The SPARC water vapour assessment II: profile-to-profile and climatological comparisons of stratospheric <i>δ</i>D(H<sub>2</sub>O) observations from satellite

Abstract

Abstract. Within the framework of the second SPARC (Stratosphere-troposphere Processes And their Role in Climate) water vapour assessment (WAVAS-II), we evaluated five data sets of δD(H2O) obtained from observations by Odin/SMR (Sub-Millimetre Radiometer), Envisat/MIPAS (Environmental Satellite/Michelson Interferometer for Passive Atmospheric Sounding), and SCISAT/ACE-FTS (Science Satellite/Atmospheric Chemistry Experiment – Fourier Transform Spectrometer) using profile-to-profile and climatological comparisons. These comparisons aimed to provide a comprehensive overview of typical uncertainties in the observational database that could be considered in the future in observational and modelling studies. Our primary focus is on stratospheric altitudes, but results for the upper troposphere and lower mesosphere are also shown. There are clear quantitative differences in the measurements of the isotopic ratio, mainly with regard to comparisons between the SMR data set and both the MIPAS and ACE-FTS data sets. In the lower stratosphere, the SMR data set shows a higher depletion in δD than the MIPAS and ACE-FTS data sets. The differences maximise close to 50 hPa and exceed 200 ‰. With increasing altitude, the biases decrease. Above 4 hPa, the SMR data set shows a lower δD depletion than the MIPAS data sets, occasionally exceeding 100 ‰. Overall, the δD biases of the SMR data set are driven by HDO biases in the lower stratosphere and by H2O biases in the upper stratosphere and lower mesosphere. In between, in the middle stratosphere, the biases in δD are the result of deviations in both HDO and H2O. These biases are attributed to issues with the calibration, in particular in terms of the sideband filtering, and uncertainties in spectroscopic parameters. The MIPAS and ACE-FTS data sets agree rather well between about 100 and 10 hPa. The MIPAS data sets show less depletion below approximately 15 hPa (up to about 30 ‰), due to differences in both HDO and H2O. Higher up this behaviour is reversed, and towards the upper stratosphere the biases increase. This is driven by increasing biases in H2O, and on occasion the differences in δD exceed 80 ‰. Below 100 hPa, the differences between the MIPAS and ACE-FTS data sets are even larger. In the climatological comparisons, the MIPAS data sets continue to show less depletion in δD than the ACE-FTS data sets below 15 hPa during all seasons, with some variations in magnitude. The differences between the MIPAS and ACE-FTS data have multiple causes, such as differences in the temporal and spatial sampling (except for the profile-to-profile comparisons), cloud influence, vertical resolution, and the microwindows and spectroscopic database chosen. Differences between data sets from the same instrument are typically small in the stratosphere. Overall, if the data sets are considered together, the differences in δD among them in key areas of scientific interest (e.g. tropical and polar lower stratosphere, lower mesosphere, and upper troposphere) are too large to draw robust conclusions on atmospheric processes affecting the water vapour budget and distribution, e.g. the relative importance of different mechanisms transporting water vapour into the stratosphere.

A reassessment of the discrepancies in the annual variation of δD-H2O in the tropical lower stratosphere between the MIPAS and ACE-FTS satellite data sets

“…The pre-screening of the individual satellite data sets is the same as described in Sect. 2 and 3 of Högberg et al (2019).…”

“…The reason for this approach is consistency since the SMR observations of HDO and H 2 O are not simultaneous and hence δD cannot be derived on a single profile basis. As shown by Högberg et al (2019) there are some differences between the different approaches to calculate δD results because of their non-commutativity. However, the main conclusions of this work are not affected by the particular approach chosen.…”

“…No constraints lead by definition to the same vertical resolution for HDO and H 2 O. This mismatch in the vertical resolution of the MIPAS HDO and H 2 O data can lead to spurious results for δD when they are combined, regardless if the combination follows the "separate" or the "individual" approach (Högberg et al, 2019). This statement is of general character for any ratio that is not retrieved directly, not only for δD.…”

“…Observations, however, show typically δD values between about −700 h and −500 h (e.g. Moyer et al, 1996;Johnson et al, 2001;Kuang et al, 2003;Webster and Heymsfield, 2003;Nassar et al, 2007;Högberg et al, 2019), clearly indicating an importance of other processes like the convective lofting of ice or the mixing of air masses and supersaturation effects.…”

“…In addition, the amplitude of the annual variation in δD was considerably smaller than that observed in the MIPAS observations. Recently, Högberg et al (2019) compared multiple δD data sets from MIPAS, ACE-FTS as well as the SMR instrument aboard the Odin satellite and concluded that all showed characteristics that typically are associated with the tape recorder. They iterated that pronounced quantitative differences exist, which would yield different interpretations of the relative importance of convection for lower stratospheric water vapour.…”