Booming or Singing Dunes - Dune Worlds: How Windblown Sand Shapes Planetary Landscapes - page 109

10

Booming or Singing Dunes

Dunes are not silent. When the sand is blowing, you can
hear a hiss. However, there is a much more dramatic and
evocative acoustic emission from dunes—so-called boom-
ing or singing. This was noted in early Chinese records
(there is a dune named Mingsha San, or Singing Sand
Mountain), although it was first documented in the Western
literature by Marco Polo in the Taklamakan desert in the
13th century. Charles Darwin noted the phenomenon in
Chile at around the same time that two American scientists,
Bolton and Julien, surveyed 'musical sands'. Physicists,
including Bagnold, have since struggled to explain the
phenomenon, which has recently become a rather active,
and bitter, arena (!) of scientific debate (e.g., Chalmers
2006).
First, it is important to define the phenomenon. Sheared
sand makes noise: walking on certain sands from places as
far afield as the Scottish island of Eigg to Barking Sands
beach on the Hawaiian island of Maui produces a squeaking
noise. Such short-lived emission can also be generated in
the laboratory with a pestle, and it usually has a frequency
of a few hundred Hertz. However, the 'booming' of dunes is
a quite distinct process. This has a much lower frequency—
typically 80-120 Hz—and lasts for ten seconds or longer, in
association with (and perhaps occasionally longer in dura-
tion than) sand avalanches on slip faces. These avalanches
can occur naturally as the slip face oversteepens with sand
saltating over the brink, or (more usually when scientists
cannot wait for that) by walking on or kicking down on the
slipface. The sound is often compared to the drone of a
propeller plane.
Some early attempts to explain the phenomenon were
made by Poynting and Thompson (1909) and Bagnold
(1966), who suggested the frequency should vary as the
inverse square root of the sand diameter. The first good
measurements (Bolton and Julien compared the sand note to
a violin to estimate frequency) were made by Lindsay and
three Criswells in 1976. However, only in the last decade
has substantial progress been made in characterizing the

process, its environment and mechanism through better field
measurements, theory and laboratory experimentation.
The Paris group of Stephane Douady and Bruno
Andreotti and student Pascal Hersen observed booming
from avalanching slip faces of barchans in Morocco in
2001. This led them to explore the mechanism and to realize
that not only must sliding sand produce sound somehow,
but this sound must be coherent in some way. As discussed
in a fascinating overview by Chalmers (2006), Douady and
Andreotti diverged, eventually bitterly so, on their inter-
pretations and became unable to work together on the topic,
such that Andreotti moved to a different lab. They have
since pursued somewhat different aspects of the problem,
e.g., Andreotti (2004) on the synchronization of the sound
emission, using novel field measurements of both the
acoustic emission with a microphone and the movement of
the surface of the avalanche with an accelerometer, while
Douady et al. (2006) explored the sound generation from
sheared sand in the laboratory.
Several prominent dune structures in the Mojave are
known to boom: Kelso, Dumont and Sand Mountain. These
are all relatively accessible from Los Angeles, and in 2007
another set of researchers entered the fray—PhD student
Nathalie Vriend and her advisor, Melany Hunt. They doc-
umented booming on a number of occasions at these dif-
ferent sites, but also measured seismic emissions with an
array of dozens of geophones and used those to measure the
structure of the dune, finding distinct layers. They argued
that these layers were key to the process, and dismissed the
grain size factor, showing a plot of emission frequency
against grain size that essentially shows no correlation.
Regardless of the interpretation, the record of the emission
is the best data of its kind so far, and highlights that the
emission is not monochromatic, but a superposition of
harmonics. This may explain why early (pre-electronic)
estimates

of the frequency appear to be rather higher

(200 Hz+) than modern measurements (80-140 Hz)

(Fig. 10.1 ).