Mapping Land Cover Using Remote Sensing Data and GIS Techniques: A Case Study of Prahova Subcarpathians

Abstract

This paper describes a methodology to obtain a land cover dataset for a selected Subcarpathian area from Prahova County, Romania. It proposes a combined Remote Sensing, GIS and Cartography approaches in order to produce an accurate land cover layer at regional and local scales for an area where the land cover features are various and complicated in configuration. The area between Prahova and Teleajen rivers is characterized by complex landforms as hills and depressions with active and dynamic geomorphic processes. This paper presents a model integrating a supervised classification derived from Landsat OLI images with currently available land cover models, as well as GIS processing results of data fusion and cartographic techniques for feature representation on the map, at different scales. Another research direction was to use some normalized differenced indices to separate the spectral classes, and to assess and validate the results from the classification. Land cover map was obtained using: 1) land cover classes derived from supervised classification of Landsat 8 OLI image data, and; b) land cover features extracted from two land cover models (CORINE Land Cover and GlobeLand30), integrated with the classification dataset. The results were confronted and validated with ground truth points generated from orthophotos and other satellite images. The layouts are generated at different scales in order to compare and validate the land cover categories.