Understanding the evolution of natural systems spatio-temporal dynamics is paramount in modern ecology. We focused on highlighting and analysing temporal and spatial dynamics of remotely-sensed chlorophyll a concentration. This pigment is linked with phytoplankton production, which in turn play a pivotal role in marine environment. Satellite platforms offer a synoptic view of surface chlorophyll a concentration for the last two decades. Coupling this source of information with statistical and Machine Learning techniques could help highlighting eventual patterns. We merged the Mediterranean chlorophyll a satellite data for the last two decades into a single dataset. We tested several techniques for reconstructing missing data and performed a general analysis. Finally, we implemented a Dynamic Time Warping Self-Organizing Map algorithm to cluster our series showing that an elastic distance measure outperforms a non-elastic one. The proposed satellite data management and analysis provided insights on spatio-temporal chlorophyll a dynamics in the Mediterranean Basin.
Mattei, F., Scardi, M. (2022). Mining satellite data for extracting chlorophyll a spatio-temporal patterns in the Mediterranean Sea. ENVIRONMENTAL MODELLING & SOFTWARE, 150, 105353.
Mining satellite data for extracting chlorophyll a spatio-temporal patterns in the Mediterranean Sea
M. Scardi
2022-01-01
Abstract
Understanding the evolution of natural systems spatio-temporal dynamics is paramount in modern ecology. We focused on highlighting and analysing temporal and spatial dynamics of remotely-sensed chlorophyll a concentration. This pigment is linked with phytoplankton production, which in turn play a pivotal role in marine environment. Satellite platforms offer a synoptic view of surface chlorophyll a concentration for the last two decades. Coupling this source of information with statistical and Machine Learning techniques could help highlighting eventual patterns. We merged the Mediterranean chlorophyll a satellite data for the last two decades into a single dataset. We tested several techniques for reconstructing missing data and performed a general analysis. Finally, we implemented a Dynamic Time Warping Self-Organizing Map algorithm to cluster our series showing that an elastic distance measure outperforms a non-elastic one. The proposed satellite data management and analysis provided insights on spatio-temporal chlorophyll a dynamics in the Mediterranean Basin.File | Dimensione | Formato | |
---|---|---|---|
mattei&scardi_env_mod_soft_2022.pdf
solo utenti autorizzati
Licenza:
Copyright dell'editore
Dimensione
4.34 MB
Formato
Adobe PDF
|
4.34 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.