The interaction of laser light with a parcel of air with a known density structure can result in one of three reactions. The simplest of these reactions is reflection. Depending on the nature of the density profile, that part of the light that is not reflected can be refracted or scattered. The extent of the refraction and scattering is determined by the density of the particles found in the air. This thesis investigates two concepts that use the above mentioned interactions. The first, the colliding shock lens (CSL) was proposed by Buccellato, Lisi and Michaelis (1993). This device uses the graded index (GRIN) lens formed by the collision of symmetrically arranged shock waves to focus a laser beam. Unfortunately, the first reported colliding shock lenses had optical apertures of the order of millimeters. This is hardly useful in realistic laser systems whose beams typically have a diameter of 10mm. The major part of this thesis involves the scaling up of the optical aperture of the CSL while simultaneously maintaining a fairly short focal length. We show how the behaviour of the CSL varies with factors such as input energy, electrical diameter, geometry and various other factors. By optimising the physical parameters a 1.5cm diameter lens is obtained having a focal length of 1.5m. We develop a simple scaling theory and run a simulation based on the fluid in cell (FLIC) method, and find good correlation in both cases between the experimentally obtained results and the theoretically predicted ones. As a further development of the work on colliding shock lenses we introduce a cylindrical colliding shock lens. This device is shown to be able to line focus a laser beam of 1cm in diameter. At this stage the focus quality is still poor and suggestions are made for further improvements. Lidar is an acronym for light detection and ranging. Such systems are based on the scattering of laser light incident on a parcel of air. We discuss the results of a campaign conducted during the period of June to November 1994 to study aerosol concentrations over Durban. Particular attention is paid to low level aerosols due to sugar cane burning over the Natal coast. These aerosols are known to influence local climate and since vertical profile studies have never been carried out, this investigation gives some useful insight into the atmospheric dynamics. We find that in June (the begining of the burning campaign) the aerosol loading in the lower atmosphere is low. However, there are very stable aerosol layers at 3km and 5km. The density of the aerosols in these layers are decoupled. In September, the turbulent atmosphere over Durban is found to destroy structure in the aerosol layers. Nevertheless, the aerosol loading is high. Scattering ratios and extinction coefficients are calculated to show the long and short term evolution of the aerosols. A new coefficient (the low altitude aerosol coefficent - LAAC) is defined as an indicator for aerosol loading in the lower atmosphere. This coefficient is compared with total column ozone values over Durban. An anti-correlatory behaviour is noticed. We also report the detection of an extremely high aerosol layer (60km) over Durban. This layer is believed to be sodium. The profiles are compared to satellite data to verify the first ever detection of a constituent at these altitudes in Southern Africa.

Thesis (Ph.D.)-University of Natal, 1996.