The water vapour self- and foreign-continua in the 1.6 µm and 2.3 µm windows by CRDS at room temperature

https://doi.org/10.1016/j.jqsrt.2019.02.016Get rights and content

Highlights

  • Water self-continuum cross-sections are measured by CRDS in the 1.6 µm window.

  • These accurate measurements confirm and extend a previous CRDS study.

  • Self-continuum cross-sections are also measured near 4720 cm1 and 5130 cm1.

  • Foreign-continuum cross-sections are reported at 4 points of the 2.3 µm window.

  • Significant underestimations of the MT_CKD foreign continuum are evidenced.

Abstract

New measurements of the water vapour continuum in the 1.6 µm and 2.3 µm windows are obtained at room temperature using highly stable and sensitive cavity ring down (CRD) spectrometers.

In the 1.6 µm window, self-continuum cross-sections, CS, are derived for 30 selected spectral points between 5700 and 6850 cm−1 using pressure ramps of pure water vapour (up to 15 Torr). Purely quadratic pressure dependence is obtained for each measurement point. Compared to our previous dataset (Mondelain et al., 2014), the retrieved Cs values are more accurate and show a general agreement. The spectral coverage is extended both on the high and low frequency edges of the 1.6 µm window. In addition, new CS values are derived in the 5120–5137 cm−1 interval extending to higher energy the coverage of the 2.3 µm window. The measurements near 5700 cm−1 and 5130 cm−1 are found lower than the MT_CKD_3.2 values by a factor of about 2 and 1.5, respectively.

Foreign-continuum cross-sections, CF, are newly obtained in the 2.3 µm window from CRDS spectra of moist air in flow regime. Spectra were recorded for different water vapour partial pressures while maintaining the total pressure constant in the high finesse CRDS cavity. After subtraction of the monomer and self-continuum contributions, CF values were derived from the linear variation of the foreign-continuum absorption with the water vapour partial pressure. CF values determined at four spectral points between 4430 and 5000 cm−1 are larger by up a factor of five than the corresponding MT_CKD values. Considering these CRDS data at room temperature and literature values at 400 K the temperature dependence of the foreign-continuum is confirmed to be weak.

Introduction

Absorption of water vapour plays a key role in the radiative budget of the Earth's atmosphere [1]. This absorption corresponds to the sum of the narrow rovibrational absorption lines, hereafter named “monomer local line”, with the broadband water vapour continuum, slowly varying in frequency. This latter contribution has its largest impact in transparency windows (or regions of weak monomer absorption) in contrast to the regions corresponding to the water vapour vibrational bands where the monomer contribution largely dominates. In a mixture of water vapour in air, like the Earth's atmosphere, two contributions of the continuum are present: (i) the self-continuum due to interactions between two water molecules and (ii) the foreign-continuum arising from interactions between water molecules and molecular oxygen and nitrogen. Despite decades of debates [2], the origin of the water continuum is unclear with different processes suggested to contribute: far wings of the rovibrational lines, water dimers (stable and metastable) and collision-induced absorption (CIA) [3].

The semi-empirical MT_CKD model (Mlawer-Tobin_Clough-Kneizys-Davies) [4], [5] is the usual pragmatic way to take into account the continuum absorption contribution of the water vapour in atmospheric radiative transfer codes. This model assumes an empirical line profile applied to all the water monomer transitions, with intermediate and far line wings fitted to reproduce a selection of laboratory and atmospheric data. The MT_CKD model also includes an additional term called “weak interaction” term.

During the two last decades large efforts have been done to complete previous experimental data [6], [7], [8] and to better characterize the water vapour continuum, particularly in the infrared windows. Several laboratory studies were dedicated to the measurement of the self-continuum cross-sections using Fourier transform spectroscopy (FTS) [9], [10], [11], [12] and cavity-enhanced absorption spectroscopy (CEAS) including cavity ring down spectroscopy (CRDS) and optical feedback CEAS (OFCEAS) [13], [14], [15], [16]. The FTS room temperature self-continuum cross-section values, in relatively good agreement to each other, were found largely higher than the MT_CKD_2.5 cross-sections in the centre of the 2.1 and 1.6 µm windows by factors greater than 20 [10], [11], [12] and 160 [11], [12], respectively. On the other hand, as a result of their inherent higher sensitivity and high baseline stability, CEAS studies of our group provided datasets with reduced uncertainties. They were found in a much better agreement with the MT_CKD model [16]. Note that our CRDS and OFCEAS measurements [14], [15], [16] rely on the careful checking of the quadratic pressure dependence of the continuum absorption which provides a key criterion to ensure the gas phase origin of the spectrum baseline variation. This is not the case for the FTS results at room temperature which relied on single pressure spectra measurements in the windows. The latest version (3.2) of the MT_CKD self-continuum was constrained to the CEAS cross-sections in the 4.0, 2.1, 1.6 and 1.25 µm windows [17].

In fact, a pure quadratic pressure dependence was achieved in all our CRDS and OFCEAS studies of the water vapour self-continuum except in the 1.6 µm window where the weak continuum signal was found to include not only a pressure squared term but also a linear term in PH2O[18], [19]. This linear term was tentatively interpreted as due to the adsorption of the water molecules on the high reflective mirrors of the CRDS cell. In this situation, the accuracy of our Cs values in the 1.6 µm window was questioned [20]. This is the reason why in the present contribution, we reconsider the 1.6 µm window where measurements are particularly demanding as a result of the weakness of the self-continuum in this window. In addition, the spectral coverage of this window will be extended on its two edges.

The foreign-continuum of water vapour in air is a major source of uncertainty to be constrained in Earth atmospheric models because, under standard conditions, it might account for a significant fraction of the absorbed incoming solar flux [21], [22]. In the laboratory, the foreign-continuum cross-sections are derived from spectra recorded with moist nitrogen (or air) which include both a foreign- and a self-contribution. As a result, the obtained foreign-continuum cross-sections, CF, are directly impacted by the self-continuum cross-sections values, derived separately with pure water vapour samples. The review included in [23], indicates that only a few experimental studies of the water foreign-continuum are available. The reference [24] describing CRDS measurements at 944 cm−1 should be added to the seven references listed in Table 6 of Ref. [23]. The FTS results obtained at NIST [25], [26], [27] and the CRDS data in the 2.1 µm window [21] match and have led to a significant increase of the MT_CKD foreign-continuum model in the transparency windows. Note that, in the limit of the reported error bars, the foreign-continuum has been found to be mostly temperature independent in the broadband CAVIAR FTS measurements between 350 and 430 K. Similar weak temperature dependence results were reported in the infrared in the 296–363 K range [25], [26] and in the 260–360 K range near 183 GHz [28].

In the following, we will present the experimental determination of the foreign-continuum by CRDS at room temperature for four new spectral points of the 2.3 µm window and the production of an extended dataset for the self-continuum cross-sections in the 1.6 µm window.

The rest of the paper is organized as follows. In the next section, the experimental setups used for the self- and foreign-continua measurements are detailed. The data analysis and the cross-sections retrievals are presented in Part 3 and 4 for the self-continuum in the 1.6 µm window and for the measurements in the 2.3 µm window, respectively. These two parts include a comparison to experimental data available in the literature and to the MT_CKD model, and a detailed analysis of the error budget.

Section snippets

The CRDS spectrometers

Three cavity ring down spectrometers were used in this work. The 2.3 µm setup has been described in details in [21], [30] while readers are referred to [31] for the two setups used in the 1.6 µm region. Briefly, a distributed feedback (DFB) laser diode (either from Eblana Photonics or NEL) is coupled to a high finesse cavity composed of two mirrors with highly reflective coatings. The different measurement points in the 2.3 µm window (near 4435, 4522, 4724, 4999 and 5130 cm−1) are listed in

Self-continuum

In a mixture of water vapour in air, the total absorption coefficient can be expressed as the sum of four terms:αtot(ν,T)=αcavity+αWML+αWCS+αWCF=αcavity+αWML+1kTCS(ν,T)PH2O2+1kTCF(ν,T)PH2OPairwhere αcavity, αWML, αWCS and αWCF are the contributions due to the cavity, water vapour “monomer local lines” (WML), water vapour self-continuum (WCS) and foreign-continuum (WCF). CS and CF cross-sections are expressed in cm2 molecule−1 atm−1. Note that the negligible contribution of the Rayleigh

Cross-section retrieval

As mentioned above the foreign-continuum measurements were performed on the basis of spectra recorded over a few cm−1 around 4435, 4522, 4720 and 4999 cm−1. The recordings were performed in flow regime with constant total pressure (generally 700 Torr) and different partial pressures of water vapour up to 6 Torr. For each water concentration, two successive spectra were recorded to check the concentration stability. All the spectra were first corrected from the background spectrum obtained with

Conclusion

In the present work, the accuracy of the self-continuum cross-sections of water vapour has been improved by CRDS in the 1.6 µm window and at the 4720 cm−1 spectral point of the 2.3 µm window. In addition, the spectral coverage of the 1.6 µm window was extended at low and high energies and a new spectral point of the 2.3 µm window was measured at 5130 cm−1. In the context of the debate about the large discrepancies between FTS and laser-based measurements at room temperature in the transparency

Acknowledgments

This project is supported by the Labex OSUG@2020 (ANR10 LABX56) and the LEFE-ChAt program from CNRS-INSU. The authors want to thank J.-M. Hartmann (LMD) for providing them classical molecular dynamic calculations of the N2 CIA in a N2single bondH2O mixture.

References (39)

  • IV Ptashnik

    Evaluation of suitable spectral intervals for near-IR laboratory detection of water vapour continuum absorption

    J Quant Spectrosc Radiat Transf

    (2007)
  • ARW McKellar

    Low-temperature infrared absorption of gaseous N2 and N2 + H2 in the 2.0–2.5 µm region: application to the atmospheres of Titan and Triton

    Icarus

    (1989)
  • U Cubasch et al.

    Introduction

  • KP Shine et al.

    The water vapour continuum: brief history and recent developments

    Surv Geophys

    (2012)
  • EJ Mlawer et al.

    Development and recent evaluation of the MT_CKD model of continuum absorption

    Phil Trans R Soc A

    (2012)
  • DE Burch

    Continuum absorption by H2O

    Report AFGL-TR-81-0300

    (1982)
  • DE Burch et al.

    Continuum absorption by H2O in the 700–1200 cm−1 and 2400–2800 cm−1 windows

    Report AFGL-TR-84-0128

    (1984)
  • DE Burch

    Absorption by H2O in narrow windows between 3000 and 4200 cm−1

    Report AFGL-TR-85-0036

    (1985)
  • DJ Paynter et al.

    Laboratory measurements of the water vapor continuum in the 1200–8000 cm−1 region between 293 K and 351 K

    J Geophys Res

    (2009)
  • Cited by (17)

    • The water vapor foreign continuum in the 8100-8500 cm<sup>−1</sup> spectral range

      2023, Journal of Quantitative Spectroscopy and Radiative Transfer
      Citation Excerpt :

      Cormier et al. [22] and Baranov&Lafferty [23] reported measurements in the 10 µm window by CRDS and FTS, respectively. The water foreign continuum was determined in the 4.0 µm window by FTS [23–25], in the 2.1 and 1.6 µm windows by CRDS [9,14,15] and near 14400 cm−1 by photoacoustic spectroscopy [26]. On the basis of Refs. [9,22], the MT_CKD foreign continuum has been significantly increased starting with version 2.8 [3].

    • The water vapor self-continuum absorption at room temperature in the 1.25 µm window

      2022, Journal of Quantitative Spectroscopy and Radiative Transfer
      Citation Excerpt :

      As a result, the disagreement reaches two orders of magnitude compared to Tomsk measurements near 1.6 µm [9,10]. In the last years, we have undertaken a characterization of the self-continuum of water vapor using cavity enhanced absorption spectroscopies (CEAS) [11–18], namely cavity ring down spectroscopy (CRDS) and optical feedback–cavity enhanced absorption spectroscopy (OF-CEAS). These CEAS self-continuum absorption retrievals were all supported by a clear pressure squared dependence of the continuum absorption in order to exclude potential experimental biases.

    • Simultaneous collision-induced transitions in H<inf>2</inf>O+CO<inf>2</inf> gas mixtures

      2022, Journal of Quantitative Spectroscopy and Radiative Transfer
    • Measurements of the water vapor continuum absorption by OFCEAS at 3.50 µm and 2.32 µm

      2022, Journal of Quantitative Spectroscopy and Radiative Transfer
    • The HITRAN2020 molecular spectroscopic database

      2022, Journal of Quantitative Spectroscopy and Radiative Transfer
    View all citing articles on Scopus
    View full text