Sand ripples constitute a common landform, spontaneously emerging virtually everywhere a fluid or gas flows over a granular bed. We gathered a consortium of field and wind tunnel workers and theoreticians from six countries to study and compare ripple formation under terrestrial and Martian atmospheric conditions. This lead to a new classification of ripples, not into one but two bedform types. Closer theoretical analysis hints at an unexpected stratification in the wind driven grain hopping (“saltation”) responsible for it, challenging also current aeolian sand transport theories.

Ripples belong to a hierarchy of aeolian bedforms (Figure) found on Earth and various extraterrestrial bodies, throughout the solar system [1]. Compared to their larger cousins, such as dunes and draas, their appearance has traditionally been attributed to a distinctly different mechanism. Namely, to a direct feedback, not mediated by the wind, between saltation and the topography of the sand bed, as manifest in the technical name “impact ripple”. But this traditional view has recently increasingly become the subject of intense debate among the geomorphology, sedimentology, planetary geology, and granular physics communities. In December 2015, NASA’s Curiosity rover found mysterious meter-scale ripples on Mars. Such large ripples are not observed on Earth, in monodisperse sand. In 2016, Lapotre et al. [2] hypothesised that they may have emerged from a hydrodynamic flow instability, much like morphologically similar subaqueous patterns on Earth and terrestrial and extraterrestrial sand dunes. This stirred up much controversy over the following years, progressively lending further support to the hydrodynamic-origin hypothesis. In 2019, Duran et al. [3] eventually proposed a mechanistic model to unify the formation of large Martian ripples with that of subaqueous ripples. They agreed with Lapotre and coworkers that the same model that successfully predicts the size of subaqueous ripples and large Martian ripples also predicts that they should have decimeter-scale aerodynamic cousins on Earth. Except that the latter would fall into a forbidden wavelength gap, delineated by the same theory. Briefly, the saturated transport conditions required for the formation of aerodynamic ripples (as opposed to impact ripples) should not actually be realised on Earth. The estimated saturation length (the distance it takes the sand flux to adapt to a change in wind speed) would exceed the predicted wavelength for terrestrial aerodynamic ripples, meaning that the latter should fall prey to a sort of “hydrodynamic suicide”.

To shed light on this paradox, we conducted a series of controlled wind tunnel experiments at ambient and Martian atmospheric conditions (at Ben Gurion University, Israel, and Aarhus University, Denmark, respectively), both demonstrating the stable coexistence of two distinct types of ripple, impact and aerodynamic (or, technically, “hydrodynamic”) ripple. Their somewhat close wavelengths but disparate growth rates may explain why they are difficult to discern on Earth and might easily have been confounded, in past studies. Our careful analysis of different wind speeds, grain sizes, and atmospheric pressures, as well as a comparison with computer simulations and theory, further corroborates the unexpected conclusion, at odds with conventional wisdom: on Earth, like on Mars, two distinct types of ripple can coexist, and the larger and slower one is hydrodynamic in nature.

Its existence also fundamentally challenges current saltation models, as it hints at a very tight coupling of the grain transport and wind speed, at a much shorter saturation length than previously thought. This implication resonates with other indirect evidence recently reported, which rests on a large compilation of data for so-called megaripples [4]. Curiously, these giant “ripples”, made of self-sorting polydisperse sands, often resemble down-scaled dunes rather than their smaller namesakes. Altogether, the theoretical analysis of our experiments to calibrate the new saturation scale lead us to the formulation of an updated state-of-the-art ripple model. It admits stable aeolian hydrodynamic ripples under terrestrial conditions and thus promises that it may become possible to reconcile all previous observations within a unified theory of aeolian structure formation. To validate this expectation quantitatively, over a wider range of ambient conditions and grain sizes, will however require further carefully controlled field and laboratory work and also further theoretical developments. In particular, constructing a bottom-up theory, based on the grain-scale transport physics, will not be an easy task. It may well compel us to question and transform the conventional understanding of the aeolian grain transport process as profoundly as that of ripples.

[1] Hayes, A. G., Dunes across the solar system. Science 360, 960–961 (2018)
[2] Lapotre, M. G. A. et al., Large wind ripples on Mars: A record of atmospheric evolution. Science 353, 55–58 (2016)
[3] Durán Vinent, O., Andreotti, B., Claudin, P. & Winter, C., A unified model of ripples and dunes in water and planetary environments. Nature Geoscience 12, 345–350 (2019)
[4] Lämmel, M. et al., Aeolian sand sorting and megaripple formation. Nature Physics 14, 759–765 (2018)