HR: 1340h
AN: A13D-1507    [Abstracts]
TI: Accuracy and Precision of CO₂ Vertical Profiles Retrieved from Thermal Infrared Spectra of GOSAT/TANSO-FTS sensor
AU: * Saitoh, N
EM: snaoko@ccsr.u-tokyo.ac.jp
AF: Center for Climate System Research, University of Tokyo, General Research Building 5-1-5 Kashiwanoha, Kashiwa, 277-8568, Japan
AU: Ota, Y
EM: ota.yoshifumi@nies.go.jp
AF: National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, 305-8506, Japan
AU: Niwa, Y
EM: yniwa@ccsr.u-tokyo.ac.jp
AF: Center for Climate System Research, University of Tokyo, General Research Building 5-1-5 Kashiwanoha, Kashiwa, 277-8568, Japan
AU: Imasu, R
EM: imasu@ccsr.u-tokyo.ac.jp
AF: Center for Climate System Research, University of Tokyo, General Research Building 5-1-5 Kashiwanoha, Kashiwa, 277-8568, Japan
AB: The Greenhouse gases Observing Satellite (GOSAT) has been developed by National Institute for Environmental Studies (NIES), Ministry of the Environment, and Japan Aerospace Exploration Agency (JAXA), and planed to be lunched in 2008 for global observation of greenhouse gases such as CO₂ and CH₄ with high accuracy/precision. This study assesses the performance of the FTS sensor on board the GOSAT, Thermal And Near infrared Sensor for carbon Observation (TANSO)-FTS, in retrieving CO₂ concentrations from spectra at around 15 μm (700-800 cm^-1). For this purpose, we computed pseudo-spectra with the spectral resolution of TANSO-FTS for various atmospheric conditions and then performed CO₂ retrieval simulations using the simulated spectra by applying Maximum a posteriori (MAP) method. CO₂ concentrations used in the simulations were derived from the transport model based on the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) [ Satoh et al., 2007]. The results show that the retrieved CO₂ concentrations agree with true concentrations to within +/-0.5% above 700 hPa every season and every latitudinal region if no temperature bias and random scatter are included. In case of temperature bias of 1 K, +/-15% or larger errors are yielded in retrieved CO₂ concentrations below 100 hPa. However, selecting channels appropriate for CO₂ retrievals on the basis of information contents computed following the Shannon�fs information theory [ Shannon and Weaver, 1949] could greatly reduce the errors attributable to temperature uncertainties; the +/-15% differences between true and retrieved CO₂ concentrations decrease less than +/-2% differences through the channel selection. We also applied the developed algorithm to real spectra obtained with a hyper-spectral sensor to confirm the validity of the method.
DE: 0300 ATMOSPHERIC COMPOSITION AND STRUCTURE
DE: 0365 Troposphere: composition and chemistry
DE: 0480 Remote sensing
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting