ИСТИНА

One of biogeographical scientific problem is detecting of plant communities and determining of its` boarders. The given problem is especially acute for small diverse patchy communities. They are typical for circumpolar natural zones like Arctic or Antarctic tundras and deserts. Consequently, the primary aim of this investigation is to develop optimal method for discrimination plant communities in ice-free areas of Antarctica. The common method is remote sensing, that is useful in severe conditions of the extreme polar latitudes. The authors analyzed the reflectance spectra of the most representative vegetation elements (different taxa of moss, lichens, alga and vascular plants) in the Barton Peninsula, King George Island, using a ground-based spectrophotometer. They used in turn method of Self-Organized Mapping and Linear Discriminant Analysis to classify the cover classes according to reflectance spectra, after reduction of bands using Principal Component Analysis. Eventually it was determined typical samples of hyperspectral patterns for representative taxa. And it was developed efficient method for discrimination of main vegetation elements in ice-free area of Antarctica. These procedures also can be applicated to the diversification of the similar plant communities with help of analyzing of airborne images in northern Russian tundras, that allows to improve quality of vegetation mapping in this region. Source: Maria Calvino-Cancela, Julio Martin-Herrero. Spectral Discrimination of Vegetation Classes in Ice-Free Areas of Antarctica. Remote Sensing. 2016, 8(10), 856.