07-17, 16:30–17:00 (Europe/Sarajevo), PA01
Urban Change Detection in Tirana, Albania (2000-2025) Using Remote Sensing and Open Geospatial Data
Tirana, the capital of Albania, has experienced rapid urbanization over the past two decades, driven by political, social, and economic transformations. This fast and often chaotic growth has led to significant changes in land use, urban expansion, and environmental conditions. Uncontrolled urbanization presents major challenges for sustainable city development, requiring a comprehensive understanding of urban change dynamics. This study analyzes the spatial and temporal dynamics of urban change in Tirana from 2000 to 2025, using open geospatial data and satellite image processing techniques to map and examine transformations in land cover and urban development.
To assess land cover changes, the study utilizes high-resolution satellite imagery from Landsat and Sentinel-2, which provides valuable information on the spatial extent of urbanization and land use transformations over the past two decades. Additionally, open data from Urban Atlas is integrated to enhance classification accuracy and enable a more detailed analysis of land cover changes. Open data from sources such as the State Authority for Geospatial Information in Albania (ASIG) is also used to validate the results, ensuring that findings reflect actual changes on the ground and providing a comparison with existing datasets.
The methodology of this study is based on change detection techniques, which are essential for understanding urban growth. Key indices such as the Normalized Difference Vegetation Index (NDVI) are used to monitor vegetation loss, while the Normalized Difference Built-up Index (NDBI) is applied to track the expansion of built-up areas. These indices help quantify land cover changes by distinguishing between urban areas, vegetation, and other land uses. NDVI is particularly useful for detecting vegetation loss, often associated with urban sprawl and land degradation. Similarly, NDBI serves as an effective indicator for monitoring the increase in built-up areas, a crucial aspect of urban expansion.
The study employs a supervised classification approach to categorize land cover into different classes using the Random Forest algorithm. This machine learning technique has proven effective in classifying land cover with high accuracy, especially in complex landscapes such as urban areas. The Random Forest algorithm combines multiple decision trees to classify pixels in satellite imagery, allowing for the differentiation of urban, vegetation, water, and other land use types. By ensuring a high level of classification accuracy, the study provides a reliable assessment of urban changes in Tirana over time.
Beyond analyzing past urban growth, the study also aims to predict future urban development trends. To achieve this, the MOLUSCE plugin in QGIS is used to model future urban growth patterns based on historical data and influencing factors such as population growth, infrastructure expansion, and policy interventions. The MOLUSCE tool enables the prediction of land use changes and urban expansion over time, helping to outline future development scenarios in Tirana. Population data from the Albanian Institute of Statistics (INSTAT) is integrated into the analysis to better understand the driving factors behind urban change. This data provides insights into population growth trends, a key driver of urbanization, and their interaction with other urban development factors.
The results of this study provide a comprehensive analysis of urban change in Tirana, offering valuable insights into the city's transformation from 2000 to 2025. The study highlights the extent of uncontrolled urbanization, vegetation loss, and the expansion of built-up areas, which are characteristic features of rapid urbanization. These findings are crucial for urban planning and policy development, as they offer a data-driven foundation for understanding the drivers of urban growth and the challenges associated with managing it. The results can aid in designing sustainable urban development strategies, helping policymakers and urban planners better anticipate and manage future growth, mitigate negative environmental impacts, and improve the quality of life for city residents.
Also, methodology and findings of this study have broader applications beyond Tirana. The approach used can be applied to other cities experiencing similar urbanization patterns, providing a valuable tool for urban planners, researchers, and policymakers globally.
By leveraging open geospatial data and advanced satellite image processing techniques, this study not only contributes to the FOSS4G community's efforts to understand urban change but also enhances the overall sustainability and informed management of urbanization. The methodology applied here is a prime example of how open geospatial data can improve urban research, as it facilitates the accessibility, transparency, and reproducibility of urban analyses. Through the use of freely available resources, the study significantly improves land cover change detection, reinforcing the growing body of research emphasizing the importance of open data in addressing global urbanization challenges.
Furthermore, the availability of open data empowers local communities to actively participate in the urban planning process, fostering public awareness and engagement. By making data accessible, the study strengthens social cohesion and builds trust between citizens and authorities, facilitating better-informed decision-making that contributes to the sustainable development of cities.
1- Hafner, S. et al. (2024) Continuous urban change detection from satellite image time series with temporal feature refinement and multi-task integration.
2- Wang, W.S. et al. (2020) Land use and land cover change detection and prediction in the Kathmandu District of Nepal using remote sensing and GIS. Sustainability, 12(9), p.3925.
3- Padma, S.P. et al. (2022) Simulation of land use/land cover dynamics using Google Earth data and QGIS: A case study on Outer Ring Road, Southern India. Sustainability, 14(24), p.16373.
4- Ramadan, G.F. & Hidayati, I.N. (2022) Prediction and simulation of land use and land cover changes using open source QGIS: A case study of Purwokerto, Central Java, Indonesia.
5- Congedo, L. (2021) Semi-Automatic Classification Plugin: A Python tool for the download and processing of remote sensing images in QGIS. Journal of Open Source Software, 6(63), p.3172.
Select at least one general theme that best defines your proposal – I make my conference contribution available under the CC BY 4.0 license. The conference contribution comprises the abstract, the text contribution for the conference proceedings, the presentation materials as well as the video recording and live transmission of the presentation – yesI am a geospatial data enthusiast with a strong background in Geoinformatics Engineering
and Computer Science. My expertise lies in combining advanced spatial data analysis with
innovative technologies to solve real-world problems. I have practical experience working
on various GIS-based projects, specializing in spatial data management, analysis, and
visualization.
As a researcher, I focus on the application of GIS systems in urban planning and other
geospatial disciplines, continuously striving to contribute to the field through both
academic and professional endeavors