KRIGING

ORDINARY KRIGING
 VARIATION OF KRIGING

• Kriging Varieties Can Be Distinguished According to the Model

Used for the Trend m(u)
• Simple Kriging (SK) Considers the Mean m(u) Known and
Constant In the Study Area. That is,
m(u) = m

• Ordinary Kriging (OK) Accounts for Local Fluctuations of the

Mean by Limiting the Domain of Stationarity of the Mean to the
Local Neighborhood W(u). That is,
m(u’) = Constant but Unknown

• Kriging with Trend (KT) Considers that the Unknown Local Mean
(m(u’)) Smoothly Varies Within Each Local Neighborhood W(u).
The Trend Component Is Modeled as a Linear Combination of
Functions fk(u). That is,
m(u’) = S ak(u’) fk(u’) with ak(u’) = ak Constant but Unknown
 SK VS OK

• Ordinary Kriging Normally Preferred Over Simple Kriging

• Knowledge of Mean Not Required
• Stationarity of Mean Over Entire Study Area Also Not
Required
• In Areas of High Values, SK Estimates Will Be Lower than OK
Estimates
• In Areas of Low Values, SK Estimates Will Be Higher than OK
Estimates
• Why? Remember, OK Uses a Local Neighborhood and the
Mean In High Value Areas Will Generally Be Greater than the
m(u)
• Note - Both SK and OK Are Exact Estimators!
 ORDINARY KRIGING

• Ordinary Kriging - Mathematical Approach

• The Unknown Local Mean m(u) Is Filtered from the Estimator
by Forcing the Weights to sum to 1!
• The Ordinary Kriging Equations:
n ( u )
Z *
O K = å
a =1
l aO K ( u ) Z ( u a )
n ( u )

å l O K
a ( u ) = 1
a = 1

where C = Covariance and µ = Lagrange Parameter

• The Lagrange Parameter (µ) Is Used to Convert a
Constrained Minimization Problem Into An Unconstrained
Minimization Problem
 ORDINARY KRIGING

• OK In Terms of C
n ( u )

å
b
lb
= 1
O K
( u ) C R (u a - u b ) + µOK ( u ) = C R (u a - u)

a = 1 ® n
n ( u )

å
b = 1
l bO K (u) = 1

• Error Variance
n ( u )
s 2
O K (u) = C R (0) - å
a
=1
la O K
( u ) C R (u a - u) - µOK (u)
 ORDINARY KRIGING

The Ordinary Kriging Equations in Matrix Form

C Matrix D Matrix
é C11 C 1 2 .. C 1 n 1 ù é l1 ù é C10 ù
ê C C .. .. 1 ú ê l ú ê C ú
ê 2 1 2 2 ú ê 2 ú ê 2 0 ú
ê .. .. .. .. . . ú ´ ê .. ú = ê .. ú
ê C .. .. C 1 ú ê l ú ê C ú
ê ú ê n ú ê ú
ë µ
n 1 n n n 0
1 0 û û
ë 1 .. 1 ë 1 û
Weights
EXAMPLE
 EXAMPLE ORDINARY KRIGING

• Consider the data configuration, 𝑥 𝑢1 =20, 𝑥 𝑢2 =50, 𝑥 𝑢3 =30,

𝑥 𝑢4 =100,
• Assume the following variogram models:

• Estimate the value and the error variance at location with the
previous three models for both configurations.
 SOLUTION - MODEL 1
• We assume that the 𝐶 1,2 = 𝐶 2,1 = 𝐶 0 − 𝛾(1,2)
variogram model is given by
𝐶 1,2 = 𝐶 0 − 𝛾(200)
3
3 200 1 200
= 100 − 100 −
2 500 2 500
3 2 1 8
= 100 − 100 −
• Configuration 1. Because 2 5 2 125
there are 4 sample points,
the covariance matrix is 5 x = 100 − 100 0.568 = 100 − 56.8
5. = 43.2
𝐶 1,3 = 𝐶 3,1 = 𝐶 1,4 = 𝐶 4,1 = 𝐶 2,4
= 𝐶 4,2 = 𝐶 2,3 = 𝐶 3,2
= 𝐶 0 − 𝛾(223.6)
3
3 223.6 1 223.6
= 100 − 100 −
2 500 2 500
= 100 − 100 0.6708 − 0.0447
= 100 − 62.61 = 37.4
 SOLUTION - MODEL 1

𝐶 3,4 = 𝐶 4,3 = 𝐶 0 − 𝛾(3,4) 𝐶 1,0 = 𝐶 2,0 = 𝐶 0 − 𝛾(1,0)

𝐶 3,4 = 𝐶 4,3 = 𝐶 0 − 𝛾(400) 𝐶 1,0 = 𝐶 2,0 = 𝐶 0 − 𝛾(100)
3 400 1 400 3 3 100 1 100
3
= 100 − 100 − = 100 − 100
2 500 2 500 −
2 500 2 500
= 100 − 100 1.2 − 0.256 3 1 1 1
= 100 − 100 −
= 100 − 94.4 = 5.6 2 5 2 125
In matrix form, we write = 100 − 100 0.296 = 100 − 29.6 = 70.4
𝐶 3,0 = 𝐶 4,0 = 𝐶 0 − 𝛾(3,0)
𝐶 3,0 = 𝐶 4,0 = 𝐶 0 − 𝛾(200)
3
3 200 1 200
= 100 − 100 −
2 500 2 500
3 2 1 8
= 100 − 100 −
2 5 2 125
= 100 − 100 0.568 = 100 − 56.8 = 43.2
 SOLUTION - MODEL 1

• In matrix form, we write • We obtain

• We estimate the value

at 𝑥∗ 𝑢0 as

• The error variance is

estimated as
 ASSIGNMENT 1

• Configuration 2???
 ASSIGNMENT – MODEL 2

• We assume that the variogram model is given by

• The only difference between models 1 and 2 is the

nugget effect.
• Configuration 1 ???
• Configuration 2 ???
 ORDINARY KRIGING

Example Data Set. Calculate Estimate for Location X.

Well X Y Value Distance to X
1 61 139 477 4.47
2 63 140 696 3.61
3 64 129 227 8.06
4 68 128 646 9.49
5 71 140 606 6.71
6 73 141 791 8.94
7 75 128 783 13.45
 ORDINARY KRIGING

Ordinary Kriging Example Solution. Estimate for Location

(64, 138) = 607.7
Well X Y Value Distance to X Well 1 2 3 4 5 6 7
1 61 139 477 3.16 1 0.00 2.24 10.44 13.04 10.05 12.17 17.80
2 63 140 696 2.24 2 2.24 0.00 11.05 13.00 8.00 10.05 16.97
3 64 129 227 9.00 3 10.44 11.05 0.00 4.12 13.04 15.00 11.05
4 68 128 646 10.77 4 13.04 13.00 4.12 0.00 12.37 13.93 7.00
5 71 140 606 7.28 5 10.05 8.00 13.04 12.37 0.00 2.24 12.65
6 73 141 791 9.49 6 12.17 10.05 15.00 13.93 2.24 0.00 13.15
7 75 128 783 14.87 7 17.80 16.97 11.05 7.00 12.65 13.15 0.00

X 64 138 ???? 10.00 5.11 0.44 0.20 0.49 0.26 0.05 1.00
5.11 10.00 0.36 0.20 0.91 0.49 0.06 1.00
Estimate 607.721 0.44 0.36 10.00 2.90 0.20 0.11 0.36 1.00
Uncertainty 7.293 0.20 0.20 2.90 10.00 0.24 0.15 1.22 1.00
0.49 0.91 0.20 0.24 10.00 5.11 0.22 1.00
0.26 0.49 0.11 0.15 5.11 10.00 0.19 1.00
D Matrix C Matrix
0.05
1.00
0.06
1.00
0.36
1.00
1.22
1.00
0.22
1.00
0.19
1.00
10.00
1.00
1.00
0.00

3.873 0.127 -0 . 0 7 7 -0 . 0 1 3 -0 . 0 0 9 -0 . 0 0 8 -0 . 0 0 9 -0 . 0 1 2 0.136

5.113 -0 . 0 7 7 0.129 -0 . 0 1 0 -0 . 0 0 8 -0 . 0 1 5 -0 . 0 0 8 -0 . 0 1 1 0.121
0.672 -0 . 0 1 3 -0 . 0 1 0 0.098 -0 . 0 4 2 -0 . 0 1 0 -0 . 0 1 0 -0 . 0 1 4 0.156
0.395 Inverse -0 . 0 0 9 -0 . 0 0 8 -0 . 0 4 2 0.102 -0 . 0 0 9 -0 . 0 0 9 -0 . 0 2 4 0.139
1.126 of C -0 . 0 0 8 -0 . 0 1 5 -0 . 0 1 0 -0 . 0 0 9 0.130 -0 . 0 7 7 -0 . 0 1 2 0.118
0.581 -0 . 0 0 9 -0 . 0 0 8 -0 . 0 1 0 -0 . 0 0 9 -0 . 0 7 7 0.126 -0 . 0 1 3 0.141
0.116 -0 . 0 1 2 -0 . 0 1 1 -0 . 0 1 4 -0 . 0 2 4 -0 . 0 1 2 -0 . 0 1 3 0.085 0.188
1 0.136 0.121 0.156 0.139 0.118 0.141 0.188 -2 . 1 8 0

99.411 0.807
0.208
315.186
Weights x 2.315
0.453
0.085 19.252 Sample Values 0.057
Lagrange C-1 x D = 0.055 35.560 0.022
Parameter Weights 0.100
60.712 0.113
31.667 0.023
0.040
45.932 0.007
0.059
-0 . 6 3 7
S u m of
Weights 1.000 607.721 The 7.293
Estimate

The
Uncertainty
 ORDINARY KRIGING

• Kriging Example (continued)

• Distance Matrix
Well 1 2 3 4 5 6 7
1 0.00 2.24 10.44 13.04 10.05 12.17 17.80
2 2.24 0.00 11.05 13.00 8.00 10.05 16.97
3 10.44 11.05 0.00 4.12 13.04 15.00 11.05
4 13.04 13.00 4.12 0.00 12.37 13.93 7.00
5 10.05 8.00 13.04 12.37 0.00 2.24 12.65
6 12.17 10.05 15.00 13.93 2.24 0.00 13.15
7 17.80 16.97 11.05 7.00 12.65 13.15 0.00

• Covariance Matrix (Assumes Exponential Model)

10.00 5.11 0.44 0.20 0.49 0.26 0.05 1 . 0 0
5.11 10.00 0.36 0.20 0.91 0.49 0.06 1 . 0 0
0.44 0.36 10.00 2.90 0.20 0.11 0.36 1 . 0 0
0.20 0.20 2.90 10.00 0.24 0.15 1.22 1 . 0 0
0.49 0.91 0.20 0.24 10.00 5.11 0.22 1 . 0 0
0.26 0.49 0.11 0.15 5.11 10.00 0.19 1 . 0 0
0.05 0.06 0.36 1.22 0.22 0.19 10.00 1 . 0 0
1.00 1.00 1.00 1.00 1.00 1.00 1.00 0 . 0 0
 ORDINARY KRIGING

• Kriging Example (continued)

• Inverse of C Matrix (=C-1)
0.127 -0 . 0 7 7 -0 . 0 1 3 -0 . 0 0 9 -0 . 0 0 8 -0 . 0 0 9 -0 . 0 1 2 0.136
-0 . 0 7 7 0.129 -0 . 0 1 0 -0 . 0 0 8 -0 . 0 1 5 -0 . 0 0 8 -0 . 0 1 1 0.121
-0 . 0 1 3 -0 . 0 1 0 0.098 -0 . 0 4 2 -0 . 0 1 0 -0 . 0 1 0 -0 . 0 1 4 0.156
-0 . 0 0 9 -0 . 0 0 8 -0 . 0 4 2 0.102 -0 . 0 0 9 -0 . 0 0 9 -0 . 0 2 4 0.139
-0 . 0 0 8 -0 . 0 1 5 -0 . 0 1 0 -0 . 0 0 9 0.130 -0 . 0 7 7 -0 . 0 1 2 0.118
-0 . 0 0 9 -0 . 0 0 8 -0 . 0 1 0 -0 . 0 0 9 -0 . 0 7 7 0.126 -0 . 0 1 3 0.141
-0 . 0 1 2 -0 . 0 1 1 -0 . 0 1 4 -0 . 0 2 4 -0 . 0 1 2 -0 . 0 1 3 0.085 0.188
0.136 0.121 0.156 0.139 0.118 0.141 0.188 -2 . 1 8 0

• The D Matrix (Covariances Between Data Locations and

Point to be Estimated) 3.873
5.113
0.672
0.395
1.126
0.581
0.116
1
 ORDINARY KRIGING

• Kriging Example (continued) 0.208

0.453
• Inverse of C Matrix (C -1) x D Matrix = Weights 0.085
0.055
• Note that Sum of Weights = 1.000 0.100
0.040
0.059
-0.637

1.000

• Weights x Data Values = z* 99.411

315.186
• z* = Estimate = 607.721 19.252
35.560
60.712
31.667
45.932

607.721

18
 ORDINARY KRIGING

• Kriging Example (continued)

• Minimized Estimation Variance, s2R, Is Calculated Using
the Equation

s2 = s2 - S(l C ) + µ = 7.293
R DATA i i0

• Weights x D Matrix (Covariance between Data

Locations and Points Being Estimated). Note that µ = -
.907 0.807
2.315
0.057
0.022
0.113
0.023
0.007

7.293