A very interesting example of sulfates detected appears in Juventae Chasma, part of the Valles Marineris canyon system. These were detected by the OMEGA instrument and were considered consistent with the thesis of there only being very old deposits of aqueous minerals based on the age of the overall region they were located.
However, the floor of Juventae Chasma may in fact be quite young. Notice the marked difference in crater counts in Juventae compared to the counts in the larger surrounding area :

Paper No. 178-0
JUVENTAE CHASMA/MAJA VALLES, MARS: IMPLICATIONS OF MOC CRATERING DENSITIES AND GEOMORPHOLOGY.
CHAPMAN, Mary G., Astrogeology Team, U. S. Geol Survey, 2255 N. Gemini Drive, Flagstaff, AZ 86001, ****@usgs.gov and RUSSELL, Andrew J., School of Earth Sciences & Geography, Keele Univ, Keele, Staffordshire, ST5 5BG, United Kingdom
"MOC data indicate a marked difference in cratering densities between the Juventae Chasma, the channel areas of Maja Valles, and the debouchment area in Chryse basin, possibly due to Amazonian-age melting of ground ice and (volcanism-induced?) floods. Cumulative densities for craters with diameters >50 m/106km2 indicate the chasma floor/chaotic areas (167K ± 48K) have a paucity of impact craters with respect to other areas. The transition area from chaos to channel (700K ± 74K) and channels directly north of the chasma (543K ± 243K) are also sparsely cratered. However, from about 300 km north of the chasma, the Maja channel area is more heavily cratered (4,524K ± 383K), as is the Chryse basin floor (6,224K ± 1,118K). The highest counts on the floor of the Chryse basin might be due to a combination of wide, thin, sediment dispersal and deposition upstream in Maja Valles. The heavily-cratered north channel area could be accounted for if (1) few young floods reached these extents, (2) the floods in these wide channel areas lacked the power to remove craters, (3) craters were protected from erosion by ice infills, or (4) target rocks were less friable. The young age at the channel head may be due to recent flood activity and/or eolian mantling (however channel features are observable through the thin dune cover). The very young age of the Juventae chasma and chaos area could be explained by a combination of friable target rocks, some material (subsequently removed) that protected the area from impacts, recent eolian activity, disruption by melting of late-stage ground ice, and/or young volcanic activity. If the interior mounds are sub-ice volcanoes, at least 2 periods of flooding may be supported by 2 stages of mound formation; i.e. the largest mound postdates chaos, but a very small interior mound, shown on M0804669, is partly buried by, therefore older than, chaos. This image may show evidence of late-stage ground ice (possible thermokarst pits on an impact crater and brittle fracture of channel-confined, dune-covered material)."
GSA Annual Meeting, November 5-8, 2001
General Information for this Meeting
Session No. 178
Planetary Geology
Hynes Convention Center: 304
1:30 PM-5:30 PM, Thursday, November 8, 2001
http://gsa.confex.com/gsa/2001AM/finalprogram/abstract_27650.htm

Accurate ages for such apparently young surfaces might be provided by the updated crater count methods of Hartmann, proposed for young surfaces such as permafrost:

MARTIAN CHRONOLOGY: TOWARD RESOLUTION OF THE 2005 “CONTROVERSY” AND EVIDENCE
FOR OBLIQUITY-DRIVEN RESURFACING PROCESSES.
William K. Hartmann, Planetary Science Institute,
1700 E. Ft. Lowell Rd., Ste 106, Tucson AZ 85719-2395 USA; ****@psi.edu
Seventh International Conference on Mars.
"Summary: Contrary to recent perceptions, the
discovery of ongoing production of decameter craters
means that researchers can date crater retention ages
(i.e., rock formation events, exhumation events, and
other generalized resurfacing events), even on small and
young formations. Table 1 summarizes some internally
consistent, direct observational data on the Martian
chronology. Further opportunities abound, and we must
continue trying to learn how to read the record of
accumulating small craters."
http://www.lpi.usra.edu/meetings/7thmars2007/pdf/3318.pdf

Detailed observations of Juventae Chasma by HiRise might provide geomorphological evidence of the youth of these deposits.

Bob Clark