Altitudinal forest-tundra ecotone categorization using texture-based classification
Remote sensing of environment.
MetadataShow full item record
Guo, W., & Rees, G. (2019). Altitudinal forest-tundra ecotone categorization using texture-based classification. Remote sensing of environment., 232 Not-Available. https://doi.org/10.1016/j.rse.2019.111312
This study proposes a new technique involving texture-based image classification to categorize altitudinal FTEs by the degree of fragmentation of the interface. This allows a) universally adaptable altitudinal FTE categorization based on widely available satellite data products and b) assessment of sensitivity of altitudinal FTEs to shift with climate change at different locations based on the spatial distribution of the corresponding categories. The FTE categorization scheme used in this study corresponds partly to the globally occurring primary altitudinal FTE ‘forms.’ Specifically, ‘diffuse’ and ‘abrupt’ FTEs are recognized and separated. Normalized Difference Vegetation Index (NDVI) calculated from Sentinel-2 imagery is used for FTE delineation and categorization. A technique named FOurier-based Textural Ordination (FOTO) is implemented to extract textural information based on NDVI variations in image windows, and supervised classification is used to further separate these windows into FTE categories based on texture. The analysis is conducted on part of the Khibiny Mountains, Kola Peninsula, Russia, and further tested on six other study areas spread across the circumarctic region. The proposed method is able to adapt to different study areas with minimum changes in parameterization, and effectively extract altitudinal FTEs and categorize them into different FTE forms with satisfactory accuracies.
Cambridge Trusts, Trinity College Cambridge, Fitzwilliam College Cambridge, China Scholarship Council
Embargo Lift Date
External DOI: https://doi.org/10.1016/j.rse.2019.111312
This record's URL: https://www.repository.cam.ac.uk/handle/1810/294673
Attribution-NonCommercial-NoDerivatives 4.0 International
Licence URL: https://creativecommons.org/licenses/by-nc-nd/4.0/