NASA selected CESBIO's land cover map of France as "image of the day"


It is a well deserved recognition for the land cover product developed at CESBIO : NASA's earth observatory blog dedicated its "image of the day" blog post for the 15th of August to this product.


This work was done in the framework of Theia, using iota2 free software also developed at CESBIO. Thanks to the information conveyed by time series,iota2 is a fully automatic processor.

Jordi Inglada's team first published in 2015 a land cover map based on LANDSAT 8 data acquired in 2014, with a 30m resolution.This production was followed, in 2017 by a new map, based on Sentinel-2 data, acquired in 2016, at 10m resolution.

Continue reading

La NASA choisit le produit d'occupation des sols OSO du CESBIO comme "image du jour "


C'est une reconnaissance bien méritée pour le produit d'occupation des sols OSO développé et produit au CESBIO : le blog "earthobservatory" de la NASA a consacré son article "image du jour" du 15 août à ce produit.


Ce travail a été conduit dans le cadre de Theia, avec le logiciel libre iota2 développé au CESBIO. Grâce à l'utilisation de toute l'information contenue dans les séries temporelles  d'images, la chaîne iota2 arrive à fournir un résultat de grande qualité, tout en étant complètement automatique.


L'équipe de Jordi Inglada a publié une première carte en 2015, basée sur les données LANDSAT 8 acquises en 2015, avec une résolution de 30m. Cette production a été suivie, début 2017, d'une carte basée sur les données Sentinel-2, acquises en 2016, cette fois à une résolution de 10m.

Continue reading

Monthly cloud free syntheses merging Sentinel-2 and Landsat 8

To compute a cloud free synthesis of surface reflectances every month, a good repetitivity of observations is necessary. The weighted average method we developed at CESBIO, and which will be part of ESA's sen2agri system was coded by Cosmin Udroiu at CS Romania. It was meant to work with both Sentinel-2 sensors and an observation every fifth day. As we are still waiting for the launch of Sentinel-2B, the monthly syntheses obtained with Sentinel-2A alone really lack cloud free data.


On the left, the Sentinel-2A monthly synthesis, above Odessa (Ukraine) in May, and on the right its flag, with, in black, the areas flagged as cloud or cloud shadow. When a pixel is flagged as cloud or cloud shadow, the monthly synthesis provides the minimum blue reflectance, which tends to avoid clouds (if possible), but often selects cloud shadows.


Fortunately, the Sen2agri L3A processor is designed to work with LANDSAT 8 too, as both satellites have similar spectral bands, and as the MACCS atmospheric correction used to produce the L2A input products works for both sensors. Of course LANDSAT 8 geometric resolution is not that of Sentinel-2, so to avoid degrading Sentinel-2 imagery when LANDSAT8 data are available, we give Landsat 8 a very low weight in the weighted average. As a result, Landsat is really taken into account only when no cloud free Sentinel-2 was available during the synthesis period.


Same result as above, but including LANDSAT 8 data. A cloud free date at least is now found for every pixel. The water mask obtained from Level 2A product is a little wrong on the Landsat 8 image due to the presence of turbid waters and thin clouds. A solution for this problem will be implemented in next MACCS L2A version. Note that the monthly synthesis of both Sentinel-2 and LANDSAT-8 leaves nearly no visible artifacts on the lands.

For a better comparison of both versions, here is a little animation of composites with and without Landsat 8.

The Sen2Agri system is still in validation phase and should be released as open source next May, 6 months from now. The L3A synthesis processor will be also implemented within Theia and monthly L3A products will be distributed by Theia as it is already the case for L2A products.

High spatial and temporal resolution optical remote sensing data to estimate maize biomass and yield


Climate variability has a strong impact on maize yield. For example, the strong drought that occurred in 2016 led to lower yields across France, even for irrigated fields. Yield estimates have a significant strategic and economic importance. High spatial and temporal resolution remote sensing data are a valuable tool for providing yield estimates at a large scale.


In a recent study (Battude et al. 2016) based on optical image time series (combination of Formosat-2, Landsat-8, SPOT4-Take5 and Deimos-1, about two images per month), CESBIO researchers have developed a new method for the estimation of maize yield. A new formulation of SAFY agro-meteorological model taking into account of the observed seasonal variation of the specific leaf area (SLA) and the effective light use efficiency (ELUE) was proposed.


Results show that these modifications improve biomass estimates at local scale.


Comparison of measured and simulated Dry Aboveground Mass (DAM) with the original version of SAFY (left) and the new model version (right)

Yield estimates are compared to annual statistical values (Agreste) on two departments in the southwest of France : the Gers and the Haute-Garonne. Results show that the model reproduces well yields (R = 0.96; RRMSE = 4.6%), even if it sometimes overestimates the values for rainfed fields.


Comparison of simulated yield and Agreste values [t.ha-1] for the Gers and Haute-Garonne departments in 2013 (left) and 2014 (right), with the distinction between irrigated and rainfed fields. Standard errors associated to simulated values are reported.

GAI thus seems to be a good indicator for estimating the irrigated maize yield at regional scale. For rainfed fields, coupling SAFY with a water balance module simulating the soil water content  may improve yield estimates. Sentinel-2 mission offers new perspectives and its data should improve the model estimates.


Reference : Battude M., Al Bitar A., Morin D., Cros J., Huc M., Marais Sicre C., Le Dantec V., Demarez V. (2016) Estimating maize biomass and yield over large area using high spatial and temporal resolution Sentinel-2 like remote sensing data. Remote Sensing of Environment 184, 668-681 DOI: 10.1016/j.rse.2016.07.030

La télédétection optique à haute résolution spatiale et temporelle au service de l’estimation de la biomasse et du rendement du maïs


La variabilité climatique a un fort impact sur le rendement du maïs. Par exemple, les fortes sécheresses de 2016 ont conduit, même pour les parcelles irriguées, à une baisse des rendements à travers la France. Les estimations des rendements présentent un enjeu stratégique et économique important. La télédétection à haute résolution spatiale et temporelle est un outil précieux pour l’estimation à large échelle de ces rendements.


Dans une étude récente (Battude et al. 2016) basée sur des séries temporelles d'images optiques (combinaison d'images Formosat-2, Landsat-8, SPOT4-Take5 et Deimos-1, environ deux images par mois), les chercheurs du CESBIO ont mis en place une  nouvelle méthode d’estimation du rendement de mais. Une nouvelle formulation du modèle agro-météorologique SAFY prenant en compte la variation saisonnière observée de la surface spécifique foliaire (SLA) et de l’efficience de conversion de la lumière (ELUE) a été proposée.


Les résultats montrent que ces modifications améliorent les estimations de la biomasse à l’échelle locale.


Comparaison de la biomasse (DAM pour Dry Aboveground Mass) simulée et  mesurée avec à gauche la version d’origine du modèle SAFY  et à droite la nouvelle version proposée.


Les estimations de rendement sont comparées à des valeurs statistiques annuelles (Agreste) sur deux départements du Sud-ouest de la France : le Gers et la Haute-Garonne. Les résultats montrent que le modèle reproduit bien les rendements (R = 0.96; RRMSE = 4.6%), même s’il surestime parfois les valeurs pour les parcelles non irriguées.


Comparaison du rendement simulé et des données Agreste [t.ha-1] pour les départements du Gers et de la Haute-Garonne en 2013 (à gauche) et en 2014 (à droite), avec la distinction entre les parcelles irriguées et non irriguées. L’erreur standard associée aux valeurs simulées est reportée.


Le GAI s’avère donc être un bon indicateur pour l’estimation du rendement du maïs irrigué à l’échelle régionale. Pour les parcelles non irriguées, le couplage de SAFY avec un module de bilan hydrique simulant le contenu en eau du sol pourrait permettre d’améliorer les estimations de rendement. La mission Sentinel-2 offre de nouvelles perspectives et les données devraient permettre d'améliorer les estimations du modèle.


Référence : Battude M., Al Bitar A., Morin D., Cros J., Huc M., Marais Sicre C., Le Dantec V., Demarez V. (2016) Estimating maize biomass and yield over large area using high spatial and temporal resolution Sentinel-2 like remote sensing data. Remote Sensing of Environment184, 668-681 DOI: 10.1016/j.rse.2016.07.030

An overview of irrigation evolution in Central Asia with Landsat

In Central Asia former soviet republics, pre-independence water allocation and irrigation system infrastructure were well maintained and operated with massive funding from the central government of the Former Soviet Union. Since independence, the situation has changed dramatically politically, institutionally and technically. Political transition from a planned to a market economy has introduced ‘new’ concepts such as land tenure, water rights and different kinds of ownership. The institutional changes are described as a transition from former state collective farms – kholkhoz and sovkhoz – to smaller private farms. (FAO report #39)
The Kyrgyz Republic is a landlocked country in Central Asia with a total area of 198 500 km2 and about 6 million inhabitants. It became independent from the Soviet Union in August 1991. Most of the land formerly controlled by the 195 kolkhoz (collective farms) and 275 sovkhoz (state farms) has been distributed to their employees and dependants in the form of certificates extending 99 years of land-use rights. 1 million hectares of fields are irrigated : almost all irrigation uses surface water, and only 4.4% of the water comes from groundwater (FAO report #39). FAO Aquastat surveys show that the Kirghiz consumption of water for agriculture has dropped from 9486Mm3 in 1994 to 7447 Mm3 in 2006 (-21%), but they also say that « These data should be used with caution, since the reason for this is not clear. It may be the result of computation methods, data quality, changed cropping pattern or improved irrigation techniques. »


We (1) have been looking at the evolution of an irrigated scheme on the southern bank of the Issyk Kul lake. The water which is coming from Karak Batak glacier melt, snow melt, and other rain runoff flows along the Chon Kyzyl Suu river which is then diverted to the Bolshoi and Polanski canals.


The statistics of water diversion (2) to those two canals show a huge drop between 1996 and 2004, followed by a flat evolution until 2012. The volume diverted in the 2000s is half the volume of the 1990s, so we would expect that the cropped area of the 90’s was much bigger than in the 2010’s.

Continue reading

Theia's Landsat data are moving


It is true that many of us can be moved by the Landsat L2A data produced by THEIA, but this time, the data were literally moved. The presently used downloading site is being prepared to host Sentinel-2A data produced by the operational MUSCATE ground segment, and meanwhile, the prototype center goes on producing LANDSAT L2A data abovce France. However, their format is not compatible with the new server.

As a result, LANDSAT data will stay hosted on the old server, whose address has just been changed. THEIA's L2A LANDSAT data are now distributed at :

Continue reading

Les données LANDSAT de THEIA déménagent


Beaucoup d'utilisateurs nous l'ont dit, "elles déménagent vos données", mais cette fois, c'est au sens propre. En effet, le site de téléchargement actuel se prépare à accueillir les données sentinel-2 produites par le centre de production MUSCATE opérationnel,. Pendant ce temps, le centre prototype continue à traiter les données LANDSAT sur la France, mais l+eur format n'est pas compatible avec le nouveau serveur.

Pour cette raison, les données LANDSAT vont rester sur l'ancien serveur, qui va changer d'adresse. Provisoirement, les données LANDSAT sont donc distribuées sur ::

Continue reading