The Valles Marineris dykes in a historical perspective

The whole material below was presented at the 7th International Dyke Conference. The corresponding abstract and poster can be downloaded on the V-MACS publications page.

Dyke emplacement: does high-resolution reality match models?

The plume tectonics model proposed earlier (Mège and Masson, 1996; Mège, 2001; Mège and Ernst, 2001) predicts that Valles Marineris opened parallel to a dyke swarm radiating from Syria Planum, the summit of the Tharsis rise, propagating in a stress field controlled by the Syria Planum magma centre overprinting a regional field (Muller and Pollard, 1977; Mériaux and Lister, 2002). The interpretation that dyke emplacement should play a significant role in the evolution of Valles Marineris has been supported by other works (Wyrick et al., 2004, 2015; Wyrick and Smart, 2009). None of these dykes, however, could be directly observed due to the poor resolution of data available in the 90's, usually 40-100 m/pixel (Viking Orbiter 1 and 2 images). Recently, the existence of this dyke swarm was supported by limited direct observations in the easternmost Coprates Chasma (Flahaut et al., 2011) and its neighbouring plateau (Huang et al., 2012). These observations justified that the plume tectonics model predictions be carefully examined in the light of the more recent data.

Insights from recent datasets

Since 2006, the HiRISE telescope orbits Mars in NASA's Mars Reconnaissance Orbiter spacecraft and takes images of Mars at a resolution of 0.25 to 1 m/pixel. Many sites in Valles Marineris have been imaged (Figure 1), some of which show eroded dykes. Flahaut et al. (2011) and Brustel et al. (2016) identified mafic dykes in eastern Coprates Chasma of orientation predicted by the plume tectonics model. Huang et al. (2012) identified several dykes exposed on the plateau south of Valles Marineris that also follow this trend.

We have been performing a systematic survey of HiRISE images in Valles Marineris in order to identify dykes exposed along the walls and in the troughs, noting their coordinates, orientation, maximum thickness, elevation above the MOLA sphere, and dip angle when not vertical. Segmented dykes are counted as single observations. Seventy-five HiRISE images of Valles Marineris walls have been surveyed to date. They have revealed the presence of more than 300 dykes, of thickness between ca. 1 m (lower detection limit) and 60 m (Figure 1). A few bodies up to 200 m thick may be either dykes or other types of intrusions, with uncertainty resulting from mantling by slope deposits.

Figure 1: Distribution of HiRISE images in Valles Marineris on April 12, 2016 (when the results presented here were obtained), of which 75 that cover the wall slopes have been examined from dyke identification.

The dykes (Figure 2) are mapped from HiRISE images only, which implies that their distribution partly reflects HiRISE image distribution. Several long portions of Valles Marineris walls and floors are not yet imaged.

Figure 2: Some representative dyke exposures on Valles Marineris hillslopes. Arrow points to thickest dyke (maximum thickness in red letters). North is to the top. HiRISE image IDs are indicated.

Figure 3: Dyke distribution in Valles Marineris. The numbers are measured dyke maximum thicknesses. The Ophir Chasma Dyke Swarm (in the yellow box) is made of ~50 thick dykes exposed on chasma floor and discussed on another poster. The white boxes indicate areas of Coprates Chasma and Candor Chasma where dyke direction analysis was performed (Figure 4).

Nevertheless, it is likely, from the current mapping, that dyke concentration is higher in some areas, especially eastern Coprates Chasma and Ophir Chasma (Figure 3), perhaps indicating that these two areas were preferred sites of local magmatic activity. Alternatively, the dykes may have propagated from the Syria Planum area to the Coprates and Ophir chasmata through the western part of Valles Marineris, as implied by Mège and Masson's (1996) plume tectonics model, but too deeply to be exposed in the western area.

Preliminary measurements indicate indeed that most dykes on the Valles Marineris walls are located at a depth > 4000-5000 m below the surface of the surrounding plateau. Although many chasma floors are deeper than this, huge debris slopes mantle the lower part of the walls and in many parts of Valles Marineris bedrock is not exposed at the such depths, which may be the explanation for the absence of exposed dykes in these areas. Wilson and Head (1994) calculated that the depth of the level of neutral buoyancy (LNB) of basaltic magma on Mars was estimated at ca. 11 km, which may be considered as a proxy for mean propagation depth. Due to uncertainty on Valles Marineris bedrock composition as well as dyke composition (most identified dykes are too thin to have their composition inferred from near-infrared spectra from orbit using the MRO/CRISM spectral imaging instrument), the LNB depth inferred by Wilson and Head (1994) only indicates the great depth of propagation of the Valles Marineris dykes, consistent with the observations.

The dyke-related landforms identified by Mège and Masson (1996) and Mège et al. (2003) are generally parallel to Valles Marineris. So are the dykes identified in the survey reported here (Figure 4), which strengthens the plume tectonics model. Nevertheless, some are not parallel to Valles Marineris, and sometimes are additionally parallel to other tectonic structures, such as narrow grabens on the plateau, or the Louros Valles sapping channels. In northern Candor Chasma, unfortunately not yet imaged at HiRISE resolution, the dykes are perpendicular to the Valles Marineris-parallel dykes, and form a true local swarm that indicates that at one moment in the evolution of Valles Marineris, dyke dilation and tectonic stretching occurred perpendicular to the tectonic stretching that produced the Valles Marineris grabens (Figures 4c and 5). Determining the chronology of dyke swarms would help reconstruct the evolution of the crustal stress field with time; unfortunately, only two sites where dyke cross-cutting relationships can be determined have been found to date, making this chronology out of reach.

Figure 4: Dyke orientation has been measured and binned in sectors 15° wide. (a) Orientation of all the dykes mapped in Valles Marineris to date. The mean orientation is subparallel (N100, arrow) to the mean Valles Marineris trough orientation (N105) and to the Syria Planum Dyke Swarm (Figure 1a). (b) Dyke orientation in eastern Coprates Chasma (box in Figure 4). More than 40% of the dykes are parallel to the main dyke trend as found in (a), although the mean orientation (N081, arrow) is parallel to the local orientation of Valles Marineris (N090). (c) Dyke orientation in eastern Candor Chasma (box in Figure 4). Half the mapped dykes trend perpendicular to Valles Marineris, suggesting that at one moment, dyke dilation (and the minimum principal stress trajectory) was parallel, not perpendicular, to the Candor Chasma graben.

Figure 5: NE-trending dyke swarm in eastern Candor Chasma (green box on Figure 4), shown here on a CTX image (5 m /pixel) due to the absence of HiRISE images here. The  white arrows highlight some of them. The dykes are parallel to the dykes in Figure 5c and their emplacement requires graben-parallel dyke dilation. CTX image B09_013086_1749_XI_05S069W; credit NASA/JPL-Caltech/MSSS.







Temporary conclusion (September 2016)

Very high resolution imaging tends to confirm earlier interpretation that the Valles Marineris troughs opened parallel to a dyke swarm, the Syria Planum Dyke Swarm (Mège and Masson, 1996). Looking at details, however, unexpectedly, in eastern Candor Chasma most dykes are perpendicular, not parallel to the troughs. This implies that at one moment in the evolution of Valles Marineris, the orientation of the stress field in the Candor Chasma area was reversed, with s3 ESE-oriented instead of NNE-oriented.

The tectonic history of Valles Marineris has been thought to be as simple as N-S to NNE tectonic stretching, perpendicular to the troughs and to regional dykes. We have here evidence that the story was substantially more complicated. Continuation of HiRISE image survey will help understand this enigmatic feature, which is likely to have implications for the whole Syria Planum-Tharsis bulge.

Cited works

Brustel C et al. (2016) Valles Marineris tectonic and volcanic history inferred from dikes in eastern Coprates Chasma. Houston, Texas, 47th Lunar Planet. Sci. Conf., Abstract #2724.

Flahaut J et al. (2011) Dikes of distinct composition intruded into Noachian‐aged crust exposed in the walls of Valles Marineris. Geophys. Res. Lett. 38, L15202, doi:10.1029/2011GL048109.

Huang J et al. (2012) Identification and mapping of dikes with relatively primitive compositions in Thaumasia Planum on Mars: Implications for Tharsis volcanism and the opening of Valles Marineris. Geophysical Research Letters 39, L17201, doi:10.1029/2012GL052523.

Mège D (2001) Uniformitarian plume tectonics: the post-Archean Earth and Mars. In: Ernst RE, Buchan KL (eds), Mantle plumes: Their Identification through Time. Geol. Soc. Am. Spec. Pap. 352, 141-164.

Mège D, Ernst RE (2001) Contractional effects of mantle plumes on Earth, Mars and Venus. In: Ernst RE, Buchan KL (eds), Mantle plumes: Their Identification through Time. Geol. Soc. Am. Spec. Pap. 352, 103-140.

Mège D, Masson P (1996) A plume tectonics model for the Tharsis province. Planet. Space Sci. 44, 1499-1546.

Mège D et al. (2003) Volcanic rifting at Martian grabens. J. Geophys. Res., 108, doi:10.1029/2002JE001852.

Mériaux C, Lister JR (2002) Calculation of dike trajectories from volcanic centers. J. Geophys. Res. 107, B4, 2077, doi:10.1029/2001JB000436.

Muller OH, Pollard DP (1977) The stress state near Spanish Peaks, Colorado determined from a dike pattern. Pure Appl. Geoph. 115, 69-86.

Wilson L, Head JW, III (1994) Mars: review and analysis of volcanic eruption theory and relationships to observed landforms. Rev. Geophys. 32, 221-263.

Wyrick DY, Smart KJ (2009) Dike-induced deformation and Martian graben systems. J. Volcanol. Geotherm. Res. 185, 1-11, doi:10.1016/j.jvolgeores.2008.11.022.

Wyrick DY et al. (2004) Distribution, morphology, and origins of Martian pit crater chains. J. Geophys. Res. 109, doi:10.1029/2004JE002240, doi:10.1029/2004JE002240.

Wyrick DY et al. (2015) Physical analogue modelling of Martian dyke-induced deformation. In: Platz et al. (eds) Volcanism and Tectonism across the Inner Solar System, Geol. Soc. Sp. Publ. 401, doi:10.1144/SP401.15.