Investigation of polar mesospheric dynamics and temperature changes by means of ground-based OH* airglow spectroscopy and model studies above Spitsbergen

Stefan Kowalewski*1, Justus Notholt1, Mathias Palm1, Christine Weinzierl1, Christian von Savginy2 *contact: [email protected]

1 Institute of Institute of Environmental Physics, University of Bremen, Germany Postgraduate International University of University of 2 Institute of Physics, University of Greifswald, Germany Programme In Physics and Environmental Bremen Greifswald pip Electrical Engineering Physics

1. Introduction & Motivation 3. OH* Airglow Temperatures and Intensities above Spitsbergen

mesospheric temperatures above Ny- lesund • The OH airglow layer is a prominent feature of the mesopause hourlyhourly averaged OH(3 temperatures-1) variability from OH(3-1) - Jan 23-24, 2011 Å MLS (alt = 80km) region at approx. 87km altitude. MLS (alt = 87km) 120HR (daily T) • Vibrationally excited OH* radicals are produced via the reaction of 120HR (hourly T)

ozone and atomic oxygen:

T[K] O3 + H → OH(ν ≤ 9) + O2 (1) intensity values of P1 branch T[K]

• Radiative deexcitation of OH* contributes to the airglow, which we can observe from space and ground-based platforms.

• For low excitation levels (ν ≤ 6) it is assumed that the OH* emission is a measure of ambient temperature.

• Vertical air mass displacements affect mesopause temperatures via Fig 3.1: Left panel – Hourly variability in temperatures and relative intensities of the P1 branch from the OH(3-1) emission on January 23rd-24th , 2011. Uncertainties are indicated by dashed lines. Middle and right panel – Temperatures and associated uncertainties derived from hourly averaged OH(3-1) adiabatic cooling or warming. In particular the polar regions are spectra (blue squares), daily temperate average (red line), daily averaged temperatures derived from thermal oxygen emissions observed from the MLS strongly affected by this process due to the meridional summer-to- instrument onboard of the EOS Aura satellite at two different height levels.

winter pole circulation. • On Jan 23rd-24th, 2011 we can find that temperature changes are leading OH(3-1) intensity changes (fig. 3.1).

• Tidal as well as gravity wave forcings are crucial to understand the • The large scattering of hourly temperatures compared to their daily averages reflects the strong dynamical dynamical variability of the mesopause region at time scales ranging variability at shorter time scales. from minutes to hours. • Two mesospheric warming events appear in our retrieved temperatures after mid-January in 2011 and 2012. These forcings can lead to harmonic modulations in airglow temperatures and intensities with pronounced phase relations. • By comparison with MLS observations our derived temperatures are generally cooler. There is a growing consensus among observational and theoretical studies that these phase relations reflect vertical wave properties of associated forcings [3,6] . 4. Modelling the OH* airglow layer

• The interplay of mesospheric variability at different time scales is still modelled vertical OH(ν) 15. Jan 2011 – 0:00 (UTC) 15. Jan 2011 – 0:00 (UTC) poorly understood. Ground-based measurements will contribute to

our general understanding of the mesospheric response to climate

change [1]. [km] [km]

altitude 2. Instrumentation & Retrieval normalised OH(ν) concentration 2 3 OH (molec/cm ) O (molec/cm ) • A Bruker 120HR Fourier-Transform Spectrometer (FTS) is Fig 4.1: Left panel – Normalised vertical distribution of OH(ν) populations based on our model approach. Middle panel – Total OH* column based on our located at the AWIPEV research base in Ny-Ålesund, model approach. Right panel – Atomic oxygen concentrations weighted with vertical OH* concentration profiles. 78°55′N, 11°56′E and was upgraded to an 125HR in autumn 2012. • The population of vibrational states according to eq (1) not only depends on the radiative lifetimes, but also on the quenching rates with other species, mainly • Measurements of OH* airglow are N2, O2 and O. available since autumn 2007. Further • Other studies suggest that O concentrations significantly affect the vertical improvements to the automatic [5] operation during the polar-night have displacements among vibrational populations.

been implemented in 2010. In addition,  Can we link this dependency to study temporal changes in O concentrations? changes in the instrumental setup have significantly improved the signal-to- • To assess this question, we updated the McDade quenching model [4] with noise ratio by the winter season latest rate constants and use chemical profiles from the SD-WACCM4 [2] 2010/11. chemistry transport model to derive vertical OH(ν) populations (fig. 4.1). Fig 4.2: Correlation between the vertical displacement at peakpositions (each • Enhancements in mesospheric O concentration (right panel of fig. 4.1) could shifted by +0.5FWHM) of simulated OH(3-1) emission spectrum, 12. Mar 2011 - 01:00 UTC reflect an enhanced vertical downward transport from the thermospheric region. OH(9)/OH(5) layers and O

concentrations at same altitudes. Scattering points are constrained to the • In our model approach we can find a significant correlation between temporal polar-night between 75°N and 80°N. changes in O concentrations and vertical OH(ν) displacements (fig 4.2).

relativeintensity 5. Summary & Outlook

relative intensity transmission wavenumber [cm-1] • In this work hydroxyl emission measurements were established at an arctic station. OH emission bands wavenumber [cm-1] • First results already show interesting dynamic features at different time scales. The implementation of additional Fig. 2.1: above: modelled atmospheric Fig. 2.2: OH(3-1) emission spectrum OH* emission bands or even different airglow sources, such as the O2 (O-l) emission, would further contribute to transmission, below: calculated emission lines observed by 120HR FTS at 1cm-1 spectral from different OH* emission bands within the resolution. The fit corresponds to a our understanding of observed temperature/intensity phase relations with regard to gravity wave activity & tidal sensitivity range of the 120HR FTS temperature of 197.0 K  1.3 K, the spectral forcing. averaging time is 1 hour • We will improve our OH* temperature comparison with spaceborne instruments by weighting temperature • The spectral sensitivity of the 120/5HR FTS covers several profiles with OH* volume-emission-rate profiles observed from the SABER and SCIAMACHY instruments. OH* emission bands, while the OH(3-1) is one of the most preferable emissions with regard to atmospheric transmission • Following up the question in Sect. 4., we will compare SABER measurements of OH* and O with our model results and its low initial vibrational excitation level (fig. 2.1). in terms of their diurnal and seasonal variability to improve our understanding on the associated processes.

Acknowledgements & References • From the relative emission peak intensities we derive OH* • We gratefully acknowledge financial support to this project by the DFG as well as the [1] G. Beig. Trends in the mesopause region temperature and our present understanding - an update. Physics and Chemistry of the Earth, Apr 2006. [2] R. R. Garcia, D. R. Marsh, D. E. Kinnison, B. A. Boville, and F. Sassi. Simulation of secular trends in the middle atmosphere, 1950;2003. JGR., May 2007 rotational temperatures by means of forward modelling as financial travel support by the ZF/Impulse and the PIP programme of the University of [3] C. O. Hines , and D. W. Tarasick, On the detection and utilization of gravity waves in airglow studies Planetary and Space Science, 1987 Bremen [4] C. McDade. The altitude dependence of the oh(X2Σ) vibrational distribution in the nightglow: Some model expectations. Planetary and Space Science, Jul 1991. shown in fig. 2.2. [5] C. von Savigny, I. C. McDade, K.-U. Eichmann, and J. P. Burrows. On the dependence of the oh* Meinel emission altitude on vibrational level: Sciamachy • We thank Douglas Kinnison and Rolando Garcia, National Center for Atmospheric observations and model simulations. Atmos. Chem.Phys., Sep 2012. [6] G.R. Swenson, and C.S. Gardner. Analytical models for the responses of the mesospheric OH* and Na layers to atmospheric gravity waves. JGR Mar 1998 Research (NCAR), Boulder, Colorado, for providing the SD-WACCM4 data