For wind-dispersed trees you can use sticky pollen traps placed around individual trees or 'in silico' phenomenological modelling. Determining patterns is more difficult with insect disposal of pollen – the best way is to develop a mechanistic SEIB model based on the phenology of the trees and behaviour of insects. I developed the latter for seed dispersal by frugivorous birds. See my paper in Ecography entitled 'Exploring the interaction of avian frugivory and plant spatial heterogeneity and its effect on seed dispersal kernels using a simulation model'.

I hope this helps :-)

Sure, Thank for your information, I'll read your paper

Regards :) 

Please check the PDF attachment.

Or you can estimate pollen dispersal distances from spatial genetic structure. Paternity analysis with microsatellites seems to be used most frequently.