A Comparison of Spectral Angle Mapper and Artificial Neural Network Classifiers Combined with Landsat TM Imagery Analysis for Obtaining Burnt Area Mapping

George P Petropoulos,Krishna Prasad Vadrevu,Marko Scholze,George Karantounias,Gavriil Xanthopoulos

doi:10.3390/s100301967

Abstract

Satellite remote sensing, with its unique synoptic coverage capabilities, can provide accurate and immediately valuable information on fire analysis and post-fire assessment, including estimation of burnt areas. In this study the potential for burnt area mapping of the combined use of Artificial Neural Network (ANN) and Spectral Angle Mapper (SAM) classifiers with Landsat TM satellite imagery was evaluated in a Mediterranean setting. As a case study one of the most catastrophic forest fires, which occurred near the capital of Greece during the summer of 2007, was used. The accuracy of the two algorithms in delineating the burnt area from the Landsat TM imagery, acquired shortly after the fire suppression, was determined by the classification accuracy results of the produced thematic maps. In addition, the derived burnt area estimates from the two classifiers were compared with independent estimates available for the study region, obtained from the analysis of higher spatial resolution satellite data. In terms of the overall classification accuracy, ANN outperformed (overall accuracy 90.29%, Kappa coefficient 0.878) the SAM classifier (overall accuracy 83.82%, Kappa coefficient 0.795). Total burnt area estimates from the two classifiers were found also to be in close agreement with the other available estimates for the study region, with a mean absolute percentage difference of ∼1% for ANN and ∼6.5% for SAM. The study demonstrates the potential of the examined here algorithms in detecting burnt areas in a typical Mediterranean setting.

Highlights

Over the past few decades, wildland fire research has been receiving increasing attention in several regions of the world, including Mediterranean regions, because of the wide range of ecological, economic, social, and political impacts of such fires
Observations from the Landsat TM sensor combined with either the Spectral Angle Mapper (SAM) or the Artificial Neural Network (ANN) classifier have been used in the past and at different geographical regions demonstrating the ability of these classifiers in land use/land cover mapping applications
The objective of the present study was to explore, for the first time, the potential use of these classifiers combined with the Landsat TM imagery analysis for deriving total burnt areas in a test site representative of a typical Mediterranean setting

Summary

Introduction

Over the past few decades, wildland fire research has been receiving increasing attention in several regions of the world, including Mediterranean regions, because of the wide range of ecological, economic, social, and political impacts of such fires. Being able to obtain accurate as well as rapid mapping of burnt areas is of key importance to both environmental scientists and policy makers. Such information is very important, for example, for estimating the economic consequences from the fire and establishing rehabilitation and restoration policies in the affected areas, assisting to avoid post-fire hazards and long-term degradation [2]. Burnt area delineation on an operational basis can provide important information on land cover changes related to ecology and biodiversity, that can in turn significantly assist in understanding post-fire recovery of an affected area [4]

Objectives

Results

Conclusion