end of the range, off Peru and Chile. This knowledge of 

 relationship between water temperature and occurrence of com- 

 mercial concentrations of tropical tunas is not of great utility 

 to California tuna fishermen at the northern end of the range, 

 since the vessels must traverse these waters in any event to get 

 to the fishing areas to the south. However, off Peru and Chile, 

 the fishermen can benefit both from consulting the temperature 

 charts which are issued at monthly intervals, and also by using 

 their own thermometers as an aid in their scouting operations. 



Investigations of the distributions of cod in relation to tem- 

 perature in the vicinity of Bear Island, between Norway and 

 Spitzbergen, by English scientists, have revealed useful rela- 

 tionships between the bottom temperatures and the location of 

 paying concentrations of cod. It has been shown that paying 

 quantities are rarely caught in water colder than 1.75°C, ex- 

 cept in summer when the fish are feeding heavily to the east of 

 Bear Island and may be found down to — 0.5°C. In early 

 summer and autumn, on grounds west of Bear Island, Atlantic 

 water touching the Bear Island banks can give good cod catches 

 with bottom temperature between 3° and 5°C. Thus, meas- 

 urements of bottom water temperature can be useful to the 

 trawlers in searching for concentrations of cod in this area. 



Another relationship of tuna to their environment which ap- 

 pears to be of tactical value to tuna fishermen is the distribu- 

 tion and behaviour of the tropical tuna ( in the eastern Pacific 

 at least) in relation to the depth of the mixed layer and the 

 structure of the underlying thermocline. The schools of tropi- 

 cal tuna occur in the upper mixed layer of warmer, low density 

 water, which may vary from 10 to 80 meters deep, and which 

 Is underlain by colder water, the sharpness of transition (from 

 the upper mixed layer to the underlying water), called the 

 thermocline, being variable. Data respecting the percentage 

 of successful purse-seine sets on tuna schools in relation to 

 these factors indicate that the schools escape through the bot- 

 tom of the net less frequently when the mixed layer is shallow, 

 especially when it is shallower than the depth to which the 

 net fishes, and when the gradient of temperature in the thermo- 

 cline is very sharp. By measuring the vertical distribution of 

 temperature, by bathythermographs or other means, the fisher- 

 men may, therefore, assist themselves in selecting situations 

 where the escape rate is minimized. 



A local phenomenon which often corresponds to fish con- 

 centrations is the occurrence of fronts, which are boundaries 

 between water masses. Along such boundaries, which can 

 often be located by sharp temperature transitions, differences 

 in water color, and occurrence of floating debris, the associated 

 vertical circulation often concentrates the plankton organism, 

 which in turn leads to concentration of forage fishes and of the 

 predatory fishes which prey upon them. Japanese long-line 

 fishermen, for example, find that laying their gear along and 

 across such fronts, which the Japanese call "siome", improves 

 their fishing success. Similarly, the near-surface schools of 

 pelagic fishes are frequently found more abundant near these 

 features. 



It is also well known that tunas, as well as some other marine 

 fish species, tend to be more concentrated in the vicinity of sea- 



mounts, which the fishermen refer to as "banks". The dis- 

 covery of new seamounts, both by the fishermen and by our 

 submarine geologists have, therefore, led to the discover) of 

 increasing numbers of good fishing spots. Bottom topography 

 charts, together with echo sounders, thus can be used by the 

 fishermen to good advantage. 



The relationships of the harvestable fish to aggregations of 

 their food organisms is also a potentially useful tool which 

 fishermen may sometimes use to improve their own fishing 

 operations, although this is not as yet very well developed. 

 For example, the relationship between herring and the copepod 

 Calanus on which it feeds is sufficiently close to assist the fisher- 

 men in locating herring by their own plankton collections in 

 at least some situations in the North Sea and in the Barents Sea. 

 Such simple instruments as the Hardy plankton indicator have 

 been developed for the use of the fishermen in these situations. 



Forecasting space and time variations 



What both fishermen and fish processors would most like to 

 have from oceanographers are reliable future forecasts of fishing 

 locations and expected catches of particular kinds of fish. 



To make such forecasts for any kind of fish, we need to have 

 useful estimates of the magnitude of the exploitable fish popu- 

 lations, understanding of the distribution and behaviour of 

 the fish in relation to measurable properties of the ocean (such 

 as temperature, salinity, depth of mixed layer, strength of cur- 

 rents and upwelling), and means of predicting the space and 

 time changes in the oceanic properties and processes. Con- 

 siderable progress has been made on all of these, and in some 

 instances useful forecasts a few to several weeks hence are pos- 

 sible. But we have yet a long way to go. 



Through the compilation and analysis of statistics on catch 

 and effort, and age composition of catches, supplemented in 

 some instances by estimates of abundance of young fish prior 

 to their entry into the stock of commercial sizes, methods have 

 been developed for forecasting the magnitude of fish popula- 

 tions which will be available to the fishery. Well known 

 examples are the New England haddock, Bristol Bay red 

 salmon, sockeye and pink salmon of the Frazer River, yellowfin 

 tuna of the Eastern Pacific, California sardines and anchovies. 



As already noted, we have also some useful, but primitive, 

 understanding of the relationships of some kinds of fish to en- 

 vironmental factors, usually temperature. One example not 

 yet mentioned is the skipjack population of the Central Pacific 

 near the Hawaiian Islands, a large component of which in- 

 habits the waters of the California Current Extension, identi- 

 fiable by temperature and salinity. As these waters shift north- 

 erly through the vicinity of Hawaii each summer, the "season" 

 skipjack appear, their availability varying with the time and 

 extent of the shift in the boundary between the California 

 Current Extension and the water mass to the north. Another 

 well known, large scale phenomenon is the "El Nino" off north- 

 ern South America, which, at irregular intervals, averaging 

 about seven years, brings abnormally warm surface waters to 

 the coast of Peru, resulting in great shifts in the populations of 



19 



