Abstract
It is important to estimate a lunar crustal thickness because it directly indicates the depth of "Magma Ocean (Warren, 1985)". Recent analyses of Apollo seismic data show that the existence of a crust seems to be significantly sharower than that of Apollo-era conclusions (Khan and Mosegaad 20002; Lognonne et al., 2003); approximately 60 km thick at the Apollo 12 site (Toksoz et al., 1974). In this study assuming that the Bouguer gravity is the consequence of variations in the thickness of a crust, we present a global structure of the lunar crust by using gravity, topography data, and such new seismic constraints.In the previous lunar crustal thickness model (Zuber et al., 1994; Neumann et al., 1996; Wieczorek and Phillips, 1998), horizontal crustal density distributions are not considered although the heterogeneity of a crustal composition inferred from the VIS and NIR multispectral properties on the lunar surface (e.g. Lucey et al., 1998) is obvious.There are correlations between the iron abundances of lunar samples and those inferred normative densities. Therefore we model a horizontally heterogeneous density model of lunar crust by using both iron abundances on the lunar surface as measured by the Lunar Prospector gamma-ray spectrometer (Lawrence et al., 2001) and iron abundance-normative density relations of lunar samples. We also model the vertical distribution of the density of crust by using the pressure dependence of elastic velocities of lunar rocks in laboratory measurements (e.g. Mizutani et al., 1974), which is resulted from pore closures of rocks due to hydrostatic pressure.