as Canton, Wuhan and Shenyang. These cities, with populations in excess 
of 2,500,000, have much higher population densities and population growth 
does not appear to have been regulated to reduce the contrast between 
rural and urban areas - a stated Chinese planning goal. The reduced slope 
of the regression line for the 1970-74 data in Figure 4 does indicate 
that the population growth of the smaller cities may be controlled. 
The above example is based on work undertaken by Lo and Welch (1977) 
and is intended to demonstrate that quantitative information related to 
the urban planning policies of a large, developing country can be obtained 
from satellite data. In the absence of more reliable data, the equations 
can be employed to predict urban population. 
Urban Energy Utilization Patterns in the United States 
In order to monitor urban energy consumption on a regional or national 
scale, a data source which provides an instantaneous, uniform overview is 
required. The DMSP vehicles operating in circular, sun-synchronous, near 
polar orbits at altitudes of approximately 830 km provide this overview 
(Colvocoresses, 1975). They are equipped with scanning radiometers which 
are employed in a high-resolution mode (3.6 km IFOV at the nadir) to 
record nighttime imagery in the visual band (0.4-1.1 um) for swath widths 
of approximately 2930 km (Blankenship and Savage, 1974; U.S. Air Force, 
1974). The data are telemetered to ground stations and transformed into 
1:15,000,000 scale transparencies which display image contrasts in 16 
shades of gray.! These images may be obtained from the Space Science and 
Engineering Center, The University of Wisconsin. 
The DMSP images reveal the nighttime illumination patterns of major 
cities on a continental basis and qualitative assessments of urban energy 
requirements are easily made from the size, shape and intensity of the 
bright spots representing the cities (Croft, 1977; Figure 5). Comparison 
of DMSP images with a dot map of the 1970 population distribution in the 
United States reveals a striking correlation between the illuminated urban 
areas (IUA's) and the dot patterns (Bureau of Census, 1973). 
If considered in three-dimensional perspective, each IUA consists of 
a basal plane (x-y plane) which roughly corresponds to the populated 
urban area and a domed surface (z-axis) for which the z-coordinates are 
proportional to energy consumption (Figure 6). Thus, the volume of the 
IUA dome is a function of urban area and energy consumption, which, in 
turn are related to population. Although these relationships are 
intuitively recognized, it is much more difficult to establish their 
quantitative validity. Attempts to do so have involved the steps briefly 
described in the following paragraphs. 
  
The value of these data for energy studies would be greatly emhanced if 
digital tapes were made available. The dynamic range of the illuminated 
urban areas portrayed against a dark background can exceed that of the 
film.