Students who want to take this course should have taken an entry-level statistics. Familiarity of at least one programming language is desired, although not definitely necessary.

Geostatistics （地理統計）
Dept. of Bioenvironmental Systems Engineering
National Taiwan University
Professor Ke-Sheng Cheng (鄭克聲教授)
Class ID - 622 U2580 Credit hours - 3

Syllabus
1.Introduction
Characterizing natural phenomena - deterministic vs. stochastic
Random variables, random processes and random fields
Stochastic simulation
The scale issue
2.Variogram modeling
Isotropic variogram modeling
Scale effect - change of support
Anisotropic variogram modeling
3.Spatial estimation
Simple kriging
Ordinary kriging
Block kriging
Universal kriging
Co-kriging
Indicator kriging
4.Assessment of spatial uncertainty
Random field simulation
Conditional simulation
5.Monitoring network design
Geostatistical approach
Maximum entropy approach
6.Random fractals and variogram analysis
Fractal dimension
Random fractals
Fractional Brownian motion ?Hurst phenomenon
7.Examples of application
Raingauge network design
Remote sensing image rectification
Contouring of design storm depths
Prediction of drought lead time
Probabilistic evaluation of reservoir trophic state
Image classification by indicator kriging
Conditional simulation of heavy metal concentrations

Grade:
Homeworks 50% (10%@5)
In-class Midterm exam 25%
Final project report 25%

Geostatistics softwares -
GS+, S-plus with spatial Stat, GSLIB

References:
Chiles, J.P. and Delfiner, P., 1999. Geostatistics - Modeling Spatial Uncertainty, John Wiley & Sons, Inc., New York.

Goovaerts, P., 1997. Geostatistics for Natural Resources Evaluation. Oxford University Press, New York.

Journel, A.G. and Huijbregts, C.J., 1978. Mining Geostatistics, Academics Press, London.