689 
respective 
ages. 
I Infrared and 
eff. = s. 
ri 
.48 
-0.16 
-54 
13 
-0.6 
-57 
18 
-0.5b 
.60 
12 
-0.49 
0 
0.5 0.9 0.7 
Graan Rod 
0-9 0.9 (ftm) Wavolongth 
Infrarad 
Figure 5« Diagrams of the mean reflectance derived 
from the mean values of the base colour components, 
a) Gl, b) G2, c)LG and d) MA. DN stands for Digital 
Numbers. 
We remark a good matching between the diagrams of the 
subframes G2 and LG; we also observe a decreasing of 
the mean reflectance from the sub-frame Gl up to MA. 
This fact would mean a high correlation between the 
response in reflectance of the surely geothermal 
zone LG and the zone G2; in both cases thermal IR 
surveys indicated zones where EV was very small, al 
most zero. 
Moreover, the sub-frame MA was the most suspected 
area, since it was the only spot indicated at the 
same time both by FATI and by EV function solutions. 
On the other hand, the sub-frame Gl shows the hi 
ghest reflectance values and that could fully be cor 
related to the highest values of FATI previously che 
cked. 
The arithmetic means of each band ratio have been 
correlated to their standard deviations (see fig.6), 
We remark here that arithmetic means of NDVI sub 
frames keep almost the same values. That confirms the 
spots selected show similar vegetational vigour and 
the mean values of actual NDVI are ranging between 
0.15 and O.bk which correspond to the middle levels 
in the grey scale of fig. 4. 
Moreover, a noticeable difference appears in the 
mean values of the IR/Green ratio between G2 and LG; 
on the contrary, their respective reflectance featu 
res shown in figure 5, resulted to be almost the sa 
me. That means the ratio IR/Green has a more discri 
minating power than the ratio IR/Red and than the 
spectral signature alone. 
In particular, the site MA shows the highest values 
in both ratios and, at the same time, it also pre 
sents the lowest reflectance mean values in according 
to the basic concept of a general decay when vegeta 
tion is submitted to the presence of geothermal fl 
uids. 
255-1 
240- 
M 
220- 
200- 
180- 
160- 
* 140- 
•K 
1*1 
Í 
120- 
<fc 
100- 
80 
60 
40- 
20- 
0- 
1 
^1 
-4 
o ^ 
O «N 
Q O 
% — 
* - 
o K, 
O Uj 
<0 1 
?£ 
JO . 
Ï £ 
e> o 
O Q 
o Î 
O 
$ 
:a 1 
Î § 
Figure 6. Diagrams of the mean values of band ratios 
at different sites, a) IR/Red, b)lR/Green, c) NDVI 
and d) standard deviation amplitude. 
These arguments authorize to take the suspected zone 
MA Poggio Mauriccia into a higher consideration than 
the other sub-frames. 
Scattergrams of data also have Deen analysed, al 
though no particular indication has outcome beside 
the high correlation that has been found between the 
Green and Red images, almost in all 4 cases. Scat- 
tergrams of band ratios IR/Green and IR/Red con 
firmed the lowest correlation is found for the 
subframe MA. 
4 CONCLUSIONS 
The application of the Derived Tbermal Quantities 
method in the Travale geothermal field has given so 
me unexpected results. 
At first, neither the old field of Lagoni nor the 
new one to the north have been particularly pointed 
out in relation to othen surrounding areas. That gi 
ves rise to some perplexity on the validity of the 
method. 
Nevertheless, some otber interesting indications 
have been provided. Although the checked sub-frames