With each new mission to Mars, the amount of available data increases dramatically. This drastic increase in data volume requires new approaches to take advantage of the available information. The goal of the work presented here is to maximize the science return from existing and future datasets.
Chapter 2 uses multiple orbital datasets to characterize Gale Crater, with a focus on the northwestern crater floor and lower mound. This work played a role in the selection of Gale Crater as the landing site for Mars Science Laboratory (MSL). It was not possible to conclusively determine the origin of the lower mound, but we interpret features on the upper mound as aeolian cross-beds.
Chapters 3 and 4 investigate methods for improving the accuracy of laser-induced breakdown spectroscopy (LIBS). In Chapter 3, the accuracy of partial least squares (PLS) and two types of neural network are compared, using several pre-processing methods including automated feature selection. We find that partial least squares without averaging typically gives the best results. Chapter 3 also investigates the influence of grain size on the accuracy of analyses, showing that >20 analysis spots may be required for heterogeneous targets. In Chapter 4, we test the hypothesis that clustering the dataset before analysis leads to improved accuracy. We observe modest improvements for five k-means clusters and with iterative application of clustering and PLS.
In Chapter 5, we use several methods to relate Mars Exploration Rover (MER) Panoramic camera multispectral observations to alpha particle X-ray spectrometer and Mössbauer spectrometer results. The correlation between the Gusev datasets is often poor although there is some improvement when only data from drilled spots is considered. The performance is better for the Meridiani data, but Meridiani PLS models are not generalizable to Gusev data. MSL ChemCam analyses and MastCam spectra may show higher correlations because the instruments have a similar information depth.
Clustering and classification methods can be used on any dataset, and as the volume of data from planetary missions continues to increase, synthesis of multiple datasets using multivariate methods such as those in this work will become increasingly important.
|School Location:||United States -- New York|
|Source:||DAI-B 73/08(E), Dissertation Abstracts International|
|Subjects:||Asian Studies, Astronomy|
|Keywords:||Landing sites, Lasers, Mars, Multiple orbital datasets|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be