Formosat-2 de-commissionned


It is with a deep nostalgia that I learned that Formosat-2 has just been decommissionned, after some of its components broke down after 12 years of acquisitions (nominal life extended by 8 years !).


When the satellite was launched, we were searching for data sets that would allow us simulate the features of VENµS satellite (which was then announced for 2008-2009, but still not launched yet :( ) and of the "GMES continental" satellite which finally became Sentinel-2. We needed high resolution images with a frequent repetitivity under constant view angles.


We were starting to use SPOT time series, limiting their view angles to +/- <10 degrees off-nadir, and we kept asking for the Indian OceanSat which was on the orbit targetted by Venµs (we never got any answer), when we first heard of Formosat-2. Formosat-2 had exactly the features we needed : 8m resolution, 24 km field of view, a one day repeat cycle (constant view angles !) and 4 spectral bands including the blue one which was lacking on SPOT.

The first dates of our first Formosat-2 time series in Morocco


With the help of Gilbert Pauc (a few months before he retired), CNES set up an agreement with SPOT Image and the Taiwan Space Agency, to trade Formosat-2 time series against radiometric calibration work from CNES. As a result, CNES took charge of Formosat-2 calibration for all its life (I started it, being still in CNES Image quality team), and we benefited from a few free time series and from a reduced price for the next ones (reduced but still expensive : 1000€/image).

Continue reading

La fin de l'exploitation de Formosat-2


C'est avec nostalgie que je vous fais part de la fin d'exploitation du satellite Formosat 2, dont un certain nombre de composants sont tombés en panne récemment, après 12 ans de fonctionnement (8 de plus que prévu).


Au moment de son lancement, nous cherchions des données nous permettant de simuler les caractéristiques du satellite VENµS (prévu pour 2008-2009, mais pas encore lancé à ce jour :( ) et du futur "satellite continental GMES" qui allait finir par devenir Sentinel-2.  Il nous fallait des données à haute résolution, acquises fréquemment, et sous des angles de prise de vue constants.


A ce moment là, les données LANDSAT étaient payantes, et nous en étions à essayer d'utiliser SPOT en limitant ses prises de vues à +/- <10 degrés du Nadir et à demander des données du satellite Indien OceanSat (situé sur la même orbite que celle prévue pour Venµs), quand nous avons appris le lancement de Formosat-2. Formosat-2 disposait des caractéristiques souhaitées : 8m de résolution, 24 km de champ, un cycle orbital d'un jour, et 4 bandes spectrales, dont le bleu.

Les premières dates de notre première série d'images Formosat-2 au Maroc

Avec l'aide de Gilbert Pauc (avant son départ à la retraite),  le CNES a mis en place un accord avec SPOT Image et l'agence spatiale de Taiwan (NSPO), pour échanger une prestation d'étalonnage contre un accès à des séries temporelles de données. C'est donc le CNES qui a pris en charge l'étalonnage du satellite sur toute sa durée de vie (c'est moi qui ai commencé, je faisais encore de l'étalonnage à ce moment là), et en échange, nous avons bénéficié de quelques séries Formosat-2 gratuites, puis d'un tarif préférentiel (quoique élevé :  1000€/image).

Continue reading

Sentinel-2A (and Landsat-8) capture a giant ice avalanche in Tibet

After reading my previous post about the Rutog ice avalanche, my distinguished colleagues Antoine R. and Olivier H. challenged me to look for a pre-event image to better highlight the avalanche area. The closest clear-sky image that I could find is a Landsat-8 image that was acquired on June 24 (23 days before the slide).


Sequence of two Landsat-8 and Sentinel-2A images. Both images are level 1 product displayed as natural color composites. Click to enlarge.

Continue reading

Sentinel-2A captures a giant ice avalanche in Tibet

The Nature News website reported yesterday on a massive ice avalanche that happened in Rutog, Tibet, on 17 July 2016. This ice avalanche killed 9 people and may be one the largest ever observed. The ice and rock mixture spread over 6 km from the collapse point up to the Aru Co lake shoreline.

Sentinel-2A image of the Rutog ice avalanche acquired on 21-Jul-2016 (4 days after the event). Click on the image to see at full resolution (1 pixel = 10m).

Continue reading

Les séries temporelles de Sentinel-2 sont bien lisses !


Dans un précédent billet, j'ai montré un exemple de série temporelle obtenu sur un pixel de forêt de pins, qui met en valeur la bonne qualité des séries temporelles que l'on obtient avec les données Sentinel-2, traitées au niveau 2A par THEIA en utilisant la chaîne MACCS. (les données de niveau 2A sont corrigées des effets atmosphériques et fournies avec un masque de nuages et d'ombres de nuages). Malgré cette bonne qualité, il restait cependant un certain bruit apparent, et j'avais attribué une partie de ce bruit aux effets directionnels, les données présentées ayant été acquises depuis deux orbites différentes de Sentinel-2, donc sous des angles différents.


Correction des effets directionnels :

Je suis à peu près sur qu'il y a quelques sceptiques parmi les lecteurs de ce blog (j'en connais même quelques uns !), qui pourraient imaginer que les effets directionnels ont bon dos, et que les erreurs proviennent d'ailleurs. Pour essayer de les convaincre, j'ai fait tourner le modèle de correction directionnelle développé au CESBIO pour  produire des synthèses mensuelles de niveau 3A pour Sentinel-2.

Continue reading

TOMCAR-Sat : Remote Sensing on the Yenisei river!

Since October 2015, the TOMCAR-Sat project aims to use high resolution satellite products to monitor Dissolved Organic Carbon (DOC) fluxes on the Arctic river Yenisei. Specifically, its objectives consist to calibrate a DOC retrieval model from time-series of satellite images and to characterize dynamics of the factors that may affect DOC concentrations, i.e. the snow cover and vegetation.


Spot 5 Take 5 (03/06/2015) - Yenisei river (Igarka, Russia)

Continue reading

Sentinel-2 time series are smooth !


In a previous post, I showed a time series of pixels showing the good data quality of time series we get with Sentinel-2, processed to Level 2A by THEIA using MACCS  processor. However, I concluded that the plot I showed still had some low noise, and that a part of it was due to directional effects coming from the use of data coming from two different orbits of Sentinel-2, and therefore acquired from different viewing angles.


Directional effect correction

I am quite sure there are some skeptic ones among the readers of this blog (I know some of them...), that could imagine the small remaining noise is due to MACCS, or Sentinel2.  So I told myself I had to prove that a large part is due to directional effects, and implemented the directional correction model we developed at CESBIO for the L3A monthly syntheses products for Sentinel-2.

Continue reading

Snow and Fire in the Dragon mountains

NASA's blog "Image of the Day" recently featured two beautiful MODIS images of the snow cover in Lesotho. In late July Lesotho experienced its heaviest snowfall in two decades. The snow is not uncommon in Lesotho given that over 80% of the country lies above 1800 m (wikipedia). However the frequency of such snow events has been reducing over the past decades due to the ongoing climate change. As a result the shepherds are less accustomed to the snow conditions so that "a severe storm like the one in July 2016 has greater potential to kill sheep and shepherds" [1]. Continue reading

Sentinel-2 MACCS L2A reflectance stability

As MACCS was used to produce a large amount of images (all the images acquired over France), we were able to make some more validation tests, such as checking the surface reflectance stability. To do that, the region of Arles in Provence (France), tile 31TFJ, is a good site, with a lot of clear skies, and the chance to be observed from two different orbits by Sentinel-2.


Here is an example for a coniferous forest (Pines probably): The reflectances of the 4 Sentinel-2 10m spectral bands are stable indeed, as it may be seen on the top plot, while the bottom plot shows the variation of the aerosol optical thickness measured by MACCS. You may have noticed that towards the summer, we can see a few couples of points, acquired 3 days apart, with small differences in surface reflectance, noticeable in the NIR infrared (black dots). These differences are probably due to directional effects as the observation from two different orbits are made with different viewing angles. These effects may be corrected. We do that for instance in our level 3 products.

Top : surface reflectances corrected with MACCS as a function of time (blue : B2 (blue), green B3 (green), red B4 (red), black B8 (Near infrared)). Bottom, Aerosol Optical Thickness measured by MACCS.

The overall stability of surface reflectances is clearly a good sign of the quality of MACCS processing.


Some summer news about Sentinel-2 and THEIA's level 2A production



I had a week of work in the middle of my holidays, which enabled be to see how the work on Sentinel-2  L2A data distribution is progressing. My colleagues who worked all July (Dominique, Joelle, Céline...) did a great job : they were able to process a large data set of L2A products with MUSCATE : all the images taken by Sentinel-2 between November 2015 and end of july 2016. The quicklooks may be seen here. It is not by pure perversity that we show the data without enabling you to download them (a few demonstration products are available for download). The distribution server is still a provisional facility, and we are still chasing the last bugs (and we found some :-( )


If the bug correction does not take too long, we should be able to deliver the L2A products generated by MACCS within MUSCATE from September, starting with France and extending progressively.

Continue reading