| 2004 
  
were 
]. ít 
ts in 
his Is 
and 
ption 
r the 
sorbs 
sible 
ough 
nd 3, 
agus 
esult 
chers 
ps in 
and7 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
  
  
  
  
  
  
  
  
Number of | number of x (Average of DNs) 
groups Ls 
piots bl b2 b3 b4 bs b6 
12 72.7778 58.0000 44.7778 117.8889 101.6667 48.6667 
24 71.4444 57.4444 43.7778 121.111 103.1111 48.3333 
Bb 25 72.1111 57.4444 44.4444 124.4444 100.1111 45.7778 
5b 27 73.0000 56.6667 43.3333 122.0000 101.3333 46.6667 
2 28 72,3333 56.2222 42.2222 122.0000 94.7778 43.1111 
32 72.5556 55.3333 42.6667 118.6667 92.3333 42.1111 
33 73.6667 S6.2222 42.6667 125.4444 97.7778 43.2222 
3 73.3333 59.7778 45.3333 131.8889 107.1111 49.1111 
©. 7 72.6667 59.8889 45.4444 126.7778 106.7778 50.7778 
3 13 72.2222 59.8889 46.3333 116.3333 105.1111 51.6667 
= ]7 72.2222 59.7778 43.8889 123.5556 106.7778 50.7778 
5 21 73.1111 57.7778 44.6667 125.3333 105.6667 48.5556 
Ra 22 72.111] 56.7778 43.0000 118.5556 106.4444 46.1111 
23 73.3333 57.8889 42.4444 122.2222 104.1111 47.2222 
4 74.2222 61.5556 46.2222 133.8889 112.6667 51.7778 
5 73. 60.2222 44.0000 124.4444 108.7778 50.4444 
e 8 74.2222 61.5556 46.1111 134.8889 112.0000 31.3556 
5 10 73.8889 60.3333 44.7778 132.4444 114.1111 53.6667 
= 19 72.1114 59.5556 45.2222 123.3333 109.4444 52.4444 
7 20 72.3333 59.8889 45.7778 123.1111 107.5556 50.0000 
29 74.3333 61.0000 40.8889 135.3333 111.8889 51.0000 
| 74.4444 62.1111 45.8889 133.6667 113.4444 51.5556 
9 74.2222 61.2222 47.1111 129.5556 109.0000 50.3333 
3 14 74.1111 58.8889 45.0000 132.7778 110.8889 50.7778 
m 16 72.2222 59.7778 43.7778 123.5556 107.8889 50.3333 
5 18 72.6667 60.8889 45.1111 130.1111 143.1111 53.4444 
= 26 73.7778 61.5556 45.7778 137.0000 114.1111 51.7778 
30 75.0000 61.5556 47.4444 133.7778 1117778 50.2222 
  
  
  
  
  
  
  
  
Table 1. Average of DNs in relation to ETM data of 24 Jun 2000. 
analyses showed that there is difference between first group 
with third and fourth groups. This band, also, revealed that there 
is interference among all other groups. Moreover, average DNs 
increases in band 5 and 7 from group 1 to 4, especially in band 
5. Therefore, the minimum DNs are related to the first group 
and the maximum DNs are related to the third and fourth 
groups. The reason for minimum DNs of first group than other 
groups is due to different spectral reflectance of Fagus 
orientalis and Carpinus betulus tree leaves. Based on empirical 
experience of experts, leaves of Fagus orientalis and Carpinus 
betulus trees of study area in accordance to date of imagery 
acquisition (July), color of Fagus orientalis leaves is darker 
than that of Carpinus betulus trees. Canopy crown of Fagus 
orientalis seems much darker than Carpinus betulus canopy 
crowns; therefore, spectral reflectance of Fagus orientalis is 
less than that of Carpinus betulus. 
To compare the averages of four different mixture groups in 
band 5 and 7, statistical analyses were applied using T-test. 
4. CONCLUSION 
Statistical analyses of this study showed that in band 5, there is 
a significant difference at level 95 and 99% probability between 
means of third group (mixture of 70% Fagus orientalis, 30% 
Carpinus betulus and others) with first group (pure composition 
of Fagus orientalis) and between means of fourth group 
435 
(mixture of 60% Fagus orientalis, 40% Carpinus betulus and 
others) with first group, but in other comparisons between 
means in bands 5 and 7, some of comparisons are not significant 
and some of them are significant. Therefore, it shows the ability 
of band 5 in separating different mixtures of all aforementioned 
groups. It means that ETM' band 5 is able to separate pure 
composition of Fagus orientalis from: i) mixture of 7096 Fagus 
orientalis, 30% Carpinus betulus and others ii) mixture of 60% 
Fagus orientalis, 40% Carpinus betulus and others in middle of 
growth season, however, it is not able to completely separate 
pure composition of Fagus orientalis from the mixture of 80% 
Fagus orientalis, 20% Carpinus betulus and others and other 
groups from each other. Result of ETM" band 7 analyses 
showed that this band is not able to significantly separate all 
means different groups. It proves that the ability of separating 
different mixtures in this band is lower than band 5. Many 
studies have shown that Landsat satellite data has relatively 
good ability in separating and inventorying of different forest 
species. Oladi (1988) used Landsat Multi Spectral Scanner 
(MSS) to separate pure forests of two species of Beech and Oak 
with 80% accuracy from each other. Poso et al. (1987) could 
provide a method of forest inventory for pine and broad leaves 
trees using Landsat-MSS and TM imagery. According to a 
study of Baker (1992) Landsat-TM imagery is able to provide 
very good information about invading plants into the 
plantations. Result of this study shows a relatively good ability