Περιλαμβάνονται, με χρονολογική σειρά, δημοσιεύσεις στις οποίες έχουν χρησιμοποιηθεί δεδομένα από το εθνικό δίκτυο παρακολούθησης λιμνών.
Δημοσιεύσεις σε επιστημονικά περιοδικά
Perivolioti, Triantafyllia-Maria; Tompoulidou, Maria; Giourieva, Viktoria; Mouratidis, Antonios; Apostolakis, Antonios
The impact of submerged aquatic vegetation on satellite-derived bathymetry in shallow waters: an accuracy assessment Δημοσίευση σε επιστημονικό περιοδικό
In: European Journal of Remote Sensing, vol. 59, iss. 1, pp. 2667786, 2026.
Περίληψη | Σύνδεσμοι | BibTeX | Ετικέτες: lake bathymetry, Lake Trichonida, machine learning, remote sensing, Sentinel-2 imagery, vegetation bias
@article{Perivolioti31122026,
title = {The impact of submerged aquatic vegetation on satellite-derived bathymetry in shallow waters: an accuracy assessment},
author = {Triantafyllia-Maria Perivolioti and Maria Tompoulidou and Viktoria Giourieva and Antonios Mouratidis and Antonios Apostolakis},
url = {https://www.tandfonline.com/doi/full/10.1080/22797254.2026.2667786},
doi = {10.1080/22797254.2026.2667786},
year = {2026},
date = {2026-05-07},
urldate = {2026-05-07},
journal = {European Journal of Remote Sensing},
volume = {59},
issue = {1},
pages = {2667786},
abstract = {The aim of this study was to evaluate the performance of four widely used Satellite-derived Bathymetry (SDB) models (Stumpf, Lyzenga, Random Forest and XGBoost) in different scenarios of varying proportions of bare substrate and submerged aquatic vegetation in lake shallow-waters. Particularly, six scenarios were created, ranging from 0% to 100% submerged vegetation within a depth range of 0–10 m. The predictive accuracy of SDB models was assessed using an updated reference digital terrain model (DTM). The performance of the models was evaluated across multiple depth ranges and bottom types (bare substrate and submerged vegetation). Evaluation metrics included root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²). One to one scatterplots demonstrated higher model performance in bare substrate, with R² values ranging from 0.56 to 0.63 for Stumpf/Lyzenga and 0.65–0.70 for XGBoost/Random Forest in bare substrate areas. Scenario-level CV of RMSE remained acceptable (<30%), decreasing from ~0.058 (0% vegetation) to ~0.022 (100% vegetation). Random Forest and XGBoost outperformed Lyzenga and Stumpf models under mixed scenarios. This study improves the understanding of SDB training requirements in mixed lake bottom types and offers valuable insights for improving the application of SDB in lakes.},
keywords = {lake bathymetry, Lake Trichonida, machine learning, remote sensing, Sentinel-2 imagery, vegetation bias},
pubstate = {published},
tppubtype = {article}
}
The aim of this study was to evaluate the performance of four widely used Satellite-derived Bathymetry (SDB) models (Stumpf, Lyzenga, Random Forest and XGBoost) in different scenarios of varying proportions of bare substrate and submerged aquatic vegetation in lake shallow-waters. Particularly, six scenarios were created, ranging from 0% to 100% submerged vegetation within a depth range of 0–10 m. The predictive accuracy of SDB models was assessed using an updated reference digital terrain model (DTM). The performance of the models was evaluated across multiple depth ranges and bottom types (bare substrate and submerged vegetation). Evaluation metrics included root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²). One to one scatterplots demonstrated higher model performance in bare substrate, with R² values ranging from 0.56 to 0.63 for Stumpf/Lyzenga and 0.65–0.70 for XGBoost/Random Forest in bare substrate areas. Scenario-level CV of RMSE remained acceptable (<30%), decreasing from ~0.058 (0% vegetation) to ~0.022 (100% vegetation). Random Forest and XGBoost outperformed Lyzenga and Stumpf models under mixed scenarios. This study improves the understanding of SDB training requirements in mixed lake bottom types and offers valuable insights for improving the application of SDB in lakes.
Tompoulidou, Maria; Karadimou, Elpida; Apostolakis, Antonis; Tsiaoussi, Vasiliki
A Geographic Object-Based Image Approach Based on the Sentinel-2 Multispectral Instrument for Lake Aquatic Vegetation Mapping: A Complementary Tool to In Situ Monitoring Δημοσίευση σε επιστημονικό περιοδικό
In: Remote Sensing, vol. 16, no. 5, 2024, ISSN: 2072-4292.
Περίληψη | Σύνδεσμοι | BibTeX | Ετικέτες: aquatic vegetation, General Earth and Planetary Sciences, GEOBIA, lake monitoring, Mediterranean lakes, remote sensing, Sentinel-2 imagery, WFD
@article{Tompoulidou2024,
title = {A Geographic Object-Based Image Approach Based on the Sentinel-2 Multispectral Instrument for Lake Aquatic Vegetation Mapping: A Complementary Tool to In Situ Monitoring},
author = {Maria Tompoulidou and Elpida Karadimou and Antonis Apostolakis and Vasiliki Tsiaoussi},
doi = {10.3390/rs16050916},
issn = {2072-4292},
year = {2024},
date = {2024-03-05},
urldate = {2024-03-05},
journal = {Remote Sensing},
volume = {16},
number = {5},
publisher = {MDPI AG},
abstract = {Aquatic vegetation is an essential component of lake ecosystems, used as a biological indicator for in situ monitoring within the Water Framework Directive. We developed a hierarchical object-based image classification model with multi-seasonal Sentinel-2 imagery and suitable spectral indices in order to map the aquatic vegetation in a Mediterranean oligotrophic/mesotrophic deep lake; we then applied the model to another lake with similar abiotic and biotic characteristics. Field data from a survey of aquatic macrophytes, undertaken on the same dates as EO data, were used within the accuracy assessment. The aquatic vegetation was discerned into three classes: emergent, floating, and submerged aquatic vegetation. Geographic object-based image analysis (GEOBIA) proved to be effective in discriminating the three classes in both study areas. Results showed high effectiveness of the classification model in terms of overall accuracy, particularly for the emergent and floating classes. In the case of submerged aquatic vegetation, challenges in their classification prompted us to establish specific criteria for their accurate detection. Overall results showed that GEOBIA based on spectral indices was suitable for mapping aquatic vegetation in oligotrophic/mesotrophic deep lakes. EO data can contribute to large-scale coverage and high-frequency monitoring requirements, being a complementary tool to in situ monitoring.},
keywords = {aquatic vegetation, General Earth and Planetary Sciences, GEOBIA, lake monitoring, Mediterranean lakes, remote sensing, Sentinel-2 imagery, WFD},
pubstate = {published},
tppubtype = {article}
}
Aquatic vegetation is an essential component of lake ecosystems, used as a biological indicator for in situ monitoring within the Water Framework Directive. We developed a hierarchical object-based image classification model with multi-seasonal Sentinel-2 imagery and suitable spectral indices in order to map the aquatic vegetation in a Mediterranean oligotrophic/mesotrophic deep lake; we then applied the model to another lake with similar abiotic and biotic characteristics. Field data from a survey of aquatic macrophytes, undertaken on the same dates as EO data, were used within the accuracy assessment. The aquatic vegetation was discerned into three classes: emergent, floating, and submerged aquatic vegetation. Geographic object-based image analysis (GEOBIA) proved to be effective in discriminating the three classes in both study areas. Results showed high effectiveness of the classification model in terms of overall accuracy, particularly for the emergent and floating classes. In the case of submerged aquatic vegetation, challenges in their classification prompted us to establish specific criteria for their accurate detection. Overall results showed that GEOBIA based on spectral indices was suitable for mapping aquatic vegetation in oligotrophic/mesotrophic deep lakes. EO data can contribute to large-scale coverage and high-frequency monitoring requirements, being a complementary tool to in situ monitoring.