From the coherent Doppler shift of the signal used by the Deep
Space Network to track Magellan [ cf. Saunders et
al., 1992], the acceleration of the spacecraft along the
line-of-sight (LOS) can be obtained. These LOS acceleration data
are used to derive gravity and geoid information for Venus.
Similar data were obtained from the Pioneer Venus (PV) spacecraft,
but were limited by solar plasma interference at the wavelength
used by PV's S-band transponder (18 cm) and the highly elliptical
orbit of the spacecraft (
km). This latter factor
greatly limited the resolution of gravity at high latitudes (far
from periapsis). Much of the Magellan LOS data was collected at an
X-band wavelength (12.6 cm), which is less sensitive to solar
plasma interference. Moreover, Magellan's initial orbit was less
eccentric (
km) than Pioneer Venus'. And in late
May of 1993 the Magellan spacecraft began the first-ever
aerobraking maneuver by a planetary probe, using Venus' atmosphere
to bring the spacecraft into a near-circular orbit. As a result,
data have been collected since August of 1993 from an orbit with
periapsis near 180 km and apoapsis between 500 and 600 km
[ Konopliv and Sjogren, 1994]. It is hoped that
LOS data will be collected for almost all of Venus at a spacecraft
altitude less than 450 km before the mission is terminated in the
fall of 1994.
These new data have allowed Konopliv and Sjogren
[1994] to produce spherical harmonic models of gravity and geoid
to degree 60 and beyond, resolving features with half-wavelengths
as small as 
km. The high resolution of such data assists
efforts to distinguish the various modes of support of topography,
helping to differentiate, for example, between support of
topography due to variations in crustal thickness, that due to
thermal anomalies in the mantle (e.g., mantle plumes), and that
due to the long-term elastic strength of the lithosphere. A
commonly used quantity is the apparent depth of compensation (ADC)
for a region, which is a rough measure of the depth of
low-density material that may be supporting high topography; large
depths (
km) tend to suggest support of topography due to
thermal anomalies in the mantle and to rule out support due to
crustal thickness variations. More precise information can be
obtained from the analysis of spectral admittance and coherence
between gravity and topography; these quantities can also be used
to estimate the thickness of the elastic lithosphere [ cf.
Dorman and Lewis, 1970; Forsyth,
1985].
Most examinations of gravity/topography relationships using Magellan data have involved hotspot features: volcanic rises and coronae. In general, these might be expected to be regions where the lithosphere is relatively thin, and where heat flow and thermal gradients are high compared to the global average. However, recent studies have found surprisingly large values of both the apparent depth of compensation and of effective elastic lithospheric thickness.
Schubert et al. [1994] investigated the gravity,
geoid, and topography of eight corona- or chasmata-related regions
on Venus. For the three largest coronae examined (Artemis, Latona,
and Heng-o) they found ADC's near 200 km, and ratios of geoid to
topography (GTR) near 30 m km
. These large values and the
locations of the gravity highs with respect to topography are most
simply explained either by flexural support of topography
(requiring an extremely thick mechanical lithosphere) or by the
presence at depth of a large positive mass anomaly such as a
subducted slab. For the chasmata segments and other coronae they
examined, ADC's were between 150 and 75 km, and GTR's ranged from
6 to 20 m km
. These features can be explained by either
thermal isostasy (variable thinning of a thermal lithosphere
approximately 100 km thick ) or by the presence at depth of large
positive density anomalies [ Schubert et al.,
1994].
A study of the gravity/topography relationships of several
volcanic rises, including Bell, Atla, Western Eistla, and Beta
Regiones, and utilizing both the global spherical harmonic field
and local inversions, found ADC's of 125, 175, 200, and 225 km,
respectively [ Smrekar, 1994]. With the possible
exception of Bell Regio, these ADC's are taken to indicate the
presence of low-density, buoyant material deep beneath the
lithosphere, perhaps providing thermal and/or dynamic support for
the high topography of these volcanic rises. Spectral coherence
and admittance of gravity and topography for Bell Regio yielded an
effective elastic thickness of 
km [
Smrekar, 1994] and was interpreted in terms of a hotspot model
in which long-wavelength topography is supported by a broad
thermal anomaly at the base of the lithosphere and
short-wavelength topography is due to volcanic construction which
loads the top of the mechanical lithosphere.
Phillips [1994] used a similar analysis for Atla Regio, assuming
that the admittance of gravity and topography at short wavelengths
(<1000 km) was due solely to volcanic loads applied at the top
of the mechanical lithosphere. This method yields effective
elastic thicknesses for the region of at least 40 km.
Perhaps the most interesting result of these studies are the large values obtained for the thickness of the mechanical lithosphere. Estimates for average heat flux for Venus range from approximately 50 to 75 milliwatts per square meter [ Solomon and Head, 1991 and references therein]. Effective elastic thicknesses of 30 km or more imply thermal gradients, and thus heat fluxes, that are considerably less than these values. Yet the obtained estimates are from volcanic rises and coronae---features that are among the best candidates for relatively thin lithosphere and high heat flow on Venus. There seem to be four possible explanations (which are not mutually exclusive): (1) Global heat flow has been significantly overestimated. (2) The modeled features are not adequately represented by static, linear flexural models and dynamic and/or non-linear models are needed. (3) Crustal and mantle rocks are considerably stronger under Venus conditions than experiments have indicated. (4) Gravity data are not adequately resolving the short wavelengths critical to coherence/admittance estimates. With the exception of the latter, these explanations all carry significant ramifications for our understanding of fundamental aspect of Venus geology and geophysics.