It becomes very clear from analysis of fisheries landings (e.g. Caddy and Gulland, 1983; Spencer and Collie, 1997) that fisheries are very dynamic and that recruitment, and hence biomass, react in a sensitive way to environmental changes and fishery impacts. Some oscillations in production are therefore probably inevitable for most fisheries. A fishery management system that is aimed at harvesting a stable yield year after year will probably have to reduce target mortalities to below F0.1, or even lower. In this respect, fisheries for oceanic resources such as tunas may have some advantages in relation to shelf demersal fisheries. This follows from the nature of the fishing process; considerable expenditures of effort, capital and time necessary for offshore operation may make it difficult in open ocean areas for most resource populations to be driven to very low levels without uneconomic expenditures of time and effort. This would suggest that fisheries economics, as well as simple biological criteria, could provide a starting point for the search for RP's, such as fMEY, the fishing effort corresponding to Maximum Economic Yield.
Empirical criteria could be useful here and may even be framed in terms of a series of arbitrary ratios, agreed minima for which might trigger a reduction in harvesting intensity. Some ideas here (especially in the case of fisheries with observer systems or satellite surveillance systems) might be to set minima for one or other of the following ratios:
(1) School sightings/Distance steamed on the fishing grounds(2) Number of sets/Distance steamed on the fishing grounds
or expressed in energetic terms:
(3) (Tons harvested/ton fuel consumed)
or based on easily available catch statistics, e.g.:
(4) % mature fish in the catch(5) [mean size in the catch]/[size at 50% maturity]
Any of these indices (or all in combination, see later) might be expected to decline as the stock declines in abundance, and any of them could be used to decide on limiting values that could lead to a reduction in fishing effort until the index in question has shown signs of a return to 'normal' conditions. Determining biomasses for oceanic and most developing country fisheries also poses significant problems. These quantities may only be derived indirectly or under certain assumptions from production model or other analyses, hence indices such as the above may be reasonable proxies for abundance.
The question of 'cryptic biomass' (e.g. Fonteneau and Soubrier, 1996) has been raised, which implies that only a fraction of the tuna biomass is available in accessible areas at attractive densities to support a fishery. This 'refugium' concept in fact provides a useful management buffer but also makes biomass-based indices or CPUE somewhat doubtful as indices of total abundance, though these may be relevant indices for local stock size.
