mav vary form person to person depending on each 
individual's experience and training. Although soils 
can be compared using one of these classifications. it 
appears from this studv that the two soil productivity 
modeling systems arc different and should not be 
compared in their ability to provide meaningful soil 
productivity classes. 
Table 3 presents the proportion of urban land usc on 
soils for cach soil productivity class derived using 
SRPG and LCC systems. For the SRPG modcl. thc 
variation of percent of soils under urbanization was 
similar in the two urban land use sources. In both 
cases. the most productive soils in Pennsylvania. that is 
those in the high producuvity category. were the most 
urbanized (9.4% for Landsat TM and 12.1% for 
DMSP/OLS). About 11% of the total urban arca in the 
high soil productivity class is urbanized. More than 
half (60.3) %of the urbanized arcas in Pennsylvania arc 
on the most productive soils (high and moderately high 
categories). The urban land use in this state covers 
9.4%. 4.8%. 2.0% and 3.8% of the total area of soils in 
the high, moderately high. moderate and low 
productivity classes. respectively. 
Table 3. Proportion (%) of urban land use for different 
soil productivity levels. 
  
  
  
  
  
Productivity Urban land source 
class 
Landsat TM DMSP/OLS 
Urban : Urban | Urban‘ : Urban 
* land : land” 
SRPG)J : LY) RASPY FIV 
High 9.4 10.8 12.1 11.9 
"Moderately | 48: 1495 Al 5:5: 190089 394 
an nien ES BCPA JAN DOM I 
Moderate 20 052324 24 "55205 
Below © PIS E. 08 ue 176 © 163 
LEO heu Done en maso. 
High 23 29.5 10.3 28.9 
"Moderately | 30 : 483 | 33° 1488 
2335] higni lob of bons i31oh oi eeepc 2 diwong 8 
Moderate 340 221712019: 13.4555 120.1 
Low Eo 167 22129. 229 ^ 280. 
  
  
  
  
  
Urban = % area of soils under urbanization. 
“Urban land = % area of soils under urbanization of the 
total urban land use in cach productivity class. 
The percent soils under urbanization increased from 
Landsat TM to DMSP/OLS. The changes were 
28.799. 14.53%. 20.0%. and 10.5 % for high, 
moderately high. moderate and low productivity 
classes. respectively. The average increase from 
Landsat TM to DMSP/OLS was 18.4%. 
The variation in the percent of soils under 
urbanization in cach soil productivitv class for the 
LCC svstem was different than that of the SRPG. In 
the LCC. the lcast productive soils were the most 
urbanized (Table 3). However. similarly to the 
SRPG. most of the urbanized land is on the most 
productive soils (16.7%). The urban land use 
productivity classcs accounted for 77.8% of the total 
urban land in the state. Similarly to SRPG, most of 
the urban land in Pennsylvania is on the most 
productive soils (high and moderately high 
catcgorics) as determined by thc LCC system. 
Vanations of proportions of soils in productivity 
classcs were similar in both urban land use class 
sources for the LCC system. 
4. SUMMARY AND CONCLUSION 
This studv provided the distributions of urban land 
usc relative to soil productivity in Pennsylvania, U.S. 
The most productive soils in Pennsylvania represent 
less than 5% of the total land. Unfortunately, these 
soils are also the most urbanized ones because the 
degrec of urbanization increased from low to high 
soil productivity classes. The LCC system was found 
to be less efficient in providing soil productivity 
classes compared to SRPG. Despite a large gap in 
ground resolution between Landsat TM and 
DMSP/OLS. the difference in urban land use class 
provided bv the two remote sensing svstem was 
relativelv small. Results of this study should be 
considered as a first step in determining the current 
status of state and regional soil resources. These 
results should help to clucidate trends and patterns in 
the human use of terrestrial ecosvstems and can also 
be used in a wide variety of other applications from 
hvdrologic modeling. land use planning and 
biodiversity assessment. 
5. ACKNOWLEDGMENTS 
This work was supported in part by the National 
Acronautics and Space Administration/Office of 
Earth. Science (NASA/OES) under Grant NAG 5- 
3856. 
462 International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 
Liu. 
Chi 
AC: 
ASF 
Niz 
plar 
GIS 
Ren 
Sinc 
Ter: 
plar 
Soil 
USL 
Vog 
Reg 
sour 
Ren