Hydrated Minerals in Northern Meridiani Planum, Mars. R.B. ...

6 downloads 34 Views 1MB Size Report
The Meridiani Planum region has some of the most enigmatic and diverse geology ... tures in Meridiani reveal that the strongest sulfate signatures appear to be ...
Hydrated Minerals in Northern Meridiani Planum, Mars.

1 2

R.B. Anderson1 and J.F. Bell III1

3 4 5 6

1

Cornell University, Dept. of Astronomy, Ithaca NY 14853; [email protected]

6 6

Abstract

7

The Meridiani Planum region has some of the most enigmatic and diverse geology on

8

Mars, and therefore has been a target for potential landing sites for lander and rover missions.

9

Mosaics of data from the OMEGA near-IR imaging spectrometer on the ESA Mars Express

10

(MEx) orbiter show widespread (thousands of km2) evidence of hydration in Meridiani Planum.

11

Much of the regional hydration does not appear to be attributable to strong sulfate or phyllosili-

12

cate absorption bands, though both are present in more localized units within Meridiani. We sug-

13

gest instead that much of the regional hydration signature is due to ferric oxyhydroxides such as

14

ferrihydrite or goethite. In addition, our maps of the localized phyllosilicate and sulfate signa-

15

tures in Meridiani reveal that the strongest sulfate signatures appear to be topographically and

16

stratigraphically beneath the strongest phyllosilicate signatures, possibly implying environmental

17

conditions conducive to phyllosilicate formation postdating those that favored sulfate formation.

18 19

Index Terms: Mars; Composition; Optical, infrared, and Raman spectroscopy;

20

Keywords: Mars; Surface; Mineralogy; Spectroscopy; Landing Sites

21

21 22

1.

Introduction

23

Meridiani Planum is recognized as geologically significant partly due to the detection of

24

a significant crystalline hematite signature [e.g., Christensen et al., 2000] as well as extensive

25

layered outcrops which comprise a stratigraphic section >800 m thick and which span an area

26

greater than the Colorado Plateau [Edgett, 2005; Arvidson et al., 2006]. The Opportunity rover

27

landed in Meridiani Planum in 2004, confirmed the existence of hematite on the surface, and

28

found sulfate minerals and other morphologic evidence of past surface- and/or near-surface liq-

29

uid water [e.g., Squyres et al., 2004; Grotzinger et al., 2005].

30

Recently, several high-priority potential landing sites for the NASA Mars Science Labo-

31

ratory (MSL) rover mission have been proposed in Northern Meridiani (Figure 1)

32

[http://marsoweb.nas.nasa.gov/ landingsites/index.html]. Here we present new regional maps of

33

hydrate, phyllosilicate, and sulfate spectral parameters in the region, based on data from

34

MEx/OMEGA [Bibring et al., 2004], and other measurements, and we speculate on the miner-

35

alogic origin, stratigraphic relationships, and inferred formative conditions of these mineral de-

36

posits.

37

2.

38

2.1

Methods OMEGA Mosaic

39

We have mosaicked multiple PDS-released OMEGA spectral image cubes in Northern

40

Meridiani. OMEGA cubes range in spatial resolution from 0.3 to 5.0 km per pixel, and cover a

41

spectral range of 350 nm to 5100 nm in 352 channels. The spectral sampling is 7 nm from 350 to

42

1000 nm, 14 nm from 1000 to 2500 nm and 20 nm from 2500 to 5100 nm [Bibring et al., 2005].

43

The cubes are divided by the solar spectrum and an atmospheric spectrum to correct for the ef-

44

fects of solar and atmospheric absorption lines. The atmospheric spectrum is derived from a

45

high-resolution pass over the summit of Olympus Mons [Mustard et al., 2005]. Phase angle ef-

46

fects were minimized by assuming a Lambertian surface and dividing by the cosine of the inci-

47

dence angle. The maps of Northern Meridiani discussed here span 6oS – 9oN and 352oE – 10oE.

48

When the OMEGA data cubes are map-projected, the data are re-sampled to a uniform

49

spatial resolution. In the case of overlapping cubes, the cube with the higher initial (pre-

50

projection) spatial resolution overwrites the lower resolution data. Seams are visible in the mo-

51

saic shown in Figure 1 due to variations in the overall brightness from cube to cube, likely due to

52

variations in atmospheric dust content, illumination geometry, and deviations from a perfect

53

Lambertian surface. A space of several pixels is present between cubes that are part of the same

54

orbit, due to anomalous rows at the beginning and end of image strips. Other artifacts are also

55

visible: most notably, the large nearly-vertical black stripe is due to anomalous pixels in the

56

original low-resolution cube that were removed. In most cases, there are multiple cubes in any

57

given location (e.g. the western portion of this mosaic), so the removal of this column is not

58

problematic. However, the eastern portion of Meridiani was not well covered in the PDS-

59

released data at the time our mosaics were generated. The one cube that covers the eastern por-

60

tion of the Meridiani region was very low resolution, and had anomalous columns, resulting in

61

the black stripe and “blurry” appearance.

62 63

2.2 Spectral Parameters

64

We measure the band depth of a feature of interest by defining a continuum value on both

65

sides of the absorption band, fitting a line to the continuum, and finding the difference between

66

the continuum value and the value at the band center. Dividing this difference by the continuum

67

value gives the band depth in percent [e.g., Clark and Roush, 1984; Bell and Crisp, 1993; Pelkey

68

et al., 2007].

69

For detection of minerals that are characterized by more than one absorption band, we

70

require more than one characteristic band to be present. For example, phyllosilicates are charac-

71

terized by a hydration band near 1900 nm, as well as by a metal-OH band near 2200 to 2300 nm.

72

To detect phyllosilicates, we used the same spectral parameters described by Loizeau [2007],

73

requiring a band depth >2% for both the hydration band and at least one of the metal-OH bands

74

for a positive detection. To detect and map sulfates, we adapted the CRISM sulfate parameter

75

[Pelkey et al., 2007] to the OMEGA data.

76 77

3. Results

78

3.1 Maps of Hydrated Minerals

79

Figure 2a shows a map of the 1900 nm hydration band depth in Northern Meridiani and

80

Figure 2b shows a map of the 2400 nm sulfate parameter. The regions with the strongest hydra-

81

tion signature reach a band depth of ~5%. Hydration band strength appears to correlate with high

82

albedo and high thermal inertia (Figure 3a), implying that the observed hydration may be due to

83

exposed bedrock. The strongest widespread regional hydration signature occurs within a ~250 m

84

thick [Griffes et al., 2007], late Noachian or early Hesperian aged [Hynek et al., 2002], unit that

85

has previously been identified as "etched" terrain due to its high relief (>100 m) outcrops. The

86

unit likely formed via aeolian erosion of indurated layered deposits [Hynek et al., 2002]. The

87

etched terrain unit has been suggested to be the result of deposition in a large body of water

88

[Hynek, 2004], although MER results also suggest the possibility of smaller-scale aqueous depo-

89

sition [Squyres et al., 2004; Grotzinger, et al., 2005].

90

Our mapped sulfate parameter is widespread at a low level in Meridiani (Fig. 2b), but the

91

strongest sulfate signature is much more localized, occurring within a moderate albedo and mod-

92

erate thermal inertia depression (Figure 3). An example spectrum from this location is shown in

93

Figure 4a. The spectrum matches well with that of kieserite (MgSO4•H2O), although pure kieser-

94

ite has significantly higher albedo at wavelengths shorter than ~1450 nm. Griffes et al. [2007]

95

interpreted the depression as part of a ~150 m thick "lower etched plains" unit and deduced the

96

presence of polyhydrated sulfates there. This unit was grouped with the etched terrain unit by

97

Hynek [2002], and was therefore interpreted as the same age (late Noachian – Early Hesperian).

98

However, Griffes et al. [2007] interpreted the lower etched plains as topographically and strati-

99

graphically below (older than) the surrounding "upper etched plains" unit materials.

100

Phyllosilicates are present in Meridiani in what appear at OMEGA resolution to be small

101

(several km2), scattered areas within the upper etched plains unit, as well as in the ejecta and

102

walls of two craters ~150 km to the north and northeast of site A (Figure 2c). Almost all of the

103

phyllosilicates are detected by the presence of a weak 2300 nm Fe/Mg-OH band, matched well

104

by nontronite (Figure 4a). Figure 5 shows a histogram of the areal occurrence of sulfates and

105

phyllosilicates over the range of MOLA elevations present in Meridiani. We find that the major-

106

ity of phyllosilicate signatures occur at higher elevation than the strongest sulfate signatures,

107

which are confined to the lower etched plains. The presence of phyllosilicates in the upper

108

etched terrain is particularly interesting because it implies that these phyllosilicates were em-

109

placed after the formation of the stratigraphically lower, sulfate-rich lower etched plains unit.

110

The formation of what may have been an extensive phyllosilicate-bearing unit (the upper etched

111

plains) significantly later than what may have been an extensive sulfate-bearing unit (the lower

112

etched plains) may be inconsistent with some recent hypotheses for Martian global climatic and

113

mineralogic evolution [e.g., Bibring et al., 2006]. Better knowledge of the relative and absolute

114

ages of these mineralogic marker units will be the key to further testing models of Martian global

115

climate evolution.

116

Our analysis indicates that the majority of the overall regional hydration signature in Me-

117

ridiani cannot definitively be attributed to either sulfates or phyllosilicates. We suggest instead

118

that ferric oxyhydroxides (e.g., goethite, ferrihydrite [Bishop et al., 1993]) are responsible for

119

this strong hydration feature in Meridiani, a hypothesis consistent with OMEGA visible wave-

120

length evidence for enhanced ferric mineral abundance in the upper etched plains unit [Griffes et

121

al., 2007]. Poulet et al. [2008] also find that the spectrum of the hydrated etched terrain could be

122

explained by ferric oxides, although they suggest that there may also be hydrated sulfates such as

123

amarantite (Fe2(SO4)2O•7H2O) present, whose spectral features are masked by the oxides. Figure

124

4b shows an average spectrum from the hydrated etched terrain in Meridiani, compared with sev-

125

eral hydrated ferric minerals and amarantite. The Meridiani spectrum matches reasonably well

126

with pure ferrihydrite, goethite, or a mixture of ferrihydrite and goethite. It is also similar to

127

amarantite at wavelengths longer than 1500 nm, but pure amarantite has a high albedo between

128

1100 nm and 1500 nm that is not observed in the etched terrain. At wavelengths >2100 nm, the

129

etched terrain spectrum shows small features suggestive of the 2200 and 2300 nm metal-OH

130

(phyllosilicate) features, and the 2400 nm drop characteristic of sulfates. However, these features

131

are near the level of the noise and in our view do not constitute definitive detections.

132 133

3.2

Landing Sites

134

Three high-priority potential MSL landing sites are within the bounds of our Meridiani

135

maps. The North Meridiani site, labeled “A” in the maps, has drawn attention because it has

136

been assessed as a very safe location similar to the Opportunity rover landing site. The site itself

137

does not have significant hydration, but areas just outside the landing ellipse are hydrated. There

138

is a moderately weak sulfate signature at this site, and a small patch of Fe/Mg-bearing phyllosili-

139

cates in the hydrated patch just to the east of the site.

140

The East Meridiani site, labeled “B”, was identified as a location worth further investiga-

141

tion in case one of the primary sites is eliminated. It is situated in the etched terrain on some of

142

the strongest hydration detected in the region. There is a weak sulfate signature within the site.

143

Phyllosilicates are present in small (single to several-pixel) patches within and near the possible

144

landing site, including our only Al-rich phyllosilicate detection in the region. Larger patches of

145

Fe/Mg-rich phyllosilicates are present ~100 km to the north.

146

The Miyamoto (formerly "Runcorn") crater landing site, labeled “C”, does not show evi-

147

dence of hydration or phyllosilicates on our maps. The sulfate parameter is also small at this site.

148 149

4. Conclusions

150

OMEGA maps of Meridiani show widespread hydration in the region, which correlates

151

with the high albedo and high thermal inertia “etched” terrain unit. Sulfates are widespread at

152

low levels, but the strongest evidence for sulfates is concentrated in a stratigraphically and to-

153

pographically lower unit, the “lower etched plains”. Phyllosilicates are present in small, several

154

km2 patches in the upper etched unit, stratigraphically above most of the sulfates. Much of the

155

hydration in Meridiani does not appear to be due to phyllosilicates or sulfates, and we suggest

156

instead that ferrihydrite, goethite, or other ferric oxyhydroxides are responsible.

157

Several proposed MSL landing sites are in the Meridiani region, although the site with the

158

strongest hydration (East Meridiani, "B" here) is not one of the six finalist sites. Whether from

159

orbit or with MSL, these kinds of measurements show that Meridiani merits significant further

160

investigation. The stratigraphy of hydrated minerals there may hold valuable clues to the history

161

of the region and of the planet as a whole, and serve as important tests of any models of the

162

planet's global mineralogic and climatic evolution.

163

163 164

References

165

Arvidson R.E. et al. (2006) Nature and origin of the hematite-bearing plains of Terra Meridiani

166

based on analyses of orbital and Mars Exploration rover data sets, J. Geophys. Res., 111,

167

E12S08, doi:10.1029/2006JE002728.

168 169 170

Bell III, J.F. and D. Crisp (1993), Groundbased Imaging Spectroscopy of Mars in the NearInfrared: Preliminary Results, Icarus, 104, 2-19, doi:10.1006/icar.1993.1078.

171 172

Bibring, J. et al. (2004) OMEGA: Observatoire pour la Minéralogie, l'Eau, les Glaces et l'Ac-

173

tivité, Mars Express: the scientific payload, edited by Andrew Wilson, p. 37 – 49, ESA SP-

174

1240, Noordwijk, Netherlands: ESA Publications Division.

175 176 177

Bibring J. et al. (2005) Mars Surface Diversity as Revealed by the OMEGA/Mars Express Observations, Science, 307, 1576, doi: 10.1126/science.1108806.

178 179 180

Bibring, J.-P. et al. (2006), Global Mineralogical and Aqueous Mars History Derived from OMEGA/Mars Express Data, Science, 312, 400, doi:10.1126/science.1122659.

181 182

Bishop, J.L. et al. (1993), Reflectance and Mossbauer spectroscopy of ferrihydrite-

183

montmorillonite assemblages as Mars soil analog materials. Geochim Cosmochim Acta, 57,

184

4583-4595.

185

186

Christensen P. R. et al. (2000) Detection of crystalline hematite mineralization on Mars by the

187

Thermal Emission Spectrometer: Evidence for near-surface water, J. Geophys. Res., 105,

188

9623-9642.

189 190 191

Clark, R. N., and Roush, T. L. (1984), Reflectance Spectroscopy: Quantitative Analysis Techniques for Remote Sensing Applications, J. Geophys. Res., 89, 6329–6340.

192 193

Clark R. N., Swayze G. A., Wise R., Livo K. E., Hoefen T.M., Kokaly R. F., and Sutley S. J.

194

(2007), USGS Digital Spectral Library splib06a, U.S. Geological Survey, Data Series 231.

195 196

Edgett K. S. (2005) The sedimentary rocks of Sinus Meridiani: Five key observations from data

197

acquired by the Mars Global Surveyor and Mars Odyssey orbiters, Mars, 1, 5-58,

198

doi:10.1555/mars.2005.0002.

199 200 201

Griffes J. L. et al. (2007) Geologic and spectral mapping of etched terrain deposits in northern Meridiani Planum, J. Geophys. Res., 112, E08S09, doi:10.1029/2006JE002811.

202 203

Grotzinger, J. et al. (2005) Stratigraphy and sedimentology of a dry to wet eolian depositional

204

system, Burns formation, Meridiani Planum, Mars Earth and Planetary Science Letters, 240,

205

11. doi:10.1016/j.epsl.2005.09.039.

206 207 208

Hynek B. M. et al. (2002) Geologic setting and origin of Terra Meridiani hematite deposit on Mars, J. Geophys. Res. , 107, 5088, doi:10.1029/2002JE001891.

209 210 211

Hynek B. M. (2004) Implications for hydrologic processes on Mars from extensive bedrock outcrops throughout Terra Meridiani, Nature, 431, 156-159.

212 213 214

Loizeau D. (2007) Phyllosilicates in the Mawrth Vallis region of Mars, J. Geophys. Res., 112, E08S08, doi:10.1029/2006JE002877.

215 216 217

Mustard J. F. et al. (2005) Olivine and Pyroxene Diversity in the Crust of Mars, Science, 307, 1594, doi: 10.1126/science.1109098.

218 219

Pelkey, S.M. et al. (2007), CRISM multispectral summary products: Parameterizing mineral di-

220

versity on Mars from reflectance. J. Geophys. Res., 112, E08S14, doi:10.1029/2006JE002831.

221 222

Pieters, C.M. and T. Hiroi, RELAB (Reflectance Experiment Laboratory): A NASA Multiuser

223

Spectroscopy Facility, 35th Lunar and Planetary Science Conference, March 15-19, 2004,

224

League City, Texas, abstract no. 1720, 2004.

225 226

Poulet F. et al. (2008), Mineralogy of Terra Meridiani and western Arabia Terra from

227

OMEGA/MEx and implications for their formation, Icarus, doi:10.1016/j.icarus.2007.11.031,

228

in press.

229 230 231

Squyres, S. W. et al. (2004)The Opportunity Rover's Athena Science Investigation at Meridiani Planum, Mars, Science 306, 1698, doi: 10.1126/science.1106171.

232

Figure Captions

233

Figure 1: OMEGA mosaic of Northern Meridiani, showing the approximate locations of three

234

high-priority MSL sites in the region. The circles are ~32 km in diameter.

235 236

Figure 2: a) Map of 1900 nm hydration band depth in Northern Meridiani. (b) Map of the 2400

237

nm sulfate parameter. Sulfates are concentrated in the lower etched plains unit [Griffes et al.,

238

2007]. (c) Map of phyllosilicates diagnosed by metal-OH bands >2% and the sulfate parameter

239

where it is >0.03. Phyllosilicates are in red and green, sulfates are in yellow. The basemap is an

240

inverted-value THEMIS day-IR map. The insets show the regions with the strongest phyllosili-

241

cate detections.

242 243

Figure 3: (a) TES thermal inertia on MOC wide-angle image of northern Meridiani. Red corre-

244

sponds to higher thermal inertia, and indicates the location of the etched terrain. (b) MOLA to-

245

pography of Meridiani. In parts (a) and (b), the sulfate-rich lower etched plains are indicated by

246

the arrow.

247 248

Figure 4: (a) OMEGA spectra of phyllosilicates and sulfates in Meridiani, compared with lab

249

spectra from USGS [Clark et al., 2007] (b) Etched terrain spectrum compared with spectra of

250

ferric minerals and amarantite from Brown University ReLAB. [Pieters and Hiroi, 2004] and

251

goethite from USGS [Clark et al., 2007]. In parts (a) and (b), Meridiani spectra are the average

252

of several adjacent spectra, divided by a relatively featureless spectrum from the same OMEGA

253

observation, to remove artifacts. Spectra have been scaled and offset by addition of an arbitrary

254

constant for clarity.

255 256

Figure 5: Histogram of sulfate and phyllosilicate areal occurrence as a function of elevation. The

257

sulfate parameter is widespread at a low level and matches closely with the overall topography

258

distribution. However, the strongest sulfate signature is much more concentrated, and is centered

259

at a lower elevation than most of the phyllosilicate detections. The median elevation for phyl-

260

losilicates (-1348 m; indicated by the vertical blue dashed line) is more than 200 meters above

261

the median elevation for the most concentrated sulfate signature (-1554 m; indicated by the red

262

dashed line).

263 264 265 266 267 268 269 270 271 272 273 274 275 276 277