The effect of salt crust on the thermal conductivity of one sample
of fluvial particulate materials under Martian atmospheric pressures
Marsha A. Presley,1 Robert A. Craddock,2 and Natalya Zolotova3
Received 2 February 2009; revised 13 July 2009; accepted 24 July 2009; published 7 November 2009.
[1] A line-heat source apparatus was used to measure thermal conductivities of a lightly
cemented fluvial sediment (salinity = 1.1 g  kg1), and the same sample with the
cement bonds almost completely disrupted, under low pressure, carbon dioxide
atmospheres. The thermal conductivities of the cemented sample were approximately
3 higher, over the range of atmospheric pressures tested, than the thermal conductivities
of the same sample after the cement bonds were broken. A thermal conductivity-derived
particle size was determined for each sample by comparing these thermal conductivity
measurements to previous data that demonstrated the dependence of thermal conductivity
on particle size. Actual particle-size distributions were determined via physical separation
through brass sieves. When uncemented, 87% of the particles were less than 125 mm
in diameter, with 60% of the sample being less than 63 mm in diameter. As much as
35% of the cemented sample was composed of conglomerate particles with diameters
greater than 500 mm. The thermal conductivities of the cemented sample were most
similar to those of 500-mm glass beads, whereas the thermal conductivities of the
uncemented sample were most similar to those of 75-mm glass beads. This study
demonstrates that even a small amount of salt cement can significantly increase the
thermal conductivity of particulate materials, as predicted by thermal modeling estimates
by previous investigators.
Citation: Presley, M. A., R. A. Craddock, and N. Zolotova (2009), The effect of salt crust on the thermal conductivity of one sample
of fluvial particulate materials under Martian atmospheric pressures, J. Geophys. Res., 114, E11007, doi:10.1029/2009JE003355.
1. Introduction
[2] Salt crusts have been observed at the Viking Lander
sites [Binder et al., 1977; Mutch et al., 1976], the Pathfinder
landing site [Rieder et al., 1997; Moore et al., 1999], and by
both Mars Exploration Rovers (MER) [Squyres et al., 2004;
Gellert et al., 2004]. Crusted particulates were first
observed on the Martian surface near the Viking 1 space-
craft where overlying sediment had been swept clear by
retrorocket exhaust (Figure 1). Polygonal fractures within
this material indicate a degree of cohesiveness that is not
present in loose, unindurated particulates. Cohesive ??clods??
obtained by the Viking Landers? sampling arms would
break up completely when the samples were sieved [Clark
et al., 1976]. Such friability suggests that the material was
only lightly cemented.
[3] Building on the data from the surface, investigation of
spacecraft data from the Viking Infrared Thermal Mapper
(IRTM) and the Mars Global Surveyor Thermal Emission
Spectrometer (TES) revealed large regional areas of inter-
mediate albedo and thermal inertia that have been interpreted
to be indurated particulate materials [Kieffer et al., 1981;
Jakosky and Christensen, 1986; Presley and Arvidson, 1988;
Christensen and Moore, 1992; Mere?nyi et al., 1996; Mellon
et al., 2000; Putzig et al., 2005]. Each of these studies
assumed that a cementing agent within the pores of the
material would increase the thermal conductivity, and there-
fore the thermal inertia, of particulate materials, but the
magnitude of such an effect was unknown. While thermal
conductivity models have been developed that include the
effects of ice [e.g., DeVries, 1952; Johansen, 1975], includ-
ing under Martian atmospheric conditions [Mellon et al.,
1997], thermal conductivity modeling efforts specifically
concerning salts under Martian atmospheres have only
recently been published [Mellon et al., 2008; Piqueux and
Christensen, 2009]. Similarly, there have been only a few
laboratory studies on the influence of salt-cements on
thermal conductivity, with none performed under Martian
conditions.
[4] Under terrestrial atmospheric conditions, Abu-
Hamdeh and Reeder [2000] measured a decrease in thermal
conductivity with increasing salt concentration when the
salt was added to the soil as an aqueous solution. In
the current Martian environment, however, liquid aqueous
solutions will not be a factor at the surface. Van Rooyen and
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 114, E11007, doi:10.1029/2009JE003355, 2009
Click
Here
for
Full
Article
1School of Earth and Space Exploration, Mars Space Flight Facility,
Arizona State University, Tempe, Arizona, USA.
2Center for Earth and Planetary Studies, National Air and Space
Museum, Smithsonian Institution, Washington, DC, USA.
3School of Earth and Space Exploration, Arizona State University,
Tempe, Arizona, USA.
Copyright 2009 by the American Geophysical Union.
0148-0227/09/2009JE003355$09.00
E11007 1 of 8
Winterkorn [1959], in contrast, find that under terrestrial
conditions the thermal conductivity of salt-cemented sand in
a ??nearly dry?? state was indeed higher than that of salt-free
sand. Similarly, Seiferlin et al. [2007] measured an increase
in thermal conductivity, under a vacuum, using wax as a
cementing agent between small glass beads. Moreover, they
observed an increase in thermal conductivity with an
increase in the amount of ??cement.?? The effect of salt
cementation under Martian atmospheric pressures, however,
has yet to be investigated in the laboratory.
[5] Mellon et al. [2008] and Piqueux and Christensen
[2009] have recently modeled the thermal effects of salt-
cement in the pore spaces between particulates under
Martian atmospheric pressures. Both studies indicated that
minute quantities of salt cement could have profound effects
on the bulk thermal conductivity of particulate materials.
Mellon et al. [2008] report that 1% by volume of a
cementing agent could increase the thermal conductivity
by 5 to 50 times. Piqueux and Christensen [2009] report
that cement fractions of 0.001% to 1% by volume will
increase the thermal conductivity by a factor of 3 to 8 times.
From 1 to 15% cement by volume the increase in thermal
conductivity is 20?50%, with thermal conductivity values
rapidly increasing to that of rock for fractions greater than
15% by volume.
[6] An ideal study of the effect of salt-cementation on the
thermal conductivity of particulate materials would include
three components: a laboratory component in which the
amount and type of cement and atmospheric pressure are
systematically varied, a field component in which naturally
occurring cemented particulate deposits are investigated,
and a theoretical component to aid the interpretation of
the experimental and field measurements. The laboratory
component increases our understanding of how variations in
physical characteristics affect the thermal conductivity of
particulate materials, and what the magnitude of those
changes are going to be. A theoretical component helps
interpret these effects, and laboratory results in turn can lead
to more accurate thermal conductivity models. The field
component increases our understanding of the geological
and environmental contexts in which these thermophysical
variations occur. Together these components would improve
our ability to interpret variations in the surface temperature
data in terms of the geology of the planetary surface.
[7] Yet the difficulties associated with measuring the
effect of salt crusts/cement on thermal conductivity are
numerous. Sampling a crusted sediment for measurement
back in the lab will break the salt bonds, particularly in a
friable, lightly cemented sediment. Transportation, particu-
larly over long distances, is likely to disrupt the cement
bonds even further. The use of a thermal conductivity probe,
whether in a lab or in situ, would break the salt bonds upon
insertion into the indurated sediment and create a layer of
disturbed soil around the probe. All of these effects would
likely lower the thermal conductivity measured relative to
the actual thermal conductivity.
[8] Formation of uniform samples of lightly indurated
particulate materials, for in situ laboratory analysis, is also
not an easy task. The texture and nature of the cement
formed will depend on the formation conditions: the amount
of water initially present, the composition and concentration
of salts within the aqueous solution, and whether the
solution evaporates or freezes and then sublimates [e.g.,
Goodall et al., 2000]. Heterogeneity in the distribution and
texture of the salt is an expected result and may bias thermal
conductivity measurements. In order to avoid disruption of
the cement, the line-heat source would need to be in place
either prior to adding the aqueous solution to the sample or
at least prior to evaporation of the solution. That sequence,
however, increases the possibility that even a small amount
of salt may form a bridge between the heating wire (or
thermal conductivity probe) and the soil immediately adja-
cent to it. Such a bridge could bias thermal conductivity
measurements to higher values than may be appropriate.
[9] While these limitations have hindered the develop-
ment of a systematic study, the effect of induration on
thermal conductivity under Martian atmospheric pressures
has been measured for one naturally occurring fluvial
sample. Although it is only one sample, under nonideal
conditions, the results demonstrate that even a relatively
small amount of cement will significantly increase the
thermal conductivity.
2. Background
[10] Measurements taken by the Viking Landers? X-ray
Fluorescence spectrometers indicated that the Martian sur-
ficial fines contained an abundance of sulfur [Clark et al.,
1976]. Initial geochemical interpretation of those chemical
analyses suggested that the Martian surficial deposits
contained 10?15% kieserite (MgSO4  H2O), 5?10% cal-
cite (CaCO3), and 1% NaCl [Baird et al., 1976]. Burns
[1987, 1988] later argued that ferric sulfates of the copiapite
and jarosite groups were also possible constituents of the
Martian surface.
[11] From an orbital perspective, much of the Martian
surface can be classified as low albedo, high thermal inertia
regions, such as Syrtis Major and Sinus Meridiani, or as
high albedo, low thermal inertia regions, such as Arabia and
Tharsis. Western Arabia Terra (aka Oxia) is one example of
a large region that does not fit that bimodal classification.
While its thermal inertia (200 J  m2 s1/2 K1) is
intermediate between that of the bright areas (130 J  m2
Figure 1. An example of a crusted material on the Martian
surface. The image is a portion of Viking Lander image
12A136, where the retrorocket exhaust had swept the
surface clear of overlying sediment. The polygonal fractures
indicate a degree of cohesiveness that is not present in
loose, unindurated particulates.
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
2 of 8
E11007
s1/2 K1) and that of the dark areas (240 J  m2 s1/2
K1), its violet reflectance (0.05?0.06) is as low as the dark
areas (0.05?0.065) and its red reflectance (0.14?0.17) is
almost, but not quite, as bright as the bright areas (0.16?
0.21) [Presley and Arvidson, 1988].
[12] When red and blue reflectances, derived from the
VikingOrbiter camera data, were plotted on a two-dimensional
histogram, a mixing trend between the dark violet
(Sinus Meridiani) material and the bright red (Arabia) material
was readily apparent, as was a mixing trend between the
intermediate (??brown,?? Oxia) material and the bright red
material [Presley and Arvidson, 1988, Figure 9]. In contrast,
a clear mixing trend was not apparent between the dark violet
end point material and the ??brown?? end point material, which
suggested that the surface material in Oxia is relatively
immobile [Presley and Arvidson, 1988].
[13] The thermal inertia of the material comprising
Oxia averaged to about 200 J  m2 s1/2 K1 [Presley
and Arvidson, 1988], which would correspond to a particle
size in unindurated material of about 90 mm [Presley and
Christensen, 1997]. On Mars, the threshold wind speed is
at a minimum for a particle diameter of 115 mm [Iversen
and White, 1982]. As such, the material comprising Oxia
should be the most easily transported on Mars. Yet eolian
bedforms are not visible in Viking images of this region
[Presley and Arvidson, 1988]. Since the material appears
to be immobile, an alternative explanation is that the
surface material of Oxia is composed of a finer grain size
that is lightly indurated [Kieffer et al., 1981; Jakosky and
Christensen, 1986; Presley and Arvidson, 1988].
[14] The color-albedo units characterized in the Oxia
region are similar to those observed in the Chryse region
(area of the Viking 1 Landing site) [Mere?nyi et al., 1996;
Arvidson et al., 1989]. Mellon et al. [2000] and Putzig et al.
[2005] also included Oxia and Chryse in the same thermal
inertia-albedo category. The similarity of thermal and opti-
cal properties with other regions suggests the possibility that
crusted soils of similar properties may be widespread across
the planet [e.g., Jakosky and Christensen, 1986].
[15] If Oxia and similar regions are composed of a salt-
crusted material, logic suggests that the increased particle to
particle connectivity must be balanced by an effective
particle size that is much smaller than 90 mm in order to
be consistent with the thermal data. One possibility then, is
that these regions have a similar effective particle size as the
bright red dust, in which case these regions would be
indurated dust deposits.
[16] Most salts, including kieserite, however, exhibit no
significant absorptions in the visible part of the spectrum. A
??transparent?? salt is therefore unlikely to specifically reduce
the blue reflectance of the deposit. Iron sulfates such as
quenstedtite [Presley and Arvidson, 1988] or jarosite
[Burns, 1986, 1987, 1988; Burns and Fisher, 1990; Burns,
1993; Presley, 1995] are ??trans-opaque,?? with absorption
features in the blue end of the visible spectrum [Hunt et al.,
1971; Bishop and Murad, 2005], and could explain both the
darkening and the loss of blue reflectance. Abundant
jarosite has indeed been identified by the MER Opportunity
[e.g., Klingelho?fer et al., 2004; Morris et al., 2006], in
addition to simpler, transparent Mg and Ca sulfates [e.g.,
Christensen et al., 2004; Squyres et al., 2004; Wang et al.,
2006].
[17] Another surficial unit has since been identified that
also has intermediate thermal inertia, but high albedo
[Putzig et al., 2005]. This unit has been interpreted as being
a very thin deposit of dust that is thick enough to dominate
the albedo, but thermally thin enough that it cannot mask
the effects of a higher thermal inertia surface underneath
[Putzig et al., 2005]. Alternatively, this region could be an
indurated deposit, but one with imperceptible amounts, if
any, of trans-opaque salts.
[18] Although the theory that salt-cements are present
over large regions of the surface seems to fit the data well,
accurate interpretation is hindered by the scarcity of knowl-
edge about the effect of salts on thermal properties under
Martian conditions. The thermal conductivity measurements
presented in this study present a small step toward clarifi-
cation of the thermal effects of pore-filling cement.
3. Experimental Procedure
3.1. Sample
[19] The sample used for this study was a cemented, but
very friable deposit collected from the Ross River flood-
plain east of the Simpson Desert in Australia [Williams,
1971; Bourke and Pickup, 1999; Hollands et al., 2006]. The
sample was obtained with a small core, approximately 8 cm
in depth with a 2.5 cm radius. Prior to extraction, an area
immediately adjacent to the sample location was measured
in situ for bulk density and moisture content with a Model
3440 portable surface moisture density gauge manufactured
by Troxler Electronic Laboratories. The moisture content
affirmed that samples collected were dry (i.e., that it had not
rained recently). The density measurements were to insure
that the appropriate bulk density would be applied in the
laboratory, as thermal conductivity may vary with bulk
density [Fountain and West, 1970; M. A. Presley and
P. R. Christensen, Thermal conductivity measurements of
particulate materials: 4A. The effect of bulk density for
granular particles, submitted to Journal of Geophysical
Research, 2009].
[20] The sample was comprised of well-rounded particles,
approximately 95% quartz, with minor amounts of potassium
feldspar and just enough amorphous or microcrystalline
hematite and goethite coating the grains to produce the red
color characteristic of the Australian desert in that region
[see Hunter et al., 2006]. The salts forming the cement
within the sample were expected to be dominated by typical
evaporite salts including calcium carbonate, gypsum and
potassium and sodium chlorides [e.g., Lindsay, 1987]. This
expectation is consistent with the ion analysis discussed in
section 3.4 (Table 1).
[21] The collection and transportation of the sample nec-
essarily broke many of the cement bonds. When the sample
was transferred into the sample holder for thermal conduc-
tivity measurement it had already taken the form of a mostly
loose particulate material. Some of the sample was more
highly indurated than other parts and the cement retained
some identity through the formation of more durable clods.
However, due to the small size of the sample holder, which is
otherwise an advantage [Presley and Christensen, 1997],
and due to the fragility of the platinum heating wire, clods
larger than about 3 mm were removed ahead of time and not
allowed in the sample holder.
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
3 of 8
E11007
[22] Once the thermal conductivity of the sample in this
condition, referred to as the ??cemented sample,?? was
measured, the sample was removed from the sample holder
and the remaining cement bonds were broken by gently
rubbing the sample by hand. A latex glove was used to
avoid transferring oils, etc., onto the sample. The procedure
was performed by hand so that the sediment particles
themselves would not be broken and reduced in size. This
state is referred to as the ??uncemented sample.??
3.2. Particle Size Analysis
[23] Particle size analyses were performed by physically
separating the particles through brass sieves corresponding
to F size intervals (F = log2d, where d is the particle
diameter in millimeters). One particle size analysis was
performed prior to breaking the salt bonds, after the thermal
conductivity of the cemented sample had been analyzed.
The sieving was done by gently shaking the sieves by hand,
in order to minimize further destruction of the cement
bonds. After the bonds had been broken and the thermal
conductivity of the uncemented sample had been evaluated,
the sample was divided into two halves. Particle size
analysis was performed on one half, using a RoTap to
facilitate the sieving [see also Presley and Christensen,
1997; Presley and Craddock, 2006]. The other half was
set aside for ion analysis.
3.3. Thermal Conductivity Measurements
[24] The line-heat source method was used to measure the
thermal conductivity of the sample, due to its simplicity and
proven reliability [Cremers, 1971; Presley and Christensen,
1997]. The laboratory set up was previously described in
detail by Presley and Christensen [1997], with upgrades as
discussed by Presley and Craddock [2006].
[25] A detailed error analysis and an assessment of
accuracy of the thermal conductivities measured in this
facility was presented by Presley and Christensen [1997,
also submitted manuscript, 2009]. The typical precision
of the thermal conductivity measurements is ?10%. For
the lowest and highest values of the thermal conductiv-
ities measured in this study, the precision errors could
reach ?20%.
[26] Measurements attained in this lab match those of
Smoluchowski [1910], Wechsler and Glaser [1965], and
Hu?tter et al. [2008], which are likely to be the most accurate
thermal conductivity measurements previously obtained
under Martian atmospheric pressures [Presley and Christensen,
1997, also submitted manuscript, 2009]. Furthermore, par-
ticle size estimates based on this laboratory?s results and
MER Rover thermal measurements match the particle size
observed with the MER Rover Microscopic Imagers
[Fergason et al., 2006; Hynek and Singer, 2007]. Although
there are no thermal conductivity ??standards?? available for
comparison, these observations suggest that measurements
obtained in this laboratory are reasonably accurate.
[27] The complete thermal conductivity data set is stored
on Compact Disk. One copy is in the possession of the
primary author. The other copy is stored at the Mars Space
Flight Facility library at Arizona State University, Tempe,
AZ, together with hard copies of abbreviated data files and a
summary of the experimental conditions. Data are available
upon request from the primary author.
3.4. Salinity
[28] In order to assess the amount of salt in the sample,
the major cations and anions were analyzed by ion chro-
matography in E. Shock?s GEOPIG laboratory at Arizona
State University.
[29] After oven-drying the sample for 2 days, nanopure
deionized water was added to the sample in a 5:1 ratio [e.g.,
U.S. Salinity Laboratory Staff, 1954], and the mixture was
shaken for 2 hours. Since gypsum was a likely component
of the salt [e.g., Lindsay, 1987] and takes longer to dissolve
than most salts, the mixture was allowed to settle overnight
in order to allow sufficient time for all of the salts to go into
solution [e.g., Schofield et al., 2001]. Salt-forming ions
were then extracted from the sample by filtering the
supernatant liquid through a 0.2 mm filter. The pH, carbon-
ate and bicarbonate concentrations were determined via
titration immediately after filtration [Buurman et al., 1996].
The remaining solutionwas divided into two parts. The soluble
anions were analyzed by anion-exchange chromatography
(Dionex DX-600) using suppressed conductivity detection
and a Dionex AS11 column. High-purity hydroxide eluent
was generated by an on-line electrolytic eluent generator.
Soluble cations were analyzed by suppressed conductivity
detection using a Dionex Cation-Exchange CS12A column on
aDionex DX-120 chromatograph. Duplicates and blanks were
run in each case to insure accuracy. Estimated analytical
uncertainties for all ions are ?1% for this procedure.
4. Results
[30] The results of the ion analysis are presented in
Table 1. The concentrations presented as mg  L1 are the
results from the ion chromatography analysis, and are with
respect to the volume of deionized water used in the
analysis. The concentrations presented as mmol  g1, are
those results recalculated to be molar concentrations per
Table 1. Concentration of Salt-Forming Ions Contained Within
the Samplea
Ions
Concentration
(mg/L)b (mmol/g)c (g/kg)c
Na+ 4.0 0.0009 0.020
NH4
+ 2.0 0.0005 0.010
K+ 22.4 0.0029 0.112
Mg2+ 8.0 0.0016 0.040
Ca2+ 20.3 0.0025 0.102
Subtotal 0.0126 0.283
Cl 4.6 0.0007 0.023
F 4.1 0.0011 0.020
NO3
 3.7 0.0003 0.018
SO4
2 8.4 0.0004 0.042
PO4
 1.0 0.0001 0.005
CO3
2 0.1 0.0000 0.000
HCO3
 143.9 0.0118 0.220
Subtotal 0.0148 0.719
Charge difference
(Cation  anion molarities) 0.0021
Total Salinity
(Sum of all ion concentrations)
1.112
aThe values are averaged results from two samples.
bConcentrations are with respect to the volume of the deionized water
used in the ion chromatography analysis.
cConcentrations are with respect to the weight of the soil sample.
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
4 of 8
E11007
gram of soil. The difference between the total of the anion
molar concentrations and the total of the cation molar
concentrations is the charge imbalance. The resultant charge
imbalance of 0.0021 is a relatively small difference and is
likely an effect of averaging the results of two separate
measurements, but also could be due to the presence of
other non-salt-forming ions, minor amounts of salt-forming
ions not analyzed for, and/or an intrinsic charge imbalance
from the surface of the soil particles. The concentrations,
presented as g  kg1, are the ion chromatography results
recalculated to be per kg of soil. The sum of the all of
the concentrations of salt-forming ions in the sample is
considered the salinity of the soil. A salinity of 1.1 g  kg1
is consistent with the friability of the sample, and its lightly
crusted appearance in situ.
[31] The particle size distributions are illustrated in
Figure 2. Figure 2a represents the cemented sample, before
the cement bonds were deliberately broken. The analysis is
indicative of the sample as it was prepared for thermal
conductivity measurement and not as it was in situ. As
previously discussed, some of the cement bonds had been
broken during collection and transport of the sample.
Conglomerate particles larger than 2?3 mm were excluded
from the sample due to the limitations of the thermal
conductivity measurement. Figure 2b is the particle size
distribution for the uncemented sample. Note that for both
samples, the smallest sieve size used was 63 mm (4F). No
attempt was made to further subdivide the <4F fraction.
Hence the statistical parameters presented in Figure 2 are
somewhat skewed by the likely incorrect assumption that
all of the particles in the <4F fraction can be assigned as 5F
(31?63 mm). Nonetheless, the cemented sample can be
roughly described as a medium- to fine-grained sand,
whereas, uncemented, the sample is more accurately
described as a silt.
[32] The bulk density of the sediment measured in situ
was 1600 kg  m3. This packing density was the same as
that measured for both cemented and uncemented forms
when the sample was poured into the sample holder for
thermal conductivity measurement. Hence no further
adjustment was required to match the bulk density to that
in the field.
[33] Figure 3 illustrates the thermal conductivities of
(a) the cemented sample, and (b) those of the uncemented
sample, measured over several atmospheric pressures from
0.5?20 torr (see also Table 2). For comparison purposes the
thermal conductivity curves for 250?275 mm, 500?520 mm
and 710?1000 mm glass beads [Presley, 1995] are included
in Figure 3a, and the thermal conductivity curves for 25?
30 mm, 70?75 mm and 90?100 mm glass beads [Presley,
1995] are included in Figure 3b.
5. Analysis
[34] As seen in Figure 2a, the thermal conductivities of
the ??cemented?? fluvial sample are equivalent to those of an
unindurated particle size of 500 mm. The thermal conduc-
tivities of the ??uncemented?? sample, after the salt bonds
had been broken, match those of glass beads with a particle
size of 70?75 mm (Figure 2b). These results are consistent
with previous studies that indicate that the larger particle
size fractions tend to control the thermal conductivity
[Presley and Craddock, 2006; see also Midtt?mme et al.,
1997]. For these two samples, however, as much as 9% of
the ??cemented?? sample is larger than 1000 mm, and at least
13% of the ??uncemented?? sample is larger than 75 mm.
Particles larger than 100 mm in both of these samples are
likely to be conglomerate particles lightly held together with
minimal cement, rather than individual solid grains.
6. Conclusions
[35] The thermal conductivities of the ??cemented??
sample are approximately 3 greater than the thermal
conductivities of the ??uncemented?? sample. This experi-
ment demonstrates that even in a lightly cemented sample
where the cement bonds have been somewhat disrupted by
the sampling process, salt cementation significantly
increases the thermal conductivity.
[36] The result of this experiment should be considered a
minimum effect that this amount of salt would have on the
thermal conductivity, as unbroken cement bonds would
undoubtedly contribute to an in situ thermal conductivity
that would be higher than what was measured after
sampling disrupted some of the cement bonding. Assuming
a specific gravity of 2.5 for the salt(s), a salinity value of
Figure 2. Particle sizes of the samples investigated in bins
of F, where F = log d, and plotted as histograms. The
sample was collected from the Ross River floodplain, east
of the Simpson Desert, Australia. (a) The particle size
distribution of the sediment as sampled, with cement bonds
broken only by the sampling process and transport. (b) The
particle size distribution of the sample once the salt bonds
have been broken as thoroughly as possible [results
originally published by Presley and Craddock, 2006].
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
5 of 8
E11007
1.1 g  kg1 should be roughly equivalent to 0.1% salt by
volume. This laboratory result is consistent with the thermal
conductivity increase predicted by Piqueux and Christensen
[2009] for low salt concentrations. This correspondence is
quite reasonable considering the relatively low salt concen-
tration and the disrupted cement bonds of the sample.
[37] The limitations of this experiment illuminate some of
the problems associated with measuring thermal conductiv-
ities of salt-cemented particulates, as discussed in section 1.
Remote measurement of temperature response is one way to
avoid these problems both in situ and in the lab. A new
thermal conductivity laboratory that will utilize a Forward-
Looking Infrared (FLIR) camera and an environmental
chamber capable of Martian atmospheric pressures and
temperatures has been constructed at the Mars Space Flight
Facility at Arizona State University. At this point, there is
no ideal laboratory or field technique to measure thermal
conductivity, particularly of salt encrusted particulates.
However, since the cement bonds will not be broken in
the process of sampling or measurement, FLIR-type instru-
ments hold a promising potential for future work on salt
encrusted particulates. As other thermal conductivity labo-
ratories capable of analysis under Martian atmospheric
conditions are designed [e.g., Hu?tter et al., 2008], efforts
should be made to establish a ??standard?? sample, so that
results between laboratories may be adequately compared,
and to encourage better quality results. Thermal modeling
should increasingly be used in tandem with laboratory and
field measurements, so that we may understand better the
properties that control thermal conductivity, as well as to
establish a more detailed understanding of the limitations of
the techniques that we utilize for measurement and of those
models.
[38] Acknowledgments. We appreciate the very thoughtful reviews
from Michael Mellon and an anonymous reviewer, as well as additional
comments from Robert Carlson and Nathan Putzig. These suggestions
allowed us to significantly improve the manuscript. We are grateful to Evert
Fruitman and Bill Coleman for upgrades and repairs to the signal condi-
tioner; to Rossman Irwin, Ruslan Kusmin, and Ted Maxwell for their help
Figure 3. Plots of thermal conductivity versus atmospheric pressure for both the (a) cemented sample
(A1A) and (b) the uncemented sample (A1B). Included in each plot are thermal conductivity
data for glass beads [Presley and Christensen, 1997] of appropriate particle sizes for comparison.
Thermal conductivities are presented here in SI units (W m1 K1) on the primary Y axis and in cgs units
(cal cm1 s1 K1) on the secondary Y axis. The thermal conductivity values for the cemented and
uncemented samples are also presented in Table 2, along with the initial temperature and power applied
across the heating wire for each measurement. Thermal conductivity data and experimental conditions for
the glass beads used in comparison were published in Appendix C of Presley [1995].
Table 2. Thermal Conductivity of the ??Cemented?? and
??Uncemented?? Samples
Atmospheric
Pressure (torr)
Initial
Temperature (C)
Power
(W/m)
Thermal
Conductivity (W/m K)
A1A ??cemented?? sample
0.5 22.1 0.5823 0.0260
1.0 22.8 0.5499 0.0414
2.0 22.2 0.5419 0.0519
3.0 22.8 0.5286 0.0619
4.0 22.1 0.5572 0.0661
5.0 21.6 0.5582 0.0778
6.0 22.3 0.5263 0.0795
7.0 22.5 0.5216 0.0908
8.0 22.6 0.5260 0.0958
9.0 23.5 0.5196 0.112
10.0 21.3 0.5501 0.113
12.0 23.3 0.5215 0.118
15.0 23.6 0.5188 0.115
A1B ??uncemented?? sample
0.6 21.0 0.6161 0.00702
1.0 22.0 0.5898 0.0101
2.0 21.1 0.5610 0.0162
3.0 21.1 0.4905 0.0216
4.0 21.8 0.5459 0.0258
5.0 21.8 0.4850 0.0291
6.0 21.3 0.4801 0.0314
7.0 21.6 0.5380 0.0342
8.0 21.3 0.5362 0.0394
9.0 21.3 0.5365 0.0382
10.0 21.6 0.5938 0.0389
12.0 21.0 0.5955 0.0461
15.0 22.4 0.5305 0.0499
20.0 21.0 0.5894 0.0557
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
6 of 8
E11007
collecting the Australian samples in the field; to Bryan MacFarlane for
particle size analysis; to Everett Schock for use of the GEOPIG Ion
Analysis Laboratory; to Tracy Lund for running the sample, duplicates,
and blanks through the ion chromatograph columns; and to Phil Christensen
for support of the Thermal Conductivity laboratory at Arizona State
University. The Smithsonian Institution provided the funds for the purchase
and transport of the Troxler 3450 moisture density gauge and for the
operator license. This study was primarily supported by NASA grant
NAG5-12180, with additional funding from grant NAG5-10214.
References
Abu-Hamdeh, N. H., and R. C. Reeder (2000), Effects of density, moisture,
salt concentration, and organic matter, Soil Sci. Soc. Am. J., 64, 1285?
1290.
Arvidson, R. E., E. A. Guinness, M. A. Dale-Bannister, J. Adams,
M. Smith, P. R. Christensen, and R. B. Singer (1989), Nature and distribu-
tion of surficial deposits in Chryse Planitia and vicinity, Mars, J. Geophys.
Res., 94(B2), 1573?1587.
Baird, A. K., P. Toulmin III, B. C. Clark, H. J. Rose Jr., K. Keil, R. P.
Christian, and J. L. Gooding (1976), Mineralogic and petrologic implica-
tions of Viking geochemical results from Mars: Interim report, Science,
194(4271), 1288?1293.
Binder, A. B., R. E. Arvidson, E. A. Guinness, K. L. Jones, E. C. Morris,
T. A. Mutch, D. C. Pieri, and C. Sagan (1977), The geology of the Viking
Lander 1 site, J. Geophys. Res., 82(28), 4439?4451.
Bishop, J. L., and E. Murad (2005), The visible and infrared spectral
properties of jarosite and alunite, Am. Mineral., 90(7), 1100?1107.
Bourke, M. C., and G. Pickup (1999), Fluvial form variability in arid
central Australia, in Varieties of Fluvial Form, edited by A. J. Miller
and A. Gupta, pp. 249?271, Wiley, Hoboken, N. J.
Burns, R. G. (1986), Terrestrial analogues of the surface rocks of Mars?,
Nature, 320, 55?56.
Burns, R. G. (1987), Ferric sulfates on Mars, J. Geophys. Res., 92(12),
E570?E574.
Burns, R. G. (1988), Gossans on Mars, in Proc. Lunar Planet. Sci. 18th,
pp. 713?721, Cambridge Univ. Press/Lunar and Planetary Inst., Houston,
Tex.
Burns, R. G. (1993), Rates and mechanisms of chemical weathering of
ferromagnesium silicate minerals on Mars, Geochim. Cosmochim. Acta,
57(19), 4555?4574.
Burns, R. G., and D. S. Fisher (1990), Iron-sulfur mineralogy of
Mars: Magmatic evolution and chemical weathering products, J. Geophys.
Res., 95(B9), 14,415?14,421.
Buurman, P., B. van Lagen, and E. J. Velthorst (1996), Manual for Soil and
Water Analysis, 314 pp., Backhuys Publishers, Leiden, Netherlands.
Christensen, P. R., and H. J. Moore (1992), The Martian surface layer, in
Mars, edited by H. H. Kieffer et al., pp. 686?729, Univ. of Arizona
Press, Tucson, Ariz.
Christensen, P. R., et al. (2004), Mineralogy at Meridiani Planum from the
mini-TES experiment on the Opportunity rover, Science, 306(5702),
1733?1739.
Clark, B. C., A. K. Baird, H. J. Rose Jr., P. Toulmin III, K. Keil, A. J.
Castro, W. C. Kelliher, C. D. Rowe, and P. H. Evans (1976), Inorganic
analyses of Martian surface samples at the Viking Landing sites, Science,
194(4271), 1283?1288.
Cremers, C. J. (1971), A thermal conductivity cell for small powdered
samples, Rev. Sci. Instrum., 42(11), 1694?1696.
DeVries, D. A. (1952), The thermal conductivity of soil, Meded.
Landbouwhogesch. Wageningen, 52(1), 1?73. (Translated by Building
Research Station (Library Communication No. 759), U. K.)
Fergason, R. L., P. R. Christensen, J. F. Bell III, M. P. Golombek, K. E.
Herkenhoff, and H. H. Kieffer (2006), Physical properties of the Mars
Exploration Rover landing sites as inferred from Mini-TES-derived thermal
inertia, J. Geophys. Res., 111, E02S21, doi:10.1029/2005JE002583.
Fountain, J. A., and E. A. West (1970), Thermal conductivity of particulate
basalt as a function of density in simulated lunar and Martian environ-
ments, J. Geophys. Res., 75(20), 4063?4069.
Gellert, R., et al. (2004), Chemistry of rocks and soils in Gusev Crater from
the Alpha Particle X-ray Spectrometer, Science, 305(5685), 829?832.
Goodall, T. M., C. P. North, and K. W. Glennie (2000), Surface and
subsurface sedimentary structures produced by salt crusts, Sedimentology,
47, 99?118.
Hollands, C. B., G. C. Nanson, B. G. Jones, C. S. Bristow, D. M. Price, and
T. J. Pietcsch (2006), Aeolian-fluvial interaction: Evidence for Late
Quaternary channel change and wind-rift linear dune formation in
the northwestern Simpson Desert, Australia, Quat. Sci. Rev., 25(1?2),
142?162.
Hunt, G. R., J. W. Salisbury, and C. J. Lenhoff (1971), Visible and near-
infrared spectra of minerals and rocks: IV. Sulphides and sulphates,
Modern Geol., 3, 1?14.
Hunter, K. A., D. S. Mackie, P. W. Boyd, and G. H. McTainsh (2006), From
desert to dessert: Why Australian dust matters, EOS Trans., AGU, Fall
Meet. Suppl., Abstract A42C-03.
Hu?tter, E. S., N. I. Koemle, G. Kargl, and E. Kaufmann (2008), Determina-
tion of the effective thermal conductivity of granular materials under
varying pressure conditions, J. Geophys. Res., 113, E12004,
doi:10.1029/2008JE003085.
Hynek, B. M., and K. Singer (2007), Ground truth from the Opportunity
Rover for Mars thermal inertia data, Geophys. Res. Lett., 34, L11201,
doi:10.1029/2007GL029687.
Iversen, J. D., and B. R. White (1982), Saltation threshold on Earth, Mars
and Venus, Sedimentology, 29, 111?119.
Jakosky, B. M., and P. R. Christensen (1986), Global duricrust on Mars:
Analysis of remote-sensing data, J. Geophys. Res., 91(B3), 3547?3559.
Johansen, O. (1975), Thermal conductivity of soils, Ph.D. Thesis, Trond-
heim, Norway, (CRREL Draft Translation 637, 1977), ADA 044002.
Kieffer, H. H., P. A. Davis, and L. A. Soderblom (1981), Mars? global
properties: Maps and applications, Proc. Lunar Planet. Conf., 12B,
1395?1417.
Klingelho?fer, G., et al. (2004), Jarosite and hematite at Meridiani Planum
from Opportunity?s Mo?ssbauer Spectrometer, Science, 306(5702), 1740?
1745.
Lindsay, J. F. (1987), Upper proterozoic evaporates in the Amadeus basin,
central Australia, and their role in basin tectonics, GSA Bull., 99(6), 852?
865.
Mellon, M. T., B. M. Jakosky, and S. E. Postawko (1997), The persistence
of equatorial ground ice on Mars, J. Geophys. Res., 102(E8), 19,357?
19,369.
Mellon, M. T., B. M. Jakosky, H. H. Kieffer, and P. R. Christensen (2000),
High-resolution thermal inertia mapping from the Mars Global Surveyor
Thermal Emission Spectrometer, Icarus, 148, 437?455.
Mellon, M. T., R. L. Fergason, and N. E. Putzig (2008), The thermal inertia
of the surface of Mars, in The Martian Surface: Composition, Mineralogy,
and Physical Properties, edited by J. W. Bell, pp. 399?427, Cambridge
Univ. Press, Cambridge, U. K.
Mere?nyi, E., K. S. Edgett, and R. B. Singer (1996), Deucalionis Regio,
Mars: Evidence for a new type of immobile weathered soil unit, Icarus,
124, 296?307.
Midtt?mme, K., J. Saettem, and E. Roaldset (1997), Thermal conductivity
of unconsolidated sediments from the V?ring Basin, Norwegian Sea,
Nord. Petrol. Technol. Ser. II, 145?197.
Moore, H. J., D. B. Bickler, J. A. Crisp, H. J. Eisen, J. A. Gensler,
A. F. C. Haldemann, J. R. Matijevic, L. K. Reid, and F. Pavlics (1999),
Soil-like deposits observed by Sojourner, the Pathfinder rover, J. Geophys.
Res., 104(E4), 8729?8746.
Morris, R. V., et al. (2006), Mo?ssbauer mineralogy of rock, soil, and dust
at Meridiani Planum, Mars: Opportunity?s journey across sulfate-rich
outcrop, basaltic sand and dust, and hematite lag deposits, J. Geophys.
Res., 111, E12S15, doi:10.1029/2006JE002791.
Mutch, T. A., et al. (1976), The surface of Mars: The view from the Viking
2 lander, Science, 194(4271), 1277?1283.
Piqueux, S. and P. R. Christensen (2009), A model of thermal conductivity
for planetary soils: 2. Theory for cemented soils, J. Geophys. Res., 114,
E09006, doi:10.1029/2008JE003309.
Presley, M. A. (1995), Thermal conductivity measurements of particulate
materials: Implications for surficial units on Mars, Ph.D. Dissertation,
Ariz. State Univ., Tempe.
Presley, M. A., and R. E. Arvidson (1988), Nature and origin of materials
exposed in the Oxia Palus-Western Arabia-Sinus Meridiani region, Mars,
Icarus, 75, 499?517.
Presley, M. A., and P. R. Christensen (1997), Thermal conductivity
measurements of particulate materials: 2. Results, J. Geophys. Res.,
102(E3), 6551?6566.
Presley, M. A., and R. A. Craddock (2006), Thermal conductivity measure-
ments of particulate materials: 3. Natural samples and mixtures of particle
sizes, J. Geophys. Res., 111, E09013, doi:10.1029/2006JE002706.
Putzig, N. E., M. T. Mellon, K. A. Kretke, and R. E. Arvidson (2005),
Global thermal inertia and surface properties of Mars from the MGS
mapping mission, Icarus, 173, 325?341.
Rieder, R., T. Economou, H. Wa?nke, A. Turkevich, J. Crisp, J. Bru?ckner,
G. Dreibus, and H. Y. McSween Jr. (1997), The chemical composition
of Martian soil and rocks returned by the mobile Alpha Proton X-ray
Spectrometer: Preliminary results from the X-ray mode, Science,
278(5344), 1771?1774.
Schofield, R., D. S. G. Thomas, and M. J. Kirkby (2001), Causal processes
of soil salinization in Tunisia, Spain and Hungary, Land Degrad. Develop.,
12, 163?181.
Seiferlin, K., M. Heimberg, and N. Thomas (2007), The effect of soil
cementation on the thermal conductivity, Geophys. Res. Abstr., 9,
EGU2007-A-02361.
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
7 of 8
E11007
Smoluchowski, M. M. (1910), Sur la conductibilite calorifique des corps
pulverizes, Bull. Int. Acad. Sci. Cacovie, 5A, 129?153.
Squyres, S. W., et al. (2004), In-situ evidence for an ancient aqueous
environment at Meridiani Planum, Mars, Science, 306(5702), 1709?
1714.
U.S. Salinity Laboratory Staff (1954), chap. 6, Methods for soil character-
ization in Diagnosis and Improvement of Saline and Alkali Soils,
edited by L. A. Richards, Agriculture Handbook 60, 83-126, USDA,
Washington, D. C.
Van Rooyen, M., and H. G. Winterkorn (1959), Structural and textural
influences on thermal conductivity of soils, Highw. Res. Board Proc.,
39, 576?621.
Wang, A., et al. (2006), Sulfate deposition in subsurface regolith in
Gusev Crater, Mars, J. Geophys. Res., 111, E02S17, doi:10.1029/
2005JE002513.
Wechsler, A. E., and P. E. Glaser (1965), Pressure effects on postulated
lunar materials, Icarus, 4, 335?352.
Williams, G. E. (1971), Flood deposits of the sand-bed ephemeral streams
of central Austrialia, Sedimentology, 17(1/2), 1?40.

R. A. Craddock, Center for Earth and Planetary Studies, National Air and
Space Museum, Smithsonian Institution, 6th Street and Independence
Avenue, S. W., Washington, DC 20560-0315, USA.
M. A. Presley, School of Earth and Space Exploration, Mars Space Flight
Facility, Arizona State University, Box 876305, Tempe, AZ 85287-6305,
USA. (swmartian@fastmail.fm)
N. Zolotova, School of Earth and Space Exploration, Arizona State
University, Box 871404, Tempe, AZ 85287-1404, USA.
E11007 PRESLEY ET AL.: SALT CRUSTS AND THERMAL CONDUCTIVITY
8 of 8
E11007