User Tools

Site Tools


Pre-registration open! (for students outside NTU, deadline December 31, 2018) Flyer (pdf file)

Numerical Methods in Community Ecology


Course focused on the analysis of community ecology data in R, organized by Institute of Ecology and Evolutionary Biology, College of Life Science, National Taiwan University

Instructor: David Zelený (澤大衛), Vegetation Ecology Lab

Language of the course: English

Course content: analysis of community ecology data, including ordination and cluster analysis, diversity analysis and analysis of species attributes. We will use real community ecology data (mostly vegetation and zoological datasets) and practice the analysis using the R project.

Target audience: senior undergraduate and graduate ecology students focused on botany and zoology, who are planning to do a study at the community level (i.e. not on a single species, but on the multiple species occurring at multiple localities). The class may be useful also for other disciplines handling multivariate data (e.g. microbiology), but the main focus is on ecological data. If you study PhD and want to join, you are welcome, but expect that some of the teaching materials will cover rather basic things.

Teaching strategy: The class includes presentation of theory, application of methods using R program, solving individual exercise during the class, homework assignments, test of knowledge (analogy of midterm and final test) and preparation of individual projects and their presentation (in English).

Schedule: 5 days during 3 weekends (in 2019 preliminarily set to 2/23-24, 3/23-24 and 5/4-5). In selected weekends, the course will start on Saturday morning at 9 a.m. and end on Sunday evening around 6 p.m.\\. First two weekends are focused on lectures and practice, the third weekend (including only Saturday) will be final exam and presentations of individual projects.

Location: Life Science Building, National Taiwan University, Taipei.


  • If you are NTU (or NTNU, NTUST) student, you can enroll for this course under the code EEB5083 (B44 U1950).
  • If you are from other university, you can tentatively pre-register for this course online and participate for free. Deadline for pre-registration is December 31. All pre-registered participants will get posibility to register for the course. In the case that you meet all criteria for final evaluation (homework assignments, midterm quiz, final test and final presentation), you will receive a certificate about the participation in the course.


  • Basic knowledge of R program: the course is not focused on advanced use of R, but we will use R as a tool to do all the analysis, so the basic skill in operating R is necessary. You can learn R by yourself in advance or participate in some of the R courses.
  • Basic knowledge of statistics is expected (correlation, regression, ANOVA, testing the significance, confidence intervals etc.)
  • You need to bring your own computer with installed R and RStudio, and access to Internet (if you are a student from outside NTU, you can use EDUROAM connection available in the classroom).

Information for students outside of Taipei

  • The course is taught during the weekend to make it convenient for students outside of Taipei to attend. Please, find an accomodation near the NTU's campus, and make sure you can be in the class on Saturday morning 9 a.m. (if you are coming from farer distance, consider arriving to Taipei already on Friday night). If you do not wish to get certificate, you do not need to attend the third weekend (only one day focused on final test, presentations and discussion), even if I would suggest you to do so. If you have any question related to the participation in the course, please feel free to contact me (!

Disclaimer: this is rather an intensive course, focused on a theory of community data analysis, and practical exercise using R on real datasets. Along to the lectures taught in the classroom, you need to also complete homework assignments, midterm quiz, final test and prepare (and present in English) the final project focused on analysis of community data. This is not an R course for advanced use of R program (I expect you have a basic knowledge of R before you enter the class). If you want to learn R, consider taking my other class Introduction to R for Ecologists (regular 3 credit class taught every winter semester).

Teaching schedule (combined theoretical and practical part)

Five days (8.5h/day) during three weekend blocks (Saturday + Sunday)

Topic Number of classes
Introduction, types of data (categorical vs quantitative, abundances, frequencies). 1
Pre-analysis data preparation (data cleaning, outliers, transformation, standardization, exploratory data analysis). 1
Ecological similarity (indices of ecological similarity and distance between samples). 1
Ordination (theory behind, linear vs unimodal, constrained vs unconstrained methods, PCA, CA, DCA, RDA, CCA, NMDS and some others, ordination diagrams, permutation tests, variance partitioning, forward selection, case studies). 3-4
Numerical classification (hierarchical vs nonhierarchical, agglomerative vs divisive; TWINSPAN) 1-2
Indicator value analysis (IndVal), diagnostic species, fidelity of species to sample groups. 1
Use of species functional traits or species indicator values in multivariate analysis (functional traits, species indicator values, community-weighted mean, fourth-corner, RLQ analysis). 1
Analysis of diversity (alpha, beta and gamma diversity, accumulation and rarefaction curves, true diversity, species abundance distribution, diversity estimators). 2
Case studies demonstrating the use of particular analytical methods. as a part of each class

Each class will be composed of two parts: theoretical introduction to the method, and practical lab, using the R program for all analyses. You need to bring your own computer with installed R and wifi access to internet.


  • Borcard, D., Gillet, F. & Legendre, P. 2011. Numerical Ecology with R. Springer.
  • Legendre, P. & Legendre, L. 2012. Numerical Ecology. Third English edition. Elsevier Science BV, Amsterdam.
  • Šmilauer, P. & Lepš, J. 2014. Multivariate Analysis of Ecological Data using Canoco 5. Second Edition. Cambridge University Press, Cambridge, UK.

numecol/start.txt · Last modified: 2018/12/10 13:08 by david