Abstract. This paper discusses the retrieval of atmospheric methane profiles from the
thermal infrared band of the Japanese Greenhouse Gases Observing Satellite
(GOSAT) between 1210 and 1310 cm−1, using the RemoTeC analysis
software. Approximately one degree of information on the vertical methane
distribution is inferred from the measurements, with the main sensitivity at
about 9 km altitude but little sensitivity to methane in the lower
troposphere. For verification, we compare the GOSAT-TIR methane profile
retrieval results with profiles from model fields provided by the Monitoring
Atmospheric Composition and Climate (MACC) project, scaled to the total
column measurements of the Total Carbon Column Observing Network (TCCON) at
ground-based measurement sites. Without any radiometric corrections of GOSAT
observations, differences between both data sets can be as large as 10 %. To
mitigate these differences, we developed a correction scheme using a
principal component analysis of spectral fit residuals and airborne
observations of methane during the HIAPER pole-to-pole observations (HIPPO)
campaign II and III. When the correction scheme is applied, the bias in the
methane profile can be reduced to less than 2 % over the whole altitude range
with respect to MACC model methane fields. Furthermore, we show that, with
this correction, the retrievals result in smooth methane fields over land and
ocean crossings and no differences can be discerned between daytime and
nighttime measurements. Finally, a cloud filter is developed for the
nighttime and ocean measurements. This filter is rooted in the GOSAT-TIR
(thermal infrared) measurements and its performance, in terms of biases, is consistent with the
cloud filter based on the GOSAT-SWIR (shortwave infrared) measurements. The TIR filter shows a
higher acceptance rate of observations than the SWIR filter, at the cost of a
higher uncertainty in the retrieved methane profiles.
A Blackbody Design for SI-Traceable Radiometry for Earth Observation