ISSN 1058-3912
Hyun-Suk Shin
Hydrologic Sciences and Engineering Program
Colorado State University, USA
ABSTRACT
Spatial Analysis Neural Network (SANN) is a specified neural network
for conducting the spatial analysis of any type of variable. It provides
a nonparametric mean estimator and also estimators of higher order statistics
such as standard deviation and skewness. In addition, it provides a decision-making
tool, including an estimator of posterior probability that a spatial variable
at a given point will belong to various classes representing the severity
of the problem of interest, and a Bayesian classifier to define the boundaries
of subregions belonging to the classes. In this paper, the use of SANN
as a decision-making tool to investigate an area contaminated by viruses
in a groundwater system is illustrated. SANN provides two pieces of information;
the contamination probability that the virus decay rate at a given point
is less than a predefined threshold value, and the classification map defining
contaminated and non- contaminated regions. The method is applied to several
cases with varying threshold levels of the observed virus decay rate values,
and the results show graphically the extent of the contaminated region
and the change of the contamination probabilities.
Reference: Shin, H.; Use of the Spatial Analysis Neural Network
(SANN) Method for Regional Groundwater Contamination Decision-Making,
Journal of Environmental Hydrology, Vol. 5, Paper 4, July 1997.