Analytics: tips and tricks for analyzing bird data in R

by Wouter Vansteelant

The number of ecologists who are tracking birds with GPS-loggers and other tracking devices is rising quickly. New technologies are thus greatly expanding the potential for studying the flight behaviour of birds across multiple scales. However, in order to understand the decisions of individual birds we must take into account the environmental conditions which they encounter during their flights. Fortunately, meteorologists and remote sensing specialists provide numerous data resources online and free of charge. Nevertheless, processing the massive amounts of data produced by tracking studies in combination with environmental data remains an arduous task for many ornithologists. Let´s admit, we would rather spend our lives chasing birds in the field rather than spending countless hours in front of a computer.

As a Ph.D. student at the Computational Geo-ecology group of the Institute for Biodiversity and Ecosystem Dynamics I´ve devoted four years of my life studying how weather shapes the migratory movements of European Honey Buzzards Pernis apivorus at multiple scales. During this time, I´ve learnt a thing or two about the analysis of bird movements in relation to environmental conditions, and I´ll continue to learn more about data analysis for some time to come. Through this blog I hope to share some of these skills with you, fellow researchers.

My aim is to share a new blog post every month or so. For starters I´ll mainly be posting tips and tricks for analysing movement data and environmental datasets in R: a wonderful and versatile free software for statistical programming and building powerful graphics. In addition, I will be writing about visualization of various other kinds of bird data in the more distant future. I honestly can´t be bothered publishing all this information in the form of scientific publication. Therefore, I open up the blog for comments, hoping for spontaneous peer-review of the R code by you, the research community. With your help we should be able to filter out bugs and to keep the code up-to-date as the R software continues to be developed in the future.

 Yours truly,

 ~ Wouter