Conclusion
Using genomic evidence and modern demographic inference procedures with machine learning, we have shown evidence for the persistence of ITR populations of both ecosystem dominant tree species Thuja plicata(western redcedar) and Tsuga heterophylla (western hemlock) throughout the Pleistocene. This is critical because both other mesic forest disjuncts and ITR endemics (e.g., endemic jumping slugs such asHemphilia camelus , H. skadei , and H. danielsi ; Rankin et al. 2019) are rainforest-dependent. Our results, as well as the recent results for western redcedar of Fernandez et al. (2021), indicate that these late-successional dominant tree species persisted throughout the Pleistocene, providing habitat for rainforest dependents, including numerous understory plant species (Björk 2010) that form the ecosystem. Further, the refugial populations in the ITR were likely small, as we show support here for Pleistocene-related population bottleneck events in both species. This evidence does coincide with the paleontological record, which suggests that the temperate rainforest did not dominate the PNW landscape until the last 5000 years, and only in the last 3500 years did the expansion of ITR begin. Coupled with the recent population expansion, we also show evidence for secondary contact at this time between the coastal and ITR populations for both species. This recent gene flow has likely muddled other genetic inferences made about western redcedar previously, which suggested that the ITR populations were a result of coastal recolonization. While our data indicate that coastal migrants contributed to the genetic architecture of the current ITR populations, our data provide strong evidence that Pleistocene refugia contributed to that architecture as well. This is supported by the high proportion of rare alleles observed in the ITR populations for Tsuga heterophylla and Thuja plicata , rare alleles that could only be the result of an ancient vicariant event with the coastal population.