using the modified gaussian model to extract quantitative …using the modified gaussian model to...

Using the modified Gaussian model to extract quantitative

data from lunar soils

Sarah K. Noble,1,2 Carlé M. Pieters,1 Takahiro Hiroi,1 and Lawrence A. Taylor3

Received 23 March 2005; revised 23 May 2006; accepted 7 June 2006; published 22 November 2006.

[1] The Lunar Soil Characterization Consortium (LSCC) has examined and characterizeda suite of lunar soils with a wide range of compositions and maturities. The purpose of thisstudy is to compare the Vis/NIR spectral properties of these lunar soils with theirpetrologic and chemical compositions using the modified Gaussian model (MGM) toobtain quantitative data about the character of relatively weak near-infrared absorptionbands. Useful compositional information can be extracted from high-quality soil spectrausing the MGM. The model had some difficulty fitting absorption bands in the 2 mmregion of the lunar spectrum, but bands in the 1 and 1.2 mm regions provided physicallyrealistic results. The model was able to distinguish high-Ca and low-Ca pyroxenes inthe LSCC suite of lunar soils in the appropriate relative abundance. In addition,unexpected insights into the nature and causes of absorption bands in lunar soils wereidentified. For example, at least two distinct absorption bands are required in the 1.2 mmregion of the spectrum, and neither of these bands can be attributed to plagioclase oragglutinates, but are found instead to be largely due to pyroxene.

Citation: Noble, S. K., C. M. Pieters, T. Hiroi, and L. A. Taylor (2006), Using the modified Gaussian model to extract quantitative

data from lunar soils, J. Geophys. Res., 111, E11009, doi:10.1029/2006JE002721.

1. Introduction

[2] The Moon is the only airless body from which wehave direct geologic samples, though even those samplesare restricted to a few isolated locales on the lunarnearside. The U.S. Apollo and Soviet Luna samples arean invaluable resource which can provide ‘‘ground truth’’for our remotely sensed data. By understanding in detailthe relationship of laboratory spectra of these samples totheir maturity, composition, and chemistry, we can testthe abilities and limitations of applying quantitativedeconvolution methods to comparable spectral resolutionremotely sensed data, such as that which will beobtained by missions like Japan’s SELENE and India’sChandrayaan-1.[3] The Lunar Soil Characterization Consortium (LSCC)

was formed to fully characterize the physical, chemical, andspectral properties of a suite of lunar soils [e.g., Taylor etal., 1999a, 1999b, 2000a, 2000b, 2000c, 2001a, 2001b,2002, 2003; Pieters et al., 2000, 2002; Pieters and Taylor,2003]. Soil samples selected for the consortium study werechosen to represent the widest variety of compositions andmaturities from the lunar environment. The soils are listedin Table 1 along with their maturity (as indicated by Is/FeO)and the measured modal abundance of selected mineral andglass phases.

[4] The soils were wet sieved (triply distilled water) intoseveral size fractions and reliable data on mineral phaseswere acquired. The consortium has concentrated on the 10–20 mm and 20–45 mm size fractions, because the opticalproperties of these sizes bear the greatest resemblance tothose of the bulk soil [Pieters et al., 1993]. The

Sunshine et al. [1990, 1999]. The MGM is a method fordeconvolving spectra into individual absorption bands. Onthe basis of crystal field theory [Burns, 1993], the MGMbeen demonstrated to be a physically realistic model ofelectronic transition absorption bands. There has beensubstantial success in applying the MGM to modelingminerals and simple mixtures [Sunshine and Pieters,1993, 1998]. The model has also been applied to Martianmeteorites and even remote data from Mars [Sunshine etal., 1993; Mustard and Sunshine, 1995]. Applying theMGM to lunar soils requires special consideration becausethe lunar continuum, which is largely a result of space-weathering processes, must first be accurately modeled andremoved.[6] We have developed a method of continuum removal

specifically for lunar soil samples [Hiroi et al., 2000],utilizing a three-component solution consisting of a linear-in-energy term (c�1), a linear-in-wavelength term (c1), andan offset (c0), as shown in equation (1) below. This three-term continuum allows more realistic solutions than the

earlier method of double-linear removal [Hiroi and Pieters,1998].

C lð Þ ¼ c�1=lþ c0 þ c1l

l ¼ wavelength

c�1; c0; c1 ¼ constants

ð1Þ

This method of continuum removal is compatible across alarge array of compositions andmaturities.Ueda et al. [2002]attempted to further account for space-weathering effects byadding an additional term to this continuum removal method,on the basis of Hapke’s space-weathering model [Hapke,2001]. However, it was found that this modified equationonly works well for very small degrees of space weathering,and thus it is not generally appropriate for lunar soils.[7] In MGM analysis, each absorption band is fit to a

modified Gaussian shape which can be fully described bythree terms: band center, band strength, and bandwidth

Table 1. Basic Characteristics of the Lunar Soils Used in This Study

Sample Relab ID Morris Is/FeOa LSCC Is/FeO Plagioclase Agglutinitic Glass Total Pyx Low-Ca: High-Ca Pyx

20–45 mm Mare10084 LR-CMP-069 78 88 16.8 52.9 16.0 27:7312001 LR-CMP-073 56 51 13.5 56.2 19.8 53:4712030 LR-CMP-065 14 12 15.3 39.4 33.8 56:4415041 LR-CMP-077 94 66 15.5 51.3 22.5 64:3715071 LR-CMP-080 52 49 18.3 47.6 22.1 61:3970181 LR-CMP-046 47 53 16.9 43.4 15.7 40:6171061 LR-CMP-057 14 9 13.9 31.4 20.8 39:6171501 LR-CMP-054 35 28 16.5 38.3 21.3 37:6479221 LR-CMP-049 81 57 16.9 46.5 13.5 39:61

20–45 mm Highland14141 LR-CMP-094 5.7 5.8 26.6 41.0 19.8 79:2114163 LR-CMP-090 57 43 18.9 56.4 16.2 75:2514259 LR-CMP-086 85 77 14.1 60.5 18.2 75:2514260 LR-CMP-083 72 80 15.6 64.0 13.7 71:2961141 LR-CMP-110 56 76 42.5 50.1 4.4 70:3061221 LR-CMP-106 9.2 8.4 58.5 28.9 7.4 71:2962231 LR-CMP-098 91 81 40.4 50.6 5.1 67:3364801 LR-CMP-102 82 83 39.3 53.6 4.5 70:3067461 LR-CMP-114 25 22 64.4 25.4 7.3 63:3767481 LR-CMP-118 31 21 61.3 27.6 6.6 68:32

10–20 mm Mare10084 LR-CMP-070 78 145 12.2 57.0 12.2 31:6912001 LR-CMP-074 56 67 13.9 56.8 17.9 53:4712030 LR-CMP-066 14 17 14.0 49.8 21.4 60:4015041 LR-CMP-063 94 92 16.2 56.7 17.0 62:3815071 LR-CMP-061 52 80 19.4 49.2 16.7 59:4270181 LR-CMP-047 47 63 18.3 51.7 8.5 45:5671061 LR-CMP-059 14 14 15.2 37.9 12.5 43:5771501 LR-CMP-055 35 50 19.8 44.8 13.7 45:5679221 LR-CMP-040 81 78 16.0 54.3 9.7 47:53

10–20 mm Highland14141 LR-CMP-095 5.7 12 28.0 48.6 10.9 80:2114163 LR-CMP-091 57 65 18.3 58.5 13.8 77:2314259 LR-CMP-087 85 102 15.4 68.7 9.1 76:2414260 LR-CMP-082 72 99 16.1 65.2 12.1 78:2261141 LR-CMP-111 56 82 41.3 49.8 5.3 72:2861221 LR-CMP-107 9.2 14 59.4 32.6 5.3 61:3962231 LR-CMP-099 91 110 40.7 51.0 5.4 66:3464801 LR-CMP-103 82 85 34.5 61.0 2.8 78:2267461 LR-CMP-115 25 24 61.0 32.4 4.1 62:3867481 LR-CMP-119 31 33 62.0 28.5 5.7 67:33

aBulk

(FWHM), as shown in Figure 1a. The starting values forthe continuum are derived using contact points at 0.75, 1.5,and 2.6 mm for each sample. The same starting parametersare given for the center and widths of absorption bands forevery sample. The fitting program allows additional con-straints to be placed on the range of centers, widths, andstrengths for each band. The model simultaneously fitsboth the continuum and the absorption bands, while mini-mizing the RMS error. Additional bands are added as neededuntil the residual is small and there are no systematic features(Figures 1b and 1c). The resulting band parameters (center,strength, and width) provide quantitative information aboutthe best fit for each absorption band that can then be directlycompared to measured compositional data.

[8] Lunar soils are composed of several minerals, as wellas volcanic glasses and weathering products. Typical spectraof the more common components of lunar soils are shown inFigure 2. Lunar spectra tend to be dominated by pyroxene,which has very prominent absorption features centered near 1and 2 mm, as well as a weaker band near 1.2 mm [e.g.,Sunshine et al., 1990]. The locations of the 1 mm and 2 mmfeatures are highly dependent on the pyroxene composition:High-Ca pyroxenes have bands centered at longer wave-lengths than low-Ca pyroxenes [Adams, 1974; Burns, 1993;Cloutis and Gaffey, 1991]. Iron content also plays a role inband location, with the 1 and 2 mm bands moving to longerwavelengths and the 1.2 mm band strengthening with in-creasing iron [Adams, 1974; Klima et al., 2005]. The 1.2 mm

Figure 1. Example deconvolution of the 10–20 mm fraction of highland soil 61221: (a) Three-banddeconvolution; arrows indicate systematic errors illustrating the need for additional bands. (b) Five-banddeconvolution; persistent error near 1.25 mm is marked with an arrow. (c) Six-band deconvolution; notethat the error has been reduced. The bands are labeled one through six as referred to throughout the text.

E11009 NOBLE ET AL.: MGM MODELING OF LUNAR SOILS

3 of 17

E11009

region is particularly complex with several soil componentsexpressing absorption bands in the region. Plagioclase feld-spar has a band at about 1.2 mm [Adams, 1975;McCord et al.,1981]. Olivine has a compound band centered near 1.1 mm[e.g., Burns, 1993; Sunshine et al., 1990]. Glasses often havebroad absorption bands in the same general region [e.g.,Maoet al., 1978]. The glass sample in Figure 2 is a quenched glasscreated from an Apollo 17 impact melt breccia [Tompkins etal., 1996]. Compared to the quenched glass, the spectrum ofagglutinates (hand picked from Apollo 11 soil 10084) is verydark and almost featureless [Pieters et al., 1993]. Aggluti-nates and agglutinitic glasses are rich in nanophase and largerblebs of metallic iron that act to reduce their albedo. Ilmeniteand other opaque minerals also darken the spectra throughoutthe region of interest.

3. Results

[9] Initially, each soil was fit with only three bandscentered at approximately 1, 1.28, and 2 mm, using theinitial parameters listed in Table 2. As demonstrated inFigure 1a, these three-band fits produced unacceptablesystematic errors, indicating that additional bands are nec-essary [e.g., Sunshine et al., 1990]. This was not unexpected,since the compositional data [Taylor et al., 2001a] indicatethat both high-Ca and low-Ca pyroxenes are present in all ofthe soils. The model was then run using five bands: a bandnear 1.28 mm, plus two bands near 1 mm and two near 2 mmto account for the high-Ca and low-Ca pyroxene. Anexample of a five-band fit is shown in Figure 1b. With fivebands, the errors have been considerably reduced.[10] However, it was noticed that even with the five-band

fit, there was still a consistent error in the results for allsamples between approximately 1.2 and 1.3 mm. The recur-ring error is highlighted in several examples in Figure 3. Asixth band is required to eliminate this error. The initialparameters for the six-band fits are listed in Table 3 and anexample is shown in Figure 1c. In addition, a seventh bandwas needed at �0.6 mm for the four Apollo 17 soils. This

additional band is required because of the presence ofilmenite-rich pyroclastic glass. The final continuum andGaussian parameters are compiled in Tables 4a, 4b, 5a,and 5b. With six (or seven) bands, the residual errorshave been reduced such that no systematic structure isapparent, as can be seen in Figure 4.

4. Discussion

4.1. One and Two mm Bands (Pyroxenes)[11] Different compositions of pyroxene appear to con-

tribute to all six of the bands identified in our study. InFigure 5, graphs are shown of the total pyroxene contentversus the band strength for each of the six bands. A strongcorrelation is expected between the total pyroxene contentand the 1 mm and 2 mm bands (bands 1, 2, 5, and 6), and infact, all four bands do show a correlation with pyroxenecontent, though the correlation with band 5 is weaker thanthe others. In addition, it appears that the 1.2 mm bands(bands 3 and 4) are correlated to pyroxene content as well.These 1.2 mm bands are discussed in further detail in thenext section. Some of the scatter seen in these graphs relatesto differences in degree of space weathering among thesamples. Because the products of space weathering result inan attenuation of absorption bands, more mature soils willtend to have weaker bands, regardless of pyroxene content.Also, the different compositions of pyroxenes will effectwhich band they correlate with; as is discussed furtherbelow, bands 1 and 5 should be dominated by orthopyrox-ene and bands 2 and 6 will be more strongly influenced byclinopyroxene.[12] Adams [1974] illustrated the regular relationship

between the band centers of the 1 mm versus the 2 mmband for a suite of pyroxenes and demonstrated how thewavelengths of the bands shift as a result of differing Ca2+

and Fe2+ contents. In Figure 6, we have superimposed theband center results from our MGM deconvolutions onto hisoriginal data along with similar data from other studies[Adams, 1975; Hazen et al., 1978; Cloutis and Gaffey,1991]. The modeled bands for the LSCC samples fallroughly within the same range as the previous works forpyroxenes. Specifically, the results of band 1 versus band5 (the low-Ca pyroxene bands) for both mare and highlandsoils fall in the middle of the previous data for orthopyrox-ene. The results for the mare soils of band 2 versus band 6(the high-Ca pyroxene bands) also fall squarely within theprevious data.[13] By contrast, band centers for the majority of the

highland high-Ca pyroxene bands (2, 6) fall slightly to theleft of the data cloud. Specifically, for the highland soils inour study, either band 2 (at 1 mm) is shifted to slightlylonger wavelengths or band 6 (at 2 mm) is shifted to slightlyshorter wavelengths than was the case for the mare soils.Adams [1974] suggested that the addition of olivine wouldcause samples to fall above this l-l trend. However, in our

Figure 2. Representative reflectance spectra of the majorcomponents of lunar soils. The following samples fromthe RELAB database were used: olv: PO-CMP-026, plag:SW-CMP-012, glass: LS-CMP-035A, cpx: LS-CMP-009,opx: SB-RGB-006, aggl: LS-CMP-045, ilm: JB-JLB-308.

Table 2. Starting Parameters for Three-Band Deconvolutions

ModifiedGaussians Center Constraints Width Constraints

1 0.97 mm/± 0.07 0.25 mm/± 0.152 1.28 mm/± 0.05 0.45 mm/± 0.053 2.10 mm/± 0.20 0.40 mm/± 0.20


4 of 17

E11009

suite, both the highland and mare samples are relatively lowin olivine, containing only about 1–5 vol.%. In addition, thehighland samples generally contain slightly less olivine thanthe mare samples. Thus this is unlikely to be the cause of theobserved deviation. The addition of Fe-bearing glass mayalso result in band centers that fall off the pyroxene trend.However, there is nothing in our data that suggests that thehighland soils would be preferentially affected by thepresence of glass. In fact, it is the four Apollo 17 maresoils that contain high concentrations of pyroclastic beads. Itis those soils that might be expected to be most influencedby glass absorptions, but the mare soils all fall on theexpected trend. Rather than band 2 (near 1 mm) beingshifted to longer wavelengths, it is more likely that band6 (near 2 mm) is shifted to shorter wavelengths for thehighland soils. The low-pyroxene contents of the highlandsoils make them particularly difficult to fit. Furthermore,

because of the method of continuum removal and band fittingutilized for these soils, we have found that the longest-wavelength band (6) is the most difficult band from whichto obtain consistent and reasonable results. While composi-tional differences cannot be ruled out, the shift seen in thehighland soils is likely to be an artifact of the approach.

Figure 3. Errors from five-band deconvolutions for several mare and highland soils. The persistenterror near 1.2–1.3 mm can be identified in all of the soils and indicates that an additional band isnecessary.

Table 3. Starting Parameters for Six- (or Seven-) Band

Deconvolutions

Modified Gaussians Center Constraints Width Constraints

1 0.91 mm/± 0.07 0.20 mm/± 0.152 1.01 mm/± 0.07 0.20 mm/± 0.153 1.25 mm/± 0.05 0.30 mm/± 0.104 1.28 mm/± 0.05 0.45 mm/± 0.055 1.83 mm/± 0.20 0.35 mm/± 0.156 2.27 mm/± 0.20 0.35 mm/± 0.157 (if needed) 0.60 mm/± 0.10 0.20 mm/± 0.10


5 of 17

E11009

[14] The combined strength of the four 1 mm and 2 mmpyroxene bands (bands 1, 2, 5, and 6) is compared to totalpyroxene content in Figure 7. As expected, there is a cleartrend of increasing band strength with increasing pyroxenecontent. The standard deviation of a linear regression fit tothis plot is ±5.17%. As in Figure 5, some of the scatter inthis graph is due to differences in degree of space weath-ering. Though it may seem counterintuitive, there is also atendency for highland soils to actually have strongerpyroxene bands than mare soils for a given pyroxenecontent. This is not due simply to differences in degree ofweathering, but rather relates to the fact that highland soilscontain fewer opaques, such as ilmenite, which tend toreduce spectral contrast overall.[15] One of the goals of this study was to test whether the

model could predict the relative abundance of low-Ca andhigh-Ca pyroxene for lunar soils. Since using only threebands produced such poor fits to the data (Figure 1a), it isclear that multiple bands are required to account for thepyroxene diversity. In determining soil mineralogy andcomposition, the pyroxenes were classified into four cate-gories: orthopyroxene, pigeonite, Mg-rich clinopyroxene(augite), and Fe-rich clinopyroxene [Taylor et al., 1999a].A pyroxene quadrilateral with the locations of these pyrox-ene types is shown in Figure 8. In order to relate themineralogy to the spectral analysis, in some cases, we havecombined the orthopyroxene and pigeonite measurementsinto a ‘‘low-Ca pyx’’ category and the Mg-rich and Fe-richpyroxenes into a ‘‘high-Ca pyx’’ category. This grouping onthe basis of Ca content necessarily neglects the role of iron indetermining band center, but some simplifying assumptionwas necessary to reduce the data for comparison with thepyroxene-related bands. We have chosen to concentrate oncalcium differences because they are more important fordetermining band centers than iron.[16] If we assume that the strength of bands 1 and 2 is

related largely to the calcium content of the pyroxene, thenthe ratio of the strengths should be directly proportional tothe ratio of low-Ca to high-Ca pyroxene in a given soil[Sunshine and Pieters, 1993]. The same should also be true

of bands 5 and 6. Comparison of the ratio of low-Ca tohigh-Ca pyroxene to the ratio of the strength of band 1 toband 2, as well as of band 5 to band 6, are shown inFigures 9a and 9b, respectively.[17] For the 1 mm region (Figure 9a), the 10–20 mm and

the 20–45mm fractions of mare soil, as well as the 20–45mmfraction of highland soils, show strong correlations betweenband strength and composition. The 10–20 mm fractionsof highland soils are, however, considerably more scat-tered. The combination of low total pyroxene contents andthe tendency for weathering products to increase in thefiner fractions likely contributes to the poor fit for thosesoils. If the highland 10–20 mm fraction is ignored, theremaining soils have an R2 of 0.74 and a standarddeviation of ±0.92.[18] In contrast to the results in the 1 mm region, no simple

relationship is seen between composition and band strengthin the 2 mm region (Figure 9b). As evidenced by Figure 6,our model has been shown to have difficulty in fitting the2 mm region, thus it is disappointing, but not entirelysurprising that such a poor relationship is observed.

4.2. The 1.2 mm Region[19] Interpretation of the 1.2 mm region of lunar soil

spectra is complex. At least two distinct bands appear tobe required to adequately fit the LSCC soil data, as indicatedby the consistent error obtained when only one band isutilized (Figure 3). With two bands, we have eliminated thesystematic errors in our fits (Figure 4); however, given thecomplexity of this region, it is entirely possible that multiplecomponents are represented by these bands, including fer-rous iron in the M1 site of orthopyroxene, pigeonite, andclinopyroxene, as well as iron in plagioclase.[20] The two bands identified by our model are labeled

bands 3 and 4. Band 3 is a stronger narrower band at shorterwavelengths (�1.22–1.26 mm). Band 4 is a wider, weakerband at longer wavelengths (�1.26–1.28 mm). The twobands are highly correlated in their band strength (R2 =

Table 4b. Highland Error and Continuum Parameters

Sample RMS

Continuum Parameters

c(�1) c(0) c(1)20–45 mm

14259 4.86E-03 �5.15E-01 �2.01E+00 2.84E-0114260 4.69E-03 �4.95E-01 �2.08E+00 3.12E-0114163 4.86E-03 �5.15E-01 �2.01E+00 2.84E-0114141 3.49E-03 �2.97E-01 �1.32E+00 1.66E-0161221 2.09E-03 �1.53E-01 �1.02E+00 1.24E-0161141 2.32E-03 �4.89E-01 �1.58E+00 2.41E-0162231 1.99E-03 �3.54E-01 �1.65E+00 2.62E-0164801 2.11E-03 �4.25E-01 �1.34E+00 2.28E-0167461 2.13E-03 �4.23E-01 �7.68E-01 1.05E-0167481 2.07E-03 �4.08E-01 �9.92E-01 1.18E-01

10–20 mm14259 3.18E-03 �6.86E-01 �1.35E+00 2.02E-0114260 3.36E-03 �7.11E-01 �1.30E+00 1.75E-0114163 2.15E-03 �5.35E-01 �1.40E+00 1.99E-0114141 2.29E-03 �3.14E-01 �1.04E+00 1.47E-0161221 1.30E-03 �2.11E-01 �7.15E-01 7.14E-0261141 2.16E-03 �5.41E-01 �1.09E + 00 1.58E-0162231 2.49E-03 �5.03E-01 �1.07E + 00 1.71E-0164801 2.06E-03 �5.30E-01 �9.91E-01 1.42E-0167461 1.51E-03 �3.91E-01 �6.13E-01 7.08E-0267481 1.60E-03 �4.00E-01 �6.75E-01 9.60E-02

Table 4a. Mare Error and Continuum Parameters

Sample RMS

Continuum Parameters

c(�1) c(0) c(1)20–45 mm

79221 3.54E-03 �3.18E-01 �2.48E+00 3.39E-0170181 2.91E-03 �3.53E-01 �2.40E+00 3.27E-0171501 3.29E-03 �2.96E-01 �2.39E+00 3.05E-0171061 3.30E-03 5.26E-02 �2.53E+00 4.37E-0110084 3.87E-03 �2.22E-01 �2.72E+00 3.67E-0112001 2.90E-03 �4.53E-01 �2.16E+00 2.72E-0112030 2.90E-03 �3.66E-01 �1.51E+00 1.35E-0115041 3.59E-03 �3.40E-01 �2.38E+00 3.62E-0115071 3.78E-03 �4.92E-01 �1.99E+00 2.71E-01

10–20 mm79221 4.28E-03 �5.09E-01 �2.19E+00 3.13E-0170181 2.89E-03 �4.88E-01 �2.05E+00 2.96E-0171501 2.69E-03 �4.09E-01 �2.23E+00 3.00E-0171061 2.95E-03 �2.32E-01 �2.09E+00 3.96E-0110084 3.27E-03 �3.93E-01 �2.36E+00 3.18E-0112001 3.40E-03 �6.31E-01 �1.82E+00 2.46E-0112030 3.08E-03 �3.26E-01 �1.59E+00 2.11E-0115041 4.06E-03 �6.34E-01 �1.88E+00 2.36E-0115071 3.50E-03 �6.03E-01 �1.79E+00 2.44E-01


6 of 17

E11009

Table

5b.HighlandSoilsDeconvolutionResults

Sam

ple

Band1

Band2

Band3

Band4

Band5

Band6

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

20–45mm

14259

0.910

0.202

0.118

1.011

0.225

0.107

1.237

0.278

0.058

1.267

0.440

0.041

1.895

0.353

0.081

2.167

0.385

0.054

14260

0.910

0.187

0.129

1.008

0.204

0.115

1.230

0.241

0.067

1.272

0.447

0.032

1.883

0.363

0.069

2.138

0.338

0.040

14163

0.913

0.181

0.169

1.013

0.202

0.131

1.237

0.257

0.045

1.270

0.444

0.024

1.898

0.334

0.058

2.203

0.436

0.084

14141

0.921

0.157

0.267

1.006

0.201

0.157

1.239

0.221

0.045

1.261

0.440

0.019

1.877

0.426

0.133

2.140

0.429

0.097

61221

0.910

0.175

0.127

1.024

0.219

0.119

1.236

0.263

0.042

1.268

0.443

0.023

1.884

0.425

0.061

2.159

0.440

0.034

61141

0.904

0.207

0.064

1.025

0.238

0.062

1.236

0.275

0.035

1.273

0.447

0.025

1.898

0.350

0.014

2.112

0.326

0.010

62231

0.911

0.181

0.054

1.026

0.221

0.057

1.242

0.277

0.026

1.271

0.445

0.023

1.881

0.347

0.030

2.149

0.338

0.022

64801

0.915

0.170

0.048

1.032

0.226

0.049

1.240

0.260

0.019

1.267

0.445

0.017

1.917

0.415

0.032

2.312

0.458

0.048

67461

0.925

0.149

0.056

0.994

0.291

0.110

1.272

0.223

0.034

1.260

0.438

0.022

1.913

0.376

0.043

2.130

0.379

0.021

67481

0.919

0.141

0.075

1.024

0.215

0.069

1.240

0.257

0.044

1.256

0.436

0.020

1.959

0.418

0.038

2.070

0.392

0.010

20–10mm

14259

0.921

0.201

0.064

1.013

0.284

0.065

1.240

0.264

0.027

1.269

0.444

0.018

1.922

0.377

0.022

2.128

0.333

0.015

14260

0.918

0.190

0.075

1.020

0.235

0.069

1.247

0.293

0.035

1.266

0.440

0.031

1.900

0.345

0.033

2.167

0.286

0.015

14163

0.932

0.160

0.080

1.006

0.296

0.078

1.244

0.258

0.025

1.269

0.444

0.018

1.904

0.387

0.040

2.154

0.373

0.026

14141

0.932

0.149

0.146

0.996

0.253

0.067

1.248

0.251

0.022

1.264

0.443

0.013

1.904

0.471

0.064

2.139

0.475

0.038

61221

0.921

0.148

0.053

1.029

0.219

0.046

1.250

0.245

0.017

1.262

0.439

0.014

1.902

0.361

0.029

2.228

0.461

0.027

61141

0.928

0.202

0.047

1.019

0.312

0.055

1.250

0.273

0.022

1.268

0.443

0.017

1.927

0.362

0.018

2.158

0.309

0.004

62231

0.929

0.238

0.039

1.020

0.294

0.051

1.248

0.257

0.019

1.263

0.440

0.017

1.901

0.331

0.011

2.137

0.438

0.013

64801

0.941

0.186

0.037

1.050

0.392

0.039

1.258

0.291

0.006

1.273

0.443

0.007

1.892

0.319

0.008

2.120

0.440

0.012

67461

0.922

0.149

0.079

1.018

0.237

0.069

1.265

0.227

0.019

1.257

0.430

0.014

1.922

0.397

0.046

2.161

0.390

0.024

67481

0.931

0.154

0.065

1.033

0.283

0.049

1.282

0.223

0.013

1.269

0.437

0.009

1.946

0.394

0.031

2.148

0.310

0.009

Table

5a.MareSoilsDeconvolutionResults

Sam

ple

Band1

Band2

Band3

Band4

Band5

Band6

Band7

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

Center

Width

Strength

20–45mm

79221

0.908

0.202

0.093

1.028

0.198

0.118

1.234

0.263

0.049

1.275

0.448

0.035

1.879

0.379

0.037

2.205

0.405

0.037

0.614

0.206

0.007

70181

0.911

0.200

0.088

1.024

0.196

0.118

1.238

0.269

0.036

1.276

0.449

0.030

1.890

0.412

0.040

2.256

0.448

0.051

0.630

0.210

0.023

71501

0.920

0.175

0.121

1.026

0.179

0.168

1.243

0.275

0.034

1.280

0.451

0.029

1.894

0.426

0.067

2.268

0.443

0.095

0.608

0.213

0.066

71061

0.914

0.146

0.171

1.028

0.160

0.208

1.237

0.255

0.068

1.273

0.446

0.036

1.876

0.458

0.100

2.331

0.531

0.114

0.589

0.200

0.124

10084

0.925

0.145

0.095

1.033

0.163

0.150

1.228

0.257

0.042

1.279

0.451

0.027

1.948

0.360

0.015

2.277

0.451

0.064

12001

0.914

0.180

0.147

1.022

0.196

0.192

1.232

0.251

0.071

1.269

0.444

0.037

1.921

0.386

0.066

2.219

0.440

0.092

12030

0.914

0.160

0.235

1.019

0.181

0.260

1.223

0.259

0.092

1.263

0.440

0.037

1.931

0.468

0.109

2.256

0.560

0.146

15041

0.909

0.172

0.139

1.030

0.199

0.143

1.231

0.253

0.059

1.273

0.447

0.029

1.937

0.374

0.059

2.250

0.453

0.066

15071

0.911

0.176

0.174

1.026

0.200

0.178

1.234

0.254

0.070

1.267

0.443

0.035

1.917

0.370

0.076

2.218

0.452

0.080

20–10mm

79221

0.895

0.306

0.101

0.998

0.202

0.127

1.225

0.261

0.048

1.274

0.448

0.036

1.780

0.338

0.026

2.179

0.374

0.016

0.608

0.203

0.005

70181

0.891

0.272

0.064

1.015

0.203

0.084

1.237

0.268

0.033

1.277

0.451

0.030

1.840

0.370

0.012

2.187

0.379

0.018

0.620

0.207

0.010

71501

0.916

0.209

0.065

1.017

0.188

0.097

1.246

0.284

0.024

1.276

0.448

0.027

1.876

0.380

0.022

2.271

0.406

0.034

0.635

0.213

0.043

71061

0.906

0.169

0.056

1.016

0.163

0.067

1.255

0.260

0.032

1.280

0.449

0.025

1.755

0.403

0.031

2.219

0.442

0.017

0.629

0.233

0.077

10084

0.905

0.231

0.052

1.023

0.184

0.087

1.231

0.261

0.034

1.276

0.449

0.024

1.832

0.353

0.001

2.299

0.396

0.020

12001

0.922

0.196

0.111

1.015

0.206

0.137

1.230

0.254

0.050

1.270

0.446

0.034

1.925

0.386

0.041

2.222

0.448

0.053

12030

0.920

0.143

0.186

1.013

0.174

0.195

1.232

0.258

0.051

1.263

0.440

0.028

1.906

0.456

0.092

2.247

0.525

0.088

15041

0.917

0.212

0.105

1.014

0.216

0.126

1.231

0.262

0.048

1.272

0.447

0.036

1.885

0.382

0.040

2.198

0.440

0.056

15071

0.922

0.198

0.130

1.017

0.225

0.138

1.232

0.255

0.054

1.266

0.442

0.032

1.902

0.400

0.053

2.217

0.463

0.063


7 of 17

E11009

Figure 4. Errors from six-band deconvolutions for several mare and highland soils. Systematic structurehas been eliminated.


8 of 17

E11009

Figure 5. Total pyroxene content versus band strength for each of the six bands. For this and allsubsequent plots: (open symbols) highland soils, (closed symbols) mare, (diamonds) 10–20 mm sizefractions, and (squares) 20–45 mm fraction. In addition: (black) mature soils, (medium gray) submaturesoils, and (light gray) immature soils.


9 of 17

E11009

0.75) (Figure 10). This correlation suggests that the cause(s)of the bands is not independent.[21] A band at roughly 1.2 mm in lunar soils has often

been attributed to plagioclase [e.g., McCord et al., 1981].However, Klima et al. [2005] have clearly documented thisfeature in Fe-Mg pyroxenes. Similarly, using the MGMmodel, Sunshine et al. [1993] found that the 1.2 mm band ina Martian meteorite was due solely to the pyroxene in thesample and was not related to plagioclase. Our data confirmthat this association with pyroxene is also the case for lunarsoils. As illustrated in Figure 11, plagioclase content of thesoil has almost no effect on the strength of band 3 or band 4,even for highland samples that are greater than 60 vol.%plagioclase. The centers of bands 3 and 4 (Figures 11cand 11d) do show slight correlations with plagioclase,suggesting that plagioclase may have some impact on thisregion, but it is clearly not a major contributor.[22] Agglutinitic glasses often constitute a significant

fraction of lunar soils, up to 70 vol.% of our most maturesamples. However, they have no influence on either thestrength or location of bands 3 or 4, as can be seen inFigures 12a and 12b). Certain lunar soils are rich in volcanicglasses, some of which also have bands in the 1.2 mmregion. For example, in our suite, the four Apollo 17 soilscontain a significant fraction of volcanic glass beads. Aswas previously reported [Noble et al., 1999], this glass doesappear to have some influence on the strength of bands inthe 1.2 mm region in these soils (Figures 12c and 12d).[23] As illustrated in Figures 5c and 5d, total pyroxene

content is correlated to the strength of bands 3 and 4. InFigure 13, we have separated the pyroxenes into their fourclasses and compared them each to the strength of bands 3and 4: Orthopyroxene content shows a very weak correla-tion for highland soils, but no obvious relationship for maresoils; pigeonite, on the other hand, is well correlated to the

strength of band 3 and, to a lesser extent, band 4 for bothmare and highland soils; Mg-rich clinopyroxene shows norelationship with either band; and Fe-rich clinopyroxene,though constituting only a very small percentage of the totalpyroxene, is surprisingly well correlated to the strength ofboth bands 3 and 4.[24] Olivine is a minor component of the soils in our

study, composing no more than 5 vol.% of any of oursamples. However, the absorption bands in olivine extendinto the 1.2 mm region (Figure 1); thus it is possible thatolivine could be contributing to one or both of the bands.Figures 14a and 14b illustrate that olivine appears to be veryweakly correlated to band strength of both bands 3 and 4.There is no correlation between olivine content and bandcenter for bands 3 and 4 (Figures 14c and 14d). The weakcorrelation with band strength is likely a consequence of thecorrelation between olivine content and pyroxene contentand probably does not represent a true relationship betweenthe 1.2 mm band and olivine, though the possibility cannotbe fully eliminated.

4.3. Maturity

[25] Is/FeO is a standard measurement of maturity (lengthof exposure time) for lunar soils. The parameter, Is, repre-sents the relative strength of the ferromagnetic resonance(due to Fe0 particles between about 4 and 33 nm) of a soil,which is then normalized by the total FeO content of the soilto account for compositional differences [Morris, 1978].The Is/FeO for each of the size fractions of the LSCC soilswere measured. Maturity, as indicated by Is/FeO, has littleeffect on the band center or widths of the deconvolved soils.Space weathering is known to reduce the strength ofabsorption bands [e.g., Pieters et al., 1993; Fischer andPieters, 1994], thus it is not surprising that the band strengthof nearly all the bands are anticorrelated to Is/FeO. Figure 15shows plots of Is/FeO versus band strength for each of thesix bands. Five of the six bands show essentially the same

Figure 6. Plot of location of the center of the 1 mm bandversus the center of the 2 mm band. The data from this studyare superimposed on data from previous studies: Adams–Adams [1974, 1975]; Hazen–Hazen et al. [1978]; C&G–Cloutis and Gaffey [1991].

Figure 7. Combined band strength of bands 1, 2, 5, and 6versus total pyroxene content. (open symbols) highlandsoils, (closed symbols) mare. In addition: (black) maturesoils, (medium gray) submature soils, and (light gray)immature soils. R2 = 0.79, s = ±5.17%.


10 of 17

E11009

trend: Immature soils may have strong or weak bands, butmature soils (those with high-Is/FeO values) never havestrong bands. Band 4 (Figure 15d) does not follow thistrend. Here we see that even mature soils can have strongband 4s; this is particularly true among the mare soils, whichgenerally have a stronger band 4 than the highland soils.Though the previous graphs indicate that pyroxene is at leastpartially responsible for band 4, Figure 15d suggests that anadditional soil component is also contributing to the band,and that component is either unaffected by space weather-ing, or it is actually created in the weathering process.

5. Conclusions

[26] Dissecting the modeling of a suite of well-charac-terized lunar soils has demonstrated both the capabilitiesand the limitations of this model utilizing the MGM methodof spectral deconvolution as a quantitative tool for lunarsoils. The model does a good job eliciting data from the1 and 1.2 mm regions. Low-pyroxene contents and highdegrees of space weathering result in lower accuracy inpredictions of pyroxene composition in the 2 mm region.[27] As a method of spectra deconvolution, the modified

Gaussian model was able to identify both high-Ca and

Figure 8. Pyroxene quadrilateral showing the location of the average composition for the four pyroxenecategories for each soil.

Figure 9. Ratio of low-Ca to high-Ca pyroxene comparedto the ratio of (a) the band strength of band 1 versus band 2and (b) the band strength of band 5 versus band 6. Thepreponderance of low-Ca pyroxene in the highlands isreadily identified.

Figure 10. The strength of band 3 versus the strength ofband 4. The bands are highly correlated (R2 = 0.75).


11 of 17

E11009

Figure 11. Plagioclase content versus properties of the 1.2 mm band (bands 3 and 4): (a) Strength ofband 3. (b) Strength of band 4. (c) Center of band 3. (d) Center of band 4.


12 of 17

E11009

Figure 12. Agglutinate and glass content versus the 1.2 mm band (bands 3 and 4): (a) Strength of band3 versus agglutinate content. (b) Strength of band 4 versus agglutinate content. (c) Volcanic glass contentversus strength of band 3 for the Apollo 17 soils. (d) Volcanic glass content versus strength of band 4 forthe Apollo 17 soils.


13 of 17

E11009

Figure 13. The 1.2-mm band (bands 3 and 4) versus the four categories of pyroxenes.


14 of 17

E11009

Figure 14. Olivine content versus the 1.2 mm band (bands 3 and 4): (a) Strength of band 3 versusolivine content. (b) Strength of band 4 versus olivine content. (c) Center of band 3 versus olivine content.(d) Center of band 4 versus olivine content.


15 of 17

E11009

low-Ca pyroxene in a suite of lunar soils with a range ofweathering degrees. In addition to the well-known pyroxeneabsorption bands near 1 and 2 mm, it was discovered that atleast two distinct �1.2 mm bands (bands 3 and 4) arerequired by the data. Plagioclase and agglutinates have been

eliminated as possible causes of the 1.2 mm bands. Rather,pyroxenes appear to be largely responsible for those bands.The Is/FeO data suggest that there is also an additionalunidentified component influencing the strength of band 4,probably related to space weathering.

Figure 15. Band strength versus maturity as measured by Is/FeO for each of the six bands.


16 of 17

E11009

[28] The strong correlation between band strength andpyroxene content (Figure 7) suggests that for high-qualityspectra data, the MGM does allow a predictive capability,even for complex soils. Even more encouraging, beyondsimply estimating total pyroxene content, the MGM mayallow us to assess the ratio of high-Ca to low-Ca pyroxenefrom high-quality lunar soil spectra.[29] Several simplifying assumptions were made in this

study, including the combining of four types of pyroxeneinto just two classes. The analysis presented here is a firststep in understanding and unlocking the potential of MGMdeconvolution for lunar soils. Our model only captures asmall part of the physical complexity of these samples.Further work exploring this complexity may help resolvemany of the issues encountered and lead to improvements inthe predictive capabilities of the model.

[30] Acknowledgments. Thank you to Jessica Sunshine for herencouragement and helpful suggestions. This manuscript benefited greatlyfrom thorough reviews by Paul Lucey and Ed Cloutis. NASA support(NNG05-GG15G, C.M.P.; NNG05-GG41G, L.A.T.) is gratefully acknowl-edged. Support of Oak Ridge Associated University through the NASApostdoctoral program to S.K.N. is appreciated. RELAB is a multiuserfacility supported under NAG5-13609.

ReferencesAdams, J. B. (1974), Visible and near-infrared diffuse reflectance spectra ofpyroxenes as applied to remote sensing of solid objects in the solarsystem, J. Geophys. Res., 79, 4829–4836.

Adams, J. B. (1975), Interpretation of visible and near-infrared diffusereflectance spectra of pyroxenes and other rock forming minerals, inInfrared and Raman Spectroscopy of Lunar and Terrestrial Materials,edited by C. Carr, pp. 91–116, Elsevier, New York.

Burns, R. G. (1993), Mineralogical Applications of Crystal Field Theory,2nd ed., Cambridge Univ. Press, New York.

Cloutis, E. A., and M. J. Gaffey (1991), Pyroxene spectroscopy revisited:Spectral-compositional correlations and relationship to geothermometry,J. Geophys. Res., 96, 22,809–22,826.

Fischer, E. M., and C. M. Pieters (1994), Remote determination of exposuredegree and iron concentration of lunar soils using VIS-NIR spectroscopicmethods, Icarus, 111, 475–488.

Hapke, B. (2001), Space weathering from Mercury to the asteroid belt,J. Geophys. Res., 106, 10,039–10,073.

Hazen, R. M., P. M. Bell, and H. K. Mao (1978), Effects of compositionalvariation on absorption spectra of lunar pyroxenes, Proc. Lunar Planet.Sci. Conf. 9th, 2919–2934.

Hiroi, T., and C. M. Pieters (1998), Modified Gaussian deconvolutionof reflectance spectra of lunar soils, Proc. Lunar Planet. Sci. Conf.[CD-ROM], 29, 1253.

Hiroi, T., C. M. Pieters, and S. K. Noble (2000), Improved scheme ofModified Gaussian deconvolution for reflectance spectra of lunar soils,Proc. Lunar Planet. Sci. Conf. [CD-ROM], 31, 1548.

Klima, R. L., C. M. Pieters, and M. D. Dyar (2005), Pyroxene spectro-scopy: Effects of major element composition on near, mid and far-infrared spectra, Proc. Lunar Planet. Sci. Conf. [CD-ROM], 36, 1462.

Mao, A. L., P. M. Bell, and H. K. Mao (1978), Crystal field spectra of Luna24 glass, Proc. Lunar Planet. Sci. Conf. 9th, abstract, 693–695.

McCord, T. M., R. N. Clark, B. R. Hawke, L. A. McFadden, P. D. Owensby,C. M. Pieters, and J. B. Adams (1981), Moon: Near-infrared spectralreflectance, a first good look, J. Geophys. Res., 86, 10,883–10,892.

Morris, R. V. (1978), The surface exposure (maturity) of lunar soils: Someconcepts and Is/FeO compilation, Proc. Lunar Planet. Sci. Conf. 9th,2278–2297.

Mustard, J. F., and J. M. Sunshine (1995), Seeing through the dust: Martiancrustal heterogeneity and links to the SNC meteorites, Science, 267,1623–1626.

Noble, S. K., C. M. Pieters, L. A. Taylor, L. P. Keller, R. V. Morris, D. S.McKay, and S. J. Wentworth (1999), The optical properties of the finestfraction of lunar soils: Initial results and implications for weatheringprocesses, Proc. Lunar Planet. Sci. Conf. [CD-ROM], 30, 1669.

Noble, S. K., C. M. Pieters, L. A. Taylor, R. V. Morris, C. C. Allen, D. S.McKay, and L. P. Keller (2001), The optical properties of the finest

fraction of lunar soil: Implications for space weathering,Meteorit. Planet.Sci., 36, 31–42.

Pieters, C. M., and L. A. Taylor (2003), Systematic global mixing andmelting in lunar soil evolution, Geophys. Res. Lett., 30(20), 2048,doi:10.1029/2003GL018212.

Pieters, C. M., E. M. Fischer, O. Rode, and A. Basu (1993), Optical effectsof space weathering: The role of the finest fraction, J. Geophys. Res., 98,20,817–20,824.

Pieters, C. M., L. A. Taylor, D. S. McKay, S. J. Wentworth, R. V. Morris,and L. P. Keller (2000), Spectral characteristics of lunar mare soils, Proc.Lunar Planet. Sci. Conf. [CD-ROM], 31, 1835.

Pieters, C. M., L. A. Taylor, and the LSCC (2002), The perplexing role ofTiO2 in the evolution of lunar soils, Proc. Lunar Planet. Sci. Conf.[CD-ROM], 33, 1886.

Sunshine, J. M., and L. A. McFadden (1993), Reflectance spectra of theElephant Moraine A79,001 meteorite: Implications for remote sensing ofplanetary bodies, Icarus, 105, 79–91.

Sunshine, J. M., and C. M. Pieters (1993), Estimating modal abundancesfrom the spectra of natural and laboratory pyroxene mixtures using themodified Gaussian model, J. Geophys. Res., 98, 9075–9087.

Sunshine, J. M., and C. M. Pieters (1998), Determining the composition ofolivine from reflectance spectroscopy, J. Geophys. Res., 103, 13,675–13,688.

Sunshine, J. M., C. M. Pieters, and S. F. Pratt (1990), Deconvolution ofmineral absorption bands: An improved approach, J. Geophys. Res., 95,6955–6966.

Sunshine, J. M., C. M. Pieters, S. F. Pratt, and K. S. McNaron-Brown(1999), Absorption band modeling in reflectance spectra: Availabilityof the modified Gaussian model, Proc. Lunar Planet. Sci. Conf.[CD-ROM], 30, 1306.

Taylor, L. A., A. Patchen, R. V. Morris, D. Taylor, C. M. Pieters, L. P.Keller, D. S. McKay, and S. J. Wentworth (1999a), Chemical and miner-alogical characterization of the 44–20, 20–10, and

using the modified gaussian model to extract quantitative …using the modified gaussian model to...

Documents