The surface distribution of composite incidental man-made radio noise power in metropolitan areas displays frequency and range variations which are predictable from a knowledge of the propagation characteristics of irregular surface terrain for low-height antennas and the noise-source spectral power densities for a Poisson distributed random noise process. The observed frequency variation of composite surface noise power in the very-low-frequency (VLF) through the lower very-high-frequency (VHF) band is porportional to the product of 1) the inverse first power of frequency, introduced by irregular terrain basic path loss dispersion for low antennas, and 2) the inverse quadratic frequency dependence of the noise-source spectral density. In the upper portion of the VHF band and in the ultrahigh-frequency (UHF) range, the basic path loss dispersion remains unchanged while the emission spectrum of the surface noise sources asymptotically approaches that of band-limited white noise. The observed first-order surface range dependence for composite man-made noise within an urban area is predictable from the measured variation of irregular terrain basic path loss for low-height antennas and displays a somewhat smaller exponent than the free space propagation function. The general range function for composite metropolitan area man-made noise power displays a quadratic exponential variation with distance from the urban center. The level of man-made noise power, observed from 1 to 5 miles above metropolitan areas, can be related to the surface noise distribution, the radiation pattern of the receiving antenna, and the coordinates of the airborne detector, relative to the urban center. The noise level, the noise spectral density, and the position dependence of the airborne noise power are predictable from this information plus a recognition that, as the receiver field of view increases with altitude or beamwidth, the amplitude distribution of the surface sources approaches a unit normal distribution in conformity with the central limit theorem for multiple independent random sources. Tests of the validity of the theoretical predictions of the range and frequency dependence of both surface and airborne man-made incidental radio noise power in the VHF-UHF bands have been performed, using accumulations of both surface and airborne noise data obtained from measurements in and above urban areas in Europe, Asia, Australia, and North and South America. Theoretical and experimental comparisons have confirmed the validity of the noise generation and propagation models and support their use for calculating the deterministic characteristics of metropolitan area incidental man-made noise.