Search

The absorption spectrum of the O-18(3) isotopologue of ozone was recorded by cavity-ring-down spectroscopy between 7920 and 7985 cm(-1) with a routine noise equivalent absorption, alpha(min), on the order of a few 10(-11) cm(-1). This permitted to detect and analyze an extremely weak vibrational band assigned as 7 nu(1) + nu(3) following predictions from the ab initio potential energy surface reported in J Chem Phys 2013;139:134307. This is the most excited rovibrational band of the O-18(3) isotopologue detected so far. In total, 259 ro-vibrational transitions were measured and modeled using an effective Hamiltonian model. The vibrational dipole transition moment parameter of the band deduced from the analysis and the fit to 177 experimental line intensities was evaluated as 2.6 x 10(-6) Debye. The absorption coefficient of the recorded pure O-18(3) spectrum was found significantly smaller compared to that of the O-16(3) in the same range. The set of 171 upper state energies experimentally determined cover the range from 92.6% (at J = 1) to 97% (at J = 26) of the first dissociation threshold D-0. The comparison with ab initio predictions for the band origin and nu-dependent rotational constants is discussed. An absorption line list containing 477 transitions is provided as Supplementary material for the 7 nu(1) + nu(3) band. (C) 2021 Elsevier Ltd. All rights reserved.

This paper describes the 2020 release of the GEISA database (Gestion et Etude des Informations Spectroscopiques Atmosphériques: Management and Study of Atmospheric Spectroscopic Information), developed and maintained at LMD since 1974. GEISA is the reference database for several current or planned Thermal and Short-Wave InfraRed (TIR and SWIR) space missions IASI (Infrared Atmospheric Sounding Interferometer), IASI-NG (IASI New Generation), MicroCarb (Carbon Dioxide Monitoring Mission), Merlin (MEthane Remote sensing LIdar missioN). It is actually a compilation of three databases: the “line parameters database”, the “cross-section sub-database” and the “microphysical and optical properties of atmospheric aerosols sub-database”. The new edition concerns only the line parameters dataset, with significant updates and additions implemented using the best available spectroscopic data. The GEISA-2020 line parameters database involves 58 molecules (145 isotopic species) and contains 6,746,987 entries, in the spectral range from 10−6 to 35877 cm−1. In this version, 23 molecules have been updated (with 10 new isotopic species) and 6 new molecules have been added (HONO, COFCl, CH3F, CH3I, RuO4, H2C3H2 (isomer of C3H4)) corresponding to 15 isotopic species. The compilation can be accessed through the AERIS data and services center for the atmosphere website (https://geisa.aeris-data.fr/), with the development of a powerful graphical tool and convenient searching, filtering, and plotting of data using modern technologies (PostgreSQL database, REST API, VueJS, Plotly). Based on four examples (H2O, O3, O2 and SF6), this paper also shows how the LMD in house validation algorithm SPARTE (Spectroscopic Parameters And Radiative Transfer Evaluation) helps to evaluate, correct, reject or defer the input of new spectroscopic data into GEISA and this, thanks to iterations with researchers from different communities (spectroscopy, radiative transfer).

This work continues the investigation of high resolution infrared spectra of ¹⁷ O and ¹⁸ O enriched ozone isotopomers in the 10 µm spectral range. Spectra are recorded by the Reims Fourier Transform Spectrometer. In a mixture of 18 isotopic species of ozone including ¹⁷ O, ¹⁶ O and ¹⁸ O oxygen atoms, two ν 3 bands are analyzed for the first time: those of ¹⁷ O ¹⁷ O ¹⁸ O and ¹⁷ O ¹⁸ O ¹⁷ O species. The analyses, performed accounting for interactions between the (100)and (001)states, permitted to assign 769 transitions for ¹⁷ O ¹⁸ O ¹⁷ O and 1290 for ¹⁷ O ¹⁷ O ¹⁸ O. The resulting sets of effective Hamiltonian parameters allow reproducing observed transitions within experimental accuracies and calculating synthetic spectra in excellent agreement with the observed ones. We also report here the comparison of main fitted parameters with recent theoretical predictions.

Internal vibrational and rotational partition functions are calculated for nine molecules that are abundant in the terrestrial atmosphere and some are also observed in planetary and exoplanet atmospheres. These molecules are often observed in non-local thermodynamic equilibrium (NLTE) conditions. Calculations are made for all isotopologues of these molecules for which data are available, generally for the temperature range T = 1–5000 K. The methods of calculation of the vibrational and rotational partition functions, Qv and Qr, respectively, the convergence, and the uncertainties for each isotopologue are discussed. Finally, a FORTRAN program, NLTE_TIPS_2021.FOR, to recall the NLTE partition sums is discussed.

Thirteen bands of the CS symmetry 16O18O18O ozone isotopic species were observed in spectra generated from 16O2/18O2 mixtures recorded between 950 and 3500 cm−1. Experimentally and theoretically this work is similar to the one on the 16O16O18O isotopologue analysis, recently published [Barbe A, et al. J. Quant. Spectrosc. Radiat. Transfer, 2018;218;231–247]. In total 11402 rovibrational transitions (corresponding to 8784 upper state energy levels) were assigned. The line positions are modelled with the root mean squares deviations varying from 0.001 to 0.002 cm−1. The final comparisons show a good agreement between experimental and theoretical band centres and rotational constants predicted from ab initio potential energy surface. Effective dipole transition moment parameters were either fitted or estimated from a comparison between observed and simulated spectra. The line lists built using effective Hamiltonian and transition moment parameters containing 61194 transitions are provided in Supplementary Materials.

Ozone formation and depletion play a key role in various physical and chemical atmospheric processes which remain to be understood in more detail. The modeling of such phenomena based on ozone dynamics calculations requires a detailed knowledge of quantum states, transition probabilities, and ozone potential energy surface (PES) near the dissociation threshold ${D}_{0}$. Knowledge of highly excited rovibrational states with energies approaching ${D}_{0}$ is necessary for a reliable interpretation of satellite measurements of the ozone absorption or emission in the upper atmosphere in nonlocal thermodynamic equilibrium conditions. In addition, these states provide an ideal probe of PESs computed by ab initio methods but their detection by absorption spectroscopy is particularly challenging due to the sharp decrease of the intensity with the energy of the corresponding vibrational combination and overtone bands. In this work, two very high-energy vibrational bands are detected by high sensitivity cavity ring-down spectroscopy providing a noise equivalent absorption ${\ensuremath{\alpha}}_{\mathrm{min}}$ on the order of a few ${10}^{\ensuremath{-}11}\phantom{\rule{0.16em}{0ex}}\mathrm{c}{\mathrm{m}}^{\ensuremath{-}1}$. The corresponding cold bands, assigned as ${\ensuremath{\nu}}_{1}+6{\ensuremath{\nu}}_{2}+3{\ensuremath{\nu}}_{3}$ and $6{\ensuremath{\nu}}_{1}+{\ensuremath{\nu}}_{2}+{\ensuremath{\nu}}_{3}$ centered at 7969 and $7993\phantom{\rule{0.16em}{0ex}}\mathrm{c}{\mathrm{m}}^{\ensuremath{-}1}$, respectively, are about nine orders of magnitude less intense than the well-known ${\ensuremath{\nu}}_{3}$ fundamental band near 10 \textmu{}m. The spectrum analysis allowed for the experimental determination of 240 rovibrational energy levels of the upper vibrational states. The corresponding rotational patterns, located between 93.1% and 96.7% of the dissociation threshold, are the most excited ones measured so far by absorption spectroscopy. The values of the band centers and rotational constants derived from the spectrum analysis provide further confirmation of the absence of an activation barrier on the minimum-energy path towards the dissociation threshold [Tyuterev et al., Phys. Rev. Lett. 113, 143002 (2014)]. The vibrational dynamics is discussed in the context of the radiative transition probabilities and related Einstein coefficients for the observed bands. Another important question in the context of the nuclear dynamics concerns the interaction among the three identical potential wells of the ozone molecule in the ground electronic state, which appears to be due to the Jahn-Teller effect. Recent theoretical results accounting for the three potential wells in a full symmetry approach predicted that vibration states should exhibit deviations from the conventional one-potential-well ab initio calculations, near the dissociation threshold. In the present work, we observe a better agreement of the experimentally determined energy value with the three-wells prediction for the (1,6,3) vibrational state involving a large simultaneous excitation of both bending (${\ensuremath{\upsilon}}_{2}=6$) and asymmetric stretch (${\ensuremath{\upsilon}}_{3}=3$) vibrations pointing towards a nearby well. This could be tentatively interpreted as a manifestation of the three-wells effect.

Fifteen bands of the most abundant asymmetric ozone isotopomer 16O16O18O were observed in new spectra generated from various 18O2/16O2 mixtures in the 950–3850 cm−1 range. These bands were fully analyzed, leading to the assignment of a total of 9976 transitions, corresponding to 7030 upper state rovibrational energy levels. The analyses include also six “dark” vibrational states perturbing upper states of the observed bands through anharmonic and Coriolis resonance coupling. The modeling of very congested spectra simultaneously containing lines of six isotopic species resulted to the root mean squares deviations that vary from 0.001 to 0.003 cm−1 in line positions, not exceeding the experimental accuracy. To this end, two different approaches both guided by the ab initio theory were employed. Below 2500 cm−1, all the interacting vibrational states involved in the usual polyad scheme corresponding to the same total stretching excitation were observed. In these cases all symmetry allowed Coriolis and anharmonic coupling terms were initially predicted from ab initio potential energy surface (PES) using the contact transformation method. The major coupling terms were then held fixed to theoretical values during the least-square fits of the observed transitions in order to avoid correlations among adjusted parameters. Above 2500 cm−1, some bands were too weak to be observed. In these cases, the observations did not permit including complete stretching polyads and we used truncated effective models, for which the theory helped to predict initial values of the band centers, their assignments and vibrational dependence of Av, Bv, Cv rotational constants. The final comparison of fifteen experimentally detected and ab initio band centers of 16O16O18O gave the rms deviation of 0.59 cm−1. This drops down to 0.09 cm−1 when using the PES that had been empirically optimized to 16O3 data. The (observed-predicted) rms deviation for Av and Bv is very small - about 0.1%. Average deviations for Cv are somewhat larger ∼0.8% as they are affected by Coriolis resonances. Effective dipole transition moment parameters were either fitted or estimated from a qualitative comparison of observed and simulated spectra involving overlapping contributions of various isotopic species. The line lists built from data reduction models containing about 51,100 transitions are provided in Supplementary Materials.

The paper presents the results of the analysis of experimental FTS infrared vibration-rotation spectra of 16O3 ozone isotopologue recorded in Reims University in the range from 900 to 5600 cm−1, which was used for the current version of the SM http://smpo.univ-reims.fr). The majority of the reported data is related to the accurate determination of line positions. New parameters of effective Hamiltonian (EH) models for three polyads of vibrational states {(001), (100)}, {(011), (110)} and {(002), (101), (030), (200)} obtained from accurate fits of observed spectra in 900 – 2200 cm−1 range, have also been described. Further, at the higher wavenumber range 2700 – 5600 cm−1, various perturbations in the observed spectra due to accidental ro-vibrational resonances and which have been at the root of a lack of accuracy in certain regions of the previously published line lists were identified. We have applied empirical corrections to line positions and to corresponding energy levels in cases where the EH models did not provide the target accuracy of 10−3 cm−1 given of more complex polyad structures, which in turn allowed the use of corrected energy levels to compute improved line positions for the hot bands. In terms of line intensities, we publish here the effective dipole moment parameters and the line lists with empirically-determined absolute intensities for the ν1/ν3 dyad which are consistent with recent ab initio calculations for the strongest bands. For the higher wavenumber range, the empirical corrections are applied for relative intensities of some series of individual transitions in order to improve the simulation of observed spectra. A significant part of the reported data is included in the 2020 release of the HITRAN database.

The spectra of ozone, enriched by 17O oxygen, were recorded and reinvestigated in the 5 μm and 10 μm regions using the Reims Fourier Transform Spectrometer. Here we focus on the analysis of the ν1, ν3 and ν1+ν3 bands of the 16O17O16O isotopologue, which were modeled simultaneously to improve the parameters of the ground state energy level. The analysis of the ν1+ν3 bands of 16O16O17O is also extended. We report here the new system of effective Hamiltonian parameters, give some examples of comparison between observed and simulated spectra, and of the improvement for the ro-vibrational transition data available in spectroscopic databases.

Accurate knowledge of line parameters, and particularly line intensities is of primary importance for satellite, ground based and balloon measurements of the ozone concentration in the atmosphere. Self-consistency in absolute band intensities of ozone in various spectral intervals is necessary to avoid discrepancies in the atmospheric ozone retrieval from different windows. Difficulties in precise experimental characterization of absolute line intensities of this unstable molecule are well known leading to many significant deviations in published intensity values and in spectroscopic compilations. Ab initio intensities computed from the dipole moment functions of our previous works (Tyuterev et al. JCP 2017;146:064304; JCP 2019;150:184303) have permitted achieving a consistency between Stark-effect measurements and various databases in 10 and 5 µm in a good agreement with the most accurate recent experiments. In this paper we extend ab initio corrections to available line lists of ozone band intensities in wider IR range up to the bands with four vibrational quanta. About 40% of intensities were corrected in the 0 - 4122 cm−1 range for thirty cold and hot bands of the main ozone 16O3 isotopologue. Most of them, but not all, were obtained using band specific scaling factors ranging from 0.75 to 1.17 making quite significant changes in the absolute values. The full line lists containing corrected line intensities in the spectral intervals considered in the paper are provided in the Supplementary material for this work. The resulting line list is now available on the web site of the “Spectroscopy and Molecular Properties of Ozone” (SM http://smpo.univ-reims.fr ). A large part of this line list including the ab initio correction reported in this work will be used in the future HITRAN20 update.

International audience; Reliable ozone spectral data consistent over several spectral ranges are a challenge for both experiment and theory. We present ab initio calculations for strong lines that lead to consistent results from the microwave to mid-infrared regions. The results agree well with established microwave line lists and our new measurements in the fundamental and first overtone regions of ozone at 5 and 10 μm. The calculations and their agreement to within 1% with measurements provide an important step toward consistent and accurate spectroscopic ozone data. The results imply that actual databases need to be corrected by about 3% in the corresponding mid-infrared spectral intervals. Appropriate recommendations for the consistency of strong line intensities of the ozone molecule in microwave, 10 and 5 μm ranges, in HITRAN (HIgh-resolution TRANsmittance and molecular apbsorption) and GEISA (Gestion et Etude des Informations Spectroscopiques Atmosphériques) databases are suggested.I. INTRODUCTION

In a mixture of 18 isotopic species of ozone including 17O, 16O and 18O oxygen atoms, four ν1+ν3 bands of 17O3, 17O17O18O, 16O17O17O, 17O16O17O in the 5 µm range are analyzed for the first time. The analyses, accounting for interactions between (101) with (002) and (200) states permitted to assign respectively 747 transitions for 17O17O18O, 705 transitions for 17O16O17O, 889 transitions for 16O17O17O and 1404 transitions for 17O3. The results of fits of effective Hamiltonian and transition moment parameters permit reproducing observed transitions within experimental accuracy. They allow to calculate synthetic spectra in excellent agreement with the observed ones. The band centers and rotational constants derived from experimental spectra are in good agreement with recent theoretical predictions from potential energy surfaces.

The spectral range from 3600 to 4300 cm−1 of ozone was reinvestigated, leading to a large improvement of the analysis of the strongest ν1+3ν3 band, visible in atmospheric spectra. Recent progress, particularly in the theoretical predictions and observation of the hot band 4ν3-ν3 has allowed identify the strong anharmonic resonance with the dark 3ν1+ν2 band. As the (310) state is also in interaction with the (211), it was necessary to include seven vibration states to correctly reproduce 3389 transitions with the root mean square deviation of 3.54 × 10−3 cm−1 for line positions and 1295 intensities within 8.6%. We report the Effective Hamiltonian parameters and the band transition moments with the corresponding statistics as well as two line lists of the vibration-rotation transitions, and associated figures of the comparisons between observed and calculated spectra. The first line list containing 15,731 transitions is directly obtained from the fitted parameters. In the second one, the calculated line positions were replaced by the observed ones, for the cases where the (νobs - νcalc) differences were larger than the experimental uncertainty. New improved information on the energy levels and transitions can be used to extend the non-LTE models of the atmospheric ozone towards the energy range above 3500 cm−1.

Three very weak bands of the 18O16O18O isotopologue of ozone have been detected by high sensitivity CW-Cavity Ring Down Spectroscopy between 6340 and 6800 cm−1. They are vibrationally assigned as 2ν1+5ν3, ν1+4ν2+3ν3 and 3ν2+5ν3 and correspond to the highest frequency bands of this isotopologue detected so far. A total of 464, 318 and 194 transitions were rovibrationally assigned, respectively. The good agreement with theoretical values achieved for the derived band centres and rotational constants confirms the accuracy of the potential energy surface recently obtained via extensive ab initio calculations. A set of line intensities was measured and fitted to derive the first transition moment parameter of the three bands. The determined sets of effective Hamiltonian parameter and transition moment operators, as well as the experimental energy levels, were used to generate a complete list of 1526 transitions, provided as Supplementary materials . The calculated line-list allows generating a synthetic spectrum which reproduces satisfactorily the experimental spectrum.

This work continues the systematic investigation of high resolution infrared spectra of 18O enriched ozone isotopologues in the 4500–6300 cm−1 spectral range. So far, the 16O18O16O and 18O16O18O spectra have been recorded and analysed using Fourier Transform Spectrometer (FTS) below 5000 cm−1 and using high sensitivity Cavity Ring-Down Spectroscopy above 6000 cm−1. In order to fill the gap between 5000 and 6000 cm−1, new infrared spectra were recorded with the Reims FTS. The analyses of the 2ν1+ν2+3ν3 band of 16O18O16O and of the 5ν3, ν1+2ν2+3ν3 and ν1+5ν3 bands of 18O16O18O are reported. New ν1+2ν2+3ν3 and ν1+5ν3 bands of 18O3 were also observed and analysed. Observed line positions were fitted using effective Hamiltonian models accounting for the dark state perturbations. The derived band centres and rotational constants are in a good agreement with new theoretical calculations from the molecular potential function.

Spectroscopy and Molecular Properties of Ozone (“SM spectroscopic applications (analysis, Fourier transform spectroscopy, atmospheric optics, optical standards, spectroscopic atlases); applications to environment studies and atmospheric physics (remote sensing); data supply for specific databases; and to photochemistry (laser excitation, multiphoton processes). The system is accessible via Internet on two sites: http://smpo.iao.ru and http://smpo.univ-reims.fr .

International audience; Our systematic investigation of the high sensitivity CW-Cavity Ring Down Spectra of 16O/18O ozone isotopologues at high vibrational excitation continues with the study of the 16O18O18O and 18O16O18O species. The first two papers of this series were devoted to the analysis of the same four bands of the 16O16O18O and 16O18O16O species in the 5930-6340 cm-1domain. Here, after a brief reminder of relevant experiment and theory, we report the analyses of two bands of 16O18O18O, vibrationally assigned as 2ν2+5ν3 and 2ν1+2ν2+3ν3_2 and three bands of 18O16O18O, assigned to 2ν1+2ν2+3ν3_1, 5ν1+ν3 and 3ν1+ν2+3ν3.They correspond to the highest vibration excitations observed so far for the 16O18O18O and 18O16O18O isotopologues. Altogether for the two new bands of 16O18O18O, 1214 rovibrational transitions were assigned up to Jmax=29 and for the three new bands of 18O16O18O, 948 rovibrational transitions were assigned up to Jmax=27. Observed line positions were fitted with root-mean squares deviations ranging from 0.005 to 0.011 cm-1, using effective Hamiltonian models accounting for dark state perturbations. The derived band centres and rotational constants are in good agreement with new theoretical calculations from the molecular potential function. The corresponding lists of 3365 lines are provided as Supplementary material.

This paper presents an extensive intercomparison and validation for the ozone (O3) product measured by the two Infrared Atmospheric Sounding Interferometers (IASIs) launched on board the MetOp-A and MetOp-B satellites in 2006 and in 2012 respectively. IASI O3 total columns and vertical profiles obtained from Fast Optimal Retrievals on Layers for IASI (FORLI) v20140922 software (running up until recently) are validated against independent observations during the period 2008-2014 on a global scale. On average for the period 2013-2014, IASI-A and IASI-B total ozone columns (TOCs) retrieved using FORLI are consistent, with IASI-B providing slightly lower values with a global difference of only 0.2±0.8%. The comparison between IASI-A and IASI-B O3 vertical profiles shows differences within ±2% over the entire altitude range. Global validation results for 7 years of IASI TOCs from FORLI against the Global Ozone Monitoring Experiment-2 (GOME-2) launched on board MetOp-A and Brewer-Dobson data show that, on average, IASI overestimates the ultraviolet (UV) data by 5-6% with the largest differences found in the southern high latitudes. The comparison with UV-visible SAOZ (Système d'Analyse par Observation Zénithale) measurements shows a mean bias between IASI and SAOZ TOCs of 2-4% in the midlatitudes and tropics and 7% at the polar circle. Part of the discrepancies found at high latitudes can be attributed to the limited information content in the observations due to low brightness temperatures. The comparison with ozonesonde vertical profiles (limited to 30km) shows that on average IASI with FORLI processing underestimates O3 by ∼ 5-15% in the troposphere while it overestimates O3 by ∼ 10-40% in the stratosphere, depending on the latitude. The largest relative differences are found in the tropical tropopause region; this can be explained by the low O3 amounts leading to large relative errors. In this study, we also evaluate an updated version of FORLI-O3 retrieval software (v20151001), using look-up tables recalculated to cover a larger spectral range using the latest HITRAN spectroscopic database (HITRAN 2012) and implementing numerical corrections. The assessment of the new O3 product with the same set of observations as that used for the validation exercise shows a correction of ∼ 4% for the TOC positive bias when compared to the UV ground-based and satellite observations, bringing the overall global comparison to ∼ 1-2% on average. This improvement is mainly associated with a decrease in the retrieved O3 concentration in the middle stratosphere (above 30hPa/25km) as shown by the comparison with ozonesonde data., SCOPUS: ar.j, info:eu-repo/semantics/published

This paper presents an extensive inter-comparison and validation for the ozone (O3) product measured by the two Infrared Atmospheric Sounding Interferometers (IASI) launched onboard the Metop-A and Metop-B satellites in 2006 and in 2012, respectively. IASI O3 total columns and vertical profiles obtained from Fast Optimal Retrievals on Layers for IASI (FORLI-O3) v20140922 software (running up until recently) are validated against independent observations during the period 2008–2014 on a global scale. On average for the period 2013–2014, IASI-A and IASI-B TOCs retrieved using FORLI are consistent, with IASI-B providing slightly lower values with a global difference of only 0.2±0.8 %. The comparison between IASI-A and IASI-B O3 vertical profiles shows differences within ±2 % over the entire altitude range. Global validation results for seven years of IASI TOCs from FORLI against GOME-2/Metop-A, Dobson and Brewer data show that, on average, IASI overestimates the UV data by 5–6 % with the largest differences found in the Southern high latitudes. The comparison with UV-vis SAOZ measurements shows a mean bias between IASI and SAOZ TOCs of 2–4 % in the mid-latitudes and tropics, and 7 % at the polar circle. Part of the discrepancies found at high latitudes can be attributed to the limited information content in the observations, due to low brightness temperatures. The comparison with ozonesonde vertical profiles (limited to 30 km) shows that on average IASI with FORLI processing underestimates O3 by ~5–15 % in the troposphere while it overestimates O3 by ~10–40 % in the stratosphere depending on the latitude. In the Northern middle latitudes, the bias varies within ±20 % for the entire altitude range. The largest relative differences are found in the tropical tropopause region; this can be explained by the low O3 amounts leading to large relative errors. In this study, we also evaluate an updated version of FORLI-O3 retrieval software (v20151001), using look-up tables recalculated to cover a larger spectral range using the latest HITRAN spectroscopic database (HITRAN 2012), and implementing numerical corrections. The assessement of the new O3 product with the same set of observations as that used for the validation exercise shows a correction of ~4 % for the TOC positive bias when compared to the UV ground-based and satellite observations, bringing the overall global comparison to ~1–2 % on average. This improvement is mainly associated with a decrease in the retrieved O3 concentration in the stratosphere (above 30 hPa/25 km) as shown by the comparison with ozonesonde data.

The activity “Absorption Cross-Sections of Ozone” (ACSO) started in 2008 as a joint initiative of the International Ozone Commission (IO3C), the World Meteorological Organization (WMO) and the IGACO (“Integrated Global Atmospheric Chemistry Observations”) O3/UV subgroup to study, evaluate, and recommend the most suitable ozone absorption cross-section laboratory data to be used in atmospheric ozone measurements. The evaluation was basically restricted to ozone absorption cross-sections in the UV range with particular focus on the Huggins band. Up until now, the data of Bass and Paur published in 1985 (BP, 1985) are still officially recommended for such measurements. During the last decade it became obvious that BP (1985) cross-section data have deficits for use in advanced space-borne ozone measurements. At the same time, it was recognized that the origin of systematic differences in ground-based measurements of ozone required further investigation, in particular whether the BP (1985) cross-section data might contribute to these differences. In ACSO, different sets of laboratory ozone absorption cross-section data (including their dependence on temperature) of the group of Reims (France) (Brion et al., 1993, 1998, 1992, 1995, abbreviated as BDM, 1995) and those of Serdyuchenko et al. (2014), and Gorshelev et al. (2014), (abbreviated as SER, 2014) were examined for use in atmospheric ozone measurements in the Huggins band. In conclusion, ACSO recommends: (a) The spectroscopic data of BP (1985) should no longer be used for retrieval of atmospheric ozone measurements. (b) For retrieval of ground-based instruments of total ozone and ozone profile measurements by the Umkehr method performed by Brewer and Dobson instruments data of SER (2014) are recommended to be used. When SER (2014) is used, the difference between total ozone measurements of Brewer and Dobson instruments are very small and the difference between Dobson measurements at AD and CD wavelength pairs are diminished. (c) For ground-based Light Detection and Ranging (LIDAR) measurements the use of BDM (1995) or SER (2014) is recommended. (d) For satellite retrieval the presently widely used data of BDM (1995) should be used because SER (2014) seems less suitable for retrievals that use wavelengths close to 300 nm due to a deficiency in the signal-to-noise ratio in the SER (2014) dataset. The work of ACSO also showed: • The need to continue laboratory cross-section measurements of ozone of highest quality. The importance of careful characterization of the uncertainties of the laboratory measurements. • The need to extend the scope of such studies to other wavelength ranges (particularly to cover not only the Huggins band but also the comparison with the mid-infrared region). • The need for regular cooperation of experts in spectral laboratory measurements and specialists in atmospheric (ozone) measurements.

The GEISA database (Gestion et Etude des Informations Spectroscopiques Atmosphériques: Management and Study of Atmospheric Spectroscopic Information) has been developed and maintained by the ARA/ABC(t) group at LMD since 1974. GEISA is constantly evolving, taking into account the best available spectroscopic data. This paper presents the 2015 release of GEISA (GEISA-2015), which updates the last edition of 2011 and celebrates the 40th anniversary of the database. Significant updates and additions have been implemented in the three following independent databases of GEISA. The “line parameters database” contains 52 molecular species (118 isotopologues) and transitions in the spectral range from 10−6 to 35,877.031 cm−1, representing 5,067,351 entries, against 3,794,297 in GEISA-2011. Among the previously existing molecules, 20 molecular species have been updated. A new molecule (SO3) has been added. HDO, isotopologue of H2O, is now identified as an independent molecular species. Seven new isotopologues have been added to the GEISA-2015 database. The “cross section sub-database” has been enriched by the addition of 43 new molecular species in its infrared part, 4 molecules (ethane, propane, acetone, acetonitrile) are also updated; they represent 3% of the update. A new section is added, in the near-infrared spectral region, involving 7 molecular species: CH3CN, CH3I, CH3O2, H2CO, HO2, HONO, NH3. The “microphysical and optical properties of atmospheric aerosols sub-database” has been updated for the first time since 2003. It contains more than 40 species originating from NCAR and 20 from the ARIA archive of Oxford University. As for the previous versions, this new release of GEISA and associated management software facilities are implemented and freely accessible on the AERIS/ESPRI atmospheric chemistry data center website., SCOPUS: ar.j, info:eu-repo/semantics/published

The infrared spectrum of 16O3 has been revisited in the ranges 3300–3600 cm−1 and 5200–5600 cm−1, recorded by the Fourier Transform Spectrometer of Reims, with an improved signal/noise ratio. In the lower spectral range the weak 2ν1+2ν2 band is observed and assigned for the first time. This allowed for completing the triad of strongly interacting (220), (121), and (022) states. The resonance interaction parameters for this triad were derived from the molecular potential energy surface using high-order Contact Transformation method. This involves Darling–Dennison resonances between (220) and (022) vibration A-states, and Coriolis resonances between (121) B-state and A-states. Altogether 18 coupling parameters were fixed to these theoretically predicted values that allowed avoiding problems of ambiguities while fitting strongly correlated parameters. The remaining small perturbations of a few Ka=5 transitions of the ν1+2ν2+ν3 band were explained by accidental Coriolis resonance of (121) with the (050) dark state. This mixed half theoretical/half empirical model (with 39 fitted and 77 theoretically constrained parameters) developed in this work for the first time allows an excellent description of 1897 line positions with the rms deviation ∼0.001 cm−1 closed to the experimental precision. The intensities of two new hot bands ν2+4ν3−ν3 and ν1+ν2+3ν3−ν1 that fall in this range were also observed and modelled, thanks to a better signal to noise ratio. In addition, the 5200–5600 cm−1 region has also been revisited, showing for the first time the (321) state which was considered as “dark” one in the previous analysis. Effective Hamiltonian parameters were derived for strongly coupled (213)/(114)/(321) upper states including (080) as a dark one. The transition moment parameters of all newly observed and assigned bands were derived resulting to average rms deviation for intensities of ∼8% and 13% for the cold and the hot bands, respectively, in the lower range and of ∼15% for the bands in the upper range that is close to experimental accuracy for weak transitions.

The absorption spectrum of the 16 O 3 isotopologue of ozone has been recorded in the 7000–7920 cm −1 region by high sensitivity CW-Cavity Ring Down Spectroscopy. This report is devoted to the analyses of the 7065–7300 cm −1 region dominated by the ν 1 + 2 ν 2 + 5 ν 3 and ν 1 + 5 ν 2 + 3 ν 3 A-type bands at 7130.8 and 7286.8 cm −1 respectively. 289 transitions were assigned to the ν 1 + 2 ν 2 + 5 ν 3 band. The corresponding line positions were modeled with an effective Hamiltonian involving Coriolis resonance interactions between the (1 2 5) upper state and the (4 4 0), (0 2 6) and (6 1 0) dark states, and an anharmonic resonance interaction with the (2 0 5) state. The very strong interaction (up to 50% mixing of the wavefunctions) between the (1 2 5) and (6 1 0) states leads to the observation of two extra lines of the 6 ν 1 + ν 2 band due to a resonance intensity transfer. 213 transitions of the ν 1 + 5 ν 2 + 3 ν 3 band were assigned and modeled taking into account a Coriolis resonance interaction with the (3 6 0) state. We take the opportunity of the present work to report the analysis of the very weak 4 ν 2 + 4 ν 3 B-type band at 6506.1 cm −1 which was assigned from previously recorded CRDS spectra. 286 transitions were modeled using the effective Hamiltonian approach. The dipole transition moment parameters of the three analyzed bands were determined by a least-squares fit to the measured line intensities. For the three studied band systems, the effective Hamiltonian and transition moment operator parameters were used to generate line lists provided as Supplementary Materials .

Infrared high resolution spectra of ozone 16O3 have been recorded in the 10 and 5 μm spectral ranges in order to derive their absolute intensities with a best achievable accuracy. Spectra have been recorded with the home made stepping mode FTS of GSMA (Reims). In the two spectral regions, we use UV–IR crossed beam cell. The quantification of ozone is achieved using UV cross section at 253.65 nm. A check of this UV calibration is also performed using direct pressure measurements of quasi-pure samples of O3. The intensities are derived from infrared spectra using multifit procedure already tested. In the 10 μm range, where 12 different spectra have been recorded, 65 well selected lines led to a good agreement (better than 0.3%) with the HITRAN 2008 (or 2004) values, confirming our previous work [De Backer-Barilly MR, Barbe A. Absolute intensities of the 10 μm bands of 16O3. J Mol Spectrosc 305:2001;43–53]. In the 5 μm range, where 18 transitions are selected, we also note a correct agreement with HITRAN 2008, despite a slightly larger averaged value between (1.9%) experimental and theoretical (HITRAN). As conclusion, authors suggest the use of current HITRAN 2008 data for atmospheric retrievals.

International audience; This paper is devoted to the third part of the analysis of the very weak absorption spectrum of the 18O 3 isotopologue of ozone recorded by CW-Cavity Ring Down Spectroscopy between 5930 and 6900 cm -1. In the two first parts [A. Campargue, A. Liu, S. Kassi, D. Romanini, M.-R. De Backer-Barilly, A. Barbe, E. Starikova, S.A. Tashkun, Vl.G. Tyuterev, J. Mol. Spectrosc. (2009), doi: 10.1016/j.jms.2009.02.012 and E. Starikova, M.-R. De Backer-Barilly, A. Barbe, Vl.G. Tyuterev, A. Campargue, A.W.Liu, S. Kassi, J. Mol. Spectrosc. (2009) doi: 10.1016/j.jms.2009.03.013], the effective operators approach was used to model the spectrum in the 6200-6400 and 5930-6080 cm -1 regions, respectively. The analysis of the whole investigated region is completed by the present investigation of the 6490-6900 cm -1 upper range. Three sets of interacting states have been treated separately. The first one falls in the 6490-6700 cm -1 region, where 1555 rovibrational transitions were assigned to three A-type bands: 3 ν2 + 5 ν3, 5 ν1 + ν2 + ν3 and 2 ν1 + 3 ν2 + 3 ν3 and one B-type band: ν1 + 3 ν2 + 4 ν3. The corresponding line positions were reproduced with an rms deviation of 18.4 × 10 -3 cm -1 by using an effective Hamiltonian (EH) model involving eight vibrational states coupled by resonance interactions. In the highest spectral region - 6700-6900 cm -1 - 389 and 183 transitions have been assigned to the ν1 + 2 ν2 + 5 ν3 and 4 ν1 + 3 ν2 + ν3 A-type bands, respectively. These very weak bands correspond to the most excited upper vibrational states observed so far in ozone. The line positions of the ν1 + 2 ν2 + 5 ν3 band were reproduced with an rms deviation of 7.3 × 10 -3 cm -1 by using an EH involving the {(054), (026), (125)} interacting states. The coupling of the (431) upper state with the (502) dark state was needed to account for the observed line positions of the 4 ν1 + 3 ν2 + ν3 band ( rms = 5.7 × 10 -3 cm -1). The dipole transition moment parameters were determined for the different observed bands. The obtained set of parameters and the experimentally determined energy levels were used to generate a complete line list provided as Supplementary Materials. The results of the analyses of the whole 5930-6900 cm -1 spectral region were gathered and used for a comparison of the band centres to their calculated values. The agreement achieved for both 18O 3 and 16O 3 (average difference on the order of 1 cm -1) indicates that the used potential energy surface provides accurate predictions up to a vibrational excitation approaching 80% of the dissociation energy. The comparison of the 18O 3 and 16O 3 band intensities is also discussed, opening a field of questions concerning the variation of the dipole moments and resonance intensity borrowing by isotopic substitution.

More than 5050 rovibrational transitions were assigned in the very sensitive CW-CRDS 18 O 3 spectra in the 5900–7000 cm −1 range leading to the first determination of 14 new vibrational band centres. Three bands correspond to the highest ozone states observed so far under high resolution. The good agreement between predicted and observed centres ( rms deviation of 1.1 cm −1 ) confirms the accuracy of the potential function previously determined from 16 O 3 data. At this energy range situated ∼20% below the dissociation energy, the small changes of masses associated with the 16 O 3 → 18 O 3 homogeneous isotopic substitution result in irregular isotopic shifts.

The absorption spectrum of SO 2 has been studied with 2 MHz instrumental resolution in the wavenumber range 1315 – 1395 cm - 1 , using a difference frequency spectrometer based on diode lasers for the 1300 and 1550 nm telecom bands. Initially, Lorentz broadening parameters are determined, at pressures up to 400 Pa, for 545 lines covering lower state J up to 60 and K a up to 14. Using this information in conjunction with spectra recorded at 100 and 200 Pa the line strength is subsequently determined for 395 lines of the ( 0 0 0 ) → ( 0 0 1 ) band of 32 SO 2 , 170 lines from the ( 0 1 0 ) → ( 0 1 1 ) hot band of 32 SO 2 , and 98 lines from the ( 0 0 0 ) → ( 0 0 1 ) band of 34 SO 2 , and dipole moments are derived for all three bands. For the hot band, line positions of 178 transitions are measured and merged with previously published data to produce an improved set of Hamiltonian parameters for the ( 0 1 1 ) state, using the most recent model for ( 0 1 0 ) .

SO2 spectra have been recorded with a quantum cascade laser spectrometer at 9 μ m . Line positions, line strengths and self-broadening coefficients of 42 lines have been precisely determined. All these parameters have shown some discrepancies with those of the HITRAN database. Moreover several lines found in this study were not present in this database. In order to have a complete line list of transitions of SO2 appearing into the recorded spectra, we have performed a new calculation of the ν 1 band of 32SO2, the ν 1 + ν 2 - ν 2 band of 32SO2 and the ν 1 band of 34SO2. This new calculation demonstrates a considerable improvement of the calculations of synthetic spectra of SO2. These new calculations improve the positions (up to 0.03 cm - 1 ) , the intensities (up to 30%) and the self-broadening coefficients (up to 100%) of the ν 1 band of 32SO2 with respect to HITRAN database and permit to obtain a complete set of data for the ν 1 + ν 2 - ν 2 band of 32SO2 and the ν 1 band of 34SO2 where the positions and strengths are in very good agreement with experimental data.

The present contribution is devoted to the analysis of three extremely weak A-type bands of 16O3 recorded near 7686, 7739 and 7860 cm−1 i.e. only a few % below the dissociation limit, D0, at about 8560 cm−1. They correspond to the most excited vibration–rotation states of ozone observed so far via high-resolution absorption spectroscopy. They were detected by high sensitivity CW-cavity ring down spectroscopy with a typical noise equivalent absorption, αmin, on the order of 2×10−10 cm−1 and modelled using the effective operator approach. The derived band centres and rotational constants show a very good agreement with recent theoretical predictions based on a new ab initio potential energy surface.

The absorption spectrum of the 18 O 3 isotopologue of ozone is recorded by CW-Cavity Ring Down Spectroscopy between 7400 and 7920 cm −1 . The typical noise equivalent absorption of the recordings is α min ≈1×10 −10 cm −1 allowing for the detection of extremely weak vibrational bands located only a few % below the dissociation limit. The absorption spectrum also includes the weak 3 A 2 (000)–X(110) hot vibronic band at 7877.53 cm −1 affected by predissociation broadening. Overall, about 1000 lines were rovibrationnally assigned to six vibrational bands. Their line positions were modelled using effective Hamiltonian models covering three vibrational interacting systems and involving several “dark” states perturbing observed transitions. Line positions could be reproduced with rms deviations on the order of 0.01 cm −1 . The dipole transition moment parameters of the six bands were determined from the retrieved line intensities. The obtained set of effective parameters and the experimentally determined energy levels were used to generate a list of 1461 transitions provided as Supplementary materials .

The absorption spectrum of 16 O 3 has been recorded between 6030 and 6130 cm −1 by Fourier Transform Spectroscopy (GSMA, Reims) and cw-cavity ringdown spectroscopy (LSP, Grenoble). The two new bands 3 ν 1 +3 ν 3 and 2 ν 2 +5 ν 3 centered at 6063.923 and 6124.304 cm −1 , respectively are observed and analyzed. Rovibrational transitions with J and K a values up to 40 and 10, respectively, could be assigned. The rovibrational fitting of the observed energy levels shows that some rotational levels of the (303) and (025) bright states are perturbed by interaction with the (232), (510) and (124) dark states. The observed energy levels could be reproduced with a rms deviation of 5×10 −3 cm −1 using a global analysis based on an effective Hamiltonian including the five interacting states. The energy values of the three dark vibrational states provided by the fit are found in good agreement with theoretical predictions. The parameters of the resulting effective Hamiltonian and of the transition moment operator retrieved from the measured absolute line intensities allowed calculating a complete line list of 2035 transitions, available as Supplementary Material. The integrated band strengths are estimated to be 1.22×10 −24 and 3.15×10 −24 cm −1 /(mol cm −2 ) at 296 K for the 3 ν 1 +3 ν 3 and 2 ν 2 +5 ν 3 bands, respectively. A realistic error for these band strengths is 15% (see text).

A set of previously unpredicted 16O3 strong absorption lines in the 5 μ m region are identified for the first time in balloon-borne high-resolution ( 0.003 cm - 1 ) solar spectra. These spectral features are consistent with recent ozone laboratory spectra, and are interpreted by accidental resonance effects. Some of the lines are also observed in high-resolution ground-based spectra.

More than 400 absolute line positions and intensities of acetylene have been measured for the two cold bands ν 3 and ν 2 + ( ν 4 + ν 5 ) + 0 , and four strong hot bands occuring in the 3- μ m spectral region, from Fourier-transform spectra. Accuracies are on the average 0.0005 cm - 1 for line positions and 5% for line intensities. Vibrational transition dipole moment values have been deduced, as also empirical Herman–Wallis coefficients, and comparisons have been made with previous results.

This work presents, for the 1-0 band of CO, the measurements of air and H 2 -broadened line widths and H 2 -induced line shifts coefficients. These values were retrieved using a multispectrum fitting procedure on spectra recorded with the Fourier transform spectrometer at Reims. These results provide confirmation of other similar measurements previously published. In addition, we present the results for self and H 2 broadened line widths for the 2-0 band.

The 2000 HITRAN edition, with the updates of 2001, contains improved and new data on the acetylene molecule. The main changes concern the 13.6- and 7.5-μm spectral regions (improved line intensities), and the 5-μm region (previously absent from the database). These changes are reviewed, and the key problem of the validation of line intensities is dealt with. The status of the currently available line parameters is critically examined, and recommendations for future improvements are given.

Continuing the systematic study of ozone high-resolution infrared spectra, we present in this paper the measurements and analyses of line positions for the 18 O 16 O 18 O isotopomer. In the range 900–5000 cm −1 , corresponding to the observed spectra, 15 bands are analysed: ν 1 , ν 3 , ν 2 + ν 3 , ν 1 + ν 2 , 2 ν 3 , ν 1 + ν 3 , 2 ν 1 , ν 2 +2 ν 3 , ν 1 + ν 2 + ν 3 , 3 ν 3 , 2 ν 1 + ν 3 , ν 2 +3 ν 3 , ν 1 +3 ν 3 , ν 1 + ν 2 +3 ν 3 , and 5 ν 3 . As in the case of 16 O 3 , 18 O 3 , and 16 O 18 O 16 O, the analysis of these bands is performed using effective rovibrational Hamiltonians for nine polyads of interacting upper vibrational states. To correctly reproduce all observed transitions, we have to account for resonance perturbations due to 13 “Dark” states: (0 3 0), (0 4 0), (2 1 0), (0 3 1), (1 0 2), (0 4 1), (1 1 2), (3 1 0), (0 3 2), (0 0 4), (3 2 0), (0 1 4), and (0 4 2). We present the range of observed transitions, the results for spectroscopic parameters (vibrational energy levels, rotational and centrifugal distortion constants, and resonance coupling parameters), as well as the statistics for rovibrational energy levels, calculations and measurements. A comparison of observed band centres with those predicted from an isotopically invariant potential function is discussed. The RMS deviation between predicted and directly observed band centres is ≈0.03 cm −1 up to 3000 and ≈0.25 cm −1 for all 16 bands up to 5000 cm −1 .

The four 5v 3 bands of 18O enriched ozone have been observed and analysed for the first time. Two species (16O18O16O and 18O16O18O) belong to the C2v symmetry group and two other (18O18O16O and 16O16O18O) to the Cs symmetry group. They have been recorded at a resolution of 0.008 cm−1 with a pathlength of 32.16 m. Despite the very weak absorptions observed, almost 250 energy levels have been derived for each of the 4 species, with J ⩽ 35 and K a ⩽ 13, and suitable sets of Hamiltonian parameters have been determined. For 3 species it has been necessary to account for the resonance between the (005) and (311) states to correctly reproduce the spectra observed. These resonances, anharmonic for C2v, and hybrid (both anhar-monic and Coriolis) for Cs symmetry confirm the accidentally extremely strong coupling between the (005) and (311) states for 16O3, due in that case to the very close distance between unperturbed energy levels. This work also confirms the excellent prediction of band centres of these four spe...

A new set of spectroscopic constants of the 16 O 3 molecule ( ω i , x ij , y ijk , γ DD , α i X , β ij X ,…), which determine vibrational dependence of band centres and rotational parameters, is derived from recent accurate analysis of high-resolution experimental ro-vibrational spectra through the theoretical approach based on second-order perturbation expansions in normal coordinates accounting for Darling–Dennison resonance interactions. These values are used to update empirical values of anharmonic coefficients ( k ijl , k ijlm ) of the potential function expansion in normal coordinates. Quadratic f rr , f rα , f rr ′ , f αα as well as cubic f rst and quartic f rstl force constants in internal (bond lengths, bond angle) coordinates are also derived. A detailed discussion is devoted to the accuracy of parameter determination for each of four steps of calculations. It is emphasised that the conventional method based on the inversion of formulae of the perturbation theory gives the largest uncertainties at the last step of calculations: the determination of the anharmonic force field in internal coordinates.

The 2900–3400 cm −1 spectral range is revisited for an accurate determination of line positions and line intensities of the 3ν 3 , ν 1 +2ν 3 , 2ν 1 +ν 3 , and 3ν 1 bands of ozone. The fit on 4520 rotational energy levels of (012), (111), (210), (130), (003), (102), (201), and (300) vibrational states determined from observed transitions of cold and hot bands in the 2400–3400 cm −1 region with J max =65 and K a max =20 gives a r.m.s. =7×10 −4 cm −1 and provides a satisfactory agreement between calculated and observed line positions (dimensionless standard deviation is χ=1.44). The set of 2580 line intensities of 7 rovibrational bands has been measured and fitted with a r.m.s. =6.9% (χ=1.2), leading to the determination of transition moment parameters for these bands. Using these parameters we have obtained more precise estimation for the integrated band intensities S(3ν 3 )=1.41×10 −19 , S(ν 1 +2ν 3 )=1.28×10 −20 , S(2ν 1 +ν 3 )=7.91×10 −21 , S(3ν 1 )=4.72×10 −22 cm −1 /mol cm −2 at 296 K, with a cutoff 2×10 −26 cm −1 /mol cm −2 . The interactions of the tetrad (003)/(102)/(201)/(300) with the (130) state and the tetrad (040)/(012)/(111)/(210) are studied.

The line intensities of the ν3 band of OCS used for atmospheric retrievals, has been measured. The spectra have been recorded with the Reims Fourier transform spectrometer. Other bands of OCS 4ν 2 , ν 1 +ν 3 , 3ν 1 and 2ν1+2ν2 were recorded to obtain the line intensities. The results are compared to usual databanks (HITRAN or GEISA). From the measured line intensities, it has been possible to estimate the square of the vibrational dipole moment matrix element ( in debye 2 ) , equal to 0.1313 1 , 7.840 9 ×10 −4 , 1.158 1 ×10 −3 , 1.218 3 ×10 −5 , 1.192 3 ×10 −5 , respectively, for ν 3 , 4ν 2 , ν 1 +ν 3 , 3ν 1 and 2ν1+2ν2.

Since the discovery of anomalies in ozone isotope enrichment, several fundamental issues in the dynamics linked to the shape of the potential energy surface in the transition state region have been raised. The role of the reeflike structure on the minimum energy path is an intricate question previously discussed in the context of chemical experiments. In this Letter, we bring strong arguments in favor of the absence of a submerged barrier from ultrasensitive laser spectroscopy experiments combined with accurate predictions of highly excited vibrations up to nearly 95% of the dissociation threshold.

Using a Fourier transform spectrometer, we have recorded the spectra of the (18)O(3) species of ozone in the region 1300-3100 cm(-1), with a resolution of 0.003 cm(-1). The large product pathlength x pressure enable us to record 18 bands, 14 for the first time. The analysis has been performed using effective Hamiltonians for polyads of strongly interacting states for ozone, accounting for Coriolis and anharmonic resonances. The spectral parameters are derived for 16 vibrational states, including the two "dark" states (040) and (130). Various resonances are studied through the mixing coefficients of rovibrational wavefunctions. Systematic intensity measurements allow determination of transition moment parameters for 16 bands. Finally, a complete list of all transitions from 1300 to 3100 cm(-1), with cutoffs 10(-26) cm(-1)/mol cm(-2) (296 K), is calculated. Copyright 2001 Academic Press.

To improve the fit of spectroscopic parameters recorded by Fourier transform spectrometer (intensities, broadening coefficients, and pressure shift), we have developed a multispectrum fitting technique, this technique being already described by other research groups (Benner DC, Rinsland CP, Malathy Devi V, Smith MAH, Atkins D. JQSRT 1995;53:705; Carlotti, Appl Opt 1988;27:3250). We describe the algorithm of the software allowing to analyse and visualise several spectra simultaneously. To validate our software, line intensities (for N 2 O and CH 3 D) and self-broadening coefficients were fitted and compared with previous studies. We have also studied the wave numbers of N 2 O and compared them with heterodyne measurements in order to estimate the relative accuracy of our results.

Measured room temperature nitrogen-broadening coefficients are given for about 60 Λ-doublets belonging to the allowed sub-bands of the first overtone of nitric oxide, between 3620 and 3790 cm−1. Whereas a slight differential broadening (3.2% on the average) is observed between the two sub-bands, no significant vibrational dependence appears when one compares these values with those available for the fundamental band. Combining the N2-broadening coefficients with the O2-broadening coefficients recently published, the air-broadening required for atmospheric applications appears to be lower than the N2-broadening by 2%.

The infrared spectra of 16O3 in the range 1300–1500 cm−1 have been recorded at high resolution (0.007 cm−1), with a large product p×l=32.4 Torr×3212 cm, and a good S/N ratio≅500. This enabled us to observed the 3ν2−ν2 band and to extend the measurements of the 2ν2 band (A. Barbe, A. Chichery, Vl.G. Tyuterev, S. Taskhun, S. Mikhailenko, Spectrochimica Acta 54A (1998) 1935–1945). Two difference bands 2ν3–ν2 and ν1+ν3−ν2 were observed as well. The positions of the lines are well known due to the studies of cold bands 2ν1, ν1+ν3, 2ν3 as well as ν2. In this region only intensity measurements of the ν1+ν3−ν2 combination band have been done so far. Here, due to the large pressure×pathlength product, we are able to observe a large range of transitions for ν1+ν3−ν2 band (up to J=51 and Ka=14), and also for the first time 2ν3−ν2 band. We have derived the transition moment parameters for 2ν3−ν2 and estimated an upper limit for 2ν1−ν2. A complete final calculation is performed for these difference bands. The principal transition moment parameters are compared with theoretical predictions.

The effective ground state potential energy function of the ozone molecule near the C(2v) equilibrium configuration was obtained in a least-squares fit to the largest sample of experimental, high-resolution vibration-rotation data used for this purpose so far. The fitting is based on variational calculations carried out with the extended Morse Oscillator Rigid Bender Internal Dynamics model. The potential function is expanded in Morse-type functions of the stretching variables and in cosine of the bending angle. The present calculation produces results in significantly better agreement with experiment than previous determinations of the potential energy surface, and the energies predicted with the new surface are sufficiently accurate to be useful for the assignment of new high-resolution spectra. The rms (root-mean-square) deviation of the fit of rovibrational data up to J = 5 is 0.02 cm(-1). For the set of all 60 band centers of the (16)O(3) molecule included in the Atlas of Ozone Line Parameters, the rms deviation is 0.025 cm(-1), and for all band centers determined so far from high-resolution spectra, including those recently observed and assigned in Reims corresponding to highly excited stretching and bending vibrations (v(1) + v(2) + v(3) = 6), the rms deviation is 0.1 cm(-1). The "dark states" that produce resonance perturbations in the observed bands are described with experimental accuracy up to the (v(1)v(2)v(3)) = (080) state. Extrapolation tests demonstrate the predictive power of the potential function obtained: rotational extrapolation up to J = 10 for the 11 lowest vibrational states results in an rms deviation of 0.06cm(-1). Also, vibrational energies measured by low-resolution Raman spectroscopy (which were not included in the input data for the fit) are calculated within the experimental accuracy (rms = 1.6 cm(-1)) of the experimental values up to the dissociation limit. The statistical analysis suggests that the accuracy of the equilibrium geometry and force constants of the molecule is considerably improved relative to previous determinations. The long-range behavior of the fitted potential at the dissociation limit O(3) --O(2) + O shows very good agreement with experimental data. The new potential energy surface was used to predict the band centers of the isotopomers (17)O(3) and (18)O(3). Copyright 1999 Academic Press.

We describe the updates to the spectroscopic parameters of ozone and its isotopes in the 1996 HITRAN compilation. Recent published studies not included in HITRAN are also summarized. Finally, we report the identification of infrared lines of the v(sub 3) bands of O-16O-16O-17 and O-16O-17O-16 in high-resolution solar spectra recorded by stratospheric balloon-borne and ground-based Fourier transform spectrometers.

The and bands of ozone, which occur in the region, have been observed for the first time using a Fourier transform spectrometer, operating at resolution and with a large pathlength pressure product . The assignment of rotation-vibration transitions has been done for J up to 35 and Ka up to 12, for the band and up to J = 40 and Ka = 2 for the band respectively. The effective Hamiltonian used in calculations accounts for the first-order Coriolis coupling between these two bands. In addition, as the levels corresponding to Ka = 5 are perturbed, it was necessary to account for anharmonic resonance with the (321) state. Also, to correctly reproduce the levels of (015) with Ka = 8,9,10 it has been necessary to include a Coriolis coupling with (080) state in the model. In this way, we have obtained a very satisfactory r.m.s. deviation of , near the experimental accuracy. Rovibrational line intensities of and bands were measured and the value of the principal transition moment constants have been recovered. Finally a complete list of line positions and intensities was calculated leading to the integrated band intensity of /molecule for the band and /molecule for the band at 296 K with a cut-off of .

Hot bands nu1 + 3nu2 - nu2 and 3nu2 + nu3 - nu2 of 16O3 in the region 2300-2600 cm-1 and the cold band 3nu2 + nu3 in the region 3050-3110 cm-1, corresponding to the v2 = 3 dyad {(130), (031)}, have been observed for the first time, using the Fourier Transform Spectrometer (FTS) at Reims and the usual experimental setup providing a large product pressure x path length, p x l. Three hundred sixty-five rovibration energy levels of the upper states were obtained with J and Ka up to 46 and 9, respectively. The fit of these data gives a r.m.s. deviation of 1.93 x 10(-3) cm-1. The v2 dependence of the rotational parameters A, B, and C for the (1v20) and (0v21) states is discussed. Copyright 1998 Academic Press. Copyright 1998Academic Press