Interpretations for the formation of the Tharsis Rise

---

Interplanetary analogies

The Tharsis Rise covers a surface area of 2 x 10⁷ km², which is one order of magnitude larger than Earth's largest large igneous province, the Ontong-Java Plateau.

The circum-Chryse outflow channel systems draw on a drainage basin across the Tharsis Rise that is approximately the size of Europe.

The lava flows of the Tharsis Montes, Alba Patera, and Olympus Mons cover up to 18 million square kilometers of area, more than twice the area of the lower 48 United States.

If superposed over the continental United States, the Valles Marineris would stretch, from end to end, between Los Angeles, California, and New York, New York.

Features on the Martian surface, particularly Alba Patera, have been analogized to the coronae of Venus. These features have collectively been analogized to terrestrial igneous features, including the High Arctic Large Igneous Province; the dikes of the Siberian Traps of Russia, the Protogine Zone of Sweden, the Brazilian dike swarms of the Paraná and Etendeka traps; the Franklin traps of northern Canada; and a number of possibly circumferential giant dike swarms observed on the North China Craton.

The rates of volcanism during the edifice-building period of Olympus Mons' history corresponds strikingly with the long-term average volcanic eruption rates of terrestrial hotspots. Both result in morphologically parallel shield volcano structures.

Holistic preconsiderations

Mantle and core processes

On Mars, Wänke and Dreibus (1994) predicted a "hot" model of an early Mars with respective mantle and core temperatures of 1973K and 2250K, by which the surface of the planet would have achieved temperatures just above the solidus of a hypothetical magma ocean on Mars until 4.5 Ga, after which the crystallization of the surface into the solid phase would have commenced. Alternatively, Lodders and Fegley (1997) counterpropose a different model in which the rapid decay of potassium-40 generates a very high initial heat flux over an additional 400 Myr ahead of the 1994 model before contractional stresses set in.

Martian core dynamo

Mars is no longer believed to contain an active core dynamo like the one currently active on Earth. Relict magnetized terrains exist when magnetic materials in situ in igneous terrains record the direction of magnetic field lines in their area when a global paleomagnetic field did exist; on Mars, magnetite and titanomagnetite are regarded as the most likely minerals accounting for the magnetization signatures observed in these terrains. Magnetized units are observed on the Tharsis Rise within the Claritas Rise and within the Thaumasia Highlands, suggesting that the shutdown of the dynamo occurred prior to the formation of the Tharsis Rise, coeval or following the formation of these two ancient provinces. Other magnetized terrains are observed across the southern cratered plains outside the major impact basins and volcanic rises. The strongest fields associated with the ancient magnetic nature of the Martian core on Noachian terrains to the southwest of the Tharsis Rise in Terra Cimmeria and Terra Sirenum, reaching an estimated intensity of 12 μT (around a quarter of the field strength of an Earth-like global magnetic field). Because neither the Tharsis Rise (excepting the Claritas Rise and the Thaumasia highlands) nor the Elysium Rise contain this magnetization signature, and none of the largest clear impact basins do either (Hellas Planitia, Argyre Planitia, Isidis Planitia), these events are generally understood to have occurred after the shutdown of the dynamo.

The transition from non-magnetized terrain in the western Valles Marineris area (Syria Planum, Solis Planum, Sinai Planum) to the magnetized ancient plateau terrains in the eastern Valles Marineris (Xanthe Planum, Lunae Planum) suggests that the lava flows that constructed the Syria Planum region erased the signature that would have typically been visible in the western extent of the Rise. This same phenomenon is observed at the peripheries of the aureole deposits of Olympus Mons as they transition into more ancient adjacent plains units.

True polar wander

The position of Tharsis is believed to have influenced the true polar wander of Mars, re-stabilizing the planet's rotation in a way that centered the Tharsis Rise on the planet's equator during the Noachian and early Hesperian periods. The growth of Tharsis is proposed to have induced an additional axial tilt of 20 degrees from azimuth. Any climate-driven geomorphology predating the Amazonian - notably including valley networks - was directly affected by the re-equilibriation of the Martian surface's positioning. Some authors have proposed that the crustal dichotomy was originally parallel to the equator, and where the north paleopole was situated underneath the Scandia Colles in the northern lowlands and the south pole in Malea Planum, near Pityusa Patera. Evidence of ice retreat has been observed and proposed in both of these areas, possibly associated with the migration of the poles during the construction of the Tharsis edifice. In the case of a wet and warm Mars, precipitation a paleotropical region of the planet could have sourced water that formed the Martian valley networks. However, the predicted distribution of putatively precipitation-sourced valley networks in this tropical band falters between Hellas Planitia and Argyre Planitia and in Arabia Terra, suggesting that the formation of valley networks is concurrent with the formation of the Tharsis Rise as the poles were migrating to their current locations.

Plate tectonics

Researchers have constrained two types of plate-tectonic regimes in the planetary sciences - that of "active-lid" convection, in which the lithosphere buckles and is systematically refolded into the mantle; or a "stagnant-lid" convective situation, where the lithosphere is strong enough to resist the action of underlying mantle convection and remains immobilized. The stresses on the lithosphere, therefore, must exceed a certain critical value for it to become overpowered by convective mantle activity, and its yield stress is controlled by the lithosphere's elastic thickness and its friction coefficient (and affected by temperature and water content), yielding the active-lid case. "Active-lid" convection as a mechanism to drive plate tectonics driven by mantle convection has only been found to occur on Earth; all other terrestrial bodies in the Solar System, including Mars, are understood to fall into this second category.

Some authors observing systematic, global-scale offsets in magnetic striping on Terra Meridiani have suggested the presence of two great faults (regional-scale transform faults) cross-cutting the southern highlands terrains in a generally north-south sense. The Valles Marineris has also been observed as perpendicular to these great faults, which is coincident with relative orientations of strike-slip and extensional tectonic features within transform faults observed on Earth. This has been offered as a means of supporting a global plate tectonics argument.

Within the Tharsis Rise, the Tharsis Montes are aligned along a common axis, as are Olympus Mons and Alba Patera. In an analogy to the formation of the Hawaiian-Emperor seamount chain in the Pacific Ocean, some authors have proposed that the drift of a tectonic plate underneath Tharsis could possibly explain the relative orientations of these volcanic features.

The putative Tharsis Plate has been analogized to Earth by means of relating the Martian Thaumasia Plateau to the Colorado Plateau in the southwestern United States and the Loess Plateau and Ordos Basin of northwestern China. All three landforms are topographically comparable, with high-standing mountain ridges bounding a central depression transitioning into a flat plains unit in the far field. Both the Colorado and Loess Plateaux formed as a result of a sustained, magmatic uplift associated with subduction - a plate tectonic phenomenon. Some researchers have proposed that the putative Tharsis Plate is bounded to the south by the Thaumasia Highlands and the southeast by the Coprates Rise, extending in parallel to the basin-and-range topographical province of the South Tharsis Ridge Belt to the southwest of the Tharsis Province, and encompassing Olympus Mons to the north.

Composition of the Tharsis Rise

Researchers mapping Olympus Mons and Alba Patera have identified relict summit lava flows that are occasionally truncated or uphill-flowing. Some have proposed that an inflation event late in the formational history of these volcanoes altered their associated stress fields, tectonically causing these inconsistencies. No evidence of such an event, however, has been observed on either the Arsia or Pavonis Montes. Furthermore, researchers have not conclusively established whether a similar event affected the summit of Ascraeus Mons. The inflation of the summit is hypothesized to occur after the summit caldera has collapsed due to a prior drainage of its magma chamber, but is later reinjected with molten material at a later point.

The basement rocks of Claritas Fossae and the Thaumasia Highlands were originally believed to be composed of basalt, norite, and anorthosite when these areas were first resolved on Mariner 9 imagery, but these initial hypotheses were later walked back due to difficulties in interpretation arising from the intensity of fracturing affecting these regions. Instead, a more generalized claim that these basement rocks were a highly resistant material was made.

Researchers assessing the composition of the exposed units on the Martian volcanic rises found that the volcanic provinces of Tharsis and Elysium are spectrally very similar, and that each province's Hesperian volcanic units have comparable mafic mineral compositions to the more recent Amazonian volcanics but starkly different from the older Noachian igneous materials. On Tharsis, most of the dust-free sites where spectral signatures registered were on the peripheries of features on the Rise, notably including Daedalia Planum.

The oldest rocks on Mars - thought to arise from the solidification of the original magma ocean, a direct indicator of composition after planetary differentiation - are understood to be enriched in low-calcium pyroxenes (among other mafic minerals), as was the case for the norites recovered from the lunar primordial crust. On Mars, low-calcium pyroxenes are a major phase of the ALH84001 meteorite (a 4.09 Gy old orthopyroxenite), and has been remotely sensed in the lava terrains of the Noachian southern highlands.

Bulk composition of the eastern Tharsis Rise

Spectral data from the Gamma Ray Spectrometer on the 2001 Mars Odyssey has been used to identify a preponderance of Th, K and Fe in the ancient Noachian-aged magnetized terrains predating the formation of the Tharsis Rise (Thaumasia Planum, the Coprates Rise, the Thaumasia Highlands, and the ancient transition zone to the southeast of the Rise leading up to Argyre Planitia). These three elements are not observed in the more recent Hesperian-aged high plain provinces of the Thaumasia Plateau at Syria Planum, Solis Planum and Sinai Planum.

The entire Thaumasia Plateau is Si-enriched and has an abundance of Al relative to the average composition of the Martian crust, although crustal aluminum composition tends to decrease moving into the older peripheral Noachian terrains. These terrains, including the more recent Solis and Sinai Plana, have additionally been found to be Cl-depleted (against the Cl-H₂O ratio), although this depletion diminishes (and, thus, H₂O depletions increase) into the older peripheral Noachian terrains to the southeast. Uniquely at regional scales on the Martian crust, the older Noachian terrains of Thaumasia Planum, the Coprates Rise, the Thaumasia highlands, and the plains leading up to the Argyre impact zone have been found to be Ca-depleted. This trend is reversed for Th, which is most enriched in the peripheral Noachian terrains rather than the Hesperian high volcanic plains. The dependence of the K/Th ratio on this transitional shift between Hesperian lava flows and Noachian basement terrains strongly suggests that the origin of the Thaumasia Plateau were principally magmatic in origin. The transition in calcium compositions across the region, in light of a magmatic hypothesis, highlights a possible shift in the nature of volcanism forming the Thaumasia region – with low-Ca pyroxenes dominating the older peripheral Noachian terrains, and high-Ca pyroxenes in the Hesperian high volcanic plains.

Although sulfates abound across the Thaumasia region, researchers have not found regionally-widespread evidence of evaporitic sulfate formation (such as gypsum) or of the widespread aqueous alteration of igneous rock with acidic (sulfur-rich) fluids. Significant concentrations of sulfate in the rock of these provinces has been attributed either to some form of explosive volcanism, or the mixing of sulfate-rich materials during periods of aeolian deposition. As these evaporitic salts are not observed on the Thaumasia Plateau, much less extensive aqueous alteration of any sort, and (furthermore) with no diapir or namakir-like apparent features (especially in fault zones), some researchers have challenged the hypothesis that the Thaumasia Plateau formed as a gigantic “megalandslide” lubricated by the dynamics of salt tectonics. However, it remains possible that evidence for salt tectonics in the Thaumasia Plateau might have been concealed by later volcanic, aeolian, or aqueous activity.

Mineralogy of the Claritas Rise

Distributions of ferromagnesic phyllosilicates, olivine, and low-calcium pyroxenes have been proposed on the southern putatively pre-Tharsis edifice of the Claritas Rise. Spectral evidence of ferromagnesic phyllosilicates are particularly observed on the crests of the promontories of the Claritas Rise.

Serpentine is a ferromagnesic phyllosilicate alteration product commonly known to accompany magnetite, which is thought to be the principal source of the magnetic anomalies on the ancient Martian terrains which formed prior to the shutdown of the core dynamo. Spectral data indicates that it is amongst the phyllosilicates present on Noachian-aged terrains, but it is exceedingly rare – it has been observed on the Claritas Rise, the only location where such an observation was noted on the Tharsis Rise. (Outside of Tharsis, serpentine has also been observed in the Nili Fossae, in western stretches of Arabia Terra and possibly within the peaks and walls of craters in the southern highlands.) Within the Claritas Rise, serpentine-bearing rock is found in raised knobs and in their debris aprons, and is associated with the presences of other phyllosilicates: chlorite, kaolinite, illite and muscovite mica. Serpentine is also known to form in association with brucite and talc, neither of which has been observed in the Claritas Rise (or elsewhere on Mars). The lack of brucite might be explained by how it weathers readily in a carbon dioxide-rich atmosphere.

Although serpentinite was hypothesized by some researchers to occur as a secondary product to magnetite, which could explain the magnetized terrains, the spatial distribution of serpentinite does not coincide with magnetized terrains outside of the Claritas Rise or in Arabia Terra. The presence of serpentine on Mars does not appear to depend on olivine abundance, its most likely progenitor; nor does it appear to depend on the availability of water for aqueous alteration. Some authors have proposed that serpentine used to be abundant but has largely weathered out; the surface composition of serpentine is not representative of its underground distribution; or that alteration conditions on Mars usually favor chlorite, or the formation of oxide minerals and smectites.

Stratigraphy of the Valles Marineris

Valles Marineris offers perspective on the stratigraphy of the Thaumasia Plateau. The walls have been coarsely divided into three geomorphic units.

The base of the Valles Marineris chasmata walls are often enriched in low-calcium pyroxenes and olivine (observed to crop out in Ganges Chasma and Eos Chasma in the eastern sector of the Valles Marineris), which are associated with highly mafic diking in the vicinity of eastern Coprates Chasma. However, these layers are spectrally heterogeneous across the entirety of their exposures. These layers have been compared against Tanaka (2014)'s global map and may be genetically linked to the pre-impact massifs exhumed by the impacts forming Argyre Planitia and Hellas Planitia. If this is true, these layers may pre-date the Noachian period and may be amongst the oldest exposed terrains on the planet. The strata's mineralogical heterogeneity supports the hypothesis that this basement layer was once the magma ocean that blanketed the incipient planetary surface following the conclusion of Mars' planetary accretion phase.

Outcrops of the ultramafic basement layer in the eastern reaches of Coprates Chasma are additionally dominated by spectral signatures of plagioclase, which has only been detected in a few other locations on Mars (Nili Patera/Syrtis Major, Xanthe Terra, northern Hellas Planitia, and most notably at Gale Crater, where it has been geochemically assessed in situ by the NASA rover Curiosity). The outcrops analyzed by Curiosity have been hypothesized to correlate to the terrestrial tonalite-trondhjemite-granodiorite (TTG) lithology, which are a major component of the Archean shields that form the most ancient cores of Earth's continents; they have alternatively been tied to felsic intraplate magmatism, which eventually yields felsic rocks that form through the fractional crystallization of initially mafic rocks across Bowen's reaction series. Some researchers have speculated, in the light of the Gale crater findings, that this area might represent a nascent craton (in the case of a TTG lithology) or a site of widespread basaltic fractional crystallization. The extent of this concentration of plagioclase does not occur again in younger strata exposed by the Valles Marineris; the conditions to form this layer have therefore likely not reappeared since these earliest phases of the Martian geochronology.

The middle layer has been dated to the Noachian and are understood by researchers to be consistent with the myriad terrains forming the dissected and heavily-cratered southern highlands of the planet. In Coprates Chasma, the northern and southern walls are asymmetric and are more given to landslides on the southern end. The cliff sides also appear much sharper and are more highstanding than they are on the southern walls, implying that the southern wall of the Coprates Chasma is more mechanically weak and thus is likely made of less structurally competent strata.

The uppermost layer is composed of a unit that is consistent with the Amazonian to Hesperian-aged volcanic plains that are associated with the formations of Syria Planum and the Tharsis Montes.

Phyllosilicates are observed within the walls of the Valles Marineris and have been dated to two events: an early Noachian event, and a global late Noachian event. More recent Al-rich clay alteration events have been observed but are almost universally associated with localized impact-driven hydrothermal activity, although most of the observed impact-related examples did occur in Noachian-aged terrains. The phyllosilicate minerals observed within the eastern reaches of the Valles Marineris (Coprates Chasma and Capri Chasma) are spectrally consistent with ferromagnesic smectites and are laterally discontinuous along the northern walls of the interlinked chasmata. The inconsistent distribution and diversity of phyllosilicate minerals suggests that they were irregularly altered by localized hydrothermal activity associated with impacts or intrusive magmatism. Alternatively, more deeply-buried phyllosilicate materials may have been exhumed or otherwise excavated.

Zeolite spectral signatures have been observed in the eastern reaches of Coprates Chasma in both the northern and southern walls and in the chasma's central horst. In the lowest stratigraphic unit, they occasionally found to intersperse with the ferromagnesic phyllosilicates observed elsewhere in the Valles Marineris. Further up in the stratigraphic column, in the middle, low-calcium pyroxene-enriched layer, zeolites are also found to intersperse with chlorite (in its lower section), with carbonate minerals (in its upper section) and with serpentine. Assemblages of these particular minerals and mineral groups have not been identified in outcrop elsewhere on Mars. All of these minerals are known to occur on Earth as a result of hydrothermal alteration of mafic rock, and (in the Valles Marineris walls) are generally associated with dikes rich in olivine. The additional presence of carbonate and of zeolites (such as analcime) suggest that the hydrothermal fluids altering minerals in this region were not acidic. The presence of carbonates can likely be explained either by dike-associated, carbon dioxide-enriched magmatic fluids, or by the effects of carbon dioxide ice buried in the country rock. Because these distinctive outcrops are highly localized and occur on both sides of the Coprates Chasma wall, but do not appear to have experienced dramatic offset on a scale of hundreds of kilometers, some authors have used them to argue against hypotheses that requires strike-slip faulting to be accommodated by the Valles Marineris when explaining the origin of the Thaumasia Plateau (and of the Tharsis Rise in general).

The heat associated with the extensive alteration of this basement layer is unlikely to be associated with the formation of the Valles Marineris, as the observed alteration minerals in this stratum are not distributed evenly against the boundary between the chasmata and the Thaumasia Plateau to the south. However, the extent of alteration is spatially consistent with the uplift of Thaumasia Planum and the Coprates Rise, suggesting that hydrothermal alteration activity took place instead during the formation of the pre-Tharsis massifs of Thaumasia, predating the formation of the Thaumasia Plateau and the Valles Marineris.

On the Claritas Rise and the Thaumasia highlands, as well as Lunae Planum in its reaches immediately north of eastern Coprates Chasma and into Juventae Chasma display spectral signatures of the following alteration minerals: ferromagnesic phyllosilicates, chlorite, carbonate, zeolites, and serpentine. These terrains are all dated to the Noachian and have been interpreted to represent uplifted ancient crustal material. These regions demonstrate the only secondary mineral assemblages that closely approximate those seen in Coprates Chasma.

The three mineralogical units of the Valles Marineris walls also notably display a vertical offset indicative of thrust faulting on a scale of 1km in the eastern sector of Coprates Chasma, in accordance with a predicted compressional stresses in this region. Explanations for thrust faulting in this area require the reduction of frictional forces, which could be explained if the fault contact was predominated by serpentinite-rich fault gouge. If this late-stage compressional faulting that occurred here depended on pre-Tharsis regional alteration processes, then this would support the hypothesis that the features of the Coprates Rise and the Thaumasia Highlands actually constitute a fold and thrust belt (which also requires this reduced friction criterion to be fulfilled). A west-dipping normal fault has also been discerned in the same way at the eastern boundary of Candor Chasma as it opens into the Melas Chasma region, in the central Valles Marineris.

Approximately 100 central-peak impact craters (notably including Oudemans crater) to the east of Noctis Labyrinthus, in the vicinity of the Valles Marineris, are found to have exhumed ancient basement terrains within their central peaks. Near to the center of the Tharsis dome's central uplift, layered ashes or lavas are preserved, but densely fractured bright rocks are exposed towards the periphery. This densely-fractured unit is not even exposed in the deepest strata of the Valles Marineris walls and may be representative of the original Martian basement dating to the cooling of the planet's original magma ocean.

Tectonic features

Extensional faulting

Martian outflow channels may have originally been routed along extensional tectonic features on the planet, including Mangala Valles and the Valles Marineris-Chryse Planitia region. The formation of Mangala Valles in particular has been associated with the formation of Aganippe Fossa, which has been proposed to be a radial dike of Olympus Mons that formed during the later stages of the volcano's formation.

Simple graben and volcanic catenae

The Tantalus Fossae near Alba Patera are the result of extensional tectonic activity, with complex graben and rifting apparent at left, and simple graben towards the right.

Where simple graben are amply distributed across the planetary surface, additional extensional stresses may be accommodated at depth according to a variety of proposed models. The uniformity, radial distributions, and significant length of Martian grabens has been analogized to the Mackenzie dike swarm in northwestern Canada, in which magmatic diking emanated across the field from a single tectonic center. The length of the graben is controlled by magma pressure; the graben ceases to expand when the underlying diking magma loses too much pressure to continue to overcome the resistance of the country rock. The width of such graben are controlled by the depth to either the intersection of its bounding faults or to the roof of the underlying dike. Strong evidence of coeval graben propagation and volcanic activity is widely expressed in Syria Planum, suggesting that at least some of the observed tensile stresses are accommodated by subsurface dikes. Alternatively, subsurface extension could be accommodated by pressurized fluids, which - upon exceeding the pore pressure of the containing rock - triggers the formation of subsurface tension cracks. Crater chains observed elsewhere can manifest from these collapsing subsurface tension cracks. This hypothesis has been evoked to explain the catastrophic upwelling of the fluids that formed the Martian outflow channels, which have sometimes been observed to begin from collapse structures or from graben. If graben on the surface are not too densely spaced together, the basement rock may be capable of elastically expanding, accommodate additional strain in the subsurface without experiencing failure. Graben are also capable of forming if a rock layer sits atop either a ductile basement or above a detachment fault. In this situation, the unfastening overlying rock can freely spread laterally. Such a phenomenon has been terrestrially observed in the Canyonlands National Park of Utah, and may explain the graben of south central Valles Marineris and the in the Olympus Mons aureole. On the Tharsis Rise, graben are typically understood to form in association with this first mechanism - subsurface diking - and are thus not purely tectonic features. They accordingly cannot be used to discern the inherently preferred tectonic style of the region (compressional, strike-slip), as the formation of a subsurface dike almost always induces an additional surficial extensional stress that cannot necessarily be regionally predicted.

A chain of collapse pit craters within the Cyane Fossae, near Alba Patera. This crater chain overlies a simple graben and is consistent with the aforementioned volcanic origination hypothesis. Imagery is HiRISE.

Extensional tectonic activity underlying graben and volcanic pit crater chains (catenae) can be attributed to diking. Most of these graben manifest surficially as narrow valleys with flat floors and are often aligned nearly shoulder-to-shoulder. Although less common, complex rifting does manifest on the planet as well.

Very large graben and collapsed pit crater chains at Aganippe Fossa (on the western upper flank of Arsia Mons) and in Syria Planum (at the curvilinear simple graben transition from the Claritas Fossae to Noctis Labyrinthus) have been identified as potential members of a giant radial dike swarm emanating from Olympus Mons. This phenomenon has been analogized to the volcanically-active eastern flank of Kilauea on the Big Island of Hawaii. The formation of these large graben are associated with the partial magma chamber drainage and collapse formation of the central calderae at the summit of Olympus Mons.

The simple concentric graben that form around summit calderae or impact craters has been attributed to a number of formation mechanisms. Some authors have proposed that the cooling and contraction of magma undermines overlying terrain, causing the terrain to subside. Others have identified another mechanism in which effusive eruptions elsewhere cause a magma chamber to deflate.

Rifting

Regions that demonstrate rifting are affected by a very high density of interdependent normal faults. The emerging pattern of deformation obscures the individual normal faults and obvious signs of elastic deformation.

Basement terrain faulting in Thaumasia

The Claritas Fossae are a Tharsis-radial region of intensive rifting, and is the densest location of extensional faulting on the Tharsis Rise. The Claritas Fossae extend southwestwards into the Thaumasia Fossae. Some of its faults may have preceded the formation of the Tharsis Rise and were reactivated in association with the first tectonic center predicted by Anderson, which was centered in this region of the planet. The Claritas Fossae likely initiated in the early to middle Noachian, experiencing a relative decline in activity by the late Noachian to the early Hesperian, and achieved its modern state by the late Hesperian or early Amazonian.

Basin and range topography

Some authors have reported on the existence of en echelon mountain ranges interspersed by valleys in Terra Sirenum, off the western flank of the Tharsis Rise, and have interpreted these structures to mirror the Basin and Range Province that persists in the Southwestern United States and in Nevada.

There are 29 large ridges that stand up to 1.5km above the surrounding plains of Terra Cimmeria and Terra Sirenum to the south and southwest of the Tharsis Rise. Authors initially interpreted these ridges to have resulted from compressional stresses driving buckling and thrust faulting. Some researchers tentatively compared this to buckling features located within the Central Indian Basin. Other researchers later noted the similarity of these ridges to the Basin and Range mountain province in the Western United States, characterized by near-parallel ridges of similar relief separated in alternation by valleys. The ranges of the Basin and Range Province are found to topographically mirror the ridges of the South Tharsis Ridge Belt nearer than possibly analogous compressive structures such as the Amenthes Rupes. The extensional ranges of the Basin and Range Province are often symmetrical horsts but may sometimes have asymmetrical reliefs. Both of these topographies are mirrored in the South Tharsis Ridge Belt, with the only major topographical differences attributable to fluvial erosion reshaping the caps of the terrestrial ranges. These terrains do not correlate with Banerdt (1992)'s stress fields of the Martian western hemisphere, which would instead predict extensional membrane stresses that have triggered radial graben swarm nucleation elsewhere on the Tharsis Rise; some researchers have speculated that they may have been formed during a tectonic event that either preceded or was tied to the earliest stages of the formation of the Tharsis Rise, such as an early stage of uplift tied to the hypothetical Tharsis megaplume. Rather than forming a triple junction as incipient plume activity is observed to do on Earth, the concentric stress signal would likely depend on the fact that since Mars is much smaller than Earth and Venus, the ridge belt forms at a sufficient distance from the heart of the Tharsis Rise as to evoke a flexural rather than a membrane stress response.

These ridges have historically been considered compressional in origin, but others have more recently argued that they are extensional because many of the requirements of a compressional origin are either not defensible given large-scale stress modeling of the Tharsis region, or are not analogous to comparable features observed on Earth. Among other stress predictions, Banerdt (1992)'s model predict extensional loading stresses in this region, although the ridges are orientationally aligned with compressionally-originated wrinkle ridge terrains in the area (which have instead been linked to a global compressional event). The terrains beneath the South Tharsis Ridge Basin are understood to be almost 2-5km thinner than the terrains of the surrounding southern highlands, a characteristic that is associated with the same process of crustal thinning that is thought to have formed the Basin and Ridge Province. This extensional activity must be offset elsewhere by compressional activity; based on the deformations of craters (which are, except in very rare cases, almost perfectly circular), some researchers have proposed that this compression could have been carried in the western reaches of the Terra Sirenum highlands, to the west of the ridge belt. This argument has not been authoritatively justified because the actual strain signals of many craters in this region are variable. However, some researchers have indicated that a coincident period of global contraction, understood to have formed the Mars-wide sets of wrinkle ridges, may have reactivated faults in this area and contributed to the more complex deformation signal. Based on cross-cutting relationships of both deformed and undeformed craters in the Terra Sirenum region, the formation of this ridge belt is thought to have begun prior to the Noachian period, and to have continued up until the beginning of the Hesperian period; they are thought to have been a favored style of rifting in regions that were structurally unweakened by magmatic activity, given the relatively low strain rates expected during the Noachian period. The en echelon distribution of the ridge belt has been speculatively linked to the migration of the hypothetical Tharsis mantle megaplume.

The extensional basins between the ridges at the western extent of the ridges within Terra Sirenum often bear the signatures of late Noachian phyllosillicates and chloride deposits, suggesting that these basins served as longstanding catchments for water originating from valley networks and outflow channels (such as Mangala Fossa), allowing them to be distinguished geomorphologically by distinctive deposits.

Although the high-standing ridges display a strong magnetization signature, the low-lying extensional basins do not, suggesting either that the basins formed after the shutdown of the dynamo, or that later resurfacing has since buried rocks that display that signature. Bouguer gravimetry of the region matches the distribution of gravitational anomalies observed in the Nevada Basin and Range province, further supporting an extensional origin of the region.

Compressional faulting

Wrinkle ridges on Mars are associated with compressional tectonic activity. They surficially express themselves as sinuous and/or crenulated ridges and are not generally linear in shape, sometimes changing by as much as 30 degrees in orientation over even short distances.

In the vicinity of Tharsis, wrinkle ridges have historically formed in concentric patterns in association with compressive peripheral hoop stresses. However they are also globally distributed, thought to be a result of a global Noachian-era contractional strain associated with the planet's secular cooling in hundreds of millions of years following the wake of the lithosphere's initial crystallization after the end of planetary accretion. Although this Noachian global contraction event coincided with the formation of the Tharsis Rise, the majority of wrinkle ridges on Tharsis are actually associated with Anderson's third stage of Tharsis' history, when the Rise's tectonic center had migrated underneath Syria Planum during the Hesperian period. However, the formation of wrinkle ridges associated directly with the formation of Tharsis has been continuous and persistent since the Noachian period, and evidence of continued Tharsis-associated wrinkle ridge formation has been also observed in the northern plains even into the early Amazonian.

Large wrinkle ridges (LWRs) have recently been identified as a second compressional morphology to the standard wrinkle ridges observed on Mars atop the Thaumasia Plateau. These ridges are generally located on Lunae Planum to the north of Coprates Chasma, and are oppositely asymmetric towards the heart of the Tharsis Bulge, relative to normal wrinkle ridges; their steeper side faces the Tharsis Rise rather than away from it. The subsurface effect of these wrinkle ridges on stratigraphic attitudes is observable due to the proximity of the observed LWRs to the northern wall of Coprates Chasma.

Fault scarps are also observed in Noachian terrains in the southern highlands of the planet and may have formed as a result of thrust faulting under similar stress conditions as the wrinkle ridges. The morphological differences between them and wrinkle ridges are attributed to differences in the material strength of the terrains upon which those compressive stresses act.

South Tharsis ridge belt contractional features

The Coprates Rise has been analogized to the Wind River Range in Wyoming, a subsidiary range of the Rocky Mountains, because the Coprates Rise demonstrates the presence of features resembling cuestas and hogbacks, and is similarly dissected by erosional valleys.

Strike-slip faulting

Strike-slip faulting occurs when the largest and also the least of the principal stresses are horizontally-aligned, and when the deviatoric stress exceeds some minimum threshold. As opposed to Martian graben, which are not purely tectonic in their nature and thus have orientations that are sensitive to the localized stress effects of diking, strike-slip faults on Mars are understood to be sensitive only to the regional stress field. Given researchers' assumptions about the fracture strength of Martian surface rock in the Tharsis area, these faults are generally understood to form at an angle of 30 degrees offset from the direction of principal stress, and thus are 30 degrees offset from coevally-formed graben or 60 degrees offset from coevally-formed compressional features (e.g. wrinkle ridges).

Thickening of the lithosphere associated with the growth of the Tharsis Rise suppresses the formation of extensional faults at depth; when compressional faults propagate upwards, a sufficiently high deviatoric stress may manifest on the surface, allowing formation of strike-slip faults. Thus, the manifestation of strike-slip faults partially depends on the thickness of the lithosphere in the region where they are observed to manifest.

Martian strike-slip faults can theoretically manifest either a positive or a negative relief. If such a fracture occurred in crystalline bedrock, the resulting fault gouge would be weaker and would erode more rapidly than its environs. However, a fault that serves as a conduit for groundwater would mineralize that fault gouge, fortifying the fault and leaving a more resistant ridge structure behind. The latter structure has been observed and studied on Mars. Strike-slip faults are required to have an extremely linear fault trace to accommodate lateral offset, morphologically distinguishing them from wrinkle ridges. They are also found to impart an asymmetric vertical deformation; the fault's leading quadrant experiences uplift and its trailing one experiences subsidence, allowing a determination of the fault's sense of slip even in the absence of obvious offset.

Except in a handful of localized applications, strike-slip faulting is not evident in most of the Tharsis region, in contradiction to the expectations of early stress models of Tharsis in which strike-slip faulting was expected in the transitionary zones between the center and peripheries of the Tharsis bulge. However, the lack of strike-slip faulting is consistent with observations on the Moon, where normal faulting occurs in places where strike-slip faulting might have otherwise been expected; in these cases, the upwelling of diking magma may have favored extensional stresses at depth, generally manifesting in extensional structural features at the surface. Furthermore, the prevalence of strike-slip faulting on Earth can be effectively be attributed to how the motion of its tectonic plates are not laterally constrained as they are on a one-plate body such as the Moon. Mars is effectively subject to the same lateral boundary constraints. In such situations, higher deviatoric stresses are required for strike-slip faulting than normal faulting; thus the crust yields extensionally before it yields to shear.

Some authors have attempted to explain the lack of immediately-visible strike-slip faulting on Mars because, as a result of these lateral boundary constraints, faults on Mars are pinned at their tips in a way that approximates terrestrial intraplate tectonic models for such faults - that is, a 100km-long strike-slip fault on Mars would be expected to impose a lateral displacement on a scale of no more than 0.6km to 3km. Strike-slip faults may sometimes yield offsets that are significant enough to notice, but are generally speaking more readily identifiable by their morphologies instead.

On the western flanks of the Tharsis Rise, based on the distribution and relative ages of strike-slip faulting morphologies, some researchers have proposed that the earliest manifestation of strike-slip activity occurred in the Noachian with the onset of activity from Anderson's stage 1 (Claritas Fossae) tectonic center, emplaced in Terra Sirenum, and migrating northwards with decreasing activity until the Memnonia Fossae region during the activity of Anderson's stage 3 (Syria Planum) center. The most recent activity appears to indicate a further shift in tectonic activity into Amazonis Planitia, likely occurring in the late Hesperian epoch. Separately, in the eastern reaches of Tharsis on the Thaumasia Plateau, in the vicinity of Thaumasia Planum and the Coprates Rise, a separate strike-slip faulting event has also been observed.

Thaumasia Plateau strike-slip faulting

One of the areas of the Tharsis region where strike-slip faulting has been observed lies in the northeastern portion of the Thaumasia Plateau, cross-cutting the wrinkle ridges of Thaumasia Planum and the Coprates Rise. A particular range of localized uplifts - morphologically distinct from neighboring wrinkle ridges - are found to be interspersed with very straight lineaments offset from each other en echelon. These uplifts have been interpreted as transpressional stepovers, with the straight lineaments interpreted as strike-slip faults. These Late Hesperian features are cross-cut by wrinkle ridges and are observed to cross-cut others, implying that their formation period preceded and overlapped that of the region's wrinkle ridges. Right-lateral strike-slip faulting predominates in the northern part of the faulted area and left-lateral strike slip faulting in the south.

When this feature was first reported in 1989, the principal stress trajectories of this region did not conform to contemporaneous isostatic models of the Tharsis Rise, suggesting additional tectonic complexities that may be associated with the Valles Marineris to the north.

Memnonia Fossae-Terra Sirenum strike-slip faulting

Strike-slip faults have been identified and studied in this region to the west of the Tharsis Rise, and are generally offset by 30 degrees from Anderson's first stage of graben formation, which is associated with the Noachian tectonic center on the Claritas Rise. They are overprinted by graben radial to Anderson's third-stage Hesperian Syria Planum tectonic center, suggesting that they mostly formed in association with first-stage faulting. (Exceptions exist in the southern reaches of the Memnonia Fossae area, where strike-slip faulting may have persisted well into the period of Anderson's stage 3 tectonics.) Two ridges of this morphology have been explicitly identified to have asymmetrical vertical throws and/or horizontal offset, with the sense of throw directionally consistent with a radial Tharsis loading. Other putative strike-slip faults in this region are morphologically similar in shape and orientation, although they do not exhibit the same clear markers of offset. The lack of obvious offset in this region's strike-slip faults has been attributed to their state of degradation, given their age. The faults are the oldest strike-slip faults yet identified on Mars.

There is exactly one instance in which a strike-slip fault in this area is found to intersect a stage 3 (Syria Planum) Tharsis-radial graben, although these graben are found in ready association with wrinkle ridges. Researchers who noted this phenomenon have suggested that the formation of a strike-slip fault relieves the deviatoric stress by increasing the circumferentially-oriented σ₃ (least principal) stress. This deflects the propagating tip of a dike downwards, burying it below the surface and terminating graben before they can intersect the strike-slip zone. In contrast, compressional tectonic activity only relieves σ₁ (greatest principal) horizontal stresses and has no bearing on the hoop stresses that control the formation of graben.

Amazonis Planitia strike-slip faults

A third region has recently been identified to have the same strike-slip morphologies previously described, existing in the Amazonis Planitia region to the northwest of the Tharsis Rise in the northern lowlands of the planet. Nearly all of these faults have asymmetrical vertical throws, and almost none of them display observable horizontal offset. In some situations, the faults appear draped by more recent lava flows, or manifesting instead as en echelon cracks rather than a continuous manifestation of brittle shear.

The strike-slip faults in this area were initially interpreted to be of an early Amazonian age. Researchers later re-examined them and were found them to be older, but located within an Amazonian unit only because they had been thinly resurfaced by early Amazonian lavas. These strike-slip faults are themselves thought to be of a late Hesperian age, at the youngest.

Complex features

Valles Marineris

Historical hypotheses for the formation of the Valles Marineris have argued that it was unfeasible for the chasmata to be bounded by normal faults, which would have manifested in modifications to the surrounding terrain (which demonstrates no topographic change or any particular mode of deformation). Most initial hypotheses envisioned a hypothesis that was not directly tectonic - the formation of subsurface voids, triggering a dominantly vertical collapse. Most tectonic explanations for the Valles Marineris anticipate normal faulting (which in the Andersonian tectonic model assumes an ideal 60° angle from horizontal, varying up to within 15° of that mean), which have conflicted with the observation of effectively vertical faults that (with the exception of the Ius-Melas-Coprates trough) abruptly terminate in rectangular fault-bound ends rather than gradually dying out in converging tips as they do on Earth. No self-consistent stress distribution can compensate for the possibility of a rectangular graben that is bound on all edges by normal faults in any case, and strike-slip faulting on this scale (in this effectively intraplate setting, which pins the tips of the faults and minimizes displacement and thus requires extremely large faults) are not observed at the regional scale. Authors have cited these issues to be particularly difficult when considering the cases of the ancestral basins, such as the neighboring Echus and Hebes Chasmata. Although the hypothesis of vertical faulting (through a collapse mechanism) neatly solves these concerns, its accompanying justifications are not well-founded - including ground ice melt, tension fractures, magma withdrawal from underlying radial dikes, or carbonate dissolution. Vertical faulting has also not been widely corroborated by researchers measuring the dip of scarps and faults exposed within the canyon walls of the Valles Marineris. However, other authors have argued that these other faults must be linked to tectonism independent of trough formation, as they are - by definition - not bounding the troughs that are observed (which would be concealed unless erosive activity cross-cuts the fault at an oblique angle). If these shallowly-dipping faults are indeed representative of trough faulting in the Valles Marineris, it is possible that the steep-walled faulting occurred first, leaving the wall rock unsupported and gravitationally driving the formation of these more shallowly-dipping faults. Alternatively the angles of the steeply-dipping faults may become shallow as they approach the surface, and may separate out into splays that explain the large variability in the dip angles observed in exposed scarps.

The free-air gravitational anomaly of the Valles Marineris system is nearly completely uncompensated by isostatic processes, and thus is supported entirely by lithospheric flexure. Such a formational mechanism is contrary to the boundary conditions that gave rise to the major rift valleys on Earth, such as Lake Baikal or the Red Sea, which are almost fully compensated by isostatic processes. Some researchers have indicated that the Valles Marineris must have formed in an isostatic equilibrium with the erosion and removal of a very thick sediment fill, inducing a flexure of the neighboring terrains, and in which extant interior layered deposits (ILDs) are the modern remnants of these eroded deposits. If the Valles Marineris were rapidly opened as a result of extensional tectonism, the inertial effects of thermal diffusion through the complex would trigger an additional stage of uplift that would settle as the thermal environment of the Valles Marineris ultimately settled upon its Pratt isostasy. The topographic expression of the Valles Marineris may have differed significantly between the modern context and the period of time shortly following the emergence of the valleys.

In accordance with this hypothesis, two centers of flexural uplift – one beneath the central depression of the Valles Marineris at the intersection of Ophir Chasma, Candor Chasma and Melas Chasma – and a smaller region beneath Eos Chasma and Capri Chasma – have been proposed. A third center of negligible net uplift was also proposed in the region between Tithonium Chasma and the Noctis Labyrinthus. Extensional tectonic features are expected to immediately parallel these centers; the Nia Fossae graben swarm lies circumferential to the south of Melas Chasma on the Thaumasia Plateau, and have been proposed to have formed in association with the flexural uplift resulting from the erosion of ILD material rather than in association with the giant Tharsis-radial dike swarm associated with Anderson's third center of tectonism (Syria Planum). Another graben swarm displays similar concentricity to the north of Ophir Chasma in Lunae Planum.

Very few of the tectonic features on the peripheral plateaux to the Valles Marineris complex are oriented in a way that would relate them to the stress fields that must have formed the chasmata, and thus either occurred before (or in the early stages of) the formation of the Valles Marineris or in unrelated later events. The only exception to this situation exists within a series of graben on the Thaumasia Plateau that formed concentrically to the southern arcuate boundary of Melas Chasma, which may have been driven by gravitational potential energies as the Melas Chasma (erosionally) expanded over time. Some of the normal faults and joints observed in the vicinity of the Valles Marineris flexural centers are far too young to be appropriately attributed to one of the swarms described by Anderson (2001)’s tectonic centers. Because of their timing it is far more feasible that they are connected to the late-stage erosion of ILDs, facilitating continually changing flexural stresses along the same orientations into the present day. Some of the faults - if they had preceded the formation of the Valles Marineris - may have experienced reactivation during the chasmata's formation, accommodating additional displacement.

In contrast to the distribution of graben on the terrains surrounding the Valles Marineris chasmata, many collapse pit chains appear to have formed as a result of the stress fields that formed the Valles Marineris as well. However, it is not clear if they were coincidentally formed from the same stress field, or by a process that was involved in the formation of the troughs themselves. These features may be simple fissures, but they are most likely underlain by magma chambers that became evacuated. They may also transition into extensionally-driven tectonic or intrusive features at depth.

The valley floors of the Valles Marineris - particularly in the Ius and Tithonium Chasmata - are dominated by landslide deposits. These landslides are often cross-cut by very recent faults, and have - in the Coprates Chasma - been constrained to dates between 50 Ma and 93 Ma, well into the Late Amazonian epoch. Older cross-cut landslide deposits are identified in Tithonium Chasma and have been relatively dated to the Middle Amazonian epoch. As there are no fresh impact craters greater than a kilometer in diameter in the vicinity of any landslide in the Valles Marineris, some researchers have ruled out the notion that impact-based tectonism can explain modern-period mass wasting on the scale observed in the chasmata, which suggests that intrinsic tectonic activity - in other words, marsquakes of moderate magnitude - must have been responsible for forming these objects.

Falling boulders will impact and imprint the surfaces upon which they track as they are detached from sections of the Valles Marineris walls. They have been used as indicators of local seismic activity across the chasmata. Given the state of preservation of the tracks of Spirit and Opportunity, some researchers have suggested that the observed boulder tracks may have formed as recently as within the past few thousands of years.

Mud volcanoes have been proposed to explain pitted cone-like features sitting on the valley floor of the Valles Marineris. Such cones are associated closely with interior layered deposits and have lower thermal inertias than basaltic ILD materials, suggesting a sandy or muddy composition. Other authors have proposed that these pitted cone features are scoria cones, which would invoke a forceful decompressive discharge of magma driven by the decompression of volatiles within that magma. Such features have been observed in Hawaii (e.g. Haleakala, Iceland (e.g. the Laki fissure), and on the flanks of Mount Etna in Italy. Mud volcanoes on Earth have been observed to form within days in the vicinity of the affected areas of an earthquake, and their prevalence in on the valley floor deposits of the Valles Marineris is supportive of the hypothesis that this correlation holds true on Mars as well.

Igneous features

Tharsis Montes

Lava flow fields

The basaltic volcanic plains on the southwestern flank of Arsia Mons (and into Daedalia Planum in the southwest) are amongst the roughest terrains on the planet (at cm-scales), apparently undiminished over time by either erosive or by depositional processes. Researchers have taken note of the unusual, extremely conspicuous thermophysical variation of the flows (as seen on infrared satellite imagery). The thermophysical properties are shown to vary significantly even between individual adjacent lava flows.

There are two general types of flow units associated with the southwestern reaches of Arsia Mons: a rugged unit and a smooth one. Thick, very rugged flows are seen to stretch up to scales of hundreds of kilometers downhill and are observed to reach up to 20km in width, and are sometimes associated with ridges, and with central channels and levee structures that are sometimes spilt-over to form laterally-extensive, “digitated” lobate structures. These widening structures have been attributed to situations in which the progression of the flow front stalled, At times, this type of flow unit takes on a fragmented, knobby distribution. Under HiRISE imagery, these rugged features have been observed to manifest as ridges of outcropping high-albedo rock interspersed with dark sands that infill the valleys between the ridges. Transverse aeolian ridges are commonly found amongst the dunes within these valleys. Other flow lobes are smooth and appear to be very dark on nighttime infrared data, and they are not generally as long as their more rugged counterparts. These tend to be far thinner and less distinct than the more rugged flow features. The same kind of digitate morphology are observed on these layers, which have been analogized to lobate pahoehoe flows on the shield volcanoes of the Hawaiian islands, relating to lava tubes and/or resulting from inflation. The effects of lava inflation in this darker, smoother flow lobe are noted to observe when the lobe is emplaced on top of an extant rugged flow feature, paralleling the observed effects of pahoehoe flows that overlay aa flows on their terrestrial Hawaiian analogues.

The darker, smoother lobate flows are seen to both embay and be embayed by the more rugged, brighter flows, indicating that many of these highly distinct flows formed coevally and that their contemporaneous parent sources were very heterogenous in nature. The boundary between different flow lobes is frequently gradational. If rugged flows formed a local depression beforehand, more recent smoother flows are observed to pool within those depressions. More recent flows sometimes incompletely bury older flows, complicating assessments of the region's stratigraphy.

As the effects of inflation in the smoother, darker layers appears confined to its superposition over the thicker, more rugged flows, they may capture some expression of the underlying flow surface's characteristics. Some authors have suggested that if the smooth darker layers have followed the channel and levee systems of underlying, more rugged flow lobes, the increased lava flow in these areas could feasibly invert the topography of those features. It more broadly explains how the dark, smooth layers sometimes become significantly more irregular in their texturations, as their flow fronts would stagnate when passing over significantly rougher terrain.

On the southwestern flank of Arsia Mons in particular, there are a collection of lobate flow features that experience no appreciable temperature change at any time in the day or night. This is extremely unusual, as areas on Mars that are extensively mantled with dust will tend to be warm during the day and cold at night (low thermal inertia); and if younger, more coherent flows are overlying older dust-mantled features, such flows will remain warm into the night and cool into the day (high thermal inertia). The authors who first examined these lava flows in detail have indicated that thermophysical data (as of 2017) is likely too coarse to resolve the specific thermodynamics of these features.

Composite volcanism

The major edifices of the Tharsis province are typically referred to as shield volcanoes and are understood to be constructed largely from basaltic lava flows, but some researchers have counterproposed as early as 1989 that the volcanoes might be stratovolcanoes (composite volcanoes) instead. In this situation, the volcanoes would be constructed from alternating layers of pyroclastic rubble and lava flows. Terrestrial stratovolcanic eruptions are significantly more explosive than terrestrial shield volcanoes and generally composed of materials that are more felsic and acidic than their basaltic shield volcano counterparts. Because Mars has a lower atmospheric pressure relative to that of Earth, explosive eruptions are expected to be exaggerated relative to their terrestrial analogues, with the effect becoming even more enhanced at higher elevations for the same reasons. Among other things, a lower atmospheric pressure is expected to cause magma fragmentation that reduces the grain sizes of pyroclasts an order of magnitude finer than what would be expected in a terrestrial situation, causing a resulting Martian stratovolcano to assume a flatter and wider profile than on Earth. Basaltic Plinian eruptions of basaltic material can be triggered with only a few hundredths of a percent increase in percent magma composition of gases due to reduction in the grain size of Martian pyroclasts, forming sub-cm-scale fine-grained mantles for tens to hundreds of kilometers from the associated vent. Because of the extraordinary height of the Martian shield volcanoes, stratovolcanic activity is not implausible, given the natural tendency to transition towards Plinian eruptions at higher elevations.

Features on Ascraeus Mons and particularly on Arsia Mons have been interpreted as extensive ash-fall deposits that were possibly sourced by an explosive (stratovolcanic) origin. These observations follow several inferred relationships dating as far back to the 1970s, in which vast dune seas were identified atop the Tharsis Montes' summits and upper flanks and within Ulysses Patera, and were speculated to have been composed of ash. Near the summit of Arsia Mons, these fines were recognized around the edges of the mountain's central caldera and around the depressions formed by pit craters, and were estimated to be up to 50m in thickness. The large thickness of this unconsolidated, featureless layer more strongly suggests a localized origin than a long-distance aeolian one, and the lack of nearby eroded features suggests that these materials are not the debris apron of a highstanding eroded feature but of volcanic activity. Few craters superpose these layers, suggesting that they are young or readily resurfaced. Researchers have alternatively noted that these deposits might be phreatomagmatic (water-lava interaction-driven), forming in association with intrusive activity in a permafrosted layer. In either case the particular vent that sourced the aforementioned deposits has not been located. The distribution of the fines observed on Arsia Mons, Pavonis Mons and Ascraeus Mons' summits were not apparently confined to elevation, consisting with hypotheses that at these high elevations, the low atmospheric pressure would almost always cause magma to erupt explosively. The association of these ash-fall deposits with pit crater chains does not hold on Olympus Mons, which lacks the pit crater chain morphologies altogether.

Tholi of the Tharsis Montes

Tharsis Tholus is a smaller volcano, dimensionally a hundred kilometers by a hundred kilometers in dimension, and is unique on Mars because it is cross-cut by large extensional faults that (possibly) stretch across the entire edifice of the volcano.

Valles Marineris

Dike swarms

The floor of Ophir Chasma is dominated by linear features interpreted as dike swarms. These have been spectroscopically identified as mafic (Mg-enriched olivines and high-Ca pyroxenes). Because of the density of dikes of this size on the floor, they likely preceded and/or were contiguous with the formation of Ophir Chasma and were exposed during the processes that formed the ancestral basin. Given the presence of the dikes in Ophir Chasma, it is likely that crustal stretching initiated the formation of ancestral basins like Ophir Chasma, which were then greatly resurfaced (perhaps on a scale of several kilometers of additional topographic reduction) by later erosive glacial activity.

Regional structural histories

Olympus Mons

Olympus Mons is the largest volcano in the Solar System, and it is a shield volcano that stands 22km from surrounding plains terrains with a profile that slopes on average at 5 degrees. The volcano is entirely flanked by an extremely stark scarp. At the base of this scarp is a flexurally-controlled moat that separates Olympus Mons from all surrounding features, including the principal edifices of the Tharsis Rise. As it is on the Tharsis Montes, Alba Patera, and Elysium Mons, and on the major Hawaiian volcano Mauna Loa, the flanks of Olympus Mons are terraced; this is understood to be a compressional effect of the volcanic cone's own elastic deformation. Certain authors have specifically associated this tectonic result with the flexure of the underlying basement rock.

Very large volcanoes such as Olympus Mons are understood to bulge outwards, spreading gravitationally along a lubricated basal detachment while inducing a downward flexure of the underlying lithosphere.

The aureole of Olympus Mons displays a number of lava flow morphologies that are not reflective of flow down the steepest gradients presently observed on the volcano's flanks. Such discordant flows have been assessed to paleotopographically reconstruct earlier stages of Olympus Mons' history. Furthermore, based on modeled lithospheric flexure, the topographic profile of Olympus Mons' peripheries strongly suggests that Olympus Mons was deposited atop a thick-skinned lithosphere (70 to 80km in depth). In conjunction with the chronology established by the aforementioned discordant lava flow morphologies, it is understood that Olympus Mons formed atop topographically flat plains. Given the topography of the landscape in the wake of flexural deformation, some researchers have estimated that up to 81% of Olympus Mons' volume lies below its base and infills the flexural trough ringing the volcano.

The aureole of Olympus Mons is separated from the volcanic edifice by a stark basal scarp which would have likely been buried if it had formed concurrently with the formation of Olympus Mons. This supports the presupposition that the aureoles are not volcanically resurfaced and thus postdate the formation of the bulk of the Olympus Mons edifice. As the aureoles have been dated in crater counts to the early Amazonian, the formation of Olympus Mons must have terminated by this epoch. As volcanic activity on Olympus Mons dropped between two to three orders of magnitude into the early Amazonian, it is likely that a continuously active magma chamber gave rise to more episodically active chambers in the early Amazonian epoch.

Tharsis Montes

Scott (1981) six-stage interpretation

David H. Scott proposed a reconstruction of the Tharsis Montes, Alba Patera, Syria Planum and Olympus Mons based on estimates for how completely draped by thin lava flows the partly-buried craters in transitional zones between the volcanic edifices are.

Scott's Stage 1 of activity involves the high-volume output of flows from an active Alba Patera and from the heart of Syria Planum, burying large-scale faulting and fracturing in the vicinity of the region. The Memnonia Fossae (to the southwest of the Tharsis province) and Ceraunius Fossae (between Alba Patera and the Tharsis Montes) are interpreted as part of the same, structural-scale graben swarm, with one of the associated graben underlying each of the incipient Tharsis Montes in a chain. Noctis Labyrinthus, the Claritas Fossae, and the modern shape of Syria Planum is believed to have already developed at this time.

Scott's Stage 2 of activity involves the peak period of volcanic output from the Tharsis Montes, with the merger of the edifices of Arsia, Pavonis and Ascraeus and the emanation of associated flood lavas, particularly those from Arsia Mons across the older terrains to the southwest, burying most evidence of extant tectonism from the pre-Noachian period and forming Daedalia Planum and its modern transition into Terra Cimmeria and Terra Sirenum. Due to the highstanding topography of the central bulge near Noctis Labyrinthus and amidst the Claritas Fossae, these terrains remain unburied. Flood lavas constituting the bulk of Olympus Mons’ edifice begin to form over lavas previously output by Alba Patera.

Scott's Stage 3 covers the maximal extent of the Tharsis Montes’ flows. The mountains reach their peak height and areal extent, and smaller tholi and highland paterae (Biblis Patera, Tharsis Tholus, etc.) begin to form across the Tharsis Rise. The reduction of the terrains around the Olympus Mons edifice into aureole material are attributed to aeolian erosion. The Ceraunius Fossae are partially obscured by eruptive activity from fissures in the area. The concentric graben forming around the Alba Patera caldera constitute the final phase of activity on the main Tharsis Rise.

Scott's Stage 4 involves the final phases of major eruptive activity from the Tharsis Montes, and the transition of volcanic activity into the Olympus Mons area. The Olympus Mons edifice is constructed during this period, overprinting the aureole. The base of Olympus Mons begins to subside and tilt, forming its conspicuous fault scarp, and creating the topographically low apron presently observed to surround the Olympus Mons edifice.

During Stage 5, major eruptive activity in Olympus Mons comes to an end, with minor volcanism continuing from the fissures of Ceraunius Fossae and from Pavonis Mons, the central of the three Tharsis Montes. Olympus Mons continues to subside. Scott's Stage 6 is the modern stage, and marks the end of the fissure eruptions at Ceraunius Fossae and around the bases of the three Tharsis Montes. Minor fault rejuvenation occurs along the extensional Stage 1 graben swarms hypothesized to that underpin the Tharsis Montes.

Later elaborations

Line-of-sight acceleration profiles involve tracking slight shifts in the acceleration of orbiting spacecraft due to changes in the gravitational constant on Mars (due to reasons other than topography, which inherently suggests density differences in underlying rock). This method has been used to confirm hypotheses that the bulk of the Tharsis Montes edifices were formed principally as a result of extensive volcanism rather than a combination of volcanism and a crustally-associated mechanism.

Valles Marineris

The Valles Marineris can broadly be subdivided into two different classes of chasmata; to the south, the majority of the complex (including the cadre of straight valleys such as the Tithonium-Ius-Coprates complex) is structurally controlled. However, the northern valleys (Ophir Chasma, Hebes Chasma, Ganges Chasma and Juventis Chasma) take on oval shapes and are referred to by some authors as "ancestral basins". The formation of this second class of chasmata has been attributed to collapse processes, including via the formation of deep fracturing or large-scale subsidence. As Ophir Chasma contains a dike swarm of a density that is typically associated with the extensional opening of other features in the Valles Marineris complex, some authors have proposed that these ancestral basins may have been initiated as a result of extensional tectonics, but have since been resurfaced and even more deeply eroded by later glacial activity in the chasmata.

Authors have proposed that the Valles Marineris formed as a result of:

• rifting
• horizontal (strike-slip) offset generated by plate tectonic activity
• accommodation of loading associated with a megalandsliding event that formed the Thaumasia Plateau, leaving the Thaumasia highlands as the landslide toe
• The undermining of the surface as melting ice left behind a subsurface void
• The opening and adjoinment of expanding tension cracks
• Lithospheric-scale failure driven by dikes radial from the Tharsis Montes
• The withdrawal of magma, resulting in a subsequent vertical collapse

Rhythmic meters-scale layering was identified on a very high resolution MOC image in 2003, identified within the basal rock layer of the northern and southern walls of the Coprates Chasma (in the east) and in the northern wall of Ius Chasma (in the west). Dark protrusive lineaments interpreted to be dikes and cumulus material from a sill were identified in the Coprates Chasma region for the first time. The outcrops in Valles Marineris were analogized to the layered mafic intrusions of the north Atlantic: Skaergaard intrusion in Greenland and its counterpart in Scotland on the other side of the Mid-Atlantic Ridge, the Rhum igneous complex. The researchers who made these analogies proposed that the initial opening of the proto-Valles Marineris could have invited the same kinds of conditions that led to the formation of the layered mafic intrusions of Skaergaard and Rhum.

Viviano-Beck (2017): mineralogical constraints

Pre-Noachian and early-Noachian crust (exposed as the basal of the three primary geomorphic units mapped in the northern wall of Valles Marineris) first formed, possibly in association with the solidification of the initial Martian magma ocean. This unit is enriched in olivine at its base and low-calcium pyroxenes towards its head. The pyroxene-enriched section of this basement layer is variably enriched with plagioclase. Due either to excavation or by inconsistent hydrothermal alteration (intrusive magmatic activity or impact events), extant mafic and feldspathic minerals became altered to ferromagnesic phyllosilicates. Regional-scale hydrothermal alteration is spatially associated with olivine-rich dikes, and has been chronologically linked to the early Noachian incipient uplift of the Thaumasia Plateau and the Coprates Rise, with activity continuing throughout the Noachian period. No later deformation or uplifting is found to cross-cut these thermally-altered zones in association with the formation of the plateau or the Coprates Rise.

In association with Anderson's Stage 3 tectonic center (Syria Planum), wrinkle ridges propagated concentrically across the Thaumasia Plateau and the Valles Marineris. A thrust fault within Coprates Chasma has been identified to the concentric compressional activity that formed these wrinkle ridges, although the actual observed vertical offset may be an artifact of the rifting of the Valles Marineris system rather than thrusting. Afterwards, Syria Planum-associated volcanic materials were deposited en masse and are observationally characterized by spectral blandness.

The modern morphology of the Valles Marineris was most likely shaped principally by glacial activity over hundreds of millions of years during and/or between the Hesperian and Amazonian periods. Some researchers have interpreted this period of glacial activity to form the so-called "ancestral basin" chasmata, particularly Ophir Chasma.

Thaumasia Highlands

Authors have proposed that the highlands in the southern reaches of the Tharsis Rise formed as:

• An orogeny associated with the uplift of the Thaumasia Plateau
• A fold and thrust belt
• The movement of Thaumasia Plateau as a detached block
• A gigantic megalandslide of volcanic materials propagating from the northwestern sector of Syria Planum, in which the highlands represent the landslide's toe
• Plate tectonism

In the 1980s, the Claritas Rise and the Thaumasia highlands were first interpreted in the 1980s to be "islands" of highstanding exposed basement rock embayed and buried by more recent volcanic activity. The vestigial province in Tempe Terra was also identified at this time as another example of this unit.

In the 1990s, researchers identified nearly three dozen compressional ridge structures across southern Tharsis. The Coprates Rise is one of these structures, and was identified as tectonically controlled because tilted rock layers were clearly observed on one of its flanks. The rise is also paralleled by other ridges at an averaged spacing of around 400km. This region was interpreted as having developed a fold and thrust belt in which the lithosphere experiences a periodic buckling. At the time of their identification, the compressional formation of these features were relatively dated to the late Noachian and early Hesperian.

The complex containing the Claritas Rise, the Thaumasia highlands, and the Coprates Rise are the only Noachian units of Tharsis that are, on average, at a higher elevation than the southern Martian highlands. They are also the only terrains on the Tharsis Rise with evidence of crustal magnetization, suggesting that they are amongst the oldest exposed terrains on Mars. These ancient terrains are thought to have formed during a pre-Tharsis orogeny.

Some authors have proposed that the loading of the pre-Tharsis Thaumasia Highlands province caused the deflection of the underlying crust which would presuppose the existence of a flexural moat lining the periphery of the ancient terrain. Such a moat would imply a negative free-air gravity anomaly indicative of infilling by detrital materials from the highlands, as the topographic uniformity of the area suggests that there is instead a density difference in the rock relative to the adjacent plains terrains. This supposition has been confirmed. Although elevated hydrogen signatures have not been spectroscopically detected within the proposed flexural moat, they may still exist if they are buried at a depth greater than 1m, and may still be invoked to explain the lower density readings that were gravimetrically used to identify the moat's extent. As the moat is dominated by late Noachian and early Hesperian units, the age of the Thaumasia Highlands - whose southeastern section is likely the oldest exposed terrain on the entire Tharsis Rise - can most likely be constrained to the early Noachian at the latest. The formation of the Thaumasia edifice partially coincided with the core dynamo, illustrated by the presence of magnetized terrains on the highlands units.

Alba Patera (Alba Mons)

Alba Patera is the northernmost of the major shield volcanoes in the Tharsis Rise, and is morphologically distinct from all other shield volcanoes on Mars because it is extremely wide (1400km east-west by 1000km north-south) and has an extremely subtle relief (7km prominence). Two swarms of graben, Alba Fossae (west) and Tantalus Fossae (east), are uniquely distributed on the volcanic surface, remaining mostly straight to the south (as Ceraunius Fossae) and the north of the rise but bowing out into an annular configuration before around the peak caldera. The distribution of this graben has not been observed elsewhere on Mars but is has been analogized to the coronae structures of Venus. These graben swarms - with a few exceptions near the summit caldera of Alba Patera - completely postdate the lava flows they cross-cut, which have been dated to the Amazonian. Researchers have subdivided the extensional tectonism on Alba Patera into at least three events:

• Ceraunius Fossae and the linear parts of Alba Fossae, Tantalus Fossae. These faults are partially embedded on the partly-buried Ceraunius Rise, proposed have been be an early Noachian pre-Tharsis massif, and may be associated with pre-existing basement structures that are aligned to a hitherto unidentified Noachian or pre-Noachian tectonic center. These faults may have been reactivated in the late Hesperian to the early Amazonian by the Alba Patera (stage 4) tectonic center.
• the annular extensions of Alba Fossae and Tantalus Fossae around the summit caldera of Alba Patera
• rimless pit crater chains (catenae), manifesting in Phlegethon Catenae, Acheron Catenae (east flank of Alba Patera), Alba Catenae, and Cyane Catenae (southwestern flank of Alba Patera), associated with giant dike swarms that are radially distributed from the central Tharsis Rise.

The concentric graben around the Alba Patera summit have been proposed to result from subsidence tied to either the plutonic cooling of or the emptying (and subsequent deflation) of the underlying magma chamber. Some authors have considered the possible effects of regional-scale stresses as well, considering peripheral flexural stresses induced by the doming of the Tharsis Rise centered near Noctis Labyrinthus. Others have proposed that a diapir from the mantle ascended underneath the present-day location of Alba Patera, inducing the annular graben morphologies, the uplift of the Alba Patera region, and its extensive volcanism. Additionally, researchers have proposed that Alba Patera lies underneath a mantle overturn regime, eroding the crustal root and inducing isostatic uplift of the Alba Patera region; the subsidence of the edifice in the wake of an initial uplift phase would then induce additional extension and compression in the region; and the injection of sills and the corresponding inflation of the overburden rock. Researchers studying the asymmetric throw of these summit-concentric graben identified their morphologies as consistent with modeled requirements for the uplift mechanisms of Alba Patera's formation (a mantle diapir, sill inflation, lithospheric root erosion, and regional Tharsis megaplume activity).

The northern and eastern flanks of Alba Patera display extensive erosion in a splayed and occasionally dendritic fashion. The units composing this section of the Alba Patera edifice has been historically identified as friable material, which had suggested that Alba Patera transitioned from the classic pyroclastically-driven highland paterae to a Tharsis Montes-style lava flow-dominated structure. However, later studies have re-interpreted these layers as non-pyroclastic in nature.

Alba Patera has been indicated as the closest feature to a corona that has been observed on a planet other than Venus, although two much smaller corona-like structures have also since been identified in Tempe Terra to the east of Alba Patera, also in the Tharsis region. Venerean coronae as characterized by their circular elevated centers buttressed by concentric and radial fractures, which are also characteristic of the morphology of Alba Patera. The formation of Alba Patera - if it is a corona - suggests that it arose due to diapirism, which induces radial extensional stresses near the cap of the plateau. Geomorphic evidences for concentric compressional stresses near the base are not observed, possibly because the higher compressional yield strength of the lithosphere was not exceeded in this situation. Such a diapir would ultimately cool over time, leading to an elastic gravitational relaxation event that ultimately settles to an equilibrium between isostatic and flexural stresses remaining in the lithosphere; this reverses the positions of the stress states expected during the uplift period the uplift period, but at a scale far too low to induce large-scale faulting. The annular extensional faults were attributed to regional fracture patterns from Ceraunius Fossae becoming directionally influenced by the network of concentric fractures that formed about Alba Patera's central caldera.

Structural history of Tharsis

Tanaka (1991) two-stage interpretation

Tanaka (1991) proposed to divide the structural history of Tharsis into two overarching periods: one involving the early Noachian to the early Hesperian, and a second involving the late Hesperian to the Amazonian.

Tanaka (1991)'s initial stage was characterized by extensive volcanism composing Lunae Planum, the northern Thaumasia Plateau such as Thaumasia Planum, in Amazonis Planitia, and on the plains of Terra Sirenum to the south of Daedalia Planum. During this period, the densely dissecting graben swarms of Claritas Fossae, Noctis Fossae, Ceraunius Fossae, Tempe Terra, and Ulysses Fossae are thought to have formed. In addition to the graben swarms, the rifts valleys of Valles Marineris and the concentric wrinkle ridges of the Thaumasia Plateau are believed to have concurrently developed. This initial period is characterized by a tensile hoop stress and compressional radial stress, with centers of tectonic activity identified on Tempe Terra, Valles Marineris, and Claritas Fossae during this phase. Volcanism during this period generally took place in association with these volcanotectonic centers.

Tanaka (1991)'s second stage stretches from the late Hesperian period into the Amazonian. During this period, intensive volcanism led to the formation of the major shield volcanoes of the Tharsis Rise, including the Tharsis Montes, Alba Patera, and Olympus Mons. The lava flows of Syria Planum were also deposited during this time, along with other flow fields within Amazonis Planitia and Ceraunius Fossae. Extremely long graben radial to these mountains formed during this period as well, and as did the Noctis Labyrinthus network. to the east of the Claritas Fossae, the downdropping of a fault block leads to the formation of a rift zone against the Claritas Rupes scarp. Hoop stretches pressed further outwards and regional-scale compressive stresses diminished; the only wrinkle ridges observed are in Chryse Planitia and Kasei Valles, and in association with the central calderas of the shield volcanoes.

Dohm (1999) five-stage interpretation

Dohm (1999) studied the Thaumasia Plateau from the south of the Valles Marineris to south of the Thaumasia highlands, and subdivided the tectonic history of the Tharsis Rise into five different events based on cross-cutting relationships between swarms of radial graben covering the western hemisphere of the planet.

Intrusions in the Thaumasia Highlands, Coprates Rise, and the Thaumasia Plateau have been proposed to have taken placed during the first two stages of the Thaumasia Plateau's history, and are associated with the formation of rifts and of erosional valleys in those provinces of Tharsis. The Valles Marineris is proposed to have begun propagating eastwards across the northern margin of the Thaumasia Plateau. Martian valley networks are associated with the hydrological aftereffects of tectonic and magmatic events across the Tharsis Rise, including the dissection of the Thaumasia highland by the Warrego Valles, principally centered around rifts, volcanoes, and impact craters.

In the early Hesperian, scattered smooth deposits are found to infill areas across the Thaumasia highlands and in cratered regions erstwhile adjacent to the Thaumasia Plateau between the second and third stages of Tharsis tectonism. Wrinkle ridges and radial graben are hypothesized to have begun propagating across the Plateau. The third stage of tectonic activity on the Thaumasia Plateau has also been linked to the abrupt deviation of graben in southern Claritas Fossae as they approach the Warrego Valles. This behavior is thought to be controlled by the presence of more resistant intrusive rock cropping out in the vicinity of the Warrego features. Furthermore, the ridged plains of Sinai Planum and Thaumasia Planum are understood to have formed during Dohm's second and third stages.

In the late Hesperian (Dohm's fourth stage), the lobate lava flow features that extend across Syria Planum and Solis Planum on the Thaumasia Plateau, as well as in Daedalia Planum to the southeast of the Tharsis Rise are thought to have formed, with extensional faulting centered on the Tharsis Montes declining as the rifting of Valles Marineris peaked.

Into the Amazonian period, geologic activity in the Thaumasia Plateau declined widely, excepting lobate volcanic flow features emanating from Arsia Mons into the adjacent Daedalia Planum province.

The impact event that formed Argyre Planitia is likely responsible for a variety of concentric and radial tectonic features extending many hundreds of kilometers away from the impact basin, which lies to the southeast of the Tharsis Rise. Isostatic readjustments in the Martian crust that likely occurred as a result of the impact are associated with the first two stages of activity that Dohm (1999) predicts on the Tharsis Rise. Dohm (1999) asserts that the lower-relief basins and the rises of the southern Thaumasia Plateau may be associated with this tectonic activity, and notes that the Thaumasia Plateau is concentric with the Argyre impact basin and may be formationally linked to the impact event.

Dohm (1999) considered the possibility that plate tectonics might have caused the extension that gave rise to the chasmata of the Valles Marineris, but noted that such extension would have to be accommodated with compression on the southern highland peripheries of the Thaumasia Plateau. However, the researchers noted that the orientations of the Coprates and Claritas Rises cannot be readily explained by this mechanism. They also noted that compressional stresses do not appear to be accommodated by Thaumasia highlands-like structures to the north of the Valles Marineris. Lastly, the uplift of the Thaumasia Plateau concluded in the early Hesperian, long before tectonic activity appears to have peaked in the Valles Marineris network (late Hesperian to Amazonian).

Anderson (2001) tectonic center geochronology

The oldest of the centers of tectonic activity lies beneath the Claritas Rise, a complex interpreted as uplifted basement rock, and it corresponds to an underlying circular magnetic anomaly. Although the massive rift system composing the Claritas Fossae overlies the rise, it is not clear whether or not the formation of the fossae is associated with the formation of the Tharsis Rise or if it predates it because of the overprinting of tectonic activity tied to later centers of activity in Syria Planum and Alba Patera. Additionally, remnants of the pre-Tharsis terrains persist in various rises that punctuate Tharsis; these notably include the Thaumasia highlands and the Coprates Rise, the Uranius Rise, the Ceraunius Rise, and an edifice in what is now southwestern Arsia Mons. Many of these terrains are characterized by extremely dense fracturing, faulting, and rifting morphologies. On the Tharsis Rise, because Noachian-period magmatic activity is often constrained to these major faulting and fracture zones, some authors have suggested that these ancient terrains are representative of sections of the Martian lithosphere that had become detached into blocks or tectonic plates as a result of superplume activity, large impacts, or pre-existing plate tectonic activity in this region. The orientation of the pre-Tharsis terrains in the initial periods of Tharsis' formation strongly suggest that an extremely large drainage basin existed underneath what is now the Thaumasia Plateau.

Into the late Noachian and the early Hesperian epochs, some researchers has posited that the growth of the Arsia Rise was the first vestige of the Tharsis Rise's future shield volcanoes. Around this time period, volcanic and hydrothermal activity was found in the central Valles Marineris and the source of the Warrego Valles outflow channels, with radial and concentric tectonic features forming relative to central Valles Marineris hypothesized to imply another major crustal dislocation. To an even greater degree, the volcanic construction of Syria Planum began around this period. Th is event is associated with even more pronounced faulting and domal uplift than that which was linked to the previous Valles Marineris center. Lava flows tied to this center of activity blanketed the Tharsis basin with the vast quantities of material constituting the presently-observed Thaumasia Plateau. The role of the Valles Marineris (Stage 2) and Syria Planum (Stage 3) tectonic centers in burying the Tharsis drainage basin has been linked to the catastrophic release of fluids associated with the development of the outflow channels surrounding Chryse Planitia and with the northwestern slope valleys (NSVs) of the northwestern Tharsis Rise. The valleys that served as the initial conduits for these catastrophic floodwaters are not clearly visible and may have been buried by later volcanic activity. However, some authors have suggested that the infilled volcanics in these valleys would be more resilient than the valley-carved brecciated basement rock, causing inverted ridge topographies to form after protracted periods of erosion in this area.

Activity associated with the Syria Planum (stage 3) tectonic activity continued to magnify into the early Hesperian, with the continued development of Syria Planum, Tempe Terra, and the Arsia Mons shield volcano accompanied by the initiation of the development of Pavonis Mons, Alba Patera, and Ulysses Patera. Additional floodwater pulses are recorded through both the circum-Chryse outflow channels and the northwestern slope valleys.

The late Hesperian and early Amazonian epochs are characterize by the migration of the tectonic center northwards beneath Alba Patera, associated with the completion of the Tharsis Montes (Arsia Mons, Pavonis Mons and Ascraeus Mons) and the formation of Olympus Mons, the largest volcano in the Solar System. Sheet lavas are observed to extend across vast swaths of the proto-Tharsis Rise, emanating from the Tharsis Montes. This migration triggered another catastrophic floodwater burst through the Chryse outflow channels. Relative to the previous tectonic center underneath northwestern Syria Planum, this tectonic center was likely shorter-lived and tectonically influenced a smaller region relative to the position of the tectonic center.

The modern Tharsis province has existed largely since the middle Amazonian period, with latent bursts of magmatic activity contributing to the continued development of the Olympus Mons and Tharsis Montes edifices, and graben propagation in the vicinity of Alba Patera, coalescing around the final tectonic center's locality between the Alba Patera edifice and Syria Planum. Late stage volcanotectonic pulses associated with this tectonic center are possible in the modern period and are certain to trigger tectonic activity in the future.

Evolution of the interpretations of Tharsis stress and tectonics

Models of stress distributions on the Tharsis Rise in the 1990s

Tanaka (1991) pointed out that most contemporaneous isostatic stress models proffered the following predictions about the stress distribution over the Tharsis Rise:

• radial normal faulting out to 2000km from the Tharsis Rise center
• a transitionary region where strike-slip faulting is to be expected
• radial compression (circumferential morphologies) out to 3500km from the Tharsis Rise center

Tanaka (1991) also noted that contemporaneous flexural loading stress models counterproposed expectations of the following stress distribution:

• radial compression (circumferential morphologies) on the central bulge
• a transitionary region where strike-slip faulting is also to be expected
• circumferential extension on the Tharsis Rise periphery

The notion of plate tectonics on Mars, and the associated implications of Martian lithospheric recycling, was first proposed by Sengör and Jones (1975) but became very controversial after the release of the publication by Sleep (1994).

Banerdt and Golombek (1990): the crustal cap hypothesis

The peripheral distribution of wrinkle ridges and central distribution of graben on the Tharsis Rise during this period is supportive of an origin rooted in isostatically-driven uplift. However, the presence of graben on the rise's periphery in Tempe Terra and Valles Marineris conflictingly supports models of Tharsis stress predicted by lithospheric flexure. Problematically, invoking a flexural model also implies a compressional stress on the center of the Tharsis Bulge where the extensional features of Noctis Labyrinthus are observed instead. Furthermore, Valles Marineris, the Sirenum Fossae, and the graben of Alba Patera stretch continuously across the transitional strike-slip-dominated areas predicted by either model.

Banerdt and Golombek (1990) proposed a model to reconcile these two ideas, explored in greater detail by Tanaka (1991), suggesting that the Martian crust is composed of two separate lithospheric units. The global unit underlying the Tharsis Rise includes a brittle upper crustal layer that becomes decoupled from the underlying upper mantle. As flexural stresses transfer from the mantle to the overlying crust, the detached unit slides freely along its boundary with the mantle and is thus not as responsive to lithospheric flexure near its center as it would be to isostatic processes. This hybrid model predicts a circumferential tensile stress throughout the Tharsis region with some radial compression at Tharsis' peripheries. When the global wrinkle ridge distributions are attributed to the aforementioned global (compressionally-stressed) contractional event, this model broadly accounts for the structural features observed across the western hemisphere of Tharsis.

Wrinkle ridges are the only global-scale structural feature on Mars whose formation does not (necessarily) primarily arise from the formation of the Tharsis Rise, and suggests that - during the late Noachian to the early Hesperian, a period of global compressional stress was in place. This global contraction follows a period of global extension, a residual aftereffect of Martian planetary accretion. Some authors have proposed that this compressional environment could have persisted for almost a billion years before waning, placing the end of wrinkle ridge formation towards the end of the early Hesperian epoch. Conversely and notably, no structural evidence for a global extensional stress field has been observed, which would have suggested that a corresponding planetwide expansion had also occurred.

McGovern and Solomon (1993): Tharsis' transition from plateau-building to volcano-building

The oldest terrains on the Tharsis Rise - particularly the Thaumasia highlands, and auxiliary terrains at its flanks, including the Claritas Rise and the Coprates Rise - have been hypothesized to mirror the formation of large igneous provinces seen on Earth (for instance, the Deccan or Siberian Traps). Such plateau-like structures are observed to extend for thousands of kilometers on Venus, and would do the same if the Thaumasia region formed along the same lines. When the elastic thickness of the crust (T_e) is very thin, any large-scale deflection in the crust will induce compressive stresses that arrest the ascent of magma to the summit of a growing volcano, forcing the magma to press outwards instead and driving the large lateral extents of such features.

In Tharsis and particularly in the Thaumasia region, this trend appears to have shifted in favor of much smaller-scale (hundreds of kilometers) shield volcano features such as the Tharsis Montes and Olympus Mons beyond the early Noachian epoch. Such features manifest when the aforementioned compressive stresses do not reach the threshold required to drive plateau formation. Over time, the elastic thickness of the Martian crust in the Tharsis area increased rapidly, terminating the plateau-building phase in Martian geologic history quickly enough to suppress the formation of the corona plateau features that have been observed on Venus, with the sole possible exception of the Thaumasia region.

Mège and Masson (1996): the mantle plume hypothesis

An underlying superplume has been invoked to explain the extremely voluminous magmatic activity during the Noachian, thought to have formed the Tharsis Rise and to have controlled virtually all of its tectonic features into the Hesperian.

A variety of landforms are observed on the Tharsis Rise that are almost diagnostically attributed to mantle plume activity on Earth. The Rise is characterized by very large flow fields, large igneous plateaux and deposits believed to originate from pyroclastic flows (both air-fall and ash-flow). Critically, the volcanic centers of radially-oriented tectonic activity observed by Anderson (2001) could manifest as distinctive episodes of increased superplume activity, and are strongly analogous to the activity of plumes studied on Earth by their associated patterns of radial diking (and graben nucleation), uplift, and concentric contractional tectonic feature formation. The Thaumasia highlands also contain evidence of explosive eruptions from volcanic fissures.

In 2012, Ezgi Karasözen (a doctoral student under the advisement of Jeffrey Andrews-Hanna) analyzed the presence of en echelon concentrically-oriented ridges to the southwest of the Tharsis Rise, stretching across Terra Sirenum and Terra Cimmeria. These features were interpreted as extensional, the outcome of the same "wide rifting" processes that gave rise to the Basin and Range Province of the southwestern United States and northwestern Mexico. The presence of such ridges has been studied as a feasible indicator of the mantle plume hypothesis, a possible peripheral flexural response to the incipient Tharsis megaplume that has since been largely overprinted by other younger features. From the pre-Noachian period, preceding the formation of the Tharsis Rise, and up towards the Noachian-Hesperian boundary the migration of this mantle plume has been evoked to explain the en echelon staggering of the large ridges. Karasözen proposed that mantle material lifted up by the plume must necessarily downwell; if this downwelling occurred in the western Terra Sirenum region, delaminating the lower crust in this region, entraining it along its flow path and recycling it. Later, the onset of widespread volcanism associated with plume activity was accompanied by the transition of the radial graben swarms characteristic of mid-stage to late-stage Tharsis tectonism.

Both Olympus Mons and the Tharsis Montes register a dramatic drop in volcanic activity into the early Amazonian. The decrease suggests a diminishing magma supply, giving rise to episodic bursts of magmatic and volcanic activity rather than a continuous and sustained output as has been observed in the Noachian and the Hesperian. If a mantle plume formed these features, this shift could be explained by the movement of the plume away from this section of Mars (as terrestrial-style plate tectonics is not possible on the planet's lithosphere), or the activity of the plume ceased around this time. The mass and breadth of this plume could be driven by convection and/or thermal blanketing as a result of the plume's activity beneath the thickened crust of the Tharsis Rise.

Anderson (2001): the five tectonic centers

Anderson (2001) proposed a model identifying five distinct formational stages of activity on the Tharsis Rise, each attributed to the activation of a tectonic center about which swarms of graben are radially oriented.

The oldest center for tectonic activity lies underneath the Claritas Fossae and affects the extant Noachian terrains of the time period, including Tempe Terra, Ceraunius Fossae, and the Thaumasia highlands. The second identified center lies underneath the Coprates Chasma and Melas Chasma region of the Valles Marineris, and is the only tectonic center that is not longitudinally aligned with the initial Claritas center. The third tectonic center exists underneath the Tharsis Montes, associated with the lava flows of the Thaumasia Plateau including Syria Planum and Solis Planum, and structures such as Noctis Labyrinthus. The fourth tectonic center is centered underneath Alba Patera. The fifth and final tectonic center migrated southwards between the onset of the Ceraunius Fossae and the Tharsis Montes.

The greatest number of tectonic features observed on Tharsis are associated with the oldest tectonic center of the Tharsis region, underneath the Claritas Fossae. The majority of the Tharsis edifice is understood to have been volcanically constructed during this time, maintaining roughly the same form throughout the remainder of its history.

In this study, Anderson and co-workers identified that the regional principal stress directions of Mars have remained relatively similar throughout the early history of Mars. When comparing gravity and topography data, it is also clear that the mechanisms controlling stress distributions on Tharsis have remained the same well into the late Amazonian.

The geographical center for stress and strain in the Tharsis Rise is understood to lie between the northeastern hinge of Syria Planum and Pavonis Mons, the central of the three Tharsis Montes. All other features in the Tharsis Rise - the Valles Marineris, Tempe Terra, Alba Patera, the Acheron Fossae, Olympus Mons, and most of the Thaumasia Plateau are peripheral features of the rise, and were constructed over time scales that were not as protracted as that of Syria Planum and the Tharsis Montes. The Stage 3 tectonic center (Syria Planum) is considered to have been the most active and longstanding of the tectonic centers proposed by Anderson (2001).

Anderson (2019): plume activity as a major driver of Tharsis activity

Anderson (2019) repositioned the stress model proposed by Banerdt (1992), which was previously centered near Anderson (2001)'s Stage 3 (Syria Planum) tectonic center, on their Stage 1 (Claritas Fossae) tectonic center and rotated it slightly. The resulting stress model became consistent with all extensional features associated with the South Tharsis Ridge Belt's basin and range topography within Terra Sirenum; the extensional features radiating off of the Tharsis Montes onto the western flanks of the Tharsis Rise. The model was not consistent with certain trends in Terra Sirenum that did match the orientations predicted by a Stage 3 centering of the Banerdt (1992) model. If the majority of Noachian-age faulting on the western flank of the Tharsis Rise was associated with the initial upwelling of Tharsis associated with the Stage 1 Claritas center, this strongly suggests that a mantle plume - initially centered on the Claritas Rise - was the major driver of tectonic activity rather than plate tectonics. Then, this mantle plume later wandered northwards towards the Stage 3 center to form these other extensional features predicted by Banerdt (1992).

Observational history

David Scott et al took crater counts of partially-buried craters to justify an interpretation on the geochronology of the Tharsis Montes, Olympus Mons, Alba Patera, and the Ceraunius Fossae, and their associated geologic units.

In 1989, Richard A. Schultz (NASA/Goddard Space Flight Center) described the presence of strike-slip faulting in the vicinity of Thaumasia Planum and the Coprates Rise in the eastern reaches of Tharsis, including transpressional stepover features. These were not found to conform to isostatic models of principal stress distributions on the Tharsis Rise.

In 1991, Kenneth L. Tanaka (United States Geological Survey), Matthew P. Golombek and W. Bruce Banerdt (Jet Propulsion Laboratory) modeled stress distributions on the Tharsis Rise in the presence of a hypothesized "detached crustal cap" and found that it corresponded well with observed tectonic structures on the Tharsis Rise.

In 2001, Robert C. Anderson, Matthew P. Golombek, Albert F.C. Haldemann, Brenda J. Franklin, (Jet Propulsion Laboratory), James M. Dohm (University of Arizona), Kenneth L. Tanaka, Juan Lias (United States Geological Survey), and Brian Peer (University of Pittsburgh) mapped radially-oriented swarms of graben across the western hemisphere of Mars and projected them in great circles across Mars to create a beta diagram. Anderson and co-workers identified five poles upon which the great circles of the graben converged, and established a new geochronology for the Tharsis Rise structured around these five tectonic centers.

In 2003, Williams et al noticed layers in the walls of Coprates Chasma and analogized it to the layered mafic intrusions of Skaergaard and Rheum, related to the hotspot underneath Iceland on the Mid-Atlantic Ridge.

In 2003, Davis published his thesis through the Colorado School of Mines, elaborating on past work relating to the formation mechanisms behind the Valles Marineris.

In 2003, Jeffrey Plescia reported on the unusual characteristics of Tharsis Tholus.

Anguita et al. performed geological and structural analyses of the greater Thaumasia region, including not just the Thaumasia Plateau but also Daedalia Planum, Aonia Terra, and the Nereidum Montes.

In 2007, James M. Dohm, Victor R. Baker (University of Arizona), Shigenori Maruyama (Tokyo Institute of Technology, and Robert C. Anderson (Jet Propulsion Laboratory) published a book chapter in a textbook called Superplumes: Beyond Plate Tectonics, describing the morphological manifestations of plume tectonics below the Tharsis Rise, offering an explanation for a plume-driven explanation for the geochronology of the range. Paleotectonic reconstructions were offered at each stage of Anderson (2001)'s tectonic center model for the western Martian hemisphere.

In 2008, Jeffrey C. Andrews-Hanna, Maria T. Zuber (Massachusetts Institute of Technology) and Steven A. Hauck, II (Case Western Reserve University) identified and characterized lineaments in Terra Sirenum and Amazonis Planitia as strike-slip faults, identified the relationships that strike-slip faults tend to have with radially-associated features such as graben and wrinkle ridges, and then compared their distributions with models of faulting on and around the Tharsis Rise. The authors used their mapping of strike-slip faults in this area to validate regional stress models at varying lithospheric thicknesses, eventually settling on a thickness range of 35km to 100km at the time of the Tharsis loading, and used their findings to constrain the period of global contraction responsible for forming the global Martian wrinkle ridge system. From there, the authors then compared and validated competing interpretations of the thermal history of early Mars, ultimately noting that the Martian surface carries no apparent signs of an early rapid cooling following accretion.

In 2009, James Dohm, Victor R. Baker, Shawn J. Wheelock, (University of Arizona), Robert C. Anderson, Lucas Scharenbroich (NASA/Jet Propulsion Laboratory, Jean-Pierre Williams (California Institute of Technology), Javier Ruiz (Museo Nacional de Ciencias Naturales, Madrid), Patrick C. McGuire (Washington University in St. Louis), Debra L. Buczkowski (Applied Physics Laboratory), Ruye Wang (Harvey Mudd College), Trent M. Hare (United States Geological Survey), J.E.P. Connerney (NASA/Goddard Spaceflight Center), Justin C. Ferris (NOAA), and Hirdy Miyamoto (University of Tokyo) presented a discussion on the composition and the history of the Claritas Rise, concluding that it predates the development of the Tharsis province. Among other projects, spectral data of the Claritas Rise from the Thermal Emission Imaging System (TES) aboard the 2001 Mars Odyssey was subjected to supervised machine learning classification (maximum likelihood classification, support vector machines, and Gaussian processing classification) and unsupervised machine learning cluster analyses to automatically map compositional differences between mountain-forming and plains-forming materials on the Rise.

In 2010, Bethany Ehlmann et al evaluated the distribution of serpentine globally across Mars based on CRISM spectral data.

In 2016, Crown et al evaluated the thermophysical and geomorphological characteristics of a section of lava flows to the southwest of Arsia Mons, towards Daedalia Planum.

In 2016, Hood et al spectrally evaluated the bulk composition of the eastern Tharsis Rise using TES data, and make comparisons using various elemental ratios to draw conclusions about the formation of the Thaumasia Plateau.

Unsorted indicators

Isostatic/flexural loading stress models of Mars were published by:

• Banerdt (1982)
• Banerdt (1991)
• Sleep and Phillips (1985)
• Phillips (1990)

Phillips (2001) estimated that the Tharsis load would have been formed by a volcanic magma outpouring close to 3 x 10⁸ km³.

Paleopole positions cannot be determinately inferred from the polarization of magnetized terrains on Mars, as the same magnetization pattern can be generated by infinitely many magnetic fields with infinitely many paleopoles.

References

This article "Interpretations for the formation of the Tharsis Rise" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Interpretations for the formation of the Tharsis Rise. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.

---