Harokopio University
Χαρτογράφηση βυθομετρίας και θαλάσσιων ενδιαιτημάτων σε παράκτια ύδατα με τη χρήση δεδομένων παρατήρησης της Γης
Abstract
dc:descriptionThe aim of this research is to explore and provide a scientifically based implementation framework of the most popular semi-empirical bathymetry retrieval algorithms from satellite data (SDB), the Lyzenga model (1978, 1981, 1985, 2006) and the Stumpf et al. al. (2003), to produce reliable, detailed and low-cost mapping of shallow coastal areas (up to 30 m depth). The study of the "conditions" of their application was carried out in three study areas in the North, Central and South Aegean (Greece). The algorithms were applied to heterogeneous bottoms in the zone up to 20-25 m in the three areas, on the one hand and to homogeneous ones - considering the type of substrate and marine habitats as derived from the application of image classification methods in the study areas, on the other. In situ data for depth and substrate type, were used to calibrate the algorithms and to verify and evaluate the accuracy of the retrieved depths. The link between the size of the training sample and the accuracy of the results was also studied. The results identified that the size and the distribution of the training sample is important and is associated with the accuracy of the estimations. Both models produce more accurate estimations when applied to a homogeneous bottom. The Lyzenga model seems to more accurately capture the bathymetry compared to the Stumpf model in any case - of homogeneous or heterogeneous sea bottom. In the case of heterogeneous seabed, both models, show weakness in the estimations of the very shallow depths - up to 2 m, whereas they seem to perform well at depths up to 20 m. However, when the models applied in homogeneous seabed, the efficiency of the very shallow depths is improved and satisfactory depths of up to 25 m are also provided. As regards the optimal size of the training sample, both algorithms follow the model of N= 10-20 per predictor for heterogeneous bottoms, where predictor is considered the number of spectral bands participating in the mathematical expressions of the algorithms.
Degree
thesis:*- Grantor dc:publisher
- Harokopio University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Drakopoulou, Paraskevi
- Δρακοπούλου, Παρασκευή
Subjects
dc:subject × 18- Δορυφορική βαθυμετρία
- Πολυφασματικές εικόνες
- Θαλάσσια ενδιαιτήματα
- Γραμμικό μοντέλο
- Μοντέλο αναλογιών
- Satellite-derived bathymetry
- Multispectral imagery
- Linear model
- Ratio model
- Marine habitats
- Γεωεπιστήμες και Επιστήμες Περιβάλλοντος
- Φυσικές Επιστήμες
- Επιστήμες περιβάλλοντος
- Γεωεπιστήμες, διεπιστημονική προσέγγιση
- Earth and Related Environmental Sciences
- Natural Sciences
- Environmental Sciences
- Geosciences, interdisciplinary
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/55306