Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
3 Planetary Surfaces INTRODUCTION Recently developed techniques promise to lead to a considerable increase in our knowledge of lunar and planetary surfaces. A much greater potential than had previously been realized has been demonstrated for investigating, with ground-based equipment, the surfaces of the planets in the visual, infrared, and radio regions of the spectrum. Techniques for investigation in other spectral regions may be expected to be developed as more extensive Earth- orbital facilities becomes available. Our best information on the age and history of the Earth is derived from study of its surface structure. A detailed knowledge of the composition and structure of the surfaces of other objects in the solar system should provide us with similar information on those objects and lead to greatly improved knowledge of the origin and evolution of the solar system. Comments on the information to be obtained from the study of planetary surfaces must be largely confined to the terrestrial planetsâMercury, Venus, and Marsâas well as Pluto and the satellites of the giant planets, whose solid surfaces are comparable with Earth's. The surfaces of the outer planetsâ if solid surfaces indeed existâare blanketed by extremely deep and opaque atmospheres and are virtually inaccessible to study. 14
PLANETARY SURFACES 15 DIAMETERS AND ROTATION Radar techniques have recently determined the diameters of Venus and Mercury to an accuracy of several kilometers, a precision far exceeding that achieved previously by telescopic observation. A very accurate upper limit to the diameter of Pluto was also determined recently by astrometric measure- ments made as the planet nearly occulted a star. A planet's rotation can be used to infer some characteristics of its internal structure (Chapter 5) and the dynamics of its atmosphere (Chapter 4). It is, of course, relatively simple to measure the rotation of those planets that present easily identifiable surface markings by noting the times of meridian passage of these markings. Despite the recent important contributions of radar to the determination of the rotation of Mercury and Venus, telescopic in- spection and photography remain the most accurate methods for studying the rotation of Mars and Saturn, while optical and radio measurements are essential for studying the rotational structure of Jupiter. The markings observed on Jupiter are cloud features. These structures appear to move relative to each other at the same latitudes, as well as to dis- play systematic differences in rotation for zones located at different latitudes. The rotation periods of these features around the Jovian axis can vary from the planetary average of just under 10 h by as much as ± 5 min; the rotational period is shortest near the equator. Radio measurements indicate rotation of the Jovian magnetic field which may be linked to the body of the planet; they are discussed in Chapter 5. No distinct clouds are normally visible on Saturn, only belts parallel to the equator. These belts have differing tonal values but no discernible structure. Occasionally, prominent clouds do appear and last long enough (a month or more) to allow their rotational period to be determined with some preci- sion. The results indicate that, as with Jupiter, the period is shortest near the equator and longest at high latitudes. This curious phenomenon has also been observed by careful analysis of the Doppler shifts in the solar spectrum reflected by the planet. The rotational period of Saturn's cloud features varies by as much as 11 percent, being roughly 10 h at the equator and 11 h at high latitudes. The very large shear motions between cloud zones at different latitudes are responsible for the comparatively short life of Saturn's spots, which are observed to be stretched out longitudinally in a matter of weeks and, eventually, to merge with the belts. Telescopic determination of the rotation of Mercury and Venus has been difficult. Nearly a century ago markings were discovered on the surface of Mercury, but these did not move as rapidly as on Mars and became confused
16 PLANETARY ASTRONOMY with the changing phase of illumination. It was assumed that Mercury had a long period of rotation, probably equal to its orbital period (as in the case of the Moon), and most optical observations were biased toward this value. Radar measurements have recently shown that the rotational period is actually about 59 days, or two thirds of the orbital period. A re-evaluation in the light of this discovery revealed that the visual observations are compatible with this unexpected result (in fact, more so than with the former assumption), and a new map of the surface markings has been constructed. Venus shows telescopic markings in visual light only on rare occasions, when features with low contrast are observed. In the ultraviolet, a cloud pattern can sometimes be seen which appears to move in a retrograde direction across the planet at a speed corresponding to a rotation period of about 4.5 days. No data on the period of rotation of the solid planet were available, however, until radar measurements were first made in 1961. Since that time, precision has steadily increased, and the rotation period is now established as close to 243 days retrograde. This result is totally unexpected both for its retrograde direction and because the period appears to agree with a resonance in which Venus presents nearly the same face toward the Earth at each inferior conjunction. The connection between the Earth's orbital revolution and the rotation of Venus is not understood and will likely prove extremely important as a clue to the origin and early history of the solar system. The significance of the different period of the moving cloud pattern as seen in the ultraviolet also remains to be determined, but the differences are probably a manifestation of the complex planetary atmospheric circulation whose nature is entirely obscure. The rotations of Uranus and Neptune have been determined spectroscop- ically in the same way as for Saturn, but the precision of current values is inadequate. Greater precision would be useful in improving the oblateness values derived from rotation for comparison with the oblateness as determined dynamically from observations of these planets' satellites. Such a comparison should yield further information on the distribution of mass between the surface and center of the planet. THERMAL PROPERTIES Temperature is a fundamental parameter of planetary surfaces that can be determined by ground-based observations. Mean temperature and temperature range are of great interest to biology because of their relevance to the existence of life. Temperature distribution and its variation with solar illumination provide information on thermal and electrical surface properties.
PLANETARY SURFACES 17 Temperature as measured by infrared or radio detectors properly applies only to the wavelength region measured; that is, the brightness temperature so obtained refers to the temperature of a blackbody that would emit the same intensity in the given spectral band. Since no surface is a perfect blackbody, measurements taken over a very wide range of wavelengths provide informa- tion about the actual radiating properties of the surface. In addition, measure- ments at the longer radio wavelengths yield data on subsurface emission, while measurements at shorter wavelengths can provide information on the degree and nature of atmospheric absorption. Radio MeasurementsâLow Resolution Planetary emission is most simply measured by including the entire disk of the planet; these measurements require less sensitivity and resolving power. Both the sunlit and unilluminated hemispheres of the inner planets can be measured; however, the geometry of the Earth's orbit limits observations of the outer planets largely to sunlit hemispheres. Wavelengths longer than 3 cm are required to penetrate the atmosphere of Venus and to measure its surface temperature. From the outset, radiometric observations proved surprising. The first measurements of the unresolved disk of Venus in 1956 were made in the range of 3 to 10 cm and showed a radia- tion temperature of nearly 600 °K. So high a temperature had not been antici- pated, and for a number of years many scientists were reluctant to believe that it represented the true temperature of the surface. As a result, these measure- ments were important considerations in designing the experiment complement of Mariner II. More recent measurements of the dark side made at a number of wave- lengths indicated a nearly constant-temperature blackbody at about 600°K but with a sharp decrease in energy at wavelengths less than 3 cm. This energy distribution was interpreted as thermal radiation from the hot surface of the planet with absorption and emission at wavelengths shorter than 3 cm in the progressively higher and cooler atmospheric layers. A gradual decrease in the disk brightness of Venus at wavelengths longer than about 20 cm has not yet been satisfactorily explained, and further work is needed. The first radio measurements of Mercury also seemed anomalous. In 1962, the planet's emission at 3 cm was found to be much higher than that ex- pected if one hemisphere were in perpetual darkness. These results were con- firmed in 1965 when measurements made at 11 cm over a wide range of illumination showed the temperature of the disk to be nearly constant. Radar observations have subsequently shown that the sidereal period of rotation, rather than the expected 88-day sun-synchronous value, is 59 days; all parts of
18 PLANETARY ASTRONOMY the planet are periodically heated by the Sun. More recent observations reveal phase effects at shorter centimeter and at millimeter wavelengths, as would be expected from radiation which originates just beneath the solid surface. Observations of Mars over the wavelength range 3 mm to 20 cm show a disk brightness temperature near 200° with a slight dependence on solar distance. Attempts to detect radiation from Mars at wavelengths as long as 1.5 m have not been successful, giving no evidence for appreciable non- thermal radiation nor for ionospheric or atmospheric effects, consistent with the findings of Mariner IV. Radio MeasurementsâHigh Resolution Although present knowledge of planetary surface temperatures has, at radio wavelengths, been based primarily on the observed emission from the entire disk of each planet, it is important to consider what has been achieved, or might be in the future, with significant improvements in radio facilities. The emphasis should be laid on increasing angular resolution and sensitivity at wavelengths which have already been used extensively. High-resolution radio mapping of planetary surfaces can define the apparent temperature distribution over the disk, as well as its variation with the observed polarization, thus permitting more detailed and meaningful determinations of surface properties. The first high-resolution radio observations of Venus used ground-based fan-beam scans at 3-cm wavelength with the narrow dimension comparable to the disk. Later, 1.9-cm scans from the space probe Mariner II were obtained with resolution of about one sixth the disk. Both indicated limb darkening consistent with the hot-surface interpretation. More recently, a significant series of observations using a high-resolution interferometer at 10 cm has measured the general features of the radio emission over the disk and revealed a brightness at the equatorial limb slightly higher than at the center of the illuminated side and about 25 percent higher than at the polar limb. As originally reduced, the observations were consistent with radiation from a smooth sphere with a dielectric constant of 2.2. Subsequently, a correction derived from the radar estimate of surface roughness has raised the dielectric constant to 2.5. Despite this correction, this value is still less than the dielectric constant of 3.7 derived from the 12.4-cm radar reflectivity, but a large part of this can likely be explained by the absorption of radio waves in the atmosphere of Venus. Discrepancy appears between values for the dielectric constant of the Moon derived from radio and radar observations and shows the incompleteness of our knowledge of the surfaces of these bodies. High-resolution observations of Mars can provide a detailed picture of the dependence of the subsurface temperature on the phase of solar illumination
PLANETARY SURFACES 19 and permit deductions on the thermal and electrical properties of subsurface layers. If the seasonal wave of darkening involves structural changes such as might accompany growth of vegetation, a simultaneous change in the polariza- tion might be detected. The small diameter of Mercury's disk requires an angular resolution of a few seconds of arc to achieve even crude resolution. High-resolution studies similar to those done on the Moon would require 300 times greater resolution than is presently available. Considerable thermal mapping of the Moon at radio wavelengths has been carried out at a much higher relative resolution than can be achieved for the planets. Past observations have given some ideas of the gross thermal and electrical properties of the lunar surface. Although in situ measurements may soon replace remote observations, ground-based data should still be valuable as a source of information on large-scale inhomogeneities in subsurface prop- erties. Also, the value of remote observations of all planetary bodies will be increased when direct calibration of Earth-based observations can be made using in situ lunar measurements. Present low-resolution observations of lunar radio emission over a wide range of wavelengths suggest a small increase of temperature with depth beneath the surface. If confirmed, these findings would be important in deter- mining the thermal budget and structure of the Moon. The observations, which require very accurate absolute calibration over a wide range of wave- lengths, should be repeated. Visual and Infrared Measurements In recent years, there has been a great increase in the number of studies of the lunar surface by infrared techniques. These measurements show that the lunar surface has a very poor ability to conduct heat. Mild and broad variations of thermal conductivity have so far correlated with an age classification system recently developed by lunar geologists and based on the degree of rounding of originally sharp surface features. Thus, perhaps the same agent of cosmic erosion-deposition that increasingly subdues crater rims with time leads to the increasing accumulation of a highly insulating crust. Hot spots on the lunar surface have been found from observations of the night side of the Moon and, even more dramatically, the eclipsed lunar surface, in the 8-14- and 20-p. atmospheric windows. These spots are indicative of differences in surface structure or composition. The surface of Mars can be studied in detail in the visible and infrared using large ground-based telescopes equipped with new detectors. A considerable improvement in resolution can be obtained if such observations are combined with area scanning (see Chapter 6). Also, a direct
20 PLANETARY ASTRONOMY improvement in resolution will result from the more favorable oppositions of 1969, 1971, and 1973, when the disk of the planet will have a maximum angular diameter of 19'.'5, 24"9, and 2 1'M, respectively. Detailed knowledge of Martian surface temperatures in critical regions such as at the edge of the waning polar caps, or in other areas when changes such as the wave of darkening occur, should add valuable clues to the nature of the surface. When area scanning is combined with a photoelectric spectral scanner, many detailed surface and atmospheric features are revealed. In fact, it may be pos- sible to separate the surface and atmospheric phenomena by means of polariza- tion-phase observations obtained in different spectral regions. A number of unsuccessful attempts have been made in the infrared to measure the temperature of the dark hemisphere of Mercury. The sunlit hemi- sphere has a temperature of about 620 °K, while the upper limit for the dark- hemisphere temperature is 150°K. SURFACE STRUCTURE The detailed structure or geography of a planetary surface, in addition to its intrinsic interest, often provides insight into the nature of internal and external forces acting to modify that surface. High-resolution photographs from the surface of the Earth and from orbiting and landing space probes have pro- vided a wealth of information about the lunar surface. With the exception of a very limited sequence on Mars from Mariner IV, however, we do not at present have a similarly detailed picture of the surface of any planet other than Earth. The difficulty in acquiring optical information on the detailed structure of planetary surfaces is directly related to their distances. For example, even at its most favorable opposition, Mars is still about 150 times farther away from the Earth than is the Moon. A resolution of 0'.'2 in telescopic photography under ideal conditions from the surface of the Earth, which corresponds to about 380 m on the surface of the Moon, thus is equivalent to about 50 km on the surface of Mars. It is clear, therefore, that telescopic study of planetary sur- faces, while extremely useful in studying large-scale variations, does not provide the detail that we would like. Earth-based radar can provide considerable data on the statistical properties and distribution over the planet of small-sized surface material. As compared with Earth-based telescopes, radar is capable of relatively high resolution and, in the case of Venus, has the advantage of penetrating the atmospheric cloud layers.
PLANETARY SURFACES 21 Two types of radar interaction with the surface of a planet can be distin- guished. One is a quasi-specular reflection from relatively smooth but inclined surface undulations. In this case, the high degree of coherence of the incident wave is largely preserved on reflection, and backscattering is confined to angles near normal incidence. The other involves a scattering from small, wave- length-sized irregularities. Such irregularities cause considerable depolariza- tion and scatter significantly over a wide range of angles of incidence. By observing the planetary radar scattering law in several polarizations and over a spectrum of operating wavelengths, it has been possible to gain information on both types of scattering for the Moon, Mercury, Venus, and Mars. The mean inclination of the relatively large-scale (as compared with a 70-cm radar wavelength) surface undulations varies from a typical value of 3° for Mars and 6° for Venus to approximately 10° for the Moon and Mercury. The average number of wavelength-sized surface irregularities per unit area is about the same for all the targets studied, although relatively less is known about this component for Mars and Mercury. By taking advantage of the coherence of radio waves reflected from the lunar surface, it has recently become possible to draw radar maps of the dis- tribution of reflected power on the Moon. The basis of the technique is a form of aperture synthesis and is related to the method currently used in airborne "side-looking" radar systems. A description is given in Chapter 6. Because of the large effective apertures that can be brought to bear, and because the atmosphere introduces far less distortion in the phase of radio waves than it does at optical wavelengths, the technique is particularly attractive as a means of mapping the planets. A result recently obtained for the Moon is shown in the Frontispiece, where a surface resolution of about 1 km has been achieved. As larger radar systems become available, it should be possible using this technique to map the distribution of reflected power across Venus and Mercury, with a surface resolution far in excess of that presently obtainable with ground- based optical telescopes. Radar mapping of Venus has already been initiated. The fine delay resolution of radar also makes available a direct determination of the planetary topography along the track of the portion of the planet lying nearest the radar. Although limited to the tropical latitudes of a planet, this type of direct measurement is invaluable in studying equatorial shape. PLANETARY COMPOSITION Detailed analysis of planetary composition requires the soft landing of pay- loads on the planets. Surveyor V made the first such analysis of a very limited area of the lunar surface by measuring the backscatter of alpha par-
22 PLANETARY ASTRONOMY ticles. Oxygen, silicon, and aluminum were identified, and the general chemical composition of the maria sampled is similar to that of a basalt. This finding has been interpreted as favoring the hypothesis that differentiation has occurred in the Moon as the result of partial or fractional melting. Our only other direct knowledge of the chemical composition of solid matter in outer space stems from meteorites. Although recovered meteorites are extremely small samples of the planetary system, their unique availability makes their study a matter of great importance, an importance which should grow as their origins are identified. Meteorites differ in composition and structure from terrestrial rocks and contain minerals unknown in the crust of the Earth. In bulk composition they range from almost pure nickel-iron, through various mixtures of metal with meta- and orthosilicates of iron and magnesium, to almost pure magnesium metasilicate. Minor amounts of the other naturally occurring elements, hundreds of isotopes, and a wide variety of minerals have also been found. The isotopic composition of most of the elements is similar to that of terrestrial rocks; where it differs, the differences are usually in the light elements where terrestrial geological processes have resulted in isotopic frartionation of a different character. Studies have revealed a complex history in meteorites of crystallization, accumulation, breakup, reaccumulation, and possibly long-term thermal meta- morphism. Isotope studies have shown that meteoritic matter was formed 4.5 billion years ago, probably shortly (100 million years) after nucleosynthesis. The meteorites are thus the oldest, and some of them the least differentiated, rocks known. They contain information about the parent body breakup which exposed them as relatively small bodies to cosmic rays; one such major event appears to have occurred about 500 million years ago. It is not clear whether meteorities were formed from one or several parent bodies. If they originated on a planet, most, if not all, were formed on or near the surface. The apparently short length of time between nucleosynthesis and meteorite formation, together with indications that all stony meteorites cooled sufficiently within about a million years to retain xenon, are of fundamental importance to an understanding of the origin of the solar system. Continued work on the xenon isotopes and other rare gases, coupled with careful petrological and geochemical studies of the meteorites, will give a more accurate time scale for the early stage of the formation of the solar system. The Prairie Meteorite Network is currently being used to obtain orbits, masses, and densities for bright meteors. No meteorite observed with this net- work has yet been recovered, but there are indications that the observed fire- balls are caused by large masses of low-density material, possibly related to
PLANETARY SURFACES 23 carbonaceous constituents. If accurate orbits of recovered meteorites can be obtained, they may give additional information on the origin of meteors and their relation to comets and asteroids. Information on the specific composition of planetary surfaces is extremely difficult to obtain from ground-based observations. Under ideal conditions, most natural rock-forming minerals either emit, absorb, or reflect radiation in a sufficiently individual pattern to permit identification by comparison with a reference standard. The potentially useful region of the electromagnetic spectrum extends from gamma rays to microwaves, but severe limitations are imposed by the opacity of the Earth's atmosphere to radiation in many of these spectral regions. The Earth's atmosphere (as well as a number of planetary atmospheres) is opaque to gamma rays, x rays, and a major region of the ultraviolet. The near ultraviolet, however, is available and is usable in mineral identification. Laboratory studies show that many natural rocks luminesce in the ultraviolet when subjected to appropriate excitation. The resulting spectra, usually broad- band and extending into the visible, indicate that the intensity of the lumines- cence is higher for granitic than for basaltic rocks. It has not yet been demonstrated that useful mineralogy can be done by this technique, but the indications are that further laboratory work would be valuable. Most of the visual work has been based on the assumption that albedo is indicative of gross composition. This assumption is, at best, dubious; several processes acting on planetary surfaces alter albedo in a manner that is to some extent independent of composition. Pulverizing tends to lighten material, sputtering may darken, and ultraviolet may bleach; the interactions and absolute effects are largely unknown. Planetary surfaces exhibit a range of colors and polarizations, but, with the exception of the vaguest generalizations, it is almost impossible to associate these data with any particular mineral or rock. Spectroscopic studies of the infrared radiation from planetary surfaces have had particular appeal, because silicate minerals exhibit several infrared absorption bands and are thought to be dominant on many planetary surfaces. Unfortunately, it has been demonstrated in the laboratory that as rock par- ticles approach micron size, band structure is substantially modified and interpretation becomes difficult. The particle size of the lunar soil is sub- millimeter, possibly tens of microns in diameter, and there is reason to believe that the soil of other planets may be characterized by similar dimensions. Meaningful comparative and absolute studies are possible, however, and more measurements in the 8-14-^ window, together with further laboratory studies, should be pursued.
24 PLANETARY ASTRONOMY The use of radar or radiometric microwave measurements does not appear promising as a diagnostic tool for surface-material identification, except with regard to its mean density. Although very little, if indeed any, positive identification of planetary surface composition can be made from Earth's surface, ground-based tech- niques can provide significant information in combination with spaceborne investigations. The use of remote sensing as a differentiating tool does not depend on the ability to account for the detailed characteristics of the sensed spectrum, but only on sufficient spectral resolution and sensitivity to find differences among areas. Since relatively few locations on planetary surfaces will be sampled in situ by men or instruments, those that are sampled will be sites of particular interest to those engaged in remote sensing. It should then be possible to correlate spectra with composition and extrapolate to regions where remotely obtained spectra, but no direct information, are available. SURFACE VARIATIONS The sources of variation within the planetary system are diverse: the changing polar caps of Mars and the associated wave of darkening are surface effects; the changing patterns of Jupiter's belts are atmospheric effects; and the chang- ing brightness of asteroids may be the effect of surface roughness, albedo, irregular shape, or all three. Even the apparently dead surface of the Moon is not without change. Occasional emissions of light in the dark areas of the Moon have been reported by very competent observers for more than 200 years. With the possible ex- ception of a single spectrogram, none has been photographed, and no satisfac- tory explanation has been advanced as to their origin or nature. Systematic searches for emissions in the visual region should be continued. If emissions occur, and especially if they can be observed spectroscopically, they would provide valuable information regarding the lunar subsurface. In contrast with the Moon, almost continual change characterizes the Martian atmosphere and surface. The seasonal growth and disappearance of the polar caps, associated changes in surface coloring, and the wave of darkening, are of particular interest, in part because they suggest biological activity. Portions of the disk are sometimes obscured by vast dust storms, and clouds are observed to form over certain regions and move over the surface of the planet. The transparency of the entire atmosphere, especially in the near ultraviolet, can change dramatically in relatively short periods of time. The surface of Venus is always hidden by clouds, but changes in the cloud
PLANETARY SURFACES 25 structure, especially in the ultraviolet, have been reported. Further understand- ing of the meteorology of Venus must wait until a long-term photographic survey of the planet has given a more detailed picture of the motion of the clouds. The cloud belts of Jupiter have long been noted for their variability. A detailed study of the changing appearance of its apparent disk is of value, not only for meteorological information but also because of possible relations between the clouds, the interior of the planet, and its magnetic field (see Chapter?). Variability is also important to studies in celestial mechanics. Light variations of the satellites of the major planets and the asteroids can assist in determin- ing the speed and sense of rotation and the direction of the rotation axis in space and in giving some indication of the shape of the object. Observations of this type have thus far been confined almost exclusively to some of the brighter minor planets and Pluto. All the variations have in common the need for observations carried out on a regular basis. For some objects, one observation every night or so may be sufficient; for others, ten or more an hour may be desirable. Regular observa- tions for a continuous period of time will require a worldwide network of telescopes, since one observatory can follow a given object only for a limited time, determined by the position of the object in the sky and by the longitude, latitude, and elevation of the observatory. These telescopes must be well distributed in longitude and devoted primarily to planetary work. Because of the scarcity of telescopes with apertures as large as 24 in., additional telescopes will have to be provided if the monitoring program is to give adequate coverage.
