218
Regional Environmental Changes : Application to Soil Erosion Modeling
'00...,------ - - - - -- - - - - - - - - - - - - - ,
aD
eo
~
~
~
~
it
~
AO
1 0
?lIo
.. .
o +--~--~-_.--._-_,--,_-_.--,_-_,--~
o
20
BO
80
'00
30
70
Figure 2. Regression between SH r % and the dominant layers.
As far as the previous hypothesis is concerned, higher SHr% values represent a
greater distribution of tree heights in the dominant layers (4 and 5), where lower
values are typical of a uniform distribution in all the different layers. The
regression (Fig. 2) calculated by taking the dominant layers and SH/7'o into
account, shows (r 2 =0.78) that the index obtained is well able to describe tree
height variability in the sample area.
2.4 Fourth step: determination of a numerical index
evaluated by using a few parameters (diameter
distribution).
Having created a structural complexity index representative of the prefixed goals,
it was then necessary to simplify the above-mentioned ground surveying
methodology by taking a great deal of time to measure the large number of tree
heights (this is not easy in a forest).
Another index, i.e. the height distribution frequency was determined by
calculating the diameter distribution frequency according to the following
equation:
Sdm =
(3.3)