MULTISPECTRAL AND HYPERSPECTRAL IMAGERY FOR SAFEGUARDS AND VERIFICATION OF REMOTE URANIUM MINES

Year
2003
Author(s)
Russell Leslie - Australian Safeguards and Non-Proliferation Office
G. A. Borstad - Borstad Associates Ltd.
R.A. Neville - Canada Centre for Remote Sensing, Ottawa, Canada
Q.S. Bob Truong - Canadian Nuclear Safety Commission
K. Staenz - Canada Centre for Remote Sensing
Peter Riggs - Australian Safeguards and Non-Proliferation Office
V. Bragin - IAEA
Abstract
In this era of strengthened safeguards measures, and especially following the recent introduction of high-resolution satellite imagery by several commercial firms, the use of imagery is becoming more common as a cost-effective means to obtain information on uranium mining activities. Such imagery is relatively straightforward to interpret, and can be used to produce valuable information for safeguards purposes, e.g. image maps for comparison with member state declarations. In addition to these open sources, there are several unrestricted “research” satellite sensors available that can provide inspectors and regulators with additional tools that might be used to confirm operations, scheduling, movement of materials and other reported details. As part of a co-operative project between the Canadian and Australian Safeguards Support Programs, very high spatial resolution panchromatic and natural colour (Red, Green, and Blue, or RGB bands) imagery of three operating uranium mines in Australia was examined [1]. This paper presents examples of the use of multispectral bands from IKONOS for multispectral classification and 220 bands from HYPERION for hyperspectral classification to demonstrate: • terrain and vegetation classification; • water usage (and by inference certain mining or milling operations); and • geographic movements of ore material on a site by spectral discrimination. The paper examines potential applications of the above findings with respect to safeguards verification of declared and undeclared activities.