Monitorowanie zasięgu roślinności o charakterze leśnym w obszarach rekultywowanych z zastosowaniem zobrazowań satelitarnych Sentinel−2
Marta Szostak , Kacper Knapik , Justyna Likus-Cieślik , Piotr Wężyk , Marcin Pietrzykowski
AbstractPresented research investigates the possibility of applying the newest, free available satellite images Sentinel−2 for the automation of land use/cover (LULC) mapping in reclaimed areas, mainly in the aspect of monitoring forested areas. The study was performed for the former sulphur mines: ‘Machów’ (871.7 ha of the dump area after the opencast strip mine) and ‘Jeziórko’ (216.5 ha of the afforested area after the borehole exploitation). These areas are characterized by a diverse terrain structure and vegetation cover as the result of reclamation. The applied directions of reclamation were agro−forestry for the Sulphur Mine ‘Machów’ and forestry for the Sulphur Mine ‘Jeziórko’. We verified whether processing of Sentinel−2 data allows for reliable LULC classification – mainly identification forested areas in relation to the LULC mapping prepared by manual vectorization of orthophotomaps. Obtained classification results for Sentinel−2 data were also compared to the results of Landsat 8 images processing. The results of Sentinel−2 images classification showed correct graphical representation of the LULC classes, especially forested areas, in the relation to the results of applied on−screen vectorization of aerial orthophotomaps – better than results of the Landsat 8 images processing. The area of the mail class ‘Forests’ as a result of classification Sentinel−2 and Landsat 8 images compared to the results of manual on− −screen vectorization of the orthophomaps shows differences: 5.4% – Sentinel−2, 12.8% – Landsat 8 for Sulphur Mine ‘Machów’ and 1.8% – Sentinel−2, 8.8% – Landsat 8 for Sulphur Mine ‘Jeziórko’. Research indicates the possibility of automation of LULC classification using Sentinel−2 images. It could be very useful for LULC changes monitoring in reclaimed areas, mainly in the aspect of forested areas mapping as a result of way of reclamation.
|Other language title versions||Monitoring the spatial range of forested areas in the reclaimed sites using Sentinel−2 images|
|Journal series||Sylwan, [SYLWAN], ISSN 0039-7660, (N/A 40 pkt)|
|Publication size in sheets||0.5|
|Keywords in English||image processing, manual vectorization, spatial analyses, reclamation|
|Publication indicators||: 2016 = 0.739; : 2018 = 0.691 (2) - 2018=0.63 (5)|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.