Date of Award
3-23-2017
Document Type
Thesis
Degree Name
Master of Science
Department
Department of Engineering Physics
First Advisor
John W. McClory, PhD.
Abstract
Previous research conducted at Lawrence Livermore National Laboratory (LLNL) and the Air Force Institute of Technology (AFIT) has shown a correlation between actinide location and elemental composition in fallout from historic weapons testing. Fifty spherical fallout samples were collected from near ground zero of a surface burst weapons test. The samples were mounted in an aluminum puck then ground and polished to a hemisphere exposing the central plane. Physical morphologies of the samples ranged from clear to opaque with inclusions, voids, and/or uniform characteristics. Spectroscopy data were collected using optical microscopes and scanning electron microscopy (SEM), with radioactivity recorded through autoradiography. Principal component analysis (PCA) was used to quantify the variations within the samples and to determine the correlations between major elemental compositions and the incorporation of unspent nuclear fuel. Principal component analysis identified four statistically significant principal components accounting for 78% of the variations within the spectroscopy data. Principal component analysis was demonstrated as a suitable mathematical approach to solving the complex system of elemental variables while establishing correlations to actinide incorporation within the fallout samples. A model was developed using spot sampling to categorize the samples, identifying three classes of samples. The model correctly identified samples with above average uniform activity, thereby identifying samples with high forensic value for recovery of unspent nuclear fuel. Final analysis of the full elemental composition and the correlation with regions of increased activity for all fifty samples is currently being completed.
AFIT Designator
AFIT-ENP-MS-17-M-096
DTIC Accession Number
AD1051622
Recommended Citation
Haws, Derek W., "Using Principal Component Analysis to Improve Fallout Characterization" (2017). Theses and Dissertations. 787.
https://scholar.afit.edu/etd/787