March 28, 2017

Infrared Imager Radiometer

The Imaging Infrared instrument is the French contribution to the CALIPSO payload.
It was developed by SODERN company with CNES as prime contractor.
The Laboratoire de Météorologie Dynamique (LMD) designed and developed the black body of the instrument and performed its ground calibration.

The IIR instrument is a 3 channel imaging radiometer in the thermal infrared at 8.65 µm, 10.6 µm and 12.05 µm.
The IIR images will provide the context of the lidar measurement by night and allow the co-registration with the MODIS multispectral radiometer on board Aqua.
IIR measurements, combined with the lidar information, will furthermore enable to retrieve the size of ice particles in semi-transparent clouds.

The number and the choice of spectral bands have been optimized to increase the efficiency of this retrieval. While the classical pair of 11-12 µm channels is sensitive to small particles, the 8.5-12 µm channel pair is more sensitive to large particles. A better discrimination of the absorption and scattering effects of complex shape crystal is expected as well as a differentiation of sphere from hexagonal shapes.

The design of the IIR instrument uses an adaptation of a single channel infrared camera (ISM) developed for the IASI project on board METOP. Is uses a non-cooled microbolometer detector.

Description of the IIR

The IIR is composed of:

  • A camera part (ISM), constituted of an objective (aperture 0.75) optimized in the thermal infrared, a microbolometer detector array, specific electronics, a passive cooler to refrigerate the whole, some mechanical pieces.
  • A filter-carrying wheel enabling to insert alternately 3 spectral filters in front of the camera.
  • A black body to calibrate the camera.
  • A pointing mirror rotating to sequentially select between the Earthview, the black body and the direction of the cold space (second source for calibration).

Main technical characteristics

  • Spectral bands: 3 bands selected by 3 spectral filters
Centre: 8.65 µm Bandwidth: 0.9 µm
Centre: 10.6 µm Bandwidth: 0.6 µm
Centre: 12.05 µm Bandwidth: 1 µm
  • Instantaneous field of view of 64 km x 64 km on the ground
  • Pixel size of 1 km x 1 km
  • Radiometric performance
NedT < 0.5 K @ 210 K
Absolute calibration accuracy < 1 K
  • Mass: 24 kg
  • Volume: 490 x 550 x 320 mm³
  • Consumption: 27 W
  • Data rate: 50 kbps


The IIR has been built by the Société anonyme d'études et réalisation nucléaires (SODERN) located at Limeil-Brévannes, under CNES contract. Among the components, the microbolometrer detector array is a BOEING U3000 detector.

On-ground calibration

Measuring infrared brightness temperatures with the accuracy needed for the scientific applications requires on-board calibration.
It is achieved by inserting between Earth images, uniform cold and warm scenes with stable and accurately known temperature, in the same way as to calibrate a household thermometer. The cold scene is obtained by aiming at deep space (which temperature is 3 K) while the warm scene is a black body.
It is an annular device of 9 cm diameter painted in black designed by LMD to have properties approaching an ideal black body. Its emissivity is estimated higher than 0.986.

A complete radiometric characterization of the instrument was performed before launch by LMD in a vacuum chamber.

Level 1 processing

The raw images of the IIR are operationally processed by NASA with a processing line specified by CNES.
This processing, named level 1, conducts:

  • the joining and the projection of the images (in the 3 channels) on a same geographical grid centred on the lidar spot position and
  • the in-flight radiometric calibration using the calibration images interleaved with the Earth views.

The radiances projected and calibrated are available for the users in an IIR level 1B product organized by half-orbits.

Level 2 processing

The level 2 processing goal is to supply the emissivity of clouds and the size of particles in semi-transparent ice clouds.
Two products are generated, one along the lidar track, and the other across the swath (64 km) of the IIR.
The processing algorithms have been specified by the IPSL and will be processed by NASA.

The classic split window technique was adapted to get the most out of the altitude measurement and of the scene classification provided by the lidar.
The cloud emissivity calculations use the cloud altitude as input, converted into temperature via the meteorological profile, which reduces a major error contribution of the method.
Moreover, the depolarization measured by the lidar provides information on the particle shape which refines our knowledge of their size.
Another improvement is in the variety of particle models used in precomputed tables and in the care with which the optical properties of these particles are calculated.

The simulations made with airborne data show a significant improvement on the retrieval of the particle size.

Aside from the lidar track, an algorithm extends the retrieval to neighbour pixels for which the lidar measurement is sufficiently representative.
The homogeneity of the scene is estimated from the IIR image itself and from the corresponding visible image, acquired by day by the WFC camera.

The IIR Technical Expertise Centre

The IIR Technical Expertise Centre enables CNES to ensure its technical expertise responsibility for the IIR, which includes 2 parts:

  • an instrumental part: technical in-flight monitoring of the instrument
  • an image quality part: in-flight commissioning and in-orbit monitoring of the IIR geometric and radiometric quality, i.e. the quality of the IIR level 1 product.

The IIR Technical Expertise Centre relies upon a computing facility named TEC which provides the operators and experts with the adequate working environment to conduct their investigations. TEC receives from MOCC the IIR housekeeping data, and from the ASDC or the SCF the raw data (level 0), level 1 and calibration products.

TEC was developed by Cap Gemini with CNES as prime contractor. It is located at the Toulouse Space Centre and will be operated by CNES teams.