Data from: Population dynamics of a long-lived pitviper inferred from capture-recapture data spanning four decades
Data files
Jan 14, 2026 version files 50.69 KB
-
C_horridus__NY_USA_1983-2019_F.inp
46.98 KB
-
README.md
3.71 KB
Abstract
Long-term studies of plant and animal populations are rare, particularly for long-lived species with slow life histories. However, such studies are necessary to understand how variation in vital rates affects population dynamics. We used a large capture-recapture dataset spanning four decades (1983–2019) to estimate several population parameters for adult females in a metapopulation of a long-lived North American pitviper (Timber Rattlesnake, Crotalus horridus) near the northern extent of the species’ geographic range. Apparent survival (ϕ, hereafter, survival) estimates were variable (0.595–0.938) among years but generally high, whereas the annual recruitment rate (f) was low (0.102, 95% confidence interval [CI]: 0.088–0.117). Survival increased by a quadratic function of the mean mass of the adult female population, resulting in survival increasing with this temporally varying population-level covariate until the population average reached approximately 768 g; survival decreased with increasing mean mass thereafter. The proximate causes of decreased survival for the largest, and thus presumably oldest, snakes remain unknown but are likely related to senescence and increased predation risk with advancing age. The best predictors for encounter probabilities were the additive effects of color morph (higher for yellow snakes than black snakes), sampling site, and year. The model-averaged estimate of realized population growth (λ) was 0.992 (95% CI: 0.980–1.005). Removal of sampling variance from the component of λ contributed by survival resulted in a 95% upper limit of 0.9997, indicating a declining population. Reflecting attributes of the species’ protracted life history (late maturation, low fecundity, and infrequent reproduction), the contribution of survival was on average 8.9 (95% CI: 7.6–10.4) times greater than the contribution of recruitment to λ, representing the slowest life history documented for any snake population to date. Our findings suggest management efforts to increase adult survival, such as reducing mortality from anthropogenic factors, should aid in the conservation of Timber Rattlesnakes and other imperiled species with similar slow life histories.
GENERAL INFORMATION
-
Author Information Name: Dr. Sasha J. Tetzlaff*, https://orcid.org/0000-0001-5889-6573
Institution: U.S. Army Corps of Engineers, Engineer Research and Development Center, Champaign, Illinois, USA.
Co-investigator 1
Name: Dr. Eric T. Hileman*, https://orcid.org/0000-0002-7044-370X
Institution: West Virginia University, Morgantown, West Virginia, USA.
Co-investigator 2
Name: Matthew G. Simon, Institution: Independent Researcher, Fort Ann, New York, USA.
Co-investigator 3
Name: Dr. Gordon W. Schuett, https://orcid.org/0000-0002-2133-3723
Institution: Chiricahua Desert Museum, Rodeo, New Mexico, USA; Department of Biology, Neuroscience Institute, Georgia State University, Atlanta, Georgia, USA.
Co-investigator 4
Name: Dr. Mark A. Davis, https://orcid.org/0000-0001-9034-9430
Institution: Illinois Natural History Survey, Prairie Research Institute, Champaign, Illinois, USA; Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, Illinois, USA.
Co-investigator 5
Corresponding Investigator
Name: Dr. William S. Brown, Institution: Department of Biology, Skidmore College, Saratoga Springs, New York, USA.
Email: wbrown@skidmore.edu
Co-investigator 6*Denotes co-lead authorship
- Dates of data collection: 1983, 1985-1988, 1992-1994, 1996, 1998, 2000, 2002, 2004-2009, 2011-2016, 2018, 2019
- Geographic location of data collection: Southeastern Adirondack Mountains, New York, USA.
- Funding sources that supported the collection of the data: Self-funded.
DATA & FILE OVERVIEW
- Description of dataset
These capture-recapture data were generated to investigate realized population growth of the Timber Rattlesnake (Crotalus horridus) using Program MARK version 11.1. - File list:
File 1 Name: C_horridus__NY_USA_1983-2019_F.inp
METHODOLOGICAL INFORMATION
Apparent survival (phi), capture probabilities (p), recruitment rate (f) and realized population growth were estimated using Pradel's time symmetric survival and recruitment model, where phi, p, and f are estimated using maximum likelihood and realized population growth is a derived parameter. All functions of the above parameters were simulated in the r package 'mvtnorm' using estimates and associated variance-covariance matrix generated in Program MARK.
DATA-SPECIFIC INFORMATION FOR: C_horridus_ NY_USA_1983-2019_F.inp. Tetzlaff et al. 2025
Line 1: Description and title.
Line 2: Authors.
Line 4: study site.
Line 5: Sampling years.
Lines 7 to 33: Years and interval lengths adjusted to annual intervals
Lines 35 to 47: Den names and groups (one factor with 12 levels), 6 dens for black morphs (B) and the same 6 dens for yellow morphs (Y).
Lines 49 to 77:
Left column: Years = the sampling intervals from which annual estimates are derived),
Middle column: Popmass = average adult female mass; a temporally varying population-level covariate)
Right column 11-yrs past_precip = Precipitation 11 years prior to the current occasion.
Lines 79 to 111: Same as above but standardize by subtracting each covariate by its mean and divided it by 2*SD so that coefficients are directly comparable to untransformed binary predictors.
Lines 113 to 819: capture histories (26 occasions) of females ≥ 500 grams, followed by the 12 den/morph groups.
Code/Software
Can be opened with any text editor