Can anyone help me out on the best package to sample from a spatial point dataset?
My dataset consists of a list of positions inside a region, and for each point I also have the name/number of the subregion where the point is located. I would like to make my sampling using only this dataset (for example I don't want to load a shape file with the subregions or a probability raster).
I have already made a "random" sampling for each subregion, however I am not taking into account the spatial distribution of the points inside each subregion.
I am looking for a package that would let me do both spatial random sampling as well as stratified random sampling, or some guidelines on how I can create my own code.
-
1I will point out that if you utilize a model that creates a "spatial distribution" of points, they are no longer random. This is because for a point process to follow some sort of spatial distribution you are rejecting the Complete Spatial Randomness (CSR) assumption. If you are attempting to capture a spatial process then use a point process model or stratify otherwise, keep with a random sample. I do have some sampling functions available in the spatialEco package, including a stratified sample, that may be of use.Jeffrey Evans– Jeffrey Evans2018年03月30日 15:15:35 +00:00Commented Mar 30, 2018 at 15:15
3 Answers 3
I'd suggest the spatstat package. Perhaps check out the quadratresample function. They also have several others to simulate random patterns that may fit your need (e.g. rstrat
and rsyst
). Random sampling should be pretty trivial to accomplish on your own (see the sample
function in base R)
If you have access to ArcGIS 10x, the Geospatial Modelling Environment from Spatial Ecology does a fantastic job of integrating ArcGIS and R functions to create very useful workflows. The random sampling tools are the best I have come across. http://www.spatialecology.com/gme
With spsample
from library(sp) it is easy to generate random points inside a polygon or spatial.
rpoints<-spsample(as_Spatial(shape),n=Numbre_point,"random")