Simultaneous Ground-Based Measurements of ClO, OClO and BrO at the NDSC-Station Ny-Ålesund Spitsbergen

B. Franke, U. Klein, J. Langer, U. Raffalski, B.-M. Sinnhuber, F. Wittrock, J. P. Burrows and K. F. Künzi Institute of Environmental Physics, University of Bremen, PO Box 33 04 40, D- 28334 Bremen, Germany e-mail: bernd@schalk.physik.uni-bremen.de

INTRODUCTION

Chlorine monoxide (ClO) and bromine monoxide (BrO) play an important role in the catalytic ozone depletion mechanism in the northern polar stratosphere. The concentration of chlorine dioxide (OClO) has been used as an indicator for the chlorine and bromine activation. We focus on the question, whether OClO is only a poor quantitative indicator for strong chlorine activation or not. We have compared ClO microwave measurements to uv/visible measurements of OClO and BrO in order to investigate their correlations. The reaction of BrO with ClO is the main source of OClO:

1)   BrO + ClO    Br + OClO                               k1

Photolysis is the major sink of OClO:
2)   OClO + hn     ClO + O                                  J2

The photolysis coefficient J2 of reaction 2 depends on the solar zenith angle. In a steady state a linear dependence of OClO and ClO is to be expected:
3)   [OClO] = k1[ClO] [BrO]/J2(SZA)

The photolysis of the dimer Cl2O2 and the inverse recation are responsible for the diurnal variation of chlorine monoxide:
4)   Cl2O2 + hn     ClO + ClO                             J4

5)   ClO + ClO + M     Cl2O2 + M                     k5

RADIOMETER FOR ATMOSPHERIC MEASUREMENTS

The ground-based passive microwave Radiometer for Atmospheric Measurements (RAM) detects the set of weak ClO emisson lines around 204.35 GHz. The subtraction of daytime and nighttime spectra eliminates instrumental effects. This procedure is feasible, because ClO in the layer of main interest around 20 km is only present during sunlight. A scaling factor method was applied to the spectra to obtain ClO peak volume mixing ratios (VMR) and column densities [1]. ClO measurements from February to the beginning of April, 1997 are presented.

DIFFERENTIAL ABSORPTION SPECTROMETER

The Differential Absorption Spectrometer (DOAS) instrument operates in the near-UV and visible spectrum. Zenith sky spectra are measured every day for solar zenith angles smaller than 95°. DOAS provides total vertical column densities of ozone and NO2 and total slant column densities of halogen oxides. For this study we used the slant column densities of OClO and BrO [2]. Both instruments are part of the NDSC-Station in Ny-&Arin;lesund (78.9°N, 11.9°E), Spitsbergen.

MEASUREMENTS

During the period from February to April 1997 Spitsbergen was located inside the polar vortex and low temperatures led to the formation of polar stratospheric clouds [3]. Heterogeneous chemical reactions on the surface of cloud particles have led to chlorine activation. For most of the observation period, we have detected enhanced ClO volume mixing ratios in the lower stratosphere [1], leading to significant ozone depletion see e. g. [4].

EXAMPLES:

March 17 (day 75): In figure 1 and 2 the diurnal variations of ClO and OClO on March 17 are shown. Due to the photolysis of the dimer Cl2O2 the ClO mixing ratio increased with decreasing solar zenith angle, whereas the amount of OClO decreased due to its photolysis.

Figure 1: The Peak VMR of ClO at an altitude of 22 km and the slant column density of OClO vs. Solar Zenith Angle (SZA). The maximum error of the OClO column density is 25%. Measured in the morning hours of March 17 1997.

Figure 2: Same as figure 1, but measured during sunset of March 17 1997.

March 20 (day 78): Three days later air parcels of higher chlorine activation were observed over Ny-Ålesund (figure 3 and 4). We measured a significantly enhanced amount of ClO as shown in figure 5. Although the ClO mixing ratio had strongly increased, the measured OClO amount did not change significantly.

Figure 3: Same as figure 1, but measured in the morning hours of March 20 1997.

Figure 4: Same as figure 1, but measured during sunset of March 20 1997.

Figure 5: OClO, ClO and BrO column densities during the measurment period from the February 9 (day 40) to April 10 1997 (day 100).

CONCLUSION

A clear correlation between the OClO and ClO cannot be seen over the whole observation period (figure 5). It appears that OClO is not a good indicator for ClO activation. These measurements confirm the model calculations of Sessler et al. [5], who stated that OClO is only a good quantitative indicator for low chlorine activation. For the correlation of OClO and BrO see [2].

REFERENCES

GoUp