Characterizing sampling and quality screening biases in infrared and microwave limb sounding

Abstract

Abstract. This study investigates orbital sampling biases and evaluates the additional impact caused by data quality screening for the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) and the Aura Microwave Limb Sounder (MLS). MIPAS acts as a proxy for typical infrared limb emission sounders, while MLS acts as a proxy for microwave limb sounders. These biases were calculated for temperature and several trace gases by interpolating model fields to real sampling patterns and, additionally, screening those locations as directed by their corresponding quality criteria. Both instruments have dense uniform sampling patterns typical of limb emission sounders, producing almost identical sampling biases. However, there is a substantial difference between the number of locations discarded. MIPAS, as a mid-infrared instrument, is very sensitive to clouds, and measurements affected by them are thus rejected from the analysis. For example, in the tropics, the MIPAS yield is strongly affected by clouds, while MLS is mostly unaffected. The results show that upper-tropospheric sampling biases in zonally averaged data, for both instruments, can be up to 10 to 30 %, depending on the species, and up to 3 K for temperature. For MIPAS, the sampling reduction due to quality screening worsens the biases, leading to values as large as 30 to 100 % for the trace gases and expanding the 3 K bias region for temperature. This type of sampling bias is largely induced by the geophysical origins of the screening (e.g. clouds). Further, analysis of long-term time series reveals that these additional quality screening biases may affect the ability to accurately detect upper-tropospheric long-term changes using such data. In contrast, MLS data quality screening removes sufficiently few points that no additional bias is introduced, although its penetration is limited to the upper troposphere, while MIPAS may cover well into the mid-troposphere in cloud-free scenarios. We emphasize that the results of this study refer only to the representativeness of the respective data, not to their intrinsic quality.

The SPARC water vapor assessment II: Assessment of satellite measurements of upper tropospheric water vapor

“…A disadvantage is that there can be significant sampling biases. For example, limb viewing infrared instruments are heavily cloud contaminated in the tropics and will show a significant dry bias compared to maps made from nadir viewing or submillimeter instruments (Millán et al, 2018). This study will show that the sampling bias is more than a factor of two in the upper troposphere.…”

A Multi-Parameter Dynamical Diagnostics for Upper Tropospheric and Lower Stratospheric Studies

“…These datasets differ greatly in geographical coverage, horizontal and vertical resolution, measurement frequency, and time period covered. To fully exploit these measurements, careful attention needs to be paid to understand how sampling differences between the instruments can affect their representation of the atmospheric state (e.g, Manney et al, 2007;Hegglin et al, 2008;Toohey et al, 2013;Lin et al, 2015;Millán et al, 2016;Miyazaki and Bowman, 2017;Millán et al, 2018;Chang et al, 2020).…”

Reply to Referee 2 Comments

“…H 2 O, O 3 , HNO 3 , CH 4 , N 2 O, and NO 2 -as well as temperature (Fischer and Oelhaf, 1996;Fischer et al, 2008). Here, the Envisat/MIPAS HNO 3 and temperature data version V5H_HNO3_20 and V5H_T_20 and V5R_HNO3_224/225 and V5R_T_220/221 (nominal mode) derived with the IMK/IAA retrieval processor covering the periods July 2002-March 2003 and January 2005-April 2012, respectively, have been used (updated version of the retrieval as described in Milz et al, 2009, andvon Clarmann et al, 2009). Comparison of the MIPAS HNO 3 data product with satellite measurements from Odin/SMR (Odin/Sub-Millimetre Radiometer), ADEOS/ILAS-II (Advanced Earth Observing Satellite/Improved Limb Atmospheric Spectrometer), SCISAT/ACE-FTS (SCISAT/Atmospheric Chemistry Experiment-Fourier Transform Spectrometer), and the Envisat/MIPAS ESA (European Space Agency) product showed good agreement, and differences were generally within ±0.5 ppbv (Wang et al, 2007;Wolff et al, 2008).…”