[R-sig-Geo] gDistance on a LARGE Spatial Object?

[Peter Versteegen]

Wouter Steenbeek <WSteenbeek <at> nscr.nl> writes:
 
Your problem suggests that the spatial extent is not very large.
I have a similar problem with over 600,000 data points.
They describe bathymetric measurements of the lake.
I'm working on making contours..

Project all of your coordinates on a plane.
Means that you can work with meters instead of lat/lon.
Euclidian distances are easily calculated, d = sqrt((x-x1)^2 + (y-y1)^2).
You have to write a few lines of basic code in R.

Setup the problem with a database or write/read a textfile.

Routine computations like this 600,000 times
looking at each of the 600,000 points,
takes under a minute on my aging Mac.


I hope this helps.