Report of the Workshop on the Decline and Recovery of cod Stocks throughout the North Atlantic, including trophodynamic effects (WKDRCS)

Atlantic cod (Gadus morhua) stocks respond to long-term climate changes, such as the warming of the North Atlantic during the 1920s and 1930s, when cod increased rapidly in abundance off West Greenland and spread far to the north. At the same time there was increased recruitment at Iceland and increased abundance and northward expansion in the Barents Sea. By the time that the waters at West Greenland cooled in the late 1960s, the cod stock biomass had declined greatly from its peak in 1949. Both climate and the fishery contributed to the subsequent collapse of the stock, but it is not possible to make a quantitative attribution and the factors interact.  In this and other cases the effective environmental factors include plankton production and other ecosystem effects.  These factors often co-vary with temperature change, making it difficult to separate them from direct effects of temperature on growth, survival and recruitment.  
Cod have been subjected to changes in climate and fishing intensity for centuries, but detailed information on declines and recoveries comes mainly from the past 30–40 years, which is a short time span relative to many natural phenomena. All stocks, with the exception of the Celtic Sea, have suffered prolonged periods of decline since 1970. Comparison between NW and NE Atlantic stocks reveals two major differences: (i) most NW Atlantic stocks share a pattern of increase and decline in biomass, whereas the NE Atlantic stocks do not, and (ii) although fishing mortality is generally higher on NE than on the NW Atlantic cod stocks, the declines in biomass were much greater in the NW Atlantic than in the NE. Directed fishing was halted during the early 1990s for all NW Atlantic shelf stocks from the eastern Scotian Shelf northward. All these stocks have since been characterized by low productivity, and several have shown no sign of recovery after more than a decade without directed fishing. The NW Atlantic stocks from the eastern Scotian Shelf northward inhabit areas with average temperatures below 4oC and in all of them the mean weight-at-age began a period of decline before the biomass declined.  The NE Atlantic stocks all inhabit areas with average temperatures above 4oC and showed less variability in mean weight-at-age.
Both fishing and climate are implicated in the declines in cod stock biomass since 1970.  In the NW Atlantic the fishing mortality increased until moratoria were imposed in the early 1990s.  The decline in biomass was caused by fishing, but changes in the productivity of the stocks contributed to the collapse and there is good evidence that the decline in biomass also caused fishing mortality to increase. Fisheries management must be sensitive to possible changes in stock productivity and must either respond quickly, to prevent increased mortality and further stock decline, or regulate fishing in a precautionary way, which is robust to uncertainties about stock productivity.   
Changes in weight-at-age are an important component of the variation in productivity of coldwater cod stocks. Variation in weight-at-age appears to be mainly due to changes in the environment. In the S. Gulf of St. Lawrence density-dependent growth and changes in the direction of size selective fishing mortality appear to be the most important factors and here size-at-age has remained low despite good conditions for growth and low fishing mortality. Age and size at maturity have declined in many stocks and there appears to be a genetic component to this change, in response to fishing (where it has been investigated – Arcto-Norwegian cod, S. Labrador, S. Gulf of St Lawrence).  Early maturity gives a selective advantage under most high mortality regimes, but reduces population productivity if fishing mortality is reduced. The reversion to older ages and larger sizes at maturity will be slow if additive genetic variance has been depleted.
The risk of stock collapse increases when stock productivity declines. Some of the life-history characteristics (growth and maturation in particular) governing productivity can be monitored by sampling commercial and research catches and may give timely indications of changes in productivity and risk of collapse.  In order to develop their routine use in assessing risk of collapse under different fisheries management strategies, indicators of possible change in productivity (weight-at-age, condition, liver index, maturation reaction norms) should be investigated using tropho-dynamic, life history and risk assessment models.  
Mean age and age diversity of spawners (and SSB) declined in many stocks in response to fishing. In many (but not all) stocks, this has resulted in a decline in recruitment rate. In Arcto-Norwegian and Icelandic cod resilience to climate change has been shown to decrease as mean age of spawners declined.
For all cod stocks, the kinds of prey and their abundance and availability vary over time. The boreal ecosystems and the Baltic Sea tend to have a narrower field of potential prey than the more southern ecosystems, and changes in the abundance or distribution of major forage species (e.g. capelin, herring) might cause food shortages for cod. This could lead to declines in condition and consequent reductions in reproductive output and even survival. Declines in prey availability have been implicated in declines in cod productivity that have lasted from one to several years, but such variability in prey has seldom been implicated as a major factor in cod stock declines. There have been suggestions that low abundance of prey may be impeding stock recovery in some areas, such as the offshore of eastern Newfoundland.
When cod stocks decline to very low abundance, the relative importance of factors governing dynamics and productivity can change. When a stock is relatively large, it may be able to sustain predation and maintain itself at relatively high abundance even when subjected to a fishery. However, if the stock has declined in abundance, for whatever reason, and predator populations have not declined, or may even have increased, then high predation mortality may impede or prevent recovery. Such impacts may occur via predation by pelagic fish on eggs and larvae of cod, as has been hypothesized for cod in the Baltic Sea, on the eastern Scotian Shelf and in the southern Gulf of St. Lawrence. It may also occur via predation on juvenile cod and perhaps even adult cod by larger predators such as seals, as has been hypothesized for the eastern Scotian Shelf, the southern and northern Gulf of St. Lawrence, and eastern Newfoundland.
In some stocks e.g. southern Gulf of St Lawrence, high natural mortality has replaced high fishing mortality, preventing recovery even when fishing pressure is low.  No substantial increase in the biomass of Baltic cod can be expected without a change in environmental conditions favouring better recruitment (even at FPA, which is 65% of the current fishing mortality).
A general conclusion from the experience off eastern Canada is that humans may have limited ability to “rebuild” cod stocks that have declined to very low levels. Simply turning off directed fishing may be insufficient to promote recovery. The properties of the stocks themselves and the state of the ecosystems in which the cod are embedded may be such that the stocks remain constrained to their new levels of low abundance for a considerable time.