Data from: The conservation value of small population remnants: variability in inbreeding depression and heterosis of a perennial herb, the narrow-leaved purple coneflower (Echinacea angustifolia)
Data files
Oct 18, 2024 version files 5.61 MB
-
echExPtBWI_14Years.csv
5.60 MB
-
README.md
5.88 KB
Abstract
Anthropogenically fragmented populations may have reduced fitness due to loss of genetic diversity and inbreeding. The extent of such fitness losses due to fragmentation and potential gains from conservation actions are infrequently assessed together empirically. Controlled crosses within and among populations can identify whether populations are at risk of inbreeding depression and whether interpopulation crossing alleviates fitness loss. Because fitness depends on environment and life stage, studies quantifying cumulative fitness over a large portion of the lifecycle in conditions that mimic natural environments are most informative. To assess fitness consequences of habitat fragmentation, we leveraged controlled within-family, within-population, and between-population crosses to quantify inbreeding depression and heterosis in seven populations of Echinacea angustifolia within a 6400-hectare area. We then assessed cumulative offspring fitness after 14 years of growth in a natural experimental plot (N = 1136). Mean fitness of progeny from within-population crosses varied considerably, indicating genetic differentiation among source populations, even though these sites are all less than 9 km apart. The fitness consequences of within-family and between-population crosses varied in magnitude and direction. Only one of the seven populations showed inbreeding depression of high effect, while four populations showed substantial heterosis. Outbreeding depression was rare and slight. Our findings indicate that local crossings between isolated populations yield unpredictable fitness consequences ranging from slight decreases to substantial increases. Interestingly, inbreeding depression and heterosis did not relate closely to population size, suggesting that all fragmented populations could contribute to conservation goals as either pollen recipients or donors.
https://doi.org/10.5061/dryad.zcrjdfnp2
Description of the data and file structure
We performed within-population, within-family, and between-population crosses for seven remnant populations of Echinacea angustifolia. We grew progeny in an experimental plot that mimics common habitat. We assessed fitness as the cumulative fruit count of progeny over 14 years of growth.
Files and variables
File: echExPtBWI_14Years.csv
Description: Main dataset containing pedigree and progeny fitness information over 14 years. Dataset is in long format and contains variables for aster analysis format.
Variables
- sirepop: Source population of the pollen donor for each cross. Seven levels: Aanenson, Eriley, Lf, Nessman, NWLF, Stevens, SPP
- matpop: Source population of the pollen recipient for each cross (e.g., the maternal plant). Seven levels: Aanenson, Eriley, Lf, Nessman, NWLF, Stevens, SPP
- mat: Unique ID of the pollen recipient for each cross.
- matfam: Unique family ID of the pollen recipient in each cross (e.g., ID of the parent of the pollen recipient). All within-family crosses (I crosses) have matching matfam and sirefam.
- sire: Unique ID of the pollen donor for each cross.
- sirefam: Unique family ID of the pollen donor in each cross (e.g., ID of the parent of the pollen donor). All within-family crosses (I crosses) have matching matfam and sirefam.
- crosslevel: Numerical identifier for crosstype. Higher number indicates a “more outcrossed” crosstyle. Levels: 1 = inbred (within-family cross), 2 = within (within-population, random family), 3 = between (between-population cross).
- crossType: Identifier for genetic cross level for each progeny. Levels: I = inbred, where two parents from the same maternal family were crossed to produce the progeny. W = within-population, where two random parents from the same remnant population, but not from the same maternal family, were crossed. B = between-population, where two parents from different populations were crossed to produce the progeny.
- yrPlanted: Year each seedling was planted in the experimental plot. All seedlings in this experiment were planted in 2006.
- row: Position (meters) relative to the center of all rows in which Echinacea were planted. There are six total rows. Row -2.5 is the furthest west, while row 2.5 is the easternmost row. Rows are spread one meter from one another.
- pos: Position (meters) relative to the center of all positions in which Echinacea were planted. Echinacea were planted in positions every 0.5 meters, and positions span ~120 meters in the plot. Position -62.5 is the southernmost position, while 62.5 is the most northern position. The location of plants within the plot relative to center may be found by its row and position coordinates.
- measureYr: Year of data collection since planting in 2006. measureYr = 1 is measurement upon planting; measureYr = 2 is measurement in fall of 2006. measureYr = 3 is measurement in fall of 2007… and measureYr = 15 is measurement in fall of 2019.
- resp: Fitness component response in aster model format. Each individual, in each year, at each fitness component (layer) has a response which indicates whether they were alive (ld = 1, 0), whether they flowered (fl = 1, 0), their count of flowering heads (hdCt = integer), and the total number of fruits (achenes) produced (achCt = integer).
- layer: Fitness component measured in each row, formatted for aster models. Four levels: ld (living during, an indicator of survival at that time), fl (flowered, whether or not a plant flowered in that year), hdCt (head count, the number of heads a plant produced that year), and achCt (achene count, or fruit count, the total number of fruits produced by that plant in that year).
- root: An indicator variable for aster format. Root is always 1, and indicates the starting node - the root node - for model specification. In this case, the root node is measureYr = 1, or when seedlings were planted in the experimental plot.
- varb: An appended form of layer and measureYr. ld06 indicates if a plant survived to measure time 6. achCt12 is the number of fruits a plant produced in measure year 12.
- fit: An variable used in aster format to indicate which layers (varbs) will be used to calculate fitness. In this dataset, fit is 0 for all layers except for achCt, where it is 1. Thus, fitness is calculated as the sum of achene counts.
- id: Individual ID for each plant in the experiment. Range is 1:1136.
- matPopSize: Population size of the maternal plant for each progeny. Population size is the mean number of flowering plants in that population from 1995-1997, when seeds were collected for this study.
- matPopType: An indicator for whether the population of the pollen recipient is a roadside or non-roadside populations. Roadside populations are subject to frequent disturbance.
- sirePopSize: Population size of the paternal plant for each progeny. Population size is the mean number of flowering plants in that population from 1995-1997, when seeds were collected for this study.
- sirePopType: An indicator for whether the population of the pollen donor is a roadside or non-roadside populations. Roadside populations are subject to frequent disturbance.
Code/software
Data were analyzed using R version 4.0.2. We used the aster package in R, version 1.3 (Shaw et al. 2008).
Shaw, R. G., Geyer, C. J., Wagenius, S., Hangelbroek, H. H., and Etterson, J. R. 2008. Unifying life-history analyses for inference of fitness and population growth. The American Naturalist 172(1):E35-E47.
We performed controlled within-family, within-population, and between-population crosses of the prairie perennial, Echinacea angustifolia (Asteraceae) originating from seven remnant populations. We then assessed the cumulative fitness of progeny over 14 years in an experimental plot using aster models.