Block Estimation: Covariance Between A Point and A Block
Block Estimation: Covariance Between A Point and A Block
Block Estimation: Covariance Between A Point and A Block
Block estimation
n
o = m1 − 1
i =1
jC (ui , u j ) = Cv (ui , u0 )
n
j =1 for i = 1, … , n
The arithmetic average of all the poin-to-point covariance between the sampled
point, ui and all the point, u0j, located within the block .
1 nb
(
Cv (u0 , ui ) C u0 j , ui
nb j =1
) nb= selected number of
points within a block
C (u1 , u1 ) C (u1 , un ) 1 Cv (u0 , u1 )
=
C (un , u1 ) C (un , un ) n Cv (u0 , un )
n
We have X ( u0 ) = 0 + i X ( ui )
*
v
i =1
2
ˆ = Cvv (u0 , u0 ) − iCv (ui , u0 )
2
E
i =1
Numerical Example 4.2 porosity value
using block kriking page 109
n
X (uo ) = o + i X (ui )
*
i =1
E X (uo ) − X (uo ) = 0
*
n
o = m(uo )1 − i
i =1
n
We force the λo to be zero, so
i =1
i =1
n
X (uo ) = i X (ui )
*
i =1
jC (ui , u j ) + = C (ui , uo )
C (u1 , u1 ) C (u1 , un ) 1 1 C (u1 , u0 )
C (u , u ) C (u , u )
=
1 n C (un , u0 )
n 1 n n
1
0
1 1
The weight λi
−1
=C c
Estimeted value at uo
n
X (uo ) = o + i X (ui )
*
i =1
n
= C (uo , uo ) − 1C (ui , uo ) −
ˆ2
E
i =1
Numerical Example 4.3. page 113
a. Isotropic Variogram
()
L = 100 M s 500
(L ) ,
b. Isotropic Variogram
()
L = 50 + 50 M s 500
(L )
and
c. Anisotropic Variogram
() ()
L = 100 M s500 L east/west direction,
and
()
L = 100 M s 250
(L ) north/south direction.
Estimate the value and the error variance at
location uo the previous three models for
both configurations !
Others exercises for Ordinary Kriging
Unbiased estimation
E X * ( uo ) − X ( uo ) = 0
n n
mX X i + mY Yk − mX = 0
i =1 k =1
n n
i =1
Xi = 1 and
i =1
yi =0
Minimum Variance
( ) ( )
n n
Var X ( uo ) − X i X u X i − Yk Y uYk
i =1 k =1
( ) ( ) ( )
n m
j =1
Xj C X u X i , u X j + Yk CC u X i , uYk + X = C X u0 , u X i for i = 1,..., n,
k =1
and
( ) ( ) ( )
n m
i =1
Xi CC u X i , uY j + Yl CY uYk , uYl + Y = CC u0 , uYk for k = 1,..., m.
l =1
CX
(u X1 , u X1 ) (
C X u X1 , u X n ) (
CC u X1 , uY1 ) (
CC u X1 , uYm ) 1 0
C X (u X1 , uXn ) (
CX uX n , uX n ) (
CC u X n , uY1 ) (
CC u X n , uYm ) 1 0
CC (u X1 , uY1 ) CC (u Xn , uY1 ) C (u
Y Y1 , uY1 ) C (u
Y Y1 , uYm ) 0 1
CC (u X1 , uYm ) (
CC u X n , uYm ) (
CY uYm , uY1 ) (
CY uYm , uYm ) 0 1
1 1 0 0 0 0
0 0 1 1 0 0
X1 C X ( u , u )
0 X1
C X
Xn
(u , u )
0 Xn
Y1 C
= C
(u , u )
0 Y1
Ym CC (u , u )
0 Ym
X
1
Y
0
Numerical Example 4.4
C X (u3 , u3 ) CC (u3 , u1 ) CC (u3 , u2 ) CC (u3 , u3 ) 1 0 X 1 C X (u3 , u0 )
C (u , u )
C 3 1 CY (u1 , u1 ) CY (u1 , u2 ) CY (u1 , u3 ) 0 1 Y1 CC (u1 , u0 )
CC (u3 , u2 ) CY (u2 , u1 ) CY (u2 , u2 ) CY (u2 , u3 ) 0 1 Y2 CC (u2 , u0 )
=
C 3 , u3 )
C (u CY (u3 , u1 ) CY (u3 , u2 ) CY (u3 , u3 ) 0 1 Y3 CC (u3 , u0 )
1 0 0 0 0 0 X 1
0 1 1 1 0 0 Y 0
C X (u3 , u3 ) = C X (0) = 25,
CC (u1 , u3 ) = CC (0) − C (u1 , u3 ) = CC (0) − C (265)
= 4 − 1 + 3M s500 (265) = 0.838,
CY (u1 , u2 ) = CY (0) − Y (u1 , u2 ) = 1.2 − C (100)
= 1.2 − 0.2 + 1M s500 (100) = 0.704.
C X (u3 , u0 ) = C X (0) − X (u3 , u0 )
= 25 − 15 + 10M s500 (200) = 4.32.
25.0 0.838 0.838 4.0 1 0 X 1 4.32
0.838 1.2 0.704 0.279 0 1 Y1 2.11
0.838 0.704 1.2 0.279 0 1 Y2 2.11
=
4.0 0.279 0.279 1.2 0 1 Y3 1.30
1
0 0 0 0 0 X 1
0 0 Y 0
1 1 1 0
x* (uo ) = 1 5 + 1.248 0.8 + 1.248 0.6 − 2.496 1.2 = 3.75,
Two variables Z1(x) and Z2(x) (such as porosity & acoustic impedance)
Use of Z1 and Z2 data to get a better interpolation of Z1
Z cok (x0 ) = Z (x ) + Z (x )
1i N Z1
i 1 i
1 j N Z 2
j 2 j
cok
2
(x0 ) = VarZ (x0 ) − Z cok (x0 )
Thanks to the extra information coming from Z2 (x):
0 cok
2
(x0 ) k2 (x0 )
Cokriging (Doyen, 1988)
Kriged porosity
Porosity by
cokriging
Conventional Estimation Techniques
Universal Kriging
Universal Kriging
Samples value X (u ) = m(u ) + R(u )
m(u ) = i)
(
Local mean is
X u
neighborhood
n
X * (u0 ) = i X (ui )
i =1
n
f (u ) = f (u )
i =1
i l i l 0 for l = 0, … , L
n
L
ˆ = C (u0 , u0 ) − iC (ui , u0 ) − l fl (u0 )
2
E
i =1 l =0
Another alternative uses the method of external drift. The trend in the data is
Provided by the secondary variable Y.
m(uo ) = a0 + a1Y (u0 )
Stationarity
E Z ( x ) = E Z (x + h ) = m
CovZ (x ), Z (x + h ) = C (h )
Field Example 4.8 – Universal Kriging To Estimate Top of Reservoir