393 
[onongahela 
inventories 
Management 
e to obtain 
ality of red 
ted species, 
res of these 
a red spruce 
es. 
and mortality 
as. 
a red spruce 
and fir re- 
hanical fac- 
mortality of 
h results of 
ted in three 
red spruce, 
ion. Survey 
northeastern 
Vermont; (2) 
Lnia; (3) and 
occupy the 
estern North 
southwestern 
of northern 
in extensive 
eas have not 
jse they have 
iation by C. 
e mortality, 
h spruce and 
range of red 
Aerial photography, using several photo scales and 
film formats, has been an integral part of these 
inventories. With one exception, color infrared 
(CIR) film has been used exclusively (Table 1). 
Aerial photos have been used to identify forested 
areas containing a spruce component and classify 
them into a series of vegetation types. Vegetation 
types were then stratified into a series of 
mortality classes. Aerial photos have also been 
used to make counts of dead and declining trees on 
fixed area photo plots. Overall survey designs and 
aerial photographic parameters varied somewhat by 
survey area. 
4.1 Northeastern United States 
In 1984, an extensive inventory of the Adirondack 
Mountains and Tug Hill Plateau regions of New York, 
and portions of Vermont and New Hampshire was 
initiated (Weiss et al. 1986) (Fig. 3). This 
inventory was based on estimates made from 55 
randomly selected 3150 ha aerial photo sample 
blocks. Each block was photographed with CIR film 
at a scale of 1:8000. Blocks were stratified into 
four vegetation classes which contained a spruce and 
fir component (Table 2) and three mortality classes 
(Table 3). A series of ca 1 ha. photo plots were 
randomly established in each of the mortality 
classes within three of the vegetation classes for 
tree counts. Aerial photo data was adjusted by a 
small sample of ground plots using double sampling 
with regression (Wear et al. 1966). 
During the following year, selected areas of the 
Adirondack Mountains of New York and portions of 
Vermont and New Hampshire, including the Green 
Mountain and White Mountain National Forests, were 
photographed. Complete coverage of these areas with 
1:24000 scale CIR film was acquired. These photos 
are being used to map the location of concentrations 
of moderate and heavy spruce and fir mortality in 
vegetation types with a spruce component. 
4.2 West Virginia 
In 1985, an inventory to estimate levels of red 
spruce decline and mortality was conducted on the 
Monongahela National Forest and adjoining private 
lands in West Virginia (Mielke et al. 1986). Aerial 
photography used for this inventory was high 
resolution panoramic photography taken the previous 
year by a NASA ER-2 high altitude earth resources 
reconnaissance aircraft (Nadir scale = ca 1:30000). 
This type of photography has been used successfully 
for a number of forest damage assessment 
applications (Ciesla et al. 1982). Complete 
coverage of all of the high elevation red spruce 
forests in the state was available. 
This photography was used to classify forests with 
a red spruce component into vegetation and mortality 
classes (Tables 2 and 3). The small scale of the 
panoramic aerial photography precluded 
identification of tree species and counts of dead 
and declining trees on small photo plots. In 
addition, exposure variations across each frame 
caused a dramatic shift in the color of coniferous 
forests and approximately 7 percent of the area 
classified as having a spruce component could not be 
stratified into mortality classes. Individual 
polygons in each vegetation/mortality class were 
subsampled with ground plots to estimate levels of 
decline and mortality. 
4.3 Southern Appalachian Mountains 
The high mountains of southwestern Virginia, western 
North Carolina, and eastern Tennessee contain six 
isolated areas of red spruce and Fraser fir. Total 
area is relatively small, ca 24000 ha, and of 
limited commercial value; however, these stands 
occur in such notable landmarks as Mt. Mitchell 
State Park, site of the highest mountain in the 
eastern United States, the Great Smoky Mountains 
National Park, and along the Blue Ridge Parkway, 
whick are areas of major recreational importance. 
Complete 1:12000 scale aerial photo coverage was 
obtained of these areas during 1984 and 1985. Each 
area of spruce-fir forest was stratified into three 
mortality classes (Table 3). These mortality 
classes differed from the classes used in the other 
survey areas and reflect the high levels of Fraser 
fir mortality caused by the introduction of the 
balsam woolly adelgid into these forests. In 
addition, a series of 1:4000 scale CIR photos was 
acquired to help monitor tree damage in selected 
intensive research sites in these areas. 
These inventories have provided a large volume of 
baseline data including statistics on the proportion 
of spruce-fir forest in each mortality class by 
state, data on volume, number of trees, and basal 
area on a unit area basis for each vegetation and 
mortality class, and data on the relative health of 
regeneration. Examples of these data are shown in 
Tables 4, 5, and 6. 
5 GEOGRAPHIC INFORMATION SYSTEMS 
Geographic information systems (GIS) provide a 
capability for storage, analysis, and display of 
spatial data. They can be used to integrate many 
kinds of thematic data and evaluate certain spatial 
relationships. 
To help identify the causal agents associated with 
forest declines and determine the potential role of 
anthropogenic pollutants in this decline complex, it 
is desirable to relate the location of areas of 
decline and mortality to certain topographic 
features such as slope, aspect and elevation. In 
addition, different forest management objectives and 
tactics on various land ownerships might influence 
the intensity of decline and mortality, as might the 
presence of pest outbreaks or other disturbances. 
Spatial data taken from aerial photographs in 
conjunction with recent inventories of decline and 
mortality in the spruce-fir forests of the eastern 
United States includes the location of vegetation 
types by damage classes. These data are presently 
being digitized for entry into a GIS. Additional 
themes, including land ownership, topographic data, 
and historical data on the location of fire, 
logging, and pest outbreaks such as that compiled by 
Pyle, et al. (1985) will also be entered into the 
data base. 
Presently, data for the northeastern states is 
being digitized and entered into a GIS developed and 
maintained by the University of Maine. Data for 
West Virginia and the southern Appalachian Mountains 
Is being stored and analyzed using the Map Overlay 
Statistical System (MOSS), a public domain GIS 
developed and maintained by the Fish and Wildlife 
Service of the U.S. Department of Interior. This 
GIS has recently been installed on a Forest Service 
Data General MV-4000 at Doraville, Georgia, for this 
work. 
GIS will also provide the capability to compare 
the present spatial distribution of forest decline 
and mortality with spatial distributions obtained 
from future inventories.