Integration of earth remote sensing data and geographic information systems in the formation of digital land use of the models for territorial communities
The article highlights the urgent problem of information support for land resource management of territorial communities in the context of reforming the land relations system, decentralization of power, and the post-war recovery of Ukraine. The article analyzes the state of the existing topographic and geodetic base and the shortcomings of the State Land Cadastre data, which often contain spatial errors and do not reflect the actual state of land use. Open multispectral satellite data from the Sentinel-2 mission (Copernicus program), aerial survey materials, and QGIS software are determined as the foundation for digital land use models of territorial communities. A concept for forming a digital model of a community's land bank is proposed, which is considered a dynamic multidimensional system rather than a static map. The technological process of creating a digital land use model is described, which includes five key stages (preparatory stage or collection and audit of input data; primary processing stage or formation of a geodatabase; analytical and synthesizing stage (spatial modeling); applied stage (implementation and monitoring)). The integration of geospatial data is considered as a technological process of bringing heterogeneous spatial information (raster remote sensing data, vector cadastral maps, aerial survey results) to a unified coordinate system and topological consistency within a geographic information system (GIS) environment for the purpose of comprehensive modeling and analysis of the territory. A structural and logical scheme for the integration of multi-sensor remote sensing data and cadastral information in a GIS environment is proposed. It is noted that topology rules are applied to eliminate errors in the geometry of objects: checking for overlaps where boundaries of one land plot cross into another, gaps between common boundaries of adjacent plots, and vertex snapping of vector objects to the nodes of the base layer within a specified tolerance. The identification of spatial changes and the structuring of land use were carried out according to the methodology of automated interpretation of Earth Remote Sensing (ERS) data based on the calculation of the Normalized Difference Vegetation Index (NDVI), which allowed for a clear delineation of vegetation cover from urbanized areas and water bodies based on the spectral brightness of pixels. A digital model of the Rivne territorial community is presented as a multilevel spatial database that contains various objects and ensures their processing, analysis, and visualization for solving tasks of spatial planning, infrastructure management, and environmental protection.