A couple of weeks ago, in the midst of the series of posts on Gulf of Mexico berms, I encountered one of those odd coincidences that are odd only because we forget the statistics behind coincidences.
The travel section of the newspaper I was reading had a piece extolling the delights of the endless expanses of dunes and beaches on the German island of Sylt - the tile of this post immediately jumped into my deranged mind. And then, browsing through the geoblogosphere, I came upon an enjoyable post by Ole Nielsen on hooked spits - which highlighted the dramatic sedimentary architecture of Sylt. Ole referred to a recent paper in Sedimentology, and so, intrigued, I took myself off to the Geological Society Library and looked it up. The story is, indeed, intriguing.
Sylt is a form of barrier island and, given the role that these shape-shifting piles of sand play in the ongoing saga in the Gulf of Mexico, they link in to the inexhaustible theme of coastal dynamics. As Ole commented, " Barrier islands are sedimentary islands separated from the mainland by bays or lagoons, and some originate as sand spits. A hooked spit is a landform created by the water currents and may be described as a curved sand bar. In fact barrier island systems often terminate with hooked spit morphologies." Barrier islands come in many shapes and sizes, and the conditions for their formation and their evolution, regardless of whether they terminate in a hooked spit or not, continue to retain many mysteries.
Sylt faces the North Sea and is the northernmost German island - indeed, the mainland across from its northern half is Denmark and the place was Danish until 1864 ; it's 38 kilometres from one end to the other (when last measured) and is connected, artificially, to the mainland. At its northern tip is a beautiful hooked spit. Sylt is geologically young, its core built from debris left over from the melting of ice sheets at the end of the last ice age: these materials were reworked by the onslaught of the rising seas and the main island began to develop around 6000 years ago. Today, the strength and complexity of tides, waves, and currents produces a dynamic playground for sand, dramatically illustrated in low and high tide images, the Wadden Sea to the east festooned with ever-shifting sand bars:
This playground has, as usual, been a difficult place to live. Historically, villages on Sylt have been swept away in storms and inundated by migrating and growing sand dunes, and lighthouses have found themselves in locations no longer effective for their original purpose. The west coast, facing the energy of the North Sea, is subject to constant erosion, the sand being flushed northwards or swept out into offshore bars. The western shoreline of Sylt is now perhaps 10 kilometres east of where it used to be, and today erosion rates reach more than a meter a year; the southern tip of the island has on occasion been severed after a violent storm. But it's an important tourist destination, and so something must be done, and that something, as usual, is beach nourishment. According to Wikipedia, which cites the reference (in German), " Since 1972 an estimated 35.5 million cubic metres of sand have been flushed ashore and dumped on Sylt. The measures have so far cost more than €134 million in total, but according to scientific calculations they are sufficient to prevent greater loss of land for at least three decades, so the benefits for the island's economic power and for the economically underdeveloped region in general would outweigh the costs." Aha - those scientific calculations again..... But these measures require annual replenishment, and so understanding the sand dynamics and sediment budgets of Sylt is important, and this where the four authors, from institutes in Hamburg and on Sylt, of the paper in Sedimentology come in. While the paper itself is not directly available, the PhD thesis of the lead author, Sebastian Lindhorst, that formed the basis for this and other publications, is available to read online.
This work has demonstrated that the sedimentary systems of the main part of Sylt and the hooked spit that forms its northern tip are quite distinct - the spit is growing at the cost of erosion further south; but the way in which the spit grows is far from simple. The work was conducted using ground-penetrating radar, a relatively simple and inexpensive technique that nevertheless has powerful applications in engineering and archaeology (there's quite a good summary on Wikipedia and various commercial companies such as GeoModel). Essentially, the reflections of the radar energy from subsurface layers and objects are recorded when they bounce back to the surface; the reflections are caused by changes in the velocity of the radar waves as the nature of the material they are travelling through changes. Use some clever computer processing and an image of what these subsurface reflections look like can be generated as a profile, a vertical "slice"; the vertical scale on such an image is actually time, in that what you are measuring is the time taken for the radar pulse to travel downwards, be reflected, and return to your recording instrument on the surface ("two-way time", TWT). But, if you know fairly well the nature of the materials underground, then this can all be converted to a depth scale. Importantly, Lindhorst and his colleagues calibrated the radar data with cores and so the character and velocity of different sands, wet, dry, fine-grained, coarse-grained, could be integrated with the radar data.
Here's an example of one of the profiles from their work:
The profile at the top is the processed radar data; TWT in nanoseconds (1 ns is a billionth of a second) is the vertical scale on the left, depth in meters, with mean sea level as zero, is on the right. In order to display an image where the components are more clearly shown and interpretable, the whole thing is vertically exaggerated by a factor of five. If it weren't displayed this way, the true-scale version would look roughly like this - accurate, but difficult to interpret:
So, back to the main figure. The profile runs at right angles to the coast of the spit, with the surface topography at the top: the higher ground on the right shows the dunes at the back of the low beach ridges that run parallel to the shore. Now, with experience, local understanding, some cores (and a little imagination), the darker layers that are the reflectors of the radar pulses can be interpreted as successive sedimentary features, and the relationships between them examined. The one significant "layer" that is not a sedimentary feature is the strong, essentially horizontal, reflector that runs across the profile just above sea level: this is the top of the groundwater table, the boundary between dry sands above and wet sands below, a velocity contrast so dramatic that it bounces back a lot of the radar energy. The lower part of the figure is the geological interpretation of the radar data, with the main components of the architecture of the spit identified (my apologies for this only being in black and white, but it's still quite clear). The main elements are all inclined (dipping) towards the right, the shoreline, and represent episodes of growth of the spit, building through time, younger swathes of sand on top of older ones, from right to left (elements G to D).
But this growth is not continuous, it takes place in episodes, so that surfaces such as F represent the sea floor following a phase of sand build-up where the actions of waves and currents gently scour away at the sand surface. The result is a discontinuity in the sedimentary pile, the next flush of sand being deposited onto that somewhat modified surface. The boundary between the two building events shows up as a radar reflection, with the geometries of the sand layers above and below being slightly different - these surfaces are what are marked on the profile as unconformities. These are vital as they divide the foundation of the spit into successive episodes of deposition and quiescence. But even more important are the features labelled "C." These mark something far more dramatic than the normal ebb and flow of sand deposition: the base of each of these packages of sand is marked by a pronounced discontinuity - the layers beneath are chopped off, scoured, removed, with new piles of sand then dumped on top of them. These "cut and fill structures" are, as the notes on the figure record, telling us of periodic major events, storm surges scouring into the edge of the spit. They are essentially scoop-shaped excavations of the pre-existing sands, filled in, and, like the succession of layers beneath them, recording sequentially younger events from right to left (shorewards).
From numerous radar profiles and the core data, Lindhorst and his colleagues construct a model of how the spit evolves, each major episode of growth, ironically, spurred by an episode of erosion. Each storm surge event leaves a scarp, a small cliff, at the back of the eroded section, and this then acts as a sand trap, leading to the development of "embryonic dunes" along its length. Sand transported along the coast builds up the beach and is periodically recycled out to the edge of the spit. Fast-growing vegetation (largely marram grass, Ammophila arenarea) stabilises the dunes which form "foredune ridges," common features of hooked spits, and a new element of the spit is now firmly established. This sequence of events repeats itself every year or so, and the spit grows.
This work also established that the hooked spit of Sylt is considerably younger than the main island: it began forming perhaps as recently as 1300 years ago. Its growth is supplied by sand eroding from the coast to the south, and that supply has increased by a factor of two or three over the last few decades - most probably as a result of the increased efforts at beach nourishment. So tourists cause landscape changes.....
I'm still not quite clear why this spit is hooked, but presumably it's a reflection of the currents swirling around the entrance to the Wadden Sea, and the dramatically complex playground of sand.
[Header image: creative commons license, from Temporalata;
satellite images from Eovision; reference for
the Sedimentology paper: Anatomy and sedimentary model of a
hooked spit (Sylt, southern North Sea), SEBASTIAN LINDHORST, JÖRN
FÜRSTENAU, H. CHRISTIAN HASS, CHRISTIAN BETZLER, Sedimentology
Volume 57 , Issue 4 , Pages935 - 955;]