Spatial variability, composition and thickness of the seasonal north polar cap of Mars in mid-spring

Abstract

Abstract The planetary fourier spectrometer (PFS) for the Mars express mission (MEX) is an infrared spectrometer operating in the wavelength range from 1.2 to 45 μm by means of two spectral channels, called SWC (short wavelength channel) and LWC (long wavelength channel), covering, respectively, 1.2–5.5 and 5.5–45 μm. The middle-spring Martian north polar cap (Ls∼40°) has been observed by PFS/MEX in illuminated conditions during orbit 452. The SWC spectra are here used to study the cap composition in terms of CO2 ice, H2O ice and dust content. Significant spectral variation is noted in the cap interior, and regions of varying CO2 ice grain sizes, water frost abundance, CO2 ice cover and dust contamination can be distinguished. In addition, we correlate the infrared spectra with an image acquired during the same orbit by the OMEGA imaging spectrometer and with the altimetry from MOLA data. Many of the spectra variations correlate with heterogeneities noted in the image, although significant spectral variations are not discernible in the visible. The data have been divided into five regions with different latitude ranges and strong similarities in the spectra, and then averaged. Bi-directional reflectance models have been run with the appropriate lighting geometry and used to fit the observed data, allowing for CO2 ice and H2O ice grain sizes, dust and H2O ice contaminations in the form of intimate granular mixtures and spatial mixtures. A wide annulus of dusty water ice surrounds the recessing CO2 seasonal cap. The inner cap exhibits a layered structure with a thin CO2 layer with varying concentrations of dark dust, on top of an H2O ice underneath ground. In the best-fits, the ices beneath the top layer have been considered as spatial mixtures. The results are still very good everywhere in the spectral range, except where the CO2 ice absorption coefficients are such that even a thin layer is enough to totally absorb the incoming radiation (i.e. the band is saturated). This only happens around 3800 cm−1, inside the strong 2.7-μm CO2 ice absorption band. The effect of finite snow depth has been investigated through a layered albedo model. The thickness of the CO2 ice deposits increases with latitude, ranging from 0.5–1 g cm−2 within region II to 60–80 g cm−2 within the highest-latitude (up to 84°N) region V. Region I is at the cap edge and extends from 65°N to 72°N latitude. No CO2 ice is present in this region, which consists of relatively large grains of water ice (20 μm), highly contaminated by dust (0.15 wt%). The adjacent region II is a narrow region [76–79°N] right at the edge of the north residual polar cap. This region is very distinct in the OMEGA image, where it appears to surround the whole residual cap. The CO2 ice features are barely visible in these spectra, except for the strong saturated 2.7 μm band. It basically consists of a thin layer of 5-mm CO2 ice on top of an H2O ice layer with the same composition as region I. A third interesting region III is found all along the shoulder of the residual cap [79–81°N]. It extends over 1.5 km in altitude and over only 2° of latitude and consists of CO2 ice with a large dust content. It is an admixture of CO2 ice (3–4 mm), with several tens of ppm by mass of water ice and more than 2 ppt by mass of dust. The surface temperatures have been retrieved from the LWC spectra for each observation. We found an increase in the surface temperature in this region, indicating a spatial mixture of cold CO2 ice and warmer dust/H2O ice. Region IV is close to the top of the residual cap [81–84°N]; it is much brighter than region III, with a dust content 10 times lower than the latter. The CO2 grain size is 3 mm and strong CO2 ice features are present in the data, indicating a thicker CO2 ice layer than in region II (1–2 g cm−2). The final region V is right at the top of the residual cap (⩾84°N). It is “pure” CO2 ice (no dust) of 5 mm grain sizes, with 30 ppm by weight of water ice. The CO2 ice features are very pronounced and the 2.7 μm band is saturated. The optical thickness is close to the semi-infinite limit (30–40 g cm−2). Assuming a snowpack density of 0.5 g cm−3, we get a minimum thickness of 1–2 cm for the top-layer of regions II and III, 4–10 cm for region IV, and ⩾60–80 cm thickness for region V. These values are in close agreement with several recent results for the south seasonal polar cap. These results should provide new, useful constraints in models of the Martian climate system and volatile cycles.