Geology Reference
In-Depth Information
Model in h
2
including Linear Model
(
h
) =
ch
<2
(23)
If = 1 the variogram model is called Linear.
Spherical model
3
3
h
1
h
c
h
'$
a
(24)
2
a
2
a
c
h
>
a
(25)
Exponential Model
45
h
67
c
1exp
(26)
89
a
!
"
Gaussian Model
4
2
5
:
h
:
c
1exp
6
7
(27)
2
a
:
:
8
9
!
"
Cubic Model
3
s
7
h
h
h
h
c
7
8.75
3.5
0.75
h
$
a
(28)
a
a
a
a
!
"
c
h
>
a
(29)
Difference between the approach of statistics and geostatistics
Statistics
Geostatistics
Geostatistics considers all the numerical values
z
(
x
),
z
(
x
') as the particular realizations of
random variables
Z
(
x
) and
Z
(
x
')
Statistics considers all the
numerical values
Z
(
x
)
, Z
(
x
') as
independent realization of the same
numerical function
Z.
In other words it does not take into
account the spatial auto-correlation
between two neighbouring values
Z
(
x
) and
Z
(
x
+
h
)
Statistics does not take the
regionalization into consideration.
In other words, the location of the
samples is ignored.
In other words it takes into account the spatial
auto-correlation between two neighbouring
values
z
(
x
) and
z
(
x + h
)
.
Geostatistics takes the regionalization into
consideration. Even when populations can have
same parameters like mean, variance, skewness
and kurtosis still they can differ considerably.
Because the regionalizations can be different.
Geostatistics also takes into consideration, the
anisotropies, the correlation with the distance.
In other words Geostatistics takes into account
the holistic view.
The variability with the distance
and the anisotropies are not taken
into consideration
