Landscape Dynamics Mapping
The tool focuses on detecting and monitoring land use and land cover (LULC) changes, as well as vegetation dynamics over time, through the integration of satellite imagery, remote sensing techniques, and machine learning methodologies. The tool’s primary aim is to provide high-resolution spatial datasets that support biodiversity conservation and sustainable land management by visualizing landscape composition and transformations in targeted areas.
It employs object-based image analysis and supervised classification to generate LULC maps and vegetation indices such as the Normalized Difference Vegetation Index (NDVI). These highlight critical patterns, such as urbanization, habitat loss, and vegetation health, providing actionable insights for policymakers, environmental planners, and conservationists.
Incorporating data from Sentinel-2 and Landsat satellites, the tool enables multi-temporal analysis at a resolution of up to 10 meters, allowing for precise evaluations of landscape dynamics.
Basic Technologies Used
The tool integrates geospatial analysis software such as QGIS and ArcGIS, which provide a foundation for managing, analysing, and visualizing geospatial data and allow for the seamless integration of satellite imagery and spatial data to produce detailed maps and insights.
For remote sensing, the tool relies on multispectral satellite imagery from Sentinel-2 and Landsat. Sentinel-2 data, with a 10-meter resolution, enables detailed analysis of vegetation and land cover changes, while Landsat imagery provides a broader temporal perspective, extending back decades, to identify long-term trends. These datasets undergo pre-processing, including geometric correction, atmospheric correction, and cloud masking, to ensure the accuracy and usability of the images.
Machine learning algorithms are used to classify LULC types. Among algorithms, Support Vector Machines (SVM) have proven most effective in distinguishing land cover classes. Other algorithms, including K-Nearest Neighbours (KNN), Random Trees, and Maximum Likelihood, are also explored to enhance classification accuracy. To further refine the analysis, Object-Based Image Analysis (OBIA) groups pixels into objects based on spectral and spatial information, improving the ability to identify small and complex features like roads and clustered habitats.
Various spectral indices enhance the tool’s analytical capabilities. The Normalized Difference Vegetation Index (NDVI) measures vegetation health, while the Soil Adjusted Vegetation Index (SAVI) is particularly effective in low-vegetation areas. Other indices, such as the Modified Normalized Difference Water Index (MNDWI) and Normalized Difference Built-Up Index (NDBI), are used to detect water bodies and urban areas, respectively.
Data hosting and distribution are facilitated by Geoserver with Web Map Services (WMS), which support the visualization and sharing of geospatial data through interactive maps.
Usefulness in GUARDEN Case Studies
In the Greek pilot case study, specifically in the regions of Evia and Thrace, the tool has been employed to assess the impact of human activities, such as wind farm development and wildfires, on local ecosystems. For Evia, the mapping tool provided detailed insights into how wind farm installations and associated infrastructure have contributed to habitat fragmentation and reductions in natural vegetation. NDVI analyses revealed significant losses in vegetation health, particularly in areas surrounding new infrastructure, while also identifying zones that remain ecologically vital. Similarly, in Thrace, the tool has documented the devastating effects of wildfires, including extensive tree cover loss and the emergence of burned areas.
In Cyprus, the tool has been used to monitor the effects of industrial expansion in the Vasilikos area. By analysing changes over time, such as the increase in built-up areas and the corresponding decline in natural habitats, the tool has provided evidence of the pressures faced by ecosystems due to urbanization and industrial development. Vegetation indices, like NDVI, have shown declining trends in vegetation health, underlining the need for sustainable land use practices.
Contact Information
- Dimitris Tassopoulos | Draxis Environmental S.A.
- dtassopoulos@draxis.gr