Abstract. Hydroxyl (OH) short-wave infrared emissions arising from OH(4-2, 5-2,
8-5, 9-6) as measured by channel 6 of the SCanning Imaging Absorption
spectroMeter for Atmospheric CHartographY (SCIAMACHY) are used to
derive concentrations of OH(v=4, 5, 8, and 9) between 80 and
96 km. Retrieved concentrations are used to simulate OH(5-3,
4-2) integrated radiances at 1.6 µm and OH(9-7, 8-6) at
2.0 µm as measured by the Sounding of the Atmosphere using
Broadband Emission Radiometry (SABER) instrument, which are not fully
covered by the spectral range of SCIAMACHY measurements. On average,
SABER “unfiltered” data are on the order of 40 % at
1.6 µm and 20 % at 2.0 µm larger than
the simulations using SCIAMACHY data. “Unfiltered” SABER data are
a product, which accounts for the shape, width, and transmission of
the instrument's broadband filters, which do not cover the full
ro-vibrational bands of the corresponding OH transitions. It is found
that the discrepancy between SCIAMACHY and SABER data can be reduced
by up to 50 %, if the filtering process is carried out
manually using published SABER interference filter characteristics and
the latest Einstein coefficients from the HITRAN database. Remaining
differences are discussed with regard to model parameter uncertainties
and radiometric calibration.