A new version of the SPOT4(Take5) products is available.

Here are the thumbnails from the China(2) site, for which several dates were missing on the version 1.0. Please note that on the server, you may download all the dates at once by clicking on the 1C or 2A buttons.

=>

The CNES teams of the THEIA Land Data Center have reprocessed the SPOT4 (Take5) data, in order to take into account a large number of images that were not processed in the first place, because some data had not been yet received or because their processing had failed due to a few little bugs.

 

The same processors and parameters were used and the only difference is the increased number of available dates, but as the L2A methods are multi-temporal and recurrent, when we add an image, the results on the subsequent images are also changed. It is thus advisable that you download again all the products of the sites you are interested in, from the following address : http://spirit.cnes.fr/take5

 

On this prototype ground segment, our management of product versions is basic, and only takes the processors into account. As the processors are unchanged, the new version 1.1 products are still identified as level 1.0 products in the Metadata. We are sorry for this inconvenience, you will need to pay attention not to mix them with the older version.

 

A few missing images

=>

I just took a work break in the middle of my holidays, but as I was away, we received a few feedbacks from users, and CNES PTSC teams, with Mireille's help at CESBIO verified the data sets released on July the 16th, in quite a rush...

 

They found out that a few scenes were missing. For some of them, it was due to the late arrival of some images (just as for planes at the airport). These images have already been added to the server.

And there were a couple of bugs that mostly affected the sites made of several SPOT images (CNES and NASA sites), and ESA Chinese site. These glitches have been corrected and the reprocessing started. The whole data set will be updated before end of August, which will constitute the version 1.1 of the SPOT4(Take5) data set.

 

Keep posted on this blog, we will update it as soon as the data are available. Meanwhile, version 1 is still accessible here, and the format described there.

 

The adjacency effects, how they work.

As explained in the post about atmospheric effects, the scattering of light by molecules and aerosols in the atmosphere brings about several effects : scattering adds some haze on the images (the atmospheric reflectance), lessens the signal from the surface (the atmospheric transmission), and blurs the images (the adjacency effects). This post is about the adjacency effects, the other aspects have already been quickly explained in the above post.

 

The figure on the right shows the types of paths that light can follow before getting to the satellite. Path 1 corresponds to the atmospheric reflectance, path 2 is path that interacts with the target, it is the one which is useful to determine the surface reflectance, paths 3 and 4 contribute to the total reflectance but interact with the surface away from the target. These paths are thus the cause of adjacency effects and they blur the images.

 

 

If not corrected, adjacency effects may cause large errors. Let's take the case of a fully developed irrigated field surrounded by bare soil. For such a case, the second figure on the right shows the relative percentage of errors for reflectances and NDVI as a function of aerosol optical thickness, if adjacency effect is not corrected.

 

 

 

An approximate correction can be applied, but it thus requires to know the aerosol optical thickness. In our MACCS processor, here is how it works :

 

  1. We first correct the images under the assumption that the Landscape is uniform. We obtain a surface reflectance under uniform absorption which is noted  \rho_{s,unif} .
  2. We compute the neighbourhood reflectance (  \rho_{s,adj} ) using a convolution filter with a 2km radius, that computes the average neighborhood reflectance weighted by the distance to the target. To be fully rigorous, this filter should depend on the optical thickness and on the viewing and sun angle (The less aerosols, the larger radius), but as we did not work on an accurate model, we used a constant radius.
  3. We correct for the contribution of paths 3 and 4 using :

 \rho_{s}=\frac{\rho_{s,unif}.T^{\uparrow}.\frac{1-\rho_{s,unif}.s}{1-\rho_{s,adj}.s}-\rho_{s,adj}. T_{dif}^{\uparrow}}{T_{dir}^{\uparrow}}

  • where  T^{\uparrow}=T_{dif}^{\uparrow}+T_{dir}^{\uparrow} is the total upward transmission, sum of diffuse and direct upward transmissions, and s is the atmosphere spheric albedo. These quantities depend on the wavelength, on the aerosol model and on the AOT. They are computed using Look up Tables based on radiative transfer calculations.

 

As this processing uses convolution with a large radius, it takes quite a large part of the atmospheric processing time.

 

Result Exemples

The images below show 3 stages of the atmospheric processing, for 2 Formosat-2 images obtained over Montreal (Canada) with a 2 days interval. The first image was acquired on a hazy day (aerosol optical thickness (AOT) of 0.47 according to MACCS estimate); and the second one on a clear day (AOT=0.1).

  • The first line corresponds to the Top Of Atmosphere images, without atmospheric correction. The left image is obviously blurred compared to the right image.
  • The second line corresponds to the atmospheric correction under uniform landscape assumption (as in step 1). The left image is still obviously blurred compared to the right image.
  • the third line show the same images after adjacency effect correction. In that case, the left image is not blurred any more, it is even maybe a little over corrected as it seems somewhat sharper that the right image.

TOA Images (On the left, the hazy image)


Surface reflectance under uniform landscape assumption (on the left, the hazy image)

 

Surface reflectance after adjacency effect correction (on the left, the hazy image)

 

The pixel wise comparison of reflectances is also a way to show the enhancement due to the adjacency effect correction. The plot below compares the images of both dates corrected under the uniform landscape assumption (on the left), and after adjacency effect correction (on the right). You may observe that the dots are closer the the black diagonal on the right. On the hazy image (May 27th), the high reflectances are a little too low, while the low reflectances are a little too high, which is the symptom of a loss of contrast.

Release of the first version of SPOT4 (Take5) data

 

Phew ! These last days were quite intense, but we did it (almost) ! We are very proud to announce that the first version of SPOT4 (Take5) will be released on Monday the 15th of July at the following site : https://www.ptsc.fr/en/products/spot4-take5. The data format is explained here.


It will be possible to download the whole time series by clicking on 1C and 2A buttons. However downloading a whole series might take some time. Several small issues have been observed but not yet corrected (see this paragraph), but we still think it is useful that you start working on our data despite the few bugs that affect them. Downloading a data set requires the user to accept a licence. With the current version of the web site, you will need to accept this license each time you download a file.

In our initial schedule, we had announced a distribution of data during the month of June, but we are just a bit late, and happy to have done it, accounting for the numerous tasks we had to fulfil.

 

  • We had to obtain the experiment decision, which was done by the 11th of December 2012, and we had to implement it. The last data were acquired less than one month ago, and they will be available with the whole time series.
  • We had to modify and tune the processing chains, and the geometry  kept us busy for several weeks (and still does over equatorial forests). The level 2A processor worked quite well from the beginning, thanks to Mireille Huc (CESBIO) !
  • It was the first production of the French Land Data Center. The ground segment was built and implemented simultaneously to the data acquisition, its development team did a very nice work (namely, Dominique Clesse from CAP GEMINI, Hassan Makhmara and Joelle Donadieu at CNES). The data processing in this changing environment was perfectly managed by the exploitation team, Nicolas Prugent, Karl Rodriguez (Steria), Eric Faucher (CNES)...
  • A simple but very nice data server was set up in a very short time by Jerome Gasperi and Bernard Specht (CNES). The result is very simple and convenient (it helped us finding the last bugs...)

Issues and bugs

The products on the data server should be considered as a preliminary version of the processing. Other versions will be distributed, because the data sets still have a few defects :

 

  • Because of a little bug, the overlapping sites (Midi-Pyrénées East and West, and Britanny) were badly processed. These sites will be processed again and released in about 10 days. For the same reason, Maricopa site (USA) could not be processed, it will be released probably in September.
  • One image at the south of Languedoc site is always missing. We do not know why yet, but there is no doubt it will be corrected shortly.
  • Some Level 1A  products were not provided by Spot Image (25 out of 1600). We will need to reprocess the affected sites.
  • For all sites, the reference image for ortho-rectification were taken from LANDSAT  5 or 7. From our point of view, their location accuracy is a little insufficient. On the next version, we will replace LANDSAT data in FRANCE by the GeoSud high resolution cover from 2009. Out of France, we will replace them by LANDSAT 8 data whenever possible.
  • Data ortho-rectification often fails on some uniform equatorial forest sites (in particular for Congo(2), Gabon, Borneo, Sumatra). As  a result several dates are missing and those available are not perfectly registered.
  • On the level 2A, atmospheric corrections were performed with a constant aerosol model globally. We will use different aerosol models depending on the location in a future version.

 

Finally, we want to draw your attention to the fact that SPOT4 data are coded on 8 bits and can be saturated. SPOT system uses a data base of histograms to determine the gains to use depending on the location and date. This database is far from perfect and saturated data are quite common. For sites resulting from the merger of several images, it is also possible that the saturation thresholds differ between the left and the right half of the site. We provide you, within Level 1C and level 2A products with a mask saturations that you really should use.

 

Despite these few defects, we wish that this data set will be useful for you, and that you will obtain good results for your experiment and for your preparation for the arrival of Sentinel-2.

 

Many of the SPOT4(Take5) team will leave for vacations in the coming weeks, this blog will b much less active as I leave today !

 

 

 

 

 

L1C registration performances for SPOT4(Take5) V1 products

Now that all SPOT4(Take5) images have been processed (pfew !), we can make an appraisal of the performances. Let's start by the geometry, which caused us a lot of trouble :

  • SPOT4 has a location accuracy around 400 mètres, but during the experiment, it went through a fifteen day period when the location errors could reach 1500 m.
  • We seek a multi-temporal registration performance of 0.3 pixel RMS. This performance is difficult to measure because the measurement technique itself (correlation image matching) is not perfectly accurate.
  • We provide as a criterion the maximum registration error observed for the 50% best results or for the 80% best results. It is likely that the last criterion includes less inaccurate measurements.

 

Here are the observed performances for 3 very different sites :

  • CMaroc site, which is an arid site with a green period in march, a lot of blue sky, and high mountains (the Atlas). performances are excellent, with errors lower than 0.3 pixels for 50% of the measurementsl.

 

  • CBretagneLoireE site, which is a rather flat coastal area with large tides, and is often very cloudy. In that case, performances are still better than 0.5 pixels. The worse dates correspond to images with a large cloud cover, for which it is not easy to automatically collect accurate ground control points.

 

  • JSumatra site is a very flat area, covered with very uniform equatorial forest, and a large river whose limits change with time. In that case, the performance is really bad, with registration errors up to 10 pixels. This uniform site does not enable to find good control points, and the ones that are found are often along the river whose contour changes with the water level.

 

Conclusions

We have obtained very good results for most sites, with registration errors below 0.5 pixels (10m) even when the initial location error reaches 1500m. However, 4 sites are resisting to this processing. These 4 sites correspond to flat forest sites covered by equatorial forest : JSumatra, JBorneo, EGabon, ECongo. The ECongo site is even so uniform that it is not possible to measure its registration performance.

These sites will be distributed with the others in a few days with the first version of the products, but you should use them cautiously.

Finally, if the registration of 95% of images is good, the location performance is inherited from our reference images, ie LANDSAT (5 et 7). The next versions will be based on Geosud (IGN) images in France and on LANDSAT 8 data elsewhere. Performances should be enhanced in the next versions.

All the quicklooks

Before the distribution starts at the land data center (before mid July, or even sooner), I have updated all the quicklooks of all the images taken during the experiment. I have checked in the catalog to see if a few images had not been forgotten. I found about 20 images (on a total of 1600). These images will be processed soon.

 

You may find all the SPOT4 (Take5)  quicklooks following the links below, or via the SPOT4 (Take5) menu.

SPOT4 acquired its last images

=>

The last imaged acquired by SPOT4 were taken on the 19th of June, in the framework of Take5 experiment. It is with some sadness that we see the end of this very intense experiment : SPOT4 de-orbitation started and the satellite will be switched off on June the 29th 2013. SPOT4 will burn in the atmosphere in a few decades.

 

First and last SPOT4 (Take5) images acquired over the Alps site, on February the 8th (left) and June the 13th (right).

 

I would like to thank once again :

 

  • Sylvia and CNES teams who contributed to the feasibility studies, to the budget negociations, to the experiment decision and to establishing the contracts and the license for use.
  • Frederic and CNES and CS-SI teams who changed the satellite orbit and then programmed it and monitored its health during its extended life.
  • Laetitia and Joel who helped us programming the gains and checking the acquisitions
  • Mickey and Astrium Geo teams (SPOT IMAGE), who downloaded, selected and processed to level 1A about 1600 images (those which had at least a small patch of clear sky) and Bruno, for his patience regarding the long, divers and complex procedures of the space agencies regarding the contracts.
  • The data users who proposed the numerous appplications of SPOT4 (Take5) data, and our partners, ESA, JRC, NASA and CCRS, whose support was essential to get the experiment decided.
  • Jean-François, and the CESBIO colleagues who kept going in the fields to collect in-situ data, and those who supported my complaints regarding the rgnenjbfhzerj weather we had in France this spring. Weather is still cloudy and cold, but now I have stopped complaining.
  • CNES and Thales geometry specialists who helped me tuning the ortho-rectification parameters
  • The PTSC development and processing teams (CNES, CAP GEMINI, STERIA), who are still working on level 1C and 2A production and distribution : the processing is well advanced, and nearly 80% of the acquired data have been processed
  • the many readers of this blog, whose regular visits cheer us up (6400 visits et 3500 different visitors, 15000 pages viewed)
  • and Mireille, who is carefully and efficiently updating and improving the Level 2 processor on which our work,is based.

This blog will go on after the end of acquisitions, we will publish here the results obtained with SPOT4 (Take5) data, and we will provide news about PTSC, Venµs, LANDSAT and Sentinel-2.

 

And also, SPOT5 will be de-orbited in a few years...

Regardez pousser les plantes/ See the plants grow

Cette série d'images a été acquise au Paraguay du 10 mars au 09 mai. On  y voit clairement le démarrage des cultures, passant du sol nu (rose) au plein développement (vert vif).

This image time series was acquired in Paraguay, from the 10th of March to the 9th of May. The start of growing season for several crops may be seen, from pink (bare soil), to bright green (when the crop if fully developped and still green).

SPOT4 (Take5) last cycle

=>

The SPOT4 (Take5) experiment begins its last 5 days cycle tomorrow (15th of June). The last images will be captured on June the 19th : we wish a nice weather for all the sites ! On June 19th, the acquisition phase of SPOT4 (Take5) will end, but the application phase will start soon. SPOT4 satellite will start its de-orbitation phase on the 29th of June.

The quicklooks of the images acquired until the end of May have been updated. You may find them following the links below, or via the SPOT4 (Take5) menu.

See the snow melting (...or not)

=>

The snow season is ending in the high Atlas mountains in Morocco this beginning of June, as you may see in this time series from SPOT-4 (Take5) experiment over the Rheraya watershed (225 km²). This watershed is one of the study sites of the Laboratoire Mixte International TREMA which CESBIO is co-leading. The Rheraya wadi provides large water resources to the populations in the arid zones downstream. The high frequency of SPOT4 (Take5) revisits and the scarce cloud cover enabled to capture the back and forths of snow cover between January the 31st and May the 26th. Continue reading