Résumé
Two major ecological transitions marked the history of the Black Sea after the last Ice Age. The first was the postglacial transition from a brackish-water to a marine ecosystem dominated by porpoises and dolphins once this basin was reconnected back to the Mediterranean Sea (ca. 8,000 y B.P.). The second occurred during the past decades, when overfishing and hunting activities brought these predators close to extinction, having a deep impact on the structure and dynamics of the ecosystem. Estimating the extent of this decimation is essential for characterizing this ecosystem’s dynamics and for formulating restoration plans. However, this extent is poorly documented in historical records. We addressed this issue for one of the main Black Sea predators, the harbor porpoise, using a population genetics approach. Analyzing its genetic diversity using an approximate Bayesian computation approach, we show that only a demographic expansion (at most 5,000 y ago) followed by a contemporaneous population collapse can explain the observed genetic data. We demonstrate that both the postglacial settlement of harbor porpoises in the Black Sea and the recent anthropogenic activities have left a clear footprint on their genetic diversity. Specifically, we infer a strong population reduction (∼90%) that occurred within the past 5 decades, which can therefore clearly be related to the recent massive killing of small cetaceans and to the continuing incidental catches in commercial fisheries. Our study thus provides a quantitative assessment of these demographically catastrophic events, also showing that two separate historical events can be inferred from contemporary genetic data.
Author SummaryFig. P1.(A) Samples of harbor porpoises, on which genetic analyses have been conducted, stranded along the coasts of the Black Sea and Aegean Sea (red dots). Inset shows area of the main map. (B) Demographic model (of 15 tested) best explaining the observed genetic variation. This expansion-decline model includes five independent parameters as follows: the time of the population foundation (TFo); the time of the population collapse (TCo); and the effective population size at the time of the postglacial foundation (NFo), after the foundation but before the population collapse (NBCo), and after the population collapse (NACo). (C) Posterior probabilities estimated under the expansion-decline model for the time at which the postglacial expansion occurred (TFo) and for the magnitude of the population size change (NFo/NBCo) that occurred after the population settled in the Black Sea (green). The probability distributions for the time at which the population recently collapsed (TCo) and the magnitude of the population collapse (NACo/NBCo) are provided in red.A general question in conservation biology, ecology, and evolutionary biology is whether population genetic approaches can detect and quantify human-driven collapses in endangered species and disentangle contemporaneous processes from more ancient ones. Our present study provides an example of how coalescent-based ABC population genetic approaches can provide important demographic information that cannot be obtained from historical and ecological modeling studies. It emphasizes that such genetic approaches will be even more valuable for species for which such historical or ecological data are not available.We screened for genetic polymorphisms of nuclear and mitochondrial DNA for 89 harbor porpoises (Fig. P1A). To determine the demographic events best explained by the observed genetic patterns, we conducted massive computer simulations under 15 alternative demographic scenarios of population evolution and analyzed the results under a flexible approximate Bayesian computation (ABC) statistical framework (5). These 15 scenarios differed in the types of changes in population size considered, such as stability, expansion, collapse, bottleneck, or expansion-decline, and by the timing of these changes, including a recent change within the past 100 generations vs. an ancient change between 100 and 1,000 generations. The ABC model choice procedure identified one class of scenarios that best explained the genetic pattern (Fig. P1B) with a probability greater than 0.99. This class of demographic scenarios consisted of a historical expansion of a small ancestral population (size NFo) to a large effective population size (NBCo), which then went through a severe population reduction within the past 100 generations to reach its current size (NACo) (Fig. P1B). We estimated that the old expansion was 15-fold and occurred within the past 5,000 y (Fig. P1C). This is compatible with the timing of reestablishment of marine conditions in the Black Sea after reconnection to the Mediterranean Sea, which allowed marine predators to recolonize this environment. More strikingly, we identified a strong genetic signal supporting a decline of ∼90% of the porpoise population in the Black Sea within the past 5 decades (Fig. P1C). This decline can clearly be related to the high mortality of small cetacean species from the 1960s to the 1990s, to the subsequent fishery collapse, and also to ongoing incidental by-catch in extant commercial fisheries.However, the magnitude of the anthropogenically induced population declines of dolphins and porpoises is, to the best of our knowledge, poorly documented, and pre- and postcollapse abundances remain undetermined (4). This information is crucial for elucidating the mechanisms behind the regime shifts observed in the Black Sea and for restoration plans. Because temporal variation in census numbers affects the genealogies of genes carried by these individuals, a powerful alternative to the analysis of historical records is the use of population genetics theory to infer past demographic history based on the observed distribution of genetic variation in natural populations (5). Here, we used such a population genetics analysis to demonstrate that postglacial colonization of the Black Sea and recent hunting activities have left clearly detectable footprints on the contemporary genetic diversity of populations of small cetaceans in the Black Sea. We focused here on one of the main apex predators in the Black Sea, the harbor porpoise (Phocoena phocoena relicta). This species provides an ideal model because it is among the species most strongly affected by anthropogenic activities and is the sole marine predator that is completely isolated in the Black Sea (4). Indeed, the closest populations of the same species are located in the North Atlantic. Therefore, from a population genetics perspective, any changes in neutral genetic diversity over time will only reflect changes in local population size, avoiding the frequently confounding effects of gene flow.The Black Sea (Fig. P1A) has had an extraordinary history over the past 18,000 y, marked first by a transition from a freshwater ecosystem during the last glacial period to a large marine ecosystem dominated by top predators, such as dolphins and porpoises, after reconnection to the Mediterranean Sea (ca. 8,000 y ago). During the past 5 decades, overfishing and hunting brought these predators close to extinction. Ecosystem modeling studies suggest that the decline of apex marine predators in the Black Sea from the 1960s to the 1990s was the key factor responsible for the loss of the resilience of its ecosystem, which triggered deep regime shifts that resulted in fishery stock collapses and intense eutrophication (1–3).The authors declare no conflict of interest.This article is a PNAS Direct Submission.Data deposition: Sequence and genotype data from this project are deposited at DRYAD (http://dx.doi.org/10.5061/dryad.km038) and at GenBank (accession nos. JX105486–JX105517).See Commentary on page 15078.See full research article on page E2569 of www.pnas.org.Cite this Author Summary as: PNAS 10.1073/pnas.1201258109.1M Llope, et al., Overfishing of top predators eroded the resilience of the Black Sea system regardless of the climate and anthropogenic conditions. Glob Change Biol 17, 1251–1265 (2011).2GM Daskalov, AN Grishin, S Rodionov, V Mihneva, Trophic cascades triggered by overfishing reveal possible mechanisms of ecosystem regime shifts. Proc Natl Acad Sci USA 104, 10518–10523 (2007).3GM Daskalov, Overfishing drives a trophic cascade in the Black Sea. Mar Ecol Prog Ser 225, 53–63 (2002).4AA Birkun, A Frantzis, Phocoena phocoena ssp. relicta. IUCN Red List of Threatened Species, ed IUCN, Version 2011.1. Available at www.iucnredlist.org. Accessed September 5, 2011. (2008).5MA Beaumont, Approximate Bayesian computation in evolution and ecology. Annu Rev Ecol Evol Syst 41, 379–406 (2010).