Research Project -- Radar Mapping of the Lunar Poles

The Moon is our nearest planetary neighbor, and has been visited six times by human explorers during the Apollo missions. Today, the Moon remains an active topic for scientific research, providing clues to the early history of the solar system and the role of meteorite impacts in shaping planetary surfaces, and serving as a proving ground for the tools of exploration we will use at Mars and beyond. Scientists at CEPS are involved with a number of lunar science experiments, using both Earth-based observatories and Moon-orbiting spacecraft. These pages focus on radar mapping of the Moon using Earth-based radio telescopes. New computing capabilities make it possible to create high-resolution maps of the lunar surface in a short period of time, and these data are being used to answer a variety of important questions about the geologic history of the Moon.

We use the 305-m radio telescope at the Arecibo Observatory, Puerto Rico, to transmit a radar signal toward the Moon, and receive the reflected echoes at the 100-m Green Bank Telescope in West Virginia. The Arecibo radar can transmit signals at a wavelength of either 12.6 cm (2380 MHz frequency) or 70 cm (430 MHz frequency). These signals penetrate into the lunar regolith - a layer of mixed rock and packed fine dust that covers the lunar surface - up to 40 m in some areas at 70-cm wavelength. It is this significant depth of penetration that makes the new observations so interesting - we can probe much farther below the surface than is possible with optical or infrared observations. This technique has been used on Earth to locate ancient river channels beneath the Sahara desert, and may someday be used to see below the layers of dust that cover much of Mars.

The radar echoes do not form a picture in the same way that a camera operates. Instead, we measure the strength of the reflected signal as a function of time and frequency. Because the Moon is a sphere, the radar return from the surface at a particular time corresponds to a circular region about the nearest point to us. The Moon's spin causes echoes to be shifted in frequency: increasing frequency on the approaching hemisphere and decreasing frequency on the receding hemisphere. Mathematical models for the motion of the Earth and Moon make it possible to convert the raw observations to maps of the lunar surface.

Due to tidal effects the Moon appears to wobble, or librate, slowly in the sky. The center of the Moon that we see slowly shifts in position, over an area about 7o wide (~210 km) in latitude and longitude. This libration allows an observer to see about 57% of the Moon's total surface over the course of several months. We use these changes in position to map the edges, or limbs, of the Moon. In many cases these areas have not been previously mapped by radar.

The radar echoes are measured in two polarizations that describe how the signal interacts with the surface and sub-surface layers of the Moon. Reflections in one of these polarizations include mirror-like bounces from smooth parts of rocks or gently rolling topography, while the other polarization is dominated by echoes from rock edges or other rough features. Using the two measurements can greatly help in understanding the geology of a target area.

We have collected almost complete coverage of the lunar nearside at 70-cm wavelength. A mosaic of southern hemisphere radar echoes, with color from the circular polarization ratio to show areas of greater rock abundance in orange and red tones, is shown here. These maps have a best spatial resolution of about 400 m per pixel. More detailed maps are being collected, for interesting sites, with about 200 m spatial resolution. For an overview of the full dataset at 1.2-km resolution, download a movie showing a circle around the limb of the Moon starting from the north pole. For detailed information and access to the complete dataset, visit the Planetary Data System: http://pds-geosciences.wustl.edu/missions/lunar_radar/index.htm

At 12.6-cm wavelength, it is possible to obtain images with spatial resolution as fine as 20 m per pixel, over areas about 300 km by 600 km in extent, in one hour of observing time. Images collected to date include coverage of the south pole, portions of the southern highlands from Tycho to Moretus craters, southern Mare Serenitatis, and the Aristarchus Plateau. Preliminary images for the south pole may be obtained from: http://www.nasm.si.edu/research/ceps/research/moon/radar_south_images.cfm

a. Ice at the lunar poles?

Despite the wobble caused by libration, the Sun's upper edge never rises more than about 2° above the horizon at the north and south poles. Craters formed by meteorite impacts near the poles can have floors that are never exposed to sunlight, even over hundreds of millions of years. Any water that reaches the floors of these craters is "cold trapped", and could remain for long periods, disturbed only by the small meteorite impacts that cause erosion and overturn of the regolith. Water could be brought to the Moon by comet impacts, or in a very slow delivery by particles from the Sun.

Water ice with internal cracks is a very strong reflector of radar signals, and the nature of the scattering process inside an ice layer leads to high values of the circular polarization ratio. These two properties occur over almost all of the permanently shadowed crater floors near the poles of Mercury, suggesting sheets of ice at least a meter or so thick. In contrast, radar studies of the Moon at 12.6-cm and 70-cm, while limited to seeing perhaps 25% of the shaded areas, have yet to find evidence for extensive, thick sheets of water ice. This work has also shown that high circular polarization ratios, once thought to be limited to scattering from ice, are actually common in rocky areas around and within younger craters, regardless of whether they ever see the Sun (for example, the bright yellow to red tones around a fresh crater in this image).

Neutron spectrometer measurements by the Lunar Prospector spacecraft, however, show that hydrogen is more abundant near the poles than in most other parts of the Moon. One explanation for the hydrogen enhancement is a 0.5-1% average abundance of ice within the upper meter of the packed dust. Such a low concentration would have no unique radar signature. Taken together, the radar and neutron data suggest that any ice within shadowed craters near the Moon's poles is in the form of small, distributed grains, making it potentially more difficult to extract as a resource.

b. Properties of ejecta from giant, basin-forming impacts.

Radar signals at 70-cm wavelength can penetrate to depths of perhaps 40 m in the low-loss, feldspar-rich regolith of the lunar highlands. This makes such observations particularly sensitive to changes in the regolith composition and rock population. In contrast, photographic and multispectral data probe only the upper few microns of the lunar surface, and are strongly modulated by shallow compositional differences (such as crater rays) and maturity (space weathering). We combine the new 70-cm data with Clementine infrared and Earth-based thermal measurements to better understand the nature of ejecta from the Moon's giant basins.

The distribution of basin ejecta is of fundamental interest for lunar geology, particularly in the context of planning new sample return missions. For example, giant basins such as Orientale excavate materials from depths of several km, serving as a probe of the upper crust (seen at 70-cm wavelength). This is the motivation for the South-Pole/Aitken Basin Sample Return Mission requested under the recent NASA New Frontiers initiative. Acquisition of topography data by the Clementine laser altimeter has identified a number of possible ancient basins in the southern hemisphere that are not well defined from photogeologic analysis. We are interested in determining the role of these basins in contributing ejecta to the near-side highlands. This study is also relevant as a precursor to orbital radar studies of large basins on Mars.

The polarization properties of 70-cm and 12.6-cm radar echoes have been used to map the extent of melt-rich material (molten rock mixed with cooler ejecta) from Orientale, the last major basin to form, that fills low areas like crater floors across the south pole (shown as shades of green to red). Because these deposits contain significant chunks or layers of the now-solid melt, craters formed by small impacts yield many more rocks than are found near craters in much older regolith. As a result, the floors of many large south polar craters are strewn with rocks, even though the original craters are quite old and eroded.

In the areas east and southeast of Orientale, the 70-cm radar data reveal the outlines of older mare basalt flows that were covered by debris when the giant basin formed. These "cryptomare" deposits cover an area almost the size of Nebraska, and may extend even farther west beneath the increasingly thick basin ejecta.

c. Distribution of material around large impact craters.

Meteorite impacts on planetary surfaces create large concentric deposits of ejecta, and some of the distant material can reach hundreds of kilometers from the crater on an airless body like the Moon. Ejecta deposits are of interest because they expose material from deep beneath the surface, and are a major source of fine and coarse sediments on some planets. We use new 70-cm radar data to study the distribution of ejecta about large lunar craters as a guide to understanding impact craters on the Earth, Mars, and Venus. Of particular interest are spoke-like and halo-like areas of low radar return associated with large craters such as Aristoteles. These dark deposits are not typically evident in multi-spectral data for the same sites, suggesting that the radar is sensing primarily the abundance of rocks within the ejecta layer. This provides a tool for mapping the patterns of coarse and fine-grained debris as a function of distance from the crater.

d. Variations in ilmenite content and age among lunar basalts.

The large impact basins on the Earth-facing side of the Moon are filled with overlapping layers of basalt lava flows. These stacks of flows can be up to several kilometers deep in the centers of the basins, but individual flows are typically thin and extend for many hundreds of kilometers. The basalt erupted soon after the basins formed, and most volcanic activity on the Moon ended by about 2 billion years ago. Lunar basalts, which are in many ways similar to lava erupted from large terrestrial shield volcanoes like those in Hawaii, have wide variations in their proportion of iron and titanium. Metallic oxide minerals such as ilmenite (a titanium oxide) could represent resources for future lunar base construction.

Radar and ultraviolet to near-infrared observations can be used to measure the variations in ilmenite among lunar basalts. The radar signals, which penetrate to depths of several meters, can be especially useful in mapping deposits that are obscured by thin surface layers of debris from nearby impact craters. A 70-cm radar image of Mare Imbrium shows echo variations due to changes in absorption of the radio signal. Darker areas have higher ilmenite concentration, while brighter regions have lower concentration. Interestingly, the age of the basalt units can also have a major effect - younger flows have thinner regolith and more abundant rocks that scatter the radar signal, while older flows are more radar-dark. This suggests that we can use the radar data to identify even small patches of very young and very old lava flows.

e. Pyroclastic deposits - A potential resource for future explorers

Lunar pyroclastic deposits reflect an explosive stage of the basaltic volcanism that filled impact basins across the lunar nearside. These fine-grained mantling layers are of interest for their association with early mare volcanic processes, and as possible sources of volatiles and other species for lunar outposts. Our 70-cm radar image of the large pyroclastic deposit that blankets the Aristarchus Plateau reveals a lava flow complex (noted by the yellow outlines) that covers a significant portion of the plateau and appears to have formed by spillover of magma from the large sinuous rille Vallis Schröteri. The pyroclastics mantling these flows are heavily contaminated with rocks 10 cm and larger in diameter. The 12.6-cm data confirm that other areas are mantled by 20 m or less of material, and that there are numerous patches of 2-cm and larger rocks associated with ejecta from Aristarchus crater. Much of the radar-detected rocky debris lies within the mantling material and is not evident in optical or infrared images. The new radar data identify thick, rock-poor areas of the pyroclastic deposit best suited for resource exploitation.