You are here: Home >> Ongoing projects >> KLIMA >> Overview
  • Print

 

 

Sensitivity analysis and application of KLIMA algorithm

to OCO and GOSAT validation

 

 

Overview

The KLIMA-IASI project is a study supported by ESA-ESRIN and lead by IFAC-CNR in collaboration with IUP, Universtiy of Bremen. The primary objective of the project is to investigate the ultimate capabilities of IASI-METOP observations for the retrieval of CO2 total abundance averaged over a monthly to seasonal time scale and over a spatial scale compatible with the requirements of a comparison with the CO2 products of the satellite missions OCO and GOSAT.

 

metop artist view.jpg

The Metop-A satellite - Artist view

[Credits: ESA AOES Medialab]

 

 

 

The study is organised in two phases. Specific objectives of Phase 1 (May 2008 – March 2009) included:

  • Adaptation of the KLIMA retrieval algorithm into a non-operational inversion code optimised for fast and accurate retrieval of CO2 average information from IASI unapodised calibrated spectra.
  • Sensitivity assessment and evaluation of the performances of the optimised KLIMA-IASI code for retrieval of CO2 information from single spectra.
  • Conclusions about the level of temporal and spatial averaging required to meet the uncertainty requirements necessary for the comparison with both OCO and GOSAT CO2 retrieval products.

 

The objectives of Phase 2 (March 2009 – December 2010) are:

  • Integration of the optimized KLIMA/IASI CO2 retrieval code into the G-POD (Grid Processing On-Demand) operational environment.
  • Inter-comparison of CO2 products retrieved from collocated IASI, OCO and GOSAT measurements for cross-validation of the three instruments, aimed at identifying possible recommendations for calibration and algorithms improvement.

 

The approach followed in Phase 1 of the study relied upon the use of the algorithm originally proposed by IFAC-CNR for the KLIMA Earth Explorer mission and exploited by the MARC retrieval code (Carli et al., 2007) for the analysis of limb-sounding observations in the millimetre-waves, to analyze the Millimetre-wave Airborne Receiver for Spectroscopic CHaracterization of Atmospheric Limb-Sounding (MARSCHALS) measurements (Oldfield et al., 2001).

The adaptation of the MARC software for the analysis of thermal infrared radiances acquired by the REFIR (Radiation Explorer in Far-Infrared) balloon-borne spectrometer in nadir viewing geometry provided the baseline version for further upgrading and optimisations of the code that were performed in the frame of this study.

The prototype software system implemented as a results of this effort represents the best trade-off between accuracy and computing time and it is referred to as KLIMA-IASI Accelerated Retrieval Model (ARM).

 

The main elements of innovation introduced by the KLIMA-IASI ARM and by its application to the retrieval of CO2 column from IASI can be recognized in the following aspects:

  • Demonstration of the feasibility of a retrieval approach exploiting all spectral channels of IASI to extract information on carbon dioxide and other target species with no need for channel selection or other data reduction techniques to handle the high dimensionality problem posed by IASI data processing.

 

  • Use of a multi-target retrieval scheme which removes the systematic errors due to the interfering parameters, in combination with a variance-covariance matrix of the residuals which takes into account the errors on forward model parameters. These features of the retrieval code are optimally suited for the estimate of the precision error on the retrieved CO2 column.

 

  • Standard output of the retrieval process including products in the data format required for calculation of the Measurement Space Solution. The Measurement Space Solution method, proposed by Ceccherini et al.(2009) is a technique that allows the optimal exploitation and representation of the information on the vertical profile of an atmospheric constituent retrieved from remote sensing observations.