Data Analysis in the Earth Sciences

Meeting: Mon-Wed-Fri, 2:00-2:50 AM,

Instructor:
Professor Jonathan Lees
(313 Mitchell Hall ; 962-0695; jonathan_lees@unc.edu)

Course Description

Level: undergraduate and graduate

Prerequisites: Basic Math; calculus may be helpful.

Requirements: 1 midterm exam, 1 final exam, final project using student provided geological data. Homework sets required, approximately 1/week. Some homework sets will involve hands on computer applications, typically using R, MATLAB, S-plus or other facilities available.

Suggested Reading List
Problem Sets
Functions in R and Matlab

Texts:
Required:

Davis, J.C., 2002, Statistics and data analysis in geology: New York Chichester, Wiley, xvi, 638 p.

Recommended:

Introduction to Geological Data Analysis, A.R.H. Swan & M. Sandilands, 1995, Blackwell Science.


Lectures and discussion will involve theoretical development and practical implementation.  Numerous examples will be presented and intuitive understanding of material will be stressed.

This course is intended for upper level undergraduates and graduate students who need experience in quantitative analysis of data commonly analyzed in the earth sciences, including solid earth, atmospheres, oceans, geology, geochemistry and paleontology. A strong emphasis is placed on applications and real world examples. Students learn how to use a standard analysis package (R, Matlab or S-plus). Homework assignments and projects require programming and creative use of these computational tools. Topics covered include: univariate statistics, statistical testing, non-parametric methods, time series, spatial analysis of maps, cluster analysis, shapes, and multivariate methods.

Approximate Schedule:

Week

Subject

1

Introduction; Probability; Statistics

2

Normal Distributions, Central Limit Theorem, testing hypotheses

3

t-test; F-test; Chi-squared test, Non-parametric methods

4

Non-parametric methods; matrix algebra, inversion, eigenvalues-eigenvectors

5

ANOVA

6

Regression

7

Regression – Time Series

8

Time Series-correlation; convolution, cross-correlation, Spectral Analysis

 

Midterm exam

 

 

9

Multivariate

10

Multivariate

11

Multivariate Analysis

12

Principal C omponents

13

Factor analysis

14-16

Factor Analysis

17

Final