Radar image tones may also vary in another systematic and controllable way. When a pulse of photon energy leaves the transmitter, its electrical field vector can be made to vibrate in either a horizontal (H) or a vertical (V) direction depending on antenna design. Most reflected pulses are parallel-polarized, i.e., return with the same direction of electric field vibration as the transmitted pulse. Thus, we get either a HH or VV polarization pairing of the transmitted and returned signals. However, upon striking the target, the pulses can undergo depolarization to some extent, so that reflections with different vibration directions return. A second antenna picks up cross-polarization that is orthogonal to the transmitted direction, leading to either a VH or HV mode (first letter refers to the transmitted signal). Some ground features appear about the same in either parallel or cross-polarized images. But, vegetation, in particular, tends to show different degrees of image brightness in HV or VH modes, because of depolarization by multiple reflecting branches and leaves. Compare the HV image of Harrisburg shown below with the previously shown (p. 8-3) K-band image of that area, obtained simultaneously in the HH mode.
Other factors contribute to the brightness or intensity of the returned signal. Two material properties provide clues about composition and surface state by the manner in which these attributes interact with the incoming pulses. One property is the dielectric constant (its symbol is the small Greek letter, κ, kappa), which is the ratio of the capacitance of a material to that of a vacuum (yielding dimensionless numbers; the value of κ is arbitrarily set to 1 for a vacuum). It is a measure of both the conductivity and reflectivity in terms of the electrical response of materials. This electrical property describes a material's capability (capacity) to hold a charge, which also measures the material's ability to polarize when subjected to an electric field. Radar waves penetrate deeper into materials with low dielectric constants and reflect more efficiently from those with high constants. Values for κ range from 3 to 16 for most dry rocks and soils, and up to 80 for water with impurities. Moist soils have values typically between 30 and 60. Thus, variation in reflected-pulse intensities may indicate differences in soil moisture, other factors being constant. Variations among rocks is generally too small to distinguish most types by this property alone.
The second material property is roughness. Materials differ from one another in their natural or cultivated state of surface roughness. Roughness, in this sense, refers to minute irregularities that relate either to textures of the surfaces or of objects on them (such as, closely-spaced vegetation that may have a variety of shapes). Examples include the surficial character of pitted materials, granular soils, gravel, grass blades, and other covering objects whose surfaces have dimensional variability on the order of millimeters to centimeters. The height of an irregularity, together with radar wavelength and grazing angle at the point of contact, determines the behavior of a surface as smooth (specular reflector), intermediate, or rough (diffuse reflector). To quantify the effect of different wave bands, a surface with a given small irregularity height (in cm) will reflect Ka band (λ = 0.85 cm), X band (λ = 3 cm), and L band ( λ = 25 cm) radar waves as if it were a smooth, intermediate, and rough surface, respectively. Other height variations produce different responses, from combinations of "all smooth" to "all rough" for the several bands used. This situation means a radar, broadcasting three bands simultaneously in a quasi-multispectral mode, can produce color composites, if we assign a color to each band (see below). Patterns of relative intensities (gray levels) for images made from different bands may serve as diagnostic tonal signatures for diverse materials whose surfaces show contrasted roughness.
As an example of multiband color composites, here is one made from SIR-C radar data that shows the Roter Kamm impact structure in western Africa:
Radar wavelengths also influence penetrability below target tops to ground surfaces. Depth of penetration increases with wavelength, λ. L and P band radar penetrate deeper than K or X bands. In forests, shorter wavelengths, such as C band, reflect mainly from the first leaves encountered, thus, from the canopy tops. At longer wavelengths, most tree leaves are too small to have much influence on backscatter, although branches will interact, so that canopies are penetrated to varying degrees. The image below, acquired by the SIR-C SAR during its Shuttle mission, shows how L-band radar has penetrated through the dense, continuous tropical vegetation cover of the Amazon Basin in Brazil to image the gently rolling terrain beneath.
Signals for all common radar bands pass through the fine droplets of moisture making up clouds, so that this condensation is effectively transparent or invisible to the beam. However, large ice crystals or raindrops do backscatter K-band radiation. Weather radar relies on this band to "picture" clouds and determine their movement using the Doppler effect (the apparent frequency shift that depends on motion toward or away from the receiver).