Do introduced North American beavers Castor canadensis engineer differently in southern South America? An overview with implications for restoration.

1. Twenty-five pairs of North American beavers Castor canadensis Kuhl were introduced to Tierra del Fuego Island in 1946. The population has expanded across the archipelago, arriving at the Chilean mainland by the mid-1990s. Densities range principally between 0.5–2.05 colonies/km. They have an impact on between 30–50% of stream length and occupy 2–15% of landscape area with impoundments and meadows. Beaver impacts constitute the largest landscape-level alteration in subantarctic forests since the last ice age. 2. The colonization pattern, colony densities and impacted area indicate that habitat in the austral archipelago is optimal for beaver invasion, due to low predator pressure and suitable food resources. Nothofagus pumilio forests are particularly appropriate habitat, but a more recent invasion is occurring in adjacent steppe ecosystems. Nonetheless, Nothofagus reproductive strategies are not well adapted to sustain high beaver population levels. 3. Our assessment shows that at the patch-scale in stream and riparian ecosystems, the direction and magnitude of exotic beaver impacts are predictable from expectations derived from North American studies, relating ecosystem engineering with underlying ecological mechanisms such as the relationships of habitat heterogeneity and productivity on species richness and ecosystem function. 4. Based on data from the species' native and exotic range, our ability to predict the effects of beavers is based on: (i) understanding the ecological relationships of its engineering effects on habitat, trophic dynamics and disturbance regimes, and (ii) having an adequate comprehension of the landscape context and natural history of the ecosystem being engineered. 5. We conclude that beaver eradication strategies and subsequent ecosystem restoration efforts, currently being considered in southern Chile and Argentina, should focus on the ecology of native ecosystems rather than the biology of this invasive species per se. Furthermore, given the nature of the subantarctic landscape, streams will probably respond to restoration efforts more quickly than riparian ecosystems. [ABSTRACT FROM AUTHOR]