CCSD3ZF0000100000001NJPL3IF0PDSX00000001 PDS_VERSION_ID = PDS3 RECORD_TYPE = STREAM OBJECT = TEXT PUBLICATION_DATE = 1995-04-18 NOTE = "Experiment description for the Clementine bistatic radar observations conducted in April. 1994. Formatted for display or printing with up to 78 constant- width characters per line." END_OBJECT = TEXT END During April 1994 a set of lunar bistatic radar (BSR) observations was carried out using the DSPSE/Clementine spacecraft. The spacecraft high-gain antenna (HGA) was aimed at the Moon's surface and radio echoes were received on Earth using 70-m antennas and receiving equipment at stations of the NASA Deep Space Network. Continuous wave S-band (13 cm wavelength) signals in right-circular polarization were transmitted from the spacecraft. Signals in both right- and left-circular polarizations were received, converted to baseband, sampled, and recorded using the open-loop receiving system at each DSN station. Data obtained from these experiments can be interpreted to improve our understanding of lunar surface physical and electrical properties on scales of centimeters to tens or hundreds of meters. Observing times for these experiments were chosen so that one of the Moon's polar regions would be visible from Earth and also so that the Earth would be close to the spacecraft's orbit plane. A primary objective of the experiments was to detect and measure radio echoes in a near-backscatter geometry from possible deposits of water ice in permanently shadowed areas on the lunar surface. Such deposits are believed to have distinct angular and polarization ratio scattering signatures and have been detected previously in the polar regions of Mercury and Mars and on the surfaces of Callisto, Ganymede, Europa, and (possibly) Titan. During observations on 9-10 April the Clementine HGA was aimed toward the south polar region on the Moon; on 23-24 April the HGA was aimed toward the north polar region. Bistatic angles near 0 degrees (the backscatter condition) were observed in each experiment; bistatic angles larger than 90 degrees were obtained in some. During a single revolution on 23 April, the HGA was aimed toward the (moving) specular point and quasi-specular echoes were obtained. In addition at least four radio occultations by the lunar limb were recorded during the 23-24 April observations. NB: The revolution number refers to an observational pass over the Moon. The revolution number was incremented by one each time the spacecraft passed over the south pole prior to the beginning of data acquisition. REVOLUTION_NUMBER was used in lieu of orbit number because of the way the orbit number was defined by the mission. The orbit number was incremented each time the spacecraft passed through the equatorial plane on the sunlit side of the Moon. Thus, the orbit number generally changed in the middle of an observational pass. This proved to be awkward in defining the data acquired by a single pass over the Moon. Background: Anomalous echoes from icy surfaces were first reported for the Galilean satellites Callisto, Ganymede, and Europa by Campbell et al. (Icarus, 34, 254,1978). They measured echo strengths comparable to what would be expected from a polished metal sphere with polarizations that were exactly opposite to what would have been expected from a smooth surface. The unusual properties were subsequently confirmed by Goldstein and Green (Science, 207, 179, 1980) and Ostro et al. (Icarus, 44, 431, 1980), but satisfactory explanations for the anomalous behavior have been elusive. Double bounces from hemispherical craters were proposed early (Ostro and Pettengill, Icarus, 34, 268, 1978) but have since been discounted on several grounds, not the least being the fact that Europa has few if any craters. Refractive scattering (Hagfors et al., Nature, 315, 637, 1985) and decoupling of electromagnetic modes (Eshleman, Nature, 319, 755, 1986; Eshleman, Adv. Space Res., 7, 5, 133, 1987) are more widely accepted mechanisms. Hapke (Icarus, 88, 407, 1990) has introduced the concept of 'coherent backscatter' and Peters (Phys. Rev. B, 46, 801, 1992) has developed the theory in detail. More recently anomalous echoes have been observed from the south residual polar cap on Mars (Muhleman et al., Science, 253, 1508, 1991) and from the region around both poles on Mercury (Harmon et al., Nature, 369, 213, 1994). Co-location of the region of anomalous radar response and the residual south polar cap on Mars heightened speculation that refractive scattering within an icy matrix was primarily responsible for the enhanced echoes; "coherent backscatter" could also play a part. Analysis of echoes from airborne radar observations over the Greenland ice sheet (Rignot et al., Science, 261, 1710, 1993) supports this interpretation. The detection of enhanced radar echoes from the polar regions of Mercury was serendipitous, but immediately precipitated further speculation that ice could be discovered in other parts of the solar system, permanently shielded from the Sun, such as on the Moon (Paige, Nature, 369, 182, 1994). The echoes from Mercury, although anomalous, are muted in comparison with other bright targets, suggesting that a certain amount of dust has been mixed with the ice. Because of the difficult observing geometry from Earth, little attention has been paid to Earth-based (monostatic) radar echoes from the Moon's polar regions until recently. The nearly grazing incidence angles give good range resolution, but shadowing is severe and can only be partly mitigated by scheduling observations to take advantage of libration. Stacy (PhD dissertation, Cornell Univ., 1993) used special processing techniques to obtain very high resolution images of polar areas from radar data acquired at Arecibo Observatory. He identified candidate locations with anomalous radar properties, but he chose not to interpret them as ice deposits. Clementine BSR Objectives: Signal-to-noise ratio in the data examined by Stacy (1993) was at least three orders of magnitude better than could be obtained from a Clementine bistatic experiment. But any Earth-based experiment is limited to the backscatter geometry. Low level enhancements of echo strength and polarization ratios which vary with bistatic angle can only be detected and measured using a spacecraft as either the transmitter or receiver. Since the Earth passed through Clementine's polar orbit plane once every two weeks, bistatic surface scattering experiments targeted on polar regions could be conducted during each of those alignments. Coincidentally, libration was such in April 1994 that nearly maximum tilt of the polar regions toward or away from the Earth coincided with times when the Earth was also in the orbit plane. Experiments were scheduled for three weekends in late March through late April. The March experiments failed when previously unknown receiver tuning limitations were discovered during the first set of observations. Those were corrected within the next two weeks, and new experiments were scheduled for 9-10 April. As the spacecraft approached the Earth-target line, the HGA was turned to illuminate the polar region, yielding backscatter echoes when the spacecraft passed the Earth-target line. As the spacecraft continued in its orbit, the bistatic angle increased, in some cases to beyond 90 degrees. If ice were present in significant amounts around the South Pole, an enhancement and polarization reversal could be expected when the Earth, spacecraft, and target region were aligned. The magnitude of the enhancement and the degree to which the echo polarization was changed could then be related to the amount of ice present and its "cleanliness". The South Pole appears to be the better candidate for long-lived ice deposits because of the existence of a large depression in that area which could not be imaged by Clementine's optical instruments (Shoemaker et al., Science, 266, 1851, 1994). Thermal infrared images, however, may be useful in further study of this area. On 23-24 April additional experiments were scheduled when the North Pole was visible from Earth. The same geometrical conditions were sought -- alignment of Earth, spacecraft, and target area; spacecraft motion would change the bistatic angle and allow measurement of any intensity or polarization dependence on that angle. DSPSE/Clementine personnel also carried out calculations to obtain quasi-specular scattering measurements during part of one revolution on this date. The specular point is the location on the target surface where mirror-like reflection would be expected. In fact, the process at radio wavelengths is "quasi-specular" because the Moon's surface is not perfectly smooth. Instead a collection of "glints" centered on the specular point would be observed if the receiver had enough resolution to form an image of the reflecting surface. For a smooth surface (relative to the 13 cm wavelength) the glints are very concentrated and the echo will appear both strong and narrow in the frequency domain. For a rough surface, the glints will be spread widely over the surface; in the frequency domain, the echo will be broad. To a first approximation, the rms roughness of the surface is proportional to the width of the echo spectrum in frequency and the integrated power in the echo spectrum is proportional to the Fresnel power reflectivity. Since the Fresnel reflectivity can be related to the dielectric constant of the surface material, quasi-specular measurements can provide quick measures of fundamental properties of the lunar regolith. Analysis: Most analyses of bistatic radar data have been conducted in the frequency domain, as noted above. When continuous wave signals are transmitted, the measurement parameters of interest are the Doppler dispersion and the strength of the echo signal. With knowledge of the spacecraft trajectory and the HGA pointing direction and radiation pattern, one can either model the expected echo for various choices of surface properties (Tyler and Ingalls, J. Geophys. Res., 76, 4775, 1971) or invert the data to obtain empirical estimates of the surface radar cross section (Parker and Tyler, Radio Science, 8, 177, 1973). Template fits based on model calculations are often preferred because they are simpler, faster, and less likely to be unstable; but recent success with bulk inversions of Magellan data (Tyler et al., J. Geophys. Res., 97, 13115, 1992) suggests that stability questions are manageable. For the data obtained by Clementine around the lunar poles, the analysis is complicated somewhat by the fact that the HGA illuminated the surface in a geometry for which the radar cross section is poorly understood. In the near-backscatter case, the incidence and reflection angles will be on the order of 85 degrees, well beyond the angles commonly encountered in nadir- and side-looking radar systems. Studies of early Earth-based backscatter data (Evans and Hagfors, Radar Astronomy, Ch. 5, 1968) suggested that an angular dependence of the form cos(phi) obtains, but early investigators were unable to specify a mechanism which would produce this result. Searching for ice-like signatures does not require use of any specific model for surface scattering. Instead, carefully calculated echo intensities in the two received polarizations can be compared and then followed as the bistatic angle changes. If the echo intensity peaks in the backscatter geometry and if there is a change toward reversing the polarization ratio at the same point, then one can assume good evidence for ice within the target. The challenge in this case is to maintain registration of Doppler contours on the surface as the spacecraft-target geometry changes (so that target points do not drift through the frequency spectrum) and to allow for the possibility that several target regions within the HGA illumination may be candidates and that each has its own angular dependence profile. Simpson et al. (1994 Annual Meeting of the Division for Planetary Sciences) showed evidence for anomalous behavior in preliminary processing of Clementine data, but the anomalies were small and not as well-correlated with predicted times and geometries as might have been expected. Although the data are presented primarily as power spectra versus frequency, there is coarse time and geometrical resolution. By careful analysis it may be possible to recover radar cross section functions and/or low-resolution images in the polar areas. These two results cannot be recovered uniquely, however. Data Set: The data set resulting from the Clementine BSR experiments includes the following: Open-Loop Digital Tapes: Data were acquired at 50000 8-bit samples per second on 6250 bpi digital tapes. Approximately 80 tapes cover the observations conducted on 9-10 April and 23-24 April. About two dozen additional tapes were made during tests preceding the actual experiments. The 6250 bpi data have been copied to 8 mm cartridges, and those have been delivered to Stanford. Four of the original 6250 bpi tapes were defective; only part of one of those tapes has been delivered to Stanford. These "original data records" (ODRs) are the primary data type for this experiment. Archival Tracking Data Files: DSN receivers can also track spacecraft signals using closed loop receivers, which lock to the carrier signal. For bistatic surface scattering experiments the carrier is, at best, intermittent; and it is often weaker than the surface echo. During periods when the HGA is pointed toward Earth, however, the closed loop receivers can provide accurate estimates of signal strength and frequency. One ATDF has been received at Stanford. These data may be useful in estimating the frequency of the on-board crystal oscillator during periods when the HGA was pointed toward Earth. SPICE Files: Files containing information on the spacecraft trajectory and attitude are needed for complete analysis of the BSR data set. Orbit reconstructions carried out at Goddard Space Flight Center have been obtained. Files of reconstructed spacecraft attitude have been obtained from the JPL/PDS Navigation and Ancillary Information Facility. HGA Radiation Patterns: Measurements of the HGA radiation pattern are needed to weight the signals expected from surface regions according to their angular displacement from the HGA boresight. Several files of calibration data have been obtained from Chris Lichtenberg, Fred Domer, and Howard Taylor of the DSPSE/Clementine Project. These patterns were obtained from an engineering model of the HGA, but that is expected to be adequate. Link Monitor and Control Logs: During many radio science experiments, the receiving configuration is known well in advance and real-time changes are not necessary. Further, the signals are often weak relative to the constant noise power in the output bandwidth. For the Clementine BSR experiments only rough estimates of signal strength were prepared in advance, so real-time changes in attenuator settings were required frequently. Calibrations, conducted in series with the actual experiment data acquisition, also resulted in wide ranges of signal levels and frequent changes. The LMC logs record operator instructions on a command-by-command basis at each receiving station. They provide the best record of operator actions and will be essential in establishing baseline reference levels during periods when signals were especially dynamic. Both LMC logs from DSS 63 (Madrid) have been captured but others have been lost.