IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, VOL. 7, NO. 2, FEBRUARY 2014 389
Comparison of Column-Averaged Volume Mixing
Ratios of Carbon Dioxide Retrieved From
IASI/METOP-A Using KLIMA Algorithm and
TANSO-FTS/GOSAT Level 2 Products
Lucia Maria Laurenza, S. Del Bianco, M. Gai, F. Barbara, G. Schiavon, and U. Cortesi
Abstract—The ESA research project “Application of KLIMA
Algorithm to CO Retrieval from IASI/METOP-A Observations and Comparison with TANSO-FTS/GOSAT Products” aims to develop a dedicated software, based on the KLIMA inversion algorithm (originally proposed by IFAC-CNR for the 6 cycle of
ESA Earth Explorer Core Missions), suited for CO retrieval and integrated into the ESA grid-based operational environment Grid
Processing On-Demand (G-POD) to process Level 1 data acquired by the infrared atmospheric sounding interferometer (IASI) and to perform a comparison with Thermal And Near-infrared
Sensor for carbon Observation Fourier Transform Spectrometer (TANSO-FTS), on board of the Greenhouse gases Observing
SATellite (GOSAT), Level 2 data. In order to obtain a reasonable capacity to bulk processing IASI data, we choose to integrate the KLIMA code into the G-POD system. For this reason, we investigated an optimized version of the KLIMA algorithm, aiming at developing a nonoperational retrieval code with adequate features for the integration on the G-POD system. The optimized version of KLIMA retrieval code has been completed and integrated on the G-POD operational environment and is available for bulk processing of IASI data. Using the KLIMA inversion code integrated into the ESA G-POD, it was possible to perform an extensive comparison of a selected set of IASI measurements collocated with TANSO-FTS observations. We performed an extensive comparison of the column-average CO dry air mole fraction (XCO ) retrieved from IASI measurements by using the KLIMA/G-POD inversion code with the operational
Level 2 SWIR products (Version 01.xx and Version 02.xx) from collocated TANSO-FTS observations. In this work, we describe the strategy adopted for the comparison and we show the results of this activity.
Index Terms—Atmospheric modeling, carbon dioxide, data validation, Greenhouse gases Observing SATellite (GOSAT), hyperspectral sensors, infrared atmospheric sounding interferometer (IASI), KLIMA algorithm, remote sensing.
Manuscript received March 27, 2013; revised May 27, 2013; accepted July 12, 2013. Date of publication August 21, 2013; date of current version February 03, 2014. The work was carried out as part of ESA-ESRIN Contract No. 21612/08/I-OL. (Corresponding author: L. M. Laurenza.)
L. M. Laurenza, S. Del Bianco, M. Gai, F. Barbara, and U. Cortesi are with the Institute for Applied Physics “Nello Carrara” IFAC-CNR, Sesto Fiorentino 50019, Italy (e-mail: email@example.com).
G. Schiavon is with the Dipartimento Informatica, Sistemi e Produzione, University of Tor Vergata, Rome I-00133, Italy.
Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JSTARS.2013.2276125
C ARBON DIOXIDE is a key constituent of the terrestrialatmosphere with both natural and anthropogenic sources.
It is one of the primary forcing agents of the greenhouse effect, as well as from being the most mobile component of the global carbon cycle, that is critically coupled to the Earth’s climate system. The short-term increase in background concentration of atmospheric CO observed in the last two centuries, from a preindustrial value of 280 to 380 ppm, originates from the steady growth of anthropogenic emissions (mostly due to combustion of fossil fuels 75% and to land use practices 25% and increased by about 80% only from 1970 to 2004 ). This increase has been unambiguously related to significant changes in the Earth’s climate, particularly the rise of global mean surface temperature. Periodical assessments of the effects of mitigation measures, targeted to limit the release of carbon dioxide from anthropogenic sources and to protect the natural sinks responsible for CO uptake, are necessary, and rely, in turn, upon accurate estimation of the strengths and spatial distribution of CO sources and sinks by inverse modeling of surface carbon fluxes.
This is based on atmospheric transport models that simulate the air–ocean and the air–land exchanges, being constrained by observed spatial and temporal gradient of CO concentration and by the associate uncertainties , .
Experimental data to constrain inverse modeling of the carbon cycle came first from surface flask sampling networks that, although highly precise and accurate, are too sparse to provide an optimal geographical coverage for inversion studies . An effective and potentially valuable source of CO data, which might become available as an alternative to surface observations (which have limited geographical and temporal coverage), is represented by remote-sensing measurements from space-borne sensors. The number of CO observations from satellite instruments can be an order of magnitudes larger than that offered by existing surface networks. In addition, remotely sensed data have the unique features of quasi-global coverage, high temporal sampling, and, in some cases, the capability to obtain vertically resolved information about the
CO distribution. Currently, the most concrete techniques for producing global maps of CO concentration from space- borne platforms exploit passive measurements in the near-infrared 1939-1404 © 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. 390 IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, VOL. 7, NO. 2, FEBRUARY 2014 (NIR) and thermal infrared (TIR). Each of the two spectral regions offers relative advantages and drawbacks, but observations in the NIR, being sensitive down to the lowermost layers of the atmosphere, are capable to retrieve the CO content in the entire air column. TIR measurements are only sensitive to