9:45 AM - 10:00 AM
[PCG19-04] Validation of a new atmospheric retrieval technique for the vertical profiles of Martian atmosphere by limb observations
Keywords:Mars, atmosphere, retrieval
Spatial-time variability of the vertical distribution of aerosols and trace gases is one of the basic information for Martian atmospheric studies. Aerosols can affect the trace gas distribution and their vertical transportation. Trace gas is also transported from the lower to the upper atmosphere. The vertical structure has strongly been investigated by solar occultation via ExoMars TGO. Solar occultation has an excellent S/N, but it is hard to cover the Local Time (LT) and horizontal distribution. Limb observations can complementarily provide wider LT and horizontal coverages for vertical profiles. However, it is not easy because the retrieval algorithms for limb observations require huge computation cost.
For Martian limb observations, we have tried to utilize the JACOSPAR package, a fast multiple-scattering retrieval tool originally developed for Earth observations. JACOSPAR considers a spherical-shell atmosphere, so that the code can simultaneously derive the vertical profile of dust and atmospheric components. It can do a fast calculation with the algorithm that computes the multiple scattering using the backward Monte Carlo and the dependent sampling approach to simultaneously simulate a radiance spectrum along multiple wavenumbers.
In this work, we did the validation of this code in the retrieval for vertical profile of dust and water ice density. We validated the forward model and the retrieval code. For the forward model, we compared it with other codes (DISORT and MITRA) with less assumption in their algorisms. In this result, JACOSPAR forward model shows the difference in brightness within 1% in limb geometry, which can be acceptable compared to the observational accuracy. In the validation of the retrieval, we checked the result of the dust and water ice densities retrieved from the model spectra formed by the forward model with model atmospheric profiles. We confirmed that the aerosol retrieval provided the accuracy within -30% to +120% when a priori aerosol density was enough large (column τ> 0.9 at 1 μm wavelength). We also confirmed that the retrieval may be difficult in the lower layers beyond nadir column of τ> 0.1 (τ along the limb: about 1 or more). We do further works in order to retrieve them with dust and water ice sizes, surface albedo, and gas densities.
We also surveyed the limb observation data of Mars Express (MEx) OMEGA and did a simple analysis without JACOSPAR. Two cases may show daily variation in the absorption depth of water vapor and water ice along the limb. In the daytime (12-16 LT) case, water ice sublimation and water vapor increase may occur. In the morning (6-8 LT) case, a rise in water vapor and water ice may be seen. The Mars Climate Database (MCD) with these cases showed that in the former the water ice and water vapor altitude change is limited in lower alitude layer, while in the latter the water vapor and water ice rises are not reproduced. These differences may be caused by the limitation of MCD. We will utilize JACOSPAR for vertical and horizontal profiles of aerosols and gases at once, which will lead to the studies of the vertical transport mechanisms.
This study is planned to contribute to the limb observations by InfraRed Spectrometer (MIRS) aboard JAXA Martian Moons eXploration (MMX) mission in the mid 2020s. Our fast retrieval tool will contribute to the MMX atmospheric observation covering wide disk and limb spectral imaging data sets, for the retrieval of the horizontal and vertical profiles of dust and water vapor over a wide range at hourly time scale.
For Martian limb observations, we have tried to utilize the JACOSPAR package, a fast multiple-scattering retrieval tool originally developed for Earth observations. JACOSPAR considers a spherical-shell atmosphere, so that the code can simultaneously derive the vertical profile of dust and atmospheric components. It can do a fast calculation with the algorithm that computes the multiple scattering using the backward Monte Carlo and the dependent sampling approach to simultaneously simulate a radiance spectrum along multiple wavenumbers.
In this work, we did the validation of this code in the retrieval for vertical profile of dust and water ice density. We validated the forward model and the retrieval code. For the forward model, we compared it with other codes (DISORT and MITRA) with less assumption in their algorisms. In this result, JACOSPAR forward model shows the difference in brightness within 1% in limb geometry, which can be acceptable compared to the observational accuracy. In the validation of the retrieval, we checked the result of the dust and water ice densities retrieved from the model spectra formed by the forward model with model atmospheric profiles. We confirmed that the aerosol retrieval provided the accuracy within -30% to +120% when a priori aerosol density was enough large (column τ> 0.9 at 1 μm wavelength). We also confirmed that the retrieval may be difficult in the lower layers beyond nadir column of τ> 0.1 (τ along the limb: about 1 or more). We do further works in order to retrieve them with dust and water ice sizes, surface albedo, and gas densities.
We also surveyed the limb observation data of Mars Express (MEx) OMEGA and did a simple analysis without JACOSPAR. Two cases may show daily variation in the absorption depth of water vapor and water ice along the limb. In the daytime (12-16 LT) case, water ice sublimation and water vapor increase may occur. In the morning (6-8 LT) case, a rise in water vapor and water ice may be seen. The Mars Climate Database (MCD) with these cases showed that in the former the water ice and water vapor altitude change is limited in lower alitude layer, while in the latter the water vapor and water ice rises are not reproduced. These differences may be caused by the limitation of MCD. We will utilize JACOSPAR for vertical and horizontal profiles of aerosols and gases at once, which will lead to the studies of the vertical transport mechanisms.
This study is planned to contribute to the limb observations by InfraRed Spectrometer (MIRS) aboard JAXA Martian Moons eXploration (MMX) mission in the mid 2020s. Our fast retrieval tool will contribute to the MMX atmospheric observation covering wide disk and limb spectral imaging data sets, for the retrieval of the horizontal and vertical profiles of dust and water vapor over a wide range at hourly time scale.