SPOT4(Take5) aerosol optical thickness validation results

We are currently preparing a data reprocessing of all SPOT4 (Take5) data, to be released before the end of 2013. For this, I tested several aerosol models and compiled all the validation results for our multi-temporal Aerosol Optical Thickness (AOT) estimation method named MACCS. Our estimates are compared to AERONET in-situ AOT measurements.

The MACCS method applied to SPOT4(Take5) data, which lacks a blue band, uses two procedures to estimate AOT :

  • either the AOT is estimated by a multi-temporal method
  • or it is gap-filled. The presence of gaps may be due to clouds, water or snow, or because the pixel reflectance is too-high for an accurate estmate, or because of a too large variation of reflectance with time is detected.


Comparison of MACCS AOT estimates with the in-situ measurements from AERONET. The blue dots correspond to cases for which the atmosphere is stable and for which there are no clouds in the neighborhood of the AERONET site. The red dots correspond to situations when the AERONET optical thickness varies around the satellite overpass time, or when clouds are detected in the image neighbourhood (20*20 km).
On the left plot, only the dates and sites for which less than 60% of the pixels were gap-filled; wheras the right plot only tolerates 20% of gap-filled pixels. The gap-filling method does not seem to introduce large amount of errors in wases when the atmosphere is stable, but it is less accurate in unstable cases..


The aerosol estimates have been obtained with MACCS prototype which is developed and maintained by Mireille Huc at CESBIO. The aerosol model is not the same as the one used for SPOT4 (Take5) first processing. This model is based on greater particles (with a modal radius of 0.2µm, compared to 0.1µm in the initial processing), as it provides a better overall agreement with AERONET measurements. We will use this model for most sites for SPOT4(Take5) reprocessing.


The RMS error of AOT estimates is 0.06, which is a state of the art performance, obtained in a very difficult condition with no blue band available. Moreover, in order to show more validation points, a few validation sites (Bruxelles, Gwangju, Ouarzazate, Wallops, NASA_LaRC) are in fact distant by more that 60 kilometers from the image footprint, which tends to degrade the performances.


The AERONET sites used in this study are :


SPOT4 Take5 Site
Aeronet Site
Belgium Brussels
South Great Plains Cart_Site
Korea Gwangju_GIST
Chesapeake NASA_LaRC
Chesapeake Wallops
Versailles Paris
Versailles Palaiseau
Tunisia Ben Salem
Maroc Saada
Maroc Ouarzazate
Sudmipy-Est Seysses + Le Fauga
Sudmipy-Ouest Seysses
Provence Carpentras
Provence Frioul


The worst results are obtained for the following sites :
  • Gwangju (Korea): The SPOT footprint in on the coast, while the AERONET site is 70 km inland, near a large town.
  • Ben Salem (Tunisia): this site was very cloudy in Spring, and large reflectance variations are observed between the remaining clear dates.
  • Palaiseau and Paris : In that case, the aerosol model seems to be inappropriate, and absorbing pollution aerosol should be introduced.

On the contrary, several sites provide very accurate results, for instance in Morocco (even the desertic Ouarzazate), Provence (including the Frioul Island where the AOT is extrapolated from the coast), and also Sudmipy, Wallops et Cart_site. Some SPOT4 (Take5) users reported inaccuracies on some tropical sites but we do not have an AERONET validation site near these SPOT4(Take5) sites.


Characterising the phenology of tropical rain forests in North Congo thanks to image time series at low and high resolution

The CIRAD research institute studies the North Congo rain forests, and uses satellite image time series to tell mainly deciduous forests from mainly evergreen forests, The analysis or 10 years of a vegetation index time series (Enhanced Vegetation Index – EVI) from MODIS images at 500m resolution, was used to produce a synthetic annual profile that characterises the phenology of the observed forests (Gond et al., 2013), and enables to separate these two forests types. It seems that these variations in leaf phenology are partly related to the geology (Fayolle et al., 2012).


SPOT4(Take5) image from June 2nd: top, with the geological limits, sandstone in the West and silt in the west ; bottom, the vegetation classes obtained thanks to MODIS (mainly evergreen in the West and deciduous in the East. EVI Temporal profiles of both forest types (evergreen and mainly deciduous) of North Congo between January and June by sixteen day periods. The black dashes mark the six clear SPOT4 (Take5) dates, and the dots mark the partially clear dates.


The SPOT4(Take5) experiment, which prefigures the data delivered by ESA's Sentinel-2, enabled to obtain 6 very clear images between February and June 2013, which is exceptional in this usually cloudy region (the clouds were in Europe this year :-( ). These images will allow us to analyse temporal profiles of photosynthetic activity with far more details than with MODIS, which is useful to understand the behaviour of these forest types, and study in detail how the are related to the geology.


This high temporal frequency data set enabled us to asses the possibilities of monitoring human activity in these rain forests. The images below, from SPOT4(Take5) show how a new logging track is opened by a company. It shows the capabilities of using Sentinel-2 data to identify and detect human interventions n the most remote places.


4 color images (SWIR, NIR and Red) in North Congo for 4 dates from March 2 June. A new forest track is being opened southward on the 4th of March, and goes further South until the 14th of April. From this date the track turns East, to access various forest sites. On the 2nd of June, the track is finished and the logging starts, it is clearly visible with a zoom.


Fayolle, A. Engelbrecht, B. Freycon, V. Mortier, F. Swaine, M. Réjou-Méchain, M. Doucet, J.-L. Fauvet, N. Cornu, G. Gourlet-Fleury, S. 2012 Geological substrates shape tree species and trait distributions in African moist forests PLoS ONE 7, e42381

Gond, V., Fayolle, A., Pennec, A., Cornu, G., Mayaux, P., Camberlin, P., Doumenge, C., Fauvet, N., Gourlet-Fleury, S., 2013, Vegetation structure and greenness in Central Africa from MODIS multi-temporal data, Philosophical Transaction of the Royal Society (serie B), 368: 20120309

Gourlet-Fleury, S. Rossi, V. Réjou-Méchain, M. Freycon, V. Fayolle, A. Saint-André, L. Cornu, G. Gérard, J. Sarrailh, J.-M. Flores, O. Baya, F. Billand, A. Fauvet, N. Gally, M. Henry, M. Hubert, D. Pasquier, A. Picard, N. 2011 Environmental filtering of dense-wooded species controls above-ground biomass storerd in African moist forests J.Ecol. 99, 981-990.


SPOT4 (Take5) users's day presentations

Voici les présentations de la Journée SPOT4 (Take5) du 2 octobre, qui a réuni une centaine de personnes au CNES pour faire un bilan des acquisitions et des produits distribués, et prendre connaissance des premiers retours des utilisateurs. J'en profite pour remercier les deux organisatrices, Sylvia Sylvander et Danielle Barrère, du CNES (DCT/ME/OT), les 23 orateurs qui nous ont proposé de brillantes présentations, et le photographe (Gérard Dedieu) qui n'a oublié que 3 orateurs.


J'ai rédigé pour le CNES un bref compte rendu en Français. Pour voir les présentations, cliquez sur les liens dans le tableau ci dessous.


Here are the presentations from SPOT4 (Take5) users day on October the 2nd. 100 people attended this meeting at CNES, which aimed at making an assessment of the acquisition and products, and obtain a first feedback from users. I would like to thank a lot the two organizers, Sylvia Sylvander and Danielle Barrère from CNES, the 23 speakers, who delivered brilliant presentations, and the photographer (Gerard Dedieu), who only forgot 3 speakers.


The table below gives access to all the slides of the presentations.



O. Marsal

CNES Introduction
S. Sylvander CNES Preparation and progress of Take 5 experiment
O. Hagolle CNES/CESBIO Justification of Take 5 experiment, site selection and data access
M. Leroy CNES The THEIA Land Data Centre
O. Hagolle CNES/CESBIO First results
M. Claverie NASA GSFC Consistency of SPOT4 (Take 5) surface reflectance data: Comparison with MODIS surface reflectance data.
C. Szczypta CESBIO Application of remote sensing to snow modelling in the Pyrenees
JP. Dedieu LTHE/CNRS Grenoble Snow cover monitoring in the French Alps
B. Koetz ESA ESA preparatory activities for Sentinel-2 exploitation - Agriculture, Land Cover Change, Costal Monitoring & Forest Mapping
C. Corbane / F. Güttler UMR TETIS Contribution of remote sensing data with high repetitivity for the identification and monitoring of natural habitats - Application to Lower Aude Valley Natura 2000 site.
A. Govind INRA/Bordeaux High resolution mapping of LAI using SPOT-4 data for spatially explicit modeling of Carbon and Water Fluxes in the Landes de Gascogne
D. Jacques UCL Preparing for the exploitation of Sentinel-2 observations for agriculture monitoring
V. Gond CIRAD Phenological monitoring of tropical forest ecosystems (North of Congo)
E. Bartholome JRC First observations from SPOT 4 Take 5 data over intertropical regions

A. Jacquin / A. Roumiguié

EI Purpan Use of multitemporal series of high and medium spatial resolution for forest and biomass monitoring
D. Courault INRA Monitoring of the evolution of agri-hydrosystems in a mediterranean region
V. Lafon EPOC, Univ. Bordeaux Contribution of Sentinel-2 to coast management
S. Battiston SERTIT First results of Take Five experiments over the Alsatian Plain (France) and Chinese lakes in term of biodiversity, forestry and hydrology
M. Le Page CESBIO, UCAM, ORMVAH, INAT Preliminary results of a real time irrigation experiment in Morocco and first results in Tunisia
M. Battude CESBIO Contribution of optical multitemporal satellite imagery for the cartography of irrigated areas
Y. Justeau Exelis VIS Take 5 GeoProcessing with ENVI Services Engine
D. Giaccobo ASTRIUM-GEO Potential use of Take5 data for the ESA DREAM Data Quality Web Service

The Level 2A products lose weight

This discreet bug was spotted by Mireille Huc a couple of weeks ago, but it had been bothering all SPOT4(Take5) data users for2 months : the L2A products were twice larger than necessary. The bug occurred when converting data to GeoTIFF format,  it happened only on the production computer and not on the test computer. Unnecessary data were added to the files, which did not prevent us from reading the useful data normally.


The data have been corrected and loaded on the distribution server on October 10th. We advise you to download them again, they will take much less space on your disks.

The terrain effect correction : how it works


Caution, this post contains formulas.


The topographic (or terrain) effects on the observed reflectances are due to several phenomena, illustrated below :

  • the closer the surface is perpendicular to sun direction, the more energy it receives per surface unit (we talk about irradiance). If the surface is parallel to sun direction, it does not receive direct sunlight. We can model it this way :
    • For an horizontal surface :  E_h= E_0.T_{dir}^\downarrow.cos(\theta_s)

      Definition of sun zenith angle and sun incidence angle

    • For a sloped surface  E_i=E_0.T_{dir}^\downarrow.cos(\theta_i)
    •  E_0 is the Top of Atmosphere irradiance, and  T_{dir}^\downarrow is the downward direct transmission, i.e. the proportion of the light that reaches directly the surface without being scattered by the atmosphere.
    • Assuming that all the irradiance is direct, the measured reflectance if the surface was horizontal is calculated from the following formula:  \rho_h=\rho_i \frac{cos(\theta_s)}{cos(\theta_i)} . However, the above assumption is not true and this formula tends to over correct terrain effects
  • The surfaces also receive a diffuse sun irradiance scattered by the atmosphere. If the surface is not horizontal, a part of the sky is obscured by the slope reducing the diffuse irradiance. Moreover, the diffuse irradiance depends on the amount of aerosols (and clouds) in the atmosphere. In addition, the surrounding terrain can also hide a part of the sky, but we do not take this effect into account here in our modelling. We use the following approximation, which is equivalent to assuming that the slope is alone in a horizontal region.
    • If surface is horizontal, the visible sky fraction is 1, if it is vertical, this fraction is 1/2
    •  \displaystyle F_{sky}= \frac{1+cos(slope)}{2}
  • Finally, the slope can receive light from surrounding surfaces, which become directly visible. In our simplified model, we always assume the entire environment of our slope is flat and,for instance, we do not take the effect of the opposite side in a valley into account :
    • If surface is horizontal, the visible ground fraction is 0, if it is vertical, it is 1/2.
    •  \displaystyle F_{fround}= \frac{1-cos(slope)}{2}


Finally, we use the following formula  to compute the reflectance that would be observed if the surface was horizontal   \rho_{h} , as a function of the slope (inclined) reflectance   \rho_{i} :

  \rho_{h}=\displaystyle \rho_{i}.\frac{T^{\downarrow}}{T_{dir}^{\downarrow}.\frac{cos(\theta_i)}{cos(\theta_s)} + T_{dif}^{\downarrow} F_{sky} + T^{\downarrow} F_{ground} \rho_{env}}

 T^{\downarrow} is the downward transmission, sum of direct and diffuse irradiances :  T^{\downarrow}= T_{dir}^{\downarrow}+ T_{dif}^{\downarrow} , and \rho_{env} is the average reflectance of the neighbourhood.


Finally, we can also account for bidirectional reflectance effects, but this correction is tricky since directional effects depend on the surface cover type. See for instance : Dymond, J.R.; Shepherd, J.D. 1999: Correction of the topographic effect in remote sensing. IEEE Trans. Geosci. Remote Sens. 37(5): 2618-2620.


It is very difficult to validate the correction of directional effects : , we could compare the correction results for satellite overpasses at different times in the day. But all the satellite optical imagers have nearly the same overpass time. A qualitative way of estimating the accuracy is to check that similar land covers on opposite slopes in a valley ( a meadow, a forest) have a similar reflectance after correction. The most suitable points are North-South valleys. Here are some examples of terrain effects correction results.

Formosat-2 image in the Alps, before (left) and after (right) terrain effect correction. The image on the right seems to have been flattened, and the opposite slope reflectances seem much more alike after correction.

Finally, an essential part of the method's accuracy is the availability of a highly accurate digital  elevation model (DEM),  up to now, only the SRTM DEM is available globally, and it only has a 90 meter resolution. Its accuracy is somewhat inadequate and sometimes leaves artefacts if the slope changes are poorly located.



Just back from Living Planet Symposium

I had the opportunity to participate to the ESA Living Planet Symposium, in Edinburgh, last week, together with 1800 other remote sensing data user, mostly from Europe. It is incredible to see that after more than 20 years in this domain, I know less than 10% of the audience !

There were 10 parallel sessions each day, and in the biggest room, 2 sessions were organised for each Sentinel mission, and these sessions have been filmed. These films are on the web, accessible from this page. As a reader of this blog, you might be interested by the sessions about Sentinel-2, available here :

  • Sentinel-2 mission, to have a complete update about the mission and its ground segment (official launch dates : 09/2014 (S2-A) and 09/2015 (S2-B)).
  • Sentinel-2 potential applications and services, in which I presented the SPOT(Take5) experiment (the third presentation, from 46' to 1h06'). Please be lenient, I was a little nervous in front of 150-200 people !

I also wrote (quickly) a seven page paper about the SPOT4 (Take5) experiment.

Feedback on the irrigation scheduling experiment using remote sensing images


CESBIO contributes to an international joint laboratory in Morocco, called TREMA, "Télédétection et Ressources en Eau en Méditerranée semi-Aride", which means "Remote Sensing and Water Resources in Semi-Arid Mediterranean". This year, this laboratory has embarked on an ambitious experiment of irrigation scheduling by satellite imagery, on a wheat plot near Marrakech. This experiment was already described in March, and it gave very promising results.

The main objective of the experiment was to see if  the logistics of irrigation scheduling by water balance model were feasible in real conditions. For this, a farmer accepted to play our game on two four hectares plots of wheat: Irrigation of the reference plot was driven by the farmer in the usual way. The test plot irrigations was driven by our tool SAMIR (FAO-56 model forced by satellite imagery).


Since the sowing late December to the harvest in early June, a weather station installed on a reference culture has given us the daily reference evapo-transpiration measurements. On the other hand, to control a posteriori the quality of our estimates of water requirements for irrigation, two flux measuring stations were set up. We also acquired a series of images SPOT5 early in the season to compensate for the slightly late start of SPOT4 experience (TAKE5) which began in February.


In addition to a clear weather throughout the season, we were able to benefit from the excellent work of the SPOT4 (TAKE5) team which provided us with the georeferenced images very quickly. The NDVI evolutions were thus available in a relatively short time. As an end user, the Office of Agricultural Haouz allowed us to perform the irrigation of the test plot in the best conditions while being subjected to the constraints of the canal system.

On the ground, everything did not work as well as we planned. Following a misunderstanding with the farmer, we completely missed the second irrigation and the fertilizer application was not timely. Indeed, the study plot is installed on a heavy clay soil that forms a crust. We were not aware that, a few days after sowing, a specific irrigation is needed to ease the emergence of plants. On the other hand, the farmer applied nitrogen fertilizer on two plots just after irrigation of the reference parcel and relatively far from the irrigation of the test plot. Under these conditions the nitrogen is relatively less soluble, and our test plot lacked fertilizers.

Our experiment has been seriously hampered by the misunderstandings with the farmer. But despite the bad start, the experiment was pursued to its end.


This plot shows the changes throughout the course of the experiment of the water supply from rainfall and irrigation, the evapo-transpiration ETobs measured in the field and the Evapo-Transpiration ET estimated by SAMIR model, using the vegetation status from SPOT4 (Take5) images. On this plot, the dates of irrigation were suggested by the model.


To our surprise, the results are extremely promising. Indeed, despite a 20% lower biomass compared to the plot driven by the farmer, we got a equivalent performance in grain yield. This can be explained by the fact that, although the average number of wheat blades was much lower on the test plot, it is very likely that the reference plot, irrigated by the traditional method, has suffered water stress in late March limiting the filling of grain.



This full-scale experiment finally turned out to be very instructive. First,  imaging/weather/irrigation logistics worked great : the weather data transmission, the reception and the geometric and radiometric correction of images, the model runs and  irrigation decision were largely automated. The SPOT4 (Take5) data, that prefigure those of Sentinel-2, proved perfectly suited to this application. Unfortunately, the clay crust has severely limited the emergence of culture. Yet this phenomenon, well-known to our farmer, taught us to cultivate humility ;-) , and we will consider the introduction of the risk in a decision support system. Finally, the functional constraints of the gravity irrigation system have taught us that our tool should be more flexible to recommend an irrigation period instead of a single date, and that we should link the service to weather forecasts.


Following this experiment, we started developing a Web service (SAT-IRR) that should shortly provide the essential functions of an irrigation decision support with a simplified interface.


A SPOT4(Take5) meeting at CNES, on October the 2nd.


CNES is organizing a one day meeting about SPOT4(Take5) on October the 2nd, in its premises in Toulouse.


This meeting is aimed at explaining how the experiment turned out and at presenting the available data set. These presentations wil take place during the morning and the afternoon will be dedicated to presentations of users projects and eventually their first results. Questions/answers sessions will also be organized.


Because of CNES security rules, this meeting is only open to persons having a passport from one of the European Union countries. Otherwise a 2 month delay is necessary, which is no possible here.If you are interested, you will need to send an email before September the 17th, to sylvia.sylvander @ (mentioning your name, nationality and affiliation). If you would like to present your project with SPOT4(Take5) data, please also mention the title of your presentation.


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 :


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.