Entity

Time filter

Source Type


Mansuy N.,University of Quebec at Montreal | Gauthier S.,Natural Resources Canada | Robitaille A.,Ministere des Ressources Naturelles et de la Faune du Quebec | Bergeron Y.,University of Quebec
Canadian Journal of Forest Research | Year: 2012

In many northern forest ecosystems, the postfire transition from a closed-crown forest to open woodland is often observed but poorly understood. This paper looks at the effect of interactions between surficial deposit, climate, and fire cycle on postfire forest recovery within a large territory (190 000 km 2) of the boreal forest of eastern Canada. Postfire recovery was estimated using the time elapsed to move from the burnt stage to the regenerated stage and the young forest stage. The main objective was to determine if forests situated in dry regions (characterized by a high proportion of dry coarse surficial deposits, low precipitation, and short fire cycle) tend to reestablish more slowly after fire, obtaining a more open stand compared with wetter regions characterized by a longer fire cycle. To identify the best explanatory model for postfire recovery, multinomial logistic regressions with the Akaike information criterion were conducted using a combination of physicoclimatic factors. Our best model suggests that the most significant predictors of postfire recovery are time since fire (χ 2 = 1370.06), surficial deposit type (χ 2 = 651.95), the Canadian Drought Code (x 2 = 247.75), and the growing season precipitation (χ 2 = 102.80). Fast recovery and dense forest regeneration are associated with subhydric till deposits only in the regions characterized by a long fire cycle (>500 years). Conversely, slow regeneration conducive to a sparse young forest was usually associated with regions characterized by a short fire cycle (<200 years) underlain by dry coarse deposits such as juxtaglacial but also mesic deposits in some cases. Our results also show that slow recovery and reduced forest regeneration are most likely to occur following fires that occurred in dry years, regardless of the deposit type and region. Source


Lesmerises F.,University of Quebec at Rimouski | Dussault C.,Ministere des Ressources Naturelles et de la Faune du Quebec | St-Laurent M.-H.,University of Quebec at Rimouski
Forest Ecology and Management | Year: 2012

Throughout the southern part of the boreal forest, timber harvesting has generated a young forest matrix interspersed with mature remnants and fragmented by numerous roads. These changes have modified the abundance and diversity of many animal species and destabilized some trophic networks. Because wolves (Canis lupus) are apex predators of the boreal food web, wolf response to cumulative disturbances could have important impacts on the entire ecosystem. Our objective was to assess the impacts of anthropogenic disturbance on wolf habitat selection in a highly disturbed landscape. Between 2005 and 2010, we tracked 22 wolves with GPS collars in nine packs inhabiting the southern fringe of Québec's boreal forest. Using resource selection functions, we assessed the synergistic impacts of anthropogenic disturbances and habitat quality on habitat selection. Wolves selected areas providing food or likely to improve hunting success, but avoided anthropogenic disturbances, especially in regions with high levels of human activity. Interestingly, wolves seemed more tolerant of infrastructure when frequenting high-quality habitats. We demonstrate how anthropogenic disturbances may influence wolf habitat selection. Wildlife managers should take into account predator responses to logging-related disturbances when planning forest management for potential prey species. © 2012 Elsevier B.V. Source


Mansuy N.,Natural Resources Canada | Boulanger Y.,Natural Resources Canada | Terrier A.,University of Quebec at Montreal | Gauthier S.,Natural Resources Canada | And 2 more authors.
Landscape Ecology | Year: 2014

The characterization of the fire regime in the boreal forest rarely considers spatial attributes other than fire size. This study investigates the spatial attributes of fires using the physiography of the landscape as a spatial constraint at a regional scale. Using the Canadian National Fire Database, the size, shape, orientation and eccentricity were assessed for 1,136 fires between 1970 and 2010 in Quebec's boreal forest and were summarized by ecodistrict. These spatial metrics were used to cluster 33 ecodistricts into homogeneous fire zones and then to determine which environmental variables (climate, topography, hydrography, and surficial deposits) influence the spatial attributes of fires. Analyses showed that 28 out of 33 ecodistricts belonging to a given fire zone were spatially contiguous, suggesting that factors driving the spatial attributes of fire are acting at a regional scale. Indeed, the orientation and size of fires vary significantly among the zones and are driven by the spatial orientation of the landscape and the seasonal regional climate. In some zones, prevailing winds during periods conducive to fire events parallel to the orientation of the landscape may favour the occurrence of very large fires (>100,000 ha). Conversely, an orientation of the landscape opposite to the prevailing winds may act as a natural firebreak and limit the fire size and orientation. This study highlights the need to consider the synergistic relationship between the landscape spatial patterns and the climate regime over the spatial attributes of fire at supra-regional scale. Further scale-dependant studies are needed to improve our understanding of the spatial factors controlling the spatial attributes of fire. © 2014 Her Majesty the Queen in Right of Canada. Source


Mansuy N.,University of Quebec at Montreal | Gauthier S.,Natural Resources Canada | Robitaille A.,Ministere des Ressources Naturelles et de la Faune du Quebec | Bergeron Y.,University of Quebec
International Journal of Wildland Fire | Year: 2010

Spatial variations in the fire cycle of a large territory (190000km 2) located in the boreal forest of eastern Canada were assessed using random sampling points. Our main objective was to determine if regions characterised by a large proportion of dry surficial depositdrainage (SDD) burn more frequently than regions with a smaller proportion. Through a regionalisation of the landscape units, we analysed the effects of SDD on spatial variations of the fire cycle. A discriminant analysis involving the SDD and other physical variables (precipitation, temperature, aridity index, water bodies, elevation and slope) made it possible to identify a combination of variables characterising each region. A considerable variation in fire cycle was observed among the different SDD types (from 144 to 425 years) and between regions (from 90 to 715 years). Through the discriminant analysis, this study suggests that a combination of possible climatic top-down (precipitation R 2=0.727, aridity index R 2=0.663 and temperature R 2=0.574) and bottom-up factors (xeric undifferentiated till R 2=0.819 and humid undifferentiated till R 2=0.691) could explain this variation at the regional scale. Implications of those results for forest protection against fire and regional development are briefly discussed. © IAWF 2010. Source


Lesmerises R.,University of Quebec at Rimouski | Ouellet J.-P.,University of Quebec at Rimouski | Dussault C.,Ministere des Ressources Naturelles et de la Faune du Quebec | St Laurent M.-H.,University of Quebec at Rimouski
Ecology and Evolution | Year: 2013

For conservation purposes, it is important to design studies that explicitly quantify responses of focal species to different land management scenarios. Here, we propose an approach that combines the influence of landscape matrices with the intrinsic attributes of remaining habitat patches on the space use behavior of woodland caribou (Rangifer tarandus caribou), a threatened subspecies of Rangifer. We sought to link characteristics of forest remnants and their surrounding environment to caribou use (i.e., occurrence and intensity). We tracked 51 females using GPS telemetry north of the Saguenay River (Québec, Canada) between 2004 and 2010 and documented their use of mature forest remnants ranging between 30 and ~170 000 ha in a highly managed landscape. Habitat proportion and anthropogenic feature density within incremental buffer zones (from 100 to 7500 m), together with intrinsic residual forest patch characteristics, were linked to caribou GPS location occurrence and density to establish the range of influence of the surrounding matrix. We found that patch size and composition influence caribou occurrence and intensity of use within a patch. Patch size had to reach approximately 270 km2 to attain 75% probability of use by caribou. We found that small patches (<100 km2) induced concentration of caribou activities that were shown to make them more vulnerable to predation and to act as ecological traps. Woodland caribou clearly need large residual forest patches, embedded in a relatively undisturbed matrix, to achieve low densities as an antipredator strategy. Our patch-based methodological approach, using GPS telemetry data, offers a new perspective of space use behavior of wide-ranging species inhabiting fragmented landscapes and allows us to highlight the impacts of large scale management. Furthermore, our study provides insights that might have important implications for effective caribou conservation and forest management. We propose an approach that combines the influence of landscape matrices with the intrinsic attributes of remaining habitat patches on the space use behaviour of woodland caribou using GPS telemetry surveys conducted on 51 females. We found that residual forest patch size and composition influence caribou occurrence and intensity of use within a patch, and that both the presence of anthropogenic disturbances and undisturbed areas in the surrounding environment influenced caribou use of residual forest patches. Woodland caribou clearly need large residual forest patches, embedded in a relatively undisturbed matrix, to achieve low densities as an antipredator strategy (spacing out behaviour). © 2013 The Authors. Source

Discover hidden collaborations