Translocations spur population growth but fail to prevent genetic erosion in imperiled Florida Scrub-Jays

Abstract

Land and natural resource use in addition to climate change can restrict populations to degraded and fragmented habitats, catalyzing extinction through the reinforced interplay of small population size and genetic decay. Translocating individuals is a powerful strategy for overcoming direct threats from human development and reconnecting isolated populations but is not without risks.¹ Habitat Management Plan analyses under section 7 of the U.S. Endangered Species Act determined that multiple subpopulations of Federally Threatened Florida Scrub-Jays (Aphelocoma coerulescens, hereafter FSJ) belonging to a metapopulation on Florida's west coast were declining demographic sinks, occupying areas where agriculture and fire suppression had degraded and fragmented the habitat.² In order to increase the viability of the overall metapopulation, 51 FSJs from five of these small subpopulations in areas to be mined were translocated throughout 2003-2010 into a larger site of more contiguous, recently restored habitat at the core of the metapopulation, which contained a small resident population.³ Prior to translocations and for nearly two decades afterward, this core population, referred to as the M4 core region (CR) population, was extensively monitored, yielding a nearly complete pedigree. We used this pedigree, along with temporal genomic analyses and simulations, to show that translocations coupled with habitat restoration generated rapid population growth, but high reproductive skew increased inbreeding and led to genetic erosion. This mechanistic understanding of mixed conservation outcomes highlights the importance of monitoring and the potential need for genetic rescue to offset consequences of reproductive skew following translocations, regardless of demographic recovery.

Keywords: Florida Scrub-Jay; conservation; fitness; genomics; inbreeding; pedigree; reproductive skew.

Figure 1.. Demographic response of the CR FSJ population to translocations

(A) Map showing M4 subpopulation sample sizes (n) for genetic analyses, with individuals depicted by colored shapes. For translocation donor subpopulations, we show the genetic sample size and total number individuals translocated from the site, in that order. Individuals were translocated into the Mosaic Wellfield (MW) site and then expanded throughout the core region (CR) outlined in red. The scale bar denotes geographic distances for the main map in kilometers (km). The geographic range of the entire M4 metapopulation is colored red in the inset map of the Florida peninsula. The time periods in which blood samples were taken from individuals for genetic analyses are displayed below the map. (B) Purple dots denoting family groups show the geographic expansion of the population throughout the CR (green area) over time. The MW site is outlined in black. (C) The solid black line shows the census size of the CR population over time with arrows showing the year and number of jays translocated into the MW site. Dashed lines indicate the number of translocated (blue) and resident (pink) founding lineages contributing to cohorts born in the CR over time. See also Figure S1.

Figure 2.. Expected genetic contributions by CR ancestors over time based on the population pedigree

(A) The total number of founding ancestors that had existed in the CR and could potentially contribute to the population is indicated by the top of the light gray shaded area, while the number of ancestors that were actually genetically represented in the population based on the pedigree is denoted by the top of the dark gray shaded area. The number of potential genetic contributors increased as jays were moved to the MW during the translocation period (2003–2010), with only a fraction contributing to the population over time. Lines show the expected proportion of the total founder genetic material originating from each of the 74 CR population ancestors over time. Genetic contributions were skewed toward a small subset of CR founders, with one breeding pair comprised of RLS-K (male, red) and WSA-K (female, blue) contributing the most (see also Table S1). Annotated arrows show the mate of RLS-K each year and the number of young that the respective pair fledged out of the total number of eggs laid per nest attempt (number fledged/total number of eggs). (B) The same expected genetic contribution by ancestors to the CR population is depicted by lines in (A) but shown as stacked bars (individuals are ordered based on ranked contribution to the 2022 population), highlighting genetic takeover by a subset of mostly translocated lineages. Sixteen founding individuals from which over 76% of the CR ancestry is expected to have derived since 2015 are labeled in (B) and denoted by the same colors in (A). The remaining ancestral lineages are colored yellow (B) or gray (lines in A; bars in B). Figures S1 and S2A–S2D show how the observed variance in ancestral genetic contributions shown here compares to levels expected from demographic stochasticity.

Figure 3.. Spatiotemporal genomic characterization of the CR and neighboring M4 subpopulations

(A) Genetic similarity among sequenced individuals based on a principal-component analysis (PCA) of genome-wide variation in our M4 sample. The percentage of the total genetic variation explained by each PC is indicated on each axis. Sequenced founding ancestors among the top 10 largest genetic contributors to the contemporary population based on the pedigree are enclosed in dashed circles and denoted by their color-band identifiers and the rank of their pedigree-based genetic contribution to the 2022 population. “+” denotes a donor site individual that was not translocated and not a member of the CR population. (B and C) Comparison of the distributions of individual inbreeding coefficients (F) and heterozygosity (H) for all sequenced individuals from different groups. Colors and shapes denoting individuals are the same as in (A). Significant (p < 0.05) group differences in F and H from tests performed on subsets of individuals for which all pairwise relatedness was below 0.4 are indicated by an asterisk (*). No statistical tests were performed involving the “other M4” group. The same highest-ranked contributors to the contemporary CR population based on the pedigree circled in (A) are also circled in the plots of F and H. See Figure S3 for distributions of individual inbreeding estimated using different approaches and Table S2 for group-level comparisons of genetic diversity based on nucleotide diversity $(\text{π})$.

Figure 4.. Characterization of reproductive skew from genomic sequencing data

(A) The distribution of the average rank-weighted relatedness statistic, $K$, for sequenced CR population founders highlights large differences in the genetic contribution by ancestors to the 2021 CR cohort. (B) A comparison between the characterization of ancestral genetic contributions to the 2021 cohort using the pedigree with all 74 ancestors versus the genetic $K$ statistic calculated for 39 sequenced ancestors shows high concordance between the two approaches for inferring reproductive skew. Resident versus translocated origin of individuals is indicated by cyan and orange colors, respectively, and the size of points is scaled to the maximum individual genetic contribution in the respective analysis. Lines connect individuals common to both analyses and are black if the individual contributed to the 2021 CR cohort according to the pedigree. The identity of the top 10 contributing jays from each analysis is displayed next to their respective points, with individuals common to both lists in black. See also Figure S2E.