Fire is the primary disturbance in the boreal forest, and variations in fire regimes have important ecological, biogeochemical, and socioeconomic implications. Recent widespread burning throughout the biome has been convincingly linked to climatic warming, with expectations of increased burning with future climate change. However, fire-regime dynamics are scale-dependent, and it is unclear whether empirical climate-fire relationships derived from the short observational record are applicable to future projections. Paleo-fire reconstructions offer a valuable extension to historical fire records by providing a context for ongoing change and offering insights to the causes and consequences of fire regime shifts over decades to millennia.
We reconstructed 10,000 years of boreal-forest fire history from analysis of macroscopic charcoal accumulation in sediment cores from 14 lakes in the Yukon Flats (YF), Alaska. Extending previous techniques for estimating regional biomass burning, we fit charcoal accumulation rates to a lognormal model that permits zero counts and untransformed data. We also estimated fire frequency by established peak-detection methods. Both metrics were validated against historical fire records, and then used to evaluate variability in Holocene fire-regime characteristics in light of past changes in vegetation and climate, and recent and projected boreal forest burning.
Results/Conclusions
Biomass burning and fire frequency increased with mid-Holocene expansion of highly-flammable Picea mariana, indicating millennial-scale vegetational control on fire regimes. However, while P. mariana attained modern abundance ~3,000 yrs before present (yBP), biomass burning remained variable, with a Holocene maximum (~850 yBP) during the Medieval Climate Anomaly when summers were likely as warm as present, followed by a local minimum (~150 yBP) during the cooler Little Ice Age. Meanwhile, fire frequency was stationary around a mean of one fire/century, suggesting little change in area burned. Thus, since its establishment, the YF boreal forest maintained relatively constant fire frequency, but variable biomass burning per unit area in response to centennial-scale climate fluctuations.
Since ~150 yBP, fire frequency and biomass burning have increased steeply in a novel combination, indicating no Holocene analog to the modern YF fire regime. Recent YF burning has been among the most extensive in Alaska, and has consumed much of the mature forest in the region. Thus, we speculate that YF is pushing the limits of the boreal fire regime through fuel depletion, and that increasingly fire-prone conditions due to climate change may provoke a dramatic regional ecosystem shift, from Picea-dominated boreal forest to an early-successional, deciduous-rich landscape.