A Rule-Based Classification Method for Mapping Saltmarsh Land-Cover in South-Eastern Bangladesh from Landsat-8 OLI

Abstract

Wetland vegetation classification often treated the saltmarsh as a single type of land-cover (LCT). Mapping the dynamic and spatially complex coastal zones using optical remote sensing is still challenging. This study firstly analyzed the spectral properties of target objects generated by Landsat 8 (OLI), formulated new spectral indices and then proposes a rule-based approach to mapping five vegetated (saltmarsh, seagrass, mangrove, non-mangrove forest, and agricultural land) and three non-vegetated (wet sand, saltpan, and built-up areas) LCT in the study area, that is, large coasts located in the south-eastern coasts of Bangladesh. The thresholds of spectral indices were selected from the newly introduced spectral indices over the method development site (Bakkhali estuary). The rule-based LCT classification process followed a set of cascade rules of image thresholding and masking, based on a hierarchical tree in order to generate detailed thematic maps of saltmarsh land-cover. Overall accuracy (OA) and Kappa coefficient (K) of rule-based approach were 84.6% and 0.821, respectively. The reliability and robustness of the approach was tested over two independent external validation test sites: Karnaphuli river estuary and Teknaf peninsula and consistent accuracy results achieved: OA = 81.7% (K = 0.787) and OA = 84.6% (K = 0.821) respectively.

RÉSUMÉ

Les classifications de la végétation des terres humides traitent souvent le marais salé comme un seul type de couverture terrestre (LCT). Il est encore difficile de cartographier les zones côtières dynamiques et spatialement complexes à l’aide de la télédétection optique. Cette étude a d’abord analysé les propriétés spectrales des objets cibles générés par Landsat 8 (OLI) et formulé de nouveaux indices spectraux. Ensuite, elle propose une approche basée sur des règles pour cartographier cinq zones végétales (marais salés, herbes marines, mangroves, forêts et terres agricoles) et trois zones non végétalisées (sables humides, salines et zones construites), d’une grande région côtière du sud-est du Bangladesh. Les seuils des indices spectraux ont été choisis à partir des indices spectraux nouvellement introduits sur le site de développement de la méthode (l’estuaire de Bakkhali). Le processus de la classification LCT a appliqué un ensemble de règles de seuillage et de masquage en cascade, basées sur un arbre hiérarchique afin de générer des cartes thématiques détaillées des marais salés. L’exactitude globale (OA) et le coefficient Kappa (K) de l’approche de classification proposée est de 84.6% et de 0.821 respectivement. La fiabilité et la robustesse de l’approche ont été testées indépendamment sur deux autres sites de validation: l’estuaire de la rivière Karnaphuli et la péninsule de Teknaf. Des précisions similaires ont été obtenues pour les deux sites, soient OA = 81.7% (K = 0.787) et OA = 84.6% (K = 0.821) respectivement.

ACKNOWLEDGMENTS

Landsat 8 OLI data were downloaded from the USGS Earthexplorer portal (http://earthexplorer.usgs.gov). For having assisted in saltmarsh land-cover type field data collection the authors thank Arannayk Foundation—Wetland conservation campaigns.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was financially supported by the National Agricultural Technology Program—Phase II, Bangladesh Agricultural Research Council (BARC), Bangladesh; the Higher Institution Centre of Excellence (HICoE) Research Grant [53209], awarded to the Institute of Oceanography and Environment (INOS), Universiti Malaysia Terengganu (UMT); Arannayk Foundation—Wetland conservation campaigns [AF-Saltmarsh-15070101].