Full text: Proceedings of the international symposium on remote sensing for observation and inventory of earth resources and the endangered environment (Volume 2)

  
  
   
   
  
     
  
  
  
     
  
  
   
  
   
   
  
   
  
  
  
  
    
  
  
  
  
   
    
   
  
  
  
  
  
  
  
   
   
Populations of China's Urban Centers 
The population of China is variously estimated at between 800 to 
950 million, and the last comprehensive census occurred in 1953. In 
the absence of current maps and census statistics, it has been possible 
to establish quantitative indicators of the relationships between urban 
growth and the stated government planning policies of the world's most 
heavily populated country from measurements on Landsat images. 
A histogram constructed from published 1953 census data for 124 
cities throughout China reveals more than 70 with populations in excess 
of 100,000 (Chen, 1967; Figure 1). As the boundaries of cities this 
size are easily delineated on Landsat images, investigations were under- 
taken to develop regression equations which would relate population to 
measurements of the built-up urban areas obtained from map or image data 
(Figure 2). Nordbeck (1965), for example, stated that the built-up area 
(A) of a settlement is related to its population (P) by an equation of 
the form: 
b 
A = aP 
This simple equation has been shown to produce correlation coefficients 
(r) of approximately *0.9 in studies of western cities for which reliable 
census data and area measurements obtained from aerial photographs were 
available (Ogrosky, 1975). A limited analysis of satellite photographs 
of non-Asian cities produced comparable results (Tobler, 1969). 
For the China study, the equation was modified so as to solve for 
population in terms of built-up urban area, i.e. 
p = aa) 
To test the validity of this equation the populations of the 124 cities 
were plotted on doubly logarithmic paper against area measurements (of 
the same cities) obtained from 1:250,000 scale U.S. Army Map Service sheets 
prepared in 1951-56. Despite the evident weakness that the built-up 
urban areas depicted on the maps may not correspond to the administrative 
boundaries of the cities, and that city shape and size may exhibit regional 
variations, the data points generally fit the prescribed straight line 
pattern (Figure 3). The regression equation for these data points as 
determined by the methods of least-squares is: 
P = 74696 A-7246 
r = +0.75 
This procedure was then extended to 13 large cities recorded on 
Landsat images in 1972-74 for which 1970 population figures had been 
previously published (Chen, 1973). Built-up areas were measured from 
1:500,000 scale color composites and plotted on doubly logarithmic paper 
with respect to population for 10 cities of 500,000 to 2,000,000 people. 
The regression line fitted to these data gives the following equation 
(Figure 4): 
P - 213993 A-4137 
r = +0.82 
Although this equation is representative of cities with populations of 
less than 2,000,000, it cannot be applied to large regional centers such 
   
Fig. 
1953 
citie 
POPULATION 
  
Fig. . 
Chine: 
censu: 
area I 
1:250. 
perio:
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.