In Retrospect the Assumption of Sustainability for Atlantic Fisheries has Proved an Illusion

Fishery landings data series from 1970 to 2002 for the Northeast (FAO Area 27) and Northwest Atlantic (Area 21) with 73 and 42 species or species groups respectively, were mapped onto single charts as a diagnostic of the overall state of fisheries in these two regions, and could be appreciated simultaneously for the entire exploited suite of species as a series of contiguous bar charts. Charts were compared with that for a similar data set of 42 species or species groups from the Mediterranean and Black Seas (FAO Area 37). A “traffic light convention” partitioned the range of landings by species over the period into four equal intervals; coloured respectively red, yellow, green and blue between zero and the highest annual landing of the species. This allowed a model-free display of changes in the timing of species’ landings at the overall ecosystem level, as represented by the important commercial resources in the three areas. The year T₅₀ by which 50% of species landings had been harvested over the period, was considered a comparable measure of the response of individual species to exploitation. For a significant proportion of groundfish and pelagic fishes, the time to reach T₅₀ occurred early in the Atlantic time series. The hypothesis was tested that this measure of the depletion schedule is related to some aspect of species biology, as judged by data from the FISHBASE database. Except for commercial invertebrates where landings peaked late in the Northwest Atlantic time series, no significant regression was found between T₅₀ values per species and any biological characteristic of the species of finfish tested, suggesting that scheduling may reflect overcapacity and targeting by fisheries and/or regime changes. A decline in peak landings proceeded sequentially over time for the large proportion of all finfish species in the North Atlantic following a broadly similar trajectory. Possible exceptions were deeper water species where fisheries began later, suggesting that “fishing down the bathymetry” has occurred. A more synchronous “pulse” of high landings occurred in the Mediterranean and Black Seas in the mid-late 1980s, which was postulated as due to an increase in system productivity. Although regime changes and quota management cannot be excluded as partly responsible for the persistent low landings late in many time series, the main conclusion is that for all three areas, the effects of fishing overshadow those resulting from differing biological characteristics, habit, or species interactions. Although quota control may in part be responsible for low landings of some species late in the time series, the fishery management regimes that applied during 1970–2002 cannot be considered sustainable. This was confirmed by comparative analysis fitting a variety of models to the raw landing data. Of the mathematical models considered, the Hubbert curve, first used to predict the trajectory of extraction of non-renewable petroleum resources globally over time, best fitted the largest proportion of species time series. This seems to confirm that a hypothesis of harvest sustainability cannot be supported by the landing data.