Aristotle University Of Thessaloniki (AUTH)
Υπολογισμός και χαρτογράφηση δεικτών στόχων βιώσιμης ανάπτυξης, με ανοιχτά δεδομένα παρατήρησης γης και υπηρεσίες υπολογιστικού νέφους
Abstract
dc:descriptionIn 2015, the United Nations established the "Sustainable Development Goals" (SDGs), which consist of specific Indicators that serve as quantitative measures for monitoring their progress. However, most studies for calculating them rely solely on statistical data without utilizing the spatial dimension, making it difficult to identify regional disparities and analyze the spatiotemporal impacts of SDG practices. Therefore, the integration of geospatial information is crucial for monitoring their progress and geospatial disaggregation is necessary, for their effective monitoring by relevant authorities .Remote sensing offers an economical, efficient, and systematic way to provide timely information about Earth's surface with spatial continuity and high spatial accuracy. Moreover, advanced processing methods, including deep learning algorithms and cloud computing, enhances the monitoring and decision-making for the SDGs. The objectives of the thesis were: a) developing a workflow for mapping open urban land use, utilizing satellite data and open geospatial products, and b) developing a workflow for national-scale land cover mapping, using satellite data and "weak" labels from open geospatial data, instead of ground data. Both workflows were designed to eliminate the need for ground reference samples and make use of deep learning algorithms as well as cloud computing technologies. The third objective of the thesis involved implementing the two developed workflows by calculating and mapping two SDG Indicators at national level for Greece. Specifically, the research focused on Indicator 11.7.1 ("Average share of the built-up area of cities that is open space for public use for all, by sex, age and persons with disabilities") and 15.4.2 ("Mountain Green Cover Index"). Open satellite data and geospatial products were mainly used, along with commercial very high-resolution satellite images. More specifically, Indicator 11.7.1 was estimated for the Athens metropolitan area 21.74%, for year 2020, while its spatial disaggregation highlighted the uneven distribution of urban open spaces among municipalities and between the city centre and the periphery. The mapping of the road extent sub-variable provided accurate results, however, the urban areas sub-variable mapping presented errors of commission. Nevertheless, the result can serve as a "potential" level for urban green spaces. As for Indicator 15.4.2, the overall accuracy of the produced land cover product was 68.65% for the six-category classification scheme and 88.21% for binary classification into "green" and "non-green" areas, while the final Indicator value for the entire country was estimated at 93.54%, for the year 2018. The spatial disaggregation of the Indicator revealed significant differences between mainland Greece and the islands, while by land cover type, the Indicator’s inability to depict the health conditions of mountain ecosystems, is highlighted. In this thesis, these indicators are calculated for the first time for the country and a new and practical approach for their monitoring and evaluation is introduced.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Verde, Natalia
- Βερδέ, Ναταλία
Subjects
dc:subject × 14- Δορυφορική τηλεπισκόπηση
- Βαθιά μάθηση
- Χαρτογράφηση κάλυψης γης και χρήσης γης
- Ανοιχτά γεωχωρικά δεδομένα
- Satellite remote sensing
- Deep learning
- Land cover and land use mapping
- Open-source geospatial data
- Περιβαλλοντική Μηχανική
- Επιστήμες Μηχανικού και Τεχνολογία
- Τηλεπισκόπηση
- Environmental Engineering
- Engineering and Technology
- Remote Sensing
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/56272