Special Reports

Laser Altimeter Observations from MESSENGER's First Mercury Flyby

See allHide authors and affiliations

Science  04 Jul 2008:
Vol. 321, Issue 5885, pp. 77-79
DOI: 10.1126/science.1159086


A 3200-kilometers-long profile of Mercury by the Mercury Laser Altimeter on the MESSENGER spacecraft spans ∼20% of the near-equatorial region of the planet. Topography along the profile is characterized by a 5.2-kilometer dynamic range and 930-meter root-mean-square roughness. At long wavelengths, topography slopes eastward by 0.02°, implying a variation of equatorial shape that is at least partially compensated. Sampled craters on Mercury are shallower than their counterparts on the Moon, at least in part the result of Mercury's higher gravity. Crater floors vary in roughness and slope, implying complex modification over a range of length scales.

Topography is a fundamental measurement to characterize quantitatively the surfaces of solid planetary bodies at length scales ranging from the long-wavelength shape to such local and regional processes as impact cratering, volcanism, and faulting. During the first flyby of Mercury by the MESSENGER spacecraft on 14 January 2008 (1), the Mercury Laser Altimeter (MLA) (2, 3) successfully ranged to the planet's surface, providing the first altimetric observations of the planet from a spacecraft.

Previous measurements of the shape and topography of Mercury had been derived from Earth-based radar ranging (4, 5) constrained by range observations from Mariner 10 (6). Because of the low inclination (7°) of Mercury's orbital plane to the ecliptic, Earth-based altimetric profiles are limited to ±12° latitude and have a spatial resolution of ∼6 × 100 km2 and a vertical precision of 100 m (5). These observations indicated a planetary reference radius of 2440 ± 1 km, an equatorial ellipticity of 540 ± 54 ×10–6, and an equatorial center of figure (COF) offset from the planet's center of mass (COM) of 640 ± 78 m in the direction 319.5° ± 6.9° W (6, 7).

The MLA profile (Fig. 1) was acquired approximately along Mercury's equator, in a region that was in darkness during the flyby, and within the hemisphere not imaged by Mariner 10. Consequently, there are no optical images of the region in which altimetry was collected, so we used an Arecibo radar image (8) to correlate the profile with surface features. The MLA began ranging ∼1 min before the spacecraft's closest approach and continued for ∼10 min. Usable returns were received up to an altitude of 1500 km, which was larger than the expected maximum of 1200 km (2). As the spacecraft velocity and range from Mercury changed during the flyby, the size of laser spots on the surface varied from 23 to 134 m and the shot spacing varied from 888 to 725 m (9). The vertical precision varied with the received signal strength and is <15 cm at the closest range, limited by the resolution of the timing electronics. The radial accuracy of ∼100 m is limited by uncertainties in the trajectory associated with errors in the ephemerides of MESSENGER and Mercury.

Fig. 1.

(Top) MLA profile (vertical exaggeration 105:1). (Bottom) Arecibo radar image [adapted from (8)] with MLA profile location (white line) superposed. Arrows at top indicate locations of craters in Table 1 interpreted from detailed analysis of MLA profile points. The locations of several of the major craters are indicated by arrows on the radar image. The two-ringed circular structure in the Arecibo image at ∼55 to 60°E is represented in part by a deep depression in the altimetry, but north-south radar ambiguities may be contributing to the structure in the image.

The profile spans ∼ 20% of the circumference of the planet and shows a 5.2-km dynamic range of topography and 930-m root-mean-square (RMS) roughness (Fig. 1). The radius of Mercury apparently decreases by 1.4 km along the equator from ∼10° to 90° E, corresponding to a 0.02° downward slope to the east. This long-wavelength surface tilt begins 30° west of the previously estimated COF/COM offset (6) and was not sampled in Earth-based radar altimetry (4). Such a long-wavelength slope, if a fundamental feature of the equatorial shape of the planet, might arise from crustal thickness or crustal density variations, global-scale mantle density variations, or topography along the planet's core-mantle boundary, which for Mercury is ∼600 km beneath the surface.

The slope can be interpreted in the context of an ellipsoidal planetary shape (10). If we suppose that the difference in principal moments of inertia, BA, is entirely a result of an ellipsoidal distribution of surface mass with density ρs and with semi-axes a > b > c, then Embedded Image(1) from which we may write Embedded Image(2) where A < B < C are the principal moments of inertia of Mercury, Cm is the moment of inertia of the mantle and crust alone, and M, R, and <ρ> are the mass, radius, and mean density of Mercury, respectively. The form of the right hand side of Eq. (2) is convenient because from measurements we have (BA)/Cm = (2.03 ± 0.12) × 10–4 (11), and from models we have Cm/C = 0.4 to 0.7 and C/MR2 = 0.31 to 0.35 (12, 13). The value of (ab) from MLA is 1.4 km. Surface shell densities of 2000 and 3000 kg m–3, which bracket likely values, yield (ab) values of 0.26 to 0.87 km. These (ab) values are less than that observed, implying that the surface topography is at least partially compensated. The simplest explanation is that support of topography occurs by variations in crustal thickness, an inference that has also been invoked to explain Mercury's COF-COM offset by analogy with the situation on other terrestrial planets (6).

MLA profiled numerous depressions interpreted as impact craters on the basis of topographic expression and appearance on the Arecibo image. As on other terrestrial planets, the geomorphological complexity of impact craters on Mercury increases with diameter (14), with craters undergoing a transition at a diameter of about 11 km from a simple bowl shape to a planform with a flat floor, slumped walls, and a central peak (14). On a given planet, the ratio of depth to diameter (d/D) is uniform for unmodified complex craters, and where the MLA profile crossed close to crater centers, the ratio is ∼1/40, less than on the Moon (d/D ∼ 1/20). Two examples are craters in the longitude range ∼45° to 50° E that have diameters (107 km and 122 km) comparable to that of Tycho (102 km), among the largest fresh craters on the Moon's nearside (Fig. 2). Whereas Tycho has a depth of 4.8 km (15), these craters have depths of 2.4 and 2.9 km, respectively. Although these craters may have undergone postformation modification, their substantially shallower depths in comparison to lunar counterparts is likely due at least in part to Mercury's higher surface gravity (16).

Fig. 2.

Close-up of two craters showing contrast in floor roughness and tilt. The vertical exaggeration is 30:1.

Crater floors may preserve evidence for modification processes that bear on geological evolution. From MLA we characterized the floors of complex craters by measuring apparent (along-track) slope, RMS roughness, and the widths of returned laser pulses, the last of which are indicative of topographic variance (due to roughness and footprint-scale slopes) within individual laser spots (17). Along-track slopes of 11 crater floors range from –10 m km–1 to +22 m km–1 (–0.57° to +1.26°) (18) (Table 1) and do not display an obvious pattern; most notably, these floor slopes do not correlate with the eastward long-wavelength slope. The RMS roughness over the approximate length scale of the crater diameter ranges from 5.7 to 110 m. Pulse widths vary considerably within individual craters, from 6 to >60 ns, indicating 2 to >20 m of vertical variability on horizontal scales of tens to hundreds of meters. For the craters studied, apparent slope, RMS roughness, and pulse widths are uncorrelated, which implies that the processes that caused tilting and created the roughness of crater floors are complex and do not operate uniformly over different length scales. Potential sources of modification include anelastic relaxation, volcanic resurfacing, tectonic subsidence, wall slumping, and ejecta emplacement from younger nearby impacts, and the variability implies that a combination of these processes operated on the profiled craters.

Table 1.

Crater apparent slopes, RMS roughness, and pulse widths.

View this table:

References and Notes

View Abstract

Navigate This Article