The variability and determinants of serotiny in Pinus yunnanensis
Data files
Jun 23, 2026 version files 421.42 KB
-
correlation_heatmap.opju
353.24 KB
-
Level_differences.opju
35.62 KB
-
Multiple_linear_regression.R
4.37 KB
-
piecewiseSEM.csv
856 B
-
piecewiseSEM.R
1.91 KB
-
README.md
10.13 KB
-
serotiny_level.csv
1.07 KB
-
trait.xlsx
14.23 KB
Abstract
Serotiny, a key fire-adaptive trait, enables certain plants to retain seeds in closed cones until triggered by fire, thus facilitating post-fire regeneration. Although serotiny has been observed in many species, the physiological and environmental mechanisms that regulate this trait remain poorly understood. In this study, we investigated the variation in serotiny level within Pinus yunnanensis, a pine species native to fire-prone landscapes in southwestern China. Serotiny level varied significantly among populations and was significantly associated with environmental factors and cone physiological characteristics, including Bio2 (mean diurnal temperature range), Bio5 (maximum temperature of the warmest month), soil pH, total soil phosphorus, and cone resin content. Cone resin content exhibited the strongest direct positive effect on serotiny. Soil pH and phosphorus concentration negatively affected resin content in cones, thus indirectly reducing serotiny level. Bio2 enhanced serotiny indirectly by acidifying soil and promoting resin synthesis, while Bio5 decreased serotiny indirectly by increasing phosphorus availability and limiting resin accumulation. Fire activity and fire radiative power showed no significant influence on serotiny level in P. yunnanensis. These findings enhance our understanding of how serotiny evolves under the combined pressures of climate and soil conditions in fire-adapted ecosystems.
Description of the data and file structure
This dataset contains the data and analysis scripts used in the associated article, “The variability and determinants of serotiny in Pinus yunnanensis”. The dataset includes serotiny measurements, cone chemical traits, environmental variables, statistical analysis scripts, and Origin project files used to generate figures.
Variation in serotiny level among Pinus yunnanensis populations was analyzed using one-way ANOVA. Multiple linear regression was used to examine the relationships between serotiny level and environmental predictors, including climate, soil properties, fire-related variables, and cone chemical traits. Piecewise structural equation modeling was used to evaluate direct and indirect pathways linking environmental factors, cone traits, and serotiny level.
Climatic variables were extracted from WorldClim version 2.1 bioclimatic raster layers based on the latitude and longitude of each sampling plot. The original WorldClim raster files, clipped raster layers, and GeoTIFF files are not included in this Dryad dataset. Only the extracted climatic values used in the analyses are provided.
Fire-related variables were derived from NASA FIRMS data and/or associated fire-related products according to the sampling plot locations. The original external raster or remote-sensing source files are not included in this Dryad dataset. Only the extracted or calculated values used in the analyses are provided.
Files and variables
File: serotiny_level.csv
Description: This CSV file contains cone serotiny level measurements for 18 sampling plots of Pinus yunnanensis. Each row represents one sampling plot.
Variables:
Variant: variety of Pinus yunnanensis.Plot: sampling plot identifier.Longitude: longitude of the sampling plot, in decimal degrees.Latitude: latitude of the sampling plot, in decimal degrees.Serotiny_level_percent: cone serotiny level, expressed as a percentage.
File: trait.xlsx
Description: This Excel file contains the trait and environmental variables used to analyze the variability and determinants of serotiny in Pinus yunnanensis. The file includes cone serotiny level, tree height, cone chemical properties, soil properties, fire-related variables, aridity index, altitude, and bioclimatic variables.
Variables:
- serotiny_level_percent (%): cone serotiny level, expressed as a percentage.
Tree_Height (m): tree height, in meters.Altitude (m): altitude of the sampling site, in meters.Resin content (%): cone resin content, expressed as a percentage.Calorific value (J/g): cone calorific value, in joules per gram.Ash content (%): cone ash content, expressed as a percentage.Cellulose content (%): cone cellulose content, expressed as a percentage.Soil pH: soil pH.Total phosphorus (g/kg): soil total phosphorus content, in grams per kilogram.Total organic carbon (g/kg): soil total organic carbon content, in grams per kilogram.Total potassium (g/kg): soil total potassium content, in grams per kilogram.Total nitrogen (g/kg): soil total nitrogen content, in grams per kilogram.Aridity index: aridity index of the sampling site. Higher or lower values indicate relative aridity according to the index used in the analysis.Wildfire activity: wildfire activity around the sampling site, derived from fire-related data.Fire radiative power (MW): fire radiative power around the sampling site, in megawatts.Bio1: annual mean temperature, in °C.Bio2: mean diurnal temperature range, in °C.Bio3: isothermality, calculated as Bio2/Bio7 × 100.Bio4: temperature seasonality, calculated as the standard deviation × 100.Bio5: maximum temperature of the warmest month, in °C.Bio6: minimum temperature of the coldest month, in °C.Bio7: temperature annual range, calculated as Bio5 - Bio6, in °C.Bio8: mean temperature of the wettest quarter, in °C.Bio9: mean temperature of the driest quarter, in °C.Bio10: mean temperature of the warmest quarter, in °C.Bio11: mean temperature of the coldest quarter, in °C.Bio12: annual precipitation, in mm.Bio13: precipitation of the wettest month, in mm.Bio14: precipitation of the driest month, in mm.Bio15: precipitation seasonality, expressed as the coefficient of variation.Bio16: precipitation of the wettest quarter, in mm.Bio17: precipitation of the driest quarter, in mm.Bio18: precipitation of the warmest quarter, in mm.Bio19: precipitation of the coldest quarter, in mm.
File: piecewiseSEM.csv
Description: This CSV file contains the subset of variables used for piecewise structural equation modeling. Each row represents one sampling plot. The file includes cone serotiny level, altitude, cone chemical traits, soil variables, and selected bioclimatic variables used to evaluate direct and indirect pathways affecting serotiny level in Pinus yunnanensis.
Variables:
serotiny_level_percent: cone serotiny level, expressed as a percentage.altitude_m: altitude of the sampling plot, in meters.resin_content_percent: cone resin content, expressed as a percentage (%).soil_pH: soil pH.total_phosphorus_g_kg: soil total phosphorus content, in grams per kilogram.ash_content_percent: cone ash content, expressed as a percentage (%).Bio2: mean diurnal temperature range, in °C.Bio5: maximum temperature of the warmest month, in °C.
This file is used as the input data file for piecewiseSEM.R.
File: Multiple_linear_regression.R
Description: This R script performs multiple linear regression analyses to examine the relationships between cone serotiny level and independent variables, including climate, soil properties, fire-related variables, and cone chemical traits.
File: piecewiseSEM.R
Description: This R script performs piecewise structural equation modeling using R. The analysis evaluates hypothesized direct and indirect relationships among serotiny level, environmental variables, and cone chemical traits.
File: correlation_heatmap.opju
Description: This Origin project file contains the correlation heatmap used to visualize relationships between serotiny level and predictor variables in the study.
File: Level_differences.opju
Description: This Origin project file contains statistical and graphical comparisons of serotiny levels across three varieties of Pinus yunnanensis.
Variables shown in this Origin project correspond to the variables defined in serotiny_level.csv.
Code, software, and workflow
The analyses were conducted using R version 4.4.1. Origin software was used to prepare the .opju project files and associated figures.
R packages
The R scripts require the following R packages. Users should install these packages before running the scripts:
readxlMuMInggplot2rdacca.hppatchworkpiecewiseSEMnlme
Please note that the package list should match the library() calls in the uploaded R scripts.
Working directory
Before running the R scripts, users should place the R scripts and input data files in the same folder and set this folder as the working directory in R. Alternatively, users can modify the input file paths in the scripts according to their local file structure.
Workflow
Multiple_linear_regression.R
This script performs multiple linear regression analyses to examine the relationships between cone serotiny level and predictor variables, including climatic variables, soil properties, fire-related variables, and cone chemical traits. The input file is trait.xlsx.
piecewiseSEM.R
This script performs piecewise structural equation modeling to evaluate direct and indirect relationships among serotiny level, environmental variables, soil variables, and cone chemical traits. The input file is piecewiseSEM.csv.
correlation_heatmap.opju
This Origin project file contains the correlation heatmap used to visualize relationships between serotiny level and predictor variables in the study. The variables shown in this file correspond to the variables defined in trait.xlsx.
Level_differences.opju
This Origin project file contains statistical and graphical comparisons of serotiny levels across three varieties of Pinus yunnanensis. The variables shown in this file correspond to the variables defined in serotiny_level.csv.Access information and data sources
The dataset was derived from the following sources:
- Original field and laboratory measurements from this study, including cone serotiny level, cone resin content, cone ash content, and other measured variables.
- WorldClim version 2.1 bioclimatic variables. Climatic variables were extracted from WorldClim raster layers based on the latitude and longitude of each sampling plot. The original WorldClim raster files are not included in this Dryad dataset.
- NASA FIRMS fire data and/or associated fire-related products. Fire-related variables were extracted or calculated for each sampling plot. The original FIRMS source files are not included in this Dryad dataset.
References for external data sources
Fick, S. E., & Hijmans, R. J. 2017. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology, 37, 4302–4315.
NASA FIRMS. Fire Information for Resource Management System. NASA Earthdata.
License compatibility statement
All files uploaded to this Dryad dataset are compatible with CC0. The uploaded files contain original data generated by this study and extracted or calculated values used in the analyses. The dataset does not include original WorldClim raster files, clipped WorldClim raster layers, WorldClim GeoTIFF files, original FIRMS source files, or other third-party files with licensing terms incompatible with CC0.
Missing values
No missing values are present in the submitted CSV or Excel data files.