Drivers and predictability of forest fire activity across the circumboreal zone of Northern Hemisphere (PREFIRE)

点击翻译按钮获取中文摘要

The ability to predict forest fire activity at the monthly, seasonal, and above-annual time scales is critical to mitigate its impacts, including the fire-driven dynamics of the ecosystem and socio-economic services. Fire is the primary driving factor of ecosystem dynamics in the boreal forest, directly affecting global carbon balance and atmospheric concentrations of the trace gases, including carbon dioxide. The resilience of the ocean-atmosphere system provides the potential for the advanced detection of the upcoming fire season intensity. Large scale anomalies involving the energy balance of world oceans and the associated atmospheric pressure patterns have been repeatedly shown to affect drought conditions, pointing to the large-scale changes in humidity and energy transport as important drivers of regional fire activity. Earlier studies showed strong links between changes in sea surface temperatures (SST) and associated pressure patterns, forcing changes in land precipitation and, subsequently, regional fire regimes. It is reasonable to expect that similar teleconnections between ocean and atmosphere, controlling fire regimes, extend over the entire circumboreal region. A likely source of such teleconnections is interactions between the Arctic climate (such as the sea-ice conditions), El Nino Southern Oscillations (ENSO), and Atlantic Multi-decadal Oscillations (AMO). In particular, the proximity of this biome to cold and dry Arctic air masses makes the boreal forests sensitive to the ongoing climate change, increasing the probability of long-lasting extreme weather events, including spring and summer droughts. It is also possible that climate extremes behind the increase in forest fire activity may be a product of global features of atmospheric features, such as amplification of planetary waves. PREFIRE seeks: (1) to identify climatic drivers controlling boreal fire activity and its predictability at the monthly, seasonal and annual timescales by relying on analyses of multiple proxies of modern and historic fire activity and climate-ocean variability, and (2) to provide decadal and century-scale predictions of future fire activity and to translate them into impacts on ecosystem services (in particular forest productivity and C storage capacity). I argue that there is a strong potential in using a large body of dendrochronological reconstructions to improve the predictability of weather extremes, such as periods of regionally increased fire activity. I propose that their analyses with independently obtained proxies of climate variability and vegetation cover should contribute towards better knowledge of modern climate drivers of forest fires and the predictability of fire activity at multiple temporal scales.