Date of Award

May 2022

Degree Type


Degree Name

Doctor of Philosophy



First Advisor

Lindsay J McHenry

Committee Members

Timothy J Grundl, Freek D van der Meer, Julie A Bowles, Mark T Harris


Deep Learning, Lake Tecopa, Mars, Paleolakes, Spectral Remote Sensing, Zeolite


This study evaluates the possible formation and evolution mechanisms of zeolites on early Mars with possible explanations for their limited detections using Earth analogs. This study focuses on the formation of zeolites in the closed basin lakes where the largest relatively pure concentrations of natural zeolites are found on Earth. Five working hypotheses were formulated to explore the limited detection of zeolites in closed basin lakes on Mars and different styles of scientific reasoning with suitable examples were used to test the independent, converging lines of inquiry. Zeolites may not be identifiable in certain locations on Mars using orbital data if, 1) they are absent, or 2) they were originally present and later removed by chemical processes (e.g. dissolution and alteration), or 3) they are present but are covered by or mixed with other materials, or 4) they are present, but the methods applied are not capable of detecting and mapping them, or 5) they are present, but we are not looking in the correct places. The first possibility was tested using geochemical modeling, while the second possibility and part of the third possibility were tested both using geochemical modeling and analog sites on Earth. A “textbook” example, paleolake Tecopa, was selected as the primary analog site. The fourth possibility was tested using hyperspectral (Hyperion) and multispectral (ASTER) orbital remote sensing data (visible to shortwave infrared and thermal infrared wavelength range) at Lake Tecopa with different spectral mapping techniques and ground truth data. Fieldwork was conducted during October 2018 and 2019 and 56 soil and rock samples were collected representing different surface materials over the area. X-ray Diffraction (XRD), X-ray Fluorescence (XRF), and Scanning Electron Microscopy-Energy Dispersive X-ray Spectrometry (SEM-EDS) were used to identify the bulk mineral composition and elemental composition of these samples. Spectral deconvolution with deep learning were applied to identify and estimate mineral abundances in zeolite-bearing mineral mixtures. A data-driven fuzzy-based weights-of-evidence method was adopted to identify favorable areas to look for zeolites on Mars, as a solution for the fifth possibility. The geochemical modeling shows that zeolites can form at low temperatures under potential early Martian conditions, both from basaltic and high silica starting materials, and some zeolites (e.g. clinoptilolite) are dissolved over time when other zeolites (e.g. analcime) precipitate as commonly observed on Earth. Field studies along with literature surveys show that most zeolite beds in paleolakes on Earth are thin, covered by other beds, or mixed with other materials due to physical weathering. Orbital and laboratory spectral studies show the difficulty of identifying non-analcime zeolites from mineral mixtures using spectral methods alone. The spectral and spectral resolution of the orbital images, atmospheric components, and dust cover also limit the detection of zeolites in orbital images. The predictive model created during this study shows favorableareas for the formation and/or presence of zeolites and this map could serve as a guide for further searches for zeolites using detailed orbital spectral image analysis and future in situ observations. The overall study shows that analcime is more likely to be found than other zeolite minerals because of it is chemically stable and easy identify using spectral data. The results imply that the paucity of detected zeolites on Mars does not preclude their wider presence, either beneath other materials, obscured by surface dust, or mixed with more spectrally dominant phases. The best way to confirm their presence is to “follow the zeolite on Mars” using future in situ observations.

Available for download on Wednesday, November 23, 2022