The Anatomy of a Wrinkle Ridge Revealed in the Wall of Melas

46th Lunar and Planetary Science Conference (2015)
Hank M. Cole1 and Jeffrey C. Andrews-Hanna1, 1Colorado School of Mines, Department of Geophysics and Center
for Space Resources, Golden, CO 80401.
Introduction: Wrinkle ridges are long (10’s to
100’s of km), quasi-linear, compressional tectonic features commonly observed on the volcanic surfaces of
the terrestrial planets, including Mars [1,2]. They have
a morphology consisting of one or more narrow (up to
6 km wide), asymmetric ridges superimposed onto a
broad arch (~20 km wide) [1]. Based on terrestrial analogs, they are thought to form when shallow volcanic
plains sequences experience horizontal shortening and
folding due to an underlying blind (not surface breaking) thrust fault which may propagate into the basement material [2-4]. The narrow wrinkles on the ridge
are likely the result of variable backthrust faulting
which causes their sinuous expression [2]. The broad
arch of the ridge has more consistent expression and
likely results from folding above the blind thrust fault
[2]. There is typically a topographic offset across the
ridges, and they often occur in a periodic spacing that
may be indicative of the maximum depth of faulting
and the brittle-ductile transition depth [3,4]. Studies of
wrinkle ridges model surface deformation by varying
the fault properties in an elastic dislocation model in an
attempt to match the surface topography. However, this
approach is indirect and may yield non-unique results,
and these studies often assume a typical thrust fault dip
of 30° [2,4-7]. In this abstract we present direct observations of a wrinkle ridge thrust fault where it is exposed in the wall of Melas Chasma in central Valles
Marineris on Mars.
Observations: The area of interest is along the
south wall of Melas Chasma where a ridge descends 6
km across a distance of 70 km from the plateau to the
valley floor (Fig. 1A). The Melas ridge is aligned with
a wrinkle ridge on the adjacent plateau that intersects
the chasma wall. A profile across the wrinkle ridge
shows a topographic step down to the east (Fig. 1B).
We interpret the Melas ridge as the expression of an
erosionally-resistant fault plane, left as highstanding
relief during the enlargement of Melas Chasma. Resistance to erosion could result from interaction with
magma or water to form an erosion-resistant surface
along the fault. If we assume that the structure of the
Melas ridge preserves the structure of the blind thrust
fault associated with the wrinkle ridge, then we can use
both ridges to constrain the properties of the fault.
Topographic Analysis: We use two approaches to
characterize the fault associated with the Melas ridge.
First, we note a difference of ~6° between the strike of
the wrinkle ridge and the that of the Melas ridge, con-
sistent with the Melas ridge representing the subsurface
expression over a range of depths of the dipping fault
plane. Since both ridges are expected to lie along the
fault plane, we can calculate a possible fault plane
normal vector by taking the cross product of two unit
vectors that align with the wrinkle ridge and the Melas
ridge. This method produces a normal vector for a fault
plane that dips 18° NW with a strike of N17°E.
Figure 1. (A) Plan view of both ridges shown in gridded MOLA topography. (B) Five cross sections in A
were taken to produce the averaged profile from west
to east (shown at 100× vertical exaggeration to display
the broad ridge and topographic offset).
46th Lunar and Planetary Science Conference (2015)
We next used Mars Express High Resolution Stereo
Camera (HRSC) digital elevation models to find the
best-fit plane to the Melas ridge. As before, we assume
that if the ridge was composed of resistant fault plane
material, then the path of the ridge crest will lie along
the fault plane. The data was analyzed by taking the
point of highest elevation in each latitude row to produce a set of points that represent the crest of the Melas
ridge. A plane was fit to this cloud of points using a
least-squares method. The strike calculated from this
plane is N21°E, which agrees with the earlier derived
strike of N17°E and the wrinkle ridge’s measured
strike of N18°E. The dip of the fitted plane of 13°NW
is in reasonable agreement with the earlier cross product approximation of ~18°. The fit of the plane is assessed using the orthogonal distance to the plane for
each point (Fig. 2), revealing an RMS misfit of 298 m
(relative to 2.5 km of relief along the profile).
Figure 2. Plot of the orthogonal distance to the best fit
plane for a fit of a section of the Melas ridge.
Figure 3. Plot of dip averaged over 50 points from
south to north along the Melas ridge. Discontinuities
come from removing spikes over 25°.
We then calculated best-fit dips for 50 point contiguous sub-sections of the data along the length of the
profile in order to test for variations of the dip with
depth. The lack of clear trends in the inferred dip along
Melas ridge over a vertical section of 2.5 km (Fig. 3)
supports the interpretation that erosion preferentially
exposed a planar surface with semi-regular dip. A linear trend in the dip plot would be present if the fault
were listric over this section [5]. The mean dip from
this approach is 11° with a standard deviation of 5°.
Discussion: The results of this analysis suggest that
the Melas ridge has a topographic expression that is
likely controlled by the preferential exposure of an
erosion-resistant fault plane associated with the coincident wrinkle ridge. Two methods using the topography
of the Melas ridge to calculate approximate fault planes
both resulted in similar dip values of 13° and 18° along
a planar scarp extending from a depth of ~1.5 km down
to a depth of at least 4.5 km below the plateau surface.
The dip direction of the thrust fault is in agreement
with that inferred from topographic profiles [2,4].
However, the dip of this thrust fault (13-18°) is more
shallow than the 30° dip that is typically assumed in
models [2,4-7]. The properties of this thrust fault could
be used to constrain models specific to this region, and
may even be generally applicable to wrinkle ridges in
other regions or planets as well. We conclude that at
least in some cases the dips of thrust faults producing
wrinkle ridges are more shallow than expected and that
listric character is not seen in the top several km. If this
dip range applies to all wrinkle ridges, this would imply that horizontal normal strain may be a factor of 1.82.5× greater than estimates based on surface relief, or a
factor of 1.1× greater than estimates based on displacement-length relationships (both assuming a dip of
30°) [7]. Analysis of global contraction using this shallower fault dip would predict that Mars has experienced a greater decrease in planetary radius from cooling and that thrust fault propogation depth may be
more shallow than previously thought.
References: [1] Watters T. R. (1993) JGR, 98,
17049-17060. [2] Schultz R. A. (2000) JGR, 105,
12035-12052. [3] Montési L. G. J. and Zuber M. T.
(2003) JGR, 108, 5048-5073. [4] Golombek M. P. et
al. (2001) JGR, 106, 23811-23821. [5] Watters T. R.
(2004) Icarus, 171, 284-294. [6] Okubo C. H. et al.
(2003) GSA Bull., 116, 594-605. [7] Nahm A. M. and
Schultz R. A. (2011) Icarus, 211, 389-400.