These days it is a rare thing for an entirely novel surface process to be discovered; two centuries of geomorphological and sedimentological studies seem to have exhausted all the basic possibilities with only a few bits and pieces to be filled in.
Go to the foot of any steep slope topped by hard rock in an arid or semi-arid area and you are sure to find a boulder field formed by a variety of mass-wasting processes, such as rockfalls. As often as not such boulders are rounded, the usual explanation being that the rounding has resulted either from chemical weathering in the up-slope colluvium or exfoliation (‘onion-skin’ formation) through physical weathering in situ. Boulders are simply too big to have been moved other than by toppling or glacial transport at high latitudes, so rounding by abrasion seems unlikely. Aeolian sandblasting tends to favour just one side of boulders and ‘scallops’ their surface.
The driest place on Earth, Chile’s Atacama Desert, has plenty of boulder fields next to areas of high relief, and sure enough they are beautifully rounded, even though it has barely rained there for around 10 million years. Jay Quade of the University of Arizona, USA, with US, Australian and Israeli colleagues noticed that many of the boulders are surrounded by moat-like depressions and their sides, but not their tops, are nicely smoothed. These features suggested that some process had caused the boulders to move around and to rub one another, but whatever that was it had not caused even quite tall boulders to topple over (Quade, J. et al. 2012. Seismicity and the strange rubbing boulders of the Atacama Desert, northern Chile. Geology, 40, 851-854). An explanation was clearly something to puzzle over, until, that is, two of the authors returned to the area to make further observations. They were caught on the exposure by a magnitude 5.2 earthquake – a not uncommon experience in the foothills of the Andes – when the ton-sized boulders began to sway, rotate and jostle together with a great deal of noise. Here was the novel mechanism of erosion and ‘granulation’: seismic rubbing.
By dating the age of the exposed surfaces using cosmic-ray generated isotopes of beryllium and aluminium, the authors have been able to estimate that over the past 1.3 Ma the boulders have experienced between 40 to 70 thousand hours of rubbing. Indeed, it is quite likely that the whole boulder field, the upslope mass wasting and the sediment in which the boulders are embedded are products of seismicity. Oddly, just such jostling and rubbing of boulders and cobbles is characteristic of Inca architecture in the Andes, whose stonework used no cement but has minimal gaps between the blocks. Who is to deny that the Incas learned their unique building method from observing seismic rubbing.