Data from: Historical reindeer corrals in northern boreal forests reveal divergent post-disturbance reorganization by forest type
Data files
Nov 26, 2024 version files 150.85 KB
-
DataFromManninenStark.xlsx
57.47 KB
-
PlotLevelDataManninenStark.xlsx
91.56 KB
-
README.md
1.82 KB
Abstract
Current theoretical frameworks on alternative ecosystem states in boreal forests do not contain a clear prediction on how the sensitivity for state transitions depends on site productivity, which forms a major gap in understanding how disturbances impact these ecosystems. To test a hypothesis that forest type may be an important determinant for forest resilience to disturbances, we analyzed tree layer, field and ground vegetation layers as well as soil physico-chemical properties in 26 historical reindeer (Rangifer tarandus L.) corrals used in animal husbandry between late 1800’s to early 1960’s, and reference forests in northern boreal forest zone in Finland. The forest sites were assigned to two forest types that represent different levels of site productivity: mesic forests (high productivity) and sub-xeric forests (low productivity), which enabled testing whether forest recovery or potential reorganization and transition to another alternative states after historical disturbance differs by forest type. The dataset includes two different datasets: the site-level data, i.e., the mean value for each site, which was used in the statistical tests (DataFromManninenStark, the total N = 26; N = 18 for sub-xeric forests, N = 8 for mesic forests), and the plot-level data (PlotLevelDataManninenStark; N = 4 for each individual site) as well as R scripts that were used for the statistical analyses.
Field data was collected during June-August 2020.
Description of the Data and file structure
Data is structured according to the individual forest sites (N = 26), using the same name as used for these historical sites in the archaeological databases in Finland. Each site contains two treatments, historical disturbance and an adjacent reference forest (historical reindeer corral = corral, reference forest = control) in two forest types (sub-xeric N = 18, mesic N = 8). In the datafile "PlotLevelDataManninenStark", we present raw data at a plot level (N = 4 for each treatment for each site). In the datafile "DataFromManninenStark", we present means for the two treatments per site; all statistical tests were conducted using this file.
In the datafiles, we present the numbers of the forest tree species per hectare, the average height and the breast height diameter of the forest trees, and cover of plant functional types in datasheet “bor_tree_veg”. Datasheet “bor_species”, which was used in the NMDS ordinations, includes all forest understorey plant species separately using a limit of a frequency of two or higher. Datasheet “bor_env” include environmental variables used as vectors in the NMDS ordination. All variables and their units are explained in “ReadMe” sheet in the Excel files.
R codes used in general linear models, generalized linear models and NMDS analyses are provided in the data file "ManninenStark_scripts.R". The model that was used for each variable is indicated in each script.
Sharing/access Information
Links to other publicly accessible locations of the data: None.
Was data derived from another source? No.
We located 26 historical reindeer corrals used in reindeer husbandry activities between the 1800's and 1960's using several different archaeological and cultural heritage databases in Finland. We analyzed tree layer (trees per hectare, average tree height and breast height thickness), field layer (the cover of plant species) and ground layer (lichens, mosses, litter, bare ground, deadwood) in four study plots in both controls and historical reindeer corrals, and calculated a mean value for each variable for the treatments for each site. Main variables analyzed in the tree layer are the densities of the main tree species (Pinus sylvestris, Picea abies and Betula spp. consisting of B. pubescens and B. pendula), the average tree height and tree diameter at breast height. The understorey vegetation data consists of the cover percentages from main plant functional types, deciduous dwarf shrubs, evergreen dwarf shrubs and herbs (the sum of forb and graminoid species). Most important soil physico-chemical properties include soil pH, carbon, nitrogen and organic matter stocks per area, soil organic matter% and isotopic enrichment of 15N and 13C, which was regarded as an index for historical animal fertilization of the sites. The database includes the tree and understorey vegetation values both per individual study plots (the name of the datafile: PlotLevelDataManninenStark) and per study site (DataFromManninenStark).
Excel, R