Ecological factors shaping postfire resilience in mature black spruce forests of eastern North America
Data files
Feb 04, 2026 version files 73.57 KB
-
dataset.csv
68.65 KB
-
README.md
4.92 KB
Abstract
This dataset encompasses 536 entries, each representing a 400 m² stand of post-fire regeneration of black spruce (Picea mariana) in the eastern North American boreal zone, affected by 21 fires that occurred between 1995 and 2016. It focuses on the post-fire recovery and resilience of mature stand of black spruce (Picea mariana), a species adapted to stand-replacing fires. The dataset includes unique identifiers, geographical locations, fire characteristics, and metrics for both pre-fire and post-fire vegetation, particularly post-fire seedling density (seedlings/ha).
The dataset allows for exploring key ecological determinants of post-fire regeneration, including seed bank conditions, fire severity, and seedbed condition. This comprehensive network facilitates an unprecedented spatial and temporal analysis of black spruce regeneration. Legal and ethical considerations involve ensuring proper data handling.
Research utilizing this dataset can inform forest management strategies and ecological assessments, underscoring the potential shifts, principally toward loss of black-spruce density, in boreal forest dynamics under climate-induced fire regime changes.
Fortin Stelsa1 and Boucher Yan2
1Corresponding author. Université du Québec en Abitibi-Témiscamingue and Université du Québec à Chicoutimi , email : fors40@uqat.ca
2Data manager. Université du Québec à Chicoutimi, email : yboucher@uqac.ca
DATA SPATIAL COPE
The study area is located in the boreal forest of eastern North America between latitudes 49°70’N-51°90’N and longitudes 73°58’W–67°68’W. The 21 fires studied had burned between 1995 and 2016 and are located in the bioclimatic domain of black spruce-feather moss of Quebec, eastern Canada. This bioclimatic domain is characterized by dominance of BS with JP, balsam fir (Abies balsamea [L.] Mill.), paper birch, and trembling aspen as companion species, depending on site conditions and successional stage. The study area is mostly covered by glacial deposits, particularly undifferentiated till. Its climate is characterized by average annual temperatures ranging between 0.05 and -2.19°C, with average annual precipitation varying between 946 and 1121 mm. The growing season is short, generally extending between May and September. The mean fire return interval varies along a longitudinal gradient, ranging from about 220 years near Lake Mistassini in the west to approximately 640 years near the Manicouagan Reservoir in the east.
SHARING/ACCESS
This work is licensed under a CC0 1.0 Universal (CC0 1.0) Public Domain Dedication license.
DESCRIPTION OF THE DATA
The following sections present the information necessary to understand each column of the dataset.csv.
Here is a list of all columns in the dataset.csv
1. id_unique
2. stocking_bs_jp
3. id_fire
4. time_since_fire
5. latitude
6. longitude
7. seasonality_real_date
8. seasonality_adjusted
9. CBI
10. sphagnum_cover
11. ericaceous_cover
12. rom_med
13. age
14. basal_area_prefire_bs
15. prefire_bs_proportion
16. seedlings_ha_bs
17. seedlings_ha_jp
18. seedlings_ha_bl
id_unique
A unique alphanumeric identifier for each plot. It consists of the measurement year, the year of the fire,
and a plot identifier. Example: 2017199512 - Plot 12, in a fire that burned in 1995 and measured in 2017.
stocking_bs_jp
Stocking value for black spruce (bs) and jack pine (jp). Calculated based on the number of future
microplots. In each plot, we had 10 microplots. Therefore, a stocking of 0 = 0%, 1 = 10%, 2 = 20%... 9 =
90% and 10 = 100%
id_fire
Unique identifier for each fire
time_since_fire
Time (years) between the fire and the forest inventory
latitude
Latitude (in degrees)
longitude
Longitude (in degrees)
seasonality_real_date
Actual date of the fire. Determine using the Modis hotspot and the Canadian National Fire Database
(CNFDB), available at https://cwfis.cfs.nrcan.gc.ca/ha/nfdb.
seasonality_adjusted
This metric is calculated from the date of the fire, adjusted to align with the start of the fire season for
each year, where the start of the fire season is designated as Day 1
CBI
Value of the Composite Burn Index (CBI), a numerical scale that quantifies the severity of the burn. The
scale ranges from 0 to 3, with 0 being unburnt and 3 being completely charred.
sphagnum_cover
The percentage of live sphagnum moss cover in each plot was determined using the point-intercept
method, where the number of stems that contacted sphagnum moss was compared to the total number of
stems, yielding a cover percentage.
ericaceous_cover
The percentage of ericaceous cover in each plot was determined using the point-intercept method, where
the number of stems that contacted ericaceous was compared to the total number of stems, yielding a
cover percentage.
rom_med
The median thickness (mm) of residual organic matter (ROM) in each plot was calculated by measuring the
thickness in each of the 10 microplots and taking the median of these values.
age
The age (years) of the plot at the time of the fire was determined by assessing a minimum of five trees present
prior to the fire event.
basal_area_prefire_bs
The basal area (m²/ha) of all black spruce (bs) stems present before the fire was calculated, providing a
measure of the pre-fire tree density in the area
prefire_bs_proportion
The percentage of black spruce (bs) in the landscape before the fire was calculated by comparing the
number of black spruce trees to the total number of all tree species present.
seedlings_ha_bs
Number of black spruce (bs) seedlings by hectare in a plot
seedlings_ha_jp
Number of jack pine (jp) seedlings by hectare in a plot
seedlings_ha_bl
Number of broadleaves (bl) seedlings by hectare in a plot
The studied fires were selected because: (i) they were located within the boreal closed crown forest in the province of Quebec; (ii) they were composed of mature (60 years) BS-dominated pre-fire stands (>50% of the basal area occupied by BS); (iii) they had burned a minimum of three years before field surveys to allow time for post-fire seedling establishment; (iv) they had burned after 1984 (v) they had not been salvage logged; and (vi) they were accessible by road. Across all selected fires, 536 circular plots of 400 m² (r = 11.28 m) were sampled during the summers of 2017, 2018, 2019, and 2020 following a systematic sampling design, with at least three plots distributed along the lower, middle, and upper portions of each slope and spaced at least 100 meters apart. Depending on the fire size and accessibility, between 4 and 152 plots were established by fire. The percentage cover of each understory plant species in each plot was assessed using the point-intercept method. This involved two perpendicular transect lines, each measuring 22.56 m, resulting in a total length of 45.12 m across the plot. At every 0.5 m increment along these lines,1.5-metre-long metal rods were placed perpendicularly. The number of rods that were touched by eachspecies across a plot was counted and divided by the total number of rods, yielding a percent cover value. The percentage ground cover of understory plants was obtained, including living Sphagnum spp. moss and ericaceous shrubs. All living post-fire tree seedlings (height > 0.01 m) of each of the main tree species, which included BS, JP, paper birch or trembling aspen (forming 99.9% of all tree stems), were counted in 10 evenly distributed 4m2 subplots. Within-plot, stocking was also calculated as the percentage of subplots occupied by at least one seedling. The thickness (cm) of the post-fire residual organic layer was measured at the center of each of the 10 subplots.