Over the past decades, the abundance and area of macrocharcoal (i.e. ≥ 150 μm in diameter) fragments from sedimentary sequences have been quantified using visual or semi-automated methods to reconstruct fire histories. However, the lack of uniformity between counting methods used in each study could introduce methodological biases influencing fire frequency reconstructions, and therefore impact their interpretation and limit their comparisons. To overcome this issue, we propose here a new automated method to quantify the number of macrocharcoal fragments and measure their areas from high-definition image capture, based on the analysis of colorimetric parameters. We tested the efficiency of our method and reconstructed charcoal influx over the last 8000 years by comparing visual and automatic counting methods along a north-south transect from eastern Canada, estimating number and size, and the associated local and regional fire frequencies. Results show that our automated method is efficient in detecting charcoal particles, except for highly minerogenic samples, and suggest that the traditional visual inspection tends to overestimate the size of macrocharcoal fragments. Local fire frequencies varied greatly depending on the macrocharcoal detection method used. At the regional scale they seem closer, and our automated method reproduces similar trends to published studies in our study area. However, it does represent a methodological advancement, particularly for recent centuries, that should be considered for future paleoecological studies.

The north-central Canadian boreal forest experienced increased occurrence of large and severe wildfires caused by unusually warm temperatures and drought events during the last decade. It is, however, difficult to assess the exceptional nature of this recent wildfire activity, as few long-term records are available in the area. We analyzed macroscopic sedimentary charcoal from four lakes and pollen grains from one of those lakes to reconstruct long-term fire regimes and vegetation histories in the boreal forest of central Northwest Territories. We used regional estimates of past temperature and hydrological changes to identify the climatic drivers of fire activity over the past 10,000 years. Fires were larger and more severe during warm periods (before ca. 5000 cal yrs. BP and during the last 500 years) and when the forest landscape was characterized by high fuel abundance, especially fire-prone spruce. In contrast, colder conditions combined with landscape opening (i.e., lower fuel abundance) during the Neoglacial (after ca. 5000 cal yrs. BP) were related with a decline in fire size and severity. Fire size and severity increased during the last five centuries, but remained within the Holocene range of variability. According to climatic projections, fire size and severity will likely continue to increase in central Northwest Territories in response to warmer conditions, but precipitation variability, combined with increased abundance of deciduous species or opening of the landscape, could limit fire risk in the future.

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.

Climate changes are expected to progressively increase extreme wildfire frequency in forests. Finding past analogs for periods of extreme biomass burning would provide valuable insights regarding what the effects of warming might be for tree species distribution, ecosystem integrity, atmospheric greenhouse gas balance, and human safety. Here, we used a network of 42 lake-sediment charcoal records across a ~2000 km transect in eastern boreal North America to infer widespread periods of wildfire activity in association with past climate conditions. The reconstructed fluctuations in biomass burning are broadly consistent with variations in ethane concentration in Greenland polar ice cores. Biomass burning fluctuations also significantly co-varied with Greenland temperatures estimated from ice cores, at least for the past 6000 years. Our retrospective analysis of past fire activity allowed us to identify two fire periods centered around 4800 and 1100 BP, coinciding with large-scale warming in northern latitudes and having respectively affected an estimated ~71% and ~57% of the study area. These two periods co-occurred with widespread decreases in mean fire-return intervals. The two periods are likely the best analogs for what could be anticipated in terms of impacts of fire on ecosystem services provided by these forests in coming decades.

Mixed?wood boreal forests are characterized by a heterogeneous landscape dominated by coniferous or deciduous species depending on stand moisture and fire activity. Our study highlights the long?term drivers of these differences between landscapes across mixed?wood boreal forests to improve simulated vegetation dynamics under predicted climate changes. We investigate the effects of main climate trends and wildfire activities on the vegetation dynamics of two areas characterized by different stand moisture regimes during the last 9000 years. We performed paleofire and pollen analyses in the mixed?wood boreal forest of north?western Ontario, derived from lacustrine sediment deposits, to reconstruct historical vegetation dynamics, which encompassed both the Holocene climatic optimum (ca. 8000–4000 a bp) and the Neoglacial period (ca. 4000 a bp). The past warm and dry period (Holocene climatic optimum) promoted higher fire activity that resulted in an increase in coniferous species abundance in the xeric area. The predicted warmer climate and an increase in drought events should lead to a coniferization of the xeric areas affected by high fire activity while the mesic areas may retain a higher broadleaf abundance, as these areas are not prone to an increase in fire activity.

We present a comparative analysis of fire reconstructions from tree rings and from wood charcoal preserved in forest soils, peat and lake sediments. Our objective is to highlight the benefits and limits of different archives and proxies to reconstruct fire histories. We propose guidelines to optimize proxy and archive choice in terms of spatial and temporal scales of interest. Comparisons were performed for two sites in the boreal forest of northeastern North America. Compared to others archives, tree-ring analysis remains the best choice to reconstruct recent fires (<1000 years). For longer periods (from several centuries to millennia), lake charcoal can be used to reconstruct regional or local fire histories depending on the method used, but the focus should be on historical trends rather than on the identification of individual fire events. Charcoal preserved in peat and soils can be used to identify individual fire, but sometimes cover shorter time periods than lake archives.

Les changements climatiques en cours semblent être à l’origine d’une intensification de l’activité de feux dans les forêts circumboréales. Des modèles prédictifs basés sur des données historiques sont couramment utilisés pour essayer d’anticiper les régimes des feux et leurs impacts pour l’horizon 2100. Cependant, les concepts alimentant ces modèles reposent sur l’interprétation de données issues de seulement quelques régions alors que de nombreuses études ont montré que des facteurs régionaux à locaux tels que la topographie, la nature des sols, la composition et la structure de la végétation, ainsi qu’un profil climatique et une météorologie particulière, peuvent impacter l’activité de feux. Au nord-est du Canada, la région de l’est du Québec et du Labrador se caractérise, comparativement aux régions de l’ouest et du centre du Québec, par un relief plus vallonné, une végétation différente (plus riche en sapin baumier (Abies balsamea) et plus pauvre en pin gris (Pinus banksiana)) et un climat plus froid et plus humide. L’étude des processus liés aux dynamiques de feux et de végétation dans cette région, et leur comparaison avec ceux des régions de l’ouest et du centre au cours de l’Holocène ont donc été réalisées dans cette thèse. L’objectif global était de savoir si l’on peut se contenter de travailler à une échelle supra-régionale plutôt qu’à une échelle régionale ou locale pour prédire les conséquences des changements climatiques en cours sur la dynamique des forêts conifériennes du Québec-Labrador. Les résultats mettent en évidence l’impact de la taille des feux, jusqu’alors sous-estimé, sur la dynamique à long terme de la végétation au sein de chacune des régions. La présence de grands feux a favorisé la propagation du pin gris depuis 7000 ans à l’ouest du Québec et autour de 2000 ans avant aujourd’hui à l’est du Québec-Labrador. Par opposition, la rareté des événements de grands feux à l’est jusqu’à 2000 ans avant aujourd’hui a engendré une densification de sapin baumier et de bouleau dans le paysage. De plus, les conditions pré-requises à l’éclosion des grands feux diffèrent entre la région de l’est et celles de l’ouest et du centre. Dans les régions continentales de l’ouest et du centre, ces événements ont été déclenchés par une augmentation des températures printanières et estivales, et une saison de feux plus longue. Dans la région de l’est, les grands feux ont été probablement causés par une forte variabilité hydrologique (précipitations/évapotranspiration). Cette divergence s’explique en grande partie par l’influence prédominante de la topographie régionale au détriment de l’impact des grandes tendances climatiques sur l’activité de feux dans l’est. xxi Au regard de nos résultats, les scénarios climatiques annoncés risquent d’augmenter l’occurrence des grands feux dans les régions de l’ouest et du centre du Québec sans qu’il y ait pour autant de conséquence significative sur la composition du couvert forestier. À l’est, les projections restent plus incertaines car les causes à l’origine des grands feux passés dans cette région n’ont pas été totalement élucidées. Cependant, toutes les hypothèses de trajectoires de végétation futures formulées dans cette thèse vont dans le sens d’un maintien du sapin baumier dans le paysage. Compte-tenu de la diversité des interactions susceptibles d’impacter significativement la taille des feux à l’échelle régionale, notre compréhension des processus liés aux dynamiques de perturbation et de végétation semble encore insuffisante pour pouvoir les prédire à large échelle. Il serait donc raisonnable, dans un premier temps, d’étudier plus finement ces processus à l’échelle de zones les plus homogènes possibles en termes de composition végétale, de topographie et de climat. C’est dans cette optique que la méhode de détection des feux locaux passés à partir des charbons lacustres présentée dans cette thèse a été développée. Elle vise à améliorer la différenciation des feux ayant eu lieu dans le bassin versant du lac étudié (local) de ceux s’étant produits à une plus grande distance (régional), et ce, à l’échelle plurimillénaire. Elle se base sur l’hypothèse qu’un feu local a engendré une plus forte séquestration de gros charbons dans le lac qu’un feu régional. Nos résultats montrent que, sur la base de cette hypothèse, notre méthode permet de détecter plus efficacement les feux locaux passés que la méthode actuellement disponible. Conjuguée à l’étude d’autres bio-indicateurs permettant de reconstruire l’environnement passé à l’échelle locale, nous devrions être capables de mieux comprendre les causes et conséquences des variations de taille des feux au regard des différentes combinaisons observées de facteurs environnementaux dans l’avenir.

Global warming could increase climatic instability and large wildfire activity in circumboreal regions, potentially impairing both ecosystem functioning and human health. However, links between large wildfire events and climatic and/or meteorological conditions are still poorly understood, partly because few studies have covered a wide range of past climate-fire interactions. We compared palaeofire and simulated climatic data over the last 7000 years to assess causes of large wildfire events in three coniferous boreal forest regions in north-eastern Canada. These regions span an east-west cline, from a hilly region influenced by the Atlantic Ocean currently dominated by Picea mariana and Abies balsamea to a flatter continental region dominated by Picea mariana and Pinus banksiana. The largest wildfires occurred across the entire study zone between 3000 and 1000 cal. BP. In western and central continental regions these events were triggered by increases in both the fire-season length and summer/spring temperatures, while in the eastern region close to the ocean they were likely responses to hydrological (precipitation/evapotranspiration) variability. The impact of climatic drivers on fire size varied spatially across the study zone, confirming that regional climate dynamics could modulate effects of global climate change on wildfire regimes.