es with new 
Idziewicz 
ition and 
(Ventura 
nventory 
ory 
uring 1991 
Park 
JORADB). 
] of 200, 
' on the size 
inland Unit 
Coch, 
atic and 
on and plot 
orporated 
INFO 
Is, each 
he 
) can be 
as 
th. 
ded 
oint 
used in 
y on- screen 
ayers such 
ition to help 
TUs) 
s several 
generated 
lencing 
Is, 
U polygon 
ispect, soil 
or the 
only where 
0 contribute 
'ailable 
on all 
the Arc/Info 
maps. 
These layers were further processed into an ITU 
coverage by polygon overlay. 
Since the vegetation types and boundaries were 
determined primarily on infrared aerial 
photography, two photo interpretation problems 
were encountered -- cases with an identified 
vegetation type but unclear boundary, or cases 
with a determined boundary but uncertain 
vegetation type. FLORADB and ITUs combined 
played crucial roles in assisting air photo 
interpretation (Ventura and He, 1993). Figure 2 
shows some of the decision logic used to 
incorporate information from these in the 
interpretation process. In a GIS environment, 
the subjectivity of some interpretation decisions 
may be reduced to a minimum by referencing 
the information such as ITUs and FLORADB. 
  
  
  
  
  
  
  
  
before processing 
  
  
  
GIS operation on point and polygon coverage 
Plot points can be linked with flora database 
indicating what species are found on the site 
and what species composition can be generated. 
ITU boundaries can be used as references when 
the vegetation boundaries on the air photo 
are difficult to delineate. 
  
   
  
  
Figure 2. GIS operation on ITU polygon and plot point coverage 
Vegetation Classification: Compositions, 
Communities, and Forests 
Species composition, plant community, and 
forest stand are the three vegetation 
classification levels established for the Apostle 
Island National Lakeshore. Species composition 
and community, as defined primarily by 
overstory tree species, were used in describing 
hierarchical southern-mesic forest of 
southeastern Wisconsin (Levenson, 1981), and 
in most previous vegetation mapping in the 
61 
Apostle Islands (Anderson et a/., 1979, 1982, 
1983; Dobie, 1977; Fraundorf, 1984; 
Harrington, 1982; Heidel, 1977; Hildebrandt, 
1978; Larsen, 1975; Stadnyk, ef al. 1974). 
In this study, the following conventions were 
used to establish a species composition, the 
lowest (most disaggregated) level in the 
classification: 
* The first-listed species is the first 
dominant, the second-listed species is 
the second. Up to three species may be 
included in a species composition.