[A-AS01] Global Carbon Cycle Observation and Analysis

[AAS01-P08] Validation of GOSAT SWIR XCO₂ and XCH₄ retrieved by PPDF-S method

*岩﨑 千沙¹、林田 佐智子²、今須 良一¹、横田 達也³、森野 勇³、吉田 幸生³ (1.東京大学大気海洋研究所、2.奈良女子大学、3.国立環境研究所)

キーワード：GOSAT, retrieval, carbon dioxide, methane

We focused on column averaged dry air mole fraction of atmospheric CO₂ and CH₄ (XCO₂ and XCH₄, respectively) retrievals from Greenhouse gases Observing Satellite (GOSAT) measurements through the photon path length probability density function (PPDF-S) based retrieval method that simultaneously retrieves target gas abundance and PPDF parameters. This method is used for an effective retrieval algorithm even under high concentration of clouds and aerosols. First, we validated PPDF-S XCO₂ and XCH₄ retrievals by comparing them with ground-based observations provided by the Total Carbon Column Observing Network (TCCON) from June 2009 to May 2014. For comparison, we also validate retrievals through another algorithm using full physics (FP)-based retrieval method. PPDF-S and FP retrieval methods are different in way to account for light scattering effect. All these XCO₂ and XCH₄ retrievals are provided by the National Institute for Environmental Studies (NIES). PPDF-S retrievals have positive biases (0.47 ± 2.11 ppm for XCO₂ and 0.76 ± 15.49 ppb for XCH₄), on the other hand, FP retrievals have negative biases (-0.28 ± 2.34 ppm for XCO₂ and -2.16 ± 13.26 ppb for XCH₄). Next, we compare global maps of XCO₂ and XCH₄ mean value, standard deviation and number of data between PPDF-S and FP retrievals. Over the ocean, PPDF-S method can retrieve large number of data whose standard deviation is larger than FP method. These PPDF-S retrievals over the ocean include data which are eliminated in post-screening process for FP method to exclude data that are strongly affected by clouds and aerosol.