Characterization and evaluation of AIRS-based estimates of the deuterium content of water vapor

John R Worden,Fredrick W Irion,Robert L Herman,Jean-Luc Moncet,Alan E Lipton,Kevin W Bowman,Susan S Kulawik,Karen Cady-Pereira,Vivienne H Payne,Igor Polonsky,Yuguang He,Dejian Fu

doi:10.5194/amt-12-2331-2019

Abstract

Abstract. Single-pixel tropospheric retrievals of HDO and H2O concentrations are retrieved from Atmospheric Infrared Sounder (AIRS) radiances using the optimal estimation algorithm developed for the Aura Tropospheric Emission Spectrometer (TES) project. We evaluate the error characteristics and vertical sensitivity of AIRS measurements corresponding to 5 d of TES data (or five global surveys) during the Northern Hemisphere summers between 2006 and 2010 (∼600 co-located comparisons per day). We find that the retrieval characteristics of the AIRS deuterium content measurements have similar vertical resolution in the middle troposphere as TES but with slightly less sensitivity in the lowermost troposphere, with a typical degrees of freedom (DOFS) in the tropics of 1.5. The calculated measurement uncertainty is ∼30 ‰ (parts per thousand relative to the deuterium composition of ocean water) for a tropospheric average between 750 and 350 hPa, the altitude region where AIRS is most sensitive, compared to ∼15 ‰ for the TES data. Comparison with the TES data also indicates that the uncertainty of a single target AIRS HDO ∕ H2O measurement is ∼30 ‰. Comparison of AIRS and TES data between 30∘ S and 50∘ N indicates that the AIRS data are biased low by ∼-2.6 ‰ with a latitudinal variation of ∼7.8 ‰. This latitudinal variation is consistent with the accuracy of TES data compared to in situ measurements, suggesting that both AIRS and TES have similar accuracy.

Highlights

Measurements of the isotopic composition of water can help identify the source of the water and provide knowledge about its condensation and evaporation history (e.g., Galewsky et al, 2016, and references therein)
The error bars shown on the difference is the error on the mean, which is the root mean square (RMS) of the differences divided by the square root of the number of co-located observations; as this error bar is a measure of precision for each latitude bin, this comparison demonstrates that there are variations in the comparison that are larger than the precision and are related to systematic errors in either the Tropospheric Emission Spectrometer (TES) data or Atmospheric Infrared Sounder (AIRS) data or both
This paper describes the vertical resolution and error characteristics of retrievals of the deuterium content of water vapor using AIRS radiances and evaluates the consistency between AIRS and TES retrievals of HDO and H2O

Summary

Introduction

Measurements of the isotopic composition of water can help identify the source of the water and provide knowledge about its condensation and evaporation history (e.g., Galewsky et al, 2016, and references therein). Near global measurements of the isotopic composition of water vapor became possible with the advent of spectroscopic techniques applied to in situ measurements (e.g., Noone et al, 2011) using lasers and for passive ground-based and satellite measurements (e.g., Worden et al, 2006; Frankenberg et al, 2009; Schneider et al, 2012; Lacour et al, 2012) These data have in turn been used to evaluate the role of convection, largescale dynamics, and evapotranspiration in the tropical water cycle (e.g., Worden et al, 2007; Frankenberg et al, 2009; Wright et al, 2017) and tropical convection (e.g., Lacour et al, 2018, and refs therein) and the role of plants in global evapotranspiration (Good et al, 2015). We compare the AIRS and TES data to evaluate the calculated uncertainties of the AIRS data

Description of AIRS and TES instruments

Description of the radiative transfer forward model

Description of the retrieval approach

Findings

Conclusion