COS 104-9 - Safe sites and seed sources: Searching for drivers of post-fire larch (Larix cajanderi) regeneration in northeastern Siberia

Friday, August 16, 2019: 10:50 AM
L006, Kentucky International Convention Center
Eric B. Borth1, Heather D. Alexander2 and Ryan W. McEwan1, (1)Department of Biology, University of Dayton, Dayton, OH, (2)Department of Forestry, Mississippi State University, Mississippi State, MS
Background/Question/Methods

The Arctic is warming at an alarming rate, a phenomenon which has been causing more frequent and severe wildfires in boreal forests. The Cajander larch (Larix cajanderi) forests of Siberia are particularly important because they cover an immense area of carbon-rich permafrost, one of the largest terrestrial carbon sinks in the world. The fate of this belowground carbon is intimately tied to the structure of these larch forests, so we seek to understand how changes in fire regimes may affect factors that drive seedling establishment. The first objective was to examine how stand structure relates to stand level seed production. Full tree cone counts were conducted on 150 trees from 18 similar-aged stands across a tree density gradient. The second objective was to evaluate the importance of microsites that favor larch establishment (deemed “safe sites”), a concept that has been anecdotally recorded, but not rigorously studied in these forests. We examined 15 Cajander larch stands in northeastern Siberia that had burned in the past 18 years. We analyzed the strength of seedling preference by comparing the percentage of seedlings that establish on safe sites to the percentage of the area that is occupied by safe sites.

Results/Conclusions

Results from objective 1 showed that medium and low density stands produce the most seeds per hectare. This pattern seems to reflect a release from competitive stress that limits cone production in the higher density stands (>10,000 trees per hectare), while also showing that stands with very low tree density (<500 trees per hectare) face diminishing returns of competition release as resources may be underutilized. This information may help predict future ecosystem structure because seed availability is likely the most important driver of forest regeneration post-fire. The results for objective 2 indicated that safe sites were indeed preferentially colonized by larch seedlings. This effect was stronger in sites with greater seed limitation. The preference for safe sites was also stronger in lowland/floodplain sites, which may be due to the differences in types of safe sites that are predominant in lowland versus upland sites. By synthesizing these results regarding seed production and seedling establishment, we can better predict the success of reforestation after wildfires which influence the future trajectory climate feedback loops.