Pollination context and abiotic stress reshape variation in floral longevity and its exposure to selection
Data files
Mar 24, 2026 version files 192.70 KB
-
README.md
7.92 KB
-
Spigler_Ostrowski_AbioticData.txt
3.22 KB
-
Spigler_Ostrowski_ClimateSoilDataExtractions.R
3.16 KB
-
Spigler_Ostrowski_FLdata_for_abiotic_data_calculations.txt
30.95 KB
-
Spigler_Ostrowski_FloralLongevityAnalysis_SAScode.txt
5.02 KB
-
Spigler_Ostrowski_FloralLongevityTrait_AbioticDataset.csv
85.19 KB
-
Spigler_Ostrowski_KS_QuantileAnalysis_Dryad.R
6.15 KB
-
Spigler_Ostrowski_PollenLoadAnalysis_SAScode.txt
2.22 KB
-
Spigler_Ostrowski_PollenLoadData.csv
7.89 KB
-
Spigler_Ostrowski_PredationData.csv
16.25 KB
-
Spigler_Ostrowski_SeedPredationAnalysis_SAScode.txt
5.88 KB
-
Spigler_Ostrowski_VisitationData.csv
1.47 KB
-
Spigler_Ostrowski_VistationAnalysis_SAScode.txt
1.51 KB
-
Spigler_OstrowskiFigure2code.R
7.77 KB
-
Spigler_OstrowskiFigure3code.R
8.11 KB
Abstract
Floral longevity is thought to evolve by natural selection imposed by pollinators and resource constraints acting on heritable phenotypic variation. In species with pollination-induced wilting, pollination rates may also influence expression of variation, raising questions of how and when floral longevity is shaped by adaptation. We created an experimental range of pollen deposition rates in a wild population of Sabatia angularis (Gentianaceae) against the backdrop of natural abiotic variation. We investigated how pollination environment affected expression of longevity, distribution of variation, phenotypic correlations, and predispersal seed predation risk. We evaluated whether any effects of abiotic factors are modified by pollination environment. Mean and variance in floral longevity declined with increasing pollen deposition rates, and phenotypic distributions became increasingly compressed. Nevertheless, we detected a flower lifespan-number trade-off and positive flower lifespan-size relationship, independent of treatment. Vapor pressure deficit (VPD) and intraseasonal changes constrained longevity. Suppression associated with VPD was greatest under low pollen deposition rates, where selection is expected to favor longer-lived flowers. Floral longevity had weak and inconsistent effects on predispersal seed predation risk. Our results demonstrate that phenotypic variation in floral longevity is conditionally expressed across ecological contexts, altering when variation is exposed to selection and subsequent evolutionary potential.
Dataset DOI: 10.5061/dryad.76hdr7t9j
Description of the data and file structure
This repository contains the datasets and analysis scripts used in the study "Pollination context and abiotic stress reshape variation in floral longevity and its exposure to selection". The files include data on floral longevity and associated traits, abiotic conditions, pollen deposition, pollinator visitation, and pre-dispersal seed predation and code used to conduct analyses and generate figures.
Files and variables
File: Spigler_Ostrowski_SeedPredationAnalysis_SAScode.txt
Description: SAS code (in text file) used to analyze risk of pre-dispersal seed predation in relation to longevity
Variables
- Uses data file: Spigler_Ostrowski_PredationData.xlsx
File: Spigler_Ostrowski_PollenLoadAnalysis_SAScode.txt
Description: SAS code (in text file) used to analyze patterns of pollen deposition
Variables
- Uses data file: Spigler_Ostrowski_PollenLoadData.csv
File: Spigler_Ostrowski_FloralLongevityAnalysis_SAScode.txt
Description: SAS code (in text file) used to analyze effects of treatment, flower number, flower size, VPD, and date of flower on longevity including interactions
Variables
- Uses data file: Spigler_Ostrowski_FloralLongevityTrait_AbioticDataset.csv
File: Spigler_Ostrowski_VisitationData.csv
Description: Pollinator visitation rates recorded as number visited within an hour for each plant.
Variables
- Plant ID: Identity of the individual plant
- PollTrt: Pollination treatment
- Date: Date observations were recorded
- NumberofVisits: Number of flower visits per hour
- NumberofVisitors: Number of flower visitors per hour (each visitor may make several visits)
- Day: Day observations recorded to use as covariate
- FloralDisplay: number of flowers open on the plant on the day of observation
File: Spigler_Ostrowski_PollenLoadData.csv
Description: Dataset for testing effect of treatment, flower age, and their interaction on pollen load (grains/stigma).
Variables
- plantID: Identity of plant
- FlowerAge: Age of flower when stigma was removed
- pollen_load: Number of pollen grains per stigma
- treatment: pollination treatment (Pollinator Exclusion [PE] and Natural Pollination [NP])
- julianday: day the stigma was collected
- Identifier: unique identity for each row of data (data point)
File: Spigler_Ostrowski_AbioticData.txt
Description: Abiotic Data for each day during the study period. These data are used to calculate averages over individual flower lifetimes.
Variables
- JulianDay: Day within the year
- MaxT: Maximum temperature (C) on that day
- MaxPAR: Maximum photosynthetic active radiation on that day
- Precip: Cumulative precipitation on that day
- MaxVPD: maximum vapor pressure deficit for that day determined from temperature and relative humidity
- Max Tsoil10: maximum soil temperature 10cm below on that day
- MaxTsoil_5: maximum soil temperature 5cm below on that day
- MaxVWC: maximum volumetric water content in soil on that day
File: Spigler_Ostrowski_KS_QuantileAnalysis_Dryad.R
Description: Code for pairwise Kolmogorov–Smirnov (K-S) tests and quantile-based analyses using bootstrapping.
File: Spigler_Ostrowski_VistationAnalysis_SAScode.txt
Description: SAS code (in text file) used to analyze pollinator visitation data
Variables
- Uses data file: Spigler_Ostrowski_VisitationData.csv
File: Spigler_Ostrowski_PredationData.csv
Description: Dataset for predispersal seed predation. Data are at the plant level, not flower level.
Variables
- PlantID: Identity of plant
- TrtCode: pollination treatment (Pollinator Exclusion [PE] and Natural Pollination [NP])
- Height: height of plant (cm)
- LNS1m: number of conspecifics within 1m of plant
- FlowerNumber: total number of flowers
- FruitNum: total number of fruits
- Predated_binary: evidence of any predation (1) vs. none (0)
- PredatedFruitNum: number of fruits with evidence of predation
- DFFOJulian: date first flower opened on plant
- DLWF_Julian: date last flower opened on plant
- meanFloLo: mean floral longevity per plant
File: Spigler_Ostrowski_FLdata_for_abiotic_data_calculations.txt
Description: Dataset for calculating mean values of abiotic factors across floral lifespans
Variables
- PlantIDfactor: Unique plant identity coded as a factor
- PlantID: Unique plant identity as a number
- TrtCode: Pollination treatment code (PE=pollinator exclusion; E=emasculated; NP=natural pollination; SP=supplemental hand pollination)
- var: Indicates identify of individual flower on the plant
- FlowerID: unique identifier for the individual flower (data row)
- Position: indicates whether flower was at secondary or tertiary position in inflorescence
- JulianOpen: Day the flower opened within the year
- JulianWilt: Day the flower wilted within the year
- FloLo: floral longevity = JulianWilt - JulianOpen
File: Spigler_Ostrowski_FloralLongevityTrait_AbioticDataset.csv
Description: Dataset used for main analysis evaluating effects of pollination treatment, flower number, flower size, and abiotic factors on floral longevity. NOTE: Columns PlantIDFactor through FloLo are the same as in Spigler_Ostrowski_FLdata_for_abiotic_data_calculations.txt
Variables
- PlantIDfactor: Unique plant identity coded as a factor
- PlantID: Unique plant identity as a number
- TrtCode: Pollination treatment code (PE=pollinator exclusion; E=emasculated; NP=natural pollination; SP=supplemental hand pollination)
- var: Indicates identify of individual flower on the plant
- FlowerID: unique identifier for the individual flower (data row)
- Position: indicates whether flower was at secondary or tertiary position in inflorescence
- JulianOpen: Day the flower opened within the year
- JulianWilt: Day the flower wilted within the year
- FloLo: floral longevity = JulianWilt - JulianOpen
- Height: height of plant (cm)
- FlowerNum: total number of flowers produced by the plant
- DFFOJulian: date the first flower opened on the plant
- DLWF_Julian: date the last flower wilted on the plant
- FlwrDuration: flowering duration = DLWF_Julian - DFFOJulian
- PlantAvgPetArea: Average petal area (average of 2 flowers)
- PlantMaxFloDisp: greatest number of flowers open on a given day across the plant's flowering duration.
- mean_MaxT_FL: mean for maximum temperature across this flower's lifespan
- mean_MaxVPD_FL: mean for maximum VPD across this flower's lifespan
- mean_MaxTsoil_5_FL: mean for maximum soil temperature at 5cm across this flower's lifespan
- mean_Max.Tsoil10_FL: mean for maximum soil temperature at 10cm across this flower's lifespan
- mean_MaxVWC_FL:mean for maximum VWC across this flower's lifespan
File: Spigler_Ostrowski_ClimateSoilDataExtractions.R
Description: Code used to calculate mean values of abiotic factors across individual floral lifespans
File: Spigler_OstrowskiFigure2code.R
Description: Code to create violin and quantile plots (Figure 2).
File: Spigler_OstrowskiFigure3code.R
Description: Code to create effect plots (Figure 3)
Code/software
R version 4.4.1 (R Core Team, 2024) within RStudio (Version 2024.09.1+394, Posit team, 2025)
Packages and data files used are clearly indicated within R scripts.
SAS version 9.4, SAS Institute Inc. SAS OnDemand for Academics is a free cloud-based version of SAS software through a web browser for non-commercial educational and research use. https://welcome.oda.sas.com/
SAS code is provided in text files and can be copied and pasted into the SAS editor.
