Did the lower crust flow on Mars?

Marin Clark

Associate Professor of Geology, Department of Earth and Environmental Sciences University of Michigan

Both Mars and Earth have major topographic features of roughly the same dimension and height. On Mars, the Tharsis rise represents a continent-scale area of approximately 5.5-6 km mean elevation. This topographic feature is second only to the planetary scale Martian dichotomy that separates the thick, elevated crust of the southern hemisphere from the thinner northern lowlands. By comparison, the Tibetan Plateau is Earth’s most extensive, high elevation landmass, also standing at about 5 km mean elevation and a few thousand kilometers in length. We’ve likely not seen anything else like Tibet in Phanerozoic time and it is generally thought that Tibet is large enough to influence both regional and global climate. Both Tibet and Tharsis contain roughly doubly thickened crust and a crustal mass of ~3 x 108 km3, although the Tharsis rise is primarily magmatic in origin and Tibet results from crustal thickening due to continental convergence.

The uplift of such wide, deformed regions depends on the material properties and deformation mechanisms of the deepest part of the lithospheric plates, which exert a primary control on how topography develops. Despite obvious differences in the geologic history and timescale between Mars and Earth, I find it striking that both planets’ most impressive topographic features are so similar in size, extent and regional slope. In particular, the southeast region of Tharsis (Thaumasia plateau) bears strong resemblance in structure and topography to eastern Tibet. High topography and thick crust of eastern Tibet produces long, low-gradient plateau margins that may be caused by the flow of weakened lower (or middle) crust. This has me thinking: if the lower crust flows in Tibet, did it do so on Mars?

Figure 1.  Smoothed elevation contour maps of Tharsis and Tibet. Contour interval is 1000 m and horizontal scales are identical in both maps. Grey shaded regions highlight lowland areas and local depressions. (A) Tharsis region of Mars, inset globe shows location. (B) Tibetan Plateau, inset globe shows location.
Figure 1. Smoothed elevation contour maps of Tharsis and Tibet. Contour interval is 1000 m and horizontal scales are identical in both maps. Grey shaded regions highlight lowland areas and local depressions. (A) Tharsis region of Mars, inset globe shows location. (B) Tibetan Plateau, inset globe shows location.

Here on Earth, the structure and tectonics community has engaged in a lively debate over lower crustal flow for the past 25 years. Lower crustal flow is compelling because it potentially explains the flatness of high standing plateaus [Bird, 1991], the absence of major subsidence above metamorphic core complexes [Block and Royden, 1990], regional compensation of highly extended terranes [Wernicke, 1990], and lateral transport of crust away from converging terranes [Royden et al., 1996]. Transport of deep crust over tens to thousands of kilometers decouples surface deformation from mantle motions. Further, flow potentially thickens or thins the crust preferentially in the lower crust relative to the upper crust. This process has consequences for relating surface deformation to isostatic elevation changes and the strength of the lithosphere. Decoupling of the surface from the mantle lithosphere challenges evidence supporting the block-like behavior of deforming continents. Because flow is not something we directly observe, the interpretations of structural, topographic and geophysical proxies that may indicate flow are not fully agreed upon. As with most intriguing scientific ideas, lower crustal flow is not without criticism from studies of deformation on both long and short times scales [Pollitz, et al., 2001; Rey et al., 2010; Lease et al., 2012]; and so the discussion on the existence of lower crustal flow as a process continues.

Eastern Tibet

The Himalayan/Tibetan orogen, the world’s highest landmass, continues to grow as a consequence of the continental collision of India with Eurasia approximately 50 million years ago and ongoing convergence to the present. During this time, the plateau developed to nearly twice the normal thickness of continental crust. The enormity and longevity of this active continental collision zone attract many studies in the field of continental dynamics. Minimal surface shortening across eastern Tibet suggests that crustal thickening occurs primarily by viscous flow of a weak, lower crust [Clark and Royden, 2000]. The uplift of a regionally continuous relict land surface and incision of major rivers define the spatial scales of crustal thickening and timing of uplift [Clark et al., 2005].

Lower crustal flow can be conceptualized as channel flow [e.g. Bird, 1991; Royden, 1996]. A channel form arises because a confined horizon in the crust becomes weakened, likely because ductile strength significantly decreases with increasing temperature. The base of a quartzo-feldspathic layer may become hot enough to fail in this manner. This layer may be sandwiched between a stronger, brittle upper layer, and a stronger, mafic lower-most crust or mantle lithosphere. Alternatively, small amounts of water or concentrated melt may act to weaken the middle or lower crust. Generally, this horizon is thought to be a few kilometers to a few tens of kilometers thick.

Figure 2.  Location maps and topographic profiles of low slope regions on Mars and Earth. (A) Regional topographic slope for the southeastern Tibetan plateau margin [Clark and Royden, 2000]. (B) Topography compared to channel flow model. Channel thickness is 15 km and flow initiates at 20 Ma. Variable channel viscosities are indicated on the profile. (C) Location map for southeastern Tibet profile. Thick black line indicates profile location. (D-E) Topography and channel flow models for southeastern Tharsis (Thaumasia plateau). Model parameters are identical to those used for southeastern Tibet in B [Clark and Royden, 2000]. Topographic profile from Montgomery et al. [2009]. (F) Location map for SE Tharsis profile. Thick black line indicates location of profile.
Figure 2. Location maps and topographic profiles of low slope regions on Mars and Earth. (A) Regional topographic slope for the southeastern Tibetan plateau margin [Clark and Royden, 2000]. (B) Topography compared to channel flow model. Channel thickness is 15 km and flow initiates at 20 Ma. Variable channel viscosities are indicated on the profile. (C) Location map for southeastern Tibet profile. Thick black line indicates profile location. (D-E) Topography and channel flow models for southeastern Tharsis (Thaumasia plateau). Model parameters are identical to those used for southeastern Tibet in B [Clark and Royden, 2000]. Topographic profile from Montgomery et al. [2009]. (F) Location map for SE Tharsis profile.  Thick black line indicates location of profile.
The flow of weak lower crust is driven by pressure gradients due to differences in crustal thickness. In the case of Tibet, these gradients exist from the high, thick plateau outward into the low elevation foreland. Forward modeling of the expansion of high topography in eastern Tibet, including dynamic topography induced by flow in the crust, leads to estimates of lower crustal viscosity based on a model of channel flow in 1- and 2- dimensions [Clark et al., 2005; Clark and Royden, 2000]. Using a model of Poiseuille flow (channel flow), we predict the necessary topographic gradient to drive flow for varying uniform channel viscosities, assuming a 15 km thick channel [Clark and Royden, 2000; Figure 2]. Material is added to the channel and flows eastward at a rate that builds a 5 km high plateau in 20 Myr. Observed topography in southeastern Tibet (also northeastern Tibet) matches model topography for Newtonian viscosity of 1018 Pa [Figure 2B]. One detail that is not well appreciated about these model results is that modeling the long, low gradient margins of eastern Tibet requires that the crust in these areas be weak prior to crustal thickening as opposed to weakening as a result of crustal thickening.

Southeastern Tharsis

The Tharsis Plateau is used here to describe the smoothed contour feature shown in Figure 1A – referring to the edifice above 2500 m that rests on an otherwise larger, elevated plain more generally referred to as the Tharsis Rise. Sustained tectonism and volcanism over much of Mars’ history was concentrated in Tharsis [Carr and Head, 2010]. During the Amazonian (< 3 Ga) Mars’ topographic history was associated with low rates of terrain building and low rates of surface degradation. So it is likely that the topographic feature we observe today, in particular southeastern Tharsis, is fossilized – a remnant from early Mars with higher heat flow and lower crustal strength [Zuber et al., 2000].

The mechanism and support of Tharsis are controversial, with debate primarily centered on either a sustained plume driven by stagnant lid convection in a one-plate planet [e.g. Zhong, 2009], or volcanic construction [e.g. Solomon and Head, 1982]. A major impact has also been suggested to have led to an early form of plate tectonics in which the structures and volcanism of Tharsis originated from subduction rollback [Yin, 2012]. In any case, topographic support of Tharsis likely originates from a combination of thermal buoyancy and crustal thickening due to magmatic accretion [Lowry and Zhong, 2003].

Tharsis volcanism initiated during the the Noachian (~ 4.0 Ga) and much of the general topography appears to have been in place since the end of this time period (~ 3.7 Ga) based on slope indicators such as valley networks and lava flows [Phillips et al., 2001]. Younger lavas of Hesperian (3.7 – 3.0 Ga) and Amazonian age (< 3.0 Ga) crop out at progressively higher elevations, respectively, making a broad trend of decreasing volcanic age from southeast to northwest [Scott and Tanaka, 1986; Tanaka et al., 1988] [Figure 1]. Noachian volcanics are concentrated in the southeast (Thaumasia Highlands and Coprates Rise). Plains of Hesperian lavas occur across the southeastern slope of Thaumasia and form the elevated regions around Valles Marineris and Claritas Fossae. Amazonian volcanic rocks are found at the highest altitude, on the ~ 5500 6500 m surface where >20 km volcanic edifices are also found (Tharsis Mons and Olympus Mons to the north).

Several existing interpretations exist for the structure, topography, and volcanism in Thaumasia. First, there is a strong radial pattern about the highest topography in Tharsis. Extension occurs at the highest altitudes (Syria Planum), trans-tension occurs parallel in the down slope direction (Caritas Fossae, Valles Marineris), and compression occurs perpendicular to the regional topographic slope at low elevations along the southeastern margin (Coprates Rise, Thaumasia highlands, Solis Dorsa). Interpretations of Thaumasia generally relate topographic loading to the radial structural pattern [Anguita et al., 2001]. Other ideas include a massive landslide of regolith including basaltic debris, salts and ice [Montgomery et al., 2009], and subduction rollback [Yin, 2012].

Potential lower crustal flow in SE Tharsis (Thaumasia)

Except near the volcanic edifices, topography decreases gently in all directions away from the central plateau. There are particularly low slopes dipping to the southeast (Thaumasia Plateau) where the elevated terrain nearly encompasses the lowest lying region of Solis Dorsa [Figure 1]. There is a regular, low-angle slope from the Syria Planum through high, smooth plains and into the Solis Dorsa. Extension in Syria Planum is observed at the top of this slope, the middle slope is undisturbed, and compression occurs at the lower slope (i.e. wrinkle ridges of the Solis Dorsa and Thaumasia Highlands. This structural pattern follows what might be expected for a mobile subsurface horizon, such as is seen in salt tectonics [Montgomery et al., 2009]. Potentially, this pattern could also arise from the presence of a weak crustal layer. Broadly  speaking, structural patterns in eastern Tibet also have extension at high altitudes with translation and crustal inflation on the eastern plateau margins [Royden et al., 2008].

The central slope and scale of Thaumasia closely mimics that of southeastern Tibet: a broad gentle slope that extends over 1500-2000 km. The Thaumasia slope does appear to match what would be predicted when applying the channel flow model of southeastern Tibet, so it at least passes some test. However, this may not be directly meaningful in terms of a lower crustal viscosity because timescales are less well known for this feature on Mars. Whether it took tens or hundreds of million years to form would dramatically affect any interpretation of channel viscosity.

One last note of interest concerns the flow-like field of the topography surrounding Thaumasia. The trans-tension in the incipient Valles Marineris area and the Claritas Fossae are superposed on the low-sloping topographic highs that surround the Solis Dorsa. Perhaps these are regions of even lower crustal viscosity that permit greater flow of material away from central Tharsis and into the surrounding region. Once again, we find a similarity with eastern Tibet, where the topography wraps around the low-lying, relatively undeformed Sichuan Basin [Figure 1]. Here, the Sichuan Basin is interpreted to be stronger than the crust that flows around it [Clark and Royden, 2000].


The scale of southeastern (and northeastern) Tibet in terms of height, length and regional slope is unique on Earth today. We observe nearly identical slopes in the Tharsis region of Mars. While constructed differently, both are continental scale regions of thickened crust that have (or had) the possibility to drive crustal flow in the presence of high gravitational potential energy. The Tharsis slope may not be unique on Mars; similar slopes can be found on at least portions of the dichotomy boundary and are proposed to be associated with the relaxation of ductile lower crust [Nimmo, 2005]. Smaller scale features on Mars may also point to a mobile lower crust, just as metamorphic core complexes do on Earth. For example, numerical modeling of extension in the Valles Marineris favors accommodation by lower crustal flow [Andrews Hanna, 2012], and crustal flow beneath impact basins early in Martian history may explain the lack of Moho relief and negative free-air gravity anomalies over these structures [Mohit and Phillips, 2007].

I favor the interpretation that special conditions existed for eastern Tibet that made the crust weak prior to crustal thickening. A pre-existing weak crust explains the unusual length of these low-sloping plateau margins. Because these conditions may be the pathological case for Earth, finding analogous features on other rocky planets may offer insight to crustal deformation of those planets as well as our own.


Much thanks to Jenny Briggs, Alison Duvall, Sean Gallen, Nathan Niemi, and Petr Yakovlev for comments.


Andrews-Hanna, J. C., 2012, The formation of Valles Marineris: 1. Tectonic architecture and the relative roles of extension and subsidence. J. Geophys. Res., v. 117, doi:10.1029/2011JE003953.

Anguita, F., Farelo, A.-F., Lopez, V., Mas, C., Munoz-Espadas, M.-J., Marquez, A., Ruiz, J., 2001, Tharsis dome, Mars: New evidence for Noachian-Hesperian thick-skin and Amazonian thin-skin tectonics. J. Geophys. Res., v. 106, doi:10.1029/2000JE001246.

Bird,P., 1991, Lateral extrusion of lower crust from under high topography, in the isostatic limit. J. Geophys. Res., v. 96, doi:10.1029/91JB00370.

Block, L. and Royden, L. H., 1990, Core complex geometries and regional scale flow in the lower crust. Tectonics, v. 9, doi:10.1029/TC009i004p00557.

Carr, M. H. and Head, J. W., 2010, Geologic history of Mars, Earth Planet. Sci. Lett., v. 294, doi:10.1016/j.epsl.2009.06.042.

Clark, M. K. and Royden, L. H., 2000, Topographic ooze: Building the eastern margin of Tibet by lower crustal flow, Geology, v. 28, n. 8, doi:10.1130/0091-7613(2000)<703:TOBTEM>2.0.CO;2.

Clark, M. K., House, M. A., Royden, L. H., Burchfiel, B. C., Whipple, K. X., Zhang, X., and Tang, W., 2005, Late Cenozoic Uplift of Southeastern Tibet, Geology, v. 33, n. 6, p. 525 – 528, doi: 10.1130/G21265.1.

Clark, M. K., Bush, J., and Royden, L. H., 2005, Dynamic topography produced by lower crustal flow against rheologic strength heterogeneities bordering the Tibetan Plateau, Geophys. J. Int., doi:10.1111/j.1365-246X.2005.02580.x.

Lease, R., Burbank, D. W., Zhang, H. P., Liu, J. H., and Yuan, D. Y., 2012, Cenozoic shortening budget for the northeastern edge of the Tibetan Plateau: Is lower crustal flow necessary? Tectonics, v. 31, doi: 10.1029/2011TC003066.

Lowry, A. and Zhong, S., 2003, Surface versus internal loading of the Tharsis rise, Mars. J. Geophys. Res. v. 108, doi: 10.1029/2003JE002111.

Mohit, P. S., and Phillips, R. J., 2007, Viscous relaxation on early Mars: A study of ancient impact basins. Geophys. Res. Lett., v. 34, doi:10.1029/2007GL031252.

Montgomery, D., Som, S. M., Jackson, M. P. A., Schreiber, B. C., Gillespie, A.R., Adams, J. B., 2009, Continental-scale salt tectonics on Mars and the origin of Valles Marineris and associated outflow channels. GSA Bulletin, v. 121, doi: 10.1130/B26307.

Nimmo, F., 2005, Tectonic consequences of Martian dichotomy modification by lower-crustal flow and erosion. Geology, v. 33, doi:10.1130/G21342.

Phillips, R. J. and 10 others, 2001, Ancient Geodynamics and Global-scale Hydrology on Mars. Science, v. 291, doi:10.1126/science.1058701.

Pollitz, F. F., Wicks, C., Thatcher, W., 2001, Mantle flow beneath a continental strike-slip fault: Postseismic deformation after the 1999 Hector Mine earthquake. Science, v.293, doi:10.1126/science.1061361.

Rey, P., Teyssier, C., and Whitney, D., 2010, Limit of channel flow in orogenic plateau, Lithosphere, v. 2, doi: 10.1130/L114.1.

Royden, L., 1996, Coupling and decoupling of crust and mantle in convergent orogens: Implications for strain partitioning in the crust. J. Geophys. Res., v. 101, doi: 10.1029/96JB00951.

Royden, L., H., Burchfiel, B. C., van der Hilst, R. D., 2008, The geologic evolution of the Tibetan Plateau, Science, v. 321, doi: 10.1126/science.1155371.

Scott, D.H., Tanaka, K.L., 1986, Geologic map of the western equatorial region of Mars. U.S. Geological Survey Misc. Inv. Map I-1802-A.

Solomon, S., and Head, J. W., 1982, Evolution of the Tharsis Province of Mars: The importance of heterogeneous lithospheric thickness and volcanic construction. J. Geophys. Res., v. 87, doi: 10.1029/JB087iB12p09755.

Tanaka, K.L., Isbell, N.K., Scott, D.H., Greeley, R., Guest, J.E., 1988. The resurfacing History of Mars: a synthesis of digitized, Viking-based geology. Proc. Lunar Planet. Sci. Conf. 18, 665678.

Wernicke, B., 1990, The fluid crustal layer and its implications for continental dynamics, Exposed Cross-Sections of the Continental Crust. NATO Adv. Stud. Inst., Ser. C, Mathematical and Physical Sciences 317: 509-544.

Yin, A., 2012, An episodic slab-rollback model for the origin of the Tharsis rise on Mars: Implications for initiation of local plate subduction and final unification of a kinematically linked global plate-tectonic network on Earth. Lithosphere, v. 4, doi: 10.1130/L195.1.

Zhong, S., 2009, Migration of Tharsis volcanism on Mars caused by differential rotation of the lithosphere. Nature Geoscience, v. 2, doi:10.1038/NGEO392.

Zuber, M. T., and 14 others, 2000, Internal structure and early thermal evolution of Mars from Mars Global Surveyor Topography and Gravity, v. 287, doi:10.1126/science.287.5459.1788.

One thought on “Did the lower crust flow on Mars?

  1. assuming the channel flow as stated in the article, what is the mean volumetric heating rate due to viscous dissipation within the channel? In other words, assuming a const viscosity of say 10^18 Pa s or whatever and given the dimensions of the channel and mean velocity, what is the J/kg s of heat generated by viscous dissipation. this is a simple calculation. How does THIS volumetric heating rate (J/kg s) compare with in situ radioactive heating rate assuming some reasonable bulk compos of U,Th and K?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s