ACKNOWLEDGEMENTS 
tepartment 
and 
. grazed 
. with 
with 
pland 
ure 
d and 
land 
4.3 Shapes and internal structures 
It is obvious from the above that the shape of the 
settlement registered on the SIR-A data may not be 
correct because of the strong backscatter. This 
is particularly so for large and less compact 
settlements. However, it was possible to identify 
the density of buildings inside a settlement under 
the favorable imaging conditions. St. Joseph, Mo. 
is the best among the five strips in displaying 
distinctly the shapes and internal structures of 
some settlements (Fig. 3). Apparently, the radar 
azimuth (N 6°30'E) has much to do with this 
result. In large settlements, it was possible to 
identify industrial areas and airfields easily, 
the former usually exhibiting bright return and 
the latter giving very low return from their 
smooth surfaces, as exemplified in Mobile, A1. 
(Fig. 4). 
5 CONCLUSIONS 
6 
I wish to thank Dr. Charles Elachi of Jet 
Propulsion Laboratory, California Institute of 
Technology, Pasadena, California and the National 
Space Science Data Center for providing me with 
the SIR-A data, which makes this research 
possible. 
REFERENCES 
Bryan, M.L. 1979. The effect of radar azimuth 
angle on cultural data. Photogrammetric 
Engineering and Remote Sensing 45:1097-1107. 
Cimino, J.B. & C. Elachi 1982. Shuttle Imaging 
Radar-A (SIR-A) experiment. Pasadena, 
California: Jet Propulsion Laboratory. 
Doyle, F.J. 1984. Surveying and mapping with 
space data. ITC Journal, No. 4, 314-321. 
ilation (P). A 
;rowth model in 
(Lo and Welch, 
le area of each 
;h 1-mm square 
; and then from 
It was found 
the individual 
correlation of 
significance. 
.1 the measured 
factor of 1.5X 
y be caused by 
but careful 
;nificantly the 
uced a glare 
size of the 
that this 
t and could be 
me, the 1980 
ettlements in 
with measured 
of a linear 
form of the 
above. The 
verall strong 
d area in the 
egions. It is 
, Al. (Fig. 4) 
much stronger 
ear regression 
wth model, a 
lement growth 
o regions than 
ndicated that 
ing settlement 
could produce 
This research has indicated that great potential 
exists in the use of SIR-A images for human 
settlement analysis because the settlements give 
strong backscatter, which renders them easily 
detectable. It was observed that the radar 
viewing geometry has a great impact on the 
settlements' detectability, and a preliminary 
observation seems to suggest the importance of 
orienting the radar look direction orthogonally to 
the settlement to give a better detection 
capability. 
Cultural factors have also affected the 
detectability. The highly compact Chinese 
settlements which were formerly restricted by 
walls in the North China Plain can be more easily 
distinguished, irrespective of size, than their 
American counterparts. 
Within the American environment, the 
detectability of settlements varies, not only 
because of differences in radar azimuths but also 
because of differences in terrain height, soil 
type, vegetation and land cover type. It appears 
that high terrain, cropland or grassland cover, 
and Mollisols soil with an irregular plain 
landform are favorable factors that enhance 
image-background contrast in settlement 
detectability. The Gulf Atlantic Coastal Plain 
emerges as the worst of all four regions in 
settlement detectability by a combination of poor 
environmental contrast and an unfavorable radar 
azimuth angle. It is also noteworthy that large 
settlements with a population over 100,000 and 
having a large spatial spread are more difficult 
to detect than small settlements. 
An important finding from this research is the 
very strong relationship that exists between map 
areas and image areas of settlements, although the 
image areas consistently exaggerate the actual 
areas of the settlements. The image areas can be 
usefully employed in the allometric growth model 
for population estimation in all four geographic 
regions. Although the shapes of the settlements 
are not always correctly recorded on the SIR-A 
images, one can still detect building density and 
other functional units of the settlement in cases 
when the optimum conditions of imaging have been 
achieved. 
To conclude, the SIR-A data in the present 
optically processed and tilt corrected 
two-dimensional image form possess adequate 
resolution and imaging quality for use in human 
settlement analysis. They also complement nicely 
other forms of high-resolution space data in 
yielding timely data of the settlements under 
adverse weather conditions. Already SIR-A data 
have been successfully comDinea with Landsat MSS 
data to provide improved spatial data for human 
settlement analysis (Welch, 1984). 
Ford, J.P., J.B. Cimino & C. Elachi 1983. Space 
shuttle Columbia views the world with imaging 
radar: the SIR-A experiment. Pasadena, 
California: Jet Propulsion Laboratory. 
Hardaway, G. & G.C. Gustafson 1982. Cardinal 
effect on Seasat images of urban areas. 
Photogrammetric Engineering and Remote Sensing 
48:399-404. 
Henderson, F.M. & Anuta, M.A. 1980. Effects of 
radar system parameters, population, and 
environmental modulation on settlement 
visibility. International Journal of Remote 
Sensing 1:137-151. 
Lo, C.P. 1984. Chinese settlement pattern analysis 
using Shuttle Imaging Radar-A data. Interna 
tional Journal of Remote Sensing 5:959-967. 
Lo, C.P. & R. Welch 1977. Chinese urban popula 
tion estimates. Annals of the Association of 
American Geographers 67:246-253. 
Welch, R. 1984. Merging Landsat and SIR-A image 
data in digital formats. Imaging Technology in 
Research & Development, July, 11-12.