A combination of wildfires and defoliating insect outbreaks play an important role in the natural successional dynamics of North American boreal forests, which, in the long term, change the post-disturbance composition and structure of forest stands. After stand-replacing disturbances (mainly wildfires), early successional hardwoods typically dominate the affected areas in boreal forests. Provided sufficient time following disturbances, the increasing recruitment of mid- to late-successional softwoods as well as the mortality of hardwoods gradually change forest composition from hardwoods to admixtures of hardwood-conifer species and conifer-dominated stands in mid and late successional stages, respectively. Such mixed woods are abundant across the southern Canadian boreal forest. In boreal Canada, mixed woods are the most structurally heterogeneous forest ecosystems, are highly productive, and form an important source of timber supply. Here we present the EASTERN BOREAL MIXEDWOODS CANADA dataset, which documents the changes in composition and structure of stands originating from eight successive wildfires representing a chronosequence of 249 years in eastern Canada. This dataset has been used in several different projects to study and model the influence of natural (e.g., insect outbreaks) and anthropogenic disturbances (e.g., harvesting) on the dynamics of post-fire stands. The data covers a high range of variability in stand composition and structure, explained by species establishment, dominance and mixture. It thus constitutes a useful source of information to trace the dynamics of the main boreal tree species of eastern north America, from their establishment to their replacement at different spatial scales (e.g., from stand to landscape level).

Note: Detailed methods will be described in forthcoming data paper.

The dataset integrates data from inventories conducted during 1990-2019 period throughout the Lake Duparquet Research and Teaching Forest (LDRTF), located in the balsam fir - white birch bioclimatic domain of Québec (Canada), where a dozen of large wildfires occurred since 1760 (Bergeron 1991). Within the LDRTF site, the EASTERN BOREAL MIXEDWOODS CANADA data have been collected in three main subsets that cover different study objectives that contribute to a better understanding of post-fire stand successional dynamics:

1.  LDRTF-TRANSECTS: These transects represent a network of 596 quadrats that covers the stands originated from eight wildfires that occurred from 1760 to 1964 (1760, 1797, 1823, 1847, 1870, 1916, 1944 and 1964). The original idea of establishing the quadrats in 1990-1991 was to study the compositional and structural variability of post-fire stands and changes over time, first with a chronosequence analysis (Bergeron and Dubuc 1989) and then by a stand-reconstruction approach (Bergeron 2000). The other objective of establishing the plots were to study the interaction between post fire stand composition and secondary disturbances (Kneeshaw and Bergeron 1998). The quadrats were re-sampled in 2009 and were used in several studies, for example for parametrizing the SORTIE-ND forest simulator to simulate the natural successional dynamics of post-fire stands in LDRTF over long periods, both in presence and in absence of natural and anthropogenic disturbances (Poulin et al. 2007, Poulin and Messier 2008, Leduc and Coates 2013, Bose et al. 2015, Maleki et al. 2020).
2. LDRTF-SEEDLING_PLOTS: In each post-fire stand (1760, 1797, 1823, 1847, 1870, 1916, 1944 and 1964) a 400 m² plot was stablished in 1997 to observe annual seedling recruitment and mortality in 16 one-m² subplots located near the plot borders.
3. LDRTF-HECTARES: In six of the post-fire stands (1760, 1823, 1847, 1870, 1916 and 1944) a one-hectare permanent plot was first established in 1991 and re-measured in 4-6 years intervals. One of the goals to establish this data was to study the spatial patterns, age structure and growth of species in each stand (Park et al. 2005).

The README.txt file contains the description of all tables and variables in the data.