MULTIVARIATE METHODS IN ECOLOGY

ADVANCED ECOLOGY (BOT 5533)

Michael Palmer,

Botany Department

Oklahoma State University

405-744-7717 carex@okstate.edu

This graduate course will focus on methods used by ecologists to analyze community data (not restricted to plants!). It will involve hands-on experience with ordination methods. Students are encouraged to bring or collect their own data. If you have a particular data set you wish to analyze, please contact me to determine whether it is appropriate. In the past, class projects have resulted in thesis chapters and published papers. Community patterns will also be discussed during one weekend field trip. General Ecology (BIOL 3034) or an equivalent course is a prerequisite; Community Ecology (BOT 5023) and prior statistical experience would be helpful. Meets Tuesdays 10:30-11:45 AM in LSE 101, AND 3:30-4:45 PM in Math Sciences 108.

Selected Bibliographies on the use of ordination methods can be found here.


LECTURES
Notes are available from the Ordination Web Page.

Classification


HOMEWORK
Prospectus: class project
Homework #1: t-tests and regression
Homework #2: multiple regression
Assignment: select a paper on indirect gradient analysis

Homework #3: distance and weighted means
Homework #4: get your species data into shape 

Homework #5: get your environmental data into shape

The Class Project - first version

The Final Symposium

Demo Data Set

Stats and Math from General Ecology lab manual

Lertzman's guide to writing

Gopen and Swan's Science of Scientific Writing

Putting things in Better Order

 

 



Here is the SYLLABUS FOR BOT 5533.  However, please note that it is extremely unlikely that we will stick precisely to schedule!  The topics and timing may be modified to suit the needs of class participants.
 


The above is a hypothetical example of a coenocline. One of the chief purposes of ordination methods in ecology is to discover which coenoclines are the most important. Sometimes this is possible even if there is much experimental error, or even if we have no measurements of the environment.