Produces apples-to-apples comparison between model and satellite columns
James East
2021-09-27
Needs to do:
- Get model data at right time step (nearest neighbor)
- Get model pressure levels
- interpolate SW to model, accounting for tropopause
- Compute model VCD (simple) and partial columns
-
$\Omega_V^M$ [ppm] = $\frac{MW_{air}Av1e-4}{1e6} \sum_{surface}^{tp} \frac{\rho (z)}{\Delta H (z)}*C_{NO_2}(z)$
-
- Apply SW to model partial columns and sum for model SCD
-
$\Omega_S^M$ =$\sum_{surface}^{tp} \Omega_V^M(z)*w(z)$
-
- Compute model AMF (see Cooper et al. 2020 ACP)
- $\text{AMF}M=M_G * \sum{surface}^{tp}w(z)\frac{n(z)}{\Omega_V^M}$
$M_G=\sec \theta_o + \sec \theta$ -
$\theta_o$ is solar zenith angle -
$\theta$ is viewing zenith angle - see OMNO2G README p. 28 (https://aura.gesdisc.eosdis.nasa.gov/data/Aura_OMI_Level2/OMNO2.003/doc/README.OMNO2.pdf)
- Compute satellite VCD with CMAQ profile (see Duncan et al. 2014 Section S.5 and Palmer et al. 2001 eqs 13-15 and discussion)
$\Omega_V^{O'} = \frac{\Omega_V^O * \text{AMF}_{\text{OMI}}}{\text{AMF}_M}$
- Add OMI VCD with CMAQ profile to file
- Add CMAQ VCD & SCD to file
- Change dims to match IOAPI expectations for easier future comparison
- 1 retrieval at a time for now, loop later
Assumptions:
- Tropopause cutoff based on OMI tropopause definition
- All sat fields regridded to CMAQ grid first
- Surface pressure set to CMAQ for SW application
- Cloud filtering based on OMI cloud fraction
- Conservative horizontal regridding (requires nco >= 5.0.1)