“…We hypothesize that these results are likely due to three nonmutually exclusive hypotheses: (1) Strong territoriality and breeding site fidelity in adults increase the similarity of the individuals within a river, (2) geographic (e.g., watershed boundary, high mountain ranges) or climatic barriers (e.g., large deserts) isolate rivers preventing the dispersal of the individuals, and driving an independent evolution of each riverine population by genetic drift, and (3) individual dispersal capacity (i.e., frequency and distance) describes an isolation-by-distance pattern that in short distance dispersal resembles a metapopulation model. For example, the year-round territorial behavior, the long-term pair bonds, and the strong site fidelity in adult torrent ducks, as mentioned in the first hypothesis, can prevent gene flow among rivers, but also contribute to the similarity of the individuals of a growing population in vacant or improved habitat, and thus increase the relatedness within each riverine population and develop an "extended family" (Alza et al, 2017;Cardona & Kattan, 2010;Eldridge, 1986;Hartl & Clark, 2007;Pernollet, Estades, & Pavez, 2012). Additionally, as referred in the second hypothesis, torrential river systems inside watersheds are analogous to islands (islands on mountain slopes) that isolate and sustain populations and communities (Black, 1997;Naiman, Magnuson, McKnight, Stanford, & Karr, 1995;Omernik & Bailey, 1997;Sullivan, Watzin, & Keeton, 2007), similarly observed in lakes (Barbour & Brown, 1974), marshes (Brown & Dinsmore, 1988), caves (Culver, Holsinger, & Baroody, 1973), mountaintops (Nores, 1995), or woodlots (Holland, 1978).…”