Conclusion
While the biogeochemical consequences of geometrid moth outbreaks in subarctic birch forests have attracted a lot of attention, less consideration has been given to background insect herbivory and its contribution to elemental cycling in this system. We showed that the leaf area loss due to background insect herbivory (~1.6%) was comparable to what was previously found in dwarf shrub tundra, yet lower than ecosystems at lower latitudes, and an increase with climate warming should therefore be anticipated. The nutrient fluxes from canopy to the soil associated with current background herbivore intensities were 1-2 orders of magnitude lower than the fluxes through senesced litter and soil input from external sources. The variation in background insect herbivory rates was substantial, however, and tended to increase with elevation at the local scale, and annual solar radiation, but showed no link to overall elevation, temperature, nor humidity variation. This supports previous reviews suggesting that most commonly recorded abiotic driver variables are not always sufficient to predict variation in insect herbivory and other ecological processes in mountainous landscapes. We speculate that accounting for context dependencies by introducing moderator variables in ecosystem models, e.g. latitude or biome, may be a fast way forward to allow better integration of data from elevational gradients, until we have sufficient understanding of the discrepancies between biotic-abiotic relationships in space-for-time substitution studies at different scales.
Acknowledgements: We thank Edward Diggory, Anders Voss, Jens Arne Antoft, Mads Engelstoft, Simon Risum, Tue Mariager, Marco Lund, Thomas Heister, Jens-Jakob Kratmann Nissen and Rasmus Svingel for valuable assistance in the field, and Femke Pijcke and Esben Nielsen for help in the laboratories in Lund and Copenhagen.
Data Accessibility Statement: All data will be made available at a public repository once the manuscript has been accepted. Until then, it is available upon request from the corresponding author (jeppe.aa.kristensen@gmail.com).
Competing interests: The authors declare no conflicts of interest.
Author contributions: JAK and DBM conceived the ideas and designed the research. JAK performed the fieldwork, while JAK and AM conducted the laboratory analyses. JAK analysed the data and wrote the first draft of the manuscript. DBM and AM made substantial contributions during revisions and approved the final version for publication.
Funding: This work was supported by the Swedish Research Council (VR 2014-05177, received by DBM), the European Research Council (ECOHERB – 682707, received by DBM), received by DBM, the Royal Physiographic Society of Lund (received by JAK) and Helge Ax:son Johnsons stiftelse (received by JAK).