a standard CMAQ simulation. These types of sources may or may not be important depending on your specific domain I suppose. For reference, see Table 1 of Varon et al. 2022 or Table 3 of Estrada et al. 2024 for a list of emission sources that are included in the GEOS-Chem methane simulation.
  4. Consider turning off deposition processes for ECH4 as I think the default deposition for methane may be unrealistically high in CMAQ. You can turn off deposition by editing the line for ECH4 in CCTM/src/MECHS/cb6r3_ae6_aq/GC_cb6r3_ae6_aq.nml. You should just be able to change the DDEP and WDEP columns for the ECH4 line from 'Yes' to ''. Then you can double check in the CCTM_DRYDEP and CCTM_WETDEP1 outputs that you no longer have ECH4 deposition. You may want to test with deposition on and off as a sensitivity to see how it impacts your results.

Hopefully this can get you off to a good start at least.

4 Likes
3

Thank you for your kind and detailed explanation.
I believe the reason why ECH4 concentrations decrease faster than expected is likely due to the unrealistically high default deposition, as you mentioned in point 4.
I’ll try modifying the code as you suggested.

Thank you again,
Jeonghyeok