The rover “Curiosity” poses for a selfie at a Martian sand dune. Credit: NASA/JPL-Caltech/MSSS

For the first time, a landed spacecraft has investigated an otherworldly dune field. A new special issue of Journal of Geophysical Research: Planets describes the results of a several-month science investigation at the Bagnold Dunes in Gale crater, using the suite of instruments carried by the Mars Science Laboratory Curiosity rover. Though the primary focus of the Curiosity mission is to investigate the ancient sedimentary rocks that likely formed in a lacustrine environment, the Bagnold dunes presented an opportunity to investigate active, modern sedimentary processes on Mars. I asked Bethany Ehlmann, who was one of the lead organizers of the science campaign, some questions about this Martian landscape and the major scientific findings of the mission.

Of all the different landscape features on Mars, why is a dune field of particular interest?

Dunes are beautiful landforms. They are also among the most dynamic landforms on Mars, responding to local weather. From orbit, we have seen dunes migrating, which is a little puzzling because the winds on Mars are weak and the atmosphere is thin, meaning it is actually hard to lift up a sand grain. Landing on the surface and visiting an active dune field lets us take a close look at the structure, form, and composition of the dunes, and investigate sand transport processes. Understanding contemporary dunes might also help to interpret ancient dunes on Mars, which are preserved in sedimentary rocks.

Which site was chosen for the investigation?

Mid-2017 map of NASA’s Curiosity Mars rover mission showing the traverse along side the Bagnold dunes. When passing through the dunes and conducting the Bagnold Dunes campaign, “Gobabeb” was the Phase 1 stop and “Ogunquit Beach” was the Phase 2 stop. Credit: NASA/JPL-Caltech/Univ. of Arizona

The location was the Bagnold Dune Field. This is a striking feature from orbit: a dark swath cutting northeast to southwest across the base of Mount Sharp in Gale crater. There are longitudinal dunes in the center of the field and crescent-shaped dunes on the margins.

We named it after Ralph Bagnold, a British army officer. Based on his explorations in North Africa, he published careful observations and physical models of dunes in The Physics of Windblown Sand and Desert Dunes [1941].

Our exploration on Mars followed 75 years after this work, and we felt it would be a fitting tribute and mark of progress in aeolian science to name the first active dune field explored on another planet after him.

What did you want to investigate at the Bagnold dune field?

We had a range of questions as Curiosity approached the Bagnold dunes. From orbit, the dunes seem olivine-rich, which is interesting because most of the other sediments filling Gale crater are not. Olivine-enriched dunes are pretty typical of Mars but, in the case of Gale crater, where did the olivine-enriched material come from? It also looks from orbital infrared spectrometer data that the margins of the dune field are of different composition or grain size from the center portion. Why would this be? Is it because minerals of different densities are separated by sand transport? These and other questions helped us to decide what measurements and samples we wanted to collect.

What data did the Curiosity rover gather while traversing the Bagnold dune field?

Curiosity exercised all of her capabilities in the dune field. We took wind measurements and images to monitor the dunes for movement. We used chemistry and mineralogy instruments on the rover mast and arm to understand what the sands were made of, including bulk sand chemistry, bulk mineralogy, and the isotopic composition and release temperature of volatiles. The rover also collected and processed samples, scuffing, scooping, and then sieving the sands into different grain size fractions to measure their compositions separately.

Did you find that the characteristics of this Martian dune field were similar to dune fields on Earth?

Mostly, the size of the grains (50 to 350 micrometers), steepness (about 30 degrees), and shape of the Bagnold dunes, as well as the centimeter-scale ripples on the flanks are similar to Earth [Ewing et al., 2017]. But an exciting discovery was of a sinuous type of ripple at approximately 1 meter spacing that we just don’t see on Earth [Lapotre et al., 2016; Ewing et al., 2017]. These ripples are at a spacing in-between the dunes themselves (hundreds of meters-scale) and the impact ripples from saltation (decimeter-scale), possibly formed by drag forces in the thin Mars atmosphere [Lapotre et al., 2016].

Two sizes of ripples on the surface of a Martian sand dune. Credit: NASA/JPL-Caltech/MSSS

The hypothesis is that in Earth’s thick atmosphere, the fluid-drag ripples are about the same size as the impact ripples, but in the thin Martian atmosphere, they have a distinct, larger spacing. Thus, looking at ripple form and spacing in sandstones may provide a new way to get a measure of how atmospheric pressure varied at different points in Mars history.

What were some of the other most interesting findings?

Mastcam telephoto of a Martian dune’s downwind face showing slumps and grainfalls from recent sand motion. Credit: NASA/JPL-Caltech

Because of the timing of the rover’s traverse, we visited the dunes during some of the weakest winds in the year [Newman et al., 2017]. Nevertheless, we did still observe slumps and grain movement [Bridges et al., 2017].

A new idea for why there is activity at lower-than-predicted wind speeds is that Mars’ low gravity allows just a handful of grains picked up by turbulent wind gusts to have a splashing effect, triggering many more grains to move [Sullivan and Kok, 2017].

Mastcam false color image of sands at Gobabeb, Namib dune, which shows subtle visible/near-infrared color differences in scuffed sands disrupted by the wheel and the discard piles for samples A (<150 μm), B (>150 μm), and C/D (150 μm-1mm). Credit: NASA/JPL-Caltech/MSSS, Ehlmann et al., 2017, Figure 6a

In terms of sand composition, we found that coarser grains are rich in mafic minerals like olivine and pyroxene [Cousin et al., 2017; O’Connell-Cooper et al., 2017].

The crystalline mineralogy of the well-sorted dune sands was the same as much finer-grained, dusty Mars soils and is characteristic of typical basaltic materials.

But the dunes also were comprised of about 40% materials that were not typical crystalline minerals, i.e., they appeared amorphous to the X-ray diffraction instrument [Achilles et al., 2017].

Comparison of the chemistry data with the mineralogy shows that volatile elements (carbon, hydrogen, nitrogen, sulphur, chlorine) must reside in this amorphous fraction.

But the dunes were found to be depleted in hydrogen, sulfur, and chlorine relative to all other materials in Gale crater.

This leads to the inference that dust and the finest soils on Mars, which are absent in the Bagnold dunes, hold the majority of water, chlorine, and sulfur found in Mars surface materials  [Ehlmann et al., 2017].

After Curiosity’s exploration, what are some of the unanswered questions that remain about this landscape?

There are several remaining questions [for a summary see Bridges and Ehlmann, 2017]. The source of the volatile elements in the sands remains a key question. Over the coming months, we will get isotopic data of hydrogen, carbon and nitrogen in the sands to understand if they are consistent with having been sequestered during a time when the atmosphere was similar to today or in a more ancient epoch or if they are from igneous processes.

We’re also interested in disentangling the source of the sands and whether they are coming from the olivine-enriched walls of Gale crater [Buz et al., 2017], an area farther up Mount Sharp, or somewhere more distant. Sands transported regionally or globally have been hypothesized for Mars, though how this might work in practice is not perfectly understood. A sand grain blown by the wind can only bounce along so many times before erosion wears it down to smaller particles that are raised into the air as dust rather than hop along near the surface to form sand dunes. At the time of this writing, the rover is analyzing samples collected during a second stop at the longitudinal dunes, so we’ll get some new compositional data.

Looking ahead to further exploration of the dunes, we’re eager to analyze the ripple structures on longitudinal dunes and assess whether they agree with the physical models that were proposed based on the data at the crescent-shaped dune we visited. Spoiler alert: there was a lot of sand motion observed at the longitudinal dunes, so stay tuned for some publications on how dunes change during higher winds on Mars.

—A. Deanne Rogers, Associate Editor for JGR: Planets and Department of Geosciences, Stony Brook University; and Bethany Ehlmann, Division of Geological & Planetary Sciences and the Jet Propulsion Laboratory, California Institute of Technology; email:


Rogers, A. D.,Ehlmann, B. (2017), A rover’s eye view of moving Martian dunes, Eos, 98, Published on 21 November 2017.

Text © 2017. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.