Density dependence during evolutionary rescue increases extinction risk but does not prevent adaptation
Data files
Jan 07, 2026 version files 264.72 KB
-
Data_CensusRemoval_NDD_DELTA_Clean.csv
163.55 KB
-
Data_Phenotyping_NDD_DELTA_Clean.csv
95.30 KB
-
README.md
5.87 KB
Abstract
Evolutionary rescue allows populations to adapt and persist despite severe environmental change. While well studied under density-independent conditions, the role of density dependence, including competition, remains unclear. Theoretical models offer conflicting predictions, with density dependence either increasing extinction risk or enhancing adaptation. We empirically tested how density dependence influences evolutionary rescue by exposing experimental populations to a stressful environment for six generations under density-dependent or independent conditions, with populations where either evolution was possible or was prevented by replacing individuals each generation. Density dependence suppressed population size and increased extinction risk, whereas density independence enabled rapid growth, especially in genetically diverse populations where evolution was possible. Although density dependence raises extinction risk, it does not prevent populations from responding to selection, since surviving density-dependent populations still exhibited increased intrinsic and realized fitness. These findings reconcile theoretical discrepancies, showing density dependence can simultaneously increase extinction risk but may favor adaptation. Our results underscore the importance of considering density dependence in conservation strategies.
Dryad DOI: https://doi.org/10.5061/dryad.rbnzs7hs7
GENERAL INFORMATION
1. Description
This research compendium describes how to analyze population size, extinction, and life-history traits data from an evolutionary rescue experiment conducted in 2022. here.
The analyses of this research compendium are described in: "Density dependence during evolutionary rescue increases extinction risk but does not prevent adaptation."
Article DOI: https://doi.org/10.1101/2024.12.12.628101v3
2. Author information and main investigator:
Name: Laure Olazcuaga
Institution: Colorado State University, USA
Email: olaz.laure@gmail.com
ORCID: 0000-0001-9100-1305
For the complete list of authors of the manuscript, see the manuscript.
3. Date and geographic location of data collection
2022
Colorado, USA.
4. Funding sources that supported the collection of the data
Data collection was supported by NSF.
5. Recommended citation for this dataset:
Olazcuaga, Laure; Melbourne, Brett A.; Nordstrom, Scott W.; Hufbauer, Ruth A. (2026). Density dependence during evolutionary rescue increases extinction risk but does not prevent adaptation [Dataset]. Dryad. https://doi.org/10.5061/dryad.rbnzs7hs7
CONTENTS
Data and file overview
- Data_CensusRemoval_NDD_DELTA_Clean.csv: Dataset containing evolutionary rescue experiment data.
- Data_Phenotyping_NDD_Delta_Clean.csv: Database containing phenotyping data.
Details of the dataset
- Number of variables: 15
- Number of cases/rows: 1316
- Variable List:
Block: Temporal block; 1, 2, 4, or 5
Gen_adults: Generation of the parents for the census measurement; [0:6]
POP_ID: Population ID from the experiment; from N1 to N192
Nb_Box_adults: number of boxes of adults counted; [0:81]
Census_adults: Number of individuals counted; [0:2827]
Individuals: Origin of the individuals; lab populations or experimental populations
Gen_eggs: Generation of the eggs for the census measurement; [1:7]
Date_start: Start date on which the eggs are laid; mm/dd/yyyy
Date_end: End date on which the eggs are laid; mm/dd/yyyy
Origin_population: ID of the ancestral individuals; Label
Density_dependence: Density of the population; Density_independence or Negative_density_independece
Genetic_variance: Genetic variance of the population; Non-bottlenecked or Bottlenecked
Evolution: Evolution treatment of the population; Evolving or Non-evolving
Treatment: Treatment of the population; Label
Treatment_all: Full treatment of the population; Label - Missing data codes:
None - Abbreviations used:
NA; not applicable - Other relevant information:
None
- Number of variables: 30
- Number of cases/rows: 527
- Variable List:
Block: Temporal block; 1, 2, 4, or 5
POP_ID_ind: ID of the population from the experiment or from the lab stock
Media: Media of the development; Standard or Extreme
Density: Density of the adults; 10ind, 50ind or 100ind
Nb_box: Number of boxes counted; [1:12]
Census1: Number of offspring in the 1st box; [1:1000]
Census2: Number of offspring in the 1st box; [1:1000]
Census3: Number of offspring in the 1st box; [1:1000]
Census4: Number of offspring in the 4th box; [1:1000]
Census5: Number of offspring in the 5th box; [1:1000]
Census6: Number of offspring in the 6th box; [1:1000]
Census7: Number of offspring in the 7th box; [1:1000]
Census8: Number of offspring in the 8th box; [1:1000]
Census9: Number of offspring in the 9th box; [1:1000]
Census10: Number of offspring in the 10th box; [1:1000]
Census11: Number of offspring in the 11th box; [1:1000]
Census12: Number of offspring in the 12th box; [1:1000]
Who_Census: Person who count the individuals in the lab; Flannery, Hailee, Harley, Morgan or Laura
Date_start: Start date on which the eggs are laid; mm/dd/yyyy
Date_end: End date on which the eggs are laid; mm/dd/yyyy
Gen_adults: Generation of the parents for the census measurement; 5
Gen_eggs: Generation of the eggs for the census measurement; 6
Density_ind: Density of the adults; [10:100]
Origin_population: ID of the ancestral individuals; Label
Density_dependence: Density of the population; Density_independence or Negative_density_independece
Genetic_variance: Genetic variance of the population; Non-bottlenecked or Bottlenecked
Evolution: Evolution treatment of the population; Evolving or Non-evolving
Treatment: Treatment of the population; Label
Treatment_all: Full treatment of the population; Label
POP_ID: Population ID from the experiment; from N1 to N147 - Missing data codes:
None - Abbreviations used:
NA; not applicable - Other relevant information:
None
HOW TO RUN IT?
This research compendium has been developed using the statistical programming language R. To work with the compendium, you will need to install the R software
itself and optionally RStudio Desktop.
You can download the compendium by cloning this repository:
- open the
.Rprojfile in RStudio - open scripts
.Rmdin reports folder and run it to produce all the analyses and associated reports. - launch the
README.htmlto be able to explore the contents on your web browser
