Lunar crustal density estimate from the GRAIL and LOLA-based global gravitational field models

We use data from the GRAIL and LRO satellite missions to estimate the horizontally varying density of the lunar crust. We determine the density model by parametrising the density using spherical harmonic functions up to degree 400. The density estimate depends on the difference between the data from the global gravitational field model generated by the topography measured by the LOLA sensor and the data from the GL1500E global gravitational field model derived from the GRAIL mission. To reduce the numerical complexity of the calculations, we approximate the topography by a sphere and test the sensitivity of the density estimates to the size of the spherical radius. We further calculate a global gravitational field model generated by the estimated horizontally varying density and the LOLA topography. We analyse the results by admittance, correlation, and Bouguer fields for degrees 150-600. The highest agreement with the input data is obtained for the approximating sphere identical to its Brillouin counterpart. Overall, the horizontally varying density model provides a more realistic gravitational field than the one from the constant crustal density. The advantages of the applied approach lie in the speed of calculation, low requirements on hardware, and ease of implementation.