Last author

Harvested fish populations can fluctuate wildly compared with those left untouched by humans, although it is not clear to what extent harvesting causes these fluctuations. This makes fish-stock management a contentious commercial and ecological issue. On page 835, George Sugihara at the Scripps Institution of Oceanography in La Jolla, California, and his colleagues explain how fishing magnifies population fluctuations. Sugihara tells Nature that the very foundation of the prevailing fishing-management theory is wrong.

Why has it been so hard to prove that harvesting destabilizes species abundances?

Historically, we've only had species-abundance data for exploited species, which came from fishery records. We really needed data on the unexploited species to serve as a control to rule out environmental causes of population collapse. The state started keeping better records of all species after the crash of the California sardine fishery in the late 1940s, because it wasn't clear whether ocean climate conditions or fishing caused sardine declines.

Current fish-stock management practices rely on maintaining specific biomass targets. Why is that wrong?

Biomass doesn't provide any indication of individual fish sizes. Picture a pond with either a single 200-kilogram fish or 200 1-kg fish; they have the same biomass but respond differently to environmentally controlled fluctuations in food supply. Depending on food availability, the single fish may or may not grow. With food, the smaller fish not only grow but also reproduce, potentially outstripping available resources and causing a population crash — so boom-and-bust cycles occur. If a management policy does not forecast these fluctuations, when abundance is high it will respond by adjusting next year's harvest targets upwards, just when the population might be poised to crash on its own. This further destabilizes fish populations.

What should fish catches be based on?

Fish populations do not exist in equilibrium — they are unstable systems that depend on ever-changing environmental conditions. Models determining harvest limits should account for the average sizes of individuals as well as how environmental conditions affect fish abundance. We need to make it clear to fisheries managers that quantitative tools exist to significantly improve fisheries' forecasting. Greater accountability is also needed. As it stands, fisheries managers are not held accountable for making inaccurate predictions — and can be wrong for generations.