I have a set of coordinates for which I want to extract temperature values from a raster. I am using the extract function from the raster library in R. Looking at the interpolation methods, there are options for simple and bilinear interpolation. I am not very familiar with raster extraction literature but I was wondering if there are other alternatives in R for other interpolation methods which give a better estimate for values at a point than bilinear interpolation. Also is bilinear the usual method used in practice? Thanks.

  • 2
    Common raster interpolation methods are mentioned in the question at gis.stackexchange.com/questions/10931. The answers there study their properties. "Better" depends on the purpose of the interpolation and the nature of the data, though: no single interpolation method is always best in all circumstances.
    – whuber
    Jun 16, 2014 at 19:29

1 Answer 1


Bilinear its a good approach when you are working with continuous data and its very usual. Simple it's useful when you are working with discrete data, because you don’t generate new values of the variable. extract() allows other methods, using buffer and fun arguments.

  • Thanks. Do you suggest any other methods that are used in practice?
    – Nitin
    Jun 16, 2014 at 17:33
  • Most of the time I work with continuous data, so i tend to use 'bilinear' Jun 16, 2014 at 18:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.