Main

At one time it was presumed from the vastness of the oceans that fishing would not drive species to extinction. There have, however, been recent sharp declines in the numbers of oceanic cod, sharks, rays, tuna, marlins, swordfish and sea turtles3,4,5,6,7. As the shelf fisheries in the northwest Atlantic began to collapse in the 1960s and 1970s, harvesting shifted to deep-sea fish species8, but many populations crashed within ten years because their recovery is so slow2,9.

Deep-sea fish are highly vulnerable to disturbance because of their late maturation, extreme longevity, low fecundity and slow growth2,9. Some deep-sea fish form spawning aggregations on seamounts and the sea floor, and this increases their susceptibility to overfishing2. Survey data collected over extended periods are limited so it has been difficult to determine the effects of deep-sea fishing on both target and by-catch species.

For our analysis, we chose five species that live on or near the bottom of the North Atlantic Ocean, on the continental slope. They range from the common to the rare: roundnose grenadier, Coryphaenoides rupestris; onion-eye grenadier, Macrourus berglax; blue hake, Antimora rostrata; spiny eel, Notacanthus chemnitzi; and spinytail skate, Bathyraja spinicauda. The species evaluated can live to 60 years of age, grow to more than 1 m in length, and mature in their late teens. C. rupestris and M. berglax have been commercially fished, and all five are taken as by-catch in fisheries that target Greenland halibut, Reinhardtius hippoglossoides, and redfish, Sebastes spp. None was taken in any substantial number, even as by-catch, before the 1970s.

We used catch data from standardized, research-trawl surveys in the Canadian waters of the northwest Atlantic Ocean over 1978–94 to determine declines in relative abundance and individual mean size (for details, see supplementary information). All species declined in relative abundance (Fig. 1): declines over the 17-year period were 87–98% and declines estimated for three generations, the IUCN benchmark, were 99–100% (see supplementary information). Survey data for an additional period (1995–2003) were obtained for C. rupestris and M. berglax. The overall declines in relative abundance for these two species over the 26-year period were 99.6% and 93.3%, respectively; estimated declines over three generations were 100% and 99.7%, respectively (see supplementary information).

Figure 1: Trends in relative abundance of five species of deep-sea fish.
figure 1

Weighted relative abundance (number per tow) over time from research-survey data, showing the estimated exponential decline (red line) and 95% confidence projections of the estimate (dashed lines) for five deep-sea species in the Canadian waters of the northwest Atlantic, 1978–94.

Full size image

According to the IUCN criteria, these five species of deep-sea fish qualify as critically endangered in the northwest Atlantic. The declines occurred on a timescale equal to, or slightly less than, a single generation of these species. All of the species apart from N. chemnitzi also declined by 25–57% in mean size over the 17-year period (see supplementary information). The survey data are not adequate for full assessment of the situation for other deep-sea fish species that may also be at risk. The largest deepwater skate in the northwest Atlantic — the barndoor skate Dipturus laevis — was driven unnoticed almost to extinction6.

Scientific investigation lags behind the collapse of deep-sea fisheries8,9. Conservation measures are necessary and lack of knowledge must not delay appropriate initiatives, including the establishment of deep-sea protected areas.