[nikipedia]

why we have non-uniform-distribution nucleation site in RPI wall boiling model?

I would be grateful if you could answer my question about RPI wall boiling model.
what parameter in RPI model in fluent cause non-uniform-distribution of nucleation site density in boiling boundary?
after using this model for a horizontal plate with a constant heat flux we have non-uniform-distribution for vapor volume fraction.
I know everything about bubble frequency, heat flux partitioning, nucleation site density and speciic area for quench part. but how on a boundary in computational domain, a cell is specified as nucleation site while the neighbor cell is not? we know that both are located on the same boundary with same total heat flux.

check this
http://imageupload.co.uk/files/j00i2b0oajiqqqm490nm.png