Tuesday, 19 November 2013

Local distribution maps for biological records in R

I thought it worth sharing this map that I created for a poster that Ambroise Baker and myself have prepared for the upcoming BSBI Annual Exhibition Meeting. It shows the occurrences of the non-native grass Polypogon viridis in Sheffield, based on records made by Ambroise and myself this year (2013). The map took a fair bit of fiddling around; working out how to correctly specify the manual scaling of the points seemed to cause me particular trouble!

Of course, R could also be used to create more traditional atlas dot maps; coastal outlines of Britain are freely available in more than one R package, and occurrence data relating to the UK grid could be manipulated in R to centre on monads, tetrads or hectads. External tools could also be used to do this before loading the data into R, e.g. see here.

The result is below. I'd be grateful to hear from anyone who can suggest any refinements to my code, or the general way in which I've gone about this.

Polypogon viridis in Sheffield, UK, 2013


  1. What an interesting blogpost, as I told you before on Twitter. You've given me an idea about sharing data of plant populations in some places in Spain. I have used R for ANOVA in statistics before and found it very difficult. It's not my cup of tea, but I want to improve.

  2. Hi Guillermo, It'd be interesting to learn more about biological data sharing in Spain -- do you have something like our National Biodiversity Network, or our Biological Records Centre? Or are things done at the regional level?

    The great thing about R these days is the number of examples online, it's usually possible to put things together from all the code out there. Or, better, find an R-savvy collaborator and just provide the ideas ; ) That way you'll have more time for your great paintings! Best wishes, Oli