Astron. Astrophys. 333, 1092-1099 (1998)

1. Introduction

Although studied for long, the problem of the stability of the CO₂ -dominated martian atmosphere remains a topical question. It was raised when it was noticed that the photolysis of CO₂ by solar light penetrating deeply in Mars tenuous atmosphere might have led to an accumulation of CO and O₂ giving an abundance for these two constituents of up to 10%. Indeed, this would have been possible as the spin-forbidden inverse reaction (CO + O CO₂) is very slow. However, observational evidences show that the amount of CO and O₂ remains very modest (approximately 0.08% and 0.12% respectively) and that therefore CO₂ is somehow efficiently recycled. McElroy and Donahue (1972) and Parkinson and Hunten (1972) showed that oxydation of CO to CO₂ can occur through a process involving the OH radicals. These two models differed in the source of OH radicals (HO₂ in the case the of McElroy and Donahue model, HO₂ and H₂ O₂ in the case of the Parkinson and Hunten one) and in the required value of the eddy diffusion coefficient K. In fact both models failed in matching the reality, the first one because it required too high values of K in the middle atmosphere and the second one because it needed too high an amount of water in the atmosphere. More recently, Krasnopolsky (1993a) and Atreya and Gu (1994) proposed models taking into account the temperature dependence of CO₂ cross-sections, for which pure gas-phase chemistry resulted in slightly too low predicted CO amounts. Along with Anbar et al. (1993a), they thus showed that the CO₂ stability problem could even be reversed ! It appeared so that there was a need for a sink of HO_x radicals, which could be provided by heterogeneous processes. However, Anbar et al. (1993b) suggested that revising the rates associated to the reactions involving HO_x, which are not precisely known, could be enough to balance the CO₂ production and loss rates, and Nair et al. (1994) demonstrated not only that heterogeneous chemistry would be inadequate to bring the CO abundance back into agreement with the observations, but also that there was no need to invoke it when reasonable modifications in a homogeneous gas-phase model can provide satisfactory agreement. Krasnopolsky (1995) reached agreement as well with the observed CO abundances assuming a reduced photolysis of water vapor due to either a temperature effect of the cross section near 1900 Å or the possible effect of impurities on the measurements of the H₂ O cross sections made by Thompson et al. (1963) near 1900 Å.

However, frequent and disk-resolved observations are necessary to assess the existence or non-existence of spatial and temporal variations of the carbon monoxide abundance. Clancy et al. (1983) have not seen clear evidence for a significant change of the CO abundance over a timescale of about 5 years, although their data allow a range of CO variability of 0 to 100%. Nair et al. (1994) suggest that, given the variability of the water vapour abundance and the critical role played by odd hydrogen in the abundances of CO and O₂, these could vary on time scales of the order of their photochemical lifetimes. This presumes that the variability of CO is comparable with the variability of the global-mean water vapour abundance averaged over a period of 5 years and the expected effect is very low. Krasnopolsky (1993b) has also shown that the solar cycle could have an effect on the CO mixing ratio, inducing a variation which could reach 35%. In all cases, the problem remains however of the apparent lack of variation of the O₂ abundance, which makes it difficult for the CO abundance to vary. Concerning the question of possible spatial variations of the CO abundance, Lellouch et al. (1991), from millimetric observations, have shown that spatial variations, if any, could not exceed 40% on spatial scale comparable to a martian radius. However, few spatially-resolved observations are available at present. In 1992, using data obtained with the instrument ISM (an infrared imaging spectrometer) onboard the Phobos spacecraft, Rosenqvist et al. (1992) found a possible depletion by a factor of about 3 above the high volcano region. But this result remained controversial, especially after Hunten (1993) showed that a small instrumental error might be at the origin of this result and also because a variable CO might be confused with a variable presence of an absorbing mineral on the surface.

In order to investigate the presence of possible temporal and spatial CO abundance variations, we have conducted a four year observation programme of CO in the atmosphere of Mars, starting in 1988. The 1988 and 1989 data have already led to CO abundance measurements from the (1-0) band at 4.7µm (Billebaud et al., 1992). The next step was to record one new spectrum in 1990 and then four spectra in 1991, corresponding to four different points on the planet, in order to test both the short-term stability of the CO abundance and the possibility for middle-scale spatial variations. These new data were recorded in the (2-0) vibrational band of CO at 2.35µm. This band was preferred to the (1-0) one because, as it is located in the solar reflected component of the planetary emission, the analysis of the data is much less sensitive to the atmospheric thermal profile.

The observations are described in Sect. 2. The model is presented in Sect. 3 and the discussion and conclusions are given in Sect. 4.

© European Southern Observatory (ESO) 1998

Online publication: April 28, 1998

helpdesk.link@springer.de