Sargassum observations from MODIS: using aggregations context to filter false detections
International audience Since 2011, the distribution extent of pelagic Sargassum algae has substantially increased and now covers the whole Tropical North Atlantic Ocean, with significant inter-annual variability. The ocean colour imagery has been used as the only alternative to monitor such a vast a...
Main Authors: | , , , , , |
---|---|
Other Authors: | , , , , , , |
Format: | Conference Object |
Language: | English |
Published: |
HAL CCSD
2022
|
Subjects: | |
Online Access: | https://insu.hal.science/insu-03952248 https://doi.org/10.5194/egusphere-egu22-2900 |
id |
ftinsu:oai:HAL:insu-03952248v1 |
---|---|
record_format |
openpolar |
spelling |
ftinsu:oai:HAL:insu-03952248v1 2024-02-11T10:06:44+01:00 Sargassum observations from MODIS: using aggregations context to filter false detections Podlejski, Witold Descloitres, Jacques Chevalier, Cristele Minghelli, Audrey Lett, Christophe Berline, Léo Institut méditerranéen d'océanologie (MIO) Institut de Recherche pour le Développement (IRD)-Aix Marseille Université (AMU)-Institut national des sciences de l'Univers (INSU - CNRS)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS) MARine Biodiversity Exploitation and Conservation - MARBEC (UMR MARBEC ) Institut de Recherche pour le Développement (IRD)-Institut Français de Recherche pour l'Exploitation de la Mer (IFREMER)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM) Université de Lille Laboratoire d'Informatique et Systèmes (LIS) Aix Marseille Université (AMU)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS) à renseigner, Unknown Region 2022 https://insu.hal.science/insu-03952248 https://doi.org/10.5194/egusphere-egu22-2900 en eng HAL CCSD info:eu-repo/semantics/altIdentifier/doi/10.5194/egusphere-egu22-2900 insu-03952248 https://insu.hal.science/insu-03952248 BIBCODE: 2022EGUGA.24.2900P doi:10.5194/egusphere-egu22-2900 EGU22 https://insu.hal.science/insu-03952248 EGU22, 2022, à renseigner, Unknown Region. ⟨10.5194/egusphere-egu22-2900⟩ [SDU]Sciences of the Universe [physics] info:eu-repo/semantics/conferenceObject Conference papers 2022 ftinsu https://doi.org/10.5194/egusphere-egu22-2900 2024-01-24T17:27:26Z International audience Since 2011, the distribution extent of pelagic Sargassum algae has substantially increased and now covers the whole Tropical North Atlantic Ocean, with significant inter-annual variability. The ocean colour imagery has been used as the only alternative to monitor such a vast area. However, the detection is hampered by cloud masking, sunglint, coastal contamination and others phenomena. All together, they lead to false detections that cannot be discriminated with classic radiometric analysis, but may be overcome by considering the shape and the context of the detections. Here, we built a machine learning model based on spatial features to filter false detections. More specifically, Moderate-Resolution Imaging Spectroradiometer (MODIS, 1 km) data from Aqua and Terra satellites were used to generate daily map of Alternative Floating Algae Index (AFAI). Based on this radiometric index, Sargassum presence in the Tropical Atlantic North Ocean was inferred. For every Sargassum detections, five spatial indices were extracted for describing their shape and surrounding context and then used by a random forest binary classifier. Contextual features were most important in the classifier. Trained with a multi-annual (2016-2020) learning set, the classifier performs the filtering of daily false detections with an accuracy of 90%. This leads to a reduction of detected Sargassum pixels of 50% over the domain. The method provides reliable data while preserving high spatial and temporal resolutions (1 km, daily). The resulting distribution on 2016-2020 is consistent with the literature for seasonal and inter-annual fluctuations, with maximum coverage in 2018 and minimum in 2016. In particular, it retrieves the two areas of consolidation in the western and eastern part of the Tropical Atlantic Ocean associated with distinct temporal dynamics. At full resolution, the dataset allowed us to semi-automatically extract Sargassum aggregations trajectories from successive filtered images. Using those trajectories ... Conference Object North Atlantic Institut national des sciences de l'Univers: HAL-INSU |
institution |
Open Polar |
collection |
Institut national des sciences de l'Univers: HAL-INSU |
op_collection_id |
ftinsu |
language |
English |
topic |
[SDU]Sciences of the Universe [physics] |
spellingShingle |
[SDU]Sciences of the Universe [physics] Podlejski, Witold Descloitres, Jacques Chevalier, Cristele Minghelli, Audrey Lett, Christophe Berline, Léo Sargassum observations from MODIS: using aggregations context to filter false detections |
topic_facet |
[SDU]Sciences of the Universe [physics] |
description |
International audience Since 2011, the distribution extent of pelagic Sargassum algae has substantially increased and now covers the whole Tropical North Atlantic Ocean, with significant inter-annual variability. The ocean colour imagery has been used as the only alternative to monitor such a vast area. However, the detection is hampered by cloud masking, sunglint, coastal contamination and others phenomena. All together, they lead to false detections that cannot be discriminated with classic radiometric analysis, but may be overcome by considering the shape and the context of the detections. Here, we built a machine learning model based on spatial features to filter false detections. More specifically, Moderate-Resolution Imaging Spectroradiometer (MODIS, 1 km) data from Aqua and Terra satellites were used to generate daily map of Alternative Floating Algae Index (AFAI). Based on this radiometric index, Sargassum presence in the Tropical Atlantic North Ocean was inferred. For every Sargassum detections, five spatial indices were extracted for describing their shape and surrounding context and then used by a random forest binary classifier. Contextual features were most important in the classifier. Trained with a multi-annual (2016-2020) learning set, the classifier performs the filtering of daily false detections with an accuracy of 90%. This leads to a reduction of detected Sargassum pixels of 50% over the domain. The method provides reliable data while preserving high spatial and temporal resolutions (1 km, daily). The resulting distribution on 2016-2020 is consistent with the literature for seasonal and inter-annual fluctuations, with maximum coverage in 2018 and minimum in 2016. In particular, it retrieves the two areas of consolidation in the western and eastern part of the Tropical Atlantic Ocean associated with distinct temporal dynamics. At full resolution, the dataset allowed us to semi-automatically extract Sargassum aggregations trajectories from successive filtered images. Using those trajectories ... |
author2 |
Institut méditerranéen d'océanologie (MIO) Institut de Recherche pour le Développement (IRD)-Aix Marseille Université (AMU)-Institut national des sciences de l'Univers (INSU - CNRS)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS) MARine Biodiversity Exploitation and Conservation - MARBEC (UMR MARBEC ) Institut de Recherche pour le Développement (IRD)-Institut Français de Recherche pour l'Exploitation de la Mer (IFREMER)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM) Université de Lille Laboratoire d'Informatique et Systèmes (LIS) Aix Marseille Université (AMU)-Université de Toulon (UTLN)-Centre National de la Recherche Scientifique (CNRS) |
format |
Conference Object |
author |
Podlejski, Witold Descloitres, Jacques Chevalier, Cristele Minghelli, Audrey Lett, Christophe Berline, Léo |
author_facet |
Podlejski, Witold Descloitres, Jacques Chevalier, Cristele Minghelli, Audrey Lett, Christophe Berline, Léo |
author_sort |
Podlejski, Witold |
title |
Sargassum observations from MODIS: using aggregations context to filter false detections |
title_short |
Sargassum observations from MODIS: using aggregations context to filter false detections |
title_full |
Sargassum observations from MODIS: using aggregations context to filter false detections |
title_fullStr |
Sargassum observations from MODIS: using aggregations context to filter false detections |
title_full_unstemmed |
Sargassum observations from MODIS: using aggregations context to filter false detections |
title_sort |
sargassum observations from modis: using aggregations context to filter false detections |
publisher |
HAL CCSD |
publishDate |
2022 |
url |
https://insu.hal.science/insu-03952248 https://doi.org/10.5194/egusphere-egu22-2900 |
op_coverage |
à renseigner, Unknown Region |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
EGU22 https://insu.hal.science/insu-03952248 EGU22, 2022, à renseigner, Unknown Region. ⟨10.5194/egusphere-egu22-2900⟩ |
op_relation |
info:eu-repo/semantics/altIdentifier/doi/10.5194/egusphere-egu22-2900 insu-03952248 https://insu.hal.science/insu-03952248 BIBCODE: 2022EGUGA.24.2900P doi:10.5194/egusphere-egu22-2900 |
op_doi |
https://doi.org/10.5194/egusphere-egu22-2900 |
_version_ |
1790604650179723264 |