The unique ecology of human predators

See allHide authors and affiliations

Science  21 Aug 2015:
Vol. 349, Issue 6250, pp. 858-860
DOI: 10.1126/science.aac4249

An anomalous and unbalanced predator

In the past century, humans have become the dominant predator across many systems. The species that we target are thus far in considerable decline; however, predators in the wild generally achieve a balance with their prey populations such that both persist. Darimont et al. found several specific differences between how humans and other predatory species target prey populations (see the Perspective by Worm). In marine environments, for example, we regularly prey on other predator species. These differences may contribute to our much larger ecological impact when compared with other predators.

Science, this issue p. 858; see also p. 784


Paradigms of sustainable exploitation focus on population dynamics of prey and yields to humanity but ignore the behavior of humans as predators. We compared patterns of predation by contemporary hunters and fishers with those of other predators that compete over shared prey (terrestrial mammals and marine fishes). Our global survey (2125 estimates of annual finite exploitation rate) revealed that humans kill adult prey, the reproductive capital of populations, at much higher median rates than other predators (up to 14 times higher), with particularly intense exploitation of terrestrial carnivores and fishes. Given this competitive dominance, impacts on predators, and other unique predatory behavior, we suggest that humans function as an unsustainable “super predator,” which—unless additionally constrained by managers—will continue to alter ecological and evolutionary processes globally.

Humans have diverged from other predators in behavior and influence. Geographic expansion, exploitation of naïve prey, killing technology, symbioses with dogs, and rapid population growth, among other factors, have long imposed profound impacts—including widespread extinction and restructuring of food webs and ecosystems—in terrestrial and marine systems (13). Despite contributions from the “sustainable exploitation” paradigm (4), contemporary humans can rapidly drive prey declines (57), degrade ecosystems (8, 9), and impose evolutionary change in prey (10, 11). Owing to long-term coevolutionary relationships that generally limit exploitation rates, especially on adult prey, these are extreme outcomes that nonhuman predators seldom impose. Meanwhile, whether present and future exploitation can be considered sustainable is hotly contested, especially in fisheries. Debate has been largely restricted to elements of the sustainable exploitation model, namely, a model of prey abundance and yields to humanity (e.g., 12, 13).

Here, we approach the notion of sustainable exploitation differently by asking whether humans—extreme in their impacts—are extreme in their predatory behavior (14, 15). Previous work has variously estimated exploitation by humans, nonhuman predators, or both, but systematic comparisons have focused on specific taxa or regions, have lumped all predators together, have been reconstructed indirectly, and/or did not include age classes (e.g., 14, 16, 17). We address these limitations with data spanning wildlife, tropical wild meat, and fisheries systems (data files S1 and S2). We examine variation in annual finite exploitation rates of marine fishes from every ocean (n = 1494 estimates, 282 species from 110 communities) and terrestrial mammals from every continent except Antarctica (631 estimates, 117 species from 179 communities) (fig. S1 and tables S1 and S2) by predator type (humans versus nonhuman), ecosystem (marine versus terrestrial), region, and trophic level. We focus on adult prey because hunters and fishers overwhelmingly target adults (18). We complement this quantitative assessment by identifying additionally unique predatory behaviors by humans that (i) facilitate the large differences in exploitation rates we detect and (ii) elicit the manifold consequences of humanity’s predatory hegemony.

Differences in exploitation rates between hunters and terrestrial predators varied among comparisons. Globally and pooled across trophic levels, exploitation rates by hunters (median = 0.06) did not differ from those of carnivores [median = 0.05; Wilcoxon test W = 46076, Padj(2) = 0.11] (Fig. 1A and figs. S2A and S3A). A paired comparison over shared prey within the same community, however, revealed that hunters exploit at higher rates than the highest-exploiting terrestrial predator [paired Wilcoxon test V = 929, Padj(2) = 0.03] (fig. S3B). Additionally, a similar paired comparison showed that the median proportion of mortality (an independent metric) caused by hunters (0.35) was 1.9 times that (0.19) caused by all other predators combined (paired Wilcoxon test V = 1605, P = 0.004) (Fig. 1B).

Fig. 1 Patterns of exploitation by human and nonhuman predators on adult prey.

(A) Complementary cumulative distribution functions showing the probability of predators exploiting prey at a rate (R) greater than or equal to a given annual finite exploitation rate (r), on the basis of the number of available individuals in populations (terrestrial mammals) or biomass (marine fishes). (B) Proportion of annual mortality caused by hunters and all other (i.e., aggregated) terrestrial predators consuming the same prey population. (C and D) Exploitation rates of human and nonhuman predators across trophic levels in (C) terrestrial and (D) marine systems. Whiskers represent distance from upper and lower quartiles to largest and smallest nonoutliers. [Art by T. Saxby, K. Kraeer, L. Van Essen-Fishman/ and K. Eberlins/]

Trophic level and regional analyses (across taxa and areas with abundant data) revealed additional patterns. Although globally pooled comparisons showed that hunters and terrestrial predators exploited herbivores (artiodactyls) at similar rates [W = 14751, Padj(9) = 1.00] (Fig. 1C), hunters in North America and Europe exploited herbivores at median rates 7.2 and 12.5 times those of hunters in Africa [both Padj(9) < 0.04]; rates did not differ statistically between hunters and terrestrial predators within any of the regions (fig. S4A). Globally, hunters exploited mesocarnivores [W = 248, Padj(9) = 0.03] and large carnivores [W = 181, Padj(9) < 0.001] at higher rates than nonhuman predators by factors of 4.3 and 9.2, respectively (Fig. 1C). Remarkably, hunters exploited large carnivores at 3.7 times the rate that they killed herbivores [W = 2697, Padj(9) < 0.001] (Fig. 1C).

Fisheries exploited adult prey at higher rates than any other of the planet’s predators (Fig. 1A and fig. S2B). Among nonhuman predators across all oceans, 50% of exploitation rates were less than 1% of annual adult biomass. In contrast, fisheries exploited more than 10% of adult biomass in 62% of cases. Overall, the median fishing rate (0.14) was 14.1 times the take (0.01) by marine predators [W = 83614, Padj(2) < 0.001] (fig. S3A). In paired comparisons, median fisheries exploitation (0.17) was 3.1 times the median rate (0.06) by the highest exploiting marine predator of the same prey [V = 382, Padj(2) = 0.02] (fig. S3B). At all trophic levels, humans killed fishes at higher rates than marine predators [all Padj(9) < 0.04] (Fig. 1D), but there were no differences in take by each predator across trophic levels [all Padj(9) ≥ 0.5]. Pooling all trophic levels, the median rate of Atlantic fisheries exploitation (0.20) was 2.9 times that of Pacific fisheries [median = 0.07, W = 6633, Padj(4) < 0.001] (fig. S4B).

Although our varied data set could impose biases in both directions (supplementary text), we reveal striking differences in exploitation rates between nonhuman predators and contemporary humans, particularly fishers and carnivore hunters. Interactions between human and natural systems likely underlie patterns. For example, global seafood markets, industrial processing, relatively high fecundity among fishes, and schooling behavior could, in part, explain the particularly high fisheries take, whereas gape limitation by piscivores and a generally species-rich marine environment might explain why marine predator rates are comparatively low. Higher human densities and reduced fish biomass (from longer exploitation) likely explain higher fishing rates in the Atlantic versus Pacific oceans. Moreover, motivations to kill typically inedible carnivores for trophy and competitive reasons [intraguild predation; (7)] are evidently powerful and drive acutely high rates. Although, in terms of numbers, it is easy to exploit high proportions of (less abundant) carnivore populations, the implications remain profound (below). In addition, whereas declines in tropical wild meat (5) might predict an opposite pattern, lower hunting rates of African herbivores could relate to simpler technology, less reporting, and/or longer adaptation to human predation.

Whereas sociopolitical factors can explain why humans repeatedly overexploit (19), cultural and technological dimensions can explain how. Human predatory behavior evolved much faster than competing predators and the defensive adaptations of prey (20). Indeed, division of labor, global trade systems, and dedicated recreational pursuit have equipped highly specialized individuals with advanced killing technology and fossil fuel subsidy that essentially obviate energetically expensive and formerly dangerous search, pursuit, and capture. Moreover, agri- and aquaculture, as well as an ever-increasing taxonomic and geographic niche, leave an enormous and rapidly growing human population demographically decoupled from dwindling prey. In fact, low prey abundance can drive aggressive exploitation, because of the increased economic value of rare resources (21).

Emerging evidence suggests that the consequences of dominating adult prey are considerable. For example, human preference for large ornaments and/or large body size has fundamentally altered the selective landscape for many vertebrates. Not only can this rapidly alter morphological and life-history phenotypes (11), the resulting changes can modify the reproductive potential of populations (22) and ecological interactions within food webs [e.g., (23)]. In addition, owing to different behavior (e.g., age-class preferences and seasonality of exploitation), hunters likely cannot substitute for carnivores as providers of ecological services [e.g., regulation of disease and wildfire (7, 9), as well as mesopredator control (8, 24)]. Finally, less explored is the potentially substantial impact of prey biomass removal from ecosystems; global trade and sanitation systems shunt energy and nutrients from food webs of provenance to distant landfills and sewers.

These implications, the high exploitation rates that drive them, and the broadest taxonomic niche of any consumer uniquely define humans as a global “super predator.” Clearly, nonhuman predators influence prey availability to humans [e.g., (25)]. But overwhelmingly these consumers target juveniles (18), the reproductive “interest” of populations. In contrast, humans—released from limits other predators encounter—exploit the “capital” (adults) at exceptionally high rates. The implications that can result are now increasingly costly to humanity (26) and add new urgency to reconsidering the concept of sustainable exploitation.

Transformation requires imposing limits of humanity’s own design: cultural, economic, and institutional changes as pronounced and widespread as those that provided the advantages humans developed over prey and competitors. This includes, for example, cultivating tolerance for carnivores (7), designing catch-share programs (27), and supporting community leadership in fisheries (28). Also key could be a new definition of sustainable exploitation that focuses not on yields to humanity but rather emulates the behavior of other predators (14). Cultural, economic, and technological factors would make targeting juvenile prey challenging in many cases. Aligning exploitation rates on adults with those of competing predators, however, would provide management options between status quo exploitation and moratoria. Recent approaches to resolve controversies among fisheries scientists reveal how distant such predator-inspired management prescriptions are now. For example, although the mean “conservative” fishing rate estimated to rebuild multispecies fisheries across 10 ecosystems (0.04) is one-fourth their maximum sustainable yield rates (0.16) (13), it remains 4 times the median value we estimated among marine predators globally (0.01). Consequently, more aggressive reductions in exploitation are required to mimic nonhuman predators, which represent long-term models of sustainability (14).

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S6

Tables S1 to S2

References (29, 30)

References and Notes

  1. Materials and methods are available as supplementary materials on Science Online.
  2. Acknowledgments: We thank M. Arseneau, L. Grant, H. Kobluk, J. Nelson, and S. Leaver for data collection; L. Reshitnyk for creating fig. S1; and P. Ehlers and J. Ehlers for statistical assistance. S. Anderson, J. Baum, T. Branch, J. Brashares, A. Calestagne, S. Carlson, T. Davies, D. Kramer, T. Levi, J. Reynolds, and the “Ecology@UVic” discussion group offered insight on drafts. We thank the Raincoast Conservation, Tula, Wilburforce, and Willowgrove Foundations. C.T.D. and T.E.R. acknowledge Natural Sciences and Engineering Research Council of Canada Discovery Grant 435683 and National Research Council Canada Operating Grant 2354, respectively. Data and R code available in Dryad (doi:10.5061/dryad.238b2). T.E.R. conceived of the idea and created the preliminary data set. C.T.D., H.M.B., C.H.F., and T.E.R. designed the research. C.T.D. led data collection and project management. H.M.B., C.T.D., and C.H.F. conducted analyses. C.T.D., C.H.F., H.M.B., and T.E.R. wrote the manuscript.
View Abstract

Navigate This Article