Data from: Seed dormancy increases population persistence in an annual plant in an environment-dependent manner
Data files
Jul 15, 2026 version files 1.07 MB
-
Processed.Data.zip
134.54 KB
-
Raw.Data.zip
918.80 KB
-
README.md
15.18 KB
Abstract
Seed dormancy can delay germination timing to more favorable growth conditions, either within years (seasonal dormancy) or across years (between-year dormancy), not only increasing seedling survival but potentially increasing lifetime fitness. As such, seed dormancy can be a form of seasonal environmental tracking. In addition, seed dormancy can act as a bet-hedging strategy by spreading the germination risk across time, within or between years. Through both environmental tracking and bet-hedging, seed dormancy is hypothesized to stabilize population demography, potentially enhancing long-term population persistence. To test whether populations that express high seed dormancy are more likely to persist than populations with low seed dormancy, we established genetically variable, experimental field populations of Arabidopsis thaliana that differ in their level of seed dormancy. Four environmental treatments were imposed to test for demographic differences in different environments and to test whether dormancy mitigates the effects of environmental variation. Seasonal seed dormancy influenced demography and population persistence primarily via early seedling or rosette mortality. High Dormancy populations had larger seedling populations and higher population persistence over the three years in the most permissive environmental treatments. However, stressful environments—namely those which were consistently unfavorable for early life stages—diminished the demographic effects of dormancy. These dynamics, in turn, resulted in High Dormancy populations unexpectedly exhibiting more variation across environmental treatments than Low Dormancy populations. Therefore, dormancy’s enhancement of demographic performance may be limited to environments that have at least some seasons or years that are favorable for early life stages. Seasonal seed dormancy may help populations persist over time, but not under all environmental conditions. In more permissive environments, dormancy can reduce population bottlenecks and maintain larger populations. Some conditions, however, may be too adverse for seed dormancy to overcome.
Dataset DOI: 10.5061/dryad.t1g1jwtjd
Description of the data and file structure
This dataset contains all the data required to replicate analyses in Quarles-Chidyagwai and Donohue 2026 (Annals of Botany). We established genetically variable, experimental field populations that differed in their level of seed dormancy to test for the demographic effects of dormancy in different environmental treatments.
Files and variables
Common Variables:
- Files: AllSBData.csv, AllYearsPopCensus.csv, AllYearsRepTraits.csv, CensusDates_All.csv, PPFD.csv, AllYearsPopCensus_PopSize.csv, NewSdlsbyDate.csv
- Block: identifies which of the 8 blocks the population cage or seed bank pot was in.
- Col and Row: identifies the location of the population cage or seed bank pot within our field setup. (not in NewSdlsbyDate.csv)
- Date: date of population cage or seed bank pot census.
- Year: year of the population cage or seed bank pot experiment. (not in PPFD.csv)
- Files: AllSBData.csv, AllYearsPopCensus.csv, AllYearsRepTraits.csv, AllYearsPopCensus_PopSize.csv, BootLambdas.csv, LifetimeFitness_supp.csv, NewSdlsbyDate.csv, PopPersistence.csv, RoseSurvtoFRMD.csv, SdlFlucts.csv, SurvtoRose.csv, TotalRep.csv, TotalRosettes.csv, TOTALSeasonalSeedlingEstablishment.csv, TotalSeedlings.csv
- Dormancy: which of the two dormancy treatments (Dormant or Non-Dormant) the population cage or seed bank pot belonged to.
- EnvtTrt or Envt: which of the 3 environmental treatments (Control, High Humidity, Native Soil) the population cage or seed bank pot belonged to.
- Files: SeedCounts.csv, SoilTemp_frostincl_no2022.csv, PopPersistence.csv, LifetimeFitness_supp.csv, NewSdlsbyDate.csv, PopPersistence.csv, RoseSurvtoFRMD.csv, SdlFlucts.csv, SurvtoRose.csv, TotalRep.csv, TotalRosettes.csv, TOTALSeasonalSeedlingEstablishment.csv, TotalSeedlings.csv
- Pop ID or PopID: Col and Row (as described above) combined.
File: Raw.Data.zip
Description: This folder includes all data collected and used for analyses in this study. The contents are listed below. "NA" or blank cells represent missing values.
File with seed bank pot seedling counts.
- DataCollector: initials of the data collector.
- Confirmed.Total: the number of seedlings confirmed to be A. thaliana.
- Max.Total: The total number of potential A. thaliana seedlings. This includes the confirmed.total and the seedlings that could not be confirmed.
File with the population census data from the population cages. These variables are also in AllYearsPopCensus_PopSize.csv.
- RILSet: which of the 3 RIL sets the population cage belonged to.
- Ident: block, RILSet, dormancy, and Envt combined to create a unique identifier for each population cage.
- CensusN: The census number for each census as Year(1-3).Census(1-29). - Also in CensusDates_All.csv
- SeedlingN.circle: the number of seedlings inside the circle.
- RosetteN.circle: the number of rosettes inside the circle.
- Bolting.circle: the number of bolting individuals inside the circle.
- Flowering.circle: the number of flowering individuals inside the circle.
- Fruiting.circle: the number of fruiting individuals inside the circle.
- mat.dead.circle: the number of mature/dead individuals inside the circle.
- SeedlingN.OC, RosetteN.OC, Bolting.OC, Flowering.OC, Fruiting.OC, mat.dead.OC: the same life stage categories as above but for outside the circle.
- TOTAL: the total number of individuals in the population at that census (based on the previous life stage counts).
File with the individual traits measured on each individual.
- RILSet: which of the 3 RIL sets the population cage belonged to (IS = Italy x Sweden).
- IndivID: the identification number assigned to each individual.
- BandColor: the band color the individual was marked with for follow-up measurements or the location of the individual if there was no follow-up. "IC" stands for "inside the circle" and "OC" for outside the circle.
- BasalBranches: the number of basal branches.
- MainBranches: the number of primary branches on the main flowering stalk.
- Height: the height of the plant in centimeters.
- Fruits: The number of fruits on the plant.
File that converts the calendar dates of each population census in "AllYearsPopCensus" to a census number. All variables described above.
File with the allele frequency information for the Italy x Sweden dormant, non-dormant, and mixed populations.
- Batch#: The batch number assigned to each line. Mixed populations got different numbers of seeds from each batch.
- ABRC Stock #: ABRC stock number of each individual line.
- Donor id: the identification number assigned to each line.
- Columns E through MN:
- First row: the ID of the genetic marker.
- Second row: the chromosome number.
- Third row: the position of the marker on the chromosome (cM).
- Rows 4-99: genotype information for each line. "A" represents the Italian genotype and "B" represents the Swedish genotype.
- Rows 100-103: identification information for the crosses.
- Rows 104-121: calculations of the population allele frequencies.
File with the data from the lab germination assays.
- RIL.SET: The RIL set of the seeds on the plate. "Bay x Sha" is the Germany x Tajikistan RIL set and "Cal x Tac" is the UK x USA RIL set.
- Plate.ID: The identification info for each plate.
- Genotype: The genotype of the seeds on the plate (matches the donor ID from the allele frequency files).
- GermTrt: The temperature treatment the plate was put in. "Above" = 26C, "Optimal" = 18C, and "Cold" = 10C.
- Rep: the replicate number of the plate.
- Dormancy: "D" for dormant and "ND" for non-dormant.
- Total.Seeds: the total number of seeds on the plate.
- Germs.4wks: the number of seeds that germinated on the plate after 4 weeks.
File with the light sensor data.
- Treatment: The same as "Envt" or "EnvtTrt."
- Time: the time of the measurement in PM.
- Sample: the sample number for each population cage sampled.
- PPFD: the Photosynthetic Photon Flux Density in micromoles per square meter per second.
File with the seed count data.
- Date Collected in Field: the date the fruit was collected.
- Block: the same as described above.
- Dormancy: "ND" for non-dormant and "D" for Dormant.
- Trt: "Drought" for the High Humidity treatment and "Soil" for the Native Soil treatment.
- Indiv #: replicate individual 1 through 6.
- Silique #: fruit number 1 through 5.
- Date Counted: the date the seed number was counted for each fruit.
- Data Collector: the initials of the person who counted the seeds.
- # Seeds: the seed number for each fruit. "." represents NA.
File with ibutton soil temperature data.
- Block: the same as described above.
- Treatment: the same as "Envt" described above.
- Year: the same as "Year" described above.
- Month: the month of the measurement, with "Start" for late May and "End" for early May to distinguish the beginning and end of each growth season.
- Date and Time: the date and time of the measurement.
- Data.Temp: the soil temperature measurements in Celsius.
File with soil moisture data.
- Station: The HOBO station number.
- PopID: the population each moisture probe was placed in, with the probe number and the Col and Row described above.
- Block: the same as described above.
- Treatment: the same as "Envt" described above.
- Year: the same as "Year" described above.
- Date, Time, and Month: the same as for "SoilTemp_frostincl_no2022."
- Data.SWC: soil volumetric water content measurements in cubic meters per cubic meter.
File: Processed.Data.zip
Description: This folder includes all processed data files. Files were either processed in R or in Excel. The contents are listed below. "NA" or blank cells represent missing values.
The same population census data as "AllYearsPopCensus" but with partial population censuses removed. All variables are listed above in the "Common Variables" and "AllYearsPopCensus" sections.
The asymptotic population growth rates for each treatment combination and model number output by the bootstrapping procedure.
- Treatment: Dormancy and Envt (as described above) combined.
- Model: the model number (see the manuscript for what the different models represent).
- Lambda: the population growth rate.
- bias: the bootstrap bias.
- std.error: the standard error around the estimated lambda.
- conf.low and conf.high: 95% confidence intervals for the lambda.
File with summary tables for the allele frequency calculations for the Germany x Tajikistan dormant, non-dormant, and mixed populations.
- Column A: Different steps of the allele frequency calculation process.
- Rows 1 through 6 calculate the allele frequencies for the dormant population.
- Rows 7 through 12 calculate the allele frequencies for the non-dormant population.
- Rows 13 through 18 calculate the allele frequencies for the mixed population.
- Columns B through BR: different genetic marker locations across the A. thaliana genome.
Fruit number from "AllYearsRepTraits" supplemented with zeros for populations with less than 20 reproductive individuals.
- Year, Block, RILSet, IndivID, BandColor: all described previously.
The number of new seedlings ("Newsdls") for each census date.
- RILSet, Ident, CensusN: all described previously.
- JulianDate: calendar date converted to Julian Date.
Population persistence data for the population cages.
- Block and RILSet described previously.
- sumY1Total, sumy2Total, sumy3Total: the total number of individuals in each population in years 1, 2, and 3.
- Y1Germ, y2Germ, y3Germ: "Established" for greater than 1 total individual and "Never Established" for zero total individuals in years 1, 2, and 3.
- sumY1Rep and sumy2Rep: the total number of reproductive individuals for each population in years 1 and 2.
- Y1Rep and y2Rep: "Persisted" for greater than 1 reproductive individual and "Disappeared" for zero reproductive individuals in years 1 and 2.
- Y1Persistence, Y2Persistence, Y3Resurrection: the full persistence categories found in Figure S2 for years 1, 2, and 3.
Survival (RoseSurv) from rosette (Totalrosettes) to reproduction (TotalFRMD) data.
- Block, RILSet, and Year described previously.
Data for the change in seedling number (Fluct) between years (Timepoint).
- Block and RILSet described previously.
Survival (SurvtoRose) from seedling (Totalsdls) to rosette (Totalrosettes) data.
- Block, RILSet, and Year described previously.
Data for the total number of reproductive (TotalFRMD) individuals in each population and year.
- Block, RILSet, and Year described previously.
Data for the total number of rosettes (Totalrosettes) in each population and year.
- Block, RILSet, and Year described previously.
Seasonal seedling establishment data.
- Season: the season as "year_season"
- Block, RILSet, Year, and Ident described previously.
- TotPopSurv: the total number of seedlings that established or survived within two weeks.
- TotPopAdjNew: the total number of seedlings.
- seasonalEst: the proportion of seedlings that established.
Data for the total number of seedlings (Totalsdls) in each population and year.
- Block, RILSet, and Year described previously.
File with summary tables for the allele frequency calculations for the UK x USA dormant, non-dormant, and mixed populations.
- Column A: Different steps of the allele frequency calculation process.
- Rows 1 through 4 calculate the allele frequencies for the dormant population.
- Rows 5 through 8 calculate the allele frequencies for the non-dormant population.
- Rows 9 through 12 calculate the allele frequencies for the mixed population.
- Columns B through DE: different genetic marker locations across the A. thaliana genome.
Code/software
To open the files in the Raw.Data and Processed.Data folders, use any software that can open .csv files, including spreadsheet software, plain text editors, and more.
To run the code in the Scripts folder of the GitHub repository linked to through Zenodo, download the Raw.Data and Processed.Data folders from Dryad and use R version 4.5.3 and SAS version 9.4. To run the R code, you will need packages: tidyverse v. 2.0.0, gridExtra v. 2.3, popbio v. 2.8, boot v. 1.3-32, broom v. 1.0.12.
Note: "PopDemog_prep.Rmd" must be run prior to "PopDemog_figs.Rmd", "SeasonalGerm_PopCages.Rmd", and "ProjectionMatrix.Rmd"; "PopPersistence.Rmd" must be run prior to "PopPers_Figs.Rmd."
Access information
Other publicly accessible locations of the data:
Data was derived from the following sources:
- The allele frequency data were derived from previous studies. We cite those studies below next to the file that used the derived data.
- GermTaj_AlleleFreqs.csv: Laserna MP, Sánchez RA, Botto JF. 2008. Light-related loci controlling seed germination in Ler × Cvi and bay-0 x sha recombinant inbred-line populations of Arabidopsis thaliana. Annals of Botany 102: 631–642; Meng PH, MacQuet A, Loudet O, Marion-Poll A, North HM. 2008. Analysis of natural allelic variation controlling Arabidopsis thaliana seed germinability in response to cold and dark: Identification of three major quantitative trait loci. Molecular Plant 1: 145–154. https://publiclines.versailles.inrae.fr/catalogue/cross-ril/33/detail
- ItalySwe_AlleleFreqs.csv: Postma FM, Ågren J. 2015. Maternal environment affects the genetic basis of seed dormancy in Arabidopsis thaliana. Molecular Ecology 24: 785–797. Dryad: doi:10.5061/dryad.k380f.
- UKUSA_AlleleFreqs.csv: Huang X, Schimitt J, Dorn L, et al. 2010. The earliest stages of adaptation in an experimental plant population: strong selection on QTLS for seed dormancy. Molecular Ecology 19: 1335–1351. https://arabidopsis.info/CollectionInfo?id=124
