Ecological theory proposes that the geometry and dynamics of suitable habitats are important predictors for the persistence of a population or metapopulation. A key finding supporting a metapopulation-like conceptualization of extinction and colonization in fragmented salmonid populations is that individuals of particular species are more likely to be absent from small patches of suitable habitat than from large patches. Extinctions from small patches occur for a few reasons that can be roughly classed as physical catastrophic causes (e.g. a major channel reorganizing event) or as biological small-population effects related to genetics and demographics. One important question with implications for land management is how strong of a role physical disturbances play in determining presence and absence of a species within a patch. If disturbance plays a key role in structuring populations, then we would expect to observe two patterns. First the spatial scale of disturbance should be on the order of the size of patches that are commonly unoccupied. The second prediction would be that the spatial scale of genetic variability should show strong gene flow at scales smaller than the scale of disturbance. We examined aerial photography over the last 40 years in central Idaho to measure the scale of stream disturbance patches. The scale of stream disturbances was on the same order of magnitude as the divide between commonly occupied and unoccupied patches, and was smaller than scales of high gene flow among bull trout populations, failing to reject the possibility that disturbance plays an important role in structuring populations in this basin. Management for persistence of the species must consider both the likelihood of decreased habitat patch sizes and increased disturbance scales with global warming. Knowing disturbance patch scales and how disturbance operates to structure populations can be used to consider a disturbance based paradigm for management of sensitive stocks using a minimum dynamic area approach in wildland systems to replace the loading based approach to cumulative watershed effects as exemplified by clean water act regulations.