The biggest pile of sand the world has ever seen

The Moine, a set of sedimentary rocks found in furthest north-west Scotland have enjoyed at least three cycles of metamorphism and deformation. My only sample from here is a migmatitic gneiss, so when I heard about people studying sedimentology in the Moine, my mind was a little bit boggled – gneisses have lost all trace of their sedimentary history. Once I read a bit more I discovered a story about sand. Lots of sand. The biggest pile of sand the world has ever seen, in fact.

High grade migmatitic gneiss with folded granite vein - Moine from Scotland

Metamorphic and sedimentary

Formally ‘the Moine’ is the Moine Supergroup of early Neoproterozoic age. This is a phrase the makes me think of a 1970s prog rock band but it is a stratigraphic term. The Moine supergroup is made up of groups like the Glenfinnan and Morar groups, each of which is a recognisable package of sediments, representing a period of time. The Moine is defined as a group of sediments that formed in a single sedimentary basin.

How did sediments end up looking like the picture above? This sample of gneiss started as mud but has since been buried and heated so it  was metamorphosed (it contains garnet and biotite), melted (the vein of granite) and deformed (the lovely folds). We know it was once mud only from its bulk chemistry. However I’m a big fan of deformation and metamorphism, so I picked a sample that shows these things. A more representative rather different sample of the Moine is seen in this picture.

Moine crossbeds from zeesstof on Flickr

This is a sample of metamorphosed sandstone, or psammite in local parlance. A practiced eye will quickly pick out the bedding (horizontal) and cross-bedding (sloping lines). These are primary sedimentary features and if you see enough of these, you can work out the conditions under which the sediment formed.

Why are these traces of an ancient river or sea-bed preserved? There are two reasons.

Firstly the rock is made up mostly of quartz, and so is not affected by metamorphic reactions under normal conditions. During metamorphism the quartz may be recrystallised, but fine structures like cross bedding can survive. Note that some of the lines above are a slightly different colour – these are rich in ‘heavy minerals’ such as Zircon.

Secondly, conditions of metamorphism and (particularly) deformation vary depending on where you are. The mylonites of the Moine Thrust zone were once sandstones, but no bedding can be seen in them now. The Moine rocks have been intensely deformed, but particular areas such as the hinges of folds still contain original sedimentary structures.

Geological paradise

NW Scotland geology map

The map above shows Northwest Scotland. The key structure is the Moine Thrust (MT) which runs parallel to the mainland coast. To the west are foreland rocks – ancient basement gneisses (Lewisian), Proterozoic sandstones (Torridonian) and a thin sliver of Cambro-Ordovician sediments. These last contain fossils that link them to rocks of similar age in North America. All these rocks formed before the Atlantic existed, in fact most formed before the ocean before the Atlantic. For the rest of this story we are in North America, geologically speaking.

East of the Moine Thrust are the Moine rocks, deformed, metamorphosed and containing igneous intrusions, in stark contrast to the sediments to the west…

Torridonian sandstone on Liathach, courtesy of Sandchem

This lovely picture of undeformed Torridonian sandstones, with a white cap of Cambrian sandstone highlights the different geological histories either side of the Moine Thrust.

How to explain the difference? One possibility is that the Moine rocks are exotic, brought in from elsewhere as part of an allochthonous terrane. Much of Alaska is made up of large chunks (terranes) which have very different histories but have all ended up being scraped onto the American continent. In the 1990s people were finding applying this then-fashionable concept everywhere, including Scotland. Or possibly the Moine and Torridonian are equivalent; twins separated at birth, who experienced very different histories as they grew older. To decide between these options, what you need is something akin to DNA for sedimentary basins.

Long-lost twins

The ‘heavy minerals’ in sediments are literally heavy. Geologists crush up samples and put them into dense liquids. The quartz floats, the heavy fraction that sinks is separated and analysed. Zircon grains are common heavy minerals and they have a very useful feature – they contain lots of Uranium which is tightly locked in. If you analyse them isotopically, you can tell the age at which they formed, even if they’ve subsequently been metamorphosed.

These ‘detrital Zircons’ will have a range of ages, reflecting the age of the rocks that were eroded to form the sediment. This may have a distinctive pattern that gives a picture of the eroding rocks surrounding a sedimentary basin. This is like DNA for sediments.

Work led by Maarten Krabbendam of the British Geological Survey strongly suggests that parts of the Moine and Torridonian are direct equivalents – twins. Rocks of the lowest part of the Moine, the Morar group have detrital zircon age profiles that are very similar to rocks from the main part of the Torridonian. They are more similarities indeed than between the Morar group and the higher parts of the Moine. Krabbendam and colleagues state that the Torridonian and lowest Moine are “similar in terms of lithology, stratigraphical thickness, sedimentology, geochemistry, detrital zircon ages and stratigraphical position on Archaean basement” which is a compelling case that they are two parts of the same basin.

Lots and lots of sand

The link between Moine and Torridonian tells us that about a billion years ago there was a sedimentary basin covering northern Scotland, containing 6-9 kilometres thickness of sand. To get that much sediment, you need to make a hole to put it in. What formed the sedimentary basin the Torridonian and Moine sediments were deposited in?

To find out, Maarten Krabbendam and Helen Bonsor at the British Geological Survey (with others) have been looking at the Moine rocks as sediments. Detailed field work allows them to create maps and sedimentological logs. Logs are vertical graphical descriptions of many things such as grain-size (mud or sand), sedimentary features or direction of water flow inferred from ripples. Remember that these are highly deformed rocks. The bedding is very rarely flat, indeed sedimentary features may be the only way of telling if the rocks are upside down or not, such is the intensity of folding in the area.

All this hard-won data can tell you a lot. Sedimentary features and patterns allow you to infer water depth. As an example, to quote Bonsor et al: “The presence of pinstripe, flaser and lenticular bedding, bidirectional ripple crosslamination and abundant mud drapes indicates that deposition was strongly tidally influenced“, in other words in relatively shallow sea-water. Other parts of the sequence contain hummocky cross stratification which is associated with deposition in deeper waters.

Add all this up and you get a picture of the sedimentary basin through time. Geologists have done this for many basins where the cause of basin-formation is known. Comparing these with the Moine/Torridonian lets us infer how this particular basin formed.

Many workers have interpreted the Torridonian and Moine as separate rift basins. These are where active faulting makes space in a relatively narrow area. The characteristics of sediments in these basins tend to change rapidly horizontally (over location in the basin) and vertically (over time). Near faults you expect coarse fan deposits. None of these things are seen in the Moine or Torridonian, which were deposited under relatively uniform conditions.

Gentle stretching of continental crust forms ‘contintental sag’ basins. Again this isn’t a good fit for the Moine/Torridonian, in part because sag basins are not often so deep.

Bonsor and Krabbendam argue that the best candidate for the Moine/Torridonian is a foreland basin, like the modern day Ganges. A mountain range pushes down onto the continental plate next to it, forming a hole that is quickly filled by sediment eroded from the mountains. These basins can be wide and deep, with long-lived stable conditions close to sea-level as the bending down of the crust and the amount of sediment arriving reach a balance.

The American connection

If the Moine/Torridonian basin formed next to a set of mountains, there should be traces of these mountains somewhere in the geological record. Which orogeny formed these mountains?

There are no fossils in these old rocks, but detrital zircons give a constraint on the maximum age of sedimentation. The sedimentary basin is around a Billion years old, and tectonic reconstructions put the basin near the active Grenville orogeny. This is a major event that created a belt of metamorphic rocks found across the North American continent.

A large range of mountains produces a lot of sand. A billion years ago there were no land plants – nothing like Rhododendrons that stabilise slopes in the Himalayas – so rates of erosion would be higher. It’s also been suggested that the Grenville mountains were located so as to get monsoonal weather patterns. The combination of high precipitation and no vegetation would have caused extremely high erosion rates and created an awful lot of sand. Perhaps the biggest pile of sand the world has ever seen.

A question: If there is a Grenvillian foreland basin in Scotland, why can’t I find anything out about an equivalent in North America? I can find references to equivalent basins in Greenland and Scandinavia but not America itself. Anyone care to suggest why not? [See interesting comments from F below]

References and further reading
H. C. Bonsor, R. A. Strachan, A. R. Prave, & M. Krabbendam (2012). Sedimentology of the early Neoproterozoic Morar Group in northern Scotland: implications for basin models and tectonic setting Journal of the Geological Society DOI: 10.1144/0016-76492011-039

H.C. Bonsor, R.A. Strachan, A.R. Prave, & M. Krabbendam (2010). Fluvial braidplain to shallow marine transition in the early Neoproterozoic Morar Group, Fannich Mountains, northern Scotland Precambrian Research DOI: 10.1016/j.precamres.2010.09.007

MAARTEN KRABBENDAM, TONY PRAVE, & DAVID CHEER (2008). A fluvial origin for the Neoproterozoic Morar Group, NW Scotland; implications for Torridon–Morar Group correlation and the Grenville Orogen foreland basin Journal of the Geological Society DOI: 10.1144/0016-76492007-076

The British Geological Survey now puts its research papers into an Open Access archive. Good for them. Encourage them by going and reading the actual papers I’m talking about.

I’ve talked most about Bonsor et al. (2012) as its the most recent. Bonsor et al. (2010) covers different ground using the same techniques. Looking at a different aspect of the Moine, Krabbendam et al. (2011) summarise some of the structural work the BGS are doing in parallel with the sedimentary studies.

Location map taken from Krabbendam et al 2008 with implicit permission of Geological Society of London

Torridon photo with kind permission of David Shand (Sandchem on Flickr).

Moine cross-beds from zeesstof on Flickr under Creative Commons

The BGS have a new feature of embedded geological maps in your web pages. I’ve had a go below, centred on the North-west Highlands.


Categories: metamorphism, Scotland, sediments

Comments (11)

  1. Andy Duncan says:

    I was about to shout at you about copyright until I saw the reference to my image near the bottom of the page. It would have been polite to have asked me first. Anyway….

    I do not know what evidence you would cite to say that the cross-bedding (which is anything but representative of the Moinian – cross bedding is very rare in high grade metamorphics such as this) is marine in origin. Your comments?

    • Metageologist says:

      Your remarks are fair, thanks for the feedback. I’ve tightened up the text a little in this area.
      About requesting permission – not required under creative commons, but certainly polite, as you say. If I put a comment on a Flickr image, is that a reliable way of contacting the owner or is there a better method?
      Thanks for the lovely image!

      • Andy Duncan says:

        Thanks, and you are most welcome.
        You are completely correct about the license, and I have added CC licenses to many of my shots. I was only momentarily concerned that the image did not directly link back to Flickr until I found the acknowledgment further down the page.
        Yes, leaving a comment as an “invitation to use” is probably the easiest way, but personally I’d only be concerned if it were a shot that was designated “all rights reserved”. I have changed licenses on request in the past.
        Nice blog by the way.

  2. F says:

    The basins have been buried and hidden by successive orogenies. Ohio is almost nothing but sediment all the way down. 😉

    It was a damn interesting question you asked, so I had to look, and these are a bit of what I got for my initial poking of the internet:

  3. Metageologist says:

    Very interesting links! Thanks.

  4. Dave Carlson says:

    Neptunism Revisited:
    (Mantled Gneiss Dome Problem, Migmatite Felsic/Mafic Differentiation and Metamorphic Folding)

    An alternative hypothesis suggests mantled gneiss domes are aqueously-differentiated 100 km and larger TNOs (trans-Neptunian objects) whose geochronology dates to the spiral in merger of their binary pairs in the inner Oort cloud (IOC).

    Dwarf planets typically arrive in the inner solar system in the form of larger core accretions of numerous TNOs and comets as they spiral in along the ecliptic due to galactic or other perturbations which erode their angular momentum, but they only have a good chance of catching up with the two inner solar-system planets with the most circular obits, Venus and Earth, since the almost-perfect hemicircular perihelia of objects in long-period orbits potentially overlap along a sizeable portion of Earth’s and Venus’ circular orbits, whereas, the more-eccentric orbits of Mercury and Mars at best only converge at two points.

    The K-Pg extinction event, 66 Ma, may have contributed the continental rock of far-East Russia, Alaska (minus the North Slope and the mountainous semicircle below Denali National Park with its contingent of terrestrial dinosaur fossil footprints washed down the Yukon River) and the balance of the North American Cordillera, complete with dwarf-planet aquatic fossils of the Burgess Shale. Cretaceous granite plutons and batholiths of the K-Pg dwarf planet core (granite plutons: aqueously-differentiated comet sedimentary cores followed by plutonic melting) as the perihelion of the dwarf planet spiraled in through the the densest concentration of comets at the inner edge of the IOC

    Felsic leucosome layers precipitate near the cold ice-water ceiling in core TNO salt-water oceans (eventually becoming ‘water logged’ and sinking onto the mafic core in leucosome layers) while mafic melanosome layers precipitate directly over the warmer growing sedimentary core, where the authigenic mineral-grain size is dependent on buoyancy of the micro-gravity oceans.

    ‘Circumferential folding’ occurs during diagenesis, as the core shrinks by expelling hydrothermal fluids, dramatically reducing the circumference and volume of the sedimentary core. Authigenic sandstone/quartzite, carbonate rock and schist mantling rock precipitates from the hydrothermal fluids, and thus the layering in the migmatite core and (mantled) gneiss dome covering rock is merely the result of sedimentary growth rings. Circumferential folding does not occur during diagenesis on Earth since Earth has such an enormous diameter, and thus the horizontal terrestrial sedimentary layers of the Grand Canyon following the Cambrian Period are not folded, but the tilted extraterrestrial layers of the Grand Canyon Supergroup exhibit circumferential folding.

    • Metageologist says:

      Thanks for you comment.

      Thinking of it as a poetic way of using some lovely scientific terms I rather like it, especially the way you link the long-ago refuted concept of ‘Neptunianism’ with Trans-Neptunian objects. However linking things because they are both named after a Greek God is not a valid scientific approach.

      I also like the way you only feel the need to describe examples of North American Geology. Ignoring the rest of the world neatly mirrors the way you ignore the vast amounts of geological evidence that is entirely inconsistent with what you propose.

      • Dave Carlson says:

        I hadn’t notice the language mirroring until you pointed it out, and therefore your semantic invalidation is merely observant. This time, how about a thoughtful response?

        For the sake of brevity I left out other hypothesized, Phanerozoic extinction-event impacts:
        – End Ordovician: Appalachian Basin (and possibly much of Western Europe, excluding Scandinavia)
        – End Silurian: Old Red Sandstone, Eastern Greenland, Scotland and most of Norway
        – End Permian: Siberia
        – End Triassic: China and South East Asia, minus North China
        – Apian Extinction (145.5 Ma): Mongolia and North China
        – End Cretaceous: Far East Russia and the North American Cordillera
        – End Eocene: Mountainous terrain from Greece to Tibet, including Turkey, Iran, Northern Pakistan and Nepal (including the young gneiss domes of Greece, Tajikistan and Nepal)
        – Middle Miocene disruption (14.5 Ma): Southern Japan with Mariana Trench tracing the impact-crater outline

        The ferocity of the P-Tr Extinction can be explained by a dwarf comet spiraling in to the inner solar system in a retrograde orbit, hitting Earth head on in its orbit around the Sun with 18.9 times the kinetic energy of dwarf planets catching up with Earth in its orbit. Then even the secondary debris trail impacts across Venezuela, North America and particularly West Africa were able to shatter the continental plates, creating faults that bled intrusive diabase in the ‘Triassic terrane’.

        Aqueous differentiation of TNOs explains concentric layering of gneiss domes as authigenic growth rings with the requirement of circumferential migmatite folding compared to the hand-waving or differential equation gloss-over treatment of conventional geology.

        Compare the competing formation mechanisms for mantled gneiss domes, and may the simplest model win out:

        1) Conventional:
        The mantled domes apparently represent earlier granite intrusions related to a orogenic period. The plutonic mass was later eroded and levelled, and thereafter followed a period of sedimentation. During a subsequent orogenic cycle the pluton was mobilized anew and new granite magma was injected into the plutonic rock at the same time as it was deformed into gneiss, causing its migmatization and granitization, or palingenesis.
        (Eskola, 1948)

        2) Alternative:
        Authigenic growth rings in an aqueously-differentiated TNO core followed by diagenesis, lithification and metamorphism of the sedimentary core

        • Metageologist says:

          My thoughtful response is that the role that extraterrestrial impacts have on earth is a fascinating theme that Earth scientists have made much progress with in the last 20 years.

          We know what a major impact looks like in detail: take the Vredefort Dome as an example. There are many distinctive structures found there that are not found in other parts of the world.

          We know what comet impacts do as well, material from North Africa has been identified as cometary material on the basis of isotopic analyses.

          So, we can spot where impacts happen. We know that the inbound material is largely vapourised, or at least melted and that it has a distinctive composition. These and other reasons are incompatible with what you propose.

          Remarkable claims require remarkable evidence. Until you provide some, and engage with the vast amounts of existing and relevant geological knowledge, your promotion of your ideas is unlikely to get you far.

          These are my last words on the matter.

          • Dave Carlson says:

            Granted I don’t have the wherewithal to prove my suppositions, but again, that’s another non-scientific argument, since concepts generally proceed proof, that is, unless one is exceedingly serendipitous.

            Agreed, all we have from the definitively-identified North Africa comet debris are some glassy tektites, no impact crater, and yet no one doubts that Earth has been pummeled by both rocky-iron asteroids and icy bodies for well over 4 billion years, so it’s a little premature to rule out alternative hypotheses.

            In the limit, our atmosphere is only equivalent to a 10 meter thick sheet of water 5.5 km above sea level, so any kilometer-scale comet or larger would punch through at interplanetary speed and not merely explode in the atmosphere as some suppose. Expecting 10 meters of water to stop a kilometer-scale comet, yet alone a TNO or dwarf planet, would be like expecting a sheet of paper to stop a bullet, except more so.

            Here’s an answer for the lack of identified icy-body impacts:
            Endothermic chemical reaction (ECR) products occurring in carbon ices (ethane, methane ices etc.) of icy-body impacts may largely clamp the impact shockwave pressure below the melting point of rock, obscuring icy-body impacts from detection.

            Experimentally, methane converts to long alkane chains and free hydrogen at 60 GPa and 4509 K (Li, Zhang et al. 2011). Thus the Michigan basin (and possibly both the Michigan and Illinois basins) may be the footprint of a Carboniferous impact that formed the long-chain hydrocarbons ECRs of the Pennsylvanian Subperiod coalfields. Then a debris flow may have bulldozed the forests, forming the primary coalfield cyclothem, complete with terrestrial vegetation and seatearth sedimentation, followed by deep burial and metamorphism into various grades of coal, complete with terrestrial vegetation fossils. So coal and oil may be signposts of icy-body impacts. And if the Michigan basin is an icy-body crater, then icy bodies don’t excavate craters like rocky impacts, but rather compress the ground since the Devonian strata is intact under the overlying Mississippian and Pennsylvanian strata.

            By comparison, the Vredefort crater may indeed be a rocky-iron impact, or at least an aqueously-differentiated icy-body impact.

  5. Tom Argles says:

    As a Himalayan field geologist, I chuckled wryly at your ‘Rhododendrons that stabilise slopes in the Himalayas’. It’s a scary thought that in the past barren slopes were even less stable than the Himalaya. Admittedly, there were fewer roads cut into them…

Leave a comment:

Your email address will not be published. Required fields are marked *

    Links (2)
  1. Pingback: Stuff we linked to on Twitter last week | Highly Allochthonous

  2. Pingback: The deceptive simplicity of a metamorphic rock | Metageologist