46th Lunar and Planetary Science Conference (2015)
STRUCTURE, CANADA. J. R. Weirich1 and G. R. Osinski1,2, A. Pentek3, J. Bailey3 1Dept. of Earth Sciences &
Centre for Planetary Science and Exploration, University of Western Ontario, London, ON, N6A 5B7, Canada,
Dept. of Physics and Astronomy, University of Western Ontario, London, ON, N6A 5B7, Canada, 3Wallbridge
Mining Company Limited, Lively, ON, P3Y 1L7, Canada. ([email protected])
Introduction: The 1.85 Ga Sudbury impact structure straddles the Superior Province to the north, and
the Southern Province to the south, and is thought to be
a multi-ring impact structure >200 km in diameter [1],
[2]. The Superior Province is an older unit (>2.50 Ga)
and consists of granitic and gneissic rocks, while the
Southern Province is younger (2.45-2.22 Ga) and contains mostly sedimentary rocks [3]. Some of these sedimentary units are still found in the Superior Province
and are referred to as Huronian outliers. Sudbury Breccia (SDBX) is a unique rock type that surrounds the
impact melt sheet, known as the Sudbury Igneous
Complex (SIC). SDBX has angular or rounded clasts
in a black aphanitic matrix, and individual examples
can be 1 cm to 100’s of m wide. SDBX typically contains clasts of the surrounding country rock, and is a
breccia at all scales. It is similar to pseudotachylite
found at the Vredefort Dome in South Africa [4]. In
this work we present geochemistry and petrographic
observations of SDBX in the Superior Province to help
elucidate its formation conditions. Here we present
preliminary data from the Halfway Lake outcrop found
30 km north of the SIC in the Superior Province; data
from outcrops along a transect towards the SIC will be
Sudbury Breccia formation: The most popular
scenarios of SDBX formation are 1) frictional melting/cataclasis during crater collapse [5], 2) shock induced melting or cataclasis [2], or 3) injection of SIC
into the surrounding footwall [6]. Cataclasis (i.e. mechanical grinding of rock into powder) and frictional
melting are typically viewed as two end-members of a
continuous progression from initial shearing (when the
rock is cold) to late stage shearing (when the rock is
hot). Scenario 1) is similar to tectonic pseudotachylite
found in earthquake fault zones, albeit on a much larger scale. Scenarios 1) and 2) require the composition of
the breccia to be a mix of the composition of the surrounding country rock, while scenario 3) requires the
composition of the breccia to have a component from
the SIC. Lafrance and Kamber [2] has shown that a
contact between sandstone and diabase near the SIC in
the Southern Province has no contribution from the
SIC, casting doubt on scenario 3). However, that study
was only at one location and may not be applicable to
the entire basin. Scenario 2) works well for small (~1
cm) SDBX occurrences, but it is unlikely that large
fluctuations in shock pressure would be present at the
scale of 100’s of meters. A criticism of formation due
to frictional melting is that once melting begins it will
lubricate the sliding surface, which would inhibit further production of melt. Melosh [7] suggests that melt
could extrude into adjacent low pressure zones in the
country rock, thus keeping the sliding surfaces unlubricated, or that the melt was viscous enough to sustain
shear, though both of these processes require strict
conditions which may not be matched by SDBX formation conditions.
Samples and Methods: Since SDBX has a large
clast content we attempted to improve upon the quality
of SDBX data presented in other studies by picking out
these clasts. Hand samples were cut into ~1 cm cubes
on a circular saw and 40–60 g of the most clast-poor
material was selected for picking and geochemistry.
Samples were sieved, and the 1–2 mm size fraction
was picked of all clasts >1 mm to obtain “matrix enriched” material. Typically about half of this size fraction had clasts >1 mm. We found that attempting to
pick all clasts >0.5 mm was slow (~1g/hr) and required
a larger initial mass to obtain enough for geochemical
analysis. Smaller size fractions of the crushed samples
were analyzed to provide a “whole rock” comparison
to the “matrix enriched” portion.
To test the variability of SDBX and country rock
across the Halfway Lake outcrop, multiple samples of
each were collected. Given that clasts in SDBX can be
10s of meters in size, it is sometimes hard to distinguish between host rock and large clasts. Here we
adopt the term “country rock” since we are attempting
to sample host rock, but acknowledge this may not be
the case. Five samples of SDBX were collected from a
50 m trek along an outcrop, and three of these samples
(the two northern most and the southern most) were
used for clast picking. We also collected two samples
of diabase, three samples of granite/gneiss, and one
sample of a rock type that is similar to siltstone found
in a Huronian outlier 4 km to the south. Further investigation of this “unknown” rock is underway.
Geochemistry: Major and trace elements, including the Rare Earth Elements (REE) were determined
for each sample. Picking had little affect on the silica
and mafic element concentrations; all other elements
had similar small differences. Hence, the “matrix enriched” samples will not be discussed further, and are
not included in any of the figures.
46th Lunar and Planetary Science Conference (2015)
Figures 1 and 2 show the mafic vs. silica content
and REE signature, respectively, of the matrix and
country rock. SDBX appears to be intermediate to the
country rocks in both major and trace elements, though
the REE have a higher concentration than both the
diabase and granite/gneiss. Since the REE are relatively immobile, they indicate that a contribution from the
“unknown” sample is probable.
To test this hypothesis, we performed a component
decomposition on Si for each SDBX sample using the
average of the two diabases, the average of the three
granite/gneisses, and the “unknown” as three end
members. We then selected the combination that gave
the lowest Chi2 for the REE, plus the immobile ele-
Figure 1. Harker diagram showing SDBX is intermediate to country rocks. Note legend.
Figure 2. Spider diagram showing SDBX intermediate to country rocks. Symbols same as Fig. 1.
ments Ti, Zr, Y, Al, Th, and Nb. We then calculated
the P-value (the probability the Chi2 would be that
poor by chance) for that combination of end members.
Th and Nb have a higher concentration in the SDBX
than any of the three end members and give large errors, resulting in a P-value <0.001 which indicates a
poor model. However, if we don’t include Th and Nb
when choosing the lowest Chi2, the P-values are >0.01
(>0.1 in four out of five cases) which indicate a good
fit. While the large Th and Nb errors are concerning, it
is remarkable how the majority of the elements are
reproduced using a combination of the three country
rocks, as can be seen in Figures 3 and 4. In these models granite/gneiss is 38–55%, diabase is 22–42%, and
Figure 3. Harker diagram of SDBX (blue) and
three component model values (green).
Figure 4. Spider diagram of SDBX (blue) and
three component model values (green).
“unknown” is 17–23% of the breccia. Replacing “unknown” with SIC as the third end member does give
reasonable P-values, however the large distance (30
km) to the SIC makes it an unlikely contributor. Future
work will explore the SIC contribution in more detail.
Conclusions: As found by other authors, diabase
is a larger component of the breccia than one would
expect given the diabase abundance at the outcrop;
perhaps the result of preferential melting of mafic material during shear [5]. We have found an unknown
rock type, perhaps sediment from a Huronian outlier,
which can explain the REE abundance in the breccia.
The abundance of this rock type at the outcrop has not
been determined, but the nearest known Huronian outlier is 4 km away. Our working hypothesis is that breccia melt has been transported large distances, either via
intrusion of melt into the country rock [7] or due to
faulting over many kilometers [8].
References: [1] Krogh T. et al. (1984) The geology and ore deposits of the Sudbury structure, 431-446.
[2] Lafrance B. and Kamber B. (2010) Precambrian
Res., 180, 237–250. [3] Rousell D. (2003) EarthScience Rev., 60, 147-174. [4] Reimold. W. (1995)
Earth-Science Rev., 39, 247-265. [5] Spray J. and
Thompson L. (1995) Nature, 373, 130-132. [6] Riller
U. et al. (2010) Geology, 38, 619-622. [7] Melosh J.
(2005) Impact Tectonics, 55-80. [8] Mungall J. and
Hanley J. (2004) J. Geol., 112, 59-70.