Station, and made available to users as publication 
AY 209 (1980). The map at a scale 1:500,000 was pu 
blished by the Cooperative Extension Service of Purdue 
University in cooperation with the state Soil and Wa 
ter Conservation Committee of the Indiana Department 
of Natural Resources and the Soil Conservation Servi 
ce of the United States Department of Agriculture. 
The soil association map was manually digitized using 
the Purdue University/LARS digitizing system. This 
system is composed of a Talos table digitizer and an 
APPLE II Plus microcomputer. A complete documentation 
of this menu-driven system was prepered by Phillips 
(1983). 
The data capture (map digitization) consisted in the 
transformation of three map primitives, i.e. control 
points, boundaries (limits of soil units), and centroids 
into a format compa 4- ! 11 Q with digital computers. After 
he proce cc dp*-- capture was completed, the computer 
.om""'--'' 1 uata were transferred from the APPLE II Plus 
mj-^rocornpucer to the host (main) computer (IBM 370/158) 
where the data were stored and the activities of editing, 
coordinate transformation, and rasterization were per 
formed. Editing the digitized data was accomplished 
by manual and automatic editing routines using a gra 
phics terminal Tektronics 4045. 
Twelve control points, as illustrated in Figure 2, 
were used to derived statistically a biquadratic re 
gression model which was used to transform the digiti 
zed values in X and Y into longitude and latitude geo-• 
graphic coordinates. These data were subsequently 
transformed into an Albers equal-area cartographic pro 
ject. This was the projection (cartographic) selected 
for the Indiana geographic information system imple 
mented at the Laboratory for Applications of Remote 
Sensing (LARS) of Purdue University. 
The final step of the map input procedure was the 
rasterization process. During this process, the boun 
dary and centroid files, stored in addresses corres 
ponding to the Albers equal-area cartographic projec 
tion, were converted into an image file. The map units 
were filled-in with cells according to a predefined 
grid (500 m x500 m on the ground) and subsequently 
each cell was assigned a class code associated with 
the centroid file (0- 255). Figure 3 illustrates the 
different steps performed during the map input proce 
dure. The coding (fill characters) assigned to each 
of the 55 soil associations existing in Indiana and 
to the portion of Lake Michigan in the state, is shown 
in Table 1. 
For the construction of the attribute database (hie 
rarchical) , extensive use was made of the available 
information generated for the state soil associations 
of Indiana (Galloway et al., 1975). Other information 
not readily available in tables or as maps, at this 
level of detail, were obtained by interpretation, ex 
trapolation and generalization of the information pre 
sent in the description of the soil series forming 
each soil association (Galloway and Stainhardt, 1981; 
Franzmeier and Sinclair, 1982). 
For displaying purposes and generation of color out 
puts of the computer generated interpretive soil maps, 
the rasterized image was transferred to the image pro 
cessing device IBM 7350 "HACIENDA". 
3 RESULTS AND DISCUSSION 
Once the input of the data is completed and the ras 
terized data set and the corresponding attribute da 
ta set are stored in the database, the spatial infor 
mation can be easily retrieved, handled, analyzed and 
displayed. The degree of the analitical capabilities 
implemented in a system depends on the nature, purpo 
se and general objectives of the user. However, a well 
thought-out system will be one that is flexible enough 
to respond to the needs for input, analysis and display 
of different kinds of data required by the main user 
of the system. 
Regardless of the objective of the principal user, 
one element seems to be present in almost every digi 
tal information system. It is the element soils, de 
picting soil types as obtained from soil surveys. It 
occurs because of its relation to the fauna, vegeta 
tion and climate, and its strong interaction with o- 
ther natural resources elements. Soils, landuse and 
infrastructure constitute the fundamental and basic 
elements forming part of the database of geographic 
information systems for natural resources. The natu 
re and types of information available in a soil sur 
vey enables the generation of several interpretive 
soil maps. These maps can be used as new variables 
for analysis or modeling of resources to predict chan 
ges that may occur through time. 
The soil associations of Indiana in digital format 
displayed in the High Level Image Processing System 
(HLIPS) device IBM 7350 "HACIENDA", is shown in Figu 
re 3. The area estimates and percentage of occurran- 
ce of each soil association in Indiana is presented 
in Table 1. Soil association Crosby-Brookston present 
on nearly level surfaces of Wisconsinan age glacial 
till plains in central Indiana, constitutes the lar 
gest association in Indiana covering an area of ap 
proximately 703,050 ha or 7.4 % of the state, followed 
by the Morley-Blount-Pewamo association, occurring on 
end moraines and on rolling areas near streams that 
dissect till plains. This association covers an area 
of 636,825 ha or 6.7 % of the state. The smallest as 
sociations are Riddles-Tracy-Chelsea on the end morai 
nes in northwesten Indiana and Lyles-Ayrshire-Princen- 
ton developed on calcareous outwash sand and eolian 
fine sand deposited in Wisconsinan time covering an 
area of 13,500 and 15,600 ha respectively, or, 0.14 
and 0.16 percent of the state. 
The potential soil erosion was calculated using the 
Universal Soil Loss Equation. The factors of the 
USLE for each soil association were estimated by 
Brentlinger et al.(1979). This information was used 
to reclassify the digital soil association map into 
four potential soil erosion groups: low, medium, high 
and very high. The potential soil erosion map is il 
lustrated in Figure 4, and the area estimates for each 
erosion group are shown in Table 2. This interpreti 
ve information can be used in conjuction with landuse 
data to predict the erosion hazard or gross erosion 
in the state. It can also be related to slope, land- 
use and proximity to streams to determine agricultu 
ral pollution due to erosion and to estimate sedimen 
tation hazards and the related dangers of floodings. 
Soil maps in Indiana are used in reassessment of a- 
gricultural land. The basic aim of any assessment ac 
tivity is the equal treatment of all individual land- 
owners. Yahner (1979) described the procedures used 
in agricultural land reassessment using estimates of 
corn yields. Each state soil association has been as 
signed an estimated corn yield ^alue. Figure 5 illus 
trates the corn yield estimate map of Indiana after 
grouping the values in high, medium and low yield va 
lues for each soil association. Because of the reso 
lution (scale) of the data and the generalization in 
volved in the creation of the soil associations, so 
me problems and difficultities may exist in the actual 
assessment of individual farm evaluation. However, 
it can be used to obtain rapid information on the ap 
proximate value of agricultural land. 
The possibility of deriving different interpretive 
maps from the soil association map can be very useful 
in creating a set of illustrative material for didac 
tical purposes . One such example is the possibility 
of showing graphically the influence of the soil for 
ming factors in determining the actual soil characte 
ristics. Figure 6 illustrates the parent material 
from which the Indiana soils were developed. It de 
picts the various kinds of materials including old 
sedimentary rocks in the southern part of the state, 
defferent thickness of loess deposits over glacial 
till, alluvial, lacustrine and eolian deposits from 
which the soils were developed. 
Topography or relief has a great influence on the 
processes of weathering and soil formation. It in-