The applet demonstrates the scintillation of the surface reflected component of a signal from a distant transmitter as the sea surface is broken by waves. This physical model gives an insight into the effects of random surface reflections on the operation of DF systems.
Sea roughness is specified by the rms slope of surface reflection areas or plates, relative to the local horizontal surface.
Transmitter /receiver antenna heights, signal frequency and spacing are adjustable.
The upper display shows the spatial distribution of reflection planes. The number are adjustable to control the presentation of the set of randomly oriented and positioned elliptical planes. Other randomly generated planes are viewed by mouse-clicking the display. The sizes of plates are automatically adjusted so that all direct the same power towards the receiver (nominally one quarter of the direct power).
The colored ellipses at the primary Reflection Point represent the Fresnel zones corresponding to the extra path lengths listed at the side of the central display. Displayed planes around this point arrive at the receiver closely in phase, whereas plates at further distances will combine with larger phase differences.
the largest phase and time delay signals arise from reflection planes close to either the transmitter or receiver. This is important as it will be noticed that reflection planes close to the transmitter and receiver are smaller to produce the same receiver power.
The lower display plots random plane data in an angle-range format.
The data above the plots reports the cursor position, the local plate area, plate width-to-length ratio, and the relative path delay in nanoseconds between the direct path and the local plate reflection path.

User Notes:
This applet clearly shows that angle scintillation occurs over most of the propagation path, but that the largest angle spreading that would be observed by a DF system is due to reflectors close to the receiver; the angle spreading reducing as a function of range. This is due to the lateral spread of useful plates never deviating far from the main path, being constrained by the practical values of lateral surface slope.
Another interesting conclusion that can be drawn from the model is that for ESM, as opposed to monostatic radar, it is inappropriate to treat the sea surface reflectivity in terms of RCS/m².
Although the fields of view of both transmitter and receiver may be very wide, the geometry ensures that only surface reflection centers, in a narrow corridor along transmitter/receiver path, can direct signals to the receiver.
For fairly calm seas, within the Rayleigh criterion, the dominant feature is specular reflection from the primary reflection point with added noise. Angle noise is more apparent around nulls, from close-in small reflection regions meeting the slope conditions.