Turbulence in the atmosphere has a broad range of consequences. Among other things it can pose a hazard to aircraft and give rise to mixing between air masses which would otherwise remain distinct. Identification of turbulent regions is therefore useful because of the importance of turbulence in the vertical diffusion of pollutants. As indicated previously intense turbulence is also a significant hazard to aviation and thus the determination of the frequency and distribution of intense turbulence relative to other atmospheric phenomena is also of significant importance. Although instrumented aircraft and balloon-borne probes have important roles to play in studies of this phenomenon, their usefulness is limited to specific events. Radars, by contrast, offer the potential for continuous observations to be made over a wide altitude extent.

Turbulent regions are usually identified by the  spread of velocities within the signal portion of a VHF radar return Doppler spectrum since they in part represent the variance of vertical velocities arising from turbulent motions. However, the spectral width is also influenced by the horizontal wind speed as a result of the finite beam width. This so-called beam-broadening component tends to dominate the observed spectral widths for high wind speeds. The technique is therefore of limited use in connection with frontal zones, for example, which are expected to be turbulent but which are characterised by high wind speeds. An additional indication of the presence of turbulence can be gained by considering the ratio of signal powers observed in the vertical direction and off-vertical direction. A small value of this so-called aspect sensitivity factor suggests that the radar signals are being scattered from isotropic irregularities of refractive index. A large value suggests that the signals are either being partially reflected from sharp, horizontally coherent changes in refractive index or being scattered from anisotropic irregularities.

An initial investigation has begun to evaluate the potential of the modified Rice parameter as a further diagnostic for the indentification of the presence of turbulence. This parameter is based on the statistical properties of the in-phase and quadrature components of the radar return signals. If the radar signals are scattered from an ensemble of roughly similar refractive index irregularities, which is expected from a region of turbulence, the signal components should have a Rayleigh distribution of amplitudes. If, on the other hand, the radar signals are predominantly returned from a single refractive index feature, such as might be expected under non-disturbed conditions, the signal components will tend to have a more Ricean distribution.