Data from: Sex-specific effects of early-life adversity on adult fitness in a wild mammal
Data files
Mar 03, 2025 version files 1.85 MB
-
AdultAnnualDF.txt
1.14 MB
-
AdultLifetimeDF.txt
185.35 KB
-
Drake_ProcB_Models.R
44.67 KB
-
LambDF.txt
469 KB
-
README.md
10.95 KB
Abstract
Early-life adversity influences adult fitness across vertebrate species. In polygynous systems with intense intrasexual competition, males may be more sensitive to conditions experienced during development. However, the importance of different aspects of the early-life environment and how their effects differ between the sexes remains poorly understood. Here, we used a long-term study of wild Soay sheep to characterise the early-life environment in terms of weather, infection risk, resource competition and maternal investment, and test the hypothesis that males are more vulnerable to early adversity. Birth weight, reflective of maternal investment and conditions during gestation, positively predicted lifetime breeding success in both sexes, suggesting a classic ‘silver spoon’ effect, though the effects were stronger in males. Males that experienced increased resource competition in their first year had lower lifetime breeding success, suggesting lasting negative consequences of nutritional stress, but there was no association in females. In contrast, challenging weather in the first winter of life was associated with stronger viability selection, with males surviving these harsh conditions having higher adult fitness. Our findings further evidence the important long-term fitness consequences of early-life adversity in wild vertebrates, demonstrating distinct aspects of the early environment may shape fitness in different and sex-specific ways.
https://doi.org/10.5061/dryad.dfn2z359k
SOAY SHEEP PROJECT DATA REUSE:
The attached files contain data derived from the long term field project monitoring individual Soay sheep on St Kilda and their environment. This is a request to please let us know if you use them. Several people have spent the best part of their careers collecting the data. If you plan to analyse the data, there are a number of reasons why it would be very helpful if you could contact Dan Nussey (dan.nussey@ed.ac.uk) before doing so.
[NB. If you are interested in analysing the detailed project data in any depth you may find it helpful to have our full relational database rather than the files available here. If so, then we have a simple process for bringing you onto the project as a collaborator.]
1) The data can be subject to change due to updates in the pedigree, merging of records, occasional errors and so on.
2) The data are complex and workers who do not know the study system may benefit from advice when interpreting it.
3) At any one time a number of people within the existing project collaboration are analysing data from this project. Someone else may already be conducting the analysis you have in mind and it is desirable to prevent duplication of effort.
4) In order to maintain funding for the project, every few years we have to write proposals for original analyses to funding agencies. It is therefore very helpful for those running the project to know what data analyses are in progress.
5) Individual identifiers may vary relative to other data archives from papers using the individual-level data.
DATASETS:
There are three datasets associated with this paper: LambDF (N=6592 observations), AdultLifetime (N=1721 observations) and AdultAnnual (N=9572 observations). The adult datasets include only individuals that survived beyond 1 May in the year following birth.
Faecal egg counts were unavailable for the earliest cohorts (1985-1987) and for >90% lambs in 2002, and counts were performed using a different method in recent years (2019-2021), so individuals belonging to these cohorts have NA values for CohortMeanFEC.
At least 60% of individuals in each cohort were captured and weighed within a week of birth, except in 2001 and 2020 where logistical challenges resulted in a high proportion of missing data (foot and mouth outbreak and Covid-19 pandemic, respectively). An NA value for BirthWeight means either the individual was not caught in first week of life or belonged to the 2001 or 2020 cohort.
The variables are as follows:
1. LambDF.txt
| Variable | Description |
|---|---|
| BirthYear | Year of birth |
| Sex | M: Male; F: Female |
| NatalLitterSize | Singleton or Twin at birth |
| MaternalLoss | MotherDied: mother died before 1 May in the year following birth; MotherSurvived: mother survived beyond 1 May in the year following birth |
| PopSizeYOB | Total Village Bay population size on 1 October in the year of birth, in number of individuals |
| FirstWinterNAO | December-March Hurrell North Atlantic Oscillation Index (station-based) for winter following birth |
| CohortMeanFEC | Mean lamb August faecal egg count in year of birth, in number of eggs present in a sample |
| BirthWeight | Residuals of a model of weight at first capture for individuals caught within the first week of life, controlling for age at capture in days |
| FirstYearSurvival | 0: lamb died before 1 May in the year following birth; 1: lamb survived beyond 1 May in the year following birth |
| RecodedID | Identity of individual - recoded |
| RecodedMumID | Identity of mother - recoded |
2. AdultLifetimeDF.txt
| Variable | Description |
|---|---|
| BirthYear | Year of birth |
| Sex | M: Male; F: Female |
| NatalLitterSize | Singleton or Twin at birth |
| MaternalLoss | MotherDied: mother died before 1 May in the year following birth; MotherSurvived: mother survived beyond 1 May in the year following birth |
| PopSizeYOB | Total Village Bay population size on 1 October in the year of birth, in number of individuals |
| FirstWinterNAO | December-March Hurrell North Atlantic Oscillation Index (station-based) for winter following birth |
| CohortMeanFEC | Mean lamb August faecal egg count in year of birth, in number of eggs present in a sample |
| BirthWeight | Residuals of a model of weight at first capture for individuals caught within the first week of life, controlling for age at capture in days |
| MeanAdultWt | Mean residuals of a model of adult (age 1+ years) August weight controlling for age in years, sex, and the interaction between the two |
| LifetimeBreedingSuccess | Number of live offspring born to or sired by an individual over their lifespan |
| Longevity | Age at death in years |
| RecodedID | Identity of individual - recoded |
| RecodedMumID | Identity of mother - recoded |
3. AdultAnnualDF.txt
| Variable | Description |
|---|---|
| BirthYear | Year of birth |
| Sex | M: Male; F: Female |
| Year | Observation year |
| Age | During the summer, in years |
| NatalLitterSize | Singleton or Twin at birth |
| MaternalLoss | MotherDied: mother died before 1 May in the year following birth; MotherSurvived: mother survived beyond 1 May in the year following birth |
| PopSizeYOB | Total Village Bay population size on 1 October in the year of birth, in number of individuals |
| FirstWinterNAO | December-March Hurrell North Atlantic Oscillation Index (station-based) for winter following birth |
| CohortMeanFEC | Mean lamb August faecal egg count in year of birth, in number of eggs present in a sample |
| BirthWeight | Residuals of a model of weight at first capture for individuals caught within the first week of life, controlling for age at capture in days |
| MeanAdultWt | Mean residuals of a model of adult (age 1+ years) August weight controlling for age in years, sex, and the interaction between the two |
| BreedingProbability | 0: didn’t give birth to or sire an offspring; 1: did give birth to or sire an offspring |
| FemaleTwinningProbability | For females that bred, 0: gave birth to a singleton lamb; 1: gave birth to twin lambs |
| MaleOffspringNumber | For males that bred, the number of live offspring sired |
| RecodedID | Identity of individual - recoded |
| RecodedMumID | Identity of mother - recoded |
MODELS
The file Drake_ProcB_Models.R contains all the necessary code to run the models included in this paper in R. This script requires the packages glmmTMB and performance. This work was orginally performaed using R 4.3.3, glmmTMB 1.1.9 and performance 0.11.0.