Lithological discrimination of accretionary complex (Sivas, northern Turkey) using novel hybrid color composites and field data


Ozkan M., ÇELİK Ö. F. , ÖZYAVAŞ A.

JOURNAL OF AFRICAN EARTH SCIENCES, cilt.138, ss.75-85, 2018 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 138
  • Basım Tarihi: 2018
  • Doi Numarası: 10.1016/j.jafrearsci.2017.11.009
  • Dergi Adı: JOURNAL OF AFRICAN EARTH SCIENCES
  • Sayfa Sayıları: ss.75-85

Özet

One of the most appropriate approaches to better understand and interpret geologic evolution of an accretionary complex is to make a detailed geologic map. The fact that ophiolite sequences consist of various rock types may require a unique image processing method to map each ophiolite body. The accretionary complex in the study area is composed mainly of ophiolitic and metamorphic rocks along with epi-ophiolitic sedimentary rocks. This paper attempts to map the Late Cretaceous accretionary complex in detail in northern Sivas (within Izmir-Ankara-Erzincan Suture Zone in Turkey) by the analysis of all of the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) bands and field study. The new two hybrid color composite images yield satisfactory results in delineating peridotite, gabbro, basalt, and epi-ophiolitic sedimentary rocks of the accretionary complex in the study area. While the first hybrid color composite image consists of one principle component (PC) and two band ratios (PC1, 3/4, 4/6 in the RGB), the PC5, the original ASTER band 4 and the 3/4 band ratio images were assigned to the RGB colors to generate the second hybrid color composite image. In addition to that, the spectral indices derived from the ASTER thermal infrared (TIR) bands discriminate clearly ultramafic, siliceous, and carbonate rocks from adjacent lithologies at a regional scale. Peridotites with varying degrees of serpentinization illustrated as a single color were best identified in the spectral indices map. Furthermore, the boundaries of ophiolitic rocks based on fieldwork were outlined in detail in some parts of the study area by superimposing the resultant maps of ASTER maps on Google Earth images of finer spatial resolution. Eventually, the encouraging geologic map generated by the image analysis of ASTER data strongly correlates with lithological boundaries from a field survey. (C) 2017 Elsevier Ltd. All rights reserved.