Earth and Planetary Sciences ETDs

Publication Date

Fall 2021


Transverse aeolian ridges (TARs) are enigmatic and largely relict bedforms on the surface of Mars. TARs are sparsely distributed but common on Mars, but their history, preservation, and past role in the sediment cycle is not well understood. First described in 2003, and detailed extensively in 2008, our study of TARs has been narrowly focused in the last decade, with more and more research noting their presence, but little investigation of the features themselves. Recent work has mostly focused on identifying Terran analogues for TARs, but TARs remain largely a unique Martian feature. In this manuscript, I clarify and refine our understanding of TAR morphologies, develop and present a new evolutionary pathway, and prototype methods for future research.

In Chapter 1 (published in Planetary and Space Science), I outline how cratering in superposition between different aeolian ridge morphologies indicated a compound development for feathered TARs. The goal of this project was to clarify the relationship between simple TARs and feathered TARs, and to refute the ‘dust-deltoid’ hypothesis put forward in the literature. Through detailed geomorphic surveys and measurements of TARs in Nirgal Vallis, I conclude that feathered TARs are compound features, formed when secondary ridges develop on the flanks of simple or forked TARs. I extrapolated these findings to suggest that feathered TARs are an intermediate form between simple/forked and networked TARs, and that the dust-deltoid hypothesis is improbable.

In Chapter 2 (published in Remote Sensing: Mars Remote Sensing), I extended on the findings from Chapter 1 and performed a broader survey across Mars for cratering in between different TAR generations. This survey produced a further sixteen examples across nine HiRISE images. These findings demonstrate that the multigenerational formation of compound TAR morphologies was not exclusive to Nirgal Vallis. Further, given that the conditions required to preserve these features in sequence were probably rare, this process likely occurred more widely. Armed with widespread observations of secondary and even tertiary ridge formation, the comparative implications for the three Terran TAR analogues (ripples, megaripples, and dunes) are discussed.

In Chapter 3 (submitted to Communications Earth & Environment) I present a unified theory for TAR evolution through the various TAR morphologies. In this study, I refuted the hypothesis that all TARs developed from the barchan form, and instead suggest that all TAR morphologies can be derived from the simple morphology. I delineated the difference between rectangular and hexagonal networked TARs, and outlined the evolutionary separation between these morphologies. I documented primary ridge reactivation by secondary ridges, and illustrated how this process can create both sinuous TARs and hexagonal networked TARs. Lastly, I combined my observations with results from previous empirical experiments to postulate how TARs could be driven to evolve from simpler 2-dimensional forms under 3-dimensional flow conditions into more complex 3-dimensional morphologies under 2-dimensional flow conditions.

In Chapter 4 (published in Remote Sensing: Deep Learning for Remote Sensing Data), I prototype an off-the-shelf RetinaNet-based neural network model for automatically classifying the Martian surface, with special attention paid to the identification of TARs. Automatically classifying the surface of Mars remains an ongoing NASA priority, and our method proved effective in identifying not only TARs but also ripples and polygonal terrain. These classes are composed of groups of features and are thus ‘fuzzy’—lacking well-defined boundaries. This model is the first application of a discrete object identification algorithm to fuzzy objects in any field. The method was optimized for reducing false-positive identifications, and had an overall precision of 92.9% across all classes using a confidence interval of 60%. As tested, the model was a proof-of-concept and lacks the infrastructure needed to systematically process the massive quantity of Martian imagery currently generated by the HiRISE camera. I hope that this work will inspire other researchers to explore applying similar algorithms and models to other problems on Earth and Mars.

In Appendix I (published in Earth Surface Processes and Landforms), I present observations of interactions between aeolian ridges and boulders in and around Proctor Crater. These observations illustrate the ability of preserved aeolian features to retain information about the wind conditions during their formation. This study produced two notable results: first, that aeolian ridges on Mars can act as surface roughness elements themselves, and second, that through an unusual process wake-like fans of crests can form downwind of obstacles on Mars. At present, there is no known analog on Earth or in numerical simulations for these wake features.

In Appendix II (published in Journal of Applied Remote Sensing), I led a team of undergraduate and graduate students to develop a simple method for extracting and analyzing boulders from HIRISE imagery. Our technique was efficient, flexible, and more accessible than other methods. We applied our new methodology to Jezero Crater, which at the time was the chosen landing site for the Mars 2020 Perseverance Rover. The methodology successfully extracted hundreds of thousands of boulders in the study area, with the highest densities found in the northern and western portions of the Jezero delta. We concluded that a small fraction of boulders migrate by downslope rockfall, but the majority of boulders remain adjacent to the scarp. The boulder distributions around the Jezero crater were thus likely the product of rock fall, scarp retreat, and creep, with scarp retreat and creep being the major processes and rock fall being less important.

Degree Name

Earth and Planetary Sciences

Level of Degree


Department Name

Department of Earth and Planetary Sciences

First Committee Member (Chair)

Louis Scuderi

Second Committee Member

Christopher Lippitt

Third Committee Member

Corinne Myers

Fourth Committee Member

Horton Newsom

Fifth Committee Member

Charles Shearer

Project Sponsors

NASA Doctoral Fellowship 80NSSC19K1676.




Martian, Aeolian, Surface processes, Transverse aeolian ridges, ripples, dunes

Document Type