STOCHASTIC INTERPOLATION OF LOGNORMAL FIELDS BY BF/MCF

Abstract

As a versatile tool to update random fields, Bootstrap filter/Monte Carlo filter is focused that is a sequential algorithm of generating a set of sample realizations of a predicted state vector and a filtered state vector respectively.
In order to clarify the potential of this method, stochastic interpolation of a lognormal spatial random field is demonstrated by using numerically simulated data.