Roadmap | The review process | Home
   Impacts on Mars
Impact-induced Martian mantle plumes: Origin of Tharsis

C.C. Reese & V.S. Solomatov

Department of Earth and Planetary Sciences, Campus Box 1169, 1 Brookings Drive, Washington University, Saint Louis, MO 63130-4899 &


Tharsis is one of the most prominent features on Mars. Volcanism and tectonism associated with the plateau by far exceed the levels of activity in other areas of the planet. This monopolar distribution of tectonism and volcanism led to the suggestion that the planform of mantle convection on Mars is dominated by a single, long-lived, thermal plume originating at the core-mantle boundary similar to a terrestrial plume but much larger (Hartmann, 1973; Carr, 1974; Phillips & Ivins, 1979). Deep mantle mineralogical phase transformations are capable of stabilizing a one-plume pattern of convection (Weinstein, 1995; Harder & Christensen, 1996; Breuer et al., 1998; Spohn et al., 1998; Harder, 2000). However, although the conventional plume model explains some features of Tharsis, there are both observational and theoretical reasons to consider alternatives.

  1. The plume model has not reproduced Tharsis development on timescales consistent with observations (Banerdt & Golombek, 2000; Phillips et al., 2001; Johnson & Phillips, 2003; Zuber, 2001).
  2. The plume model only accounts for dynamic support of Tharsis. Crustal thickening (Zuber et al., 2000) and observations of layered sequences of rock in the walls of Valles Marineris (McEwen et al., 1999) suggest that constructional volcanism is a major contributor to Tharsis elevation (Solomon & Head, 1982). Also, a recent analysis suggests that Tharsis is predominantly supported by crustal thickening and lithospheric flexure while a thermal plume would contribute only a fraction to the present-day topography and areoid (Zhong, 2002; Lowry & Zhong, 2003; Zhong & Roberts, 2003).
  3. The plume hypothesis implies an actively convecting mantle and a sufficiently large heat flux from the Martian core. However, in the absence of plate tectonics, Martian mantle convection can be very sluggish (Grasset & Parmentier, 1998; Reese et al., 1998; 2002; Solomatov & Moresi, 2000). Scaling relationships (Solomatov & Moresi, 2000) for olivine rheology (Karato & Wu, 1993) indicate that a wet mantle convects only in a narrow subsolidus temperature range while a dry mantle is convectively stable even above melting temperatures. Widespread melting would, in turn, differentiate heat producing elements (Turcotte, 1989; Spohn, 1991; Schubert et al., 1992), increase viscosity (Karato, 1986), and produce compositional stratification (Zaranek & Parmentier, 2004) all of which tend to suppress convection.
  4. Rapid development of a conventional thermal plume at the core-mantle boundary after planetary formation could be problematic because of mantle heating. Although a transient episode of early surface recycling (Sleep, 1994) can facilitate mantle cooling, the duration might be short (Tajika & Sasaki, 1996) and the subsequent transition to stagnant lid convection would result in mantle heating shutting off any core heat flux and associated plume activity (Nimmo & Stevenson, 2000). The liquid state of the Martian core recently inferred from solar tidal deformation is consistent with such a possibility (Yoder et al., 2003).
  5. Geochemical evidence also argues against vigorous mantle mixing. Martian meteorite Hf/W and Sm/Nd systematics suggest that core and crust formation were contemporaneous and occured within ~30 Myr of planet formation (Harper et al., 1995; Lee & Halliday, 1997; Halliday et al., 2001; Kleine et al., 2004) and also indicate limited large-scale convective mixing of the upper mantle. Survival of isotopic heterogeneity in the Martian upper mantle since the time of core formation and early vigorous convection are difficult to reconcile (Zuber, 2001). Large variations in Sm/Nd and Lu/Hf ratios among shergottites also suggest a heterogeneous mantle which retains an isotopic signature of initial differentiation (Albarede et al., 2000). Finally, cessation of Xe degassing early in Mars history suggests that large scale mantle magmatism, and presumably vigorous convection, have been limited since ~300 Myr (Marty & Marti, 2002) after formation.

Alternative hypothesis: Impact plume

An alternative hypothesis is that Tharsis could have experienced a large impact early in Martian history (Schultz & Glicken, 1979; Schultz, 1984; Schultz & Frey, 1990). Recent studies suggest that this hypothesis is plausible from a geodynamical point of view (Reese et al., 2002; 2004). Large impacts at the end of planetary formation are a statistically inevitable consequence of accretional dynamics (Wetherill, 1985,1990; Weidenschilling et al., 1997; Chambers & Wetherill, 1998; Agnor et al., 1999) and result in melting beyond the excavation zone producing an intact melt region with radius of the order of several impactor radii (Melosh, 1990; Tonks & Melosh, 1992;1993; Pierazzo et al., 1997).

Three major processes following impact-generation of an intact melt volume are crystallization, isostatic adjustment and melt percolation. Melt distribution in the impact-heated region varies strongly, from superliquidus conditions near the surface to subsolidus conditions deeper in the mantle. Cooling and crystallization of regions with the highest degree of melting to about 60% crystal fraction can be very fast, on the scale of 1000 years (Solomatov, 2000). For Mars, isostatic adjustment of the region with melt fraction varying from 40% to 0% and melt percolation through the partially crystallized matrix are much slower processes (Reese et al., 2004).

A qualitative description of the process is as follows. Fast initial crystallization produces a region with melt fraction varying from about 40% near the surface to 0% several hundred kilometers below the surface. Because of its positive buoyancy, this region floats up. Hot, partially molten material rises and spreads out on the surface of the planet and is replaced by colder, less molten material from below. All processes slow down as the melt fraction and driving density difference decrease and the viscosity increases.

At some point, the dynamics merge with the solid state convective evolution of the planet. Further upwelling results in decompression melting and production of compositional buoyancy (through mantle depletion and melt retention). The end result is a long-lived (compared to initial dynamical timescales) localized mantle upwelling and associated magmatism – an "impact plume". Production of Tharsis by an impact plume requires neither globally occurring convection nor generation of conventional plumes at the core-mantle boundary.

Numerical simulation

A fully three-dimensional spherical shell simulation was used to study thermal, magmatic, and compositional aspects of Martian impact-plume evolution (Reese et al., 2004). A strongly temperature-dependent exponential viscosity law and rigid upper surface boundary condition is assumed ensuring stagnant lid convection (Solomatov & Moresi, 2000). There is no radiogenic internal heating. The average, pre-impact, mantle temperature is 1800 K with a fixed surface temperature of 220 K. The initial intact melt region is hemispherical with radius R related to impactor size and velocity (see, e.g., Tonks & Melosh, 1993). For the case presented here, R=1500 km, corresponding to an impactor radius of ~900 km (Reese et al., 2004). After fast initial crystallization it is assumed that this region has a uniform thermal anomaly of 300 K. Because any core heat flux was probably short lived (Nimmo & Stevenson, 2000), the bottom boundary is assumed to be insulating throughout evolution. The initial cold boundary layer thickness is resolution-limited to ~300 km. For these initial conditions, there is a partially molten sublithospheric layer ~500 km thick. This layer is compositionally buoyant due to the presence of melt and the fractional density difference is assumed to be ~2%. Subsequent decompression melting due to upwelling mantle results in the same compositional buoyancy (via mantle depletion and melt retention). Only dry melting below 8 GPa is considered.


The initial upwelling velocity decays with time from a maximum that scales inversely with interior viscosity. Subsequent evolution depends on interior viscosity. For high interior viscosity (1023 Pa s), thermal evolution is dominated by conduction. The stagnant lid thickens conductively but remains convectively stable. Interior velocities are very low and approximately constant throughout evolution. For low interior viscosity (1021 Pa s), lid thickening leads to development of small scale instabilities at the lid base. The mantle heat flux and velocity decay with time as the spherical shell cools (Figure 1).


Figure 1. Convective mantle heat flux (left) and maximum interior velocity (right) as a function of time (heavy lines). For the high interior viscosity case (1023 Pa s), the heat flux due to conductive cooling of a half space is also shown (thin line). For the low interior viscosity case (1021 Pa s), the heat flux and cold plume velocity (thin lines) are shown for a parameterized convection calculation based on time-dependent stagnant lid convection scaling laws (Solomatov & Moresi, 2000).

The duration of the magmatic episode depends on interior mantle viscosity. Buoyant, depleted mantle which has undergone melting spreads out at the bottom of the viscous lid. For all cases, melt production decays with time from an initial maximum to very low levels. As interior viscosity decreases, the decay rate and total melt volume decrease and increase, respectively. The surface distribution of volcanism is directly related to the radially integrated melt fraction. Since magma transport to the surface in the stagnant lid regime is poorly understood, no attempt is made to discriminate between intrusive and extrusive flux. Instead, the total melt volume per unit surface area (crustal thickness if all melt contributes to crustal growth) is calculated. In all cases, widespread volcanism is suppressed by lid thickening, and volcanism is concentrated in the impact plume region. For low interior viscosities, there are two spatial scales in the distribution. The outer scale is that of the impact plume. The inner scale is associated with localized, small scale convection within the plume (Figure 2).

Figure 2. Magmatic evolution of impact plume. Magmatic rate as a function of time (left). Final spatial distribution of the melt volume per unit surface area (center). The grid line and frame intervals are 15. Final distribution of mantle which has undergone partial melting along a cross section passing through the impact plume axis (right). Color (white through red) indicates volume fraction of material which has undergone partial melting and the solid line indicates the radius of the initial intact melt region.


Scaling relationships suggest that for sufficiently large impacts the transient crater is formed completely within the melt region making retention of a final crater unlikely (Tonks & Melosh, 1992). For smaller impacts, elimination of initial crater structure formed outside the melt region can occur during the processes of isostatic adjustment and melt extrusion onto the surface. The impact plume mechanism is capable of producing additional magmatism associated with the extended period of upwelling and decompression melting. In this case, extensive magmatism would not only erase transient crater structure but might result in the development of a large igneous province.

Mars Global Surveyor (MGS) topography (Smith et al., 1999) plus material contained within a depression due to Tharsis loading and lithospheric flexure correspond to ~3 x 103 km3 of igneous material (Phillips et al., 2001). The melt volume associated with the initial impact can be compared to the decompression melt volume produced by the upwelling impact plume. For the low interior viscosity (1021 Pa s) case, the impact plume melt volume is of the order of the impact-generated, retained melt volume, ~109 km3 (Reese et al., 2004). It seems likely that, in this case, the impact plume mechanism can contribute significantly to the total extrusive melt volume. Large-scale melting ceases by the end of the Noachian consistent with timing of Tharsis formation. Perhaps a large impactor, via extrusion of initial shock melt and impact plume decompression melting, has the potential to result in development of a large volcanic construct such as Tharsis.


  • Agnor, C.B., R.M. Canup, H.F. Levison, On the character and consequences of large impacts in the late stage of terrestrial planet formation, Icarus, 142, 219-237, 1999.
  • Albarede, F., J. Blichert-Toft, J.D. Vervoort, J.D. Gleason, and M. Rosing, Hf-Nd isotope evidence for a transient dynamic regime in the early terrestrial mantle, Nature, 404, 488-490, 2000.
  • Banerdt, W.B., and M.P. Golombek, Tectonics of the Tharsis region of Mars: Insights from MGS topography and gravity, Lunar Planet. Sci. Conf., 31, #2038, 2000.
  • Breuer. D., D.A. Yuen, T. Spohn, and S. Zhang, Three dimensional models of Martian mantle convection with phase transitions, Geophys. Res. Lett., 25, 229-232, 1998.
  • Carr, M.H., Tectonism and volcanism of the Tharsis region of Mars, J. Geophys. Res., 79, 3943-3949, 1974.
  • Chambers, J.E., and G.W. Wetherill, Making the terrestrial planets: N-body integrations of planetary embryos in three dimensions, Icarus, 136, 304-327, 1998.
  • Grasset, O., and E.M. Parmentier, Thermal convection in a volumetrically heated, infinite Prandtl number fluid with strongly temperature dependent viscosity: Implications for planetary evolution, J. Geophys. Res., 103, 18171-18181, 1998.
  • Halliday, A.N., H. Wanke, J.-L. Birck, and R.N. Clayton, The accretion, composition, and early differentiation of Mars, Earth, Moon, and Planets, 96, 197-230, 2001.
  • Harder, H., Mantle convection and the dynamic geoid of Mars, Geophys. Res. Lett., 27, 301-304, 2000.
  • Harder, H., and U.R. Christensen, A one-plume model of Martian mantle convection, Nature, 380, 507-509, 1996.
  • Harper, C.L., L.E. Nyquist, B. Bensal, H. Wiesmann, and C.-Y. Shih, Rapid accretion and early differentiation of Mars indicated by 142Nd/144Nd in SNC meteorites, Science, 267, 213-217, 1995.
  • Hartmann, W.K., Martian surface and crust: Review and synthesis, Icarus, 19, 550-575, 1973.
  • Johnson, C.L., and R.J. Phillips, Constraints on the evolution of the Tharsis region of Mars, Lunar Planet. Sci. Conf., 33, #1360, 2003.
  • Karato, S.-I., Does partial melting reduce the creep strength of the upper mantle?, Nature, 319, 8151-8176, 1986.
  • Karato, S.-I., and P. Wu, Rheology of the upper mantle: A synthesis, Science, 260, 771-778, 1993.
  • Kleine, T, K. Mezger, C. Munker, H. Palme, and A. Bischoff, 182Hf – 182W isotope systematics of chondrites, eucrites and Martian meteorites: Chrononlogy of core formation and early diffentiation of Vesta and Mars, Geochim. et Cosmochim. Acta, 68, 2935-2946, 2004.
  • Lee, D.C., and A.N. Halliday, Core formation on Mars and differentiated asteroids, Nature, 388, 854-857, 1997.
  • Lowry, A. R., and S. Zhong, Surface versus internal loading of the Tharsis rise, Mars, J. Geophys. Res., 108, 10.1029/2003JE002111, 2003.
  • Marty, B., and K. Marti, Signatures of early differentiation of Mars, Earth Planet. Sci. Lett., 196, 251-263, 2002.
  • McEwen, A.S., M.C. Malin, M.H. Carr, and W.K. Hartmann, Voluminous volcanism on early Mars revealed in Valles Marineris,
    Nature, 397, 584-586, 1999.
  • Melosh, H.J., Giant impacts and thermal state of the early Earth, in Origin of the Earth, edited by H.E. Newsom and J.H. Jones, Oxford Univ. Press, New York, 1990.
  • Nimmo, F., and D. Stevenson, The influence of early plate tectonics on the thermal evolution and magnetic field of Mars, J. Geophys. Res., 105, 11969-11979, 2000.
  • Pierazzo E., A. M. Vickery, and H. J. Melosh, A reevaluation of impact melt production, Icarus, 127, 408-423, 1997.
  • Phillips, R.J., and E.R. Ivins, Geophysical observations pertaining to solid-state convection in the terrestrial planets, Phys. Earth Planet. Inter., 19, 107-148, 1979.
  • Phillips, R.J., M.T. Zuber, S.C. Solomon, M.P. Golombek, B.M. Jakosky, W.B. Banerdt, D.E. Smith, R.M.E. Williams, B.M. Hynek, O. Aharonson, and S.A. Hauck, Ancient Geodynamics and global scale hydrology on Mars, Science, 291, 2587-2591, 2001.
  • Reese, C.C., V.S. Solomatov, and L.-N. Moresi, Heat transport efficiency for stagnant lid convection with dislocation viscosity: Application to Mars and Venus, J. Geophys. Res., 103, 13,643-13,657, 1998
  • Reese, C.C., V.S. Solomatov, and J.R. Baumgardner, Survivial of impact induced thermal anomalies in the Martian mantle, J. Geophys. Res., 107, 10.1029/2000JE001474, 2002.
  • Schubert, G., S.C. Solomon, D.L. Turcotte, M.J. Drake, and N.H. Sleep, Origin and thermal evolution of Mars, in Mars, edited by H.H. Kieffer, B.M. Jakosky, C.W. Snyder, and M.S. Matthews, pp. 147-183, Univ. of Ariz. Press, Tucson, 1992.
  • Schultz, P.H., Impact basin control of volcanic and tectonic provinces on Mars, Lunar Planet. Sci. Conf., 14, 728-729, 1984.
  • Schultz, P.H., and H. Glicken, Impact crater and basin control of igneous processes on Mars, J. Geophys. Res., 84, 8033-8047, 1979.
  • Schultz, R.A., and H.V. Frey, A new survey of large multiring impact basins on Mars, J. Geophys. Res., 95, 14175-14189, 1990.
  • Sleep, N.H., Martian plate tectonics, J. Geophys. Res., 99, 5639-5655, 1994.
  • Smith, D.E., M.T. Zuber, S.C. Solomon, R.J. Phillips, J.W. Head, J.B. Garvin, W.B Banerdt, D.O. Muhleman, G.H. Pettengill, G.A. Neumann, F.G. Lemoine, J.B. Abshire, O. Aharonson, C.D. Brown, S.A. Hauck, A.B. Ivanov, P.J. McGovern, H.J. Zwally, T.C. Duxbury, The global topography of Mars and implications for surface evolution, Science, 284, 1495-1503, 1999.
  • Solomatov, V.S., Fluid dynamics of magma oceans, in Origin of the Earth and Moon, edited by R. Canup and K. Righter, Univ. Arizona Press, Tucson, 2000.
  • Solomatov, V.S., and L.-N. Moresi, Time dependent stagnant lid convection on the Earth and other terrestrial planets, J. Geophys. Res., 105, 21795-21817, 2000.
  • Solomon, S.C., and J.W. Head, Evolution of the Tharsis province of Mars: The importance of heterogeneous lithospheric thickness and volcanic construction, J. Geophys. Res., 82, 9755-9774, 1982.
  • Spohn, T., Mantle differentiation and thermal evolution of Mars, Mercury, and Venus, Icarus, 90, 222-236, 1991.
  • Spohn, T., F. Sohl, and D. Breuer, Mars, Astr. and Astrophys. Rev., 8, 181-235, 1998.
  • Tajika, E., and S. Sasaki, Magma generation on Mars constrained from 40Ar degassing model, J. Geophys. Res., 101, 7543-7554, 1996.
  • Tonks, W.B., and H.J. Melosh, Core formation by giant impacts, Icarus, 100, 326-346, 1992.
  • Tonks, W.B., and H.J. Melosh, Magma ocean formation due to giant impacts, J. Geophys. Res., 98, 5319-5333, 1993.
  • Trompert, R.A. and U. Hansen, On the Rayleigh number dependence of convection with a strongly temperature-dependent viscosity, Phys. Fluids, 10, 351-360, 1998.
  • Turcotte, D.L., Thermal evolution of Mars and Venus including irreversible fractionation, Lunar Planet. Sci. Conf., 20, 1138-1139, 1989.
  • Weidenschilling, S. J., D. Spaute, D. R. Davis, F. Marzari, and K. Ohtsuki, Accretional evolution of a planetesimal swarm. 2. The terrestrial zone, Icarus, 126, 429-455, 1997.
  • Weinstein, S.A., The effects of a deep mantle endothermic phase change on the structure of thermal convection in silicate planets, J. Geophys. Res., 100, 11,719-11,728, 1995.
  • Wetherill G.W., Occurrence of giant impacts during the growth of the terrestrial planets, Science, 228, 877-879, 1985.
  • Wetherill G.W., Formation of the Earth, Annu. Rev. Earth Planet. Sci., 18, 205-256, 1990.
  • Yoder, C. F., A. S. Konopliv, D. N. Yuan, E. M. Standish, and W. M. Folkner, Fluid core size of Mars from detection of the Solar tide, Science, 300, 299-303, 2003.
  • Zaranek, S.E., and M.E. Parmentier, Convective cooling of an initally stably stratified fluid with temperature dependent viscosity: Implications for the role of solid state convection in planetary evolution, J. Geophys. Res., 109, B03409, doi:10.1029/2003JB002462, 2004.
  • Zhong, S., Effects of lithosphere on the long wavelength gravity anomalies and their implications for the formation of the Tharsis rise on Mars, J. Geophys. Res., 107, 10.1029/2000JE001589, 2002.
  • Zhong, S., and J. H. Roberts, On the support of the Tharsis Rise on Mars, Earth PLanet. Sci. Lett., 214, 1-9, 2003.
  • Zuber, M.T., S.C. Solomon, R.J. Phillips, D.E. Smith, G.L. Tyler, O. Aharonson, G. Balmino, W.B. Banerdt, J.W. Head, C.L. Johnson, F.G. Lemoine, P.J. McGovern, G.A. Neumann, D.D. Rowlands, S. Zhong, Internal structure and early thermal evolution of Mars from Mars Global Surveyor topography and gravity, Science, 287, 1788-1793, 2000.
  • Zuber, M.T.,The crust and mantle of Mars, Nature, 412, 220-227, 2001.
last updated November 29th, 2004