Centre de documentation
scientifique

Sea surface temperature prediction model for the Black Sea by employing time-series satellite data: a machine learning approach / Hakan Oktay Aydınlı in Applied geomatics, vol 14 n° 4 (December 2022)  

Public [article]  inApplied geomatics > vol 14 n° 4 (December 2022) . - pp 669 - 678 Titre : Sea surface temperature prediction model for the Black Sea by employing time-series satellite data: a machine learning approach Type de document : Article/Communication Auteurs : Hakan Oktay Aydınlı, Auteur ; Ali Ekincek, Auteur ; Mervegül Aykanat-Atay, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 669 - 678 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] analyse diachronique [Termes IGN] apprentissage automatique [Termes IGN] classification par réseau neuronal récurrent [Termes IGN] détection de changement [Termes IGN] données Copernicus [Termes IGN] image Aqua-MODIS [Termes IGN] méthode des moindres carrés [Termes IGN] modèle de simulation [Termes IGN] Noire, mer [Termes IGN] optimisation (mathématiques) [Termes IGN] série temporelle [Termes IGN] température de surface de la mer Résumé : (auteur) High temporal resolution remote sensing images provide continuous data about the marine environment, which is critical for gaining extensive knowledge about the aquatic environment and marine species. Sea surface temperature (SST) is one of the basic parameters that can be obtained with the help of remote sensing. Long-term alterations in the SST can affect the aquatic environment and marine species, such as the life expectancy of anchovies in the Black Sea. Forecasting the dynamics of SSTs is crucial for detecting and eliminating the SST-oriented impacts. The goal of the current study is to construct a predictive model to estimate the daily SST value for the mid-Black Sea using a machine learning approach by employing time-series satellite data from 2008 to 2021. Turkey’s mid-Black Sea coastal line, comprising Ordu, Samsun, and Sinop stations, was chosen as the study area. The SST predictive model was represented by applying the recurrent neural network (RNN) long- and short-term memory (LSTM). Adam stochastic optimization was used for validation, and the mean square error (MSE) for each location was found to be 0.914, 0.815, and 0.802, respectively. The findings indicate that our model is significantly promising for accurate and effective short- and midterm daily SST prediction. Numéro de notice : A2022-894 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1007/s12518-022-00462-y Date de publication en ligne : 23/08/2022 En ligne : https://doi.org/10.1007/s12518-022-00462-y Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102242 [article]

Mainstreaming remotely sensed ecosystem functioning in ecological niche models / Adrián Regos in Remote sensing in ecology and conservation, vol 8 n° 4 (August 2022)  

Public [article]  inRemote sensing in ecology and conservation > vol 8 n° 4 (August 2022) . - pp 431 - 447 Titre : Mainstreaming remotely sensed ecosystem functioning in ecological niche models Type de document : Article/Communication Auteurs : Adrián Regos, Auteur ; João Gonçalves, Auteur ; Salvador Arenas-Castro, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 431 - 447 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] carbone [Termes IGN] écologie forestière [Termes IGN] écosystème forestier [Termes IGN] habitat animal [Termes IGN] image Aqua-MODIS [Termes IGN] image Terra-MODIS [Termes IGN] indice de végétation [Termes IGN] niche écologique Résumé : (auteur) Biodiversity is declining globally at unprecedented rates. Ecological niche mod-els (ENMs) are one of the most widely used toolsets to appraise global changeimpacts on biodiversity. Here, we identify a variety of advantages of incorporat-ing remotely sensed ecosystem functioning attributes (EFAs) into ENMs. Thedevelopment of ENMs that explicitly incorporate ecosystem functioning willallow a more holistic and integrative perspective of the habitat dynamics. Thesynergies between the increasingly available open-access satellite images andcloud-based platforms for planetary-scale geospatial analysis offer an unprece-dented opportunity to incorporate ecosystem processes and disturbances (suchas fires, insect outbreaks or droughts) that have been so far largely neglected inecological niche characterization and modelling. The most paradigmatic exam-ple of EFAs is the application of time series of spectral vegetation indicesrelated to primary productivity and carbon cycle. EFAs related to surface energybalance and water cycles derived from remote sensing products such as landsurface temperature or soil moisture enable a fine-scale characterization of thespecies’ niche—eventually improving the predictive performance of ENMs. Allthese advantages confirm that a new generation of ENMs based on such EFAswould offer great perspectives to increase our ability to monitor habitat suit-ability trends and population dynamics. However, despite the technicaladvances and increasing effort of remote sensing community to develop inte-grative EFAs, ENMs have yet to make full profit of the most recent develop-ments by integrating them in ENMs. A coordinated agenda for remote sensingexperts and ecological modellers will be essential over the coming years tobridge the gap between remote sensing and ecology disciplines and to take full(and timely) advantage of the fast-growing body of Earth observation data andremote sensing technologies—with special emphasis on the development andtesting of new variables related to key processes driving ecosystem functioning. Numéro de notice : A2022-715 Affiliation des auteurs : non IGN Thématique : BIODIVERSITE/IMAGERIE Nature : Article DOI : 10.1002/rse2.255 Date de publication en ligne : 15/02/2022 En ligne : https://doi.org/10.1002/rse2.255 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101614 [article]

Monitoring and analysis of crop irrigation dynamics in Central Italy through the use of MODIS NDVI data / Marta Chiesi in European journal of remote sensing, vol 55 n° 1 (2022)  

Public [article]  inEuropean journal of remote sensing > vol 55 n° 1 (2022) . - pp 23 - 36 Titre : Monitoring and analysis of crop irrigation dynamics in Central Italy through the use of MODIS NDVI data Type de document : Article/Communication Auteurs : Marta Chiesi, Auteur ; Luca Angeli, Auteur ; Piero Battista, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 23 - 36 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] bilan hydrique [Termes IGN] carte agricole [Termes IGN] cultures irriguées [Termes IGN] image Aqua-MODIS [Termes IGN] image Sentinel-MSI [Termes IGN] image Terra-MODIS [Termes IGN] irrigation [Termes IGN] Normalized Difference Vegetation Index [Termes IGN] Toscane (Italie) Résumé : (auteur) A recent study has proposed and tested a semi-empirical method to estimate crop irrigation based on a water balance logic and Sentinel-2 Multi Spectral Instrument (MSI) NDVI imagery. The current paper aims at extending the same approach to the analysis of the main irrigation patterns occurred in Tuscany (Central Italy) during the 2000–2019 period. This operation was made possible by feeding the irrigation water (IW) estimation method with 250-m spatial resolution Terra/Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI images. The results of this operation were first assessed versus various reference datasets available for the region; next, the annual maps of IW estimated for the 20 study years were analyzed at province scale in conjunction with relevant agricultural statistics. The use of MODIS in place of MSI images reduces the IW estimation accuracy irregularly at local scale, depending on the size and spatial arrangement of irrigated and non-irrigated fields; the reduction in accuracy is, however, marginal over relatively large areas. Irrigated crops are decreasing throughout most Tuscany provinces, while they are increasing in the most southern and driest province. The possible reasons and implications of these findings are finally discussed in relation to the main environmental issues affecting the region. Numéro de notice : A2022-099 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/IMAGERIE Nature : Article DOI : 10.1080/22797254.2021.2013735 Date de publication en ligne : 05/01/2022 En ligne : https://doi.org/10.1080/22797254.2021.2013735 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99549 [article]

Studying informativeness of satellite image texture for sea ice state retrieval using deep learning methods / Clément Fougerouse (2022) 

Public Titre : Studying informativeness of satellite image texture for sea ice state retrieval using deep learning methods Type de document : Mémoire Auteurs : Clément Fougerouse, Auteur Editeur : Champs-sur-Marne : Ecole nationale des sciences géographiques ENSG Année de publication : 2022 Importance : 47 p. Format : 21 x 30 cm Note générale : Bibliographie Rapport de projet pluridisciplinaire, cycle ING2 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications [Termes IGN] données d'entrainement (apprentissage automatique) [Termes IGN] glace de mer [Termes IGN] image Aqua-AMSR [Termes IGN] image C-SAR [Termes IGN] image radar moirée [Termes IGN] inférence [Termes IGN] optimisation (mathématiques) [Termes IGN] réseau neuronal convolutif [Termes IGN] restauration d'image Index. décimale : PROJET Mémoires : Rapports de projet - stage des ingénieurs de 2e année Résumé : (Auteur) De nos jours, la détermination des glaces de mers se fait manuellement et est réalisée par des experts, les cartes obtenues ne sont donc pas bien précises et peuvent comporter des erreurs. L’objectif de l’étude est de pouvoir automatiser la classification des différents types de glaces de mer à partir d’images satellitaires SAR et AMSR2, en utilisant des réseaux de neurones convolutifs et d’améliorer la précision des réseaux déjà existants. Pour cela, nous partons des réseaux existants et nous rajoutons de nouvelles données d’apprentissages et nous modifions la structure du réseau de neurones convolutif. Puis nous étudions la texture des images pour pouvoir prendre en compte les formes des glaces et ainsi de créer plusieurs classes pour les glaces de mers. Que ce soit avec l’ajout de nouvelles données ou la modification de la structure du réseau, la précision des prédictions du réseau de neurones a grandement été amélioré. Nous passons d’une précision de 74% en moyenne sur les quatre classes utilisées à une moyenne de 95% après toutes les améliorations réalisées. Notons également, que la détection de la présence ou non de glace est très précise 98%. Quant à l’ajout des nouvelles classes et à la prise en compte de la texture des images satellitaires, nous obtenons des résultats très intéressants : le classificateur permet de distinguer certaines combinaisons, mais a du mal pour d’autres, notamment pour les glaces qui ont des petites formes. Ainsi, cette étude a permis d’améliorer considérablement la précision des réseaux existants pour classer la glace dans les quatre types habituels bien qu'ils restent moins performants pour classer en prenant en compte la forme des glaces. L’étude du caractère informatif a permis de connaitre les combinaisons détectées par la texture des images SAR. Note de contenu : 1. Introduction 2. Data used for training the CNN 2.1 NetCDF files 2.2 SAR data 2.3 AMSR2 data 2.4 Ice Chart 3. Processing 3.1 Overview 3.2 Statistical analysis 3.3 Preprocessing 3.3 Training 3.4 Inference 3.4 Baseline binary CNN 3.5 Baseline continuous CNN 3.6 Adding the larger area SAR data 3.7 Adding the AMSR2 data 3.8 Optimization 3.9 Experiments with informativeness 4. Results 4.1 Statistics 4.2 Baseline Binary 4.3 Hugo continuous 4.4 Extended SAR sub-image 4.5 AMSR2 4.6 Optimization 4.7 Informativeness tests 5. Conclusion and discussion Numéro de notice : 26868 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Mémoire de projet pluridisciplinaire Organisme de stage : Nansen Environmental and Remote Sensing Center NERSC Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101688

Documents numériques

en open access Studying informativeness of satellite image texture for sea ice state retrieval using deep learning methods - pdf auteur Adobe Acrobat PDF

Snow cover change assessment in the upper Bhagirathi basin using an enhanced cloud removal algorithm / Mritunjay Kumar Singh in Geocarto international, vol 36 n° 20 ([01/12/2021])  

Public [article]  inGeocarto international > vol 36 n° 20 [01/12/2021] . - pp 2279 - 2302 Titre : Snow cover change assessment in the upper Bhagirathi basin using an enhanced cloud removal algorithm Type de document : Article/Communication Auteurs : Mritunjay Kumar Singh, Auteur ; Renoj J. Thayyen, Auteur ; Sanjay K. Jain, Auteur Année de publication : 2021 Article en page(s) : pp 2279 - 2302 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] analyse diachronique [Termes IGN] analyse spatio-temporelle [Termes IGN] bassin hydrographique [Termes IGN] bilan de masse [Termes IGN] changement climatique [Termes IGN] eau de fonte [Termes IGN] filtrage spatiotemporel [Termes IGN] glacier [Termes IGN] Himalaya [Termes IGN] image Aqua-MODIS [Termes IGN] image Terra-MODIS [Termes IGN] Inde [Termes IGN] manteau neigeux [Termes IGN] MNS ASTER [Termes IGN] nébulosité [Termes IGN] nuage [Termes IGN] variation saisonnière Résumé : (auteur) This research paper proposes a new five-step protocol to enhance the result of existing cloud removal algorithms using Moderate Resolution Imaging Spectroradiometer (MODIS) daily snow cover products (SCPs). The study has been carried out for the upper Bhagirathi basin (up to Maneri Hydropower Project) located in the Western Himalaya. Gafurov and Bárdossy test employed to validate the performance of the proposed method, followed by comparing with the field observed snow cover duration (SCD) data. The result shows that the mean overall accuracy of the proposed method for cloud removal is about ∼95%. However, the cloud removal method by Gafurov and Bardossy also achieved similar mean overall accuracy but with the higher variability within the individual images as compared with the variability within the results obtained by the proposed method. SCD computed from cloud removed SCPs matched significantly with the field observed SCD for a point location, supporting the accuracy achieved by the cloud removal method. This study also examines the spatiotemporal variability of the snow cover in the study area during the past 18 years (2000–2018). During the observation period, no specific trend was observed for annual maximum snow cover, while yearly minimum snow cover in the basin showed an increasing trend since 2010. Seasonally, December and June month witnessed significant changes. December experienced a declining trend in snow cover between 3000–6000 m a.s.l. covering 88% of the basin area, whereas, June showed an increasing trend between 4500 to 6000 m (a.s.l.). This elevation range covers 61% of the basin area, including core 86% of the glacier area within the basin. September and October experienced the highest inter-annual snow cover variability. Maximum snow cover month of February and minimum snow cover month of August experienced the least variability. The present study suggests significant elevation-dependent increasing as well as the decreasing trend in the snow cover with seasonal contrast, which may affect the glaciers as well as the hydrological behavior of the basin. Numéro de notice : A2021-832 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2019.1704069 Date de publication en ligne : 19/12/2021 En ligne : https://doi.org/10.1080/10106049.2019.1704069 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99005 [article]

Estimating regional soil moisture with synergistic use of AMSR2 and MODIS images / Majid Rahimzadegan in Photogrammetric Engineering & Remote Sensing, PERS, vol 87 n° 9 (September 2021)  

Permalink

Retrieval of ultraviolet diffuse attenuation coefficients from ocean color using the kernel principal components analysis over ocean / Kunpeng Sun in IEEE Transactions on geoscience and remote sensing, vol 59 n° 6 (June 2021)  

Permalink

Detection of rainstorm pattern in arid regions using MODIS NDVI time series analysis / Mohamed E. Hereher in Geocarto international, vol 36 n° 8 ([01/05/2021])  

Permalink

Validation and analysis of Terra and Aqua MODIS, and SNPP VIIRS vegetation indices under zero vegetation conditions: A case study using Railroad Valley Playa / Tomoaki Miura in Remote sensing of environment, vol 257 (May 2021)  

Permalink

Precipitable water vapor fusion based on a generalized regression neural network / Bao Zhang in Journal of geodesy, vol 95 n° 4 (April 2021)  

Permalink

Urban heat island formation in greater Cairo: Spatio-temporal analysis of daytime and nighttime land surface temperatures along the urban–rural gradient / Darshana Athukorala in Remote sensing, vol 13 n° 7 (April-1 2021)  

Permalink

Geospatial analysis of September, 2019 floods in the lower gangetic plains of Bihar using multi-temporal satellites and river gauge data / C.M. Bhatt in Geomatics, Natural Hazards and Risk, vol 12 n° 1 (2021)  

Permalink

Monitoring tree-crown scale autumn leaf phenology in a temperate forest with an integration of PlanetScope and drone remote sensing observations / Shengbiao Wu in ISPRS Journal of photogrammetry and remote sensing, vol 171 (January 2021)  

Permalink

Semantic segmentation of sea ice type on Sentinel-1 SAR data using convolutional neural networks / Alissa Kouraeva (2021) 

Permalink

Cross-calibration of MODIS reflective solar bands with Sentinel 2A/2B MSI instruments / Amit Angal in IEEE Transactions on geoscience and remote sensing, vol 58 n° 7 (July 2020)  

Permalink