Compositional variations in the Vestan Rheasilvia basin

Compositional variations in the Vestan Rheasilvia basin

Icarus 259 (2015) 194–202 Contents lists available at ScienceDirect Icarus journal homepage: Compositional variation...

8MB Sizes 10 Downloads 42 Views

Icarus 259 (2015) 194–202

Contents lists available at ScienceDirect

Icarus journal homepage:

Compositional variations in the Vestan Rheasilvia basin E. Ammannito a,b,⇑, M.C. De Sanctis b, J.-Ph. Combe c, A. Frigeri b, R. Jaumann d, A. Longobardo b, H.Y. McSween e, E. Palomba b, F. Tosi b, C.A. Raymond f, C.T. Russell a, the Dawn Science Team a

University of California Los Angeles, Earth Planetary and Space Sciences, Los Angeles, CA 90095, USA INAF, Istituto di Astrofisica e Planetologia Spaziale, Area di Ricerca di Tor Vergata, Roma, Italy c Bear Fight Institute, Winthrop, WA 98862, USA d DLR, Institute of Planetary Research, Berlin, Germany e Planetary Geoscience Institute and Department of Earth & Planetary Sciences, University of Tennessee, Knoxville 37996, TN, USA f Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA b

a r t i c l e

i n f o

Article history: Received 20 January 2015 Revised 26 April 2015 Accepted 13 May 2015 Available online 28 May 2015 Keywords: Mineralogy Asteroids, composition Asteroid Vesta Asteroids, surfaces

a b s t r a c t We present and describe the maps of spectral parameters such as pyroxene band centers and depths, reflectance at 1.4 lm and 2.8 lm band depth in the Rheasilvia quadrangle. We found a broad anti-correlation between pyroxene band centers and depths while the reflectance is not correlated with the pyroxene spectral parameters. In addition, we found that the Rheasilvia quadrangle is free of OH absorption signatures. We also derived lithological maps with improvements in the spatial resolution with respect to previous lithological maps of the same region. We confirm that the central mound is dominated by eucritic/howarditic pyroxene while diogenitic lithology has been found mainly in a region delineated by Tarpeia, Severina and Mariamne craters. We found small scale variations in the composition of pyroxene. These variations identify lithological units that extend for tens of km, although small units of less than 1 km have also been found. We consider this fact as an indication of a high level of compositional heterogeneity within the Vestan crust. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Asteroid 4/Vesta has been intensively studied since its discovery in 1807 by Heinrich Wilhelm Matthias Olbers. Telescopic observations (ground-based and Hubble Space Telescope (HST)), revealed that this asteroid exhibits pyroxene absorption bands similar to the howardite–eucrite–diogenite (HED) meteorites (McCord et al., 1970). It was also revealed that there are longitudinal variations in albedo and surficial composition (Gaffey, 1997; Binzel et al., 1997; Vernazza et al., 2005; Carry et al., 2010; Li et al., 2010; Reddy et al., 2010). In addition, the existence of a large basin at the Vestan South Pole was identified (Thomas et al., 1997). The discovery of a family of small asteroids dynamically and compositionally linked to Vesta – the Vestoids (Binzel and Xu, 1993) – provided the missing link in the interpretation of the Vesta–HED connection: Vestoids are likely samples of the missing mass ejected from Vestan South Polar basin during the cratering event, while the HEDs are small fragments of Vestoids that have reached the inner Solar System and the Earth (Marzari et al., 1996). In this ⇑ Corresponding author at: University of California Los Angeles, Earth Planetary and Space Sciences, 595 Charles Young Drive East, Los Angeles, CA 90095, USA. E-mail address: [email protected] (E. Ammannito). 0019-1035/Ó 2015 Elsevier Inc. All rights reserved.

scenario, the HEDs are considered to be excavated samples of a stratified internal structure (McSween et al., 2011). Petrologic and geochemical studies of HEDs, and thermal evolution models, have generated a hypothesis to explain the internal structure of Vesta, called the magma ocean model (Righter and Drake, 1997; Ruzicka et al., 1997; Mandler and Elkins-Tanton, 2013; Toplis et al., 2013). According to this picture, a body with an initial chondritic composition experienced widespread or global melting induced by the heat produced by decay of 26Al and possibly other minor short-lived radionuclides. Crystallization of this magma ocean generated a layered structure. Metals (Fe, Ni) concentrated in the central core, while olivine and orthopyroxene formed a mantle of harzburgite and orthopyroxenite (diogenite). If melting were incomplete, an olivine-rich (dunite) lower mantle may have formed (Neumann et al., 2014) below harzburgite and orthopyroxenite. Continuing crystallization formed a crust of gabbro and basalt (sampled respectively by cumulate and basaltic eucrites). The most external layer is composed of regolith breccias (sampled by polymict eucrite and howardite). Although the magma ocean model remains popular, another scenario envisions continuous melt extraction from the mantle, so that a global scale magma ocean never formed (Wilson and Keil, 2012).

E. Ammannito et al. / Icarus 259 (2015) 194–202

The Dawn spacecraft accomplished a major step forward in understanding Vesta. From July 2011 to August 2012, the spacecraft orbited Vesta (Russell et al., 2013), and during this time the Visible InfraRed Mapping Spectrometer (VIR) acquired spectra of its surface from 0.2 to 5 lm (De Sanctis et al., 2011). The instrument operated during Survey (VSS), High Altitude Mapping (VSH) and Low Altitude Mapping (VSL) orbits as well as during Approach and Departure phases, providing an almost global coverage of the surface. Data from VSL have the highest spatial resolution. 70 m/px is the nominal resolution in this orbit, in comparison with 170 m/px and 700 m/px that are the typical resolutions during the VSH and VSS orbits, respectively. While VSL coverage is limited to less than 1% of the total surface, that dataset provides a detailed view of some localized areas. The Dawn spacecraft confirmed the existence of an approximately circular basin in the South Pole, named Rheasilvia, superimposed on an older basin called Veneneia (Russell et al., 2012; Jaumann et al., 2012; Marchi et al., 2012; Schenk et al., 2012). In addition, spectroscopic measurements confirmed that the spectrum of Vesta is dominated by pyroxene absorption bands with variations of band center position, band depth and other band parameters at both large and small scales (De Sanctis et al., 2012a, 2013; Ammannito et al., 2013a; McSween et al., 2013). In particular, there is a strong indication that the Rheasilvia basin has its own spectral characteristics: the pyroxene absorption bands are deeper and wider, and center positions are shifted toward shorter wavelengths with respect to the average values. Rheasilvia’s central mound has a relatively low spectral diversity (McSween et al., 2013). These spectral behaviors indicate the presence of Mg-pyroxene-rich terrains in Rheasilvia, an occurrence confirmed by the Gamma-Ray and Neutron Detector (GRaND) (Prettyman et al., 2012) and the Framing Camera (FC) color data (Reddy et al., 2012), the other two instruments on the Dawn spacecraft. The Rheasilvia basin has attracted considerable interest because its diameter (500 ± 20 km; Jaumann et al., 2012) with respect to the size of Vesta (rotational ellipsoid with semi-axes of 280.9 km and 226.2 km with the origin at the center of mass of the body; Ermakov et al., 2014) is such that in a magma ocean scenario, ultramafic rocks from the upper mantle should have been exposed (Ivanov and Melosh, 2013; Jutzi et al., 2013). Therefore, this particular location has been considered as a potential window into the internal structure (Pieters et al., 2011). Interestingly, Dawn has found no evidence so far of the presence of olivine within the Rheasilvia basin (Ammannito et al., 2013b; Ruesch et al., 2014; Clenet et al., 2014). It is worth noting that the identification of olivine, when combined with pyroxene, by reflectance spectroscopy in the VIS/NIR range has always been controversial (Beck et al., 2013). Both minerals have a diagnostic feature at about 1 lm, but the crystal structure of olivine has a lower absorption coefficient relative to that of pyroxene. The direct consequence is that it is difficult to detect olivine in concentrations <25% in the presence of abundant orthopyroxene (Beck et al., 2013). However, also the locations with pure diogenitic-like terrains, thought to be lower crust material, are concentrated in only a few spots, and do not correspond to the uplifted central mound, which is instead dominated by howarditic and eucritic compositions (McSween et al., 2013; Ammannito et al., 2013a). The lack of both olivine detections and widespread ultramafic minerals within Rheasilvia has implications for the Vestan internal structure.

2. Maps of the Rheasilvia basin Here we describe the characteristics of the distribution of spectral parameters in the Rheasilvia quadrangle, which includes all latitudes below 65°S (Roatsch et al., 2012). Details of the map


production process, as well as the computation of spectral parameters, are provided in Frigeri et al. (2015) and Combe at al. (2015). In this paper, all the maps and coordinates are given using the Claudia Prime Meridian Reference System (Russell et al., 2012; Reddy et al., 2013). Products in this paper can be converted in the IAU Reference System by adding 150° from longitudes values (Li, 2013b). The spectral position of the centers of the two major absorption bands in the VIR sensitivity range, at 1 lm (BI) and 2 lm (BII), are especially useful for identifying variations in pyroxene chemical composition (Burns, 1993). We computed the pyroxene band centers following the method described by Ammannito et al. (2013a). The values obtained are shown in the maps in Fig. 1. In these maps, some regions have both BI and BII centers shifted to shorter wavelengths (blue in the map) with respect to the average values on Vesta which are 0.926 lm for BI center and 1.971 lm for BII center. This implies the presence in these locations of pyroxenes with higher Mg versus Fe concentration (Adams, 1974; Klima et al., 2007). For both BI and BII, there is a general trend that longitudes between 0°E and 100°E have shorter spectral position of the centers of the two pyroxene bands. In addition, there are localized regions with particularly short values of band centers. These regions are associated with some of the lowest topography on the Vestan surface (Jaumann et al., 2012). These are presumably outcrops of the deep crust or upper mantle exposed at the base of Rheasilvia’s central uplift (Reddy et al., 2012). Another characteristic of BI and BII centers is that their values are correlated, meaning that spectra with particularly high or low values of BI center position also have a high or low BII center position. This is an indication that the mineralogy in this region is dominated by a pyroxene chemical composition with a minor or absent influence of other components such as carbonaceous chondrites, which tend to increase the values of the BII center (De Sanctis et al., 2015). Projected maps of the values of pyroxene band depth are shown in Fig. 2. The values in the maps have been corrected for photometric effects using a method described by Longobardo et al. (2014). In analogy with the band centers, the band depths of BI and BII have the same trend, meaning that low values of BI depth correspond to low values of BII depth, and the same for high values. However, the distributions of band centers and depths do not have a clear correlation. In general, the region below 65°S has deeper bands when compared with the average Vestan surface. This distribution seems to correlate with the pyroxene band centers that, as noticed earlier, have shorter values in the same latitude range. However, this broad correlation is not supported when we analyze the distribution at smaller scale (tens of km and below). In particular, there is a region in the middle of the map with deeper bands that roughly corresponds to the central mound (topographic high). This may be a consequence of additional photometric effects not completely taken into account in the correction applied or of a different physical state, such as grain size distribution, in the mound. Fig. 3 (left) shows a map of the reflectance generated at 1.4 lm, as described in Combe et al. (2015). The photometrically-corrected 1.4 lm reflectance provides a contextual view of the surface optical properties. Although in the map there are some residual artifacts, mainly due to poor illumination conditions in the polar areas, it is possible to identify a hemispherical dichotomy with the region roughly from 135°E to 315°E longitude, showing generally lower values in comparison to the opposite region. This dichotomy does not correlate with either pyroxene band centers or depths, indicating that it is not associated with pyroxene chemical composition. This is particularly clear when we notice that the highest values in reflectance are in the region of longitudes between 335°E and 20°E, while minimum values of band centers are between 0°E and 100°E, and the depths increase closer to the South Pole.


E. Ammannito et al. / Icarus 259 (2015) 194–202

Fig. 1. Band centers in stereographic projection centered on the South Pole. The units of the color bars on the top corners are microns. Left (a) BI centers, right (c) BII centers. The projections are based on the Claudia prime meridian reference system. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Band depths in stereographic projection centered on the South Pole. The values in the maps are expressed in relative units (depths computed on continuum removed spectra (Longobardo et al., 2014)). Left (a) BI depths, right (c) BII depths. The projections are based on the Claudia prime meridian reference system.

Fig. 3 (right) shows a map of the depth of the 2.8 lm OH absorption band generated as described in Combe et al. (2015). As is clear from the range of values (0.015–0.025 in comparison with global values, which range from 0.015 to 0.060) of this parameter in the map, the entire southern region is poor in OH. This result is in agreement with early findings on the distribution across Vesta’s surface of either OH (De Sanctis et al., 2012b) and H (Prettyman et al., 2012). Therefore we confirm that the Rheasilvia impact event may have erased the OH that was evenly delivered on Vesta by infall of carbonaceous chondrites.

When the photometrically-corrected reflectance at 1.4 lm is combined with the 2.8 lm band depth (Fig. 4), it shows that dark materials on Vesta are generally hydrated (McCord et al., 2012; De Sanctis et al., 2012a,b), from the contamination by carbonaceous chondrite meteorites. Although Rheasilvia is bright and anhydrous compared to the rest of Vesta, small variations occur. In the two-dimension scatter plot (Fig. 4a) the two-lobed data cloud reveals two slightly distinct composition units that correspond to the before mentioned hemispherical reflectance dichotomy. Both units are characterized by a diffuse anti-correlation of

E. Ammannito et al. / Icarus 259 (2015) 194–202


Fig. 3. Reflectance at 1.4 lm (left) and depth of the 2.8 lm OH absorption band depth (right). The values range in the map is from 0.278 to 0.375 for the reflectance and from 0.01 to 0.03 for 2.8 lm band depth.

Fig. 4. Surface reflectance and hydrated materials for the Rheasilvia quadrangle region. (a) Two-dimensional scatter plot of the photometrically-corrected 1.4 lm reflectance vs. 2.8 lm absorption band depth. The color scale in the background and the level of shade are the key plot for the map (b) on the right. (b) Color composite polar map of 1.4 lm reflectance vs. 2.8 lm band depth, as defined by the color scale on the left (a). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

reflectance and 2.8 lm absorption band depth. The first one has relatively low reflectance (<0.32) with some variations in the 2.8 lm band depth, and it is color-coded in blue and magenta on the map. The other one has higher reflectance (>0.32) and smaller variations in the 2.8 lm band depth and is color-coded in yellow and green on the map. The southwestern wall of Tarpeia (70°S, 30°E) is especially bright and anhydrous, probably because the impact that formed the crater further depleted the surface in OH.

3. Variations in pyroxene composition To better describe the chemical composition of pyroxene, we developed a method to measure the variation in Mg and Fe content using spectral parameters associated only with BII rather than the combination of BI and BII. In this way, we can consider only the IR section of each spectrum. There are several advantages to this approach. The first is that in the 1 lm region, there are spectral

features of other minerals like olivine and plagioclase that are likely present on Vesta, whereas the 2 lm region is affected only by pyroxene composition. Olivine and plagioclase are considered minor spectral components in the VIS/NIR spectral range because their absorption features are masked by the pyroxene, since this mineral has a stronger absorption coefficient. However, they can have small effects that may alter the relationship between Mg/Fe content in pyroxene and the spectral positions of BI and BII. In addition, with this approach we do not use the 1 lm region of the spectrum which can have some effects that are not completely accounted for in the Instrument Responsivity that was applied to calibrate the dataset used in this work (Ammannito et al., 2014). Another advantage is the fact that our method uses only the IR detector, one of the two detectors on the spectrometer. In this way we do not need to correct for the different resolutions of the two channels, as described in Frigeri et al. (2015). The method described here relies only on BII and assumes that for each spectrum, this pyroxene band is the result of the


E. Ammannito et al. / Icarus 259 (2015) 194–202

combination of two components which we refer to as short and long wavelength component (Fig. 5, left). We have modeled the region of the spectrum between 1.5 and 2.5 lm with a custom version of the Modified Gaussian Model (MGM, Sunshine et al., 1990; Sunshine and Pieters, 1993). In the model, in the space defined by the natural logarithm of reflectance and energy, we simulated the continuum with a straight-line function while the two components of the pyroxene band have been simulated with two different Gaussian-like functions. This approach has been first tested on HED spectra and then applied to VIR data (see Supplementary material). Fig. 5 (right) is a scatter plot of the band center of the long wavelength component versus the band center of the short wavelength component. The plot shows the position of the points computed from HED and VIR spectra. Both populations plot along roughly the same line. From HED values we can see that diogenites (diamonds, red box) plot on the lower end, eucrites (triangles, blue box) plot on the higher end, and howardites (squares, green box) plot in the intermediate region. VIR points (purple cloud) nicely over-plot the trend defined by the HEDs. We used this diagram to associate a lithology (diogenite, howardite or eucrite) to each VIR spectrum, considering in which box the value computed from that spectrum is located. This is the same association process used to produce previous lithologic maps based on VIR spectra (McSween et al., 2013; Ammannito et al., 2013a), but starting with different parameters (i.e. spectral position of the two pyroxene components). It is worth to notice that the combination of parameters we use in this work has a distribution of the VIR derived points with less dispersion and better overlap with the meteorites in comparison with the similar plot in McSween et al. (2013). It is a consequence of the fact that the pyroxene BII is less subject to instrumental artifacts and potential presence of other compositional phases in addition to pyroxene. Fig. 6 (left) is the lithologic map of the Rheasilvia quadrangle, produced by using the results of the MGM applied to BII. This kind of map provides a way to link the pyroxene chemical composition measured by VIR with the layer from which the mineral is coming, assuming a magma ocean-like internal structure. Without entering into the details of compositional variability among each class of HED meteorites and focusing on a scheme based on quantities that can be measured by VIR, here we consider diogenite as Mg-rich pyroxene derived from the most external layer of the mantle, eucrite as Fe-rich pyroxene from crust, and howardite as a mixture of Mg-rich and Fe-rich pyroxene that may indicate intermediate depth or surficial regolith. According to these definitions, the Rheasilvia basin has, on average, a howarditic composition (green in the map) with an enhancement of eucrite on the left side (blue in the map). Diogenite-rich regions (red in the map) are located near the South Pole on the bottom right and within the larger impact craters Tarpeia, Severina and Mariamne. We note that the Rheasilvia central mound does not have Mg-rich pyroxene, but is mainly composed by howardite and eucrite. Fig. 6 (right) is a lithologic map of the southern hemisphere produced by comparing the center positions of BI and BII on Vesta with the positions in HED meteorites (McSween et al., 2013). Both maps in Fig. 6 connect VIR spectra with HED-defined lithologies, the difference being the analysis process used; the map on the left uses only BII, whereas the map on the right uses a combination of BI and BII. The color indices in the two lithologic maps are the same so that the results of the two different methods can be directly compared. While the maps show a general agreement in the trend of higher diogenite/howardite ratio on the right side with the left with being more howardite/eucrite-rich, there are some relevant differences, especially at a small scale. In particular, we

note that the BI/BII-based map (on the right) is dominated by an intermediate composition (i.e. cyan and yellow), while the BII-MGM method has the tendency to enhance the extreme compositions (blue and red). This can be a consequence either of the difficulty in determining the center of BI, since it is spectrally close to the junction of the two VIR detectors, or of the shorter variability range of pyroxene chemical composition with respect to BII. The BI/BII map identifies the general trends, while the BII-MGM map provides a better direct comparison with the HED meteorites.

4. Specific regions of interest The portion of the Rheasilvia basin below 65°S generally has a howarditic composition, with a higher concentration of diogenitic versus eucritic material in the region between 45° and 225°E-lon. However, there are several locations with lithological characteristics different from the surrounding regions. Combining the analysis of the maps presented in previous paragraphs with the geological map of the region (Williams et al., 2014), some specific locations have been identified for further investigation. The major craters in the mapped area are located within an undifferentiated crater material unit, as defined by Williams et al. (2014). Among them are Tarpeia, Severina and Mariamne, which have a considerable variability in pyroxene composition. Alypia crater is an exception, since it is entirely covered by bright material, according to the unit in the geological map of Williams et al. (2014). Another feature of interest is the Parentatio Rupes, a mass-wasting unit near the Rheasilvia mound, because of its steep slope and vertical exposure material. Tarpeia crater, with a diameter of 40 km, is centered at 70°S and 29°E. According to the geological map, this crater and its surroundings are entirely within a single unit likely made of relatively fresh and unmodified impact crater deposits (Williams et al., 2014; Yingst et al., 2014). In the same geological unit, another crater, Paculla (diameter of 22 km centered at 64°S and 5°E) occurs. Despite the geological homogeneity, in the Tarpeia/Paculla region there are lithological heterogeneities, in particular within Tarpeia crater, as shown in Fig. 7. Most of the left side of this crater is composed of eucritic like material (blue1 in Fig. 7, left), while moving toward the right side, there is an increasing presence of diogenitic material, with the highest concentration found close to the crest of the rim (red in Fig. 7, left). On the right side of Fig. 6 are the average spectra of each lithological class identified within Tarpeia. The continuum of the spectra has been removed by dividing two straight lines: the first one from 0.67 lm to the maximum of each spectrum between 1.2 lm and 1.5 lm, and a second one from the same maximum to 2.5 lm. The plot shows that in this particular location band depths are anti-correlated with band centers, supporting the idea that Fe-rich pyroxenes tend to have less pronounced absorption bands (De Sanctis et al., 2012a,b). The Fe-rich pyroxene unit identified within Tarpeia extends beyond the crest of the right side rim to the internal wall of Paculla crater. Paculla is of interest because its internal rim is amongst the brightest locations across the Vestan surface (Li et al., 2013; Schröder et al., 2013). Such bright locations correspond to spurs formed by large blocks or crustal segments protruding from the crustal rim (Otto et al., 2013). However, the variation in the values of reflectance within the crater (from 0.30 to 0.37, in comparison to 0.27 to 0.37 in the rest of the quadrangle) are not correlated with a substantial variability in the composition of pyroxene, since most of the internal part of the crater is made of a single eucritic unit. On the contrary, Zambon et al. (2014) noted that the bright locations in the 1 For interpretation of color in Figs. 6–8, the reader is referred to the web version of this article.

E. Ammannito et al. / Icarus 259 (2015) 194–202


Fig. 5. MGM on the 2 lm pyroxene band. On the left an example: black points are VIR acquired data, in red the modeled spectrum and in purple and orange the two components. On the right the scatter plot on the centers of the two components when the MGM was applied to HED and VIR data. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Lithological map in stereographic projection centered on the South Pole. On the left lithology classes defined with the MGM applied to the pyroxene 2 lm band. On the right the classes are defined with the BI BII center position. Boxes in the map on the left define regions described in the next paragraph.

Fig. 7. On the left, lithological map of Tarpeia, Paculla and Alypia craters on top of a FC clear filter mosaic. On the right, average spectra of the lithological units within Tarpeia crater.


E. Ammannito et al. / Icarus 259 (2015) 194–202

rim of Paculla have pyroxene band shallower than the surroundings; they interpreted this characteristic as the possible presence of plagioclase. Alypia is a crater with a diameter of 15 km centered at 70°S and 350°E at the beginning of the slope of the Rheasilvia central mound, a few kilometers away from Tarpeia and Paculla (Fig. 7, left). This crater and its ejecta are located within the only bright crater material unit of the region, according to the geological map (Williams et al., 2014). This kind of unit is characterized by a continuous, high albedo ejecta blanket and pronounced continuous crater rims elevated relative to surrounding materials (Yingst et al., 2014). Interestingly, this bright unit does not have a different composition with respect to the nearby units. However, a significant variability is present within the crater itself: the crest of the rim is mostly howarditic material (green in Fig. 7, left) while the floor and the walls toward Tarpeia are enriched in eucritic material (blue in Fig. 7, left). This difference is also supported by the distribution of BI and BII depths, which are less pronounced where the concentration of eucritic material is higher. However, as already noted in the global maps, the correlation between pyroxene composition and band depth is not always present, as evident in the region just outside Alypia on the top/left. This region has a Fe-rich pyroxene patch which also has band depth similar to the values in the rim of crater with more Mg. In this specific location, the presence of elongated features with a straight or slightly curved shape has been identified and interpreted as the result of the propagation of a pressure wave through the regolith (Otto et al., 2013, and references therein). Therefore, it is possible that the weak correlation between pyroxene band depth and composition is masked by mass wasting surficial processes which, through vertical mixing of regolith, can expose on the surface new uncontaminated material. Mariamne (diameter of 30 km centered at 68°S and 204.73°E) has the same compositional dichotomy of Tarpeia (Fig. 8). It is located in an undifferentiated crater unit made of fresh and unmodified impact crater deposits (Williams et al., 2014; Yingst et al., 2014) completely surrounded by Rheasilvia terrain units. The internal walls are howarditic (green in Fig. 8) with two diogenitic units (red in Fig. 8) while in a section of the crater (toward the South Pole), there is a eucritic unit (blue in Fig. 8). This eucritic unit roughly corresponds to dark mass wasted material that extends from the rim of the crater to its floor. The diogenitic and howarditic units, on the contrary, correspond to bright fresh surface. Mariamne and Tarpeia lay in the same kind of geological unit (undifferentiated crater material), they have the same latitude but they are separated by roughly 180° in longitude. As noticed in the description of the lithological map, the area considered in this paper is compositional divided in two parts, one richer in diogenite and the other richer in eucrite. The boundary between these two major compositional regions roughly corresponds to the line connecting these two craters. Interestingly, both craters have diogenitic units in the internal part of the rim, close to the crest, on the same side of the major compositional region richer in diogenite. This observation suggests that these two craters have exposed in the internal walls a diogenitic lower layer which in nearby locations has been covered by a layer with higher concentration of howardite and eucrite. The Rheasilvia central mound has attracted much interest because, according to hydrodynamic codes, the central uplift is the region where the minerals from lower internal layers should occur (Ivanov and Melosh, 2013). However, applying our method based on the analysis of pyroxene BII, we have found that the regolith in this location has a low concentration of diogenitic material confirming early results based on Dawn spectral measurements (McSween et al., 2013). On the contrary, in the mapped area, most of the Mg-rich pyroxene is found in correspondence with the

Fig. 8. Mariamne crater. Lithological map superimposed to the FC clear filter mosaic.

lowest topographic values. In particular, Severina crater (diameter of about 35 km centered at 75°S and 121°E) lies in a region with a high concentration of diogenitic materials (red yellow and green in the map in Fig. 6, left) which, together with Matronalia Rupes, has subunits formed by pyroxenes with the highest amount of Mg versus Fe on Vesta’s surface. It is worth noting that spectral data from all mission phases confirm the presence of a lithological unit formed by pure diogenite in this crater. It is also worth noting that while this region is dominated by Mg-rich pyroxene, there is a Fe-rich pyroxene unit of roughly 0.4 km2 (Fig. 9). This unit is located on the wall of a small crater that lies on the floor of Severina crater. A possible explanation is that the second impact, after the one that generated Severina, exposed part of the eucritic crust. This interpretation is supported by the fact that the eucritic unit is on the internal rim of the small crater and that this specific location is among lowest topographic values of all across Vesta. In addition in the floor of the small crater there is a pond of fine grained material likely generated by downhill movement of the regolith. The presence of mass wasted material in the floor reinforce the possibility that the eucritic unit is crustal bedrock. We cannot rule out, however, the possibility that also the eucritic unit itself is part of the mass wasting within the crater. The Parentatio Rupes is a system of scarps and ridges that partially surrounds the Rheasilvia center. This feature is a topographic high that is elevated for roughly 7 km above a reference ellipsoid with axis 285 km and 229 km (Otto et al., 2013). It is about 90 km long and it is centered at about 74°S and 253°E. In the geological map, the Rupes is identified as a mass wasting unit likely related to the Rheasilvia formation event (Williams et al., 2014). In this location, pyroxene has a generally uniform composition with an Mg–Fe ratio in line with the average values across Vesta. However, there is a spot of roughly 6 km2 centered at about 74°S and 315°E in a scarp where pyroxene has a higher concentration of Mg (Fig. 10). Within Rheasilvia a slump of material with different pyroxene chemical composition has already been identified (Stephan et al., 2014). However, the shape of the spot that we have identified in the Parentatio Rupes does not support the interpretation of this diogenite as coming from a slump, since the highest concentration of diogenite is found halfway up the scarp. It seems more likely that the downhill movement of regolith, as a consequence of the slope, has exposed the walls of the scarp. It is worth notice that the difference in elevation between the topographic top and bottom of the Rupes, in this particular location, is about 4 km; therefore it is possible that the layer of regolith, supposed to be less than 1 km in this region (Jaumann et al., 2012), has been completely removed as a consequence of the downhill movement.

E. Ammannito et al. / Icarus 259 (2015) 194–202


either the validity of the impact simulations or the models describing the internal structure. Spectra of several locations within the Rheasilvia quadrangle have been analyzed, and small scale variations in the composition of pyroxene have been found. These variations identify lithological units that extend for tens of km, although small units of less than 1 km have also been found. In particular, we notice that the small units of less than 1 km are located in regions of opposite lithological class: the eucritic unit is inside Severina, which is generally enriched in diogenite, while the small diogenitic unit is in the Parentatio Rupes, which is dominated by eucrite. This fact, associated with the limited extent of such units, is likely an indication of a high level of compositional heterogeneity within the Vestan crust, which is difficult to reconcile with the conventional three-layer model for Vesta’s petrogenesis. Acknowledgments Fig. 9. Severina crater. Lithological map superimposed to the FC clear filter mosaic. The black arrow indicates the eucritic unit described in the text.

The authors gratefully acknowledge the contribution of the Dawn Instruments and Operations Teams. This work is supported by NASA through the Dawn project and by an Italian Space Agency (ASI) grant. The VIR spectrometer is funded by ASI. It was built by Selex-Galileo, Florence, Italy and is now managed by INAF – Istituto di Astrofisica e Planetologia Spaziali, Rome, Italy. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at 017. References

Fig. 10. Parentatio Rupes. Lithological map superimposed to the FC clear filter mosaic.

5. Conclusions We presented and described the maps of spectral parameters such as pyroxene band centers and depths, reflectance at 1.4 lm and 2.8 lm band depth in the Rheasilvia quadrangle. We found a broad anti-correlation between pyroxene band centers and depths: pyroxenes with bands at shorter wavelengths (Mg-rich) generally have deeper bands. However, surficial processes like mass wasting, weathering and grain size variability could easily alter this link. On the contrary, the reflectance is not correlated with the parameters of the pyroxene bands. In addition, we found that the Rheasilvia quadrangle is free of OH absorption signatures. We derived lithological maps using a method based only on the pyroxene 2 lm band. We found results that are broadly consistent with previous lithological maps of the same region, with improvements in the spatial resolution and capability of discriminating between the possible lithological units. We confirm that the central mound is dominated by eucrite. Pyroxene native to the upper mantle (diogenite) has been found mainly in a region delineated by Tarpeia, Severina and Mariamne craters. In terms of relative depths of the origin of pyroxene, the distribution we have found is compatible with numerical models describing the Rheasilvia impact event. However, the absence of a detectable amount of olivine, in combination with the diogenite found in the basin, is threatening

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 (32), 4829–4836. Ammannito, E. et al., 2013a. Vestan lithologies mapped by the visual and infrared spectrometer on Dawn. Meteorit. Planet. Sci. 48 (11), 2185–2198. http:// Ammannito, E. et al., 2013b. Olivine from Vesta’s mantle exposed on the surface. Nature 504, 122–125. Ammannito, E. et al., 2014. Updates in the IR responsivity of VIR, the spectrometer of the Dawn mission. 40th COSPAR Scientific Assembly. Held 2–10 August 2014, in Moscow, Russia, Abstract B0.4-56-14. Beck, A.W., McCoy, T.J., Sunshine, J.M., Viviano, C.E., Corrigan, C.M., Hiroi, T., Mayne, R.G., 2013. Challenges in detecting olivine on the surface of 4 Vesta. Meteorit. Planet. Sci. 48 (11), 2155–2165. Binzel, R.P., Xu, S., 1993. Chips off of Asteroid 4 Vesta – Evidence for the parent body of basaltic achondrite meteorites. Science 260, 186–191. Binzel, R.P. et al., 1997. Geologic mapping of Vesta from 1994 Hubble Space Telescope images. Icarus 128, 95–103. Burns, R.G., 1993. Mineralogical Applications of Crystal Field Theory, second ed. Cambridge University Press, Cambridge, UK. Carry, B. et al., 2010. First disk-resolved spectroscopy of (4) Vesta. Icarus 205, 473– 482. Clenet, H. et al., 2014. A deep crust–mantle boundary in the Asteroid 4 Vesta. Nature 511 (7509), 303–306. Combe, J.-Ph. et al. 2015. Reflectance and hydrated materials of Vesta. Icarus 259, 21–38. De Sanctis, M.C. et al., 2011. The VIR spectrometer. Space Sci. Rev. 163, 329–369. De Sanctis, M.C. et al., 2012a. Spectroscopic characterization of mineralogy and its diversity across Vesta. Science 336, 697–700. De Sanctis, M.C. et al., 2012b. Detection of widespread hydrated materials on Vesta by VIR imaging spectrometer on board the Dawn mission. Astrophys. J. 758. Article ID L36, 5 pp. De Sanctis, M.C. et al., 2013. Vesta’s mineralogical composition as revealed by the visible and infrared spectrometer on Dawn. Meteorit. Planet. Sci. 48 (11), 2166– 2184. De Sanctis, M.C., 2015. Eucritic crust remnant and the effect of in-falling hydrous carbonaceous chondrites characterize the composition of the Marcia region on Vesta. Icarus 259, 91–115. Ermakov, Anton I. et al., 2014. Constraints on Vesta’s interior structure using gravity and shape models from the Dawn mission. Icarus 240, 146–160. http:// Frigeri, A. et al., 2015. The spectral parameter maps of Vesta from VIR data. Icarus 259, 10–20.


E. Ammannito et al. / Icarus 259 (2015) 194–202

Gaffey, M.J., 1997. Surface lithologic heterogeneity of Asteroid 4 Vesta. Icarus 127, 130–157. Ivanov, B.A., Melosh, H.J., 2013. Two-dimensional numerical modeling of the Rheasilvia impact formation. J. Geophys. Res.: Planets 118 (7), 1545–1557. Jaumann, R. et al., 2012. Vesta’s shape and morphology. Science 336, 687–690. Jutzi, M. et al., 2013. The structure of the Asteroid 4 Vesta as revealed by models of planet-scale collisions. Nature 494 (7436), 207–210. 10.1038/nature11892. Klima, R.L., Pieters, C.M., Dyar, M.D., 2007. Spectroscopy of synthetic Mg–Fe pyroxenes I: Spin-allowed and spin-forbidden crystal field bands in the visible and near-infrared. Meteorit. Planet. Sci. 42, 235–253. Li, J.-Y., 2013. Body-fixed Coordinate Systems for Asteroid (4) Vesta. . Li, J.-Y. et al., 2010. Photometric mapping of Asteroid (4) Vesta’s southern hemisphere with Hubble Space Telescope. Icarus 208, 238–251. Li, J.-Y. et al., 2013. Global photometric properties of Asteroid (4) Vesta observed with Dawn Framing Camera. Icarus 226 (2), 1252–1274. 10.1016/j.icarus.2013.08.011. Longobardo, A. et al., 2014. Photometric behavior of spectral parameters in Vesta dark and bright region as inferred by the Dawn VIR spectrometer. Icarus 240, 20–35. Mandler, B.E., Elkins-Tanton, L.T., 2013. The origin of eucrites, diogenites, and olivine diogenites: Magma ocean crystallization and shallow magma chamber processes on Vesta. Meteorit. Planet. Sci. 48 (11), 2333–2349. 10.1111/maps.12135. Marchi, S. et al., 2012. The violent collisional history of Asteroid 4 Vesta. Science 336, 690–694. Marzari, F., Cellino, A., Davis, D.R., 1996. Origin and evolution of the Vesta asteroid family. Astron. Astrophys. 316, 248–262. McCord, T.B., Adams, J.B., Johnson, T.V., 1970. Asteroid Vesta: Spectral reflectivity and compositional implications. Science 168 (3938), 1445–1447. http:// McCord, T.B., Li, J.-Y., Combe, J.-P., McSween, H.Y., Jaumann, R., Reddy, V., Tosi, F., Williams, D.A., Blewett, D.T., Turrini, D., Palomba, E., Pieters, C.M., de Sanctis, M.C., Ammannito, E., Capria, M.T., Le Corre, L., Longobardo, A., Nathues, A., Mittlefehldt, D.W., Schröder, S.E., Hiesinger, H., Beck, A.W., Capaccioni, F., Carsenty, U., Keller, H.U., Denevi, B.W., Sunshine, J.M., Raymond, C.A., Russell, C.T., 2012. Dark material on Vesta from the infall of carbonaceous volatile-rich material. Nature 491 (7422), 83–86. McSween, H.Y. et al., 2011. HED meteorites and their relationship to the geology of Vesta. Space Sci. Rev. 163, 141–174. McSween, H.Y. et al., 2013. Composition of the Rheasilvia basin, a window into Vesta’s interior. J. Geophys. Res. 118, 335–346. Neumann, W., Breuer, D., Spohn, T., 2014. Differentiation of Vesta: Implications for a shallow magma ocean. Earth Planet. Sci. Lett. 395, 267–280. Otto, Katharina A. et al., 2013. Mass-wasting features and processes in Vesta’s south polar basin Rheasilvia. J. Geophys. Res.: Planets 118 (11), 2279–2294. http:// Pieters, C.M. et al., 2011. Surface composition of Vesta: Issues and integrated approach. Space Sci. Rev. 163, 117–139.

Prettyman, T.H. et al., 2012. Elemental mapping by Dawn reveals exogenic H in Vesta’s regolith. Science 338, 242–246. Reddy, V. et al., 2010. Compositional heterogeneity of Asteroid 4 Vesta’s southern hemisphere: Implications for the Dawn mission. Icarus 210, 693–706. Reddy, V. et al., 2012. Color and albedo heterogeneity of Vesta from Dawn. Science 336, 700–704. Reddy, V. et al., 2013. Comparing Dawn, Hubble Space Telescope, ground-based interpretations of Vesta. Icarus 226 (1), 1103–1114. Righter, K., Drake, M.J., 1997. A magma ocean on Vesta: Core formation and petrogenesis of eucrites and diogenites. Meteorit. Planet. Sci. 32, 929–944. Roatsch, Th. et al., 2012. High resolution Vesta High Altitude Mapping Orbit (HAMO) Atlas derived from Dawn framing camera images. Planet. Space Sci. 73 (1), 283– 286. Ruesch, O., Hiesinger, H., De Sanctis, M.C., Ammannito, E., Palomba, E., Longobardo, A., Zambon, F., Tosi, F., Capria, M.T., Capaccioni, F., Frigeri, A., Fonte, S., Magni, G., Raymond, C.A., Russell, C.T., 2014. Detections and geologic context of local enrichments in olivine on Vesta with VIR/Dawn data. J. Geophys. Res.: Planets 119 (9), 2078–2108. Russell, C.T. et al., 2012. Dawn at Vesta: Testing the protoplanetary paradigm. Science 336, 684–686. Russell, C.T. et al., 2013. Dawn completes its mission at 4 Vesta. Meteorit. Planet. Sci. 48 (11), 2076–2089. Ruzicka, A., Snyder, G.A., Taylor, L.A., 1997. Vesta as the HED parent body: Implications for the size of a core and for large-scale differentiation. Meteorit. Planet. Sci. 32, 825–840. Schenk, P. et al., 2012. The geologically recent giant impact basins at Vesta’s south pole. Science 336, 694–697. Schröder, S.E. et al., 2013. Resolved photometry of Vesta reveals physical properties of crater regolith. Planet. Space Sci. 85, 198–213. j.pss.2013.06.009. Stephan, K. et al., 2014. Small fresh impact craters on Asteroid 4 Vesta: A compositional and geological fingerprint. J. Geophys. Res.: Planets 119 (4), 771– 797. Sunshine, J.M., Pieters, C.M., 1993. Estimating modal abundances from the spectra of natural and laboratory pyroxene mixtures using the modified Gaussian model. J. Geophys. Res. 98, 9075–9087. Sunshine, J.M., Pieters, C.M., Pratt, S.F., 1990. Deconvolution of mineral absorption bands: An improved approach. J. Geophys. Res. 93, 6955–6966. Thomas, P.C. et al., 1997. Impact excavation on Asteroid 4 Vesta: Hubble Space Telescope results. Science 277, 1492–1495. Toplis, M.J. et al., 2013. Chondritic models of 4-Vesta: Implications for geochemical and geophysical properties. Meteorit. Planet. Sci. 48 (11), 2300–2315. Vernazza, P. et al., 2005. Analysis of near-IR spectra of 1 Ceres and 4 Vesta, targets of the Dawn mission. Astron. Astrophys. 436, 1113–1121. Williams, D.A., Yingst, R.A., Garry, W.B., 2014. Introduction: The geologic mapping of Vesta. Icarus 244, 1–12. Wilson, L., Keil, K., 2012. Volcanic activity on differentiated asteroids: A review and analysis. Chem. Erde 72, 289–321. Yingst, R.A. et al., 2014. Geologic mapping of Vesta. Planet. Space Sci. 103, 2–23. Zambon, F. et al., 2014. Spectral analysis of the bright materials on the asteroid Vesta. Icarus 240, 73–85.