Partial sub-pixel and pixel-based alteration mapping of porphyry system using ASTER data: regional case study in western Yazd, Iran

ABSTRACT

This study uses remote sensing image-processing to enhance recognition of areas with mineral abundance halo, which can be associated with porphyry mineralisation of the Urumieh-Dokhtar Magmatic Belt (UDMB) in western Yazd. Granitoid intrusions of Oligocene-Miocene age in UDMB hosts the most important Porphyry Copper Deposits (PCDs) of Iran such as Darezereshk and Aliabad PCDs. Diagnostic spectral absorption features of six shortwave infrared (SWIR) and three visible and near-infrared (VNIR) bands of the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data were applied to determine the alteration minerals and their abundance. In this research reflectance data from various representative alteration types throughout the study area were measured by the FieldSpec-3 spectroradiometer and applied for mapping minerals. Different remote sensing analytical techniques, including the Principal Component Analysis (PCA), band ratio, relative absorption band depth, and MTMF, were applied to enhance mapping the alteration minerals. Field works including rock sampling, petrography and field spectroradiometer followed up the alteration mapping. Field observation implies that, the vein-type mineralisation above the cupola of this porphyry system is preserved in the western part, although it was eroded in the eastern area. Consequently, PCDs potential is expected to be higher in the western area.

KEYWORDS:

• Remote sensing
• ASTER image
• Porphyry Copper Deposit (PCD)
• Yazd
• Iran

Acknowledgments

Row images data were provided by Geomatics office of Geological Survey of Iran (GSI), and field studies and thin sections preparation were also carried out with the help of GSI. The GSI management’s staff are gratefully acknowledged, especially Dr Aida Mohebi who reviewed the manuscript and Dr Ali Bayatani (both experts of remote sensing in GSI) who worked with us by use of software. Many thanks are given to Prof. Lorence Collins from California State University Northridge for editorial assistance.