Isolation by distance is the phenomena that (even partial) reproductive isolation due to geographical distance between reproductive populations will result in greater genetic distance between those populations. This phenomena is not the only thing that can affect genetic distances, including other forms of reproductive isolation.
Given that the example plot of genetic distance on wikipedia appears to be a linear-ish relationship, perhaps the effect that geographical distance has on genetic distances could be subtracted via a linear regression model. I'm not sure how common linear relationships are for this type of data, but you can find examples of non-linear relationships such as found in "Genetic and phenotypic divergence in an island bird: isolation by distance, by colonization or by adaptation?". Other studies, such as "Direct and indirect measures of dispersal in the fairy shrimp Branchipodopsis wolfi indicate a small scale isolation-by-distance pattern", show pretty marginal linearity.
Is there a conventional procedure in the literature for removing the isolation by distance effect on the genetic distances?