Dealing With the Complexity of Effective Population Size in Conservation Practice

Abstract

Effective population size (Ne) is one of the most important parameters in evolutionary biology, as it is linked to the long-term survival capability of species. Therefore, Ne greatly interests conservation geneticists, but it is also very relevant to policymakers, managers, and conservation practitioners. Molecular methods to estimate Ne rely on various assumptions, including no immigration, panmixia, random sampling, absence of spatial genetic structure, and/or mutation-drift equilibrium. Species are, however, often characterized by fragmented populations under changing environmental conditions and anthropogenic pressure. Therefore, the estimation methods' assumptions are seldom addressed and rarely met, possibly leading to biased and inaccurate Ne estimates. To address the challenges associated with estimating Ne for conservation purposes, the COST Action 18134, Genomic Biodiversity Knowledge for Resilient Ecosystems (G-BiKE), organized an international workshop that met in August 2022 in Brașov, Romania. The overarching goal was to operationalize the current knowledge of Ne estimation methods for conservation practitioners and decision-makers. We set out to identify datasets to evaluate the sensitivity of Ne estimation methods to violations of underlying assumptions and to develop data analysis strategies that addressed pressing issues in biodiversity monitoring and conservation. Referring to a comprehensive body of scientific work on Ne, this meeting report is not intended to be exhaustive but rather to present approaches, workshop findings, and a collection of papers that serve as fruits of those efforts. We aimed to provide insights and opportunities to help bridge the gap between scientific research and conservation practice.

Keywords: Kunming‐Montreal global biodiversity framework; Ne; biodiversity monitoring; bridging science‐to‐application gap; effective number of breeders; genetic diversity; genetic indicators; species conservation and management.

FIGURE 1

Schematic representation of a metapopulation evolving through time as an example, intended to highlight the possible ambiguity of Ne estimations. The X‐axis indicates populations' spatial distribution (sites) along one spatial dimension, for simplicity. As we go up and forward in time (Y‐axis), subpopulations disappear, and others are formed, but the metapopulation as a whole is maintained. (A) Sampling a single subpopulation (IIb) in the present (t0) and applying different Ne estimation methods may result in vastly different Ne estimates representing different aspects of the effective size, which are all commonly called “Ne”. Disambiguation of these different meanings is essential in conservation. Depending on the approach, one can estimate from the same sample local contemporary Nex (e.g., using linkage disequilibrium, kinship, or a temporal method), contemporary NeMeta (when based on heterozygosity (He) decay across time; this requires two samples across time; pink), coalescent Ne (when based on its current He, and assuming mutation‐drift equilibrium; chartreuse), NeMeta at different times in the past (blue, seagreen), but never past Nex. (B) Methods that estimate recent Ne trajectories (0–200 generations ago) will initially reflect local Ne, but will increasingly reflect metapopulation Ne, and samples taken in different subpopulations but with origins in the same metapopulation will eventually converge on the same NeMeta, which is the sum of the past Nex (here at t ‐ 2). The risk is that this is interpreted as a population decline, whereas it represents a confounding effect of spatial scale (Novo et al. 2023). (C) A landscape‐level schematic depiction of the processes occurring in (A) and (B).

FIGURE 2

Participants of the international workshop, funded by the COST Action 18134, Genomic Biodiversity Knowledge for Resilient Ecosystems (G‐BiKE), August 2022 in Brașov, Romania. Participants who joined virtually are not shown.

FIGURE 3

A schematic overview to evaluate various methods used to estimate the Ne by incorporating different techniques and underlying assumptions: The two approaches involved in parameter manipulation during the workshop were constructing hypothetical datasets using simulations to test a range of alternative scenarios on Ne estimation (scenario A) and manipulation of existing empirical datasets to mimic some of the likely biases that may occur (scenario B). Genetic markers such as short tandem repeats (STRs or microsatellites) or single nucleotide polymorphisms (SNPs) are used independently or in combination. Ne estimation involves different methods with divergent assumptions, including LD‐based approaches and kinship analysis. Similar to the first scenario (A), these methods may use STR, SNP, or a blend of both. Finally, Ne estimation strategies may involve consistent methodologies with variations in including or excluding spatial and/or temporal samples. These variations allow for an evaluation of how sample selection influences the accuracy of Ne estimation. These analyses may use STR, SNP, or a combination of genetic markers.

FIGURE 4

A general representation of the complexity of estimating effective population size (Ne). The figure presented is a conceptual model illustrating the complex framework used to assess genetic indicators in the context of conservation genetics. It details the interactions between the different levels of analysis and the factors that influence the assessment of genetic diversity in various species. At the highest level, ‘spatial genetic clustering of populations’ is highlighted, distinguishing between isolated populations, metapopulations, and continuous populations. This classification is key to understanding gene flow and genetic structure, which are fundamental to conservation strategies. At the middle level, the “life history level” includes aspects such as demographic history, generation time, reproductive strategies, differences in reproductive success, and hybridization. These elements influence the effective population size and the Ne/Nc ratio, which are essential parameters in the study of population genetics; this life‐history data primarily provides insights into local Ne. Genetic markers and time‐series data are essential for comprehending the structure and dynamics of populations at a spatial scale, yet they have the potential to introduce biases that could diminish the effectiveness of the analysis. The framework underscores the necessity of synthesizing data from genetic markers with an in‐depth understanding of species' life histories. Additionally, it highlights the imperative to secure the availability of extensive genetic data, which is critical to informing effective conservation strategies.