No abstract

No abstract

A novel aerosol lidar inversion method based on the use of multiple-scattering contributions measured by a multiple-field-of-view receiver is proposed. The method requires assumptions that restrict applications to aerosol particles large enough to give rise to measurable multiple scattering and depends on parameters that must be specified empirically but that have an uncertainty range of much less than the boundary value and the backscatter-to-extinction ratio of the conventional single-scattering inversion methods. The proposed method is applied to cloud measurements. The solutions obtained are the profiles of the scattering coefficient and the effective diameter of the cloud droplets. With mild assumptions on the form of the function, the full-size distribution is estimated at each range position from which the extinction coefficient at any visible and infrared wavelength and the liquid water content can be determined. Typical results on slant-path-integrated optical depth, vertical extinction profiles, and fluctuation statistics are compared with in situ data obtained in two field experiments. The inversion works well in all cases reported here, i.e., for water clouds at optical depths between ~0.1 and ~4.

Classical optics holds that the extinction cross of particles should be equal to twice their geometric cross section, in the limit where the particles are much larger than the wavelength. It follows that the extinction coefficient of such large scatterers should be independent of wavelength. Snowflakes are much larger than the wavelengths of visible and infrared radiation, yet many investigators have found that the visible and infrared extinction coefficient of falling snow measured with transmissometers is wavelength dependent. This dependency is known to be a result of the scattering contribution to the transmissometer signal. Furthermore, many measurements in the visible and infrared show that extinction values measured simultaneously with two transmissometers are linearly related up to at least 12 km(-1). The slope depends on the wavelengths and optical characteristics of the transmissometers. We show that for small values of extinction, the observations can be explained by taking into account single-scattering contributions to transmissometer signals. For high values of extinction, a multiplescattering model gives good agreement with measurements.

No abstract