New version of PEPS (French Sentinel mirror site)

As you probably know, PEPS is the French Collaborative ground segment for Copernicus Sentinel program. And, first of all, it is a mirror site that distributes all the Sentinel data in near real time. These last weeks, real time was not available for Sentinel-2, as the data format and structure of Sentinel-2 products had deeply changed, and the software needed adaptation. PEPS team created a new collection, named "Sentinel-2 Single Tiles", coded "S2ST" to separate the old format from the new one. Now that the new version has been installed and validated, the PEPS mirror site is once again up to date.

Continue reading

Festival time with sentinel-2

Burning Man, black rock desert, USA.
Area : about 5 km2
Official attendance in 2016:: about 70000 people per day during 8 days.
1 people each 72 m2

Conclusion about my search of the event on Sentinel-playground: EASY! The event is iin the middle of the desert, high contrasts, long preparation time, no cloud, .

Burning Man

Festival des Vieilles Charrues, Carhaix, France..
Area: about 8 hectares for the music festival and 20 hectares for the camping area
Official attendance in 2016: about 70000 people / day during four days.
Density: 1 people each 4 m2

Conclusion about my search of the event on Sentinel-playground: LUCKY! I searched for other big european music festivals and found only this one. Most of the dates appeared to be cloudy or without image during the festival..This one was perfectly clear.

Conclusion: I still can't hear the music

Sentinel-2A unveils two new lakes in Borneo

The remote sensing company Planet has published images, which were captured by its fleet of nano-satellites, of Lake Bakun on the island of Borneo [1]. Planet tells us that this artificial lake is the largest lake in Southeast Asia since its flooding in 2011. Yet it is not visible on mapping web services like Bing Maps or Google Maps! It does show in OpenStreetMap at low resolution. This is due to the fact that the island of Borneo is one of the most cloudy places in the world so the usable optical remote sensing images are very rare.
Continue reading

Release of Theia Sentinel-2 L2A data over la Réunion

Because of the little issues we stumbled upon when producing Sentinel-2 level 2A data over France with MUSCATE, we had not yet demonstrated an end-to-end production, with a distribution on the operational server. While waiting for the end of the production of data over metropolitan France, we did this test the processing of Réunion Island.


September 2016 eruption witnessed by Sentinel-2A

The island is covered by only one Sentinel-2 tile, but the processing worked very smoothly, which is a very good sign after all teh difficulties we met. The data are now released on :


We processed all the available data, strating from January the 18th, 2016, until end of November. To produce the data acquired later on, we are waiting for the new version of PEPS, compatible with the new Sentinel-2 data format. We hope it will be possible as soon as February. .
The data set includes a nice image of the eruption of the Piton de la Fournaise volcano, in September. Regarding the data quality, you may find that the cloud shadows are a little exaggerated. We will try to fix that with a better tuning of the threshold for the next versions.


Global sea ice area : a scary plot

When I say "a scary plot", I don't mean a Chinese plot aimed at confusing some naive North Americans, it just mean a figure (an American one, by the way, made with data from the National Snow and Ice Data Center) that provides the total sea ice area over the globe. And this figure is scary :

The red curve corresponds to last year's sea ice area.

Let's hope it is partly due to the el-niño phenomenon we had last year, and that next years will come closer to the average curves. However, as a cold wave is about to reach western Europe, let's record it for later !

Pine Island Glacier rift

The rift of Pine Island Glacier's floating tongue is a cause of concern among glaciologists because it suggests that this part of the ice shelf is not stable and may collapse in the future. As explained in this EOS article [1]

Rifts usually form at the sides of an ice shelf where the ice is thin and subject to shearing that rips it apart. But this particular rift originated in the center of Pine Island Glacier’s ice shelf and propagated out to the margins.This implies that something weakened the center of the ice shelf. The most likely explanation is that a crevasse melted out at the bedrock level, driven by a warming ocean, according to the researchers."

Online survey: can we detect an expansion of the rift on the Sentinel-2 images only?

Sentinel-2 images of the Pine Island Glacier ice shelf rift. Each subset shows the eastern portion of the rift on a different date. Which one shows the longest rift according to you? Click on the image to vote!


The results of the survey are given in the next post.


[1] Lipuma, L. (2017), West Antarctic ice shelf breaking up from the inside out, Eos, 98, doi:10.1029/2017EO064743. Published on 04 January 2017

White Christmas in Aleppo

This is a Sentinel-2A image of Aleppo on Dec 22, showing some snow through the clouds. I do not know if this sunny interval lasted a long time, and I wonder if the inhabitants of Aleppo had wished snow for Christmas. I walked through the streets of Aleppo in July 2003 at the end of my bachelor's internship which was held mainly in Damascus and Palmyra. Aleppo is the most fascinating city that I have visited in Syria. I remember the alignments of soap, cloth and hardware stores that narrowed even more the narrow alleys of the old city. I remember the Persian carpets hanging from the balustrades of the ancient caravanserais. I remember countless smiles exchanged with people whose language I did not speak, in the souk, in the mosques, where tourists like me were welcome. I remember the kids playing soccer in front of the walls of the citadel. I wonder if kids played snowballs on December 22, 2016 in Aleppo. I'm sad and I'm a little ashamed too.

Aleppo (Sentinel-2A image of 22-Dec-2016)

The importance of a good cloud mask for the operational use of Sentinel-2 data


Top of Atmosphere reflectance in the 4 high resolution channels of Sentinel-2

The above plot shows the TOA reflectance time series gathered by Sentinel-2 over a pixel chosen randomly in a tile in the centre of France (tile 31TDK, pixel 3000-7000), from L1C products.  Looking at the time series, it is rather difficult to tell what kind of surface was observed, even if a vegetation cycle seems to be present.  As we will see below, most of the observed noise is due to the presence of clouds or cloud shadows.

Continue reading