Spectral Discrimination of Uranium-Mineralized Breccia Pipes in Northwestern Arizona

A. Y. Kwarteng, P. C. Goodell, N. E. Pingitore, Jr., K. J. Wenrich

The price of uranium is currently the lowest in more than a decade. The only type of uranium deposit that is economically viable in the depressed uranium market is such high-grade ore as the unconformity type found in Canada and Australia. Exploration for uranium-bearing breccia pipes in northwestern Arizona by both domestic and foreign companies is currently active because of the relatively high-grade ore they contain and their tendency to be polymetallic. In the United States, uranium-mineralized breccia pipes are one of the few deposits that can compete in the current market.

A stepwise discriminant analysis was performed on spectral data acquired from the field, laboratory, and Landsat thematic mapper (TM). The principal objectives were (1) to investigate the fundamental differences in the spectral properties of outcrops on the surface of breccia pipes and the background, (2) to choose TM bandpasses that were statistically optimum for distinguishing between breccia pipes and the background, and (3) to compare the results of the field, laboratory, and TM digital data which were acquired by different instruments having different spatial and spectral resolutions.

The discriminant analysis of the three data sets indicated TM bands 1, 3, and 7 were the best for distinguishing breccia pipes from their background outcrop. Between 62% and 79% of the samples used were correctly identified as members of the groups to which they actually belong. However, scatter diagrams constructed from the discriminant functions and the ternary plots of the three TM bands showed that the samples in the groups were only partially separable visually. In all three cases, the first discriminant function, DF1, mapped the contrast between the two main lithologic units, Kaibab and Moenkopi Formations. The higher discriminant functions, DF2 and DF3, were difficult to interpret and did not correspond to any obvious physical geologic parameter. This problem does not, however, preclude the development of an exploration technique based on discriminant analysis of field, laboratory, and remote-sensing data.

AAPG Search and Discovery Article #91022©1989 AAPG Annual Convention, April 23-26, 1989, San Antonio, Texas.