Predicting the genetic purity of Cutthroat trout in Rocky Mountain National Park

Background/Question/Methods

Introduced and invasive species have major ecological, evolutionary, and economic impacts on native species and ecosystems.  Trout have been moved within and beyond their native habitat at least since European-Americans arrived in the American West.  Trout were introduced to both populated and historically fishless waters for recreation, subsistence, and even conservation purposes.  The introductions happened on a massive scale: in Rocky Mountain National Park (RMNP) alone, nearly 20 million eggs, fry, or fingerlings of Oncorhynchus (the genus that includes rainbow and cutthroat trout) were stocked into 1500 km of stream or lakes between 1886 and 1964.  Previous work has established that putatively pure populations of cutthroat trout in RMNP are actually admixtures of native and non-native cutthroat trout as well as rainbow trout.  Research on the spatial distribution of non-native trout suggests that the distance between habitat and potential introduction points is a predictor of genetic purity.  Hence, we tested whether distance to anthropogenic features could be used as a proxy for propagule pressure, a proxy that would be useful in the absence of detailed stocking records for predicting the distribution of genetic purity, using molecular genetic and morphological assessments of purity in combination with publicly-available GIS layers covering RMNP. 

Results/Conclusions

We found that distance to roads or trails as potential introduction points was negatively but insignificantly correlated with genetic purity (percent pure cutthroat) (r = -0.3023, R² = 0.0914).  Density of roads or trails was also not a strong predictor of genetic purity (r = 0.2026, R² = 0.0411).  We also used stocking records collected from the state archives to test our assumption that stocking compromises the genetic integrity of natural populations.  Percent purity was not correlated with number of recorded stocking events, number of individuals stocked, or years since most recent stocking.  These results indicate that there is no spatial signal in the distribution of genetic purity in RMNP in relationship with stocking.  This could be due to the persistence of pure populations intentionally stocked into historically fishless lakes.  On the other hand, the area of RMNP might be too small to detect a relationship between distance and purity.  A state or region-wide analysis could reveal such a pattern.  Overall, the lack of results imply that trout stocking has been so extensive that the signal of purity and the distribution of genetic material within historical ranges has been lost.