Satellite detection by satellite



Airplanes largely disrupt our remote sensing images, because of the ice contrails they leave behind them, which often turn into cloud cover. We had to set up a method for detecting and correcting aircraft contrails.

LANDSAT 8 image acquired over Paris on 14/04/2013. On the left, RGB color composition, on the right, image of the 1.38μm band. Given the number of traces of planes,  we might have to choose between flying or observing the earth.

But a new nuisance is appearing: the satellites themselves. More than 4000 satellites orbit around the earth, and with the nanosatellites mode, launches of space objects have multiplied. 450 new objects appeared last year, more than 500 are expected in 2019. As most of these satellites are launched in low orbit, between 400 and 600 km altitude, they orbit between our favorite observation satellites and the Earth.


And the future is quite worrying (generally speaking, the future is more worrying than the past): according to my colleague from CNES, Christophe Bonnal: "The US company One Web has the ambition to deploy 600 satellites within three years three to offer broadband internet access from space. Several companies have similar projects in drawers : Boeing has announced the sending of 2400 satellites, Samsung sits at 4000, while Elon Musk speaks bluntly of 12,000 spacecrafts ".


The 12,000 satellites in the Starlink constellation would be located at 3 different altitudes (340 km, 550 km and 1,200 km). Two of these altitudes will therefore be visible from the Sentinel-2 orbit. And already, the company Planet has about 200 satellites at an altitude of 400 km.


Given the large number of satellites, I wondered if it was possible that the images of Sentinel-2 were disturbed by the presence of satellites located a little lower. The possibility is quite high, because finally, most optical observation satellites seek to make their observations around 10:30 in the morning. With a good orbit propagator, and thanks to Norad's data, it's pretty easy to find the moments when one of the Sentinel-2 passes over one of the Planet satellites just below. And with that information, accurate to a few tenths of meters, we can search for the satellite in the image


Here are three examples of results obtained, two on very recent images, and one older. Although the planet satellites are quite small, their metal surface reflects the sun well and therefore leaves a visible mark on Sentinel-2 images.


Sentinel-2 image from March 28th, over China. The satellite can be seen within the red circle. See the zoom on the image on the right. The bright point, just right of the image center is satellite Planet Flock 1C-11,

Sentinel-2 image from March 27th, over Spain. The satellite can be seen within the red circle. See the zoom on the image on the right.

The bright point, just right of the image center is satellite Planet Flock 3R-8. On both images, my computation predicted it would be in the image center. There must be a bias.

This interesting case was observed just after Flock 3P launch with PSLV on the 12 the of January 2018. Three satellites, whose obits a re still quite close, can be seen just left of the image center.

(Sentinel-2 image observed over New Caledonia, on 13th Janury 2018.)


Artist view of Flock 3P launch from PSLV on January 12th.


Well, the phenomenon is still modest, thanks to the small size and low orbit of Planet satellites, but if large constellations are launched at an altitude closer to that of Sentinel-2, they can cover many pixels.Will we have to resort to the technique recently developed by the indian government to  avoid multiplying white spots on our images ?


And of course, this text was published on the first of April, and the "satellites" shown in the images are just white spots, probably not satellites, but who knows :)


SPOT5 est sur l'orbite Take5/ SPOT5 is on Take5 orbit


Tout s'est très bien passé ce matin, et SPOT5 a atteint sa nouvelle orbite, 2,5 km plus bas. Félicitations et un grand merci à toute l'équipe des opérations du CNES ! Les premières images seront acquises la semaine prochaine. Ne quittez pas !


On m'a de nouveau demandé comment un petit changement d'altitude de 2 km pouvait produire un changement aussi important de cycle orbital (de 26 à 5 jours). C'est expliqué ici.


Everything went well this morning, and SPOT5 is now on its new orbit, 2.5 km lower. Congratulations and many thanks to CNES operations teams ! The first images will be acquired next week. Stay tuned !


I had that question again : why does an altitude change by only 2 kms result in an orbit cycle change from 26 days to 5 days ? It is explained here.


Take5 goes to the movies


How to go from 1 image every 5 days to 24 images per second ?

It was possible, thanks to CNES funding, thanks to an imaginative producer, Gérard Dedieu (who does not smoke cigars yet), thanks to a talented film director and scenarist, Thierry Gentet (the only film director who understands space mechanics), and thanks to his team, Mira Production, who are even able to shoot beautiful images in our  ... splendid CESBIO offices, and thanks to a series of promising actors and actresses Anne Jacquin, Valérie Demarez, Virginie Lafon, Valery Gond, Jean-Pierre Dedieu, and another one, the last one, who cannot say a full sentence before the 5th take.


We hope this little film will help you understand or explain the possibilites and opportunities offered by multi-temporal images at a high resolution, and that it will give you ideas to use the new SPOT5 (Take5) data.


Phased orbits, how do they work ?


As we are working to set a new Take5 experiment with SPOT5, here are some explanations of how it is possible to change the repeat cycle of a satellite from 26 days to 5 days, by just changing the satellite altitude by a couple of kilometres. There is nothing complicated behind that, just some simple arithmetic.

A phased orbit is an orbit for which the satellite repeats the same trajectory periodically. From its orbit at an altitude of 822 km, SPOT5, like its predecessors, has a cycle of 26 days. Every 26 days, it overflies the same places on earth. In 26 days, SPOT5 makes 369 revolutions around the earth. In 24 hours, a SPOT satellite runs through 369/26=14.19 orbits. Lowering its altitude by 2 km, the satellite slows a little, but the length of the circle it has to run along is reduced. It takes a little less time to make a revolution around the earth. The satellite does exactly 14.2 orbits per day.


Here are some of the orbits of SPOT4 (Take5), with some of the sites observed in France and North Africa during the experiment. The satellite started with the Cyan track, then the green one on the day after, then the yellow one on the next day and so on. 5 days later, it came back to the cyan orbit. You may see that it was possible to acquire a site on on the green track from the adjacent one on the cyan track.


14.2 orbits per days, is equivalent to 71 orbits in 5 days. After 71 orbits and 5 days exactly  SPOT4 was always at the same place during the Take5 experiment, and its cycled was changed from 26 to 5 days.


I have been also asked how the initial 26 days repeat cycle of SPOT5 was defined. The CNES engineers who designed it wanted to make it possible to observe each point on the earth from the vertical. As the SPOT satellites had a field of view of 116 km using both instruments, with a 26 days repeat cycle we had 116x26x14.19 = 43000km, just a little more than earth equator length. However, it was quickly seen that users did not ask for exactly vertical images and that the instruments were programmed mostly independently looking in different directions. However, the 26 days cycle was kept for all the SPOT satellites just as the High Speed Trains rail separation is related to the width of the hindquarters of a horse.

Finally, nothing would prevent from using the SPOT satellites from a 5 days repeat cycle orbit, which would not really change the ability to use the images how they are used now, but would allow new possibilities thanks to the possibility to observe users from constant viewing angles.


It is a little funny to observe that SPOT6 and SPOT7 do not use the initial SPOT orbit, and only fly at an altitude of 694 km but still with a 26 days phased orbit, this time obtained with 379/26=14.58 orbits per day. However, the justification cannot be the field of view, as this field of view is only 60 km.  But just by rising the orbit by a few kilometers, a 5 days orbit could be obtained




SPOT4 (Take5) : what's next ?

(Version Française)
Orbit Change

Tomorrow, January the 29th at night, the altitude of SPOT4 will be lowered by 2.5 km. This operation is not far from the regular manoeuvres of orbit control and should not be risky. During the night, the orbit parameters will be measured and checked by our colleagues of the Operations Sub Directorate at CNES (DCT/OP).

Programming first acquisitions

Image programming schedule will be uploaded on the satellite on the afternoon of January the 30th by our CNES DCT/OP colleagues. If everything works well, first images will be acquired on Wednesday, January 31st (Day 5 in the 5 days cycle). The images will be recorded on-board and downloaded on Toulouse receiving station on the2nd or 3rd of February, along with all the images collected in between.

Data Inventory

Astrium Geo (ex-Spot-Image) will upload the images in its internal catalogue (4th or 5th of February). We will be able to check them then.

Image production

First level 1A should be produced shortly afterwards by Astrium.  For the MUSCATE Production Center Teams at CNES and CESBIO, it will be the start of the final integration of level 1C and level 2A processors. As Take 5 experiment was only decided on the 11th of December,  the integration of the elementary processors (ortho-rectification, calibration, cloud detection, atmospheric correction), and their fine tuning will take a while.

Stay Tuned !

Quand est-ce qu'il passe, SPOT4 ?

(english version)

(mise à jour du 29 janvier 2013, Vous pouvez aussi consulter le calendrier des observations.)

Quel jour ?

Sur chacun des 42 sites de l'expérience Take5, SPOT4 fera une acquisition tous les 5 jours. La manœuvre de changement d'orbite aura lieu le 29 janvier et dès les premiers jours de février, probablement même le 1er, les premières données seront acquises. Plusieurs utilisateurs ont déjà demandé à connaître le jour de passage afin de coordonner les acquisitions sur le terrain, simultanément au passage du satellite.

Pour savoir quel jour du votre site sera acquis, téléchargez le fichier kmz ci-joint, chargez le dans Google Earth et cliquez sur l'emprise du site qui vous intéresse. Vous pourrez lire le jour du cycle et le pas miroir de SPOT4 utilisés. Le pas miroir vous permettra de prédire l'heure de passage.


De gauche, à droite, les orbites des jours 1 à 5 du cycle, et les sites observés avec le même code de couleur.

Sachant que le jour 1 du premier cycle de Take5 aura lieu le 31 janvier, on peut en déduire le jour de passage du satellite :

  • Si le site qui vous intéresse est programmé le 1er jour du Cycle, les passages auront donc lieu, , le 31 janvier, en février le 5, le 10, le 15, le 20, le 25, en mars, le 2, le 7...
  • Si c'est le 3eme jour du cycle, ce sera les 2, 7, 12, 17, 22, 27 février, le 4 mars, le 9 mars...


A quelle heure ?

L'heure de passage est un peu plus compliquée à calculer, car l'inclinaison de l'orbite de SPOT4 n'est plus maintenue depuis quelques années pour économiser des ergols et prolonger la durée de vie du satellite. Cela se traduit par une dérive de l'heure de passage,  qui devient de plus en plus matinale.:

  • En février, le satellite passe à 9h25 TU à l'équateur
  • En juin, le satellite passera vers 9h10 TU à l'équateur.
  • Dans les deux cas, à 45 degrés de latitude Nord, il faut encore enlever 12 minutes.

Cette heure de passage est valable si votre site est sous la trace :

  • si votre site est observé depuis l'Ouest (pas miroir supérieur à 46), rajouter quelques minutes : 15 minutes si le site est observé avec un pas miroir proche de 91 (avec un angle de 27 degrés)
  • Si votre site est observé depuis l'Est sous un angle de 27 degrés (pas miroir inférieur à 46), il faut soustraire quelques minutes (15 minutes pour un pas miroir proche de 1, avec un angle de 27 degrés depuis l'Est).

Le CNES a essayé (Merci Frédéric), autant que possible, dans la programmation, d'acquérir les sites depuis l'Est, pour retarder l'heure de passage et avoir un soleil plus élevé.

En fait, si vous avez vraiment besoin de connaître l'heure exacte, le plus simple sera de nous demander l'heure de passage des premières images acquises sur votre site, et d'appliquer une dérive linéaire de l'heure de passage de 15 minutes en 4 mois.

When will SPOT4 observe my site ?

(version Francaise)

(updated version on January 29th, you may also have a look at the observation calendar)

What date ?

SPOT4 will observe each of its 42 sites every 5th day. The orbit change will be done January 29th and on one of the first days of February, perhaps even the first, the Take5 data acquisition will start. Several users have already asked about the date of acquisitions to schedule ground measurements simultaneously to the satellite overpass.

To know on which day of the cycle your site will be observed, download this kmz file and open it with google-earth. Click on the footprint of the site you are interested in. You may read the value of the day number in the 5 days cycle, as well as the value of the mirror step.

From left to right, above France, orbits of days 1 to 5, and sites observed with the same colour code

Knowing that the day 1 of the first Take5 cycle will be January the 31st, you can easily compute the overpass date :

  • If you are interested in a site observed on the first day of the cycle, the observations will therefore take place in January, the 31st, in February,  the 5, 10, 15, 20, 25, March, 2, 7 ...
  • If the site is observed on the 3rd day of the cycle, it will be, in February, 2, 7, 12, 17, 22, 27, March 4, March 9 ...

What time ?

The overpass time is a little more complicated to calculate since the inclination of SPOT4'orbit is no longer maintained to save propellants and increase the satellite lifetime. This causes a drift of overpass times, towards earlier overpasses.

  • In February, the satellite passes the equator at 9:25 UTC,
  • In June, the satellite will pass the equator at 9:10 UTC.
  • In both cases, at 45° North latitude, it will be 12 minutes earlier.

This overpass time is valid if your site is below the satellite track.

  • If your site is viewed from the West (mirror step greater than 46), it will be observed a few minutes later. For instance, add 15 minutes if the site is observed at an angle of 27 degrees (mirror step close to 91).
  • If your site is viewed from the East (mirror step lower than 46), it will be observed a few minutes earlier. For instance, subtract 15 minutes at an angle of 25 degrees (mirror step close to 1)

We have tried, whenever possible to program site acquisitions from the West, to have higher sun elevations.

In fact, if you really need to know the exact overpass time, the easiest way is to ask the overpass time of the first images acquired on your site, and apply a linear drift of 15 minutes in 4 months.

Changing SPOT4 orbit : easy ?

(French Version)

When we submitted the SPOT4(Take 5) experiment to CNES, we knew that CNES would not accept it easily, since a similar proposal made by Gérard Dedieu before SPOT2 de-orbitation had been rejected. But we did not imagine the amount of work we were requesting from our colleagues at CNES. To show the project feasability, our CNES colleagues had to :

  • find a project manager who coordinated the study: Sylvia Sylvander
  • choose the new orbit (2 to 6 days repeat cycle) minimizing fuel consumption : We must keep enough fuel to be able to reduce the altitude of the satellite, so that it burns in the atmosphere after a period of less than 25 years. The finally chosen orbit provides a 5 days repeat cycle, resulting in a very low fuel consumption. This orbit also provides the exact repeatability of the two satellites Sentinel-2.
  • choose the strategy change orbit. The date of optimal maneuver is January 29, but it corresponds to the end of a full moon, and it is prohibited to maneuver SPOT4 during the full moon. It is not due to superstition, but only because of a potential blinding of star sensors used to determine the orientation of the satellite. But a detailed analysis of recent full moons in the same period last year showed that the maneuver could still be executed on January 29 without any risk.
  • check that the ground segment (designed 15 years ago) can handle the new orbit. The ground segment programs the satellite and the acquisitions, manages the old tape recorders, coordinates the data download to the receiving station, while avoiding interference with other satellites. Because the satellite is no longer on its nominal orbit, all the conditions of interference are to be recalculated.
  • check that the ground segment is able to ingest and process products. The products are usually referenced by their orbit number, which will be different...
  • test the system interfaces: a one-week trial on a simulator of the spacecraft and its system showed that everything should work fine
  • the usual programming system at SpotImage will not work on this orbit, we will have to use CNES programming system, more flexible but less automated. It takes one hour and a half to program the 42 sites observed over 5 days, but someone will have to do it every 5 days.
  • find internal and external staff (and budgets) to extend the life of SPOT4 for 5 months.
  • negotiate with SpotImage (Astrium Geo), the cost of producing Level 1A products and maintaining the cloud notation
  • prepare the MUSCATE production center that will provide level 1C and 2A users. This production facility will be implemented within the Land Data Centre.

Many thanks to Didier Roumiguières, Sylvia Sylvander, Laurence Houpert, Jean-Marc Walter, Jordane Sarda (CS-SI), Aurélie Moussy-Soffys, Frédéric Daniaud (CS-SI), Michel Moulin, Benoît Boissin,  Selma Cherchali, Françoise Schiavon, Marc Leroy, Jerôme Bijac (Astrium geo) and to all CNES and Astrium-Geo people who contributed to the acceptation SPOT4(Take5) experiment.