Do harvest retention patches in the boreal forest emulate those resulting from wildfire? A comparison of understory vegetation a decade after disturbance
Data files
Jun 27, 2025 version files 149.81 KB
-
Linear_Models.csv
22.39 KB
-
ordination.csv
60.77 KB
-
README.md
7.53 KB
-
Species_Codes.csv
3.06 KB
-
Trait_Data.csv
3.60 KB
-
Trait_Site_Data.csv
9.06 KB
-
Trait_Species_Data.csv
43.39 KB
Abstract
To sustain a breadth of ecosystem services, Ecosystem-based forest management (EBM) aims to reduce differences between managed and natural forests. Based on observed structural complexity of forests following natural disturbance, retention of forest structure at harvest is being implemented globally. Despite a decade of including retention patches in managed forests, it remains unclear if patches in forests disturbed by fire and harvest exhibit similar structural characteristics and biodiversity as intact forest. Such knowledge is critical within an adaptive management framework.
We present the first study comparing understory vegetation of post-fire and post-harvest remnants in the boreal. A decade following disturbance, we examine forest structure and plant diversity in key locations of harvests and burns: the disturbed matrix, island remnants within the disturbance, adjacent undisturbed forest, and edges of these. We utilize a trait-based framework for plant diversity to test hypotheses about how traits vary in relation to disturbance type, environmental variables, and with variation in structure.
Edge effects were minor and didn’t differ between burns and harvests. Both harvest and wildfire remnants maintained understory communities and forest structure similar to adjacent forest, but remnant and reference areas exhibited some compositional differences between fire and harvest sites. Analysis of plant functional traits revealed similar patterns across burns and harvests, with colonization traits associated with disturbed areas, while persistence traits were associated with island remnants.
Synthesis and applications: The similarity of the understory community and forest structure of island remnants to adjacent reference forest in burned and harvested areas suggests remnants are effective reservoirs of undisturbed forest. This provides evidence that implementation of aggregated retention at harvest helps achieve a key objective of retention, i.e. maintaining structural heterogeneity to support biodiversity.
Dataset DOI: 10.5061/dryad.12jm63z9f
Description of the data and file structure
This datset contains the data needed to replicate the analyses in Schneider et al. (in review), examining the vascular understory plant composition and forest structure of post-harvest and post-fire island remnants in the Boreal forest of Alberta. Data collected includes plant cover data, leaf trait data, and tree measurements and also includes trait data sourced from the TOPIC and TRY databases.
Files and variables
File: Species_Codes.csv
Description: A list of species codes (column A) and their associated Latin Binomial (column B)
File: Trait_Site_Data.csv
Description: Dataset with explanatory forest structure variables to be used in the fourth corner analysis
Variables
- BADR: Ratio of living tree basal area to basal area of recently dead trees. Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- CONBAAR: Ratio of living coniferous tree basal areal to living deciduous tree basal area. Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- canopy: Canopy cover (%). Empty cells indicate the measurement was not collected in the plot type.
- BAA: Basal area of living trees (m^2/ha). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- olayer: Thickness of the soil organic layer (cm)
- litter: Ground cover of leaf litter (%)
- saps: Number of saplings (stems/ha). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
File: Trait_Data.csv
Description: Dataset with trait variables for vascular understory species to be used in the fourth corner analysis
Variables
- raunkier: Raunkiaer life form. Categorized as: ch, chamaephyte; g, geophyte; h, hemicryptophyte; mc, micro-phanerophyte; mg, mega-phanerophyte; t, therophytes
- lateral extension: Categorized by distance of lateral spread: 0, absent; 1, limited; 2, extensive (non-phanerophyte); 3, phanerophyte intermediate; 4, phanerophyte extensive
- resprouting: Resprouting capability present: 1; absent 0
- seed production: 1, few; 2, some; 3, abundant
- height: Maximum plant height (cm)
- seed weight: Weight of 1000 seeds in grams
- SLA: Specific leaf area with petioles, cm2 g-1
- reproduction: s, reproduces only by seed; svp, reproduces primarily by seed, some vegetative reproduction; vp, reproduces primarily vegetatively
- foliage: de, decumbent or prostrate stem; e, erect leaves; el, erect leafy stem; fa, arched stem; g, graminoid leaves; me, mono stem erect phanerophyte; ms, mono stem spread out; mu, multi-stem; r, rosette; sr, semi-rosette; um, umbel-shaped stem; vi, vine
File: Trait_Species_Data.csv
Description: Dataset with vascular understory species covers to be used in the fourth corner analysis
Variables
- location: Unique code for each plot. First three letter represent region, net two numbers represent site within region, next two letters represent plot location
- All other variables are cover of species (%). Species codes are detailed in the species code spreadsheet
File: ordination.csv
Description: Data required to replicate PerMANOVA and NMDS analyses.
Variables
- location: Unique code for each plot. First three letter represent region, net two numbers represent site within region, next two letters represent plot location
- region: Six regions, three harvest, three fire
- color: Six level factor, a combination of disturbance type and location (disturbed, island, reference)
- treat: Twelve level factor, a combination of disturbance type (fire/harvest) and plot location (DE,DI,IE,II,RE,RI)
- f1: Two level factor representing disturbance type (fire/harvest)
- f2: Three level factor representing location (disturbed, island, reference)
- f3: Two level factor representing edge or interior plots
File: Linear_Models.csv
Description: Data Necessary to replicate any mixed effect linear models
Variables
- location: Unique code for each plot. First three letter represent region, net two numbers represent site within region, next two letters represent plot location
- site: An identifier for site, a grouping of six plots located within an island remnant (x2), the adjacent disturbance (x2), an unharvested reference site (x2)
- disturbance: Two level factor representing disturbance type (fire/harvest)
- region: Six regions, three harvest, three fire
- plot: Six level factor, plot location within the site (DE,DI,IE,II,RE,RI)
- cov: Total cover of vascular understory plants (%)
- rich: Number of unique vascular understory plant species in the plot
- shann: Hill diversity, derived from Shannon’s entropy
- grass: Cover of graminoids in the understory (%)
- shrub: Cover of shrubs in the understory (%)
- forb: Cover of forbs in the understory (%)
- olayer: Thickness of the soil organic layer (cm)
- litter: Ground cover of leaf litter (%)
- saps: Number of saplings (stems/ha)
- RCV: Volume of recently dead downed trees (m^3/ha).
- RTV: Volume of recently dead downed trees and recently dead standing trees (m^3/ha). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- canopy: Canopy cover (%). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- RSV: Volume of recently dead standing trees (m^3/ha). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- BAA: Basal area of living trees (m^2/ha). Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
- RBADR: Ratio of living tree basal area to basal area of recently dead trees. Empty cells indicate the measurement was either not collected or not applicable for analysis in this plot type.
Access information
Trait data was partially acquired from the from the following sources:
-
TOPIC database:
Aubin, I., Messier, C., Gachet, S., Lawrence, K., McKenney, D., Arseneault, A., Bell, W., De Grandpré, L., Shipley, B., Ricard, J.-P. & Munson, A.D. 2012. TOPIC - Traits of Plants in Canada. [Internet] Natural Resources Canada, Canadian Forest Service, Sault Ste. Marie, Ontario, Canada [Cited date] Available from: https://natural-resources.canada.ca/science-data/science-research/research-centres/topic-network-traits-plants-canada-database-topic-canadian-repository-invertebrate-traits-trait-ecological-records-critter.
-
TRY database:
Kattage, J., Boenisch, G., Diaz, S., Lavorel, S., Prentice, I. C., Leadley, P., Tautenhahn, S., Werner, G., & et al. (2020). TRY plant trait database – enhanced coverage and open access. Global Change Biology, 26(1), 119–188.