Thermal and visible studies of Mars using the Termoskan data set

Citation

Betts, Bruce Harold (1994) Thermal and visible studies of Mars using the Termoskan data set. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/M9T5-G181. https://resolver.caltech.edu/CaltechTHESIS:02112013-110255064

Abstract

In February and March, 1989, the Termoskan instrument on board the Phobos '88 spacecraft of the USSR acquired the highest spatial resolution thermal data ever obtained for Mars, ranging in resolution from 300 m to 3 km per pixel. It simultaneously obtained broad band visible channel data. The panoramas cover a large portion of the equatorial region from 30°S to 6°N. New and unique analyses facilitated by Termoskan are presented here. In addition, this thesis describes the instrument, data, and validation. Termoskan thermal data shows good temperature agreement with Viking IRTM. However, conversion of Termoskan visible data to bolometric albedo is problematic.

Utilizing the Termoskan data, I recognized a new feature on Mars: ejecta blanket distinct in the thermal infrared (EDITH). Virtually all of the more than 100 such features discovered in the Termoskan data are located on the plains near Valles Marineris. I compiled a data base of 110 EDITH and non-EDITH craters ranging in diameter from 4.2 km to 90.6 km. EDITHs have a startlingly clear dependence upon terrains of Hesperian age, and show almost no other correlations within the data base. The Hesperian terrain dependence cannot be explained by either atmospheric or impactor variations. Wind patterns or locally available aeolian material cannot provide a single overall explanation for the observed variations. I postulate that most of the observed EDITHs are due to excavation of thermally distinctive Noachian age material from beneath a relatively thin layer of younger, more consolidated Hesperian volcanic material. The plausibility of this theory is supported by much geological evidence for relatively thin near-surface Hesperian deposits overlying massive Noachian megabreccias on the EDITH-rich plains units. I suggest that absence of thermally distinct ejecta blankets on Noachian and Amazonian terrains is due to absences of distinctive near-surface layering. Thermally distinct ejecta blankets are excellent locations for future landers and remote sensing because of relatively dust free surface exposures of material excavated from depth.

Also included in the thermal images are observations of several major channel and valley systems including significant portions of Shalbatana, Ravi, Al-Qahira, and Ma'adim Valles, the channel south of Hydraotes Chaos, channel material in Eos Chasma, and small portions Simud, Tiu, and Ares Valles and channel material in Gangis Chasma. Simultaneous broad band visible data exists for all but Ma'adim Vallis. I find that most of the channels and valleys have higher inertias than their surroundings, consistent with previous thermal studies of martian channels. I show for the first time that thermal inertia boundaries closely match all flat channel floor boundaries. Using Viking albedos, Termoskan temperatures, and thermal modelling, I derive lower bounds on typical channel thermal inertias ranging from 8.4 to 12.5 (10^(-3) cal cm^(-2) s^(-1/2) K^(-1). Lower bounds on inertia differences with the surrounding heavily cratered plains range from 1.1 to 3.5. Atmospheric and geometric effects are not sufficient to cause the inertia enhancements. I agree with previous researchers that localized, dark, high inertia areas within channels are likely aeolian in nature. However, thermal homogeneity and strong correlation of thermal boundaries with the channel floor boundaries lead me to favor non-aeolian overall explanations. Small scale aeolian deposition or aeolian deflation may, however, play some role in the inertia enhancement Channel floor inertia enhancements are strongly associated with channels showing fretted morphologies such as wide, flat floors and steep scalloped walls. Therefore, I favor fretting processes over catastrophic flooding for explaining the inertia enhancements. Fretting may have emplaced more blocks on channel floors or caused increased bonding of fines due to increased availability of water. Alternatively, post-channel formation water that may have been preferentially present due to the low, flat fretted floors may have enhanced bonding of original fines or dust fallout The coupling of both EDITHs and channel inertias to morphology is unlike most sharp Martian inertia variations which are decoupled from observed surface morphology.

Termoskan observed morning limb brightening in the thermal channel, but not in the visible channel. The thermal morning limb brightening is likely due to a water ice or dust haze that is warmer than the surface at the time of the observations. A water ice haze with a scale height of 5 km could match the observations. Visible scattering is observed to be significant on morning and evening limbs out to 60 or 70 km. Localized high altitude stratospheric clouds are observed in the visible channel.

The Termoskan data show that the highland-lowland boundary in the Aeolis Quadrangle appears strongly correlated with a high-low thermal inertia boundary. The sharpness of that boundary varies from less than 4 km to more than 50 km. In all cases, inertias continue to decrease gradually for many tens of km into the lowlands. Several other large scale thermal boundaries are also observed in the data.

Termoskan observed fine thermal structure on the flanks of Arsia Mons and elsewhere, which represent examples of interesting and significant thermal variations seen at the limit of Termoskan's spatial resolution. Sharp variations and boundaries imply there cannot be global scale dust blanketing deeper than about one centimeter, if that.

Termoskan obtained the first ever thermal images of Phobos' shadow on the surface of Mars, along with simultaneous visible images. The best observed shadow occurrence was on the flanks of Arsia Mons. For this occurrence, I combined the observed decrease in visible illumination of the surface with the observed decrease in brightness temperature to calculate thermal inertias of the Martian surface. Most of the derived inertias fall within the range 0.9 to 1.4, corresponding to 5 to 10 micron dust particles for a homogeneous surface. Dust at the surface is consistent with previous theories of Tharsis as a current area of dust deposition. Shadow derived inertias are sensitive to mm depths, whereas diurnally derived inertias are sensitive to cm depths. The shadow derived inertias are very similar to Haberle and Jakosky [1991] atmospherically corrected Palluconi and Kieffer [1981] Viking IRTM diurnally derived inertias. Thus, if near surface layering exists at all in this region, it is not very significant.

Thesis Files