Optical and physical properties of stratospheric aerosols from balloon measurements in the visible and near-infrared domains. III. Presence of aerosols in the middle stratosphere.

The aerosol extinction measurements in the ultraviolet and visible wavelengths by the balloonborne spectrometer Spectroscopie d'Absorption Lunaire pour l'Observation des Minoritaires Ozone et NOx (SALOMON) show that aerosols are present in the middle stratosphere, above 25-km altitude. These observations are confirmed by the extinction measurements performed by a solar occultation radiometer. The balloonborne Laboratoire de Météorologie Dynamique (LMD) counter instrument also confirms the presence of aerosol around 30-km altitude, with an unrealistic excess of micronic particles assuming that only liquid sulfate aerosols are present. An unexpected spectral structure around 640-nm observed by SALOMON is also detectable in extinction measurements by the satellite instrument Stratospheric Aerosols and Gas Experiment III. This set of measurements could indicate that solid aerosols were detected at these altitude ranges. The amount of soot detected up to now in the lower stratosphere is too low to explain these measurements. Thus, the presence of interplanetary dust grains and micrometeorites may need to be invoked. Moreover, it seems that these grains fill the stratosphere in stratified layers.

Determination of polar stratospheric cloud particle refractive indices by use of in situ optical measurements and T-matrix calculations.

A new algorithm to infer structural parameters such as refractive index and asphericity of cloud particles has been developed by use of in situ observations taken by a laser backscattersonde and an optical particle counter during balloon stratospheric flights. All three main particles, liquid, ice, and a no-ice solid (NAT, nitric acid trihydrate) of polar stratospheric clouds, were observed during two winter flights performed from Kiruna, Sweden. The technique is based on use of the T-matrix code developed for aspherical particles to calculate the backscattering coefficient and particle depolarizing properties on the basis of size distribution and concentration measurements. The results of the calculations are compared with observations to estimated refractive indices and particle asphericity. The method has also been used in cases when the liquid and solid phases coexist with comparable influence on the optical behavior of the cloud to estimate refractive indices. The main results prove that the index of refraction for NAT particles is in the range of 1.37-1.45 at 532 nm. Such particles would be slightly prolate spheroids. The calculated refractive indices for liquid and ice particles are 1.51-1.55 and 1.31-1.33, respectively. The results for solid particles confirm previous measurements taken in Antarctica during 1992 and obtained by a comparison of lidar and optical particle counter data.

Optical and physical properties of stratospheric aerosols from balloon measurements in the visible and near-infrared domains. II. Comparison of extinction, reflectance, polarization, and counting measurements.

The physical properties of stratospheric aerosols can be retrieved from optical measurements involving extinction, radiance, polarization, and counting. We present here the results of measurements from the balloonborne instruments AMON, SALOMON, and RADIBAL, and from the French Laboratoire de Météorologie Dynamique and the University of Wyoming balloonborne particle counters. A cross comparison of the measurements was made for observations of background aerosols conducted during the polar winters of February 1997 and January-February 2000 for various altitudes from 13 to 19 km. On the one band, the effective radius and the total amount of background aerosols derived from the various sets of data are similar and are in agreement with pre-Pinatubo values. On the other hand, strong discrepancies occur in the shapes of the bimodal size distributions obtained from analysis of the raw measurement of the various instruments. It seems then that the log-normal assumption cannot fully reproduce the size distribution of background aerosols. The effect ofthe presence of particular aerosols on the measurements is discussed, and a new strategy for observations is proposed.