Engineering topics
Publications and source records attributed to Frey, Herbert.
Evidence supporting an intense early bombardment on the Moon in addition to the traditional Late Heavy Bombardment at approx. 4 BY ago include the distribution of N(50) Crater Retention Ages (CRAs) for candidate basins, a variety of absolute age scenarios for both a "young" and an "old" Nectaris age, and the decreasing contrasts in both topographic relief and Bouguer gravity with increasing CRA.
For more than a decade we have used GRIDVIEW to help analyze topographic and related data for Mars and more recently for the Moon. Our focus has been to employ the stretching, contouring, profiling, circle-fitting and other capabilities of GRIDVIEW to search for Quasi-Circular Depressions (CTAs) in MOLA, LOLA and other topographic data, and for Circular Thin Areas (CTAs) in Mars and Moon model crustal thickness data. Both QCDs and CTAs likely represent buried or obscured impact craters not readily visible in image data. We found clear evidence for a much larger population of buried impact craters in the northern lowlands of Mars (Frey et al. 2002), suggesting that part of the Red Planet is not significantly younger than the southern highlands. Edgar and Frey (2008) found that the N(300) crater retention ages of both areas were essentially identical, a conclusion confirmed by Wyatt (unpublished data) using more recent crustal thickness data for Mars. MOLA topographic data and MOLA-derived crustal thickness data were used to both identify a large number of previously unrecognized very large impact basins (D> 1000 km) on Mars and to determine relative crater retention ages for them (Frey, 2008). The distribution of N(300) CRAs suggested most formed in a relatively short interval of time. This dating also suggested the main magnetic field of Mars disappeared during this period (Lillis et al., 2008), because only the youngest basins systematically lack a remagnetized signature. Similar QCD and CTA analysis of first Clementine (Frey, 2011) and more recently LOLA topographic and LOLA-derived crustal thickness data for the Moon (Frey et al., 2011) revealed a significantly larger population of impact basins > 300 km in diameter than previously known. N(50) CRAs suggest a two-peak distribution of ages (Frey, 2012). An improved counting process confirms the two peaks, perhaps indicating both a pre-Nectaris Early Heavy Bombardment (EHB) as well as a Late Heavy Bombardment (LHB) on the Moon (Frey and Burgess, 2012, this meeting), with obvious implications for the early bombardment history of the Earth.
LOLA topography and LOLA-derived crustal thickness data provide evidence for a population of impact basins on the Moon that is likely a factor 2 larger than the classical lists based on photogeology. Frey (2012) determined N(50) crater retention ages (CRAs) for 83 candidate basins > 300 km in diameter by counting LOLA-identified craters superimposed over the whole area of the basins. For some basins identified in topography or model crustal thickness it is not possible to unambiguously identify the crater rim as is traditionally done. Also, Quasi-Circular Depressions (QCDs) > 50 km in diameter are recognizable in the mare-filled centers of many basins. Even though these are not apparent in image data, they likely represent buried impact craters superimposed on the basin floor prior to mare infilling and so should be counted in determining the age of the basin. Including these as well as the entire area of the basins improves the statistics, though the error bars are still large when using only craters > 50 km in diameter. The distribution of N(50) CRAs had two distinct peaks which did not depend on whether the basins were named (based on photogeology) or recognized first in topography or crustal thickness data. It also did not depend on basin diameters (both larger and smaller basins made up both peaks) and both peaks persisted even when weaker candidates were excluded. Burgess (2012, unpublished data) redid the counts for 85 basins but improved on the earlier effort by adjusting the counting area where basins overlap. The two peak distribution of N(50) ages was confirmed, with a younger peak at N(50) ~ 40-50 and an older peak at N(50) ~ 80-90 (craters > 50 km diameter per million square km). We suggest this could represent two distinct populations of impactors on the Moon: one producing an Early Heavy Bombardment (EHB) that predates Nectaris and the second responsible for the more widely recognized Late Heavy Bombardment (LHB).
There are about 30 very large impact basins on Mars, > 1000 km in diameter, most of which are revealed by their topographic and/or crustal thickness signatures. Crater retention ages and model absolute ages suggest these all formed in a relatively short time (100-200 million years?), perhaps during a "Late Heavy Bombardment" (LHB) caused by the evolution of the orbits of the giant planets. This so-called "Nice Model" of planetary formation may explain the LHB on the Moon at about 3.9 billion years ago and would have produced a similar bombardment throughout the inner solar system. The formation of 30 very large impact basins would have had catastrophic environmental consequences for Mars, which were further complicated by the demise of the global magnetic field at about the same time. If there are no very large basins on Mars older than the 30 we see and the LHB really lasted everywhere only a short time, there may have been a relatively longer time (~400 million years?) during which Mars and the Earth suffered no major impact trauma and during which conditions on both worlds may have been far more habitable than during the LHB. However, if the formation of the Mars crustal dichotomy was due to an even larger giant impact that predated the very large basins, all record of this earlier and possibly more clement time on Mars may have been erased. Ages of the smaller but still very large basins can be used to approximately date the giant impact (if it occurred). Even the very large basins appear to have reset the crater retention ages of the entire crust of Mars and may have by themselves erased any record of an earlier time.
During the period we calculated the atmospheric data sets related to its mass and angular momentum distribution. For mass, we determined the various harmonics from the NCEP-NCAR reanalysis, especially the low-order harmonics that are useful in studying the gravitation distribution as will be determined from the GRACE mission. Atmospheric mass is also related to the atmospheric loading on the solid Earth; we cooperated with scientists who needed the atmospheric mass information for understanding its contributions to the overall loading, necessary for vertical and horizontal coordinate estimation. We calculated atmospheric angular momentum from the NCEP-NCAR reanalyses and 4 operational meteorological centers, based on the motion (wind) terms and the mass (surface pressure) terms. These are associated with motions of the planet, including its axial component causing changes in the length of day, more related to the winds, and the equatorial component related to motions of the pole, more related to the mass. Tasks related to the ocean mass and angular momentum were added to the project as well. For these we have noted the ocean impact on motions of the pole as well as the torque mechanisms that relate the transfer of angular momentum between oceans and solid earth. The activities of the project may be summarized in the following first manuscript written in December 2002, for a symposium that Dr. Salstein attended on Geodynamics. We have continued to assess ocean angular momentum (OAM) quantities derived from bottom pressure and velocity fields estimated with our finite-difference barotropic (single layer) model. Three years of output (1993-95) from a run without any data constraints was compared to output from a corresponding run that was constrained by altimeter data using a Kalman filter and smoother scheme. Respective OAM time series were combined with corresponding atmospheric series and compared to observed polar motion. The constrained OAM series provided slightly better variance reduction than the unconstrained series. Analysis provided a check on the estimation scheme and pointed to further work to improve the determination of OAM using this method. A significant effort was also devoted to quantifying effects of uncertainties in high frequency winds on the mean and seasonal momentum exchange between atmosphere and oceans.
The large population of buried impact basins found in MOLA elevation data on Mars provides compelling evidence for a pre-Noachian crust below the oldest visible Early Noachian surface units, and lowland crust below the younger plains that is Early Noachian in age, older than much of the visible highlands, but not as old as the buried pre-Noachian highlands. The large (D greater than 200 km) buried basins are suggested by Quasi-Circular Depressions (QCDs) that are not apparent in image data, and include features up to 3000 lun diameter in both the lowlands (Utopia) and highlands (a newly found "Ares Basin"). There are about a dozen QCDs larger than 1000 km diameter. We have placed these large features in a relative age sequence based on superimposed smaller QCD. Only the youngest and most obvious of these (Hellas, Argyre, Isidis) lack magnetic anomalies within their main rings. These all have an N(200) cumulative crater density of less than 2.5. Somewhat older lowland-making basins (Utopia, Chryse, Acidalia) with an N(200) age of approximately 3.0, have weak magnetic anomalies, and the oldest, most subdued basins (including Ares) with N(200) greater than 3.5 have many strong magnetic anomalies within their main ring. These older basins likely formed before the main magnetic field died. We have compared our inventory of large QCDs with the distribution of gravity anomalies and with a crustal thickness model which shows many roughly circular areas of thinner crust completely or partly surrounded by narrow regions of thicker crust. These have the structure expected for impact basins, and many of them do correspond to the visible or buried QCDs we previously identified. But there are cases where the crustal thickness feature is offset from the QCD found in topography alone, and there are also several, sometimes large examples of such features which do not coincide with QCDs previously identified. For example, we find several likely buried basins revealed in the crustal thickness data in the Arcadia and Amazonis regions which we did not previously identify, including several features in the 600-1200 km diameter range.
This report describes research directed toward understanding the response of volcanic lahars and lava flows to changes in the topography along the path of the flow. We have used a variety of steady-state and time-dependent models of lahars and lava flows to calculate the changes in flow dynamics due to variable topography. These models are based on first-order partial differential equations for the local conservation of volume. A global volume conservation requirement is also imposed to determine the extent of the flow as a function of time and the advance rate. Simulated DEMs have been used in this report.
Coverage of the northern hemisphere of Mars by the Mars Orbiter Laser Altimeter (MOLA) during the aerobraking hiatus and the two Science Phasing Operation periods provides improved definition and characterization of large impact basins. Gridded MOLA data show the Utopia Basin has a pronounced bowl-like structure, as opposed to the interior rises suggested by the earlier USGS DEM. The elevation structure is concentric about the basin center as mapped by McGill. In particular, the proposed inner ring closely follows the -4 km contour over much of the southern, western and northwestern sides. Higher topography along portions of the dichotomy boundary aligns with the basin's outer ring. High topography in the polar region also occurs where the outer ring should lie, raising the possibility that perhaps some of the polar topography is due to basin structure as well as ice. Two MOLA passes near Phison Rupes provide evidence for a large "stealth" hole where Viking imagery show little evidence of any major structure. The 2 km deep, 600 km wide depression at 31OW, 3ON is as large as the Cassini impact basin 1000 km to the SW. While Cassini is easily recognized in image data, the "MOLA Hole" is not. If this depression is a deeply eroded and buried impact basin (as perhaps suggested by a decrease in the crater density and somewhat smoother terrain than in adjacent areas), it is not clear why it has managed to maintain its great depth. In Tempe at the dichotomy boundary a 300 km wide impact basin is revealed by pronounced bowl-like topography centered at 87W, 47N, even though only about 1/3 of the basin rim structure is obvious. The basin lies on a sloping boundary zone, with the more buried N rim up to 2 km below the rugged S rim. A similar N-S asymmetry in basin ring structure occurs for the much larger Isidis Basin, where the S rim rises 6 km but the subdued N rim rises barely 2 km above the floor. There is essentially no topographic expression of the main ring in the NE quadrant of Isidis where, if it exists, it lies below Hesperian-age plains.
The distribution of ancient massifs and old cratered terrain in the southern Daedalia region indicate the presence of at least two and probably three impact basins of large size. One of these is located near where Craddock et al. placed their center for the Daedalia Basin, but it has very different ring diameters. These basins have rings exceeding 1000 km diameter and overlap significantly with centers separated by 500 to 600 km at nearly identical latitudes of -26 to -29 deg. The smaller westernmost basin appears slightly better preserved, but there is little evidence for obvious superposition that might imply a temporal sequence. Recognizing the improbability of random impacts producing aligned, nearly contemporaneous features, we suggest these basins may have resulted from clustered impactors.
Many ancient, highly degraded large impact basins exist on Mars. In many cases these basins overlap or are overlapped by more easily observed, presumably younger, impact basins. While impact basin overlap is becoming more recognized, the effects of such overlap have only occassionally been described. Such effects will depend on a variety of factors including the absolute and relative size of the basins, the degree of overlap, the state of the lithosphere and its thermal gradient at the time of impact, and the time between impacts. There now exists enough evidence for overlapping basins of different sizes that some of these can be discussed. This paper highlights some examples of the obvious effects of basin overlap.
We propose a new center (35.5 W, 32.5 N) and ring assignment for the original Chryse impact basin based on photogeologic mapping and re-examination of the published geology. Noachian features in the Chryse Planitia area are the best indicators of the original ancient multiringed impact structure. While other workers have centered the Chryse impact on the topographic low associated with Hesperian volcanic and fluvial deposits, we suggest that the center of the original Noachian-age excavation cavity was located 800 km farther NE, and the basin topography was significantly modified over time.
Among the outstanding problems in Martian geology are the cause of the off-axis and asymmetric distribution of the southern polar layered terrain and residual ice deposits and the cause of the orientation of scarps, valleys, and re-entrant canyons which occur there. A perhaps related problem region is the apparently small number of large (D greater than 500 km) impact basins seen in the relatively well-preserved cratered terrain of the south polar region. Previously only the 850 km wide South Polar Basin was easily recognized. The south polar region was mapped in detail, searching for evidence of ancient, highly degraded impact basins that may have escaped earlier notice, for two reasons: (1) to determine whether the apparent absence of large impact basins is due to incomplete mapping and recognition or a fundamental characteristic of the Martian crust related to the origin of the Martian crustal dichotomy, and (2) to determine whether ancient impact basins, if they exist, exert some control on the distribution of volcanic and polar deposits in the southern hemisphere and on the topography on which these deposits lie. Several promising candidates, including a large pre-Hellas basin in the Malea Planum region and an older but comparably sized basin overlapping South Polar, were previously described. Concentration is on the possible influence of the candidate basins in localizing the asymmetric distribution of polar deposits and in controlling the orientation of structures found within these deposits.
Free-air and Bouguer gravity anomalies from a 50x50 field, derived from re-analysis of Viking Orbiter and Mariner 9 tracking data and using a 50x50 expansion of the current Mars topography and the GSFC degree 50 geoid as the equipotential reference surface, with the Martian crustal dichotomy are compared. The spherical harmonic topography used has zero mean elevation, and differs from the USGS maps by about 2 km. In this field the dichotomy boundary in eastern Mars lies mostly at -1 to -2 km elevation. Bouguer gravity anomalies are shown on a map of Noachian, Hesperian, and Amazonian age terrains, simplified from current geologic maps. The map is centered at 300 deg W to show the continuity of the dichotomy boundary. Contour interval is 100 mgals. Gravity and topography were compared along approximately 40 profiles oriented parallel to the dichotomy boundary topographic gradient, to determine how the geophysical character of the boundary changes along its length and what this implies for its origin and development.
Free-air and Bouguer gravity anomalies from a 50x50 field (MGM635), derived at the Goddard Space Flight Center, with global topography, geology, and the distribution of large impact basins was compared. The free-air gravity anomalies were derived from re-analysis of Viking Orbiter and Mariner 9 tracking data and have a spatial resolution of 250-300 km. Bouguer anomalies were calculated using a 50x50 expansion of the current Mars topography and the GSFC degree 50 geoid as the equipotential reference surface. Rotational flattening was removed using a moment of inertia of 0.365 and the corrections from Table B2 of Sleep and Phillips. Crustal density and mean density were assumed to be 2.9 and 3.93 gm/cm(sup 3). The spherical harmonic topography used has zero mean elevation, and differs from the USGS maps by about 2 km. Comparisons with global geology use a simplified map with about 1/3 the number of units on the current maps. For correlation with impact basins, the recent compilation by Schultz and Frey was used.
Among the problems in Martian geology are the cause of the off-axis and asymmetric distribution of the southern polar layered terrain and residual ice deposits and the cause of the orientation of scarps, valleys, and reentrant canyons that occur. A problem region is the small number of large (D greater than 500) impact basins seen in the relatively well preserved cratered terrain of the south polar region. This region was mapped in detail, searching for evidence of ancient, highly degraded impact basins for two reasons: To determine whether the apparent absence of large impact basins is due to incomplete mapping and recognition or a fundamental characteristic of the Martian crust related to the origin of the Martian crustal dichotomy, and whether ancient impact basins exert some control on the distribution of volcanic and polar deposits in the Southern Hemisphere and on the topography on which these deposits lie. The possible influence is examined of the candidate basins in localizing the asymmetric distribution of polar deposits and in controlling the orientation of structures found within these deposits.
Many passive margins not complicated by nearby anomalous crustal structure have satellite elevation crustal magnetic anomaly contrasts across them that are recognizable in reduced-to-pole versions of the Magsat and POGO data. In the Labrador Sea region this contrast is particularly well developed with strong positive anomalies overlying the continental crust of Greenland and eastern Canada and prominent negative anomalies situated over the Labrador Sea and Baffin Bay. In this work, forward modeling of the large-scale crustal bodies in this region (continental, oceanic, passive margin, several anomalous structures) was used to show that the Magsat anomaly contrast is due simply to the change in crustal susceptibility and thickness at the continental/oceanic crustal transition. Because the thickness varies more than the average susceptibility from continental to oceanic crust, the strong anomaly contrast is essentially an edge effect due mostly to the change in crustal structure.
This review, discusses publications during the last quadrennium (1987-1990) that used aeromagnetic data, marine magnetic data, satellite magnetic data, and rock magnetic and petrologic data to provide information on the sources of magnetic anomalies. The publications reviewed reflect increased integration of rock magnetic property and petrologic studies with magnetic anomaly interpretation studies, particularly in deep crustal magnetization, exploration for hydrocarbons, and inversion of marine magnetic anomalies. Interpretations of aeromagnetic data featuring image display techniques and using the horizontal gradient method for locating magnetization boundaries became standard.
Retracking of overland returns from the Seasat altimeter using algorithms originally developed for recovering elevations over ice has led to the successful recovery of high quality continental topography over Australia and other continents. Cross-over analysis both before and after orbit adjustment shows the altimetric data over land to have a 2-3 m quality. Direct comparison of gridded Seasat data with surface data re-averaged in the same way shows excellent agreement except where Seasat data are sparse, due either to poor track spacing or to dropouts caused by loss of tracker lock over steeply sloping ground. These results suggest that useful topographic data can be derived from Seasat and the more recent Geosat altimeters for parts of the world where surface data are few or of poor quality.