Gravity/topography admittance inversion on Venus using niching genetic algorithms

We used niching genetic algorithms (NGAs) to invert localized Venus gravity/topography admittance over lowland regions Atalanta and Lavinia Planitiae, as well as volcanic rise Atla Regio for comparison. Assuming both top (topography) and bottom (mantle density anomalies) loads, we calculated theoretical admittance using thin elastic shell models. We inverted admittance for crustal thickness, elastic lithosphere thickness, mantle density anomaly thickness, and ratio (p(z)) of mantle density anomaly to topographic load. NGA inversion provides an efficient means of finding globally optimal and sub-optimal solutions. Error analyses of all three regions show that p(z) is a robust estimate; there is significant trade-off between elastic lithosphere and mantle anomaly thicknesses, while crustal thickness is ill-constrained. Optimal models suggest that mantle density anomalies are similar to+1 to 2% underlying lowland regions and similar to-3 to -4% underlying Atla Regio.