Odd Oxygen Research Articles

A test of our understanding of the ozone chemistry in the Arctic polar vortex based on in situ measurements of ClO, BrO, and O3 in the 1994/1995 winter

We present an analysis of in situ measurements of ClO, BrO, O3, and long‐lived tracers obtained on a balloon flight in the Arctic polar vortex launched from Kiruna, Sweden, 68°N, on February 3, 1995. Using the method of tracer correlations, we deduce that the air masses sampled at an altitude of 21 km (480 K potential temperature), where a layer of enhanced ClO mixing ratios of up to 1150 parts per trillion by volume was observed, experienced a cumulative chemical ozone loss of 1.0±0.3 ppmv between late November 1994 and early February 1995. This estimate of chemical ozone loss can be confirmed using independent data sets and independent methods. Calculations using a trajectory box model show that the simulations underestimate the cumulative ozone loss by approximately a factor of 2, although observed ClO and BrO mixing ratios are well reproduced by the model. Employing additional simulations of ozone loss rates for idealized conditions, we conclude that the known chlorine and bromine catalytic cycles destroying odd oxygen with the known rate constants and absorption cross sections do not quantitatively account for the early winter ozone losses infered for air masses observed at 21 km.

A climatology of total ozone mapping spectrometer data using rotated principal component analysis

The spatial and temporal variability of total column ozone (Ω) obtained from the total ozone mapping spectrometer (TOMS version 7.0) during the period 1980–1992 was examined through the use of a multivariate statistical technique called rotated principal component analysis. Utilization of Kaiser's varimax orthogonal rotation led to the identification of 14, mostly contiguous subregions that together accounted for more than 70% of the total Ω variance. Each subregion displayed statistically unique Ω characteristics that were further examined through time series and spectral density analyses, revealing significant periodicities on semiannual, annual, quasi‐biennial, and longer term time frames. This analysis facilitated identification of the probable mechanisms responsible for the variability of Ω within the 14 homogeneous subregions. The mechanisms were either dynamical in nature (i.e., advection associated with baroclinic waves, the quasi‐biennial oscillation, or El Niño‐Southern Oscillation) or photochemical in nature (i.e., production of odd oxygen (O or O3) associated with the annual progression of the Sun). The analysis has also revealed that the influence of a data retrieval artifact, found in equatorial latitudes of version 6.0 of the TOMS data, has been reduced in version 7.0.

Odd oxygen measurements during the Noctilucent Cloud 93 rocket campaign

The NLC‐93 campaign at Esrange, Sweden, addressed the chemical, dynamical and electrical environment of the summer mesopause in the presence of noctilucent clouds (NLC). A major topic was the detailed investigation of odd oxygen abundances and their structure in the vicinity of the cloud layer. Applying independent in situ measurements, we have obtained consistent profiles of atomic oxygen, ozone, and mesospheric total density. Atomic oxygen was measured on two rocket flights by means of a newly developed probe utilizing the 1304 Å resonance fluorescence / absorption technique. Detailed Monte Carlo simulations of the radiative transfer and the instrument aerodynamics are included in the data analysis. We find relatively low atomic oxygen densities with a peak of 2 × 1011 cm−3 at 96 km. An airglow radiometer recorded the O2 IR‐atmospheric band emission, allowing the retrieval of ozone densities by means of an inversion model that considers both direct and indirect sources of O2 (a1Δg). The odd oxygen measurements are compared to a steady state photochemical model. While good agreement is found in the mesosphere, the results deviate close to the cold mesopause. Significant vertical structures are found in the atomic oxygen profiles; sharp gradients and distinct minima coincide with the observed NLC layer. Possible influences of heterogeneous chemistry on the odd oxygen abundance are discussed.

Implications of upper stratospheric trace constituent changes observed by HALOE for O3 and ClO from 1992 to 1995

Measurements from the Halogen Occultation Experiment (HALOE) of the upper stratosphere show increases in HCl and H2O and decreases in CH4 and O3 during the period 1992–1995. These changes all coincide with the decline of solar cycle 22. Using a simple photochemical model, we find that 4 major components contribute to the O3 decrease: 1) an increase in total chlorine as indicated by increasing HCl 2) an additional increase in reactive chlorine due to repartitioning of Cly by the decreasing CH4 3) a decrease in odd oxygen production due to decreased solar flux and 4) an increase in odd hydrogen loss due to increasing H2O. At 2 mbar, the Cly repartitioning is the largest cause of O3 changes. Because the Cly repartitioning coincides with decreasing solar flux, some recent observational estimates of the long‐term response of upper stratospheric O3 to solar UV irradiance variability may be too large. Compared with the HALOE O3 data, the model O3 exhibits a larger negative trend. This appears to be because the model O3 is more sensitive to the increased H2O than is the observed O3. The 20–30% decline in CH4 also implies a large increase in ClO which exceeds that expected from CFCs. Observations of ClO from the Microwave Limb Sounder (MLS) support this inference.

Localized rapid ozone loss in the northern winter stratosphere: An analysis of UARS observations

Observations of low‐ozone air pockets forming during northern winter in the middle stratosphere outside the polar vortex provide an opportunity to test models of the photochemistry of ozone at several altitudes, as the trajectories of the associated air parcels are well defined by Lagrangian transport codes over the periods of interest, and vertical profiles of key species, including ozone, are available from instruments aboard the Upper Atmosphere Research Satellite (UARS). We find that a Lagrangian photochemical model, where the chemistry within an isolated parcel of air is simulated as it travels along a specified trajectory, does reproduce the observed formation of low‐ozone pockets in the 6–10 mbar altitude range within the measurement uncertainties but overestimates the ozone loss rate at higher altitudes. The rapid ozone loss localized in these pockets is due to the isolation of air at high latitudes (and high solar zenith angles). Thus the low ozone levels are due to a decrease in the odd oxygen production rate and not to an increase in the loss rate by reaction with halogen species, as in the “classical” ozone hole.

Peroxyacetyl Nitrate (PAN) Measurements During the POPCORN Campaign

During the POPCORN campaign between 3 and 24 August 1994 we measured peroxyacetyl nitrate (PAN) in a rural area of Mecklenburg-Vorpommern (North-Eastern Germany) above a corn field. A total of about 5000 PAN measurements were carried out within the three weeks of the campaign. Measured PAN mixing ratios ranged from below the detection limit of 10 ppt up to an afternoon maximum of 1 ppb. The mean value of all data was 140 ppt. The daily mean PAN mixing ratios were typically in the range of 50 to 250 ppt, but during a clean air episode PAN mixing ratios of well below 40 ppt were observed. The characteristic relative diurnal variation of the PAN mixing ratios with a late night/early morning minimum and an afternoon maximum persisted during these episodes. The daily averages of the PAN mixing ratios showed clear episodic variations which coincided with the duration of typical synoptic episodes of two to six days duration. Based on the measurements of the various parameters determining the PAN formation and destruction rates, the local budget for PAN was calculated. During daytime the calculated net photochemical formation rate of PAN was nearly always significantly higher than the observed change of the PAN concentration. This demonstrates that substantial amounts of PAN (often in the range of several hundred ppt/h) were exported from the corn field. The resulting removal of NOx to some extent effects the budget of nitrogen oxides (NOx), but the export of odd oxygen radicals in the form of PAN during daytime often amounted up to 30–50% of the OH-radical formation by ozone photolysis. Thus the importance of PAN as reservoir and transport medium for odd oxygen radicals can be very substantial and may have a significant impact on the budget and distribution of odd oxygen radicals.

Vertical nitrogen dioxide and ozone concentrations measured from a tethered balloon in the Lower Fraser Valley

A series of vertical profiles of temperature, relative humidity, NO 2 and O 3 were determined in the Lower Fraser Valley, British Columbia, as part of the PACIFIC '93 field study. Data from one day show very structured vertical distributions of all parameters in the morning, as expected from the limited vertical mixing under the nocturnal inversion. NO 2 concentrations of 20 ppbv were observed 300m above the surface, while the surface concentrations were ∼ 2 ppbv. Ozone and nitrogen oxide chemistry were observed at all altitudes throughout the PBL. Titration of O 3 by NO to produce NO 2 was observed in layers above the ground, under the influence of the NBL. An increase in odd oxygen throughout the PBL, during the morning and early afternoon, shows “smog chemistry” is occurring even though the ground-based O 3 measurements suggest this day was not particularly chemically active. However, once the NBL has dissipated, the ground-based measurements seem representative of the entire PBL.

Latest Pliocene pollen and leaf floras from Bernasso palaeolake (Escandorgue Massif, Hérault, France)

Bernasso palaeocanyon (Hérault, France) is in the Escandorgue volcanic massif, south of the Massif Central. The first glacial-interglacial cycles of the Plio-Pleistocene period are recorded in the infilling of this canyon and are rich in macroflora (mainly leaves) and microflora (mainly pollen). Pollen zones I and III represent a steppe environment, in contrast to pollen zone II, a forested one. This study concentrates on pollen zone II for which new pollen spectra and abundant leaves are described owing to new outcrops and a 8 m long drillhole (BN I) through the deposit. Pollen zone II-a poorly illustrates an open oak forest that could correspond to a pioneering phase after a steppe period (pollen zone I). No sampling of the sediment between those yielding pollen zones II-a and II-b was possible. Therefore little is known of the intermediate vegetational phases. Pollen zones II-b to II-d show a diversified mixed temperate forest. Pollen zones II-a to II-d are probably steps in vegetation successions of an entire interglacial period such as the Nogaret interglacial, defined in a neighbouring diatomite deposit of similar age. The macrofloral study is based on 800 samples. Most abundant specimens are Carpinus suborientalis, Parrotia persica, Acer and Carya. The Bernasso macroflora reflects a mesothermic forest similar to the present vegetation south of the Caspian sea. The difference between the dominant genera in the macroflora and microflora leads to an improvement in the interpretation of pollen percentages. There is an under-representation of pollen from Carpinus orientalis, Parrotia persica and Acer sp. and over-representation of Carya minor, Zelkova, Pinus and Tsuga. The Hyrcanian region and to a lesser degree the Euxinian region, serve as models for the Pliocene flora. Compared to present-day Bernasso, the Late Pliocene climate was milder and had more precipitation throughout the year with a reduced decrease during the summer leading to reduced seasonal contrast. The estimated duration for pollen zone II, part of an odd oxygen isotope stage driven by an obliquity cycle (41 ka), is less than 10 ka. The age estimates (based on K/Ar dates, palaeomagnetism and cyclopalynostratigraphy) indicate that Bernasso floras developed during the time interval between ca. 2.16 and 1.96 Ma. Therefore, Bernasso pollen zones II-b to II-d could correlate to part of an oxygen isotope stage between stage 81 and stage 73.

Detailed balance limits the rate of odd oxygen production in the reaction of vibrationally excited oxygen with O2

Application of detailed balance permits estimation of the rate coefficients for the reaction of highly vibrationally excited O2 with ambient O2 to yield odd oxygen, a process of potential importance in the stratosphere, O2(υ=26, 27) + O2 → O3 + O. The rate coefficients calculated in this manner are much smaller than those reported from experiment and below any level of atmospheric significance.

Model studies of the influence of O2 photodissociation parameterizations in the Schumann-Runge bands on ozone related photolysis in the upper atmosphere

Abstract. A new parameterization for atmospheric transmission and O2 photodissociation in the Schumann-Runge band region has been developed and tested with a 1D radiative-photochemical model. The parameterization is based on the O2-column along the line of sight to the Sun and the local temperature. Line-by-line calculations have served as a benchmark for testing this method and several other, commonly used, parameterizations. The comparisons suggest that differences between the line-by-line calculations and currently accepted parameterizations can be reduced significantly by using the new method, particularly at large solar zenith angles. The production rate of O-atoms computed with this method shows less than 6% deviation compared to the line-by-line calculations at any altitude, all solar zenith angles and in all seasons. The largest errors are found toward the shorter wavelengths in the Schumann-Runge region at low altitudes. Transmittance is approximated to better than 4% at any altitude and/or solar zenith angle. The total O-production rate above 20 km is approximated to better than 2%. The new parameterization is easily implemented in existing photochemical models and in many cases it may simply replace the existing algorithm. The computational effort exceeds that of other parameterizations but in view of the total computation time needed for the actual calculation of the parameterized Schumann-Runge bands this should not lead to significant performance degeneration. The first 14 coefficients of the parameterization are included in this study. Both the complete sets of coefficients and a simple algorithm can be obtained by contacting the authors. A photochemical model study shows the largest effect of the parameterization method is on odd hydrogen concentrations. Subsequent interaction with an odd oxygen family causes differences in the ozone concentrations between the different parameterizations of more than 10% at selected altitudes. Although it is already established that deficiencies in the treatment of Schumann-Runge band absorption are unlikely to explain the current underestimation of ozone concentration at the stratopause in a variety of photochemical models, this study does show that the choice of parameterization has a large impact on the accuracy of the results at large solar zenith angles and in different seasons.

UV/visible and IR absorption cross sections of BrONO2

The absorption cross sections of BrONO2 between 200 and 500 nm were measured over the temperature range 298 to 220 K using a diode array spectrometer. The BrONO2 absorption cross sections are weakly dependent on temperature at wavelengths < 450 nm (less than 10% change between 298 and 220 K) but decrease rapidly with decreasing temperature at wavelengths > 450 nm; at 480 nm the cross section decreased by ∼35% in going from 298 K to 220 K. Our room temperature absorption cross sections are in good agreement with the measurements of Spencer and Rowland (1978) over the common wavelength range of the measurements, 200 to 390 nm. We show that wavelengths longer than 390 nm must be included in the calculation of the BrONO2 atmospheric photolysis rate. We also show that photolysis of BrONO2 could be a significant atmospheric loss process for odd oxygen (due to halogen chemistry) below about 25 km. The infrared absorption cross sections for the BrONO2 band centered at 803.3 cm−1 were also measured. The integrated band strength was (2.7±0.5) × 10−17 cm2 molecule−1 cm−1.

Production and vibrational kinetics of nitric oxide in the disturbed polar thermosphere

We study mechanisms of NO excitation in the (quiet and disturbed) polar thermosphere using a one-dimensional non-steady set-up. Main mechanisms are: 1) TV-excitation in the atom-molecular interactions, 2) chemical reaction of odd nitrogen and oxygen molecules. We find that the vibrational kinetics of molecular nitrogen plays an important role in the excitation of NO(v) during very disturbed periods. The collective relaxation effects of precipitating electron beams and the process of VV′-energy transfer are principal in the generation of NO(v) in very disturbed conditions. The role of vibrationally excited molecular nitrogen in chemical kinetics of NO is discussed. The energy efficiencies of the mechanisms in the aurora are assessed. Good agreement is obtained when comparing our results to rocket and aircraft data of infrared radiations of vibrationally excited NO.

Photochemistry and stability of the atmosphere of Mars

An understanding of the composition, structure, transport and photochemical processes of the present atmosphere of Mars is essential for addressing such fundamental questions as the stability, evolution and the origin of the Martian atmosphere. This paper discusses our current knowledge of the photochemistry and stability of the Martian atmosphere, with special emphasis on new models and work in progress. The atmosphere of Mars is composed mainly of CO 2 (95.3% by volume), N 2 (2.7%), and 40Ar (1.6%). Trace constituents, especially H 2O vapor (150–200 ppm, average), CO (0.07%), O 2 (0.13%), and ozone (0.03 ppm average) make up the rest. Vertical mixing, characterized by an eddy diffusion coefficient, is of the order of 10 6cm 2s −1 up to 40 km/1,2/, rising to 10 8cm 2s −1 at the homopause (125 km). Photochemistry initiated by the solar ultraviolet dissociation of CO 2 and H 2O, results in the formation and distribution of the above-mentioned trace constituents. O 2 and CO remain well-mixed and virtually constant because of their long photochemical lifetimes. Ozone shows an anticorrelation with water vapor amounts because of the odd oxygen loss on HO x and H. An intermediate photochemical product in the form of hydroxyl radicals, OH, plays a catalytic role in maintaining the balance between the photodissociative loss of CO 2 and its subsequent recycling through a reaction of CO with OH. The homogeneous gas phase chemical reactions, however, do not entirely explain the stability of CO 2, O 2, or CO, thus necessitating the inclusion of heterogeneous adsorption as a loss mechanism for certain species, such as H 2O, OH, H 2O 2, O or O 2 on aerosols of dust or ice suspended in the Martian air/3/. Chlorine catalyzed reactions may be important in maintaining the stability of the Venus atmosphere. Recent detection of formaldehyde (CH 2O) at the 0.5 ppm level in the atmosphere of Mars /4/ is puzzling, since the conventional mechanism for its formation — oxidation of methane — would require 0.5–1% CH 4 which could not have escaped detection. Even 2-D models require at least a ppm level of CH 4, which also would have been detected, if present. Reaction of H with CO to form HCO, and subsequent reaction of HCO with HO 2, also would produce CH 2O only at a ppb level. Perhaps, a heterogeneous reaction involving long wavelength ultraviolet irradiation of a COH 2OCO 2 mixture could help; however, laboratory experiments do not yet quantify the yield of formaldehyde. Should the presence of formaldehyde be confirmed after further observations, it would hold a tantalizing possibility of the existence of more complex organics in the interior, surface and the atmosphere of Mars. The virtual lack of organics on Mars could be attributed to the presence of oxidants such as hydrogen peroxide.

Wave‐modified mean exothermic heating in the mesopause region

We employ a model of wave‐driven OH nightglow fluctuations to calculate the effects of gravity waves on the chemical exothermic heating due to reactions involving odd hydrogen and odd oxygen species in the mesopause region. Using a model based on time means and deviations from those means, it is demonstrated that gravity waves contribute to the time‐average exothermic heating. The effect can be significant because the fractional fluctuations in minor species density can be substantially greater than the fractional fluctuation of the major gas density. Our calculations reveal that the waves mitigate the exothermic heating, demonstrating their potential importance in the heat budget of the mesopause region.

The "Ozone Deficit" Problem: O 2 ( X, v ≥ 26) + O( 3 P ) from 226-nm Ozone Photodissociation

Highly vibrationally excited O(2)(X(3)sigmag(-), v >/= 26) has been observed from the photodissociation of ozone (O(3)), and the quantum yield for this reaction has been determined for excitation at 226 nanometers. This observation may help to address the "ozone deficit" problem, or why the previously predicted stratospheric O(3) concentration is less than that observed. Recent kinetic studies have suggested that O(2)(X(3)sigmag(-), v >/= 26) can react rapidly with O(2) to form O(3) + O and have led to speculation that, if produced in the photodissociation of O(3), this species might be involved in resolving the discrepancy. The sequence O(3) + hv --> O(2)(X(3)sigmag(-), v >/= 26) + O; O(2)(X(3)sigmag(-), v >/= 26) + O(2) --> O(3) + O (where hv is a photon) would be an autocatalytic mechanism for production of odd oxygen. A two-dimensional atmospheric model has been used to evaluate the importance of this new mechanism. The new mechanism can completely account for the tropical O(3) deficit at an altitude of 43 kilometers, but it does not completely account for the deficit at higher altitudes. The mechanism also provides for isotopic fractionation and may contribute to an explanation for the anomalously high concentration of heavy O(3) in the stratosphere.

Energy released by recombination of atomic oxygen and related species at mesopause heights

This paper studies the total chemical energy release of seven exothermic reactions involving odd oxygen and odd hydrogen species at mesopause heights. Monthly and zonally averaged heating rates are calculated on the basis of atomic oxygen and atomic hydrogen densities as inferred from the SME satellite measurements in the altitude regime from 80 to 93 km. Trace constituents not available from SME were determined by means of a photochemical model. The resulting heating rates are found to be fairly high at all seasons, altitudes, and latitudes accessible to SME satellite measurements. Very large heating rates occur at high latitudes in early winter, reaching a maximum value of approximately 28 K per day. Consequently, chemical heating appears to play an important role in the energy budget and the dynamics of the upper mesosphere and lower thermosphere.

Comet Shoemaker‐Levy‐9 impact with Jupiter: Aeronomical predictions

The fragments of comet Shoemaker‐Levy‐9 will enter the atmosphere of Jupiter during July 20–26, 1994. Significant amounts of water vapor will be injected into the upper atmosphere of Jupiter either from the comet itself or from the lower atmosphere of Jupiter. The photochemistry of both the neutral gas and the ionosphere will be greatly altered by the influx of this water vapor or the atomic oxygen generated by the dissociation of the water. Enhanced abundances of H2O (or other species such as NH3) in the atmosphere above the homopause should persist for at least a year and should be globally distributed. The odd oxygen (i.e., O or H2O or OH) associated with the cometary water influx alters the ion chemistry by removing H+ ions, which also has the effect of reducing the ionospheric electron density because H+ is ordinarily the main ion species in the Jovian ionosphere. The density of H3+, both in the auroral and non‐auroral ionosphere, will also be reduced due to presence of water. This ion species has been detected spectroscopically at Jupiter, and a drop in its abundance should be detectable. The major ion species will become H3O+ which could reach a peak density as high as 104 cm−3 in the non‐auroral ionosphere and 105 cm−3 in the auroral ionosphere. It should be possible to detect this ion species spectroscopically from Earth‐based observatories.

Inhibition of odd oxygen production in the carbon bond four and generic reaction set mechanisms

Abstract Outdoor smog chamber experiments have been conducted by researchers in Australia at lower initial NOx and VOC-to-NOx ratios than those chamber experiments which were used to formulate photochemical reaction mechanisms such as the Carbon Bond Four (CB4). When the Australians simulated their experiments with CB4, they found that CB4 underpredicted ozone production at low VOC-to-NOx ratios, i.e. in the NOx-rich region of the ozone isopleth diagram above the ridge line. CB4 predicts a significant NOx inhibition in this region whereas the empirical General Reaction Set (GRS) mechanism, which was developed by the Australian researchers to fit their data, does not. In this study, a mass balance and process analysis technique is used to explain the origins of the NOx -inhibition in CB4 simulations. The causes are determined to be in the inorganic reactions of the CB4, which are believed to be both complete and well-known. The primary cause was a strong negative feedback of new radical production from the photolysis of ozone which occurred because of the longer delay needed to oxidize the higher initial NO. The CB4 mechanism could still be incorrect if it is missing a strong radical source, for example, in the poorly understood aromatics chemistry. The GRS, which lacks representation of the inorganic processes responsible for NOx-inhibition, may be achieving its predictive accuracy by compensating errors. If this is true, then the GRS would be unsatisfactory for ambient air use. Additional smog chamber experiments at low VOC-to-NOx ratios and better characterization of the Australian chambers are needed to resolve the discrepancy in CB4 and GRS predictions.

Effect of vibrationally excited oxygen on ozone production in the stratosphere

Photolysis of vibrationally excited oxygen produced by ultraviolet photolysis of ozone in the upper stratosphere is incorporated into the Lawrence Livermore National Laboratory two‐dimensional zonally averaged chemical‐radiative‐transport model of the troposphere and stratosphere. The importance of this potential contributor of odd oxygen to the concentration of ozone is evaluated based on recent information on vibrational distributions of excited oxygen and on preliminary studies of energy transfer from the excited oxygen. When energy transfer rate constants similar to those of Toumi et al.(1991) are assumed, increases in model ozone concentrations of up to 4.0% in the upper stratosphere are found, and the model ozone concentrations are found to agree slightly better with measurements, including recent data from the Upper Atmosphere Research Satellite. However, the ozone increase is only 0.3% when the larger energy transfer rate constants indicated by recent experimental work are applied to the model. An ozone increase of 1% at 50 km requires energy transfer rate constants one‐twentieth those of the preliminary observations. As a result, vibrationally excited oxygen processes probably do not contribute enough ozone to be significant in models of the upper stratosphere.

Absorption of solar radiation by O2: Implications for O3 and lifetimes of N2O, CFCl3, and CF2Cl2

An accurate line‐by‐line model is used to evaluate effects of absorption in the Schumann‐Runge bands of O2 on transmission of ultraviolet radiation. Allowing also for absorption in the Herzberg continuum, the model is shown to provide a reliable simulation of observed transmission in the spectral interval 192 to 200 nm. The model is used to evaluate rates for photolysis of N2O, CFCl3, and CF2Cl2, and to infer global loss rates (1.22×l010 kg N yr−1, 7.21×107 and 3.04×l07 kg Cl yr−1, respectively) and instantaneous lifetimes (123, 44, and 116 years, respectively) appropriate for 1980. A parameterized version of the line‐by‐line model enabling rapid evaluation of transmission in the Schumann‐Runge region is described. Photochemical calculations employing the parameterization and constrained by data from the Atmospheric Trace Molecule Spectroscopy experiment are used to examine the budget of odd oxygen. Consistent with previous studies, it is shown that photochemical loss of odd oxygen exceeds production by photolysis of O2 for altitudes above 40 km. The imbalance between production and loss is shown to be consistent with a source of odd oxygen proportional to the product of the mixing ratio and photolysis rate of ozone, which suggests that processes involving vibrationally excited O2 may play an important role in production of odd oxygen.