What comes after the new pest? Ecosystem transitions following insect pest outbreaks induced by climate change in the European high North

Moth outbreaks of unprecedented extent have affected 1 mill ha of boreal birch forest in northern Fennoscandia, and is arguably among the most abrupt, and large-scale terrestrial ecosystem disturbance attributed to recent climate change in Europe. Our research has targeted how the ecosystem in the most disturbed region may transform into new ecosystem states under continued climate warming, recurrent pest outbreaks, the action of wild and domestic herbivores and forest management. Such new ecosystem states may imply loss of biodiversity and changed functions within the ecosystem, that involves feedback to the climate system and provisional ecosystem services. With a combination of large-scale comparative and small-scale experimental studies, we have targeted the following topics: 1) Regional ecosystem state transitions following moth outbreaks, 2) The importance of large herbivores as drivers and maintainers of state changes, 3) Salvage logging as a management action, and 4) Ecosystem cascades. We have shown that the intensified outbreaks that occurred during the 2000?s have exceeded the resilience of the birch forest and caused large-scale mortality, locally above 90% of stems. We have identified a strongly non-linear relationship (a threshold) between defoliation intensity and stem mortality, indicating that this ecosystem is prone to abrupt transitions from a forested to a non-forested state. Low recovery on a regional scale, likely both due to herbivory from mammalian grazers/browsers and a positive feedback between the density of surviving stems and sapling establishment, suggests that this transition will be long-term, or even permanent, in many areas. The role of mammalian herbivores has been further quantified by means of experimental exclosures. We have shown particularly 1) that a higher proportion of trees die during the outbreak in areas which are subject to summer grazing by reindeer, and 2) that summer grazing impairs the establishment of new forest. Four years after the outbreak (2012) the proportion of surviving stems was <2% and ~5% in summer and winter grazed areas respectively. Six years later (2018), this was unchanged in summer grazed areas but had increased to ~40% in winter grazed areas. This shows that the two grazing regimes are on two completely different trajectories, one towards recovery, and one towards a non-forested state. Forest management can potentially direct these trajectories. In collaboration with forest management authorities, we have tested how salvage logging of dead and damaged trees shortly after an outbreak, can increase forest recovery. On a short term, logging is a promising action, with approx. 4 times more basal shoots being produced in logged stands relative to unlogged controls. The long-term benefits, however, will depend both on the intrinsic state of the forest, mainly site conditions, and on the local browsing pressure from large herbivores. As a management action, salvage logging should be carefully targeted towards those stands and areas where it is most likely to be successful. Moth outbreaks, and the short- and long-term state changes following, are likely to influence other parts of the birch forest ecosystem, including biodiversity such as bird and insect communities, game species such as moose, ptarmigan and hare, and associated small predators, for instance the red fox. These cascading effects are poorly known, and often difficult to quantify. We have targeted this with a diverse array of methods, incl. field surveys (snow tracking, acoustic surveys), camera trapping, and insect flight interception traps. We have shown that bird communities so far seem robust towards the changes, and that communities of dead wood associated insects, although species rich, respond numerically less than might have been expected to the drastic increase of dead wood available in the system. The suggests that these insects may not play a strong functional role in the decomposition of the dead wood, at least at the present state of succession. Further, we have shown that herbivores in the system, in particular ptarmigan ? but also moose, are responding to forest damage through lower presence in heavily damaged areas, and that this cascade also to the generalist predator red fox. The long-term state changes and implications of continued warming and recurrent moth outbreaks cannot be elucidated in a short-term project, and this project has benefitted from both research infrastructure and time series established by us during previous NFR-projects. Despite of research conducted over a decade, we are still in an early phase of understanding the mechanisms behind, and the implications of the intensified moth outbreaks for the future of the birch forest. We therefore aim at extending the present research and partnership with managers into the long-term through the adaptive monitoring system COAT (www.coat.no).

Prosjektet har bidratt til et varig samarbeid med forvaltningsmundigheter, grunneiere og næringer i Finnmark samt til oppbygningen av overvåkningssystemet COAT.

The proposal targets one of the most pervasive, large-scale drivers of ecosystem change in the circumpolar North - the spread and intensification of forest insect pest outbreaks resulting from climate warming. Specifically, we target the impact of moth outbreaks of unprecedented extent and severity that recently have devastated 1 million ha of boreal birch forest in northern Fennoscandia - arguably the most abrupt, severe and large-scale ecosystem disturbance that can be attributed to recent climate change in Europe. Our research aims to unravel how the ecosystem in the most disturbed region at the forest-tundra interface may transform into new ecosystem states under continued climate warming, recurrent pest outbreaks, the action of wild and domestic herbivores and forest management. We target state variables that quantify overall ecosystem structure - including sub-arctic biodiversity, ecosystem functioning - including processes that feed back to the climate system and provisional ecosystem services. We employ an approach that allows us to interactively test and refine hypotheses about drivers of long-term state transitions by a combination of large-scale comparative and small-scale experimental studies. We will use a battery of methods ranging from remotely sensed measurements of landscape-scale primary productivity and surface reflectance (albedo), to ground and below-ground estimates of key species and functional groups. State-of-the-art multivariate statistical modeling will be used throughout to test predictions. In cooperation with stakeholders we will investigate management actions that may mitigate detrimental ecosystem states and prevent loss of ecosystem services and biodiversity. We aim to extend the present research and partnership with managers into the long-term through the adaptive monitoring system COAT. Extensive outreach from our research will include interaction with pupils and teachers through TUNDRA schoolnet.