0

I have a matrix of size (Mx3).Column 1 & 2 are latitude and longitude, respectively. Column three is temperature value. I want to do area binning of the data so that in every 100 meter square bin show 1 point instead of a few points.

Any one has an idea how to do area binning geographically?!

Masterminds
  • 1
  • Are you wanting an average of all the temperature values within each square? And assuming that's true, your resolution is better than 10m by 10m right? i.e., you have at least one point per 100m^2 section? Is it over a small enough area that you can assume it to be flat, or will projection be a problem? – chrisb2244 Aug 27 '14 at 05:04
  • Adding to @chrisb2244's question - what is the largest distance between points you have there? Do they all fit in a single UTM slice? – Dev-iL Aug 27 '14 at 09:15
  • This problems have too many dimensions. You need to specify the problem to reduce the scope of the question. How large is the area, does it fit in one UTM zone, do you have mapping toolbox? If you want to do it in a simple way you need mapping toolbox. Otherwise check [wikipedia](http://en.wikipedia.org/wiki/Universal_Transverse_Mercator_coordinate_system), to find out about lat,lon -> UTM transformation. And to @chrisb2244, you may start getting dimension problems even after a few bins, so if it must look good and an xy-grid is wanted it is recommended that you interpolate the coordinates. – patrik Aug 27 '14 at 12:05

0 Answers0