Many of the large-scale, plume-related features on Venus can be modeled by a buoyant viscous drop, or plume head, as it rises and spreads laterally below a free fluid surface. The drop has arbitrary density and viscosity contrast and begins as a sphere below the surface of fluid half space. The boundary integral method is used to solve for the motion of the plume head and for the topography, geoid, and stress at the fluid surface. As the plume approaches the surface, stresses in the fluid above it cause it to spread and become thin below the surface. During the spreading, the surface swell above evolves through various stages whose morphologies resemble several different plume-related features observed on Venus. When the plume head first approaches the surface, a high broad topographic dome develops, with a large geoid, and radial extensional deformation patterns. At later stages, the topography subsides and becomes plateau-like, the geoid to topography ratio (GTR) decreases, and the dominant stress pattern consists of a band of concentric extension surrounded by a band of concentric compression. We find that a low-viscosity model plume head (viscosity that is 0.1 times the mantle visocity) produces maximum topography that is 20% lower, and swell features which evolve faster, than for an isoviscous plume.

We compare model results with both large-scale highland swells, and smaller-scale features such as coronae and novae. The dome-shaped highlands with large GTRs such as Beta, Atla, and Western Eistla Regiones may be the result of early stage plume motion, while the flatter highlands such as Ovda and Thetis Regiones which have lower GTRs may be later stage features. Comparison of model results with GTR data indicates the highlands result from plume heads with initial diameters of about 1000 km. On a smaller scale, an evolutionary sequence may begin with novae (domes having radial extensional deformation), followed by features with radial and concentric deformation (such as arachnoids), and end with coronae (with mostly concentric deformation). The model predicts that the highlands evolve on a timescale of order 10 Ma, and the smaller-scale features evolve in a 100 Ma timescale.