Modern spatiotemporal geostatistics provides a powerful framework for generation of predictive maps over a spatiotemporal domain by accounting for general knowledge to define a space of plausible events and then restricting this space of plausible events to be consistent with available site-specific knowledge. The Bayesian maximum entropy (BME) method is one of the most widely used modern geostatistics methods. BME results from assigning probabilities of plausible events based on general knowledge through information maximization and then applying operational Bayesian conditionalization that can explicitly assimilate stochastic representations of various uncertain (soft) data bases. The paper demonstrates that fuzzy data sets can be indirectly assimilated by BME through a two-step process: (a) reinterpretation of the fuzzy data as probabilistic through a generalized defuzzification procedure, and (b) efficient assimilation of the probabilistic results of generalized defuzzification by the BME method. A numerical demonstration involves site-specific probabilistic results obtained from the generalized defuzzification of a simulated fuzzy data set and general knowledge that includes the spatial mean trend and correlation structure models. The parameters of these models can be inferred from the hard data equivalent values of the probabilistic results. Accordingly, details of inference based on probabilistic soft data are also considered.
[1]
George Christakos,et al.
On the assimilation of uncertain physical knowledge bases: Bayesian and non-Bayesian techniques.
,
2002
.
[2]
George Christakos,et al.
BME representation of particulate matter distributions in the state of California on the basis of uncertain measurements
,
2001
.
[3]
Wolfgang Näther,et al.
Linear regression with random fuzzy observations
,
1990
.
[4]
Clayton V. Deutsch,et al.
GSLIB: Geostatistical Software Library and User's Guide
,
1993
.
[5]
George Christakos,et al.
Random Field Models in Earth Sciences
,
1992
.
[6]
George Christakos,et al.
Modern Spatiotemporal Geostatistics
,
2000
.
[7]
G. Christakos,et al.
Spatial statistics of clustered data
,
2004
.
[8]
Timothy C. Coburn,et al.
Geostatistics for Natural Resources Evaluation
,
2000,
Technometrics.
[9]
Kazuo Tanaka,et al.
An Introduction to Fuzzy Logic for Practical Applications
,
1996
.
[10]
Ricardo A. Olea,et al.
Geostatistics for Engineers and Earth Scientists
,
1999,
Technometrics.
[11]
Patrick Bogaert,et al.
Temporal GIS: Advanced Functions for Field-Based Applications
,
2002
.
[12]
R. Daley.
Atmospheric Data Analysis
,
1991
.