Selenophysical parameter inversion in the Lunar Southern Hemisphere Highland based on mutant particle swarm optimization

Publication date: Available online 14 May 2019Source: Physics of the Earth and Planetary InteriorsAuthor(s): Zhen Zhong, Jianguo Yan, Shuanggen Jin, Menghua Zhu, J. Alexis P. Rodriguez, Huaqiang Zhu, Yi LiAbstractSelenophysical parameters can be estimated based on admittance between gravity and topography, especially high-precision GRAIL gravity and LOLA topography data. These parameters include load ratio between subsurface and surface loads, crustal thickness, crustal density, and effective elastic thickness. Considering non-negligible membrane stress, the lunar lithosphere is best modeled as a thin and elastic spherical shell. Taking into account of the nonlinearity of the governing equation of the shell and premature convergence of PSO, we introduced an updated algorithm of MPSO which considers a self-adaptive inertia weight and a mutation operator as commonly used in Genetic Algorithms (GA). Results indicate that MPSO is relatively more flexible in the choice of tuning parameters than the general algorithm of PSO. A low mutation probability of 0.005 is found to promote globally optimized convergence, and a low value of 0.002 is employed in parameters estimation for 20 areas in southern hemisphere highland. Using the updated algorithm of MPSO, it is found that a well-constrained crustal density for most of the studied regions is less than or equal to the mean crustal density (2550 kg·m−3) of the entire lunar highland. The relatively small crustal density is likely a ...

Source: Physics of the Earth and Planetary Interiors - May 15, 2019 Category: Physics Source Type: research