Figure 3. Single orbit DMSP/OLS swath. No thresholding or 
other processing. See Figure 4 for effects of thresholding. 
  
3.12 Landsat-based Urban Land Cover Change 
Multitemporal mosaics were made of four each Landsat MSS 
[1976] and Thematic Mapper [1987] to test the utility of 
unsupervised classification approaches for urban land cover 
calculations. 
After classification we found that even though there was 
approximately a four-fold difference in the spatial resolution 
of the sensors [ca. 0.36 ha versus 0.09 ha] the results of the 
classification were reasonable. The results of the 
classification [Table 1] showed a stable area of bare soil + 
desert classes across the 11 year period [30.5 vs 29.9% of 
study area], an increase in urban land cover from 6.6 to 12%, 
a slight decrease in vegetated area from 58.2 to 55.6 % and a 
relatively stable amount of water and other classes at < 2%. 
We found considerable difficulty in separation between arid 
land cover types and urbanized land covers, especially in 
areas where little vegetation was mixed with the urban land 
cover. There was no direct field validation so the 
classification scheme may have considerable error, but it 
does present a viable method when appropriate data sets are 
available. The latest data we could acquire was 1987, so this 
estimate may not accurately depict current conditions. As a 
counterpoint, very recent nighttime images are available, for 
this area, and at a much lower cost in terms of both 
acquisition of the data and analysis. Figure 5 shows the 
typical landscape we were dealing with in this classification 
procedure. This Thematic Mapper image shows urban, 
irrigated agriculture and arid land cover classes in the 14,400 
km? study area. 
Table 1. % Land Cover Change 1976-198 
Land Cover Type MSS76 TM87 
Bare Soils 4.15 2.37 
Desert 26.33 27.52 
Urban 6.58 11.95 
Urban/mixed 1.69 1.48 
Vegetation 58.16 55.60 
Water 1.78 0.93 
Unclassified 1.31 0:15 
Total 100.00 100.00 
446 
  
     
   
Figure 4. Raw DMSP/OLS, thresholded and saturated pixels 
only. 
    
  
3.4 Urban Occupation of FAO Soils - The 
determination of the occupation of specific FAO soils 
classes by an enhanced "city lights" [Imhoff et al, 
1997a,b] product made from the raw NOAA NGDC 
DMSP/OLS ‘stable lights’ product yielded results in close 
agreement with literature references which have stated that 
about 4% of the area of Egypt is urbanized. Our analysis 
shows 3.7% urbanization of the whole country. The analysis 
was carried out at the country level. The nine soil and cover 
types in Egypt and the percentage of each of those soil types 
covered by “city lights” urban land cover is shown in Table 
2. Figure 6 shows the spatial distribution of soils in Egypt. 
The soil types are ranked in descending order of area 
coverage. Of these soil types, the Fluvisols are by far the 
most appropriate soils for agriculture. According to our 
analysis, almost 33% of the best soils in Egypt are under the 
urban land cover type, thereby, according to our hypothesis, 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998