Data from: Conservation through the lens of (mal)adaptation: concepts and meta-analysis

Derry, Alison, Département des sciences biologiques Université du Québec à Montréal Montreal Quebec Canada

Fraser, Dylan, Quebec Center for Biodiversity Science Montreal Quebec Canada

Brady, Steven, Southern Connecticut State University

Astorg, Louis, Département des sciences biologiques Université du Québec à Montréal Montreal Quebec Canada

Lawrence, Elizabeth, Concordia University

Martin, Gillian, Département des sciences biologiques Université du Québec à Montréal Montreal Quebec Canada

Matte, Jean-Michel, Concordia University

Negrín Dastis, Jorge Octavio, Département des sciences biologiques Université du Québec à Montréal Montreal Quebec Canada

Paccard, Antoine, McGill University

Barrett, Rowan, Quebec Center for Biodiversity Science Montreal Quebec Canada

Chapman, Lauren, Quebec Center for Biodiversity Science Montreal Quebec Canada

Lane, Jeffrey, University of Saskatchewan

Ballas, Chase, Pace University

Close, Marissa, Pace University

Crispo, Erika, Pace University

Published Mar 12, 2019 on Dryad. https://doi.org/10.5061/dryad.76dc375

Cite this dataset

Derry, Alison et al. (2019). Data from: Conservation through the lens of (mal)adaptation: concepts and meta-analysis [Dataset]. Dryad. https://doi.org/10.5061/dryad.76dc375

Abstract

Evolutionary approaches are gaining popularity in conservation science, with diverse strategies applied in efforts to support adaptive population outcomes. Yet conservation strategies differ in the type of adaptive outcomes they promote as conservation goals. For instance, strategies based on genetic or demographic rescue implicitly target adaptive population states whereas strategies utilizing transgenerational plasticity or evolutionary rescue implicitly target adaptive processes. These two goals are somewhat polar: adaptive state strategies optimize current population fitness, which should reduce phenotypic and/or genetic variance, reducing adaptability in changing or uncertain environments; adaptive process strategies increase genetic variance, causing maladaptation in the short term, but increase adaptability over the long term. Maladaptation refers to suboptimal population fitness, adaptation refers to optimal population fitness, and (mal)adaptation refers to the continuum of fitness variation from maladaptation to adaptation. Here we present a conceptual classification for conservation that implicitly considers (mal)adaptation in the short-term and long-term outcomes of conservation strategies. We describe cases of how (mal)adaptation is implicated in traditional conservation strategies, as well as strategies that have potential as a conservation tool but are relatively underutilized. We use a meta-analysis of a small number of available studies to evaluate whether the different conservation strategies employed are better suited toward increasing population fitness across multiple generations. We found weakly increasing adaptation over time for transgenerational plasticity, genetic rescue, and evolutionary rescue. Demographic rescue was generally maladaptive, both immediately after conservation intervention and after several generations. Interspecific hybridization was adaptive only in the F1 generation, but then rapidly lead to maladaptation. Management decisions that are made to support the process of adaptation must adequately account for (mal)adaptation as a potential outcome and even as a tool to bolster adaptive capacity to changing conditions.

Usage notes

Table S1 database effect sizes - (Mal)adaptation meta-analysis

The following keywords were used in our literature searches of conservation strategies: `assisted colonization`, `assisted migration`, `translocation`, `genetic rescue`, `evolutionary rescue`, `transgenerational plasticity`, `hybridi*ation AND outbreeding`, `hybridi*ation OR hybrid AND viability AND conservation`; (January 2018 using Web of Science or Google Scholar). To determine if an article had usable data on maladaptation, we used the following four criteria: (1) data must include a fitness metric, including survival, fecundity or egg/seed size, or abundance or recruitment; correlates of fitness such as growth or body size were rejected; (2) either data from a control without a conservation intervention (to measure relative fitness), or data from the population before conservation was implemented (to measure absolute fitness), must be available; (3) data from at least two different time periods after the conservation intervention must be available; (4) sample size, and a metric of sampling variance (standard error or deviation, or confidence intervals) for the fitness metrics must be reported. A total of 15 articles on a total of 15 species were selected based on these four criteria. The species covered a wide taxonomic breadth, including yeast, plants, invertebrate animals, and vertebrate animals. A total of 95 entries were included, with multiple entries from most studies. The type of conservation intervention (i.e., the conservation “strategy”) was gleaned from study abstracts. We also recorded whether the experimental conditions imposed were stressful (40 entries, experimental stress treatment, e.g., high salt or heat shock) or benign (55 entries, control treatment, e.g., low salt or no heat shock, or no implicit stress treatment imposed).

Table S1 database effect sizes.csv

Table S2- Meta-analysis: larger dataset of conservation strategies based on categorical data (fitness improved, declined, or did not change)

Using a larger dataset than Table S1 that also included studies without sampling variability or sample size, we were table to extract data from 35 studies on a total of 33 species. This dataset included all 95 entries from the smaller dataset described above, and an additional 35 entries, for a grand total of 130 entries. For each entry, we recorded whether the longer time period increased or decreased fitness relative to the short time period, using the equation: (w2 – w0 or wcontrol2) – (w1 – w0 or wcontrol1) = difference This difference was recorded as “positive”, “negative” or 0. This was done for each of the 130 entries. We then created a collapsed database, with only one entry per combination of study, species, and conservation strategy (noting that some studies included multiple species or strategies). If all entries for a study or species or strategy had differences (as above) that were positive, we assigned that entry as “positive”, and similarly for differences that were all “negative”. If some entries were positive and others were negative for a given study/species/strategy, we assigned that entry a 0, meaning we were unable to conclusively determine whether the conservation implementation was beneficial over the long term.

Table S2.csv