Dry and warm conditions have exacerbated the occurrence of large and severe wildfires over the past decade in Canada’s Northwest Territories (NT). While temperatures are expected to increase during the 21st century, we lack understanding of how the climate-vegetation-fire nexus might respond. We used a dynamic global vegetation model to project annual burn rates, as well as tree species composition and biomass in the NT during the 21st century using the IPCC’s climate scenarios. Burn rates will decrease in most of the NT by the mid-21st century, concomitant with biomass loss of fire-prone evergreen needleleaf tree species, and biomass increase of broadleaf tree species. The southeastern NT is projected to experience enhanced fire activity by the late 21st century according to scenario RCP4.5, supported by a higher production of flammable evergreen needleleaf biomass. The results underlie the potential for major impacts of climate change on the NT’s terrestrial ecosystems.

Wildland fire is the most important disturbance in the boreal forests of eastern North America, shaping the floral composition, structure and spatial arrangement. Although the long-term evolution of the frequency and quantity of burned biomass in these forests can be estimated from paleo-ecological studies, we know little about the evolution of fire sizes. We have therefore developed a methodological approach that provides insights into the processes and changes involved over time in the historical fire-vegetation-climate environment of the coniferous forests (CF) and mixedwood forests (MF) of eastern boreal North America, paying particular attention to the metric of fire size. Lacustrine charcoal particles sequestered in sediments from MF and CF regions were analyzed to reconstruct changes in estimated burned biomass, fire frequency, and their ratio interpreted as fire size (FS-index), over the last 7,000 years. A fire propagation model was used to simulate past fire sizes using both a reference landscape, where MF and CF compositions over time were prescribed using pollen reconstructions, and climate inputs provided by the HadCM3BL-M1 snapshot simulations. Lacustrine charcoals showed that Holocene FS-indices did not differ significantly between MF and CF because of the high variability in fire frequencies. However, the estimated burned biomass from MF was always lower than that from CF, significantly so since 5,000 BP. Beyond the variability, the FS-index was lower in MF than CF throughout the Holocene, with slight changes in both forests from 7,000 to 1,000 BP, and simultaneous increases over the last millennium. The fire model showed that MF fires were consistently smaller than CF fires throughout the Holocene, with larger differences in the past than today. The fire model also highlighted the fact that spring fires in both forest types have always been larger than summer fires over the last 7,000 years, which concurs with present-day fire statistics. This study illustrates how fire models, built and used today for forecasting and firefighting, can also be used to enhance our understanding of past conditions within the fire-vegetation-climate nexus.

Context
The Canadian boreal forest provides valuable ecosystem services that are regionally and globally significant. Despite its importance, the future of the Canadian boreal forest is highly uncertain because potential impacts of future climate change on ecosystem processes and biomass stocks are poorly understood.

Objectives
We investigate how anticipated climatic changes in coming decades could trigger abrupt changes in the biomass of dominant species in Canada's boreal forests. Methods Using the dynamic global vegetation model LPJ-LMfire, which was parameterized for the dominant tree genera in Canada's boreal forests (Picea, Abies, Pinus, Populus) and driven by a large range of climate scenarios grouped by two forcing scenarios (RCP 4.5/8.5), we simulated forest composition, biomass, and the frequency of disturbance, including wildfire, from Manitoba to Newfoundland.

Results
Results suggest that responses of this region to a warmer future climate will be very important, especially in southern boreal areas and under the RCP 8.5 forcing scenario. In these areas, reductions of total aboveground biomass incurred by fire and heat-induced tree mortality events are projected; the fertilizing effect of increasing atmospheric CO2 on forest productivity is unlikely to compensate for these losses. Decreases in total forest stocks would likely be associated with forest cover loss and a shift in composition in particular from needleleaf evergreen (softwood) to broadleaf deciduous (hardwood) taxa.

Conclusion
The simulated future reduction in softwood biomass suggests that forest management strategies will have to be adapted to maintain a sustainable level of forest harvest and tree density that meets demands for wood products, while maintaining other ecosystem services.

Wildland fires are the main natural disturbance shaping forest structure and composition in eastern boreal Canada. On average, more than 700?000?ha of forest burns annually and causes as much as CAD 2.9 million worth of damage. Although we know that occurrence of fires depends upon the coincidence of favourable conditions for fire ignition, propagation, and fuel availability, the interplay between these three drivers in shaping spatiotemporal patterns of fires in eastern Canada remains to be evaluated. The goal of this study was to reconstruct the spatiotemporal patterns of fire activity during the last century in eastern Canada's boreal forest as a function of changes in lightning ignition, climate, and vegetation. We addressed this objective using the dynamic global vegetation model LPJ-LMfire, which we parametrized for four plant functional types (PFTs) that correspond to the prevalent tree genera in eastern boreal Canada (Picea, Abies, Pinus, Populus). LPJ-LMfire was run with a monthly time step from 1901 to 2012 on a 10?km2 resolution grid covering the boreal forest from Manitoba to Newfoundland. Outputs of LPJ-LMfire were analyzed in terms of fire frequency, net primary productivity (NPP), and aboveground biomass. The predictive skills of LPJ-LMfire were examined by comparing our simulations of annual burn rates and biomass with independent data sets. The simulation adequately reproduced the latitudinal gradient in fire frequency in Manitoba and the longitudinal gradient from Manitoba towards southern Ontario, as well as the temporal patterns present in independent fire histories. However, the simulation led to the underestimation and overestimation of fire frequency at both the northern and southern limits of the boreal forest in Québec. The general pattern of simulated total tree biomass also agreed well with observations, with the notable exception of overestimated biomass at the northern treeline, mainly for PFT Picea. In these northern areas, the predictive ability of LPJ-LMfire is likely being affected by the low density of weather stations, which leads to underestimation of the strength of fire–weather interactions and, therefore, vegetation consumption during extreme fire years. Agreement between the spatiotemporal patterns of fire frequency and the observed data across a vast portion of the study area confirmed that fire therein is strongly ignition limited. A drier climate coupled with an increase in lightning frequency during the second half of the 20th century notably led to an increase in fire activity. Finally, our simulations highlighted the importance of both climate and fire in vegetation: despite an overarching CO2-induced enhancement of NPP in LPJ-LMfire, forest biomass was relatively stable because of the compensatory effects of increasing fire activity.