THERMOSPHERIC NITRIC OXIDE INFRARED EMISSIONS MEASURED BY CRISTA

K. U. Grossmann , M. Kaufmann , and K. Vollmann

Physics Department, University of Wuppertal, Gauss-Str.20, D-42097 Wuppertal, Germany

ABSTRACT

In November 1994 the CRISTA (Cryogenic Infrared Spectrometers and Telescopes of the Atmosphere) experiment measured mid and far infrared spectra from the Earth limb. One of the spectral channels covered the 5.3 µm (delta v = 1) band of nitric oxide. A first an alysis of the data of this channel is presented. Large enhancements of the 5.3 µm band radiances were observed at high northern and at high southern magnetic latitudes. The measured intensities were a factor of 4 - 6 above those found at low latitudes.

Nitric oxide concentrations in the lower thermosphere have been subject to extensive research over the past decades. Most of the experimental data were collected by ultraviolet sensing instruments on rockets and on satellites. In these experimental techniques the resonance fluorescence of solar UV from NO bands is analysed. The data are therefore limited to daytime conditions. As a consequence there is much less experimental data available for nighttime. Remote sensing of NO can also be performed by measuring the NO (delta v = 1) band in the infrared at 5.3µm. NO emission contributes considerably to the radiative cooling of the lower thermosphere (Kockarts, 1980; Gordiets et al., 1982). A number of experiments have been carried out in the past which quantitativel y measured this emission globally by satellite instruments (Ballard et al.,1993; Smith and Ahmadjian, 1993; Wise et al., 1995), and locally by rocket-borne instruments (Ulwick et al., 1985; Stair et al., 1985; Grossmann et al., 1994).

The conversion of the measured infrared radiances to NO densities, however, requires the explicit determination of the vibrational level populations of the NO molecules. The level populations are not in local thermodynamic equilibrium (NLTE) in the thermos phere, like other molecules at these altitudes. In the case of nitric oxide, such an NLTE model is comparatively simple if higher order effects like chemical excitation or rotational non-equilibrium are neglected. Under these conditions NO (v = 1) is populated essentially by collisions with ground state atomic oxygen and, to a much lesser extent, by radiative pumping (Wise et al., 1995). At high altitudes above 150 km and for average or low atomic oxygen densities (compared to CIRA - 86), the 5.3 µm volume emission rate depends nearly linearly on the oxygen concentration. In the lowest part of the thermosphere or during times of high oxygen atom abundance, the 5.3 µm emission is less sensitive to the O - concentration. Furthermore, the 5.3 µm band is opt ically thin even for limb viewing geometries. The band integrated volume emission rate from the 5.3 µm band is, therefore, directly proportional to the NO column density along the line of sight of the instrument. The collisional excitation rate is highly temperature dependent due to the relatively high vibrational energy of the v = 1 levels. Thus, in order to derive NO densities from these infrared measurements, the thermospheric temperature must be precisely known as well as the local atomic oxygen densit y. The latter can be derived from the measurements of the 63 µm line of ground state atomic oxygen which was done in one of the CRISTA channels (Grossmann et al., 1997) .

MEASUREMENTS

Thermospheric 5.3 µm band emissions were measured by the CRISTA experiment (CRyogenic Infrared Spectrometers and Telescopes for the Atmosphere) which was flown as part of the Shuttle mission STS-66 in November 1994.

Fig. 1 : Nitric oxide 5.3 µm R - branch band intensities on day 313, 1994. The dashed line shows the 60° magnetic latitude contour.

Details about this experiment are given by Riese et al. (1997). The CRISTA sensor was optimized for high spatial resolution measurements in the stratosphere and mesosphere. During one of the CRISTA measurement modes (mode 1), every fourth altitude scan of the center telescope of CRISTA allowed measurements of the NO emission up to about 130 km. Measurements in this mode were taken for about 55 hours. An example of the data is shown in Figure 1. Plotted are the radiances integrated over the wavenumber interval from 1870 cm^-1 to 1980 cm^-1 for day 313 (Nov 9, 1994) at an altitude of 120 km.

Fig. 2 : Same as Figure 1 except for day 312 and CRISTA measurement mode MT (for details see text).

This wavenumber interval essentially covers the R-branch of the NO fundamental band. High signals were found at high northern latitudes over Canada as well as at high southern latitudes south and south-west of Australia. In these areas high magnetic latitudes were reached as demonstrated by the 60° magnetic latitude contour in Figure 1. The high radiance values seen in these regions were caused by either high nitric oxide densities or higher temperatures. A first analysis of the atomic oxygen data from the CRISTA instrument (Grossmann et al., 1997) indicates rather low atomic oxygen densities at all latitudes, but especially at high latitudes in the southern hemisphere. The fluctuations of the data points at mid and low latitudes are mainly caused by the instrumental noise.

On day 312 the CRISTA instrument was operated for about 4 hours in a different scanning mode which allowed it to take data with the NO channel up to about 160 km (measurement mode MT). The data obtained during mode MT at 120 km altitude are shown in Figure 2. The NO emissions were again integrated fom 1870 cm^-1 to 1980 cm^-1. The noise equivalent radiance of a single data point in Figure 2 is about 1×10^-8 Wcm^-2sr^-1. Thus, the signal to noise ratio for low intensities, as encountered at low latitudes, is only in the vicinity of one.

In Figure 3 altitude profiles are plotted which were obtained by averaging all profiles measured at high northern latitudes (40°N to 65°N) in the longitude sector from -20° to -100° (profile N), at low latitudes within ±20° to the equator (profile E), and at high southern latitudes (40°S to 55°S) in the longitude sector from 100° to 180° (profile S).

Fig. 3 : NO band intensities averaged over ±20° latitude (profile E), from 40° to 60° latitude and -100° to -20° longitude (profile N), and from -40° to -55° latitude and 100° to 180° longitude (profile S).

As in Figure 2, the data are wavenumber integrated in the interval 1870 to 1980 cm^-1. The profiles generally peaked in the 120 to 130 km range. The peak radiances at high latitudes exceeded those at low latitudes by factors of 4 - 6. On day 313 (Figure 1) similar enhancements were recorded. These values were less than the values found by Wise et al. (1995) and may be explained by the low geomagnetic activity on both days. Of note was day 312 which was the fifth quietest day in November 1994.

RADIANCE MODELLING

A simple NLTE model was set up in which the NO (v = 1) level populations are calculated by balancing the excitation rates (radiative pumping and collisions with atomic oxygen) with the quenching rates (spontaneous emission and collisions with atomic oxygen). These processes are the only ones of importance above 100 km (Wise et al., 1995). The thermospheric temperature and the atomic oxygen densities were taken from the CIRA-86 global mean model. The values of A_P = 4 and F_10.7 = 70 were chosen because these simulate the conditions during the CRISTA mission. In Figure 4 the vibrational temperatures of the first vibrational level are plotted vs altitude together with the respective kinetic temperature of the CIRA model . Above 130 km the vibrational temperature is essentially constant, due to the fact that the decreasing atomic oxygen densities are roughly compensated by the increase in kinetic temperature.

Fig. 4 : Vibrational temperature of NO (v = 1) for CIRA-86 global mean atmospheric conditions (for details see text).

The contribution of the radiative excitation to the population of the v = 1 level is negligible above about 115 km (i.e. 1%). The drop in the vibrational temperature below about 130 km reduces the volume emission rate of NO (5.3&181;m) drastically. Therefor, the limb intensity reaches a maximum near 130 km (Figure 3), and falls off gradually towards lower altitudes, due to the changing ray path geometry. As result, information about NO densities can only be obtained above the emission maximum where the limb intensity profile is sensitive enough to the NO concentration. The absolute value of the vibrational temperature depends strongly on the kinetic temperature wich has to be evaluated very thoroughly before NO densities can be derived. The agrement between the vibrational and the kinetic temperatures near 100 km is accidental and does not mean that LTE conditions are reached.

OUTLOOK

The UV spectrometer MAHRSi (Middle Atmosphere High Resolution Spectrograph Investigation) was flown on the same satellite as the CRISTA experiment (Offermann and Conway, 1995). MAHRSI was boresighted with the CRISTA center telescope (which includes the NO channel) and during part of the mission MAHRSI measured daytime nitric oxide densities by analysing the NO gamma-band resonance fluorescence. A detailed future comparison of derived NO densities of both MAHRSI and CRISTA will allow establishment of a validated NLTE retrival model for the 5.3µm NO band.

ACKNOWLEDGMENT

The CRISTA experiment and its data evaluation is funded by the Bundesminister für Bildung und Forschung through DARA GmbH, Bonn.

REFERENCES

Ballard, J., B. J. Kerridge, P. E. Morris, and F. W. Taylor, Observations of v = 1-0 emission from thermospheric nitric oxide by ISAMS, Geophys. Res. Lett., 20, 1311 (1993)

CIRA-86, COSPAR international Reference Atmosphere : 1986, Adv. Space Res., 10 ( 1990 )

Gordiets, B. F., Yu. N. Kulikov, M. N. Markov, and M. Ya. Marov, Numerical modeling of the thermospheric heat budget, J. Geophys. Res., 87, 4504 (1982)

Grossmann, K. U., D. Homann, and J. Schulz, Lower thermospheric infra-red emissions from minor species during high-latitude twilight - A. Experimental results, J. Atmos. Terr. Phys., 56, 1885 (1994)

Grossmann, K. U., M. Kaufmann, and K. Vollmann, The fine structure emission of thermospheric atomic oxygen, Adv. Space Res.,this issue (1997)

Kockarts, G., Nitric oxide cooling in the terrestrial atmosphere, Geophys. Res. Lett., 7, 137 (1980)

Offermann, D., and R. R. Conway, CRISTA-SPAS mission searches for effects of dynamics on the composition of Earth's middle atmosphere, Eos, 76, 337 (1995)

Riese, M., P. Preusse, R. Spang, M. Ern, M. Jarisch, K. U. Grossmann, and D. Offermann, Measurements of trace gases by the CRyogenic Infrared Spectrometers and Telescopes for the Atmosphere (CRISTA) experiment, Adv. Space Res., this issue (1997)

Smith, D. R., and M. Ahmadjian, Observation of nitric oxide rovibrational band head emissions in the quiescent airglow during the CIRRIS-1A space shuttle experiment, Geophys. Res. Lett., 20, 2679 (1993)

Stair, A. T., R. D. Sharma, R. M. Nadile, D. J. Baker, and W. F. Grieder, Observations of limb radiance with cryogenic spectral infrared rocket experiment, J. Geophys. Res., 90, 9763 (1985)

Ulwick, J. C., K. D. Baker, A. T. Stair, W. R. Frings, R. Hennig, K. U. Grossmann, and E. R. Hegblom, Rocketborne measurements of atmospheric infrared fluxes, J. Atmos. Terr. Phys., 47, 123 (1985)

Wise, J. O., R. L. Carovillano, H. C. Carlson, R. G. Roble, S. Adler-Golden, R. M. Nadile, and M. Ahmadjian, CIRRIS 1A global observations of the 15 µm CO2 and 5.3 µm NO limb radiance in the lower thermosphere during moderate to active geomagnetic activity, J. Geophys. Res., 100, 21357 (1995)