Assessment of the quality of the Version 1.07 temperature‐versus‐pressure profiles of the middle atmosphere from TIMED/SABER

Abstract

The quality of the retrieved temperature‐versus‐pressure (or T(p)) profiles is described for the middle atmosphere for the publicly available Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) Version 1.07 (V1.07) data set. The primary sources of systematic error for the SABER results below about 70 km are (1) errors in the measured radiances, (2) biases in the forward model, and (3) uncertainties in the corrections for ozone and in the determination of the reference pressure for the retrieved profiles. Comparisons with other correlative data sets indicate that SABER T(p) is too high by 1–3 K in the lower stratosphere but then too low by 1 K near the stratopause and by 2 K in the middle mesosphere. There is little difference between the local thermodynamic equilibrium (LTE) algorithm results below about 70 km from V1.07 and V1.06, but there are substantial improvements/differences for the non‐LTE results of V1.07 for the upper mesosphere and lower thermosphere (UMLT) region. In particular, the V1.07 algorithm uses monthly, diurnally averaged CO2 profiles versus latitude from the Whole Atmosphere Community Climate Model. This change has improved the consistency of the character of the tides in its kinetic temperature (Tk). The Tk profiles agree with UMLT values obtained from ground‐based measurements of column‐averaged OH and O2 emissions and of the Na lidar returns, at least within their mutual uncertainties. SABER Tk values obtained near the mesopause with its daytime algorithm also agree well with the falling sphere climatology at high northern latitudes in summer. It is concluded that the SABER data set can be the basis for improved, diurnal‐to‐interannual‐scale temperatures for the middle atmosphere and especially for its UMLT region.