183 - 
3TS , crown 
smaller 
ee classes 
eating 
d degraded 
a rated 
jht and 
resources. 
ation a 
was even 
by V.K. 
as a 
than 
( 8 ). 
a classes 
a closure 
7 TIWARI 
ion into 
t area in 
sample 
is on the 
test and 
the volume 
ipplied 
ve saving 
sive 
n for 
tern is 
are st 
One of 
le weight 
various 
survey. 
r such 
)e a 
Lties of 
to be 
is the 
rom 
all the 
¡.cable to 
the 
shape to 
s end in 
sation has 
they were 
sses was 
QUANTITATIVE INVESTIGATIONS 
The degree to which quantitative information on forest areas 
from aerial photographs could be obtained depended upon the nature of 
the forest and scale. Large scale photographs provided the information 
with sufficient accuracy while on medium and small scales the accuracy 
had wider limits. 
"Tree Height Measurements and Estimations" 
Most popular method is Parallax method with the use of either 
parallax bar with mirror stereoscope or parallax wedge with lens stereo 
scope. In the initial stages of forest inventory the differences between 
photo measured heights and ground height were wide (FAO Technical 
Report 1 : SF/IND 23) . The differences were attributed to insufficient 
training and lack of coordination between photo interpretation and ground 
data. JOSHI ( 4 ) indicated no significant difference between photo 
heights measured on 1:5000 to 1:2 0000 scale and photo and ground 
heights were within + 2 meters. The two important considerations with 
parallax method, that the base of the tree should be visible and sufficient 
measurable parallax should be achieved, are met with in most of the forest 
areas of the country. The worst cases came from tropical evergreen 
types where either the base is not visible or surrounded by high scrub 
growth which makes the setting of floating mark on the ground difficult. 
In a few cases such a situation is also met with in tropical moist deciduous 
or some coniferous forests but it is not so serious in that identical tall 
trees where the base is seen are met with nearby. Investigations as to the 
differences between interpreters were carried out by VERSTEEGH ( 15 ) 
indicating no significant difference for dry deciduous forests of Bastar. 
"Crown Diameter Measurements" 
This parameter has not been popular amongst the photo inter 
preters of tropical forests. The reason is obvious with the fact that 
the form of the crowns of broadleaved species is very varied without 
any uniformity. Secondly the measurements necessarily require a larger 
scale which are either not available or are costly and are not justified 
when used in forest inventories. Nevertheless the parameter has always 
tempted the photo interpreters in getting some information on diameter 
breast height which is not directly obtainable from aerial photographs. 
GUPTA ( 3 ) measured the crown diameters of 148 teak trees on 1:10000 
scale photographs and tried to establish a correlation between crown 
diameter and stem diameter (1.37) and ended in a correlation coefficient 
of 0.4. He, however, tried to correlate the frequency distribution and 
cumulative frequency distribution curves between crown diameter and 
d.b.h. He concluded that with suitable class interval and stratification 
into height classes a reasonable prediction of stem diameter distribution 
could be made with crown diameter distribution. JOS HI ( 4 ) found 
that crown width measurements on aerial photographs in irregular misce 
llaneous forests where trees were in clusters and crowns had a tendency 
to fuse with the crowns of adjoining trees were not very reliable. This 
parameter has been used in tree aerial volume table for teak of Allapalli 
forests. In fact the nature of the forests where JOSHI worked do 
represent a mass of tropical deciduous type of forests in India and in