The final list of sites for SPOT5 (Take5)

After the preliminary list of sites issued last week, we are now able to release the final list of SPOT5 (Take5) sites, after CNES checked that the satellite is able to visit each of them every 5 days.

We had to do 5 changes compared to the preliminary list of sites. Three sites had to be removed because the density of sites was too high around them. We are very sorry for those of you who proposed these sites and were counting on using the data, we had technical constraints and choices had to be made. The unlucky sites are :
- Mariazell in Austria (but Eisenwursen site is now twice longer in compensation)
- KleinAltenDorf and BlackForest in Germany


2 sites have been moved slightly so that they can be merged in one segment of two images, with no need to move the mirror between them :
- Alsace and Mulhouse in France.
In the new file, we have added a colour code (and a comment provided by each site) which provides the observation day in the 5 days cycle. The 2nd of April should be a day 1 in the cycle, but it will have to be confirmed as it is only the nominal option. It will only be confirmed on the 3rd of April, after the orbit change operations have ended. From then on, you will be able to compute the observation days until the end of the experiment.

Here is the colour code :

  • Blue is day 1
  • Green is day 2
  • Red is day 3
  • Yellow is day 4
  • Pink is day 5
  • Black is Sodankyla site in Finland (which is observed every day except day 2)

A first look at SPOT5 (Take5) sites


Update (March 19th) : the final list of sites has been released.


In November, ESA released a call for proposals to select the sites to be observed during the SPOT5 (Take5) experiment. 62 proposals were received, with more than 100 sites submitted. ESA appointed a committee to review the proposals, a first list of 110 sites was issued, including the sites proposed by CNES and ESA for their own projects. CNES did a first assessment of the possibility to acquire all these data, taking into account the motion speed of the mirrors, the memory and download capacity, the amount of time necessary to prepare the programming for each 5 days cycle in this special context, far from the nominal orbit.. Following these tests, CNES and ESA agreed to add 40 more sites in the list.


The final feasibility study for the 40 additional sites has started at CNES and should be completed by mid March. But, since we are only one month from the start of experiment, CNES and ESA decided to release it in its provisional state, so that users can get ready, especially if they intend to collect field data. This list is now available below. The likelihood of its feasibility is quite high, although a couple of sites might be cancelled, where the density of sites is high. We hope the proposers of the few sites which might not be kept finally will understand. Areas of Interest of concern are: Austria, the region near the French German border, and Luxembourg.


From now on, and until the start of the experiment next month,  all the ancillary data necessary to do the processing need to be set-up, which requires :

  • defining the footprint of the products.
  • finding reference image for the ortho-rectification, in the LANDSAT 8 data base.
  • finding digital elevation models for each site. SRTM will be used in the latitudes between -60 and +60, but elsewhere, we will have to find other data sources.
  • tuning the gains for the sites usually covered by snow, in order to avoid to saturate the snow.

How MACCS estimates Aerosol Optical Depth.


I already explained in this blog  the principles of estimation of the aerosol optical thickness that we use to process the LANDSAT or SPOT5 (Take5) data, and soon Venµs or Sentinel-2 data, within the MACCS method developed at CESBIO, and used at CNES by THEIA. We started writing an article in 2010 to explain the method details and show the validation results, but we only found a sufficiently quiet period this autumn to finish it. The paper has just been published in remote sensing (MDPI), with open access. Enjoy your reading !


Hagolle, O.; Huc, M.; Villa Pascual, D.; Dedieu, G. A Multi-Temporal and Multi-Spectral Method to Estimate Aerosol Optical Thickness over Land, for the Atmospheric Correction of FormoSat-2, LandSat, VENμS and Sentinel-2 Images. Remote Sens. 2015, 7, 2668-2691.

From left to right, validation results for Aerosol Optical Thickness (AOT) measures by the multi-temporal method, the multi-spectral method and the combination of both. The combination of both methods allows to measure AOT in a much larger range of cases without degrading accuracy.

Venµs 140 - 75 SPOT5 (Take5)


It might look like a basket ball score, but it is only the number of proposals received for VENµS and SPOT5 (Take5) calls. One day before VENµS's deadline, on February the 16th, it looked like it would end as a draw, but 60 proposals were received on the last 24 hours by Gérard Dedieu, VENµS's PI.


In both cases, the number of proposed sites is even higher, since many proposals ask for several sites, sometimes up to ten, and I am afraid we will not be able to select all of them.

  • The 62 proposals received at ESA, plus the 13 received at CNES, were analysed and their accessibility is now being checked.
  • For VENµS, we are going to start reading. Great, I only have 139 proposals to read left...

Multitemp 2015



Although there is no link between this blog and the Multitemp 2015 symposium,
I am happy to support this international symposium dedicated to the applications and techniques to use the remote sensing image time series. (And maybe they will not sue us for using a similar name for this blog...).


That's the reason why I accepted to chair a session about SPOT4 (Take5) and the preparation of Sentinel-2 (and not just because Annecy is a very nice town in the French Alps). Many of you produced nice results from SPOT4 (Take5),data, or submitted  a site at the SPOT5 (Take5). call for sites.

Your extended abstracts are awaited on the symposium site  before February the 27th..

SPOT5 (Take5) is getting ready


The SPOT5 (Take5) project, lead by Sylvia Sylvander at CNES et Bianca Hoersch at ESA is getting ready. Here a a few news about the agenda of the coming months :

  • SPOT5 ends its operational life on March the 31st

SPOT5 tracks during Take5 experiment. In blue, the tracks from day 1 (every 5th day, starting from the 5th of April), day 2 is in green, day free in yellow, day 4 in orange and day five in red.


  • The orbit change manoeuvre should take place on April the Second. We know now the satellite tracks for the whole experiment, thanks to Jean-Marc Walter from CNES. It is not exactly the same as SPOT4 (Take5). For the lucky users who will get a site for both experiments, the observation angles will be different.
  • After the manoeuvre, a margin of a few days is kept in case a correction is necessary. Then a first image will be taken to check that all the processing works, as we did for SPOT4 (Take5).
  • Then we will get some rest for the Easter Week-end, and the first acquisitions will be programmed for the 8th of April. However, if something prevented us from doing the manoeuvre on the Second of April, this whole schedule will be delayed by a few weeks, as we lack manpower to prepare a backup solution. But it is really not likely !
  • F. Daniaud (CS-SI for CNES), is now assessing the possibility to observe the sites that were asked by ESA, by ESA's call and by CNES. If we are unlucky, some modifications might be necessary, by moving or removing sites. This task is heavy, because the satellite is not on its nominal orbit, and must be programmed manually, site per site. And this time, we will observe more than 100 sites !
  • Martine Béhague and the SPOT5 (Take5) operations team (CNES DCT/OP) are preparing simulations and tests to check that all the operations will work.
  • The interfaces with Airbus D & S (ex Spot Image) are being prepared.
  • The ground processing software preparation has started with the THEIA CNES team, with support from Cap Gemini and CESBIO. Mireille Huc, at CESBIO, just released the new version of MACCS prototype, which will handle the SPOT5 (Take5) data.
  • Vincent Poulain (Thales-IS) is now developing some tools to help us prepare the auxiliary data needed to process each site. For each site, we need to define its footprint, choose its projection, prepare its DEM, find one or several reference images (mostly LANDSAT 8 ) without clouds and merge them. Multiply that by more than 100 !
  • studies are starting to define the ESA-CNES SPOT5 (Take5)  distribution server

The first SPOT5 (Take5) picture will be taken only two months from now, and we have a lot of work ahead !



Deadlines / Echéances

Please be aware of 3 deadlines in February :



Attention ! Attention ! Plusieurs échéances approchent :

  • la date limite pour répondre à l'appel à communications pour le colloque Multitemp 2015 à Annecy est le 27 février. J'y organise une session sur les expériences SPOT (Take5).

All LANDSAT 8 data acquired over France in 2014 are now available at Level 2A


My colleagues from the THEIA production center at CNES just processed the last Level 2A LANDSAT 8 products acquired in 2014 over France. Weather was quite good during the last days of 2014 and a large number of cloud free pixels are available.


To download them  :

To access the documentation :



THEIA now distributes all the LANDSAT 5 and LANDSAT 8 data acquires over France since 2009, after atmospheric correction, and with a good cloud mask.


Happy 20th birthday, CESBIO !

Le CESBIO vient juste de fêter son vingtième anniversaire ! Pour notre grande fiesta, Simon Gascoin a édité de beaux posters artistiques, dont l'un est basé sur la série SPOT4 (Take5) du Paraguay, qui illustre ce blog. Nous vous l'offrons pour impression...

CESBIO just turned 20 ! For our big fiesta, our colleague Simon Gascoin edited very nice artistic posters, one of which was based on the SPOT4 (Take5) series above Paraguay. If you want to print it...