Size-sorted sand creates uncommon length scale in windblown formations

When wind sweeps across beaches and deserts, it carries and redeposits sand grains, creating molehill-sized ripples and mountainous dunes. Occasionally, the interaction of wind and sand produces intermediate-scale formations called megaripples. ¹ An explanation of how the rare formations develop has long eluded scientists. Finding one could provide insight into ancient wind and soil conditions on Earth and other planets where megaripples appear in images.

Ripples and dunes arise from different wind-driven, or aeolian, processes. In the case of ripples, sand grains lifted centimeters high by the wind fall back and strike the ground, causing stationary grains to bounce up and hop small horizontal distances, usually about 10 cm. Those grains pile up in a corrugated pattern whose crest-to-crest distance is established by the hop length.

Dunes are much larger than the hop length. They form when wind accelerates over a topographic obstacle, picks up more sand as it goes, and deposits the sand just at the crest while the pile of sand grows. The sand does not travel continuously in the wind, but in discrete hops. For would-be dunes less than 10 m across, the hopping grains overshoot the crest, so the structure cannot grow. Instead, it will be eroded, which explains why intermediate-sized sand waves between ripples and dunes are rarely observed. ²

Now Klaus Kroy, Marc Lämmel, and colleagues at the University of Leipzig in Germany and Ben‐Gurion University of the Negev in Israel have explained why megaripples are so rare. ¹ The researchers propose that the unusual intermediate-sized sand waves, separated by more than 10 cm, form in a similar manner to giant sand dunes but in a narrow range of wind speeds.

Mountains or molehills?

Sand formations are difficult to study in situ, due to their large spatial and temporal scales (see Physics Today, March 2014, page 19 ). Kroy’s team combed the literature for measurements of morphology and grain size from the southern Negev desert in Israel, where ripples, dunes, and megaripples coexist (see figure 1), to evaluate their mathematical models. ³

Figure 1.

Bulk sand contains a mixture of grain sizes. As moderate winds carry off small grains, the coarse grains left behind eventually form a layer that armors the sand bed. When the wind is just below the strength needed to shift the heavy grains, the fine grains will sometimes strike the stationary heavy grains and give them enough energy to roll forward—but only in tiny steps the size of the grains’ diameter. The heavy grains respond to changes in wind speed on much shorter length scales than the longer-hopping light grains. As a result, heavy grains sometimes build dunes that are much smaller than dunes dominated by fine grains. Those tiny dunes are what geomorphologists have dubbed megaripples.

Sand beds evolve over centuries. Well-defined surface layers, like the coarse grains of a megaripple, become part of the bulk grain distribution. Local differences in the bulk composition may indicate wind strength or direction. Change in the bulk may indicate wind strength or direction. Figure 2 shows the coarse grains of the megaripple on top of the mixed bulk layer. A steady, moderate wind rarely blows for the hours to weeks necessary to form a megaripple. Gusts and storms commonly disrupt the process and will quickly erode the megaripple by causing the grains to make jumps that overshoot the pile’s crest.

Figure 2.

Until the wind changes

In 2009, Edgar Manukyan (ETH Zürich) and Leonid Prigozhin (Ben-Gurion University) proposed that wind sorts inhomogeneous sand into layers of small and large grains and that a critical wind strength sets large grains into motion so that megaripples form. ⁴ Kroy’s team has now extended that proposal to create quantitative descriptions, predict how and when the structures form, and explain their fleeting nature. A crucial component of their model explores how different wind speeds influence sand distribution.

Key to forming a coarse layer is separating coarse grains from the initial bulk. Figure 3 shows a modeled mixture of sands, with coarse grains 0.4 to 4 mm in diameter. A constant wind entrains some of the lighter particles and sweeps them away. The wind leaves the bigger grains to accumulate. After 200 hours, the coarse grains have separated from the bulk and form a megaripple. That time scale is extremely sensitive to the bulk grain distribution and the wind strength. A moderate wind could be considered 60 km/h.

Figure 3.

A megaripple created by a constant wind is quickly demolished when that wind exceeds the threshold value required to mobilize large grains. The time series of figure 3a is extended in figure 3b, but with wind strength increased by a factor of 2.5. In just three hours, the strong wind sweeps away the coarse grains and returns the sand to the normal distribution.

Of course, actual wind speeds are variable. The variable wind profile shown in figure 3c takes almost 100 days to erode the initial megaripple. By 200 days, another megaripple has begun to form. The average wind speed is assumed to be 25% of the moderate wind in figure 3a, and 10% of the wind speed during a storm. That grain distribution is similar to that observed in the Negev region (figure 3d).

Geologists have commonly treated megaripples as “ripples” because of their morphology. Like ripples, megaripples lack the steep leeward slope characteristic of dunes. But like dunes and unlike ripples, megaripples are irregularly spaced rather than periodic. “If megaripples are dunes, they exist independently of their neighbors,” says Kroy. The study offers a paradigm shift for thinking about megaripples as small dunes rather than as large ripples.

More field testing is needed to establish the effects of changing wind speeds on grain-size distribution. Because megaripples have been classified as ripples, most field studies that attempt to probe their dynamics have recorded the distances between nearby crests and have rarely collected grain-size data in a way suitable for interpretation by the new theory. “Perhaps the most useful part of our work is suggesting what kinds of measurements are needed to reveal meaningful correlations. In the past, people collected sand [data] in many different, sometimes inconsistent ways,” says Kroy.

Climate history near and far

In his 1941 book, The Physics of Blown Sand and Desert Dunes, Ralph Bagnold calculated how long it had taken the prevailing winds on North Africa’s Mediterranean coast to push dunes from the seashore into the interior, where they turned the region into an uninhabitable desert. His answer was 7000 years. His work opened the possibility of evaluating past climate conditions by determining how long ago sand dunes formed.

Earth’s largest megaripples are found on Argentina’s Puna Plateau. Kroy’s study could similarly help geologists infer past ambient wind conditions that prevailed in the decades to centuries over which those megaripples formed. Analyzing the grain composition in current sand beds could also reveal past conditions. “In the future, petrified megaripples could tell us about climate conditions long ago,” says Kroy.

Megaripples—or at least aeolian bed forms at the intermediate scale rarely found on Earth—are fairly common on Mars. In an analysis of observations from the Mars Science Laboratory’s Curiosity rover and the Mars Reconnaissance Orbiter, Mathieu Lapotre (Harvard University) speculated that megaripples might be the dominant aeolian bed form on Mars. ⁵ “[Kroy’s] idea may prove useful in teasing out the relationship between various bed forms on Mars,” says Lapotre.

Recently, Matthew Telfer (University of Plymouth), Eric Parteli (University of Cologne), and colleagues reported on several hundred ridges of wavelengths 0.4 to 1 km in images from the New Horizons spacecraft’s July 2015 flyby of Pluto. ⁶ They proposed that the features are small dunes and are best explained by wind deposition of sand-sized particles of methane ice.

Kroy and other scientists are working on understanding the implications for the present-day wind regime on Mars, Pluto, or other planets. The work could lead to refined models of how sand dunes form and propagate.