Seasonal differences assist in mapping granite outcrops using Landsat TM imagery across the Southwest Australian Floristic Region

G. Alibegovic, A.G.T. Schut, G.W. Wardell-Johnson, T.P. Robinson

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)

Abstract

Knowledge of the location and extent of granite outcrops (GOs) in the Southwest Australian Floristic Region is important to understand their role as refugia. We present a methodology to map GOs using biannual Landsat TM imagery. An adaptive vegetation cover mask capitalising on seasonal differences, combined with a supervised classification, allowed differentiation of granite from other land covers on five GOs across the rainfall gradient. This methodology provided high classification accuracy (Overall Kappa ranged from 0.83 to 0.91) irrespective of location. Therefore, there is potential to use these methods to compile a more complete GO inventory over the region.
Original languageEnglish
Pages (from-to)37-49
JournalJournal of Spatial Science
Volume60
Issue number1
DOIs
Publication statusPublished - 2015

Keywords

  • machine learning algorithms
  • spaceborne thermal emission
  • reflection radiometer aster
  • climate-change
  • western-australia
  • band ratios
  • biodiversity
  • vegetation
  • refugia
  • cover

Fingerprint

Dive into the research topics of 'Seasonal differences assist in mapping granite outcrops using Landsat TM imagery across the Southwest Australian Floristic Region'. Together they form a unique fingerprint.

Cite this