The generation over two-dimensional grids of normally distributed rand
om fields conditioned on available data is often required in reservoir
modeling and mining investigations. Such fields can be obtained from
application of turning band or spectral methods. However, both methods
have limitations. First, they are only asymptotically exact in that t
he ensemble of realizations has the correlation structure required onl
y if enough harmonics are used in the spectral method, or enough lines
are generated in the turning bands approach. Moreover, the spectral m
ethod requires fine tuning of process parameters. As for the turning b
ands method, it is essentially restricted to processes with stationary
and radially symmetric correlation functions. Another approach, which
has the advantage of being general and exact, is to use a Cholesky fa
ctorization of the covariance matrix representing grid points correlat
ion. For fields of large size, however, the Cholesky factorization can
be computationally prohibitive. In this paper, we show that if the da
ta are stationary and generated over a grid with regular mesh, the str
ucture of the data covariance matrix can be exploited to significantly
reduce the overall computational burden of conditional simulations ba
sed on matrix factorization techniques. A feature of this approach is
its computational simplicity and suitability to parallel implementatio
n.