Employing both full and partial sub-pixel mapping methods to delineate hydrothermal alteration zones associated with porphyry copper deposits

The southeastern portion of the Urumieh–Dokhtar magmatic arc (UDMA), known as Kerman Cenozoic magmatic arc (KCMA), is a major host to world-class giant to subeconomic small porphyry copper deposits (PCDs) in Iran. As the KCMA is characterized by well-exposed rocks and sparsely vegetated surfaces, it is an intriguing region for geological remote sensing studies. In particular, mixed pixels are a key source of annoyance in traditional image classification because of a sensor’s immediate field of view restriction and the variety of land cover classes. By evaluating the observed spectrum of mixed pixels, sub-pixel mapping techniques can decompose each mixed pixel and determine the proportion of each component class, and so a classification map with a finer resolution is attainable. This paper endeavors to assess the capability and accuracy of linear spectral unmixing (LSU), multiple endmember spectral mixture analysis (MESMA), and mixture tuned target constrained interference minimized filter analysis (MTTCIMF) to investigate how well these sub-pixel algorithms could identify and map key hydrothermal alteration zones linked with PCDs in the Pariz–Chahargonbad area. Previous works have applied these algorithms widely to hyperspectral data, but few previous works have applied them to multispectral data such as ASTER. In this work, these algorithms were found helpful in the accurate identification of argillic, phyllic, and propylitic alteration zones per validation with field observations, petrographic studies and X-ray diffraction analysis of rock samples.