SEARCH

SEARCH BY CITATION

Keywords:

  • Mercury;
  • Mariner 10;
  • multispectral images;
  • impact craters;
  • crater rays;
  • surface composition

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[1] We have used recalibrated Mariner 10 color image data (UV and orange) to examine spectral trends associated with crater features on the “incoming” hemisphere of Mercury. Spacecraft images of the Moon (Galileo and Clementine) and lunar sample laboratory spectra provide a framework for interpreting the Mercurian data. Two spectral parameter images were constructed. One reveals opaque mineral abundance, and the other provides a map of variations related to optical maturity and FeO content. The crater Kuiper and its rays are bright mainly because they are fresh (immature) but also because Kuiper has excavated material with a lower opaque content than the surroundings. Under the influence of space weathering, Kuiper will eventually lose its rays and evolve into a bright-floored crater like Lermontov. Lermontov and nearby smaller craters are examples of craters that are likely mature, but remain bright because of the low-opaque material exposed on their floors. In this portion of the planet, it appears that a surface layer with moderate-opaque abundance overlies low-opaque material found at depth. On the basis of estimates of crater excavation depths, we suggest that the moderate-opaque surface layer is ∼3–4 km thick. This two-layer stratigraphy is reminiscent of the lunar highland crust and may have originated in similar magma ocean processes. We also describe a feature that exposes dark material with a relatively high opaque content. This feature could be a Mercurian analog to a lunar dark-halo impact crater or perhaps a pyroclastic deposit. This implies that opaque-rich, possibly mafic, magmas were generated in the interior and reached the surface through effusive or explosive activity.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[2] Compared to the other terrestrial planets, relatively little is known about the surface composition of the planet Mercury. An inferior planet, Mercury's angular separation from the Sun as seen from Earth is never very large, making reliable telescopic observations problematic. Some early visible to near-infrared (Vis-NIR) spectral measurements of Mercury displayed an absorption near 1 μm (1000 nm) (e.g., see the review by Vilas [1988]; also Blewett et al. [1997b]) that was interpreted to indicate the presence of ferrous iron, which is responsible for prominent absorption bands in spectra of the Moon and some asteroids. However, these spectra may suffer from incomplete removal of telluric absorptions. More recent Vis-NIR spectra of Mercury are in some cases featureless [Vilas, 1985, 1988; Warell, 2003; Warell and Blewett, 2004]. Spectra of other locations exhibit weak absorptions near 1.1 μm that have been attributed to calcic clinopyroxene [Warell et al., 2006]. Warell [2002] obtained images of Mercury in six Vis-NIR wavelengths at roughly 300 km/pixel spatial resolution. His analysis of spectra and ratio images found no evidence for the “1 μm” absorption feature. This lack of clear absorption bands suggests that Mercury's silicate regolith is low in ferrous iron.

[3] Thermal emission spectroscopy in the midinfrared has also been employed to probe the planet's composition. Sprague et al. [1994] measured 7.3–13.5 μm spectra of three locations on Mercury, primarily intercrater plains. They found the wavelength location of the Christiansen emission feature to be indicative of intermediate or mafic lithology, and a resemblance between the Mercury spectra and laboratory spectra of terrestrial basalt and anorthite. Sprague and Roush [1998] reported a comparison of previous Mercury spectra with additional laboratory emission spectra (∼7–13 μm). These workers concluded that the Mercury spectra exhibit features common to feldspars and feldspar-pyroxene mixtures, including lunar highland breccias. Emery et al. [1998] collected the first spectra of Mercury from 5 to 7.5 μm. They attributed some spectral features to physical effects such as near-surface thermal gradients and grain size variations, but noted a resemblance between Mercury's 6-μm emission peak and features characteristics of feldspathic and feldspathoidal minerals and olivine. A series of 8.1–13.25 μm images of Mercury were obtained by Sprague et al. [2000], who found emissivity features indicative of the pyroxene bronzite, with the overall spectrum roughly similar to picrite, an ultramafic rock. These workers also found spectral evidence for the mineral sodalite. Cooper et al. [2001] collected 8–12.5 μm spectra of Mercury, and concluded that the transparency features were consistent with intermediate, mafic and ultramafic rock types. Sprague et al. [2002] obtained 3–13.5 μm spectra of the planet. At one location they observed a 5 μm emission peak attributed to low-Fe clinopyroxene, and a 7.9 μm emissivity maximum consistent with material of intermediate silica content. Another location exhibited three emissivity maxima, and therefore must possess a composition distinct from the other location. Thus the thermal-infrared emission results provide evidence for compositional variety from place to place on Mercury. At longer wavelengths, observations of Mercury's microwave emissions [Mitchell and de Pater, 1994] reveal that the surface is much more transparent at centimeter wavelengths than the lunar highlands, attributed to lower Fe and Ti on Mercury. Jeanloz et al. [1995] extrapolated this microwave data to lower frequencies in order to make comparisons with laboratory measurements of dielectric loss tangent, and predicted FeO + TiO2 < 6 wt.% in the Mercurian regolith.

[4] From the existing Earth-based measurements there is a general consensus that Mercury's surface consists of low-iron silicates, perhaps grossly similar to the composition of the lunar highland crust. In particular, Blewett et al. [1997b, 2002] performed a comparison between the spectral properties of Mercury and those of lunar pure anorthosites. Lunar pure anorthosite is a highland rock type consisting of >90% plagioclase feldspar and containing <2–3 wt.% FeO. Pure anorthosite on the Moon is exposed in a number of nearside basin rings, and occurs over large expanses of the northern farside [Hawke et al., 2003]. The pure anorthosites are thought to be a remnant of the Moon's original crust, formed as the magma ocean cooled and less dense plagioclase rockbergs accumulated by flotation [Hawke et al., 2003, and references therein]. Nearside pure anorthosites identified in telescopic spectra are found in massifs, the walls of large craters, or in crater central peaks, and are relatively fresh. Blewett et al. [1997b] compared the spectral slopes of lunar pure anorthosites with slopes of spectra for Mercury, and found the Mercurian spectra to be steeper (redder). This is consistent with the integral disc of the planet being dominated by mature areas, as would be expected. Blewett et al. [2002] used Clementine images of optical maturity [Lucey et al., 2000b] and FeO content to identify small farside regions that are highly mature and very low in FeO (about 3 wt.%). The spectral properties of these areas have similarities with Mercury, and suggest that Mercury is lower in FeO than even these very low iron lunar areas. Warell and Blewett [2004] performed Hapke modeling of telescopic spectra of Mercury. Their favored model was a 3:1 mixture of feldspar and enstatite, with a bulk FeO content of 1.2 wt.%.

[5] Only one spacecraft has visited Mercury. Mariner 10, which made three flybys in the mid-1970s, carried no instruments capable of making direct compositional measurements of Mercury's surface. However, Mariner 10 was equipped with twin multispectral cameras that imaged about 40% of the planet's surface in high-resolution monochrome and moderate resolution multispectral modes. Compositional inferences can be made from multispectral data, but difficulties in calibrating Mariner 10's vidicon images resulted in limited studies of the color data, focusing on qualitative comparison of selected regions [e.g., Hapke et al., 1975; Rava and Hapke, 1987]. Beginning in the early 1990s, extensive work was done to recover, reprocess, and perform new calibrations of Mariner 10 images of both the Moon [Robinson et al., 1992] and Mercury [Robinson et al., 1997; Robinson and Lucey, 1997]. Note that the recalibrated Mariner 10 images of the Moon are for a different wavelength set than the Mariner 10 Mercury images.

[6] Since the time of the Mariner 10 mission, our understanding of the factors that control the spectral characteristics of a lunar-like surface at ultraviolet to near-infrared wavelengths has progressed substantially. The conceptual framework of B. Hapke [Hapke et al., 1975; Rava and Hapke, 1987] has proven useful in interpreting lunar and Mercurian reflectance. Silicate minerals or glasses lacking Fe or Ti have absorptions only at wavelengths shorter than ∼200 nm, leading to a sharp dropoff in reflectance from the visible toward the ultraviolet (UV). Departures from this “baseline” spectrum are largely controlled by three components: (a) ferrous iron (Fe+2) in silicate minerals and glasses, (b) submicroscopic metallic iron particles (SMFe, also termed “nanophase iron,” npFe0) on and within regolith grains, and (c) opaque phases such as ilmenite (FeTiO3, the major carrier of Ti in lunar materials [Papike et al., 1991]) or large-grained metallic iron (i.e., grains larger than the wavelength of light). Small amounts of Fe+2 (and/or Ti+4) produce charge-transfer absorptions in the UV that may encroach into the visible [Wells and Hapke, 1977; Burns, 1993], but it is the Fe+2 crystal field absorption near 1000 nm that is of most value for mineralogical identification [e.g., Adams, 1974]. The SMFe is a product of space weathering, and results from two processes [Noble and Pieters, 2003]. Vapors generated by micrometeorite impact and solar wind sputtering of FeO-bearing material are deposited to form SMFe coatings and rims on soil grains. Larger SMFe blebs are produced by reduction of FeO in melts; these blebs are found throughout agglutinate particles. As a planetary regolith is exposed to space, it accumulates weathering products and thus “matures” as a function of time. The spectral effects of space weathering have been reviewed by Fischer and Pieters [1994, 1996], Pieters et al. [2000], Noble et al. [2001], and Hapke [2001]. Cintala [1992] and Noble and Pieters [2003] have specifically considered space weathering in the Mercurian environment. The accumulation of SMFe with continued surface exposure is mainly responsible for the optical changes that occur as a soil matures. These changes include reddening of the spectrum, darkening (a decrease in overall reflectance), and a reduction in the strength of absorption bands. Opaques are dark, have a shallow or “blue” spectral slope and lack strong absorption features. The presence of opaques may therefore be detected by darkening and a decrease in the normally reddish lunar spectral slope.

[7] P. G. Lucey and coworkers [Lucey et al., 1995, 1998, 2000a, 2000b; Blewett et al., 1997a] have attempted to quantify B. Hapke's qualitative three-component description of regolith reflectance using methods applicable to multispectral images, in particular those from Clementine. The Lucey method for FeO mapping examines spectral variations described by both the visible reflectance and a reflectance ratio [see also Fischer and Pieters, 1994, 1996]. The ratio, of reflectance at a near-infrared wavelength near 1000 nm to that at a wavelength in the visible continuum, is related to the strength of the Fe+2 absorption band. On a plot of the ratio versus reflectance constructed for laboratory-measured samples or a remotely observed surface, trends related to changes in the bulk FeO content are orthogonal to changes caused by variations in SMFe, i.e., the maturity of the surface. This ratio-reflectance dependence on FeO and SMFe permits a remotely measurable parameter sensitive only to ferrous iron abundance to be defined. By correlating the spectral FeO parameter value with compositional information from returned lunar samples, a calibration to wt.% FeO may be developed. A complementary parameter, OMAT, is a measure of optical maturity separated from FeO composition [Lucey et al., 2000b; Grier et al., 1999, 2001].

[8] The vidicon detector used by the Mariner 10 cameras did not allow imaging at wavelengths near 1000 nm [Benesh and Morrill, 1973], needed for characterizing the strength of the ferrous iron absorption band. Therefore the Clementine FeO and OMAT parameters, which rely on spectral reflectance measurements at 950 nm, cannot be derived from Mariner 10 data. However, Lucey et al. [1998, 2000a] developed a method for mapping opaque minerals such as coarse-grained metallic iron or ilmenite [see also Charette et al., 1974; Rava and Hapke, 1987]. The Lucey algorithm for opaque mineral mapping utilizes a measure of the spectral slope in the ultraviolet-visible portion of the spectrum along with the visible reflectance. This spectral space separates a trend caused by variation in the abundance of opaque phases from an opposing trend due to the combined effects of maturity and FeO. When the absolute abundance of FeO is small (less than a few wt.%), as is expected to be the case on Mercury, the Maturity+FeO trend is essentially measuring optical maturity alone. Thus the Mariner 10 data can be used to produce a parameter image related to opaque abundance and another parameter image related to optical maturity.

[9] Mariner 10 spectral parameter images have been used to identify probable volcanic deposits on Mercury [Robinson and Lucey, 1997; Robinson et al., 1998; Robinson and Taylor, 2001]. In this paper, we address fundamental questions about the color properties of Mercurian crater rays and other crater-related deposits, building on earlier work by Hapke et al. [1975], Rava and Hapke [1987], and Robinson and Lucey [1997].

2. UV-Visible Color Trends on the Moon and Mercury

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

2.1. Color in Relation to Composition on the Moon

[10] The modern era of lunar color studies is over thirty years old. The color difference photographs produced by E. A. Whitaker [e.g., Whitaker, 1966, 1972] dramatically illustrated that significant color variations and strong color boundaries were present on the nearside of the Moon. Charette et al. [1974] found that the ratio of spectral reflectance at 410 nm to 560 nm (“UV/Vis”) increases with the TiO2 content of mature mare basalts. The 410 nm to 560 nm ratio serves as a measure of the visible spectral slope, which is controlled by both the opaque mineral content of a surface and its state of maturity. A number of workers have applied the Charette relationship to Earth-based telescopic and Galileo spacecraft images to make TiO2 maps of the lunar maria [e.g., McCord et al., 1976; Johnson et al., 1977, 1991a; Greeley et al., 1993; Melendrez et al., 1994; Williams et al., 1995].

[11] The work of P. G. Lucey and coworkers [Lucey et al., 1998, 2000a; Blewett et al., 1997a] extended the ability to map TiO2 abundance to the lunar highlands. The Lucey method for mapping TiO2 makes use of the ultraviolet (UV) to Vis ratio coupled with the Vis reflectance to separate trends related to both composition and maturity. For Clementine, the UVVIS camera data at 415 nm (UV) and 750 nm (Vis) are used. Spectral data for lunar sample return sites extracted from Clementine images (Figure 1) reveal two major trends, one related to variations in opaque mineral abundance (predominantly ilmenite, and hence strongly correlated to TiO2 content) in the mare sites, and another trend, primarily in the highlands, owing to differences in regolith maturity and FeO content. In order to exploit these spectral trends, Lucey et al. [1998, 2000a] defined a parameter sensitive to the abundance of opaque phases as the angle θOP:

  • equation image

Here R415 and R750 represent the reflectance at 415 nm and 750 nm. The location of the origin (xo, yo) depends on the particular data set being analyzed, and was chosen to maximize the correlation of θOP to the laboratory-measured TiO2 content of lunar samples returned from the sites and stations, subject to a qualitative assessment of the degree to which maturity effects visible in images of θOP were suppressed. Increasing values of θOP are proposed to dominantly correspond to a greater abundance of opaques [Lucey et al., 1998, 2000a].

image

Figure 1. UV/Vis ratio versus Vis reflectance plot for sample return sites and stations observed by Clementine. Pure ilmenite, being dark and spectrally flat, would plot off the graph at Vis reflectance of ∼0.05 and a UV/Vis ratio of ∼1.0. Lucey et al. [2000a] defined an angular opaque mineral parameter θOP, such that stations with larger values of θOP plot closer to ilmenite and hence have a greater content of opaque phases. The trend from lower left to upper left tracks changes in opaque mineral abundance, which for the Moon is closely related to TiO2 content. The trend from lower left to upper right represents changes in maturity and ferrous iron content, with lower reflectance and lower ratio corresponding to greater maturity or higher FeO. After Lucey et al. [2000a].

Download figure to PowerPoint

[12] In the opaque analysis scheme of Lucey et al. [1998, 2000a], the distance of a point from the origin is a measure of the combined effects of maturity and FeO content:

  • equation image

where the Maturity+FeO parameter is based on the 415 nm to 750 nm ratio, distinguishing it from the original OMAT maturity parameter of Lucey et al. [2000b] and Grier et al. [1999, 2001], which is insensitive to FeO content and is based on a 950 nm to 750 nm ratio. Higher values of Maturity+FeO correspond to a lesser degree of maturity and/or to a lower FeO content. The two factors cannot be uniquely separated with the existing Mariner 10 image bands.

[13] The parameterizations represented by equations (1) and (2) are derived from a general model for the spectral effects of opaques in a silicate regolith [Hapke et al., 1975; Rava and Hapke, 1987]. The parameter θOP has a reasonably good correlation with the TiO2 content of returned lunar samples [Lucey et al., 1998, 2000a]. In some areas of the lunar surface there is disagreement between TiO2 predicted by the Lucey method applied to Clementine data and Lunar Prospector neutron spectrometer measurements [Elphic et al., 2002]. It has been noted that UV-Vis spectral trends vary among the major maria [e.g., Staid and Pieters, 2000, 2001; Gillis et al., 2003], and that ilmenite or TiO2 content can be predicted on the basis of reflectance spectra of lunar mare soil samples with only modest success [Pieters et al., 2002]. Alternate TiO2-mapping methods have been suggested [Gillis et al., 2003; Shkuratov et al., 1999, 2003]. However, in the current work we are not attempting to make accurate measurements of TiO2 abundance on Mercury, but rather seek to study first-order spectral trends that can be linked to geologically interesting properties such as opaque content and optical maturity.

2.2. Developing Mariner 10 Spectral Parameter Images for Mercury

[14] The Lucey opaque mineral analysis was derived for the Clementine 415-nm and 750-nm filters. Lunar images somewhat closer in wavelength to the Mariner 10 Mercury data were collected by the Galileo spacecraft during two Earth-Moon flybys. Galileo filters included 410 and 560 nm [Pieters et al., 1993]. The Mariner 10 images of Mercury were obtained at different wavelengths (355 nm and 575 nm), so it is prudent to check if the different band passes substantially change the outcome of the analysis. Lunar sample reflectance spectra do not contain sharp absorption features relative to the width of the band-pass filters, so major differences would not be expected. To perform this check, we obtained directional-hemispherical laboratory spectra of 36 lunar samples from the John Adams database maintained on the Reflectance Experiment Laboratory (RELAB) Web site at Brown University (http://lf314-rlds.geo.brown.edu/) (Table 1). The spectra were measured from bulk particulates in either <250 μm or <1000 μm size fractions. Ratio-reflectance diagrams were constructed for three sets of wavelengths: 415 and 750 nm (Clementine), 410 and 560 nm (Galileo), and 355 and 575 nm (Mariner 10 Mercury). The data are shown in Figure 2. Spacecraft images are reduced to bidirectional reflectance, so strictly speaking Figure 1 and Figure 2 do not depict the same reflectance quantity. However, both the directional-hemispherical and bidirectional reflectances are controlled largely by the single-scattering albedo and differ only because of geometric factors [e.g., Hapke, 1993]. Here we are interested in general spectral behavior, not in an exacting quantitative comparison, so second-order effects are not relevant to our analysis.

image

Figure 2. Color ratio versus Vis reflectance plots for lunar samples from the John Adams spectral collection listed in Table 1. Triangles, Apollo 11, 12, and 14; squares, Apollo 15; stars, Apollo 16; pluses, Apollo 17. (a) Clementine wavelengths. (b) Galileo wavelengths. (c) Mariner 10 Mercury wavelengths. The two spectral trends apparent in Figure 1 are present for all three wavelength sets.

Download figure to PowerPoint

Table 1. Lunar Sample Spectra Used for the Plots of Figure 2
Lunar SampleRELAB Spectruma
  • a

    RELAB, Reflectance Experiment Laboratory.

10084CLS720
12042CLS476
12070CLS107
14259CLS036
15021CLS739
15081CLS294
15091CLS296
15101CLS232
15211CLS087
15231CLS254
15271CLS260
15471CLS467
15501CLS408
15601CLS086
61141CLS357
61221CLS429
61241CLS438
61281CLS182
62231CLS188
63501CLS019
64421CLS252
64501CLS029
64801CLS023
65501CLS401
67481CLS378
67941CLS218
68501CLS430
69961CLS398
71501CLS011
73221CLS369
73241CLS002
73261CLS635
74241CLS027
76261CLS048
78481CLS053
79511CLS375

[15] In all three plots of Figure 2 there are two general trends: one from lower left to upper right representing compositional (FeO) and maturity variations in highland samples, and another trend from lower left to upper left corresponding mainly to changes in opaque (and hence TiO2) abundance in mare samples. The y-axis (ratio value) range for the Galileo data is less than that of the other two examples since the wavelength difference for Galileo (560–410 = 150 nm) is narrower compared to Mariner 10 (575–355 = 220) and Clementine (750–415 = 335). The relative position of some samples shifts slightly in the three wavelength sets. This is probably a consequence of the Clementine 750-nm channel being on the shoulder of the “1000-nm” ferrous iron absorption band, and perhaps the location of the filters relative to the ultraviolet-visible charge transfer absorption edges. The fact that the plots are similar gives confidence that color contrast in the visible can be represented by any pair of wavelengths with separation >∼150 nm (see Johnson et al. [1991b] for an evaluation of wavelength pairs for use in the Charette relationship). In other words, if these lunar samples were present on Mercury, then Mariner 10 images would show color trends like those revealed in Clementine and Galileo images of the Moon.

[16] In order to construct a plot similar to Figure 2 for spacecraft image data, calibrated Galileo mosaics from the second Earth-Moon encounter [Belton et al., 1994] were retrieved from the U.S. Geological Survey's Lunar Consortium Web site (http://astrogeology.usgs.gov/Projects/LunarConsortium/#Galileo; the “violet” filter is 410 nm, “green” is 560 nm). The images, which predominantly cover the nearside, were resampled to a spatial resolution of 3 km/pixel. A two-dimensional scatterplot was constructed from the 410-nm/560-nm ratio and 560-nm reflectance images, excluding portions of the images toward the terminator to avoid shadows (Figure 3a). A similar figure constructed from Clementine 415-nm/750-nm versus 750-nm images has been presented by Lucey et al. [1998]. The scatterplot displays the same trends present in the ratio-reflectance diagram for lunar samples at these wavelengths (Figure 2), with distinct mare and highland modes. Hence the same spectral characteristics found in the laboratory for millimeter-scale returned lunar samples are seen in spacecraft remote measurements of the Moon at multikilometer-scale resolution.

image

Figure 3. Two-dimensional image histograms of color ratio versus reflectance. The frequency of image points is indicated by the color code, which increases as light blue–blue–pink–red. (a) Observation of the nearside of the Moon by Galileo. The highland and mare spectral trends evident in the laboratory lunar sample ratio-reflectance plots of Figure 2 are found in the lunar image data. (b) Mariner 10 observation of Mercury's incoming hemisphere. This portion of Mercury does not display the two clear spectral trends seen on the Moon, but some elongations in the directions of the lunar trends are apparent. The plot can be viewed in terms of three spectral end-members: “a,” bright material with intermediate ratio (fresh, low in opaques); “b,” low-reflectance, low-ratio material (mature or higher in FeO, low in opaques); and “c,” low-reflectance, high-ratio material (high in opaques).

Download figure to PowerPoint

[17] For Mercury, 3-km/pixel recalibrated Mariner 10 images of the first-encounter incoming hemisphere were photometrically normalized using the Hapke parameters given by Robinson and Lucey [1997]. In order to approximately adjust the 575-nm (orange) mosaic to the correct absolute value, the image was scaled so that its average bidirectional reflectance (excluding pixels with extreme incidence or emission angles) is equal to 0.138, the average visible hemispherical albedo determined from ground-based telescopic measurements [Veverka et al., 1988]. The 355-nm mosaic (UV) was normalized to an average value of 0.6 times the orange. This factor of 0.6 was estimated by inspecting the slopes of disc-integrated UV-Vis spectra of Mercury given by Vilas [1988]. The absolute calibration of the orange or UV mosaics is not important to our analysis, since we are interested only in qualitative trends. The two-dimensional histogram of UV/orange versus orange, excluding areas of shadowing and poorer registration, is presented in Figure 3b. The distribution of image pixels in this diagram does not display the strongly bimodal structure of the lunar case. The particular shape of the distribution seen on such a plot, whether for the Moon or Mercury, is a function of the type and extent of the geologic units present in the image and the degree of mixing between spectral components or geologic end-member units. The part of Mercury seen by Mariner 10 lacks large expanses of low-reflectance, high-ratio (bluish) material analogous to the high-opaque (high-Ti) lunar maria. Still, the scatterplot is not perfectly circular, and there are suggestions of the “Maturity+FeO” and “opaque mineral” trends. The plots in Figure 3 can be interpreted as representing three end-members: “a” represents fresh, low-opaque material at high reflectance and intermediate ratio, “b” represents mature material with higher abundance of SMFe (or higher FeO) at low reflectance and low ratio, and “c” represents material richer in opaques at low reflectance and high ratio (compare Figure 1 of Robinson and Lucey [1997]). The Mariner 10 scatterplot has a roughly triangular shape, suggesting that on Mercury mixing has taken place among all three components. This contrasts with the situation on the Moon, where variation is chiefly along two lines: a mixing line between end-members “b” and “c” (representing variations in opaque (ilmenite) abundance between low-Ti and high-Ti mare basalts), and a mixing line between end-members “a” and “b” (variations in maturity and FeO content in highland material). On the Moon, mixing along the third side of the triangle (highland material with dark, high-opaque mare material, i.e., “a”–“c”) is found to a limited degree only through lateral transport at some mare-highland boundaries, and hence is not apparent in the lunar scatterplot of Figure 3a. The majority of the Mercurian surface, represented by the peak of the histogram in Figure 3b, corresponds to mature material of intermediate opaque abundance.

[18] The next step was to construct opaque-mineral abundance (θOP) and Maturity+FeO parameter maps from the Mariner 10 images using equations (1) and (2) (using 355 nm in place of 415 nm, and 575 nm in place of 750 nm). Selection of the origin (xo, yo) for defining the two parameters was done by trial and error. The best origin produces a smooth-looking opaque-parameter image in which small fresh craters are less visible than in a reflectance or UV/orange ratio image, indicating that maturity differences have been largely compensated for. A good choice of origin also results in a Maturity+FeO image where nearly all brightness variation appears to be related to maturity, giving confidence that compositional differences have been suppressed. Recall that for Mercury, the regolith is expected to be low in ferrous iron, so that the Maturity+FeO parameter should mainly be a measure of maturity. The orange reflectance, UV/orange, Maturity+FeO, and θOP, images are shown in Figure 4. The images reveal interesting variations that correlate with geomorphologic boundaries. Robinson and Lucey [1997] drew attention to a plains unit with low opaque abundance compared to the surroundings, and argued that this spectral difference, along with the embayment relations, was consistent with the presence of effusive volcanic materials. In the present work, we concentrate on impact craters and related features.

image

Figure 4. Mariner 10 mosaics of the first-encounter incoming hemisphere in orthographic projection, centered at 0° latitude, 55°W longitude, with north toward the top. (a) Orange filter (visible, 575-nm) image, linear contrast stretch 0.80–0.22. The prominent bright feature near 12°S, 30°W, is the Kuiper-Murasaki crater complex. Lermontov (152-km diameter) is the bright-floored crater near 16°N, 48°W. (b) Mariner 10 “UV/Orange” ratio image (355-nm/575-nm). Linear contrast stretch 0.53–0.67. Brighter tones indicate bluer color. (c) Spectral parameter image, Maturity+FeO, constructed from the images of Figures 4a and 4b using equation (2). Linear contrast stretch 0.16–0.30. Brighter tones correspond to fresher (less mature) material, or lower FeO. Brightness differences caused by compositional contrasts are suppressed compared to the reflectance or ratio images. (d) Opaque-phase abundance image, θOP, constructed from the images of Figures 4a and 4b using equation (1). Linear contrast stretch 0.5–1.1. Brighter tones indicate a greater abundance of dark, spectrally neutral opaque phases in the regolith material. In the θOP image, variations caused by differences in maturity (e.g., small fresh craters), seen in Figures 4a and 4b, have largely been erased.

Download figure to PowerPoint

3. General Observations

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[19] Much of the area seen in Figure 4 falls within the “H6” (Kuiper) quadrangle of the planet. The major mapped units [DeHon et al., 1981] are intercrater plains, considered to be ancient crustal material; cratered plains, rough material embaying or filling many large old craters; smooth plains, which are more lightly cratered and found within some craters and topographic low areas; and crater deposits.

[20] Mercury's surface features exhibit a variety of color relationships, revealed by the UV/orange and spectral parameter images of Figure 4. Some areas with high orange reflectance, including the near-rim deposits southeast of Kuiper and the floor of Lermontov, are dark in the ratio image, corresponding to redder color. Other features that are bright in the orange reflectance image, such as the rays of Kuiper and some small craters, are bright also in the ratio image (indicating relatively bluish color). These color-reflectance properties can be understood in terms of variations in composition (specifically, the abundance of spectrally neutral opaque phases) and the state of maturity of the regolith [Robinson and Lucey, 1997], and are discussed further in the sections below.

[21] The ratio and spectral parameter images of Figure 4 do not show any obvious latitude dependence, although there is some reason to expect latitude-dependent reflectance or color on Mercury. Hapke [1977] considered the effects of Mercury's magnetic field (assumed to be a dipole) causing the solar wind flux striking the surface in the polar regions to be much greater than at the equator. Hapke [1977] concluded that the lack of observed polar darkening indicates that solar wind saturation of regolith grains is not a major contributor to the reduction of Fe+2 to metallic Fe blebs (SMFe) in the space-weathering process [see also Hapke, 2001]. Noble and Pieters [2003] have calculated the rates at which SMFe particles produced by space weathering will coarsen in a process known as Ostwald ripening. They concluded that temperature differences on Mercury would promote coarsening near the equator relative to the poles, such that the spectrum should become redder with higher latitude. Warell [2003] reported an increase in spectral slope with latitude in his disc-resolved telescopic spectra of Mercury, suggesting that Ostwald ripening may be taking place. The photometric normalization procedure applied to the Mariner 10 images by Robinson and Lucey [1997] may have removed latitude dependence in reflectance, but this procedure would not be expected to affect color ratios. Thus the issue of latitude-dependent spectral changes on Mercury awaits clarification with new data.

4. Kuiper

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[22] The portion of Mercury shown in Figure 4 is dominated by the very bright Kuiper-Murasaki crater complex and rays. Murasaki is a 130-km diameter crater located at 12.6°S, 30.2°W. The younger crater Kuiper, 62 km in diameter and centered at 11.3°S, 31.1°W, is superimposed on Murasaki. Rays extending from Kuiper are visible for many crater diameters in the orange reflectance, UV/orange ratio, and parameter images. The Kuiper interior, near-rim and distal rays are bright in the Maturity+FeO image. This is consistent with the presence of fresh material associated with a relatively recent impact (Figure 5a), although it is possible that lower FeO abundance contributes to the brightness in the Maturity+FeO image. In the opaque parameter image (Figure 5b), the interiors of Kuiper and Murasaki are much darker than the background. This indicates that material with a lower abundance of opaques has been exposed by the impacts, as noted by Robinson and Lucey [1997]. The exterior surface within a few crater diameters is also relatively dark in the opaque abundance image and the distal rays from Kuiper are faintly visible as dark streaks. Therefore the rays of Kuiper are interpreted to be predominantly immaturity rays as defined by Hawke et al. [2004]. The rays are visible because they are less mature than the background, similar to the rays of the lunar crater Tycho (Figure 6a). Kuiper's rays will fade as space weathering proceeds, but the opaque parameter image suggests that there is some compositional component to the rays, particularly close to the crater. Therefore the Kuiper rays may retain some visibility even when they have reached full optical maturity because they may have a slightly lower opaque content than the terrain on which they were deposited. In this sense, the Kuiper rays are somewhat analogous to the rays of the lunar crater Copernicus. Copernicus's ray system is fully mature [Pieters et al., 1985], but remains bright because the ejecta contains lower-FeO highland material that is intrinsically higher in albedo than the surrounding mare basalt substrate (Figure 6b). Eventually, Kuiper's rays will disappear as they are gardened and physically mixed by impacts of all sizes. Kuiper's rays can thus be described as “combination rays” [Hawke et al., 2004] because their visibility is largely due to immaturity, with a lesser contribution from compositional factors.

image

Figure 5. The Kuiper-Murasaki crater complex. (a) Maturity+FeO image, linear stretch 0.18–0.26. The dotted circle shows the approximate rim of Murasaki, 130 km in diameter. (b) Opaque parameter image, linear stretch 0.54–1.07. Kuiper crater is centered on the northwest rim of Murasaki.

Download figure to PowerPoint

image

Figure 6. Galileo 756-nm images of prominent lunar rayed craters. (a) Tycho, 85 km in diameter. The rays and ejecta are similar in composition to the surrounding highlands, but the rays are visible because they are immature (fresher or less weathered) than the background mature highlands [Hawke et al., 2004]. The dark halo concentric to the crater rim is impact melt [Hawke et al., 1986]. (b) Copernicus, 93 km in diameter. The crater's highland-rich ejecta has reached optical maturity (and hence disappears in a lunar OMAT image), but remains visible because of the compositional contrast with the dark mare basalt substrate on which it was deposited [Hawke et al., 2004, Pieters et al., 1985].

Download figure to PowerPoint

5. Bright Crater Floors

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[23] Several workers, including Dzurisin [1977] and Schultz [1977], have discussed craters on Mercury that have bright floors or bright patches on their floors. Dzurisin [1977] examined Mariner 10 color ratio images and stated that the bright patches were bluish relative to the surrounding floor. He suggested that fumarolic, or some other type of physicochemical alteration, was responsible for the high albedo. Figure 7 shows several craters with bright interiors. The floors of Lermontov (152 km diameter, 15.2°N, 48.1°W), the southeast floor of Mistral (110 km diameter, 4.5°N, 54°W), and an ∼40 km crater on the western rim of Mistral (indicated by horizontal arrows in Figure 7) have Maturity+FeO values similar to the surroundings, indicating that the brightness of the material is not caused by immaturity. In the map by DeHon et al. [1981], the floor of Lermontov is occupied by smooth plains and bright ray material, and the interiors of Mistral and the crater on its rim are mapped as undivided plains and bright ray material. The dark appearance of Lermontov and the two features associated with Mistral in the opaque-phase image reveals that they have exposed material that has a low abundance of opaques. Similar to the situation at Kuiper, low-opaque material at depth appears to have been exposed by these impacts. We infer that Lermontov was once a bright-rayed, bright-floored crater like the present-day Kuiper, but that space weathering and impact gardening have removed Lermontov's rays. Hence Kuiper and Murasaki can be expected to eventually evolve into bright-floored craters similar to today's Lermontov. Other craters similar in size to Lermontov but lacking bright floors are found in this region, and presumably would have also excavated to the low-opaque layer. These craters are older than Lermontov [DeHon et al., 1981], and we suggest that they have lost their rays and bright floors through a combination of space weathering and lateral transport.

image

Figure 7. Images of the area near Mercury's Lermontov crater. (a) Orange reflectance image, linear stretch 0.09–0.19. (b) Maturity+FeO image, linear stretch 0.18–0.26. (c) Opaque parameter image, 0.62–1.15. Lermontov (152-km diameter), enclosed by the white square, and two smaller craters indicated by horizontal arrows have bright floors or floor patches exposing material with a lower abundance of opaques than the surroundings. The lower horizontal arrow indicates a bright patch on the southeast floor of Mistral crater, and the upper horizontal arrow indicates a crater on the western rim of Mistral. The two craters marked by vertical arrows are immature, as suggested by their bright appearance in the Maturity+FeO image. The feature enclosed by the grey box appears to be both immature (bright in Figure 7b) and high in opaques (bright in Figure 7c).

Download figure to PowerPoint

[24] A rough estimate of the thickness of the moderate-opaque surface layer that has been penetrated to reveal subjacent lower-opaque material can be made from the size of the bright-floored craters. The smallest crater in this area that clearly has a bright, low-opaque floor is the one on the western rim of Mistral. It is about 40 km in diameter, and is centered at ∼5.5°N, 55.5°W. Croft [1985] provided the following relationship between the transient crater diameter, Dtc, and the final crater diameter, D,: Dtc = (DQ0.15) × (D0.85), where DQ is the diameter of the simple-to-complex transition for craters on the planet. Pike [1988] gives the transition diameter for Mercury as 10.3 km. Therefore, for a final diameter of 40 km, Dtc = 32.6 km. Next, Cintala and Grieve [1998] suggest that the excavation depth is ∼1/6 of the transient crater diameter. Thus the excavation depth of the observed 40-km crater would be ∼5 km. This represents an upper limit on the thickness of the surface layer at this location, since the entire interior of this crater appears to have a low content of opaques (dark in Figure 7c). The presence on the floor of impact melt originating at shallower depths, or of wall slumps, would complicate this simple estimate. Larger craters such as Lermontov would easily penetrate a 5-km-thick surface layer.

[25] On the Moon, the nearside highland crustal stratigraphy has been shown to consist in many places of a noritic surface layer somewhat richer in pyroxene overlying a layer of low-pyroxene, low-FeO pure anorthosite [Hawke et al., 2003]. As described in section 1, the pure anorthosite is thought to have formed by plagioclase flotation in the lunar magma ocean. Hawke et al. [2003] hypothesize that the more mafic lunar surface layer formed by impact mixing that penetrated the still-forming flotation crust and emplaced material with the bulk magma ocean composition on top. It could be that the low-opaque material exposed by craters such as Kuiper, Murasaki, Lermontov, Mistral and the smaller crater on Mistral's rim is analogous to the lunar pure anorthosite layer [Robinson and Lucey, 1997] and formed through similar magma-ocean processes. The moderate-opaque surface layer seen on Mercury would then correspond to the lunar noritic highland crust present at the surface. The pure anorthosite on the lunar nearside is found primarily in basin rings and the central peaks of larger craters, implying an origin at depths of >5 km [Hawke et al., 2003; Tompkins and Pieters, 1999]. The thickness of the Moon's more mafic surface layer may be 5–30 km [Pieters, 1986, 1993].

[26] An alternative hypothesis for the bright floor of Lermontov was presented by Rava and Hapke [1987]. They suggested that the crater floor appears to have a lower crater density than the surroundings, and pointed out that the northeast part of Lermontov's floor contains an irregularly shaped rimless pit. Rimless depressions are indicative of endogenic modification [Schultz, 1977], and hence Rava and Hapke [1987] proposed that the bright floor deposit could have been emplaced by pyroclastic activity. If this is the case, then the pyroclastic material is bright and low in opaques, as demonstrated by the spectral parameter images. Lunar pyroclastic deposits are associated with the maria, and are dark and high in FeO (see review by Gaddis et al. [2003]). Some pyroclastic particles separated from lunar soils, such as the Apollo 15 green glasses, are relatively bright in the visible. However, the green glasses are high in FeO (∼20 wt.%, see Table A6.3 of Taylor et al. [1991]) and hence a surface deposit dominated by green glass would darken significantly under the action of space weathering. If the bright floor material found at Lermontov and the other craters is of pyroclastic origin, then nature of the material (bright and low in opaques) is in contrast to that of the dark, high-opaque material described here in section 7 and by Robinson and Lucey [1997] and Robinson and Hawke [2001] as possible Mercurian pyroclastics. While the possibility of bright, low-opaque pyroclastics on Mercury cannot be eliminated, we prefer the hypothesis that these craters have excavated into a common subsurface layer.

6. Other Small Craters

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[27] Several other small craters appear bright in the Maturity+FeO image of Figure 7b. In contrast to the bright-floored craters discussed in section 5 above, which are mature and hence similar to the mature background in Maturity+FeO value, these craters are probably young and fresh. Two such small craters are marked by vertical arrows in Figure 7. We attribute their brightness in the Maturity+FeO images to the presence of immature material, although the possibility that a lower FeO content is partly responsible cannot be ruled out. These two craters are mostly unremarkable in the opaque parameter image of Figure 7c, indicating that they expose material with an opaque-phase abundance similar to the surroundings, although the northern crater appears to have very slightly lower values. Therefore we suggest that these are examples of “normal” fresh craters. We can use these craters to further constrain the thickness of the moderate-opaque surface layer. The crater north of Lermontov is ∼25 km in diameter. Applying the calculation described in section 5 to this crater gives a depth of excavation of 3 to 4 km. This crater may have just begun to tap into the lower-opaque layer at depth. The crater marked by the vertical arrow south of Lermontov is very small, probably <5 km in diameter as estimated from Figure 6-B of Davies et al. [1978] (available at http://history.nasa.gov/SP-423/sp423.htm), and would excavate to depths of <1 km. It does not appear to expose low-opaque material from depth. Therefore the moderate-opaque surface layer in this area is roughly 3–4 km thick.

[28] A third area with interesting characteristics is shown in Figure 7. This feature lies west of Lermontov and is enclosed in a grey box in Figure 7. It is dark in the orange reflectance image but bright in both the Maturity+FeO and opaque parameter images. If the bright appearance in the Maturity+FeO image is caused by immaturity, then this represents a relatively fresh crater(s) that has deposited material that is dark (i.e., has low orange reflectance) and is high in opaques. These characteristics are analogous to those of a lunar dark-halo impact crater. Lunar dark-halo impact craters are associated with cryptomaria, where mare basalt has been buried beneath a highland-rich ejecta unit [e.g., Head et al., 1993; Blewett et al., 1995; Antonenko et al., 1995]. A later impact may then excavate down to the mare basalt and deposit it to form a dark halo around the crater. The identification of a dark-halo-like deposit at this location on Mercury provides evidence for opaque-rich material at depth. We note that the morphology of this feature is not clear in the Mariner 10 images used here or in a higher-resolution black and white photomosaic given by Davies et al. [1978, Figure 6-B], and it is difficult to estimate the diameter of the putative crater. Robinson and Lucey [1997] and Robinson and Hawke [2001] discussed possible pyroclastic deposits that have color properties similar to the feature considered here. Therefore the dark opaque-rich material may have been emplaced by pyroclastic activity rather than excavated by impact. In either case, the presence of dark, high-opaque material is suggestive of mafic volcanic material, emplaced as lava flows and then covered with less mafic crater or basin impact debris (dark-halo impact crater interpretation) or emplaced through explosive activity (pyroclastic interpretation). Therefore it would appear that despite the compressive stress regime in the Mercurian crust, some (basaltic?) partial melts generated in the planet's interior were able to migrate upward and erupt onto the surface [cf. Jeanloz et al., 1995].

[29] For completeness, we describe one more interpretation for the dark, opaque-rich feature west of Lermontov in Figure 7. It may be that the bright appearance in the Maturity+FeO image is caused not by immaturity, but instead the high values in this parameter are a result of low FeO. Under this interpretation, the material exposed here is lower in FeO than the typical surrounding terrain, but higher in opaques (from its high values in the opaque parameter image of Figure 7c). However, on the Moon, material higher in opaques is high in FeO. Furthermore, material with low FeO exhibits higher reflectance rather than lower. Hence the alternative explanation for the location in the grey box (low FeO with high opaques) is probably less likely than the interpretations in the preceding paragraph.

7. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[30] The Hapke framework [Hapke et al., 1975; Rava and Hapke, 1987] for understanding the general reflectance characteristics of silicate regoliths permits spectral parameter images to be constructed from multispectral imaging in the visible and near infrared, as demonstrated by P. G. Lucey [Lucey et al., 1995, 1998, 2000a, 2000b; Blewett et al., 1997a]. For Mercury, Mariner 10 images taken through ultraviolet and orange filters are available. An exercise in using spectra of lunar samples to plot ratio-reflectance diagrams demonstrates that the Mariner 10 wavelengths yield the same spectral-compositional trends found with the well studied Clementine and Galileo filter sets. Recalibrated Mariner 10 mosaics [Robinson and Lucey, 1997] were used to create two spectral parameter images similar to those of Robinson and Lucey [1997]: one depicting variations in the abundance of spectrally neutral opaque phases, and one controlled by differences in degree of maturity and/or FeO content. We used these spectral parameter images to examine impact crater-related features on Mercury. Because Mercury is generally believed to have low FeO in its surface materials, we interpret variations in the Maturity+FeO parameter image to be caused predominantly by maturity differences.

[31] The prominent Kuiper-Murasaki crater complex is bright because fresh material with low opaque content has been exposed. Material with low opaque content exists at depth at this location [Robinson and Lucey, 1997], in contrast to the moderate-opaque content of most of the surface in this hemisphere. The rays of Kuiper are visible mainly because they are fresher (less mature) than the mature background on which they were deposited. However, a compositional difference (lower opaques in the rays) may partially contribute to the visibility of the rays. Therefore Kuiper's rays are “combination rays” as defined by Hawke et al. [2004].

[32] The crater Lermontov is an example of a class of Mercurian craters with bright floors or bright floor patches [e.g., Dzurisin, 1977; Schultz, 1977]. The bright floor of Lermontov is mature and low in opaques. The bright floor has been attributed to fumarolic alteration along floor fractures [Dzurisin, 1977], and to pyroclastic activity [Rava and Hapke, 1987]. However, we suggest that the crater has excavated a low-opaque zone that exists at depth, similar to Kuiper-Murasaki. With continued space weathering, Kuiper's ray system will eventually fade and Kuiper will become a bright-floored crater similar to Lermontov. Several other craters in the vicinity of Lermontov also have mature floors that are bright owing to a low opaque content.

[33] Some other small craters, which are probably fresh on the basis of their bright appearance in the Maturity+FeO image, do not possess floors with a low abundance of opaques compared to the surroundings. On the basis of crater depth/diameter considerations, the moderate-opaque surface layer may be ∼3–4 km thick. The inferred two-layer structure of Mercury's crust could be analogous to the lunar highland crust described by Hawke et al. [2003], with the Mercurian low-opaque layer at depth corresponding to the lunar pure anorthosite layer and Mercury's moderate-opaque surface corresponding to the more mafic lunar highland upper crust. Mercury's stratigraphy may have originated from magma ocean processes similar to those thought to have produced the Moon's.

[34] Elsewhere in the Mariner 10 incoming hemisphere, there is evidence for fresh material with high opaque abundance. We have identified a dark feature that may be analogous to a lunar dark-halo impact crater, or possibly a pyroclastic deposit. Thus it appears that partial melting within the Mercurian interior produced relatively opaque-rich, possibly mafic magmas that were able to reach the surface through effusive or explosive activity. As described by Robinson and Lucey [1997] and Robinson and Hawke [2001], such volcanic activity has implications for the planet's volatile inventory and thermal history.

[35] We have been able to learn a substantial amount from multispectral images of Mercury in just two colors (Hapke et al. [1975], Rava and Hapke [1987], Robinson and Lucey [1997], Robinson and Hawke [2001], and the present work). The conclusions in this paper are based on the three-component model for a silicate regolith, as described in the Introduction. If Mercury happens to violate this model (at global, regional, or local scales), perhaps because of the presence of strongly colored minerals/glasses, unanticipated space weathering effects, or unusual particle properties (size, shape, scattering efficiency, etc.) then our interpretations may be incorrect. So little is known about Mercury's composition that we will not be surprised if new data returned by future spacecraft missions reveal that our current conceptual framework is too limited. NASA's MESSENGER mission [Solomon et al., 2001] carries a multispectral imaging camera (MDIS) with ten filters covering 415–1020 nm, and a UV-Vis-NIR spectrometer (MASCS) covering 115–1450 nm [Gold et al., 2001]. The high spectral resolution of MASCS will provide a detailed characterization of the surface spectral properties. This characterization can be extended by defining useful new spectral parameters, which may be mapped at higher spatial resolution using the MDIS images.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[36] This work was made possible by financial support from the NASA Planetary Geology and Geophysics Program (NASW-02007 and NNH05CD11C, D.T.B.; NAG5-13208, B.R.H.; NAG5-11516, P.G.L.; and NAG5-12018/S2, M.S.R.). The authors thank Mark Cintala for advice on the subject of crater excavation depths. Jeff Gillis-Davis helped to produce the latitude-longitude grid shown in Figure 4. Detailed comments by Sarah Noble and an anonymous reviewer led us to make substantial improvements to this paper. This is HIGP publication 1463 and SOEST contribution 6994.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information
  • Adams, J. B. (1974), Visible and near-infrared diffuse reflectance spectra of pyroxenes as applied to remote sensing of solid objects in the solar system, J. Geophys. Res., 79, 48294836.
  • Antonenko, I., J. W. Head, J. F. Mustard, and B. R. Hawke (1995), Criteria for the detection of lunar cryptomaria, Earth Moon Planets, 69, 141172.
  • Belton, M. J. S., et al. (1994), Galileo multispectral imaging of the north polar and eastern limb regions of the Moon, Science, 264, 11121115.
  • Benesh, M., and M. Morrill (1973), MVM 73 TV subsystem calibration report, document 615148, Jet Propul. Lab., Pasadena, Calif.
  • Blewett, D. T., B. R. Hawke, P. G. Lucey, G. J. Taylor, R. Jaumann, and P. D. Spudis (1995), Remote sensing and geologic studies of the Schiller-Schickard region of the Moon, J. Geophys. Res., 100, 16,95916,977.
  • Blewett, D. T., P. G. Lucey, B. R. Hawke, and B. L. Jolliff (1997a), Clementine images of the lunar sample-return stations: Refinement of FeO and TiO2 mapping techniques, J. Geophys. Res., 102, 16,31916,325.
  • Blewett, D. T., P. G. Lucey, B. R. Hawke, G. G. Ling, and M. S. Robinson (1997b), A comparison of Mercurian reflectance and spectral quantities with those of the Moon, Icarus, 129(1), 217231.
  • Blewett, D. T., B. R. Hawke, and P. G. Lucey (2002), Lunar pure anorthosite as a spectral analog for Mercury, Meteorit. Planet. Sci., 37, 12451254.
  • Burns, R. G. (1993), Origin of electronic spectra of minerals in the visible to near-infrared region, in Remote Geochemical Analysis, edited by C. M. Pieters, and P. A. J. Englert, chap. 1, pp. 129, Cambridge Univ. Press, New York.
  • Charette, M. P., T. B. McCord, C. M. Pieters, and J. B. Adams (1974), Application of remote spectral reflectance measurements to lunar geology classification and determination of titanium contents of lunar soils, J. Geophys. Res., 79, 16051613.
  • Cintala, M. J. (1992), Impact-induced thermal effects in the lunar and Mercurian regoliths, J. Geophys. Res., 97, 947973.
  • Cintala, M. J., and R. A. F. Grieve (1998), Scaling impact melting and crater dimensions: Implications for the lunar cratering record, Meteorit. Planet. Sci., 33, 889912.
  • Cooper, B., A. Potter, R. Killen, and T. Morgan (2001), Midinfrared spectra of Mercury, J. Geophys. Res., 106, 32,80332,814.
  • Croft, S. K. (1985), The scaling of complex craters, Proc. Lunar Planet. Sci. Conf. 15th, Part 2, J. Geophys. Res. 90, suppl., C828C842.
  • Davies, M. E., S. E. Dwornik, D. E. Gault, and R. G. Strom (1978), Atlas of Mercury, NASA Spec. Publ. SP, SP-423.
  • DeHon, R. A., D. H. Scott, and J. R. Underwood (1981), Geologic map of the Kuiper (H6) quadrangle of Mercury, U.S. Geol. Surv. Misc. Invest. Ser. Map, I-1233.
  • Dzurisin, D. (1977), Mercurian bright patches: Evidence for physio-chemical alteration of surface material? Geophys. Res. Lett., 4(10), 383386.
  • Elphic, R. C., D. J. Lawrence, W. C. Feldman, B. L. Barraclough, O. M. Gasnault, S. Maurice, P. G. Lucey, D. T. Blewett, and A. B. Binder (2002), Lunar Prospector neutron spectrometer constraints on TiO2, J. Geophys. Res., 107(E4), 5024, doi:10.1029/2000JE001460.
  • Emery, J. P., A. L. Sprague, F. C. Witteborn, J. E. Colwell, R. W. H. Kozlowski, and D. H. Wooden (1998), Mercury: Thermal modeling and mid-infrared (5–12 μm) observations, Icarus, 136, 104123.
  • Fischer, E. M., and C. M. Pieters (1994), Remote determination of exposure degree and iron concentration of lunar soil using VIS-NIR spectroscopic methods, Icarus, 111, 475488.
  • Fischer, E. M., and C. M. Pieters (1996), Composition and exposure age of the Apollo 16 Cayley and Descartes regions from Clementine data: Normalizing the effects of space weathering, J. Geophys. Res., 101, 22252234.
  • Gaddis, L. R., M. I. Staid, J. A. Tyburczy, B. R. Hawke, and N. E. Petro (2003), Compositional analyses of lunar pyroclastic deposits, Icarus, 161, 262280.
  • Gillis, J. J., B. L. Jolliff, and R. C. Elphic (2003), A revised algorithm for calculating TiO2 from Clementine UVVIS data: A synthesis of rock, soil, and remotely sensed TiO2 concentrations, J. Geophys. Res., 108(E2), 5009, doi:10.1029/2001JE001515.
  • Gold, R. E., et al. (2001), The MESSENGER mission to Mercury: Scientific payload, Planet. Space Sci., 49, 14671479.
  • Greeley, R., et al. (1993), Galileo imaging observations of lunar maria and related deposits, J. Geophys. Res., 98, 17,18317,205.
  • Grier, J. A., A. S. McEwen, P. G. Lucey, M. Milazzo, and R. G. Strom (1999), The optical maturity of ejecta from large rayed craters: Preliminary results and implications, paper presented at Workshop on New Views of the Moon II: Understanding the Moon Through the Integration of Diverse Datasets, Lunar and Planet. Inst., Houston, Tex.
  • Grier, J. A., A. S. McEwen, P. G. Lucey, M. Milazzo, and R. G. Strom (2001), Optical maturity of ejecta from large rayed lunar craters, J. Geophys. Res., 106, 32,84732,862.
  • Hapke, B. (1977), Interpretations of optical observations of Mercury and the Moon, Phys. Earth Planet. Inter., 15, 264274.
  • Hapke, B. (1993), Combined theory of reflectance and emittance spectroscopy, in Remote Geochemical Analysis, edited by C. M. Pieters, and P. A. J. Englert, chap. 2, pp. 3142, Cambridge Univ. Press, New York.
  • Hapke, B. (2001), Space weathering from Mercury to the asteroid belt, J. Geophys. Res., 106, 10,03910,073.
  • Hapke, B., G. E. Danielson Jr., K. Klaasen, and L. Wilson (1975), Photometric observations of Mercury from Mariner 10, J. Geophys. Res., 80, 24312443.
  • Hawke, B. R., P. G. Lucey, and J. F. Bell (1986), Spectral reflectance studies of Tycho crater: Preliminary results, Lunar Planet. Sci., XVI, 9991000.
  • Hawke, B. R., C. A. Peterson, D. T. Blewett, D. B. J. Bussey, P. G. Lucey, G. J. Taylor, and P. D. Spudis (2003), Distribution and modes of occurrence of lunar anorthosite, J. Geophys. Res., 108(E6), 5050, doi:10.1029/2002JE001890.
  • Hawke, B. R., D. T. Blewett, P. G. Lucey, G. A. Smith, J. F. Bell, B. A. Campbell, and M. S. Robinson (2004), The origin of lunar crater rays, Icarus, 170, 116.
  • Head, J. W., S. Murchie, J. F. Mustard, C. M. Pieters, G. Neukum, A. McEwen, R. Greeley, and M. J. S. Belton (1993), Lunar impact basins: New data for the western limb and farside (Orientale and South Pole-Aitken Basins) from the first Galileo flyby, J. Geophys. Res., 98, 17,14917,182.
  • Jeanloz, R., D. L. Mitchell, A. L. Sprague, and I. de Pater (1995), Evidence for a basalt-free surface on Mercury and implications for internal heat, Science, 268, 14551457.
  • Johnson, J. R., S. M. Larson, and R. B. Singer (1991a), Remote sensing of potential lunar resources: 1. Near-side compositional properties, J. Geophys. Res., 96, 18,86118,882.
  • Johnson, J. R., S. M. Larson, and R. B. Singer (1991b), A reevaluation of spectral ratios for lunar mare TiO2 mapping, Geophys. Res. Lett., 18(11), 21532156.
  • Johnson, T. V., R. S. Saunders, D. L. Matson, and J. L. Mosher (1977), A TiO2 abundance map for the northern maria, Proc. Lunar Sci. Conf. 8th, 10291036.
  • Lucey, P. G., G. J. Taylor, and E. Malaret (1995), Abundance and distribution of iron on the Moon, Science, 268, 11501153.
  • Lucey, P. G., D. T. Blewett, and B. R. Hawke (1998), Mapping the FeO and TiO2 content of the lunar surface with multispectral imaging, J. Geophys. Res., 103, 36793699.
  • Lucey, P. G., D. T. Blewett, and B. L. Jolliff (2000a), Lunar iron and titanium abundance algorithms based on final processing of Clementine UVVIS data, J. Geophys. Res., 105, 20,29720,306.
  • Lucey, P. G., D. T. Blewett, G. J. Taylor, and B. R. Hawke (2000b), Imaging of lunar surface maturity, J. Geophys. Res., 105, 20,37720,386.
  • McCord, T. B., C. Pieters, and M. A. Feierberg (1976), Multispectral mapping of the lunar surface using ground-based telescopes, Icarus, 29, 134.
  • Melendrez, D. E., J. R. Johnson, S. M. Larson, and R. B. Singer (1994), Remote sensing of potential lunar resources: 2. High spatial resolution mapping of spectral reflectance ratios and implications for near-side mare TiO2 content, J. Geophys. Res., 99, 56015619.
  • Mitchell, D. L., and I. de Pater (1994), Microwave imaging of Mercury's thermal emission at wavelengths from 0.3 to 20.5 cm, Icarus, 110, 232.
  • Noble, S. K., and C. M. Pieters (2003), Space weathering on Mercury: Implications for remote sensing (in Russian), Astron. Vestnik, 37, 3439, (Sol. Syst. Res., Engl. Transl., 37, 31–35.).
  • 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, 3142.
  • Papike, J., L. Taylor, and S. Simon (1991), Lunar minerals, in Lunar Sourcebook, edited by G. H. Heiken, D. T. Vaniman, and B. M. French, chap. 5, pp. 121181, Cambridge Univ. Press, New York.
  • Pieters, C. M. (1986), Composition of the lunar highland crust from near-infrared spectroscopy, Rev. Geophys., 24(3), 557578.
  • Pieters, C. M. (1993), Compositional diversity and stratigraphy of the lunar crust derived from reflectance spectroscopy, in Remote Geochemical Analysis, edited by C. M. Pieters, and P. A. J. Englert, chap. 14, pp. 309339, Cambridge Univ. Press, New York.
  • Pieters, C. M., J. B. Adams, P. J. Mouginis-Mark, S. H. Zisk, M. O. Smith, J. W. Head, and T. B. McCord (1985), The nature of crater rays: The Copernicus example, J. Geophys. Res., 90, 12,39312,413.
  • Pieters, C. M., et al. (1993), Crustal diversity of the Moon: Compositional analyses of Galileo SSI data, J. Geophys. Res., 98, 17,12717,148.
  • Pieters, C. M., L. A. Taylor, S. K. Noble, L. P. Keller, B. Hapke, R. V. Morris, C. C. Allen, D. S. McKay, and S. Wentworth (2000), Space weathering on airless bodies: Resolving a mystery with lunar samples, Meteorit. Planet. Sci., 35, 11011107.
  • Pieters, C. M., D. G. Stankevich, Y. G. Shkuratov, and L. A. Taylor (2002), Statistical analysis of the links among lunar mare soil mineralogy, chemistry, and reflectance spectra, Icarus, 155, 285298, doi:10.1006/icar.2001.6749.
  • Pike, R. J. (1988), Geomorphology of impact craters on Mercury, in Mercury, edited by F. Vilas, C. R. Chapman, and M. S. Matthews, pp. 165273, Univ. of Ariz. Press, Tucson.
  • Rava, B., and B. Hapke (1987), An analysis of the Mariner 10 color ratio map of Mercury, Icarus, 71, 397429.
  • Robinson, M. S., and B. R. Hawke (2001), Low albedo, blue, and opaque-rich spectral anomalies in the Mercurian crust, paper presented at Workshop on Mercury: Space Environment, Surface, and Interior, Lunar and Planet. Inst., Houston, Tex.
  • Robinson, M. S., and P. G. Lucey (1997), Recalibrated Mariner 10 color mosaics: Implications for Mercurian volcanism, Science, 275, 197200.
  • Robinson, M. S., and G. J. Taylor (2001), Ferrous oxide in Mercury's crust and mantle, Meteorit. Planet. Sci., 36, 841847.
  • Robinson, M. S., B. R. Hawke, P. G. Lucey, and G. A. Smith (1992), Mariner 10 multispectral images of the eastern limb and farside of the Moon, J. Geophys. Res., 97, 18,26518,274.
  • Robinson, M. S., B. R. Hawke, P. G. Lucey, G. J. Taylor, and P. D. Spudis (1997), The color of Mercury, Lunar Planet. Sci. [CD-ROM], XXVIII, abstract 1689.
  • Robinson, M. S., B. R. Hawke, P. G. Lucey, G. J. Taylor, and P. D. Spudis (1998), Compositional heterogeneity of Mercury's crust, Lunar Planet. Sci. [CD-ROM], XXIX, abstract 1860.
  • Schultz, P. H. (1977), Endogenic modification of impact craters on Mercury, Phys. Earth Planet. Inter., 15, 202219.
  • Shkuratov, Y. G., V. G. Kaydash, and N. V. Opanasenko (1999), Iron and titanium abundance and maturity degree distribution on the lunar nearside, Icarus, 137, 222234.
  • Shkuratov, Y. G., et al. (2003), Composition of the lunar surface as will be seen from SMART-1: A simulation using Clementine data, J. Geophys. Res., 108(E4), 5020, doi:10.1029/2002JE001971.
  • Solomon, S. C., et al. (2001), The MESSENGER mission to Mercury: Scientific objectives and implementation, Planet. Space Sci., 49, 14451465.
  • Sprague, A. L., and T. L. Roush (1998), Comparison of laboratory emission spectra and Mercury telescopic data, Icarus, 133, 174183.
  • Sprague, A. L., R. W. H. Kozlowski, F. C. Witteborn, D. P. Cruikshank, and D. H. Wooden (1994), Mercury: Evidence for anorthosite and basalt from mid-infrared (7.3–13.5 μm) spectroscopy, Icarus, 109, 156167.
  • Sprague, A. L., L. K. Deutsch, J. Hora, G. G. Fazio, B. Ludwig, J. Emery, and W. F. Hoffmann (2000), Mid-infrared (8.1–12.5 μm) imaging of Mercury, Icarus, 147, 421432.
  • Sprague, A. L., J. P. Emery, K. L. Donaldson, R. W. Russell, D. K. Lynch, and A. L. Mazuk (2002), Mercury: Mid-infrared (3–135 μm) observations show heterogeneous composition, presence of intermediate and basic soil types, and pyroxene, Meteorit. Planet. Sci., 37, 12551268.
  • Staid, M. I., and C. M. Pieters (2000), Integrated spectral analysis of mare soils and craters: Applications to eastern nearside basalts, Icarus, 145, 122139.
  • Staid, M. I., and C. M. Pieters (2001), Mineralogy of the last lunar basalts: Results from Clementine, J. Geophys. Res., 106(E11), 27,88727,900.
  • Taylor, G. J., P. Warren, G. Ryder, J. Delano, C. Pieters, and G. Lofgren (1991), Lunar rocks, in Lunar Sourcebook, edited by G. H. Heiken, D. T. Vaniman, and B. M. French, chap. 6, pp. 183284, Cambridge Univ. Press, New York.
  • Tompkins, S., and C. M. Pieters (1999), Mineralogy of the lunar crust: Results from Clementine, Meteorit. Planet. Sci., 34, 2541.
  • Veverka, J., P. Helfenstein, B. Hapke, and J. D. Goguen (1988), Photometry and polarimetry of Mercury, in Mercury, edited by F. Vilas et al., pp. 3758, Univ. of Ariz. Press, Tucson.
  • Vilas, F. (1985), Mercury: Absence of crystalline Fe2+ in the regolith, Icarus, 64, 133138.
  • Vilas, F. (1988), Surface composition of Mercury from reflectance spectrophotometry, in Mercury, edited by F. Vilas et al., pp. 5976, Univ. of Ariz. Press, Tucson.
  • Warell, J. (2002), Properties of the Hermean regolith: II. Disk-resolved multicolor photometry and color variations of the “unknown” hemisphere, Icarus, 156, 303317.
  • Warell, J. (2003), Properties of the Hermean regolith: III. Disk-resolved vis-NIR reflectance spectra and implications for the abundance of iron, Icarus, 161, 199222.
  • Warell, J., and D. T. Blewett (2004), Properties of the Hermean regolith: V. New optical reflectance spectra, comparison with lunar anorthosites, and mineralogical modelling, Icarus, 168, 257276.
  • Warell, J., A. L. Sprague, J. P. Emery, R. W. H. Kozlowski, and A. Long (2006), The 0.7–5.3 μm spectra of Mercury and the Moon: Evidence for high-Ca pyroxene on Mercury, Icarus, 180, 281291.
  • Wells, E., and B. Hapke (1977), Lunar soil: Iron and titanium bands in the glass fraction, Science, 195, 977979.
  • Whitaker, E. A. (1966), The surface of the Moon, in The Nature of the Lunar Surface: Proceedings of the 1965 IAU-NASA Symposium, edited by W. N. Hess et al., pp. 7998, Johns Hopkins Univ. Press, Baltimore, Md.
  • Whitaker, E. A. (1972), Lunar color boundaries and their relationship to topographic features: A preliminary survey, Moon, 4, 348355.
  • Williams, D. A., R. Greeley, G. Neukum, R. Wagner, and S. D. Kadel (1995) Multispectral studies of western limb and farside maria from Galileo Earth-Moon Encounter 1, J. Geophys. Res., 100, 23,29123,299.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. UV-Visible Color Trends on the Moon and Mercury
  5. 3. General Observations
  6. 4. Kuiper
  7. 5. Bright Crater Floors
  8. 6. Other Small Craters
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information
FilenameFormatSizeDescription
jgre2197-sup-0001-t01.txtplain text document1KTab-delimited Table 1.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.