[R-sig-Geo] Comparison between two marked point pattern data sets

[Andrew Niccolai]

Dear Spatial R group,

I disclose that I have been more active in image segmentation than spatial
statistics of late so please bear with me if this question is too pedantic.
I have a field collected marked point pattern of tree crown size classes.
The point data consists of the X,Y coordinates of the center of the tree
crown projected onto CT State Plane Datum and the estimated crown size class
(described as a radius measured in feet).  There are 8 possible size
classes: 4, 6, 8, 10, 12, 14, 16 and 20 feet.  I have produced a marked
point pattern data set with the same crown size class attribute from an
automatic crown detection algorithm using remotely sensed data.  I would
like to test how similar (or different if that turns out to be the case) the
two marked point patterns are to each other, ie, how well the size class
spatial location and distributions match.  I know this is not a perfect
match as there are 516 field measured points and 568 remotely sensed points.
However, I am looking for a metric to assess how closely the simulated size
class distribution patterns emulate the field measured patterns.

Any suggestions are truly appreciated.
Thanks,

Andrew Niccolai
Doctoral Candidate
Yale University
office (203) 432-5144
cell (860) 402-6079