Long-term demographic data on the red-cockaded woodpecker
Data files
Oct 19, 2023 version files 5.78 MB
Abstract
In cooperative breeding systems, inclusive fitness theory predicts that non-breeding helpers more closely related to the breeders should be more willing to provide costly alloparental care, and thus have more impact on breeder fitness. In the red-cockaded woodpecker (Dryobates borealis), most helpers are the breeders’ earlier offspring, but helpers do vary within groups in both relatedness to the breeders (some even being unrelated) and sex, and it can be difficult to parse their separate impacts on breeder fitness. Moreover, most support for inclusive fitness theory has been positive associations between relatedness and behavior, rather than actual fitness consequences. We used functional linear models to evaluate the per capita effects of helpers of different relatedness on eight breeder fitness components measured for up to 41 years at three sites. In support of inclusive fitness theory, helpers more related to the breeding pair made greater contributions to six fitness components. However, male helpers made equal contributions to increasing pre-fledging survival regardless of relatedness. These findings suggest that both inclusive fitness benefits and other, direct benefits may underlie helping behaviors in the red-cockaded woodpecker. Our results also demonstrate the application of an underused statistical approach to disentangle a complex ecological phenomenon.
README
Manuscript citation: Kerr, William, and Walters (2023) Inclusive fitness may explain some but not all benefits derived from social behavior in a cooperatively breeding bird. American Naturalist.
Archive citation: Kerr, Natalie; Morris, William; Walters, Jeffrey (Forthcoming 2023). Demographic data on the red-cockaded woodpecker [Dataset]. Dryad. https://doi.org/10.5061/dryad.3bk3j9kqs
Affiliated authors: Natalie Z. Kerr, William F. Morris, and Jeffrey R. Walters
Corresponding author details:
- Email: natalie.kerr@duke.edu,
- Location: Department of Biology, Duke University, Durham, NC 27705
To run the code file ("Kerr-et-al_FLMs.rmd"), you will need to install R and RStudio, as well as install the packages (using install.packages()) listed below and at the beginning of the RMarkdown file.
List of packages and their versions
- mgcv (Version 1.8.42)
- bbmle (Version 1.0.25)
Note that we used these versions of these two packages for Kerr et al. 2023.
The RMarkdown file and datafiles need to be contained within the same folder for the code to run. To run our FLM analyses, we archived two datasets: reproductive data (for the five reproductive fitness components) and the survival data (for the three survival fitness components). Both are part of the long term demographic monitoring.
For the reproductive data (rep_dat), each row is a group in a particular year. For the survival data (surv_dat), each row is a particular individual (that is either a nestling N and breeder B).
List of variables present in both datasets:
- Pop - Three populations or "sites" (i.e., Bragg, Camp Lejeune, and Eglin)
- Territory - Unique number ID for each territory
- TerritoryByManag - Unique number ID for territory x management period
- Year - Year data was collected
- mother.age - Age of the mother (in years) for that territory in that particular year
- father.age - Age of the father (in years) for that territory in that particular year
- MR - For the rds file, MR is a matrix where each column represents a summed relatedness value for male helpers, rows represent unit of observation which is group for rep_dat and individual birds for surv_dat. Note that the columns of the MR matrix are populated into individual columns of the data frame for the csv file.
- MN - For the rds file, MN is a corresponding matrix for MR that represents the number of male helpers of that particular summer relatedness value present in the group. Note that the columns of the MN matrix are populated into individual columns for the csv file.
- FR - Similar to MR, the rds file contains a FR matrix where each column represents a summed relatedness value for female helpers, rows represent unit of observation which is group for rep_dat and individual birds for surv_dat. Note that the columns of the MR matrix are populated into individual columns of the data frame for the csv file.
- FN - Similar to MN, the rds file contains a corresponding FN matrix for FR that represents the number of female helpers of that particular summer relatedness value present in the group. Note that the columns of the FN matrix are populated into individual columns for the csv file.
List of variables present in the reproductive dataset:
- firstnestattempt - Binary variable representing whether the breeding pair attempted a first nest (i.e., 0 or 1)
- firsteggattempt - Number of eggs in the first clutch
- eggsuccess - Binary variable representing whether one or more eggs successfully hatched in the first nest (i.e., 0 or 1)
- nestlingsuccess - Binary variable representing whether one or more nestlings successfully fledged in the first nest (i.e., 0 or 1)
- firstfledgling - Number of fledglings in the first nest
List of variables in the survival dataset:
- sex - Sex of the individual (i.e., F for female and M for male)
- age - Age of the individual (in years)
- status - Status of the individual bird (i.e., J for juvenile and B for breeder)
- surtonext - Binary variable representing whether the bird survived to the next year ("1") or did not ("0")
Note that columns with cells containing "NA" values are groups or individuals without data for that associated column.
Please reach out to Natalie Z. Kerr, for any questions regarding these data and code.
Methods
We used long-term demographic monitoring data collected over 28 to 41 consecutive years at three sites: the Sandhills region in south-central North Carolina (1980–2020), Marine Corps Base Camp Lejeune on the central coast of North Carolina (1986–2020), and Eglin Air Force Base in the western panhandle of Florida (1993–2020). Monitoring methods are described in detail by Walters et al. (1988) (see also Appendix A for more details on monitoring). See Walters and Garcia (2016) for how individuals are assigned breeder and helper status.
Usage notes
You will need both R and RStudio to use the dataset (and corresponding code).