THEORY

In ordinary kriging, which estimates the unknown value using a weighted linear combinations of the available sample[2]:

(1) The error of i-th estimate, ri, is the difference of estimated value and true value at that same location: (2) The average error of a set of k estimates is:

(3) The error variance is:

(4) Unfortunately, we can not use the equation because we do not know the true value V1,...,Vk. In order to solve this problem, we apply a stationary random function that consists of several random variables, V(Xi). Xi is the location of observed data for i > 0 and i <= n. ( n is the total number of observed data). The unknown value at the location X0 we are trying to estimate is V(X0). The estimated value represented by random function is:

(5) The error variance is :

(6 )

is the covariance of the random variable V(X0) with itself and we assume that all of our random variables have the same variance. is the Lagrange parameter[2]. In order to get the minimum variance of error, we calculate the partial first derivatives of the equation (6) for each w and setting the result to 0. Here is the example of differentiation with respect to w1:

(7) All of weight Wi can be represented as:

For each i, 1 <= i <= n (8) We can get the each weight Wi through the equation (8). After getting the value, we can estimate the value located in X0.