Possible Segregated Ice at the Phoenix Landing Site: Was Liquid Water Involved?

Lander cameras on the Phoenix mission revealed polygonal terrain at the landing site. Areas identified by topography within the work area of the arm included a polygon and a surrounding trough. Two trenches were dug, the first (Goldilocks) on the shoulder of a trough area exposed a bright, hard material and the second (Snow white) in the center of the polygon exposed hard material, but with multispectral properties indistinguishable from soil. Visibile-NIR spectra of the Goldilocks bright material are consistent with slightly dusty ice. When first exposed, a 2 cm chunk of material broke off and was observed to completely disappear in 3 sols, an implied sublmation rate of 100 micrometers per hour. We hypothesize that the Goldilocks bright material is segregated ice. The material is hard, localized, has distinct edges, and was initially covered with only 3 cm of soil, thus was 2cm shallower than the hard layer in the Snow white trench in spite of a more south-facing exposure. A trench dug 40 cm further south of Goldilocks, with similar orientation, reached 18 cm depth without encountering hard material. Plausible mechanisms for emplacement of segregated ice include liquid water pooled into a thermally-produced crack analogous to terrestrial ice wedge polygon formation, snowparticles depositing preferentially in the troughs, and vapor deposition preferentially into cracks (D. Fisher, Icarus 179, 387, 2005). Mission observations were performed relevant to evaluating these formation mechanisms. Wet chemistry analyses of soils suggest they contain Mg(ClO4)2, a soluble hygroscopic salt with a eutectic freezing point of /- 68C. If liquid water moved though the soil and formed the bright deposit in Goldilocks trench, a higher concentration of perchlorate would be expected in the area of the ice. Mg(ClO4)2. 6 H2O would crystallize when the salty water froze, forming white rhombohedral crystals. After scraping away the surface soil, approximately 500 cm2of bright material was exposed in Goldilocks trench and left undisturbed for 79 sols. During this time, the brightness of the material slowly faded and, by sol 99, a sublimation lag covered the bright deposit with nearly the same spectral properties as soil. It was not possible to obtain a large enough sample of the lag to directly measure salt concentration with a wet chemistry cell. Instead, a small sample of the lag was examined with the Optical Microscope to look for morphological evidence of salts. The material was stickier and more cohesive than previous soil samples examined with the microscope, and a population of light colored particles up to 30 microns in diameter with evidence of angularity consistent with microcrystallinity was found. This observation is suggestive of possible salts more concentrated in this area. In conclusion, the microscopy results are consistent with a liquid water formation mechanism but inconclusive without a direct measurement of the composition of the material.