State of California Marine Research Committee

Home , CalCOFI, Gat (landform), Pelagic cod

STATE OF CALIFORNIA MARINE RESEARCH COMMITTEE

STATE OF CALIFORNIA DEPARTMENT OF FISH AND GAME MARINE RESEARCH COMMITTEE

CALIFORNIA COOPERATIVE OCEANIC FISHERIES IN V EST IGAT I0 N S

Volume Vlll

1 July 1959 to 30 June 1960

Cooperating Agencies: CALIFORNIA ACADEMY OF SCIENCES CALIFORNIA DEPARTMENT OF FISH AND GAME STANFORD UNIVERSITY, HOPKINS MARINE STATION U. S. FISH AND WILDLIFE SERVICE, BUREAU OF COMMERCIAL FISHERIES UNIVERSITY OF CALIFORNIA, SCRIPPS INSTITUTION OF OCEANOGRAPHY

1 January 1961 EDITORIAL BOARD GARTHI. MURPHY(Chairman) E. H. AHLSTROM J. D. ISAACS JOHNRADOVICH LETTER OF TRANSMITTAL January 1,1961 EDMUND G. BROWN Governor of the State of California Sacraniento, California DEARSIR: We respectfully submit the eighth report on the work of the Cali- fornia Cooperative Oceanic Fisheries Investigations. The report consists of two sections. The first contains a brief review of the administrative and research activities during the period July 1, 1959 to June 30, 1960, a description of the fisheries, and a list of publications arising from the program. The second section is comprised of original scientific contributions. These papers are either the direct results of the CalCOFI research programs, or represent research directly pertinent to resource development in the pelagic realm off California. Respectfully, THEMARINE RESEARCH COMMITTEE J. G. BURNETTE,Chairman D. T. SAXBY,Vice Chairman J. J. BOGDANOVICH HAROLDGARY W. M. CHAPMAN MAX GORBY JOHNHAWK ARTHURH. MENDONCA LEEP. PAYNE CONTENTS Page I. Review of Activities 1. Review of Activities July 1, 1959-June 30, 1960- ______5 2. Review of the pelagic wet fisheries during the 1959-60 season----_ 13 .. 3. Publications ...... 15 II. Scientific Contributions 1. Symposium on Fisheries Oceanography Maurice Blackburn, Editor --______...... 19 Problems in Fish Population Fluctuations Oscar E. Sette______-_--______-_--___81 Fisheries Oceanography in Japan, especially on the Principles of Fish Distribution, Concentration, Dispersal and Fluctua- tion .. Michitaka Uda ...... 25 Fisheries Oceanography in Europe Maurice Blackburn ______-______--______33 Tuna Oceanography Programs in the Tropical Central and Eastern Pacific Milner B. Schaefer______-_-_------_------__-_ 41 Oceanography and North Pacific Albacore James W. McGary, Joseph J. Graham, and Tamio Otsu--_--- 45 Oceanography and Variations in the Pacific Sardine Popula- tion Garth I. Murphy--____------_--___------_--___------55 Relationships of Variable Oceanographic Factors to Migration and Survival of Fraser River Salmon L. A. Royal and J. P. Tully--__ -___-_-- _ __--- _ ____--_____ 65 Observations on the Ecology of the Pacific Cod (Gadus macrocephalus) in Canadian Waters Keith S. Ketchen ______--_ _ ...... ____ 69 Fisheries Oceanography Elbert H. Ahlstrom -___ _ --______-______71 Fisheries Oceanography Warren S. Wooster ______...... 73 2. Evidence of a Northward Movement of Stocks of the Pacific Sardine Based on the Number of Vertebrae Robert L. Wisner__-______-______----75 3. The Water Masses of Sebastian Vizcaino Bay John G. ~~yllie______-----_---T’ 83 4. On Nonseasonal Temperature and Salinity Variations along the West Coast of the United States Gunnar I. Rod~n______-______--___~95 5. The Use of a Common Reference Period for Evaluating Climatic Coherance in Temperature and Salinity Records from Alaska to California Margaret I(. Robinson____----______------_------121 PART I REVIEW OF ACTIVITIES July 1, 1959 -June 30, 1960

REPORT OF THE CALCOFI COM4MITTEE

INTRODUCTION so immediately important as is: what can be expected of the sardine stocks in the more distant future?; It is the purpose of this revicw by the CalCOFI and: are there under-utilized stocks of fish that are Committee to eliwidate the present position of the coiiiplemeiitary to the sardine, that is stocks that in- program and to point out tlie directions in which the crease at a time of diminution of the sardine stocks program is developing and our thinking is evolving. and vice versa. Surely the changing answers that so- The ohjwtivcs aiid statements of objertives of the caiety iteeds in this area can best be drawn from the CalCOF I prograiii have b~engiven much thought most general kiiowledge, extensive yet intimate knowl- aiid expression during the life of the program, and, edge 011 the stocks of pelagic fishes off our coast, their naturall.v, have evolved. In August 1960, the CalCOFI Iiuiiibers, biology, needs, habits, fluctuations, and Committee reformulated the objecatives as follo1 their fishery. To acquire knowledge and understanding of the This matter of the dual obligatioiis of the scientist factors governing the abundaiiw, distributioii, and who must integrate the fiiidings, the scientific synthe- variation of the pelagic marine fishes. The oceano- skt, has been discnssed above at some length for it graphic and biological factors affecting the sardine is this position of research that is occupied by the and its ecological associates iii the California Cur- CalCOFI Committee, with an attendant duality of rent Systeni will be given resmrch emphasis. is It ‘(masters. As the research progresses, the nienibers the ultimate aim of the investigations to obtain an ’’ of the CalCOFI Coinmittre niiist look in both direc- uiitlerstanding sufficient to predict, thus permitting tions aiid attempt to answer both questions : how does eficieiit ultilization of tlie species, arid perhaps a fiiicliiig or line of inquiry help us understand the iiiaiiipulation of the population. pelagic enviroiinieiit of the California Current Sys- This statement formalizes some aspects of such re- tem? and how docs it help 11s answer the questions search that have long been recognized: first, that no of society ? pelagic fish, such as the sardine, can be studied in Rephrasing these thoughts to the conduct of re- nature as a creature isolated from his natural associ- search, the following emerges. Specialized research atrs ; second, that there is ail ultimate responsibility to answer an immediate question posed by society is of such research to the ncwk of society and that this a coiisiimer of basic knowledge and understanding. responsibility takes the form of an ultimate goal of This specialized research will be most successful when attaining an understanding sufficient for the guidance there is adequate basic knowledge, and, as a corol- of society in the utilization of the resource. lary, failure of the specialized research to supply so- examine this matter farther, it is a responsibility To ciety with significant practical results is most likely of the scientist to ask qixestions of nature. In this case, ascribable to inadequate basic understanding. It fol- to ask questions about pelagic fish : what is their food? lows that the principal tasks of the CalCOFI Com- how does it vary? what ultimately kills them? how mittee are to: develop projects to answer questions is their spawning carried out ? how successful is their posed by the pitblic to the extent that there is suffi- spawning? what is the water like in which they live cient basic kiiowleclge ; and encourage research in dis- and how does it affect these creatures? aiid a myriad ciplines and areas where lack of basic knowledge of similar important inquiries. In addition, scientists in a program such as this blocks the siiccess of the program. have two further essential tasks. The first is to weld When the program was initiated it was realized their findings with previous knowledge, thus further- that the principal lack was basic knowledge, so even ing the general or basic understanding of the way though there was a most alarming and serious fisher- nature functions in order that future questions ran ies crisis, the originators and sponsors established the be formulated more meaningfully. At the same time, research on a broad fundamental basis. Because of the scientist must be in a position to answer the this we are now approachiiig the answers to the orig- questions of society to the best of his knowledge. Yet inal questions. Even more important we are also ap- the needs of society change, and it is essential that proaching R position from which we can answer a these questions change to fit the changing needs. For broad spectrum of questions about our pelagic re- example, what was once the important question to sources. This is an iniportaiit and happy aspect of the CalCOFI program: how large will be the avail- this program of investigation, and its truth will be able sardine stock next year ?-perhaps is not now increasingly apparent in the discussion to follow. 6 CAJJIFORNIA COOPERATIVE OCEANIC FJSHERIES ISVESTICATIONS FINDINGS wew produced during the years 1949-56, when the California Current was strong and steady and the What have been the major findings and what is the good year classes wr’e produced when the California picture that has emerged and is emerging froni this Current was often wcak aiid unsteatly. In the period inquiry ? when sardine spawning was general along the entire We know that the sarclilie has not recently been a I’ac.ific Coast. the sardine larvae had a much improved major component of the pelagic fish fauna of the opportuiiity to remain in the California Current arid California Current. It lias often bem a coiispicuous to reach thc iiursery grounds, much to the same degree component because of its habit of schooling relatively its the anchovy now does, and sardine spawning was close to shore, but many other species of fish have been 1110re successful. and are now greatly more abundaiit in the California Other teiitativc fiiidiiigs indirectly support this gcan- Current waters. This iiicludes several spwies of eco- c>ral pic+urc. It appears that large sardine larvae are logirally important deep-sea smelts, lantern fish, very spottily tlistributd. Thus, it is possible that and the anchovy, hake, and jack niackercil, which good survival of tlw sardine larvae is not the result are also commercially importatit. of geiimil eiivironniental conditions, but the combilia- Froni the years up to 1949 the sardinr niai~~tained tioii of nialiy favorable c~onclitionsand fcw unfavor- a population comparatively larger than in more ablr conditions iii sniall patches in the ocean. One recent years, although the years of the earlier period woiilil espc‘ct, siich patchincw of favorable and nn- were rnarkeclly varying in char;tctc.r, with warin and favorable cdoiiil it ions to bv inore rorrinion in rapiclly (.old years and years of other rhararter in alternation. varying yc’ars thiiii ill steady ywrs. Daring this period the populatiotls of sardinrs and IIow does this hypothesis stand up under the most anchovies were about the same size, with anchovies rcwiit history! \VP brlieve that the yars 1957 and possibly in slightly greater abundance. 19.X w(~r(~years niii(:h like those of the early period The years of the decade beginning about 1948 to whii sarciiiic. rc~cruitiiieiit was strong. That is, they 1957 were, however, of a different charactcr, relatively w’re variahlr years, iv;trni(~rthan average, with a colder, but not so cold as some of the cold years of thc tlecvased streiigth of the California Current, and a previous two c1ecadt.s that produccd good ycm calassrs tlecrrase of thc iiorthcrn forms. Yet in these two of sardines. The ontstaiiding charactcv-istic2 of this yc’ars the sardine apparently had only modest year recent decade was its relatively 1nonoto11ous steadi- classes ! ness of temperature, winds and salinity. Dnring this i2pparently a rewrsion to 111~11y of the conditions period, tlie population of sardines off California ft.11 wider whicdh the sartline previonsly m.as successful to new lows while the population of anrlioviw ap- dit1 iiot result ill a retiirn of SII~C parently reached new highs. l’lankton volnnies The coiltinlied presriice of a large popidation of reached very- high values during this latter (le(’ade, iiIlchovi(1s was one conspicuous condition that did not but it is doubtful if the food material iii the plwIikto11 rc’verse, and it is possible that this and other competi- that is of value to the sardine increasetl siniilarly. In tors are obstacles to the returii of the sardine’s SIIC- other words, much of the increased planktoil ronsistetl cess. of jellies, salps, doliolids atid pyrosonies that have Kuperiniposed on these physical aiid biological con- little food value in themselves and, that as a matter of tlitions is? of wiirse, thc eflect of the fishery. ,\lthough fact, compete with tlie sariline and anchovy for 1)lank- the fishuy niay haw little rffcct on the total amount tonic food. of all forage fish, thc selectirr nature of the fishery Why then should the aiivhovy population increasc probably favors the survival of competitors of the while that of the sardine dec~lined,and why should sardiiie to soiiie niikiiown clcgree. this occur when previonsly an occasional nioderatc1ly- This experienw has led also to a greater under- cold year produced a good year class ? stwiitliiig of tlir relationships between tlie large preda- This is possibly so because the anchovy spawiis at a woiis fishes that cwiistitutr one trophic level and the somewhat 1owc.r temperatnre than the sartliiie and at sartlinr aiitl other plaiiktoii-feediii~forage fish that a more favorable season with respwt to zooplankton constitiite a 1ow.r trophic level. Thc yellowtail, bar- food. In addition the anchovy may bc bettcr fitted to racuda, and bonito appear to be as abundant as ever. feed oii phytoplankton than is the sardine, and, henc~), \\’hen tlie waters w~rniediip in 195i these three can better compete with salps and other cold-water spwies became available by moving into the range plankton. of thc sportfisher)-, ~v~iithouyh the sardine popula- In addition an important factor in the survival of tioii aiid sarcliiie fishery rcmaincd at a low level. The sardine larvae is their opportunity to niigratc. to the sardiiic. populatioii shiftcd iiorth~vardone year later coastal nursery grounds. That is, iiiii(*ti of their larval iii 1958 and the conniiercial ratch rose to 107,000 tons. mortality may coli t of their being snept away into Thiis the predators iii 195i inovt~dnorth in the virtual inhospitable waters from which they cwiiiot return. abstwce of sardines, presiiniahly content with the The aiichovy with its niore northerly distribution other forage fish and the increased temperatures. along the l’acific Coast, lias a greater opportunity for Apparently these predators caii do well if the total its spawn to remain in the California Current as lar- foragt>-fish population is adequate, that is they are vac cluring periods of strong currents. not specifitl in their forage-fish associations. That these factors may be important ones i? sup- Perhaps it is not inappropriate to reiterate here ported by a number of findiugs: Snialler year classes that it takes more than a large population of fish to REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 7 guarantee good fishing, the fish must be where they food supply available to them and other effects. On can be caught. Apparently the populations of yellow- the other hand it may divert predators to the com- tail, barracuda and bonito were just as high during peting species. the cool years, but they were abundant only off Mex- Thus the problem on the relation between numbers ico. Similar to many of the game species, sardines of parents and numbers of progeny is very complex, are located farther north during warm years and and it has never been successfully resolved with farther south during cool years. For example, the respect to any marine fish. Most of the studies of outstanding 1939 year class of sardines may have ap- marine fish, and the regulations, involve managing the peared larger than it actually was because it was so age at which recruits are harvested, not managing the available as two-year old fish during the hot year, numbers of recruits, and our research clearly shows 1941. Clearly, availability of pelagic fishes is influ- there is nothing to be gained by managing the age of enced by water temperatures. harvesting the sardine. Theory states there must be Other vital parts of this picture are emerging; some relation between spawning stock size and the marly of them long anticipated. numbers of recruits. Yet, me know there is no simple A particularly important finding is the verification relation. We need to define the extent of this relation of the existence of two genetically-separated popula- and to define the extent to which it will yield point to tions of sardines-a northern group and a southern point predictions. This latter will, of course, involve group. Tt has long been suspected that sardines off the interplay of the population and the environment. the central and lower Baja California Coast and the The attempts to understand the environment have fish from the northern Baja and California Coasts already been discussed. In order to focus on the effect were distinct stocks. Subsequently it has been found of the fishery we need to know the catch in detail, a that some of the sardines in the southern area were routine project for a number of years. In addition we fall spawners while those in the north spawned in the need to know the size of the spawning population and spring and summer. These may be the stocks that the size of the recruited population. The present pro- constitute the two genetically-distinct groups of sar- gram of sampling the fishery provides much of this in- dines as determined by blood type. formation and can perhaps provide more in the future There is reason to believe that it is the northern if additional information is gathered. Spawning stocks stock which has decreased most over the last decade. have been estimated from egg surveys, and have and The southern group also migrates north, but in are being estimated from “fish surveys.” When all greater or lesser amounts depending on oceanographic these data are fully available, which will be soon under conditions. The migrants contribute at least to the present plans, we should have a fresh look at the popu- southern California fishery. lation problems insofar as the direct influence of man Also, it has been found that the major fluctuations is concerned. of the oceanographic coiiditions of the California Current undoubtedly are a part of fluctuations involv- SUMMARY ing most of the North Pacific if not the entire Pacific. These, in turn, appear to result from or to be associ- The picture that emerges from these results is that ated with, major changes in the atmospheric circula- the sardine, a rather minor part of the biomass, can tion. Apparently, the events that trigger these changes prosper when there is much variation in the environ- take place some distance from the California Current ment, but that under steady conditions, or at least System and as much as a year passes before the under steady cool conditions, the sardine, perhaps changes are fully felt on this coast. This gives real abetted by the pressure of man, gives way to its com- hope for prediction of oceanographic conditions, in a petitors. However, with the firm establishment of com- general way perhaps a year ahead. petitors in the habitat of the sardine, a brief return There exists a basic aspect of the sardine problem of -otherwise favorable environmental conditions is in- that requires other answers intimately related to the adequate to re-establish stocks of sardines in the more public need. The most frequent question is: can man northerly portions of its range, although in warm influence the fnture size of his catch or the population years the southern stocks become available to Cali- by voluntarily changing the harvest? That is, for fornia fishermen. example, by catching fewer or more sardines. This It is also clear that shifts in the relative abundance has been studied longer and with less satisfactory among sardines and other forage fish has little effect results than any other aspect of the sardine problem. on the population of predators. The problem is very difficult because it involves pri- With this picture developing from the inquiries of marily the relation between the numbers of spawners the CalCOFI program we must slowly change the pro- and the numbers of progeny. Obviously, an animal gram to make the most of our new understanding. that spawns many thousands of eggs per year does Since we must understand the broad changes of the not have a simple, steady relation between numbers oceanic conditions, the routine surveys are being ex- of parents and numbers of young. An additional com- panded. Siiice we can recognize the types of oceano- plexity is that removal of adult fish does more than graphic changes from fewer data, the major surveys decrease the spawning stock of the species. Such re- are being made less frequently (quarterly) and the moval tends to enhance the survival rate of its own lines farther apart. The ship time so released is being young, but similarly, it may also allow competitors used for more direct studies of the environment, the of the species to be more successful by increasing the conditions and competitors in the sardine spawning 8 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS area, spottiness of larvae and plankton, countercur- are aged and measured, live samples are saved for rents, etc. genetic studies by the U. S. Bureau of Commercial This intensification of effort also is being devoted to Fisheries, and other fish are preserved for further the sardine itself, the conditions of its eggs, its light study ashore by the Department. preferences, its distribution and associates as larvae These two routine projects for four species repre- and juveniles, and its growth rate as associated with sents most of the Department’s effort in the CalCOFI the strength of year classes. program. Although routine, they are basic to the needs It is clear that we soon will have considerable de- of the entire program, since before causes of variation tailed understanding of the actual conditions under in abundance and availability can be investigated, the which sardines and perhaps other pelagic fishes ex- variation must be described. perience high survival. In addition to the routine work, measurements To date in this program there have been extensive were made for morphometric studies on about 2,500 scientific results. There has been a long period of pi- sardines. This is part of a study to determine if suffi- oneering and exploration and the slow development of cient differences can be recognized in fish from differ- a picture of the place of the pelagic fishes in the ent regions of the coast to enable one to recognize California Current System. With this development of “types” and thereby, follow sardine movements from perspective, there is increasing opportunity to feed year to year. One other aspect of the morphometric the findings back toward more meaningful research study is to look critically at samples of fish whose and greater opportunity to answer the question of blood has been typed to see if phenotypic differences society. can be detected which correspond with the genetic Perhaps the most important product of the pro- separation obtained from serological work. gram is a perspective of the sardine and its associates. The live bait catch sampling was continued as was We know that the sardine was once a conspicuous and the airplane surveys of the anchovy population. Stud- valuable component of the population. Well may it be ies on the effect of water temperatures on fish distribu- again. But we now realize that the sardine is like the gold of California, a conspicuous, valuable, easily- tion continued and the results will be published. harvested element in the midst of less-conspicuous riches of far greater potentialities but requiring pain- staking development.-E. H. Ahlstrom, J. D. Isaacs, G. I. Murphy, and J. Radovich.

AGENCY ACTIVITIES California Academy of Sciences Studies on the behavior and reactions of sardines and related species were continued, utilizing one investigator and assistance as needed. Specific investigations during the year included studies on the responses of anchovies to ultraviolet and infra-red radiation, and their responses to gradients of white light. In addition, further tests were completed on the responses of anchovies, jack mackerel, Pacific mackerel, and sardines to various electrical fields. Studies on the schooling of anchovies were conducted at Marineland.

California Department of Pish and Game The Department of Fish and Game continued its basic monotoring of the pelagic wet fish fisheries of California. This includes : sampling sardines, anchovies and Pacific and jack mackerel, to determine the age and length composition of the catch; inter- viewing fishermen to obtain information on fishing localities and fishing effort ; and working with the Department of Fish and Game statistical unit to in- sure that the catch records are accurate. The department also continued its fish survey work which, in effect, extends the type of information obtained from the fishery to areas beyond those presently 122000’ occupied by the fishing fleet. The pelagic species are ~ ~~ - sampled at sea and information is obtained on the Monterey Bay, showing cruise pattern and distribution and relative abundance of fish. The fish station locations. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 9

Hopkins Marine Station, Stanford University the preceding two years. At Santa Cruz the tempera- Throughout the year the Hopkins Marine Station ture differences were much less; they were about equal has continued to keep a finger on the pulse of the to those of 1957 and averaged only 0.2" C colder than Pacific in the central California area through a pro- those of 1958. During each of these first five months gram of three correlated activities: 1) Daily shore the temperatures were equal to or warmer than those temperatures were taken at Pacific Grove and Santa of one of the preceding two years, and in August they Cruz. 2) At approximately weekly intervals, depend- were warmer than both. ent upon the weather, cruises were made covering six During the next six months from December, 1959 stations in Monterey Bay. At each of these stations re- through May, 1960, the situation changed markedly. versing thermometer temperatures and water samples At both extremes of Monterey Bay, the average tem- for salinity determinations were taken at the surface perature in each of these months was markedly colder and 15 meters. A bathythermograph was also lowered than that of the corresponding months of the two pre- to 50 meters, except at stations 1 and 5 where shallow ceding years. At Pacific Grove this period averaged water limited the depth of the casts to 30 and 20 1.30" C colder than that of 1957-58 and 0.7" C colder meters respectively. Vertical hauls for phytoplankton than that of 1958-59. The situation was similar at and zooplankton were also made at each station. 3) Santa Cruz ; the corresponding values were 1.2" C and Each month 26 shore temperature stations were occu- 0.9" C June provided what might be considered a re- pied between Monterey and Islay Creek, 20 miles south turn to warm water conditions with temperatures of Morro Ray on a day during or close to the period above those of 1959, but below those of 1958. It should when CalCOFI ships were operating off-shore in the be mentioned here, that although winter and spring same latitudes. were characterized by water colder than that of the prior two years, temperatures between October and February were still higher than those prevailing in 1956-57 just before the onset of warmer conditions along the California coast. The temperature situation in Monterey Bay, as outlined above and based on daily observations, was extremely well corroborated by the averages of 26 shore stations occupied at monthly intervals along the coast- SHORE TEMPERATURES - PACIFIC GROVE line to the south. Here, too, the summer and fall of 1959 appeared to be typical of the recent warm-water period, but the following winter and spring yielded temperatures markedly colder than in the preceding two years. The warming noted in June in Monterey - = 1959-60 Bay, took place somewhat earlier toward the south and __ = 1958-59 the curve of the coastal run approached that of 1957- __.-= 1957-58 58 in May as well as in June. The abrupt decline in surface temperatures during I,,,,,:,,,, JISONDJCMIY? the winter and sprii1.g is perhaps more strikingly SHORE TEMPERATURES - SANTA CRUZ shown by monthly averages of the data collected on the weekly cruises. These are not so strongly influenced by shallow-water warming as are those taken along shore. They show clearly that the surface temperatures are depressed well below those of 1957-58 and 1958-59. In fact, the surface temperatures during these months are comparable to the very cold period of 1955-56. Bathythermograph temperature averages showed little change over those of the previous year in the first half of the period under discussion, although there was a slight lowering at all depths in September, JLSONDJFMAYJ October and January. However, with the onset of up- COASTPL TEMPERATURES-MONTEREY to MORRO BAY welling in February, there was a concomitant overall During the early part of the period covered by this lowering of temperatures to approximately 1" C below report (J~ly1959-June 1960), monthly averages of those of the previous year. Although this condition daily shore temperatures taken at Pacific Grove and prevailed until June, at which time the temperature Santa Cruz indicated conditions only slightly cooler rise rioted at inshore area5 prevailed in the upper 10 than those of the warm-water periods prevailing dur- meters, the persistence of relatively low values was ing the previous two years. At Pacific Grove the probably not due to continued upwelling. Lowered period averaged about 0.6" C colder than in 1957 and temperatures of late winter and early spring are nor- about 0.4" C colder than 1958, but during August and mally the result of upwelling. Fpwelling is typically October the water was actually somewhat warmer than indicated by a spread between the maximum and mini- during the corresponding month in one or the other of mum values for any given month. This temperature 10 CALIFORNIA COOPERATIVR OCEANIC FISHERIES INVESTIGATIONS __ - 17- A n --

1956 I 1957 I 1958 I 1959 I 1960 JASONDJ FMAMJJASONDJF NAMJ JASONDJfMAMdJASONdJPMAMd

SURFACE TEMPERATURES - MONTEREY BAY, CALIFORNIA MONTHLY AVERAGES

difference is due to the presence of cold freshly up- ographic program is being modified. To a great extent welled water at stations over or adjacent to Monterey this modification is the result of the perspective and Canyon, and of surface-warmed eddies at the extrmii- the broader recognition of problems presented and ties of the Ray. It is noteworthy that between Jlarch summarized in th(1 CalCOFI’s report of last year and June, 1960, a period which in normal years is Volume VIT. In particnlar, the fact that the changes characterized by a coiispicnons spread in the inaximiini in the local waters of the California Coast are related and minimum curves, thrse lines ran close together. to fluctuation in conditions over much of the Pacific, Although it must be admitted that this niay throw is receiving increased recognition. The surveys are doubt on the validity of the spreatl as an index of iip being extended to reach farther north and farther to welling, it niay also indicate that the cooling vas tlne sea, and the iiuniber of station lines occupied is being to factors other thaii upn.r~lling.111 this c.onnec.tioii, it rednced. Thew changes stem from : the need to obtain may be pointed out that the only sizwble planlcton a broach and more general look at the California. bloom (another iiidicater of iipv elling) occiircd in Currcint ; the ability to recognize change from fewer March, just siibseqiient to the upwelling of February data as a result of understanding from previous as shown by the spread of the maxininln and miiiiinuin studies ; and thc necessity to carry out concentrated in that month. No Airther sizcablc plankton prodnc- studies on speeial featiires of the environment that tioii was noted and it is probablc that no cltirichment have been pointed up. Thc modified field program from deep waters took place. Hvidently the cooling will thus consist of a basic extensive survey plan, as was not due to upwelliiig but was a result of a wide- sho~vn,ant1 a sericJs of special oceanographic stndies spread phenomenon affecting a large area of the Sorth concerned with such subjects as: the nature of the Pacific. countercurrriit circulation, tlirect current measure- Salinities in the Bay were slightly highcr from ,July ments, the ecology of spawning areas, and the nature through September over the previons year. They I\ ere of plankton sncces\ion. somewhat lowered in the periocl from eJaniiary to I11 the area of interpretation of the oceanographic March, especially in February whc~iiraiiifall and run- data, the program is becoming inore involved with off affected surfare salinities. Tt 11 it\ iiot pos\iblc to pretlictioii. The basic correlatioii of temperature, interpret differences in saliiiit iw of corre5poiiding wind, and sunshilie has been worked out and the months in various years as indicators of iipwelliiig next step will involve dynamic prediction, that is, Phytoplanktoii vo1umt.s hare remaincd low through- predictioii that considers currents. Studies of the out the past year, aiid the total volume of saniplcs was broad climatology of the Eastern North Pacific are less thaii half of that obtili1ld ill eavh of tht. previous receiving increased emphasis. two years. The most noticeable bloom was in ,Ilarc.h, Correlation bctn een the oceanographic conditions when hauls were far abovc tliosc. of 19% and 1!)T,9, wild and the plankton of the areas is recaeiving attention. comparable to 1957 volume~,the 1attc.r bcing the last Work is proceeding on the zoogeography of the region, year when phytoplaiikton was preseiit in niar1w(l qnan- aiid all major groups of zooplankton are now receiv- tity in the Bay. inc attention. It is now pos5ible to examine associations of zooplankters, the as5ociation of zooplankton, Division of Marine Resources Scripps Znstitution of and- for example, fi4i larvae, and the association of Oceanography University of California these organisms with the oceanographic coiiditions, TTTith the complction and publication of almost tm and this is iincler way. years of ineasiirenients of the oceanic conditions of In this direction, preparation of charts displaying the California Current System, the Cal(’OE’I’s ocean- the oceanographic conditions throughout the years is REPORTS VOLUJIE VIII, 1 JUTiY 1959 TO 30 JUNE 1WO 11

being undertaken, to permit a inore facile comparison investigations of population dynamics, population size, of biological conditions with the oceanographic year-class size, availability, age and growth, fecundity, weather. This is also a step in the preparation of a genetics (subpopulations) and physiology ; the tangen- comprehensive description of the California Current tial programs include plankton studies (data on System. plankton volumes are issued anniially ) and erologi- In the field of instruments, progress has been made ca11y associated fishes (including anchovy, jack inack- toward substituting moored stations for ship observa- me1 and Pacific mackerel). tions, although these are not yet routinely in use. Nets Perhaps the most exciting research results w’re ob- to solve special problems of fish and plankton behavior tained in the genetic studies of snfipopulations. The and distributioii have been constructed, the deep-free tern, one of three blood systems that have been vehicle has been expanded to include carrying ai1 ex- distinguished, has proven of \ralue in separating sub- perimental current meter, and equipment for routine populations of sardines. A rnorc iiorthrrly distributed direct observation of the ecology of pelagic areas are group of sardines (off California aid at times north- under developnient. ern Baja California) arerage over 133, “ C” positive individuals, while a inore southrrly distrihutcd group U.S. Bureau of Commercial Fisheries (BCF) (Magdaleiia Ba>-, Raja California, to as far north as The program of research of the La Jolla Laboratory, San Diepo) averages less than 6% “C” positive fish. Bureau of Commercial Fisheries, is macle up of 10 in- The most recent iiivestigatioii initiated by the TAL vestigations, 8 of which center directly on the sardine Jolla Laboratory, physiology studies, is obtaining ill- and 2 that are tangential. In the former category are teresting results in several fields : osmoregulation 12 CAT,IFORSIA COOPERATIVE OCEANIC FISHERIES ISVESTIGATIONS and energy requirements of sardine eggs and larvae, studies of the distribution dynamics of sardine food organic constituents (fatty acids) of the ovary, nu- organisms. trition of sardine-related species and rearing of post- Sardine spawning in 1960 had the type of distribu- yolk sac sardine larvae. tion that has been typical of warm years since 1957. Availability studies are concerned with investigat- Spawning was less widely distributed and closer ing the manner in which the environment influences inshore than previously, with major centers in the the distribution and behavior of the sardine. A phase Channel Island area of Southern California and in of the behavior project completed during the year Sebastian Viscaino Bay, Baja California. Based on compares, on the basis of eye structure, the relative preliminary scanning of samples, the amount of visual capacities of the Pacific sardine, northern an- spawning appears to be low. Fecundity studies have chovy, Pacific mackerel, jack mackerel and some other shown that during the recent warm years, sardines off fishes. The sardine has the highest visual acuity and Southern California mature at much smaller sizes, the lowest sensitivity, thus should be able to discrimi- have a more prolonged spawning season and are much less uniform with regard to stage of ovarian develop- nate objects better than the other species in bright ment. A preliminary examination of sardine larval light, but riot as well as others in dim light. Explora- data for 1952-1957 indicates that larval survival after tory behavior experiments are being started, and the the yolk sac stage has been relatively constant-larger experimental facilities are being improved. An im- larvae (15.5 mm and longer) comprising between proved plankton pump is being built to be used in 1.5% and 3% of the total larvae. REVIEW OF THE PELAGIC WET FISHERIES During the 1959-60 Season SARDINE end of the 1959-60 season. the catch of sardines totaled The enthusiasm generated by the 104.000 ton 1958- O1lly about 37.OoO tons . 59 sardine fishing season was short lived. as a soft During the summer of 1959. before the season be- market kept the catch from reaching a higher value . @n. fish surl'eys off the California fishing grounds The poor foreign market continued into 1959-60 and conducted by the California Department of Fish and as the season unfolded. it became apparent that catch Game and egg and larvae surveys conducted by the would be limited even further by the supply. At the U . S. Bureau of Commercial Fisheries. revealed the

TABLE 1

SEASONAL CATCH IN TONS OF SARDINES ALONG THE PACIFIC COAST-EACH SEASON INCLUDES JUNE THROUGH THE FOLLOWING MAY'

1 PACIFIC NORTHWEST CALIFORNIA2 I- I- NORTHERN CALIFORNIA

British Wash- Reductio1 San Southern Total Bajan Grand Season Columbia ington Oregon Total Ships Francisco Monterey Total California hliornia 11 California 11 Total I- I- //I 1916.17 ...... 7. 710 7. 710 19.820 27. 530 .... 27,530 1917.18...... 80 ...... 8c .... 70 23. 810 23. 880 48.700 72.580 .... 72,660 191819 ...... 3. 640 ...... 3.64C .... 450 35.750 36.200 39.340 75.540 .... 79, 180 1919.20 ...... 3. 280 ...... 3.2W .... 1. 000 43. 040 44. 040 22. 990 67. 030 .... 70.310 1920.21 ...... 4.400 ...... 4.40C .... 230 24. 960 25.190 13.260 38. 450 .... 42. 850 1921-22.- ...... 990 ...... 99c .... 80 16.290 16.370 20.130 36.500 .... 37. 490 1922.23 ...... I.020 ...... 1.020 .... 110 29.210 29. 320 35.790 65. 110 .... 66. 130 1923.24 ...... 970 ...... 970 .... 190 45.920 46. 110 37. 820 83.930 .... 84.900 1924.25 ...... 1.370 ...... 1.370 .... 560 67.310 67. 870 105.150 173.020 .... 174;390 1926.26 ...... 15.950 ...... 15.95C .... 560 69. 010 69. 570 67. 700 137.270 .... 153,220 1926.27 ...... 48. 500 ...... 48.500 .... 3.520 81.860 85.380 66.830 152.210 .... 200.710 1927.28 ...... 68.430 ...... 68.43C .... 16.690 98.020 114.710 72.550 187.260 .... 255,690 192829...... 80.510 ...... 80.51C .... 13.520 120.290 133.810 120.670 254. 480 .... 334,990 1929.30 ...... 86.340 ...... 86.34C .... 21. 960 160.050 182. 0 10 143.160 325.170 .... 411,510 1930-31._...... 75.070 ...... 75.076 IO,960 25. 970 109.620 146.550 38. 570 185.120 .... 260, I90 1931-32...... 73.600 ...... 73.600 31,040 21.607 69. 078 121.725 42.920 164.645 .... 238,245 1932.33 ...... 44. 350 ...... 44.350 58.790 18.634 89.599 167.023 83.667 250. 690 .... 295,040 1933.34 ...... 4. 050 ...... 4. 050 67,820 36. 336 152.480 256.636 126.793 383.429 .... 387,479 1934.35 ...... 43.OOO ...... 43. OOC 112,040 68. 477 230. 854 411. 371 183.683 595. 054 .... 638,054 1935-36-...... 45.320 10 26.230 71. 560 150,830 76.147 184.470 411.447 149.051 560.498 .... 632.058 1936-37 ...... 44. 450 6,560 14,200 65. 210 235,610 141.099 206. 706 583.415 142.709 726. 124 .... II 791:334 1937.38...... 48.080 17,100 16,660 81.840 67,580 133.718 104.936 306.234 110.330 416. 564 .... 498. 404 1938.39 ...... 51.770 26,480 17,020 95.27C 43,890 201. 200 180.994 426.OS4 149.203 575. 287 670.557 1939.40 ...... 5. 520 17,760 22,330 45. 610 .... 212. 453 227. 874 440. 327 96. 939 537. 266 1::: 582.876 .... j 1940.41...... 28.770 810 3, 160 32. 740 118.092 165.698 283.790 176.794 460. 584 .... 493.324 1941.42 ...... 60.050 17,100 15,850 93.OOC .... 186.589 250.287 436. 876 150.497 587.373 .... 680.373

1942.43...... 65. 880 580 1,950 68.41C .... 115.884 184.399 300.283 204.378 504.661 .... ~ 573.071 1943-44- ...... 88.740 10,440 I,820 101.000 .... 126.512 213. 616 340. 128 138.001 478. 129 .... 579. 129 1944.45 ...... 59. 120 20 .... 59. 14C .... 136.598 237.246 373.844 181.061 554.905 .... 614.045 1945-46...... 34.300 2,310 90 36.70C .... 84.103 145.519 229. 622 174.061 403.683 .... 440. 383 1946.47 ...... 3.990 6,140 3,960 14.09C .... 2.869 31. 391 34.260 199.542 233. 802 .... 247. 892 1947-48-...... 490 1,360 6.930 8.78C .... 94 17.630 17.724 103.617 121.341 .... 130.121 194849...... 50 5,320 5.37c .... 112 47.862 47. 974 135.752 183.726 .... 189.096

1949.50...... -...... 17.442 131.769 149.211 189.714 338.925 .... 338.925 1950-51_-...... _...... 12.727 33. 699 46.426 306.662 353.088 .... 353.088 1951.52 ...... _...... 82 15.897 15.979 113.125 129. 104 16.184 145.288 1952-53-...... __ ...... 49 49 5.662 5.711 9.162 I 14.873 1953.54...... 58 58 4.434 4. 492 14.306 18.798 1954.55 ...... _.~...... 856 856 67.609 68.465 12.440 80.905 1955.56 ...... __. .... _ ... 518 518 73.943 74. 461 4. 207 78. 668 1956.57...... _...... 63 63 33.580 33.643 13.655 47.298 1957.58 ...... __ ...... 17 17 22.255 22. 272 32.196 1958.59...... _ ... 24.701 24.701 79. 270 2;:::; 11 126.305 1959-60' ...... _._ .... 16. 109 16.109 21. 146 21.424 58.679

1 Data for British Columbia were supplied by the Canadian Bureau of Statistics and 2 Prior to the 1931-32 season fish landed in Santa Barbara and San Luis Obispo the province of British Columbia; those for Washington by the Washington Counties are included in Southern California . Beginning with the 1931-32 1)ep;rrtment of Fisheries. and for Oregon by the Fish Commission of Oregon . season fish landed north of Pt . Arguello are included in the Monterey landings Tonnages delivered to the reduction ships and data for Baja California were and those south of Pt . hrruello are included in the Southern California compiled by the United States Fish and Wildlife Service from the hooks of landings . the companies receiving fish . California landings were derived from the records 3 The amount of sardines landed in Baja California prior to the 1951-52 season are of the California Ikpartment of Fish and Came . not known . Beginning with 1951.52 . the period of fishing approximates that of the rest of the Pacific Coast . Figures are preliminary . Preliminary records . 14 CL4LIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS presence of sardines in sufficient numbers to yield only ANCHOVY a slightly smaller catch than that of the preceding season. This was with the assumption that the sardines Commerci+l landings of anchovies in 1959 dropped would behave in the same manner as in 1958. In to a total of only 3,600 tons from 5,800 tons landed in addition, the blood genetic studies of the U. S. Bu- 1958. These were the poorest catches since 1951, de- reau of Commercial Fisheries showed that a “southern spite a large anchovy population present off the Cali- fornia coast. The low catch reflects market conditions type ” of sardine extended northward from Mexico into southern California during the summer, rather than availability of anchovies, since the anchovy Although the season officially opened August 1, for catches are generally low when sardines are abundant central California and September 1, for southern and high when sardines are scarce. The scarcity of California, price negotiations prevented the fleet from sardines in 1959, did not increase the demand for fishing off central California during August and off anchovies, however, because of the large inventories southern California during September. After the of sardines left over from the previous year. price was finally settled at $35 a ton, the central Cali- The catch of anchovies for live bait was about 5,000 fornia fleet began fishing on September 1, and the tons compared with approximately 4,000 tons in 1957 southern California fishery began on October 4. and 1958. The live bait fishery continued to be depend- When the fleets finally ventured forth, sardines were ent on the very small “pinhead” (fish of the year) somewhat scarce. This appears to have been the result and one year old anchovies. This was no indication of scarcity, however, as large anchovies apparently of two things : sardines were not schooled densely, probably because the “southern type” present off moved offshore and deeper to “cool off” when the southern California were spawning at that time ; and ocean warmed up. Egg and larvae surveys of the U.S. the fish moved southward as winter arrived. By Febru- Bureau of Commercial Fisheries indicate that a large ary the line separating the “northern type” from the spawning population was present during 1958 and “southern type” was about 150 miles south of the In- 1959. In addition, offshore sampling by the California ternational Border. Department of Fish and Game revealed the presence of large anchovies between 20 and 60 miles off the The 1959-60 season resulted in landings of 16,110 coast, and some large anchovies were killed during off - tons off central California and 21,150 tons off south- shore seismic operations. Another piece of evidence ern California, for a total California catch of 37,260 tons. As in 1958-59, most of the catches in southern TABLE 3 California were made north of San Pedro. The fishing fleet totaled 128 vessels, 38 from central California CALIFORNIA SEASONAL CATCH IN TONS OF and 90 from southern California. Again two-year-olds PACIFIC AND JACK MACKEREL (1957 class) dominated the catch. The 1956 class which (Each Season Includes May through the Following April) made up the bulk of the 1958-59 season’s landings did not appear outstanding, reaffirming the earlier con- ‘ercentage Pacific Jack Pacific clusion that the 1956 class was not large but extremely Season Mackerel Mackerel Total Mackerel available during 1958. I- 1926-27 ...... 1,797 183 1,980 90.8 TABLE 2 1927-28...... 3,228 213 3,441 93.8 192829.--_._..--_...------19,703 278 19,981 98.6 COMMERCIAL LANDINGS AND LIVE BAIT CATCH OF ANCHOVIES 1929-30...... 28,347 337 28,684 98.8 IN TONS IN CALIFORNIA, 1939-1959 1930-31...... 6,403 155 6,558 97.6 1931-32 ...... 7,576 336 7,912 95.8 (Live Bait Catch 1943-4s Not Recorded) 1932-33...... 5,425 233 5,658 95.9 1933-34 ...... 36,437 553 36,990 98.5 1934-35 ...... 56,732 827 57,559 98.6 Commercia Percent 1935-36...... 73,194 4,925 78,119 93.7 Year Landings Live Bait Total Live bait 1936-37...... 50,373 2,879 53,252 94.6 1937-38 ...... 35,223 4,121 39,344 89.5 1938-39...... 38,032 1,948 39.980 95.1 1939...... 1,074 1,503 2.577 58.3 193940 ...... 49.980 559 50,539 98.9 1940...... 3,159 2,006 5,165 38.8 194041...... 53,777 875 54,652 98.4 1941...... 2,052 1,582 3,634 43.5 1941-42 ...... 35,877 959 36,836 97.4 1942...... 847 258 1,105 23.3 1942-43...... 24,110 4,897 29,007 83.1 1943...... 785 1943-44 ...... 38.902 4.228 43,130 90.2 1944...... 1,946 1944-45...... 40,393 6,871 47,264 85.5 1945...... 808 1945-46...... 26,001 4,635 30,636 84.9 1946...... 961 2,748 3,709 74.1 1946-47 ...... 29,448 15,573 45,021 65.4 1947...... 9,470 2,854 12,324 23.2 1947-48...... 19,814 71,330 91,144 21.7 1948...... 5,418 3,725 9,143 40.7 194849...... 19,101 27,845 46,946 40.7 1949...... 1,661 2,802 4,463 62.8 1949-50 ...... 25,031 32,494 57,525 43.5 1950...... 2,439 3,824 6,263 61.1 1950-51...... 16,945 68,187 85.132 19.9 1951...... 3,477 5,142 8.619 59.7 1951-52 ...... 15,953 37,495 53,448 29.8 1952...... 27,892 6,810 34,702 19.6 1952-53...... 10,109 75,750 85,859 11.8 1953...... 42.918 6,391 49,309 13.0 1953-54 ...... 4,415 18,369 22,784 19.4 1954...... 21,205 6,686 27,891 24.0 1954-55 ...... 13,605 9,417 23,022 59.1 1955...... 22,346 6,125 28,471 21.5 1955-56...... 13.448 29,674 43,122 31.2 1956..-----..--_-..-______28,460 6,332 34,792 18.2 1956-57- ...... 28,592 48.173 76,765 37.2 1957...... 20,274 4,110 24,384 16.9 1957-58...... 28,119 19,917 48,036 58.5 1958...... 5,801 4,236 10,037 42.2 1958-59...... 12.388 11,352 23,740 52.2 1959-....-...-..-_--..---- *3,587 4,737 8,324 56.9 1959-60* ...... _I 20,641 33,280 53,921 38.3

Preliminary estimate. * Preliminary estimate. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 15 which indicates that the large anchovies moved off- PUBLICATIONS shore, comes from examination of stomachs of bluefin 1 July 1959 - 30 June 1960 tuna caught in purse seines between Santa Catalina hhlstrom, E. H., 1959. Vertical distribution of pelagic fish eggs and Sari Clemeiite Islands. These stomachs were and larvae off California and Raja California, C.S. Fish and gorged with freshly ingested large (7 inch) anchovies. Wildlife Service, Fishery Bulletin 161, 60: 107-146. Therefore it is apparent that the low commercial Vertical distrihution of 46 kiiids of pelagic fish larvae and 8 anchovy catch was due to a lack of demand, and the kinds’of ews are discussed. Most fish eggs and larvae were found to occur in the upper mised layer and the upper Dart scarcity of “hook size” anchovies in live bait catches of tlie thermocline, netween the surface and apprbximatelg was a result of an offshore movement of the larger fish. 123 meters deep. All the more common kinds of larvae showed marked differences in vertical distribution from series to series. Information is given on differences between day and MACKEREL night catches. Comniercial landings of Pacific mackerel during the Ahlstrom, E. H., 1959. Distribution and abundance of eggs of 1959-60 season rose to 21,000 tons from 12,000 tons in the Pacific sardine, 1952-1956. U.X. Fish and Wildlife Scrv- ice, Fishery Bulletin 165, 60: 185-213. 1958-59, while the jack mackerel catch of 33,000 tons During the period covered, a major change occurred in the tripled that of the previous year. distribution of sardine spawning. In 1952 and 19’53 sardine The higher price paid for both species of mackerel spawning was mostly confined to the waters off central Baja in 1959 ($3.00 per toil compared with $35.00 per ton California. In 1954, sardine spawning spread northward to waters off southern California, and this distribution has for sardines) reflected a better market, and un- continued through 1965 and 1956. Estimates of total eggs doubtedly contributed to the increased mackerel spawned during these years ranged from 136 x 1012 to catches in 1959-60, augmented by the scarcity of 436 x 10*?. sardines causing more fishing effort to be directed to- &4hlstrom, E:. H.,1959. Sardine eggs and larvae and other fish ward the higher priced mackerel. larvae, Pacific Coast, 1967. U.8. Dept. Interior, Fish and Pacific mackerel less than one-year-old dominated Tl-ildlife Service, Spec. Sei. Report: Fisheries No. 388, 67 pp. the catch with fair numbers of one- and two-year-olds, Iklser, W., 1959. Bioassay of organic micronutrients in the also present. sea, I’roced. of the n’otioncil dcadeniy of Sciences, 45(10). I3eriier, L. D., 1959. The food of the larvae of the northern auchovy h’ngraztlis nrordaz. Iiiter-Aniericnn Tropical Tuna TABLE 4 Commission Bulletin, 4(1). ANNUAL COMMERCIAL LANDINGS IN TONS OF THE PELAGIC Ihytl, C. JI., 1960. The larval stages of Pleiironcodes plnnipes WET FISH IN CALIFORNIA FROM 1926 THROUGH 1959 Stimpsoii (Crustacea, Decapotla, Galatheidae) , Biologicol

~ ~ ~ ~ ~ ~ ~ ___ ~ ~ Bitl!., 118(1) : li-30. 1 Sar- An- Pacific Jack Clausseii, 1,. G., 1959. h sonthrwi range esteiisiou of the Amer- Year dines :hovies lackerel Iackere Ierring Squid Total ican shad to Todos Haiitos Hay, Iinjii (’aliforiiia, Mesico. .. .- __- ---- __- __- ____ California Fish utrd Qunie, 45(3) : 217-218.

1!126 -~... 143,371 30 1,805 118 227 1,568 147,119 Coiitois, L). E., 1959. Riuetics of bacterial growth : relation- 1W7. .. 171,138 I84 2,364 231 584 3,007 177,508 ship hetween population tleusitg :ind specific growth rate 1!128.. .. . 210.133 179 17.626 269 570 676 329.455 of coiitiiiuous cultures. Jour. of Gen. dficrobiology, 21 (1): I!J29..--. 325,886 191 28,987 349 479 2,330 358,222 40-50. 1!130 .. 231,031 1A0 8,266 184 359 5,485 365,485 I):iuglierty, A., aiid R. S. Wolf, 19GO. Age mid length composi- 1931. ... 182,176 1,54 7,127 282 343 869 190,951 tiori of tlie sardine catch off the Pacific Coast of the United IO32.. .. . 21 1,305 150 6.237 268 383 2,115 220,458 States and Mexico in 1957-58. Calif. Fish and Game, 46(2) : 1933. ~..313,19!) 159 34,807 505 30 1 412 349.383 189-193. 1934-. .. . ,j39.9Gfi 129 56.924 791 40 1 765 518,976 193.5. ... . 347,879 90 73,214 4,992 464 408 527,047 Farris, D. A,, 1960. Failure of ail anchovy to hatch with continued growth of the larva. Lin??rology and Oceanography, 1936. ..~731,772 98 50,271 2,300 420 473 785,334 5 (1): 107. 1!137.. .-. 533.743 113 30,468 3,270 316 251 570,163 1938 ...~.jll,l~!l5 368 39,924 2,067 252 800 555,106 Fink, E., 1969. Observation of porpoise predation on a school 1939...-. 380.3!17 1,074 40,455 1,880 151 581 524,538 of Pacific sardines, Calif. Fish and Gnnre 45(8) : 216-217. 1940. . 432,!187 3,159 60,252 716 227 900 518,241 Hand, C. H., and L. D. Iierner, 1959. Food of the Pacific sar- 1941.. .. 631,240 2.053 39,084 1,034 395 716 574,522 dine (Sardinops caern7ea), U.S. Fish and Wildlife Service, 1942. ... 484,874 847 26,277 2,674 95 476 515,239 E’ishery Bulletin 164, 40: 175-184. 1943. ... 486,135 785 37,607 6,350 315 4.582 535,774 1944.. .. . 573.604 1,946 41,828 6,388 211 5,468 629.445 Hgatt, H.,1960. Age composition of the southern California 1945..... 422,531 808 26,858 4,516 230 7,613 462,556 catch of Pacific mackerel, Puenmatophoriis diego, for the 1957-58 season. Calif. Fish and Game, 46(2) : 183-188. 1946.. .. . 255.380 961 26,938 7,547 241 19,012 310,079 1947. .__127.757 9,470 23,239 64,524 827 7,271 233.088 .Johnson, 11. TV., 19C;O. Production ant1 distribution of larvae 1948..... 181,018 5,418 19,693 36.449 4,001 9.628 256.207 nf the snriiix lobster, I’u?iitlirus interruptus (Randall) with 1949.. .. . 316,690 1,661 24,886 25.625 190 3.430 372,482 records on P. gracilis, Ritll. of the Scripps Inst. of Oceano. 2,998 446,364 1950...-. 337,261 2,439 16,325 66,628 713 C.C”. 7(6) : 413-462. 1951..... 164,450 3,477 16,759 44,919 2.462 6,191 238,258 MacGrrgor, J. S., 1959. Relation between fish condition and 1952..... 7,165 27,891 10,302 73,261 4,748 1,836 125.203 populatioii size in the sardiue (Sardinops caeruleal. U.R. 3,751 27.875 3,901 4,459 87,638 1953..... 4.734 42,918 Fish and Wildlife Fishery Rulletin 166, 60: 215-230. 1954...._ 68.252 21.205 12,696 8,667 456 4,078 115,354 Service, 1955.. .. . 72,804 22.346 11,655 17,877 973 7,136 132,791 A high degree of inverse correlation was shown between con- clitioii factor and catch, niid :I high positive correlation be- 1956.. .. . 34,777 28,460 25,006 37,881 868 9.742 136,734 t\veei: length and conditiou factor for the seasons 1941-2 1957.. .. . 22,931 20,274 31,022 41.006 594 6,225 122,052 through 193-7. 1958.... 103.723 5,801 13,824 11,033 1,200 3.729 139,310 1959*.-.. 37,183 3,587 18,801 18,754 864 9,826 89,015 JlcGowiin, J. A,, 1960. The relationship of the distribution of the planktonic warm, Poeohius nieseres Heath, to the water mases of the Sorth Pacific. Deep Research, 6: 125-139. * Preliminary estimate. Sea 16 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIOXS

Reid, C. F., 19Cfi. Notes on four specimens of I’acific sardine Itotlen, G. I., and G. JV. Groves, 1960. On the statistical pre- taken in Auguat 1957 off British Columbia and Oregon. Cafi- diction of ocean teniperntures. Jour. of Geophy. Iieseurch fornia Pish and Game, 46(2) : 195-198. Wiil--\-,. i Reid, J. 1,. Jr., 1939. Evidence of South Equatorial Counter- current in the I’acific Ocean. Nature 184, July 18, 1939. Sweeney, 13. AI., F. T. Haao and .J. JV. Hastings, 19.59. Action spectra for two effects of light in luminescence in Gony- Robinson, 31. K., 1960. Indian Ocean vertical temperature sections. Deep Sea Research, 6: 249-255. aitln~polyedra. Journal of Geneivl Physiology, 43 (2). Rotlen, G. I., 1{)J9. On the heat alltl salt halance of the Cali- l’hrailkill, R., 1RfiO. ~OoplaIlktOll volumes off the Pacific Coast, forniu Current Region. Nears Foundation, Journal of Xarine 1957. U.S. Dept. Interior, Pish and Wildlife Service, Spec. IZesearch, 18 (1). S‘ci. Report: Fisheries KO.326, 57 pp. PART II SCIENTIFIC CONTRIBUTIONS

SYMPOSIUM ON FISHERIES OCEANOGRAPHY

MAURICE BLACKBURN, Editor

INTRODUCTION At the Eastern Parific Oceanic Confereiice of 1957, Mr. Towiiseiid Croniwell (Inter-American Tropical Tuna Commission) explained the need for a sym- posiuni for exchange of iiiforinatioii and views about the usefulness of oceanography in fisheries investigations, and was asked to suggest a scientific meeting in which such a symposium could appropriately take place. Mr. Cromwell was killed while on duty iii Julie 1958. At the Eastern Pacific Oceanic Conference of 1958, the matter was raised again by Dr. Maurice Black- burn (Scripps Institution of Oceanography, Iiiiiversity of California) and discussed by coiifereiice participants. As a result, arrangements were made to include a “Symposium on Fisheries Oceanography ” in the programs of the forthcoming meetings of the Americaii Society of Lininology and Oceanography (Pacific Division) and the American Society of Ichthyologists and Herpetologists, to be held as part of the 40th annual meeting of the Pacific Division of the American Association for the Ad- vancement of Science. Dr. Blackburii was invited to organize the Symposium and preside over it. The Symposium took place at San Diego State College, California, on the afternoon of June 17 and the morning of Julie 18, 1959. It took the form of presentation of papers (recorded), questions and discussion from the floor (not recorded), and summaries and comments by persons who had been invited to present them (recorded). Publication in this medium was arranged by one of the participants, Mr. Garth I. Murphy, with the consent of all concerned. The organizer of the Symposium wishes to thank all who participated in it, the office bearers of the societies who sponsored it, and the Marine Research Committee of the California Department of Fish and Game which published the proceedings. ____ 1 The heading of each paper states the organization to which the author belongs. An additional contribution of a general nature presented by Richard H. Fleming, Department of Oceanography, University of Washington, Seattle, is not available for publication.

PROBLEMS IN FISH POPULATION FLUCTUATIONS

OSCAR E. SETTE U.S. Bureau of Commercial Fisheries Biological laboratory, Stanford, California

Our chairman has defined ‘ ‘fisheries oceanography ” Of course the trouble with the theory of fishing is as meaning “any kind of oceanography required for that a fish population reacts to other things besides the appraisal or exploitation of any kind of organism fishing-things in the variable environment. We as- useful to Man”. This, as intended, is a very broad sunie these things average on-and sometimes they definition. My contribution to this symposium will be do. But often the time dimension for averaging out is confined to a discussion of only one particular set of illtolerably loiig. Aid of course the same thing is true problems that I believe to be most important in the of the ocean. Over a long enough period of time the utilization of fishery resources and afford most ex- average oceaii probably is an equilibrium ocean, but citing opportunities for oceanographic research- the measurements we take must be mostly those of an namely, fishery fluctuations. ocean in disequilibrium. I believe it call be said for By fishery I mean any harvesting of any living fishery biology, as it has been said for oceanography, ocean resource, and by fluctuations I mean any irregu- that there is a peculiar dreamlike quality character- larity in the amount or the quality of the harvest. izing our descriptions and discussioiis of the things Since most of our present coinmercial harvest from the that we are studying. sea consists of fish, my discussion, terminology aiid ex- As a fishery biologist I venture the further thought amples pertain to fishes, but most of the concepts are that it mould be easy to relax and eiijop this dream also appropriate to the invertebrates and some even to were it not for fluctuations. They have a nightmare-ish the flora of the sea. quality that jolts one into reality. Aiid I am reminded In considering the problem of fishery fluctuations I that this could be true for oceanographers as well. A have come to realize that there are a great many year ago this month a group of oceanographers (in- parallelisms in the discipline and the state of knowl- cluding the one who first voiced the phrase “peculiar edge in fishery biology with those in oceanography. dreamlike quality ’ ’) , meteorologists, marine biologists, By fishery biology I iiiean the study of fish populations and fishery biologists met at Rancho Santa Fe, briiig- as dynamic system-and by oceanography I mean the ing with them meteorological, oceanographic, biolog- parallel thing: The study of the oceaii as a dynamic ical and fishery data, to describe what happened in system. In defining each of these in this way I am the Northeast Pacific Ocean in 1957 and 1958 and ignoring a great deal that could be included, but I why. It was abundantly clear that there had been a believe I am retaining the central core of each. sudden and marked warming of at least the eastern The first parallelism is that there are only a few margin of the Pacific from Alaska to Peru, with an things we caii measure. In fishery biology we measure assortment of consequences to the oceanic biota. The only the fisherman’s harvest aiid the effort he puts results of pondering over these events mill be pub- into makiiig it, when we really want to know the size lished atid I will not here attempt to put them before of the fish population and how fast it is changing. In you, except to remark that we still do not know, with oceanography we caii only nieasure the temperature any precision, what happened, nor with any degree of and salinity when we want to know which way the assuraiice, why. water is running, aiid how fast. To be sure, in each Henry Stoinmel, in his book 011 the Gulf Stream, field we caii, by a one-time special effort, pet a direct has this to say about fluctuations: (Stommel 1958, measure (estimate). We can lay a tag-and-recapture p. 136) project-or a multiple-ship drogue expedition. But these are not routine, every-day events. For the most “ Many catastrophes of an economic kind, such as part we have to deduce what we want to know from the failure of the rice crop in Japan, or of a certain something else that we can measure. fishery, or years of unusual numbers of icebergs iii Second-and this is largely a consequence of the shipping lanes, are attributed to fluctuations in first-for fish populations we have only a theory to ocean currents. Very little is really known about tell us how a fish population reacts to fishing. This such fluctuations. It takes years of careful and ex- relates to a fish population living in a steady enviroii- pensive observation to produce even a very crude ment. In some few iiistaiices, and for a limited period description of them. The scientific programs of our of time, we find that a population really does react oceanographic institutioiis are not geared to long- approximately as the theory says it should. For oceaii- term prohlenis of this kind; there is much pressure ography we have more theories. But practically all re- for novelty, much temptation to follow the latest late to ail equilibrium oceaii. For some parts of the fad, aid a persistent though erroneous notion that ocean, sometimes, one of these theories, or a combina- all worth-while problems will ereiitually be solved tion of them, seems to explain approximately what by some simple, ingenious idea or clever gadget. A actually happens. well-planned long-term survey designed to reveal 22 CAT,IFORSIA COOPERATITE OCEASIC' FISHERIES ISTESTIGATIOKS

fluctuations in ocean currents would be expensive limits, the spawning population is large but the back- aiid time-consuming. It might even fail, because of pressure from the environmental limit depresses re- inadequacies of the tools we have at hand. But until production or increases natural mortality, or both, so this burdensome and not immediately rewarding that the annual increase is again small. At some level task is undertaken, our information about the fluc- of population size intermediate between these ex- tuations of ocean currents will always be fragmen- tremes, where the spawning stock is moderately large tary." aiid back-pressure from the environment moderately gentle, the annual increase is maximal. Of course, at Henry Stommel seems somewhat over~vhehned by any level, the population size will be in equilibrium the task of making the observations needed for de- when the annual catch equals the aiinual increase, but scribing and elucidating fluctuations iii the Gulf the aiinual harvest that can be sustained without dis- Stream. The task probably is even more difficult for turbing the equilibrium will be maximal at the level clescribiiig aiid elucidating the fluctuations in fish of population abundance that affords the maximum populations, because it is likely that the effects on annual increase. fish populations of physical and chemical changes in Fishery biological research has been directed mainly the enviroiirnent are mediated through several trophic toward determining this level of niaximum sustainable levels in the biota. We will come back to this later. yield, using the concepts embodied iii this theory. A It seems clear that the problem of fluctuations lies iiumber of mathenlatical models have been developed at the research frontier in both fishery biology aiid to express these concepts. They assume that the en- oceanography, and that it is going to be difficult to vironmental capacity is (mistant or fluctuates moder- break through the frontier. ately aiid randomly, and that the observed or com- Ilet us consider again, for a moment, the anatomy puted changes in recruitment and mortality and hence of fishery fluctuations. We can discern three principal annual yield are fuiictioiis of population size (or elements operating to determine the amount of the density) alone. These models have proved very useful catch : The abunclancr of the organism, its availability in studying the population dynamics for some fisheries. to the harvester, and the amount of harvesting effort. Teshall not here be concerned with the last of these. In other populations, including some that support The amount of harvesting effort is determined by very important fisheries, the effective birthrate, that economic conditions. Changes in the amount of effort is, the relative numbers reaching recruitment age, can be measured, though not easily, and its effects on varies widely from year to year, apparently without the amount of catch can be determined and dis- relation to the size of the spawning stock. This is counted. It is iiot a problem in biology or oceaiiog- knowii by studying the age composition of samples raphy. of the catch. From the data on age composition Abundance and availability, in contrast, present through sufficient number of years, it is possible to problems both in biology and oceanography. estimate the relative number of individuals surviving to fishing size or age from each year's spawning. I We suspect that availability varies widely. For in- shall call this relative number " year-class strength". stance, the failure of the albacore fishery from 1928 to 1938 was, we tltink, an availability phenomenon. I hare looked up a few data on relative year-class strength : The albacore population probably was as large as Successive The largest usual, bat did iiot approach close enough to the coast year classes was to enter the range of the fishing fleet. Western Atlantic Availability is a matter of distribution and behavior, m:ickerel ______among 14 15,000 times the smallest lh\tern Sort11 both on a coarse-grained pattern as in the albacore I'acific sardine ~__ "1 'io0 failure, and on a fine-grained pattern as in the school- Kotliak (hlnsli:~) ing of fish. It is highly iinportaiit to the strategy of herring -_~-_~___ 28 34 fishing and to the economy of the fishing industry. Southeast d1:isli:i herring _-_____ ~- Since I'rofessor T'da will, I think, discuss availability 20 13 in some detail later in this symposium. I shall coiifiiie These are all very gross estimates, and should be myself to fluctuations in abundance (population size). taken only to indicate that year-class strength often Fishing theory says that the annual increase in a varies through several orders of magnitude. population is a function of population size and en- Year-class strength obviously must be a function of vironmental capacity. If a popalatioll ' ' fills ' ' its ell- number of eggs spawned, or of survival through the vironnieiit, births and deaths are equal and the popu- egg stage, through the planktoiiic larval sta,o'es or lation is in equilibrium. When fishing takes place, through the post-planktonic juverlile stages, or a catch mortality is imposed, the populatioll is rednced conibination of these. No doubt irregularities occur below the environrneilt 's capacity, births exceed " nat- in all stages, but the evidence, still regrettably scanty, ural " deaths aiid the population tends to increase points to the survival through planktonic egg and toward the environmental limits. With very intense larval stages as being the most likely critical one in fishing and a very low population level, the reproduc- determining year-class strength. tive increase is near its maximum, natural mortality Where year-class strength fluctuates widely it is near its minimum, but the annual increase is loly be- not a function of population size and we must look cause there are few spawners. Then fishing is very to environmental causes. light and the population near the envirolillielltal How shall we do this? REPORTS YOLUAlE VIII, 1 JULY 1959 TO 30 JUNE 1960 23 Because it is difficult to speak in generalities, I a sense, we could think of a water mass as undergoing shall employ an hypothetical exatnple. Let 11s con- an evolution as to its physical and chemical proper- uider a fish population of a species, like the mackerel, ties and also as to the trophic succession in its biota. the sardine or the anchovy, that spawns in the waterb Perhaps more frequently than not, the water mass off California. may undergo dissolution instead of evolution and Through evolutionary time, in the members of suc.h such cases may indeed lead to failure of survival of a fish population, there must have evolved an internal its larval fish population. biochemical system and a related pattern of behavior In this connection, it is interesting that there have that dett>rmiiiesthe responses to the things sensed in been reported two instances in wltic*h it appears that the eiivironment. The behavior pattern must have a water mass was cItanged so rapidly that the con- bren so adjusted that it lead to suc(~essfu1reproduc- taiiied fish larvae died almost intniectiately. One in- tion through all of the time inyolved in its evolution. statice, reported by Joh!~Coltoil (1!159), was of larvae Otherwise thc species would be extiitct. Tf I may speak contained in a mass or parc~4of watrr at the southern tclcologically, thc members of a living fish population, edge of Georges Bank in whirh the temperature, ap- or at least the vast majority of them, must work out parently by mixing with Gulf Stream water, had problems of navigation and prediction to enable them risen rapidly enough to kill the fish larvae it CO~I- to fiiid and occupy a specific kind of water mass for taiiied. The other instance. reported by Donald Stras- spawning. burg (1959), was the occurrence of dead larval Frig- This water mass must have physical and chemical ate mackerel in plankton tows in Jiawaiian waters uti- properties which are tolerable to the biovhemistry and der circumstances suggesting water niixillg as the biophysics of the eggs to be spawned and the larvae cause. We do not know whether or not such sudden to be hatched from the eggs. Further, by the time changes can occur in really large enough water nlasses larvae have hatched out, and by the time they have to determilie the failure of a year class, but these fnlly absorbed the yolk sac, thiu water inass must (mi- reports are suggestive. taiii other orgaiiisnis suitable for the larvae to fced However this may be, it appears that one mode iipoii aid iti such coi~ceiitrati~iisas will permit the of attack 011 the problem of year-class strength Ivould larvae to get eiiough food for maintenance of metabo- be the study of the source, the life history, the move- lic wtivity and growth. 1)oubtless thtl larvae will ments aiid, perhaps the dissolution of water masses. require larger food particles or more of them as they This infers sea work to observe. with continuity grow larger. So this water iiiass must continue to through space and time, many properties of the water afford ail iiwreased and probably different diet aiid its biota. It suggests, further, that a joining of throng h the subsequent nioiitlis of planktonic larval the physical and biological disciplines might facili- cxisteiice. Finally, when the larvae iiictarnorphose into tate such a study. The joining of the laboratory ex- the juvenile stage and take up an active, rather than perimentalists would probably facilitate the study a passive, drifting cxisteiice, this water mass must still more. If we know what properties of the ell\’won- hale reached a place or a condition which is suitable merit the fish is able to sense, and how the fish reacts for juvenile existence. J2any marine species spawn to the things it senses, we would be led more quickly in the open sea at a co~tsiclerabledistance from the to observing the things which cause the fish popula- juvenile nursery groiiitds. For such a species, the tion to maintain or change its distribution, and if we partxiit must have prcdictrd the trajectory of the could identify in the laboratory the survival require- water mass over a period of weeks or inonths. ment of fish larvae, we would be led more quickly to To be sure, mtmy individuals of the populatiou limy our observing the critical events in the sea. fail to na~+ptcproperly or to predict accurately aiid Til effect, I ani proposing the joining of four dis- it seem5 that ofteii the whole population, or most of cipliiies : oceanography, fishery biology, marine biit, may be in error. Or, alternatively, the conditions ology and experimeiital biology. Possibly additional itt the sca may iii some scasons become so aitoinaloiis disciplines may bc reqnired. Certainly, instrument that thcw are no water inasscs snitable for the spwies systems for recording automatically, with time and spawtiiig or the survival of the eggs and larvae after space continuity, a number of properties of sea water spawiiing. 111 any event, there are many year-class aiid its motions would be of tremendous importaiwe. failnres or iic~~r-failiirc~sin our n.idely-fliictnating. fish- On the biological side, it is quite possible that fish cries. Such failures, of course, must be intc.rspersec1 pathology, especially the study of fish larva disciases, with enough SLKTCSSPS for the popiilatioii to be per- may be germane to this study. prtuated. For species with an adult life of several To speak in more general terms, it is my be1ic.f years, several aiinnal failures can happen in sucws- that in fishery oceanography the challenge and the sioii, ant1 apparently do happen in our widely fliictn- opportunity lie\ in studying the changing sea rathrr ating fish populations, without exterminating the than the equilibrium ocean, and in stiidying thc bio- spec. ies. logiiral conseqnences of the changes at the various I have spokcii of the water mass as a thing which trophic lex els. In speaking of “ cwnseqiiences” I mean maintai~isits integrity over considerable periods of to inclnd(3 not only the effects on the population niiin- lime. This probably do~snot happen often for water bers, which I have dwelt upon at some length, but masses in the surface la) er. Thry uiidoubtcdly iin- also on population distribution and behavior which dcrgo changes through insolation and through mixing Professor I’da will soon discuss with 11s. In the aggre- horizontally and vertically with adjacent waters. In gate this implies the necessity of observation of physical arid chemical properties of sea water, its motions therefore, that progress in the field does not depend and mixings, and the numbers, kinds and perhaps alone on research activities in the interests of devel- stages of the biota inhabiting the waters, all with oping and utilizing fishery resources, but will benefit spacc’ and time coiitinuity sufficient to describe the also from activities undertaken in the iiiterrst of events that take place and to investigate their inter- other important activities of mankind. relationships. The biological determinations probably An indication of trends in this direction is the would niorr readily be made with the aid of labora- reeeiit report of the National Academy of Sciences tory experimental determinations of eiiviroiinieiital Committee on Oceanography, which, among other requirements critival for sarvival of the various or- things, proposes a great augmeiitatioii of effort, much ganisms, particularly pelagic fish larvae. of it along the lines we have considered in this dis- It would seem that this amounts to an undertaking cussion. of such vast scale that it may be quite out of line with the importance of fishery resources to ow society. However, fluctuations of the ocean and of its plant REFERENCES and animal populations is a problem of significance Colton, J. R., Jr., 1959. A field observation of mortality of marine fish larvae due to warming. Limnology and Oceanogra- not only to fisheries. Modern military systems no phy, 4 (2) : 219-221. longer earl be based on the average state of the sea, Stominel, H., 1958. “The Gulf Stream. A physical and dyiiami- and modern meteorology is no longer uninterested in cal description.” ITnit ersity of California Press : 202. the possibility that the feed-back to the atmosphere Strashurg, I). W.,19.59. An inytance of natural mass mortality from anomalous sea conditions may have to be con- of larval frigate mackerel in the Hawaiian Islands. Journal sidered in extended weather forecasts. We can hope, du (’onseil, 24 (2) : X6-263. FISHERIES OCEANOGRAPHY IN JAPAN, ESPECIALLY ON THE PRINCIPLES OF FISH DISTRIBUTION, CONCENTRATION, DISPERSAL AND FLUCTUATION

MICHITAKA UDA Tokyo University of Fisheries INTRODUCTION PROPOSED PRINCIPLES Oceanographic studies of existing fishing grounds (1) Marine organisms are distributed according to and exploratory expeditions into new areas to dis- the variable enviroiimental (hydrobiological) condi- cover new fishing grounds have aided very materially tions whic.li they require for successful development. in increasing the efficiency of fishing, and have con- (2) The ba.;ic pattern of the fish shoaling curve in tributed greatly to the tremendous yield of the Japa- respoiise to normal enviroiimeiital conditions is shown nese fisheries in recent years. After the first investiga- by the probability curve which is modified by special tion of oceanic currents around the Japanese Islands sea conditions, such as a cold front, etc., or by the by drift bottle experiments started by Yuji Wada in composition (size, etc.) of the fish schools. 1893, Tasaku Kitahara founded a fisheries ocean- H. 0. Bull (1952) found ~iiarinc.fishes to be sen- ographical organization in Japan in 1909 and pro- sitive to temperature changes of 0.03" in his experi- posed (11118) the principle of fish assemblage near a ments. line of convergence (' ' Siorne "). This principle was further developed by M. Uda (1927-58), K. Kimura (1940-58), and others. A fishing ground may be defined as a locality or Alaskapol lock area where a fishery is economically conducted on large forms such as whales or on populations of fish, such as salnion, herring, sardine, and tuna, etc. Fisheries science, in dealing with fish shoals (or fish schools) on the fishing grounds, must determine the reasons for their general distribution, their eoncentra- tions in certain areas and their general dispersal. It is important, of course, that cwough research be available and that equipment be highly efficient in tracking down and capturing the school of fish. The fishing operations, both in amount of area covered and actual fishing done, are restricted by weather and ocean conditions. There are a iiiiniber of problems associated with the safe and efficient conduct of fishing operations such as methods of locating schools-fish finders, aerial scouting, observation of the oceanographic structure of water, etc. ; maritime meterological kaowl- edge arid information to prevent or at least effectively reduce the damage due to storms, abnormal currents, etc.; processing and transport to market of the fish caaught; the management of the fishery to assure continued maximum sustained yield ; and safe navigation. Fisheries science, from the view point of fisheries FIGURE 1. Optimum water temperature spectra of important fishes in oceanography, seeks to find the principles controlling Japan (Uda 1957). the abundance of fish in the four dimensional (x, y, z, t) field of fisheries and includes the theory of fishing Uda (1940, 1957) prepared a diagram (Fig. 1) of conditions, fishing areas or grounds, fishing periods or optimum temperature spectra for some of the import- seasons in any time section, as well as the variations in ant commercial fishes of Japan, in which the optimum fishing conditions in any spatial section. As a practical temperatures (range about 4"-5" C) are shown as extension, it seeks to develop fisheries forecasts or each characteristic "proper value" (e,) in the following predictions, not only of the fishing grounds and fishing equation. seasoils, but also of the distribution and the extent of (e - eo>2 N = N,e - the catvhes, in both good and poor years. With refer- 262 ence to the progress acquired by fisheries during the first half of this century, the following principles of where 6 is the standard deviation, a measure of the fish distribution, concentration, dispersal, and fluctu- range, e is the water temperature and e, is its mode, ation proposed by the writer, based on his research N is the number or catch of fish, with N, representing (1927-1958), are illustrated. its peak. When 6 is large it is called eurytherm, when 26 CA1,IFORSI.i COOPER-kTIVE OCE-iSIC FISHERIES ISVESTIGATIOSS small, stenotherm, arid in the range (0,,&2&)98 per- wliale grounds jiist north of the pack ice zone (11. cent of N are included. Kcla, 1931, K. Kasu, 1959) (Fig. 2). A similar probability curve for salinity ean be pre- (4) Warm and c~oldwater iiitrnsioiis into well-de- pared. veloped suitable water temperature zoiies bring about (3) The localizatioii of fish coilcentrations is deter- concentrations of fish and produce good fishing areas. rniiied by the narrowness of the optimuni watrr zone By the intrusion of mifavourable water niasses (e$., and each special three-dimensional oceanographic abnormally warm, saline, or cold, frc\h \I ater) the structure. Especially good fishing grounds correspoiitl fish also become conc~eiitrated.For example, the mete- to the zone of the oceanic fronts (boundaries) of water orologically abnormal onshore currcwt “ Ryutyo ’) masses, including “ siome ” or lines of coiivergenve, (rapid current or storm current) pr‘oclnces a heavy and to the zone of up-welling (area of divergence) and catc.h of vcllowtail, sardine, tuna, etc., to the coastal other factors, as stated below such as ridging, entraiii- set-net fishery. ment, eddies, turbulent mixing, etc. (3) According to IC Rranclt ’s theory (1899) of the (a) Kitahara’s law (first postulatd by T. Kitahara, development of the organic life cycle of the food chain, 1918, and extended by &I. Vcla, 1936, 1938) states that productive zoiies TI liicli are fertilized either naturally the “oceaiiic front” corresponds to the area where or artifically by certain physical and chemical proces- marine life is concentrated and where fishing is good, ses becwiiie potential areas for fishiiip or for accjui- having a “sionie” (line of convergence) on the ocean culture by having available ample supplies of the surface as an indicator. priniary nutrient substanws. (b) Kathaiisohn’s lam (first laic1 do\vn by A. Sathan- Physical factors important in cleyeloping productive sohii, 1906, and extended by others), states that highly areas iiiclude : productive arid consequently good fishing grounds are (a) Tnrbuleiit mixing, due. to waves aiicl currents, found in areas of up-welling. The Scripps Institution which has been observed in coastal areas or arouiicl of Oceanography has conducted hoiiie basic work on oceanic banks and islands (e.p., the shrimp grounds, the riiechauics of productive iip-welling in the Cali- etr., along the S.K. coast of Iiitliil ill the sniiiiiier fornia Current System. nionsooii period, as pointed out by S. I<. Panilrkar, (e) Topo=raphicall~--develope(lback-eddy systems 1!138). Productivity of the fishing grouiids of her- (near a strait, channel, peninsula, cape, island, estu- ring, cod, etc., in the coastal straits of British Colurn- ary-mouth, etc.) are rich feeding areas, and are good bia may be attributed to the misiiig effect of the fishing groiuicls for mackerel, squid, yellowtail, etc. stroiig tidal currents, similar to the straits in the (Uda, 1958). Iiilarid Seas of Japan. (d) Thermoantirlines or ridpings occur in some (b) Tiiternal mares (of the tidal period) as studied sub-surface fishing grounds (e.g.. tuna fishing areas in b>- C. Cox, IC. lToshida adT. JIorita in 1938 in the the Equatorial Parific) as pointed out by T. Croninell, southern sea of ?Japan. The frrtilizatioii of the tmia 1958 This is due to a variety of uiiclerwater up-well- growicls on the deep oceanic baiiks or sea momits may ing or some type of entrainment. be due to internal \I aves. I,. N. Cooper (1937) has sug- (e) Entrainment (J. P. Tnlly, 1952, 1936) in estii- gested that internal wa~s011 the coiitiiieiital shelf aries, oil the coiitinental shelf edge, insular shelf edge rnay contribute to the fertilization of the niack- edges, or near fishing baiiks, produces a highly fer- crel fisliilig grounds in the Crltic Sea aiid A. 12. Jliller tilized zone and produces good areas for fishing or (1930) has studied the mixing processes orer the shelf other aqniculture. edge. ( f ) Dviiamically-produced eddies alonp oceanic (c) Mixing by coiiwction due to winter cooling, fronts are rich feecling areas, supplied with an abun- which is important in the seas of higller latitucics. The dance of planktonic food aiid small fish. These attract productivity of the Okhotsk Sea, Bering Sea, Japan the fish which teiid to rerriain thhre and consequently Sea, Y(~1IowSea and China Sea delwnds largely on develop into suitable fishing regions. the winter rnoiisoon. (g) In the northern hemisphere, cyclonic (counter- clockwise) eddies representing cwld eddies. coiistitiite (6) Schooling of fishes (IL’) is a function of a good fishing areas in the marginal zones of up-welling special group of hydrobiological conditions (S), the ie.g., saury, whales, squids, etc., in the Polar Froiital spatial gradient (OS) and the rate of time variation Zone; albacore in the Kuroshio Front) ancl are asso- (S). Thus, N = fn (S,VS,S).Where continuowly sta- ciated with favourable water teniepartures. ble or uniform ocean conditions prevail, the conren- In the southern hemisphere clockwise eddies develop along the Antarctic Coiivergance as favourable tration of fish and consequently the development of good fishing areas cannot he expected. A marked (N. Lat.) (S. Lat.) spatial gradient in water temperature, etc., or a sudden change in thcse featurcs is a promising indication of a good fishing area or of a good catch. (7) Diiriiig the feeding migration, schools of fish seek out areas \there the food organisms are abuiidaiit aiid arrive nornial1~-at the time ~~lie11the food is FIGURE 2. Eddy formation at polar fronts, schematic (Udo 19590). abtiiidant. The food forms vary according to the par- migrations of fish. For example, the yellowtail fol- Sea herring areas) drive away fish schools and dis- lows along the 50-100 in. isobaths of the continental rupt fishing. Near the niargin of such discoloured shelf ; the sea bream prefers a fine sand and shell bot- areas we sometimes find good catches. tom near a bank or reef. Some fish tend to assemble (21) Long term cyclic fluctuations in coinmercial on banks or reefs, on the margin of sea-valleys or fisheries are the result of changes in the reproduc- canyons, on sea mountains, close to the continental tion, development, distribution or availability of fish shelf edge or near the coast (ground fishes and stocks as caused by cyclic environiental changes. The pelagic fishes). effect of such changes depends on the degree to which Such topographical or geological irregularities the conditions depart from those laid down by Uda bring about the localization or concentr*ation of fish, (1958) as the optimum conditioiis and defined by the as cmitrasted to the smooth, unbroken flat sea bottom. temperature spectra of Fig. 1. Fluctuations occur in Therefore, the exploration of the sea bottom by means all fishes whether they br associated with cold cur- of echo sounders, dredges, core samplers, etch., is im- rents (Group A in Fig. l),intermediate cold-warm portant in fisheries. cnrrents (Group K), warin curreiits (Group C) or (13) Fish which migrate in mid-water areas, i.e., coastal water areas (Group I), not shown in Fig. 1). iii the optimum temperature zone, are concentrated therc by the vertivally confining influences of the FLUCTUATIONS IN FISHERIES: LONG TERM upper and lower unfavourable temperature strata and TRENDS AND PREDICTION also horizontally by flows of less favourable waters (e.g., squids, mackerel, tuna). A. General Fish populations ( 1’) in nature repeatedly increase (14) Fish tend to move upward into relatively and decrease (acconipanied by extension and contrac- shallower strata for feeding and collect in those areas tion of their ranges) under some unknown and com- where the prey-food forms are concentrated. Feeding plex natural enviroiinierital c*onditions. The fishing in thew arms nsiially extends from twilight or dusk conditions and the resultiiig amount of catch (N) for (‘ ‘ Pu Mazume ’ ’) to dawn (‘ ‘ hsa Mazume ”) or sun- each fish species (sardine, herring, tiuia, etc.) fluc- rise, and during the late evening, night, and early tuate from year to year (in a periodic or some irreg- nioriiing is the best period for fishing (angling, net- ting, lining). The tinw of the turn of the tide (ebb to flood, or flood to ebb) is another very good fishing period, espclcially for angliiig. These above-mentioned periods correspond to the time whcn the fish are actively feeding, as indicated by echotraces and catch samples (e.g., I’da, 1937). (15) Generally speaking, coincident with the approach of meteorological disturbances such as ty- phoons, c+ycloiies and fronts, the fish present in coastal waters swim up c1osc.r to the surface and feed very actively. Consequently, both before and after such atmospheric distnrbanc good fishing occurs. On the high seas, howevw, the fish teiid to scatter widely at such times aid seek out new feeding areas. To follow such fish or locate them again is a difficult problem. (16) In accordance with the state of devcJlopnic.nt of the c.onr~ergenceand the temperatures prevailing FIGURE 3. Long period variations of the catches of important fisher (i.e., whether favourable or not) the productiveness (Uda 1952). of the fishing grounds and the development of the fish thereon nil1 differ for each specks present. TABLE 1 (17) During the season of the spring tide (full RECENT FLUCTUATIONS OF THE FISHERIES IN THE JAPAN SEA niooii, dark) the fish schools teiid to approach closer (b-bad; bb-very bad; g-good; gg-very good) to the coast (whales, tuna, etc.). (18) Stormy gales or severe inonsooils cause fish Period (interval of to collect to the leeward of islands and headlands, years) 1 Sardine 1 Tuna 1 Squid 1 Herring etc., to avoid the rough seas (yellowtail, saury, flying fish, etc.). 1 1868-1905. - ~.~.. . ~ ~

2 1906-1912- -..~. .. ~ ~

(19) After severe earthquake, hcavy storms, vol- 3 (1913-1917).-.- ~ ~ ~..~ 4 1917-1921.. ~.~..~~. canic eruptions, “ tunami ’ ’, etc., the fishing areas and 5 1923-1931 -...--..-- fishing conditions may become quite disorpariized and 6 1932-1940 .._....~~. altered. 7 1941-1948 ...... ~~~. 8 1949-1955.. ~ ~... ~. . (20) “Red Tides” and other abnormal “bloom- 9 195G-1957...... ~... ings” of plankton (eg., Phaeocystis in the North REPORTS VOT,T,'JIE T'III, 1 JT7LY 19.79 TO 30 JUNE 1960 29 ular way) with intervals of nearly a century, or (A), the intermediate group (E), the warm current 20-30 years, or 50-60 years (See Fig. 3 and Table 1.) group (C) and the cwastal water group (D) etc., and In general, the periodic. decay (or growth) of yield for each of them there are representative indicator can be expressed as: organisms. Groups A, R, C and appear in siicces- sion, corresponding to the order of optimum tempera- N = Noecrt cos (wt - /.I) . . . . (1) 2 tures arranged as " optimum temperature spectra " in recent years. The growth and flowering of the C The actual fluctuations of a fish population (or group appears in company with the derelopment of a catch) is the result of its natural fluctuation modified warm current. Contrariwise, the A group flourishes by some artificial factors (e.g. fishing, or some bad along with the development of a cold current (Table effects of industrial wastes) (M. ITda, 1957). 2). Following M.B. Schaefer (1954) : Considering the fluctuations over a long period of dP years, we can distinguish unstable periods and rela- -= f(P) -KPF ...... dt . (2) tively stable periods of at least several years' duration and at most several decades when referred to (where K, P, P, are all functions of environmental some standard level of fisheries yield, and moreover factors). periods when the sea regions are correspondingly un- In company with the movement of predominant stable and relatively stable. warm currents and cold currents, the zones of favour- The conditions in the regions of the sea on the route able and uiifavonrable catch shift and undulate from of feeding migrations, especially near the oceanic north to south or south to north meridionally (y- fronts, fluctuate greatly and show unstable (rich or direction) along the coast, and froin offshore to near- poor) catches. Regioiis of the sea near the spawning shore, or nearshore to offshore laterally (2-direction). grounds and along spawning migration routes show As the function of the movements in each direction, comparatively stable catches. 2, y, with velocities cl, c2 respectively, we have: The spawning grounds and nursery grounds move not only from south to north or from north to south, N = @ (x - clt, y - c2t)----__ - -___- (3) but niay vary in the relative percentage abundance The migration route of fish shoals approaches the of fish present on the grounds as we see in the cases coast in rich years and moves further away from the of sardine, herring, etc. The sudden change in the coast in poor years (eg. sardine, yellowtail, etc.) . long-term variation is brought about by the sudden The fishing gears suited to each fishing locality growth or decav of warm and cold currents (e.g. in change from set-net near the coast to purse seine in the years of 1923, 1941 and 1931). the intermediate waters, to drift gilliiet and longliiies Apart from artificial causes (overfishing, water offshore. pollution, etc.) , the ultimate cause of the fisheries We can find groups of iniportant fish species which fluctuations map be found in variation of reproduc- show positive and negative variations, respectively, tion potential, survival, recruitment and availability and also indicator species for each group (euphau- in relation to variations in environmental conditions. siids, copepods, Sagitta, porcupine-puffers, tiny horse Fluctuations in fisheries for some important pelagic mackerel, some kinds of jelly-fishes, etc.). fishes (clupeids, tunas, salnions, etc.) seem to occur Apparently there are certain biological associations 011 a world-wide scale which may correspond to global in the seas and oceans, i.e. the cold current group geophysical variations (climatic change or the fluctuation of oceanic currents, polar ice, precipitation and evaporation over the oceans or zonal regions). TABLE 2 Herring (F.1I.C. Taylor, 1955) along the Pacific PHASE LAG OF THE MODE (MAXIMUM) OF CATCH coast of Canada exhibited heavy larval mortality from CURVES AND THEIR OPTIMUM TEMPERATURE the offshore current when northeasterly winds pre- I vailed in April. Even in the case of bottom fishes Period I Period I1 Optimum water Fish species (years) (rears) temperature (haddock, cod, etc.) the effects of environment. especially the wind conditions at the larval stage, are very serions, as shown by H. B. Hachey (1955) __j A Herring ...-_---1944-45 3O-8'C and J. N. Carruthers et aZ. (1951). Dominant year- A Atka Mackerel..iT 1944-48 2°-60(-130C, young) classes of important fish populations (e.g. sar- 1951 ('52 dine, herring, bluefin tuna, etc.) and favourable en- 1947-50 vironmental conditions may occur over several (10 to 1951-52 1954 ('51- about 20) years, and the appearance of large-sized '52 Japan (old fish) population\ without the succeeding year- Sea) 1955 classes (small-sized fish) give warning of the ap- 1954-57 proaching end of good fisheries. Conversely, the increase of young and adults year by year may foretell C Bluefin tuna.. .. 1936-41 1 1956-57 C Albacore ...... 1935-40 1954-57 a growing fishery. C Skipjack.---.-- t 1936-38 1 1955-57 The inverse relation between sardine and anchovy D Anchovy...... ~- 1955-58 or others may suggest some law of balance of law of - alternate predominance in the seas, or correlational balance sheet between pisci-predator, plankton feeder, REFERENCES and benthos feeder. Brantlt, K., 1S99. iil)er der Stoffwechsels im Jleer. TViss. Bfeer- Concerning sardine, mackerel, herring, Pacific euitntei’s. K. 8’. .4bt. Iiiel, 4: 21:3-30. saury or other fish populations, we can recognize the Brett, .T. Ii., l!).SX. We must 1e:irn niore ;ilmut salmon. Wester?% fluctuation of the spawning grounds in the course of Fisheries 56(1 ). spawning migration routes, and the corresponding en- Ilull, 11. O., 1!)52. An ev:iIu:ition of our knowledge of fish be- hnviow in rel;ition to hydrogr:~phy. Xnpp. (lons. I’ernz. vironmental fluctuation may be the fundamental cause Internot. Esplor. Yer, 131 : X-23. of this fluctuation. A special phase for each prosperity Currnthers, J. K.,A. I,. I,;iwfortl, ant1 Y. 13‘. C. Yeley, 1!)51. of fish populations can be pointed out, as we see in Fishery hydrography : I)rootl-streiigth fluctuntions in v;irious Fig. 3, Tables 1 and 2. In conclusion, we should de- North Sen fish, with suggested methotls of prediction. fiiel fine the abundance and maximum sustainable catch as .Ileereuforsch. 8: 5-l.i. a function of environmental factors (W. E. Riclrer, Cooper, I,. €1. N., 19,Z.i. T)e;ith of ti (Yiannel herring fishery. 1958). Sew Ncietitist, May. Cromwell, T., 19iS. Thermocline tol)ogr;iphy, horizontal cnr- B. Examples of fluctuation rents :ind “ridgiug” in the eastern tropical I’acific. Iiiter- .-I rnericcrn Y’rop. Z’trizn. CO~IJI.,Nii11. 3(3) : 13.764. (a) Hachey, H. 13., l!)Xi. n’iiter rep1:iceineiits iind their significxnce Cold yenrs \17flrlJlyenis to a fishery. Z’crp. Mcrr. Hiol. Ocetriiogr., UeepSecc I(es., sni)pl. 1900-1910-1915 l!E&40 to 3: 68-is. (peal, : 1!)41-4!) 1!132-35) Hjort, .J., 19%. Fluctu:ition in year- (1!)40Ll9. (Title not given). A’ci Rep. Tl-hole Res. Itist. and herring fisheries in the past 50 years suggests to 14. us the effect of Arctic and sub-Arctic warming on Sath:insohn, A,, 1906. Uher die lietleutung vertilialer Wasser- land atid in the sea. I~e.wegnngenfur die l’rotlnktion des l’laiilitons im >leer. Ziyl. The historically big catch of sockeye salmon and Siichs. Gescllsch. Il‘iusrnsch., dbh. Math.-Phys. Iilasse, herring in British Columbia (Fravr River) in 1958 29(5). suggests local variation of environniental conditions. Okatla, AI., 192!). (Title not given). -1. Imp. Fish. Inst. I’anikknr, S. K., 1958. Ocr:inoprnpIiic:il and fisheries research Because of the pressures of fisheries and man-made in India. I’roc. 8th. I’oc. Sei. C”oti!/ress,3: 294-302. civilizatiou (water pollution, deforestation, traffic and Hicker, IV. E.,1!).5X. JIaxinium sustained yields from flactant- construction, disturbances, etc.) the complex fluctua- ing environnients and niisetl stocks. b. Fish. Res. Hd. Cnncrdn tion of fisheries in relation to environments (natural 15: !)!)1-1006. and artificial) should be studied. Seh:tefer, AI. i3., 1!64. Some aspects of the dynnmics of popn- Ititions irnport~itto the m~in:igement of commercial fisheries. Inter-Anier. Z’rop. l’uiia. Poi~t~ti.,Hnll. 1 (2):%i-T,G. CONCLUDING REMARKS Taylor, I”. H. C., l!) The 1’;icific herring (Clupen pnllnsij The effect of environmental conditions on the early along the I’iicific cowt of C:in:ida. Intcrnnt. Xorth I’acif. stages of life cycles of fishes is a very important factor Fish. C‘o)ii?~i.,Bull. 1: 105-28. in the study of fisheries and should be studied inten- Tully, J. I’., 1952. Sotes on the Iwhnviour of fresh water enter- sively in the future. Within recent years such studies iiig the sea. Proc. 7th I’ocif. Sei. Congress, 3: 267-88. concerning mackerel, sardine, herring and tuna, etc. Tully, T. I’., and A. J. I)otliniead, 1956. L’roperties of the water have been started. Fluctuations in fisheries for im- in the Strait of Georgiii, Iiritish C‘olunibin, and influencing portant pelagic fishes seem to owur on a world-wide factors. 6. Fish. Xes. Hd. (‘nnctdn, 14(X) : 241-31!). scale. There is a need for interiiatiorial co-operation in {Jtla, JI., l!U7. Relation Iwtwecw the d:iily catch of fish and the study of environmental variations. the meteorological elements. I. Rtntistic:il studies on the in- flr~enceof the motiou of cyclones npon fishing. J. Ivtp. B’ixh. Zmt. 23 (3) : 80-88. ACKNOWLEDGMENT 1936%.1,ocality of fishing center and s1io:ils of “k‘:ituwo,” cor- This paper is based on my fisheries seminar notes related with the contact zone of coltl and warm current. R~tll. and oceanographic notes pwsented at the Biological Jnp. NOC. Sei. Fish. 4(6) : 385-!)0. Station, Nanainio, R.C., during October 1958-May 19361). Fishing center of “Sanmu” (f‘ololnhis snirn) correlntetl with the head of 0y;isiwo coltl current. Bull. Jap. Soc. S’ci. 1959 (Uda 1959a, 1959b). I am deeply indebted to Fish. 5(4). Dr. R. E. Foerster, Dr. J. P. Tully, Dr. W. E. 1940. On the principle of concentration and dispersal of fishes. Ricker, Dr. F. Neave and Ilr. A. J. Dodimead for Syokubutu to IXhuto (I’lants ant1 animals), 8 (8-9). (In their suggestions and help. Japanese ) REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 31

19.72. On the relittion 1)etween the variation of the important 1938. General consideration on the fluctuation of Tusima cur- fisheries conditions and the oceanogrnphicnl conditions in the rent-s3 stem and its related fisheries conditions. Rep. Tusima adjacent waters of Jnpan. 1. b. Tokuo T-tiir. Fish. 38(3) : ('urrent Research Project, Japanese Fisheries Agency, 1: 363-X!). 501-38. 1959a. The Fisheries of Japan. Fish. Res. Bd. Canada, MS. 1!)54. Studies of the re1;ition between the whtlling grounds and Rep. Series (Hiol.) So. GSG:96, mimeo. the hydrogrnphicnl conditions. 1. Rci. Rep. T17haZe Res. Ittst. 9: 17:Mi. l!G9t). Oceanographic seminars. Fish. Res. Bd. Canada, YE. Rep. Series (Ocennog.) No. 51 : 110, mimeo. 1!)X Researches 011 the fisheries grounds in relation to the ITda, JI., and ill. Ishino, 1%8. Enrichment pattern resulting scattering layer of supersonic ware (Introductory report). from eddy systems in relation to fishing grounds. J. Tokyo J. Tokyo t-itiv. Fish. 42 (2). Cnit,. Fish. 44(1-2). 1957. A consit1er:rtion on the long years' trend of the fisheries Vda, JI., and E. Tokunaga, 1937. Fishing of Germo germo fluctu:itioti in rehition to sea conditions. Bull. tJap. SOC.Sci. (I,nc@p&de)in relation to hydrography in the north Pacific Fish. 23 (7-8). waters. 1. Bull. Jap. SOC.Sci. Fish. 5(5) : 295300.

FISHERIES OCEANOGRAPHY IN EUROPE

MAURICE BLACKBURN Scripps Institution of Oceanography, University of California, La Jolla Any review of the status and acconiplishmellts of of the currents, etc., was of the greatest importance fisheries oceanography would be incomplete without for the investigation of the problems connected with an account of fisheries oceanography in Europe, where life, that on the other hand, the study of the floating the science began and is still vigorous after more than organisms had particular worth for the solution of 50 years. hydrographic problems, and consequently that a sharp Fisheries oceanography may be said to have begun line should never be drawn between these two main with the establishment in 1902 of the International divisions . . ." Council for the Exploration of the Sea, by representa- Most marine scientists know in a general way holT, tives of north-west European governments. The form mainly under the aegis of this Council, the north-west that fisheries oceanography was to take was foreseen European countries have eo-operatively and individu- and described by the Council in its first Administra- ally planned, performed, discussed, and published an tive Report (1903), in the following words : immense aniount of oceanographic work which was, or " It was thereby shown that the research work might was hoped to be, pertinent to fisheries probleius. The best be divided into two main divisions, of which the remarks in this paper will be confined to the investigations in the region between the English Channel, the Barents Sea, and Greenland. There has, of course, been much work also off the Atlantic coasts of France, Spain, and Portugal, which are in the International Council's area, and iii the Mediterranean and Black Seas, which are not. Figure 1, the well-known surface current chart of IIelland-Hansen and Nansen (1909), shows how the warmer more saline North Atlantic water enters the region, mainly through the Faroe-Shetland Channel, as the Norwegian Current, while colder less saline Arctic water flows out, between Greenlaiid and Ice- land, as the East Greenland Cnrrent. An ontflow of lowsalinity water from the Baltic contributes to the Norwegian Current. The shallow North Sea receives Atlantic water froin the north, around Shetland ; and to a less extent froin the south, through the English Channel, as indicated in Figure 3. Another kind of suinniary of the hydrography of part of this region is the chart of Dietrich (1950) showing natural regions based on distribution of salinity and temperature (Fig. 2). These figures deinonstrate the great differences in the environment of the fishes of the region. Much of the early fisheries oceanographic work consisted of col- lection, description, and analysis of material in snf- ficient mass to show what species or populations of fish reacted to what properties of the rnvironnient and in what ways. The resnlts have been suniinarized by Tait (1952, Chapters 1, 2). They gave rise, among other things, to a few rather useful, though not entirely dependable, rules ; e.g. that fish populations breed at FIGURE 1. Surface currents in the Norwegian Sea (from Hellond- particular temperatures (Orton 1920) and that breed- Hansen and Nansen 1909). ing migrations are contranatant (E. S. Iiiissell 193i). one had in view the physical conditions of the sea, the Meiition ninst be made at this point of the aqiiariinn other the biological-more especially in regard to the experiments of Ball (1952), n hich denionstrated the animals most useful as humail food. Katurally it was ability of fishes to perceive and act purposively to seen from the beginning that the study of the physical changes as low as 0.03" C., 0.2 O/oll of salinity, and 0.05 conditions, of the chemical nature of the ocean waters, uiiits of pH, in the water arouiid thein ; this niade the co~iclusionsfroin the field work more acceptable. 1 Contribution from the Scripps Institution of Oceanography. I I - - -mom

FIGURE 2. Hydrographic regions in the North and Baltic Seas (from Dietrich 1950).

This kind of work was more straightforward than theory were : that year-classes may vary greatly in size that which came later. In the period between the wars because of differing natalities and mortalities of young several lines of fisheries work were progressing in a from year to year ; that the progression of these rich, way that seemed to leave the hydrographers and plank- iiiediuni, and poor year-classes through the older fish- tologists with a minor role to play as far as practical able part of the population has a decisive effect upon fisheries problems were concerned. The hydrographers the numbers present and catchable each year ; and and planktologists found plenty to do on their own that this effect is predictable in detail if the initial account, but their connection with fisheries biology year-class strengths, aiid average growth-rate of the was temporarily weakened. fish, are known. Althongh IIjort perceived the prob- One reason was that it had become clear that the able significance of eiiviroiimental factors in causing condition of many important fish populations, espe- the differences in the. year-(.lasses, he pointed out that cially demersal populations, could be improved by the causes woulcl have 110 great importance for prac- restricting man's fishing (E. S. Russell 1942). This tical forecasting if oiie c.oiild niea\iire the differences meant: that there was no need of oceanography to themselves before the year-classes became iniportaiit help fishermen find these fish, since the fishernien in the fishable segment of the poprilatioii. It turned were already too efficient at that for their own good; out that the differences could be so measiiretl ill sev- and that oceanography's promise to understand and eral populations, by study of .age-c.onipohitioii of 1111- predict changes, nothing more, seemed less attractive inerous samples of fish taken before or just at their than the promise of fisheries biology to control changes entry into the fishery, and that useful predivtioiis by regulating fishing. This affected the work on plaice, coiild be made. This was done particularly for herring, haddock, hake and other species, and the programs of in south-east England (Hodgson 1932) and elsewhere. the countries coiiceriied-especially England, Holland, Til proposing his theory of natural fluctuations, Belgium, Denmark, and Germany-and continues to Hjort ( 1914) had considered and virtually rejected do so. The work on the pelagic herring and the semi- the idea that the coastal food-fish penetrate the ocean pelagic cod mas not greatly affected. waters. Coastal fishermen everywhere tend to believe Another reason was the well-merited popiilarity of that fish must be offshore when they cannot be found the theory of iiatnral fluvtuations in fish populations inshore, and persistently ask their scientists to investi- according to Hjort (1914). The essential points of the gate these waters for them. We will see later that the REP(3RTS VOLUME YIII, 1 JU 5

Norwegian fishermen were partly right in this belief, and that Hjort was wrong when he disagreed with it. However his conclusion was reasonable in the light of the evidence available to him, and it is salutary to give his words :

l1In the course of the Norwegian fishery investigations carried out under my supervision, I have en- deavoured in various ways to discover whether any fish move out beyond the coastal banks . . . Numerous experiments have been made with drift net and floating lines, for the most part, however, with negative result. ” He then explains that he did find Sebastes, the red- fish, in many places; some herring in deep water between the North Sea and the Faroe area, but virtually none in the main part of the Norwegian Sea; and little or nothing in the way of cod, haddock, coal- fish, and catfish ; and he continues : “Experiments of this nature are, however, by no means easy to carry out, and negative or mainly negative results are scarcely sufficient to warrant the conclusion that the species in question do not occur in any quantity in the waters investigated; there is always the possibility that the fish might occur in shoals, which it would be a matter of merest chance to encounter in so great an expanse of sea. One thing at least is certain; we have no other grounds for sup- posing the existence, in any considerable numbers, of coastal fish in the deeper parts of the Norwegian Sea, FIGURE 3. Generolized picture of distribution of plankton indicators beyond (occurrences of Sebastes and restricted oc- in the North Sea area: showing occurrence of Sagitta elegans, S. sefosa (SET.), and other species; arrows indicate general water circulation currences of herring noted above). ” (from F. S. Russell 1939). This helped to divert attention from the ocean waters, and was another disservice to the kind of fish- or mixed upwards, in the Channel approaches, during eries oceanography that the founders of the Inter- the nineteen-twenties. Observing that this would be national Council had envisaged. facilitated by the elevation of a layer fairly rich in We now turn to what some fisheries oceanographers, phosphate by a few hundred metres, he supposed the physical and biological, did while some fisheries biol- arrival of a suitably large body of water at greater ogists were specializing in the above-mentioned ways ; depth, formed by cooling and sinking of water be- and how some common ground has now been reached tween Faroe and Greenland. Winter temperatures in again, especially in herring and cod research. those regions were particularly low in several years There were workers in England and Scotland who ending about 1921, a year before the beginning of believed that the varying influx of Atlantic water to the phosphate record at Plymouth. The effect of the the North Sea and English Channel was the key to heavy water in keeping the nutrient-rich water ele- many fluctuations in fish populations. vated in the mouth of the English Channel is supposed Plymouth, in the western English Channel, had a to have continued for some years after the recruit- herring fishery until about 1938 ; its extinction began ment of the heavy water ceased. in 1931 when the first of an unbroken series of poor Figure 3 shows that Ragitta elcgans, the plankton- year-classes appeared. The same year marked the indicator of good conditions for fish in the English beginning of a fall in the standing crop of planktonic Channel, is also distributed in the North Sea in a young fish generally, and the beginning of a large way that suggests an association with Atlantic water drop in the winter maximum of. inorganic phosphate. entering through the Faroe-Shetland Channel (F. S. It was also the period at which 8ugittu eleyans was re- Russell 1939). The idea of plankton organisms as placed in the plankton by &‘. setosa. It was concluded indicators of hydrographic conditions has been so that the influence of Atlantic ocean water ~ponthe popular in England and Scotland that plankton has Channel, and with it the possibility of renewal of nu- been routinely collected, along regular steamship and trietits to the levels of the nineteeii-twenties, had re- weathership tracks radiating out from Britain, in crdrd (F. s. Russell 1939). There has beell no definite most peace-time years since 1932. The convenient improvemeiit siricde that time. Hardy Continuous Plankton Recorder is used. Figure Cooper (1955), rioting no evidenee that Atlantic 4 shows the coverage now being obtained (Hardy surface waters have ever been rich enough to account 1956). for the phosphate levels found off I’lymonth before The surface currents of the Northern North Sea 1931, coiisitlcrcd various possibilities by which deeper region are shown, from the drift-bottle work of Tait nutrient-rich Atlantic water could have been carried (1937), in figure 5. Tait (1955) has also studied vol- L

3 6 CA1,IFORR‘IA COOPEIiAI’IVE OCEANI(: FISHERIES IXVBSTICATIONS

FIGURE 5. Surface currents in the northern North Sea (from Tait 1937).

ery arid herriiig-oceanography program in the middle of the Norwegian Sea, from Iceland to Spitzhergen (Fig. I), in the very region where IIjort (1914) thought no appreciable amount of herring existed. Location of the resource was primarily the result of equipping a new research ship with asdic (sonar) and using it in regions where a ne157 generation of Norwegian and Icelandic biologists had concluded, from tagging experiments, that herring must occur. FIGURE 4. Scope of Continuous Plankton Recorder survey in 1953 But oceanographic work accompanied these surveys (from Hardy 1956). from the start, and is producing a picture of herring shoals concentrated, in summer, along the eastern Lime transport of oceanic water (salinity > 34.99 O/Oo) boundary of the cold East Iceland and East Green- through 68 sections made across the Faroe-Shetland land Currents (Devold 1952, Marty 1956). Plankton Channel from 1927 through 1952, noting various sea- and productivity ( CI4 method) observations indicate sonal and annual changes. Scottish biologists have that this is the best part of the region for production been trying to link such changes with fluctuations in of herring food (Berge 1958). Much has been added distribution of plankton organisms on the above- to our knowledge of currents in the region (Alekseev mentioned sampling lines around the north of Britain, and Istoshin 1956). and with events iii the herring fishery off the northeast coasts of Scotland (Glover 1955). Another aspect of plankton work related to herring 2000 ’ must be mentioned, the aggregation of the herring on to the plankton, especially the copepod Culanzcs, which it eats. This has been investigated several times in different areas, frequently in the hope that the relationship would prove sufficiently close to enable fishermen to locate herring from their own plankton surveys with simple collecting instruments. This has happened sometimes but not almays (e.g. Hardy, Lucas, Henderson, and Fraser 1936, in the North Sea). It is now clear that a close positive relationship between herring and food is much more likely to occur in some situations than in others, as shown for example from the work of Manteufel (1941) in the Karents Rea, where the study was simplified by the fact that only one brood of Calaniis occurs per year. The relation- showtog how Ihc changmg bloma5$ o$ CaldnUS 16 rtlAtrd ships between this brood and the herring are sum- to the cewel&t80* bccwcen kiAnu.5 dndht Catch of herrmgc marized diagrammatically in figure 6. An4 +o the scafhrtng And 5halmg of thhrHw&g

Finally, for the herring, we must note the appear- FIGURE 6. Summory of Calanus-herring relationships in the Barents ance in the last decade of an important herring fish- Sea (from Manteufel 1941). The amount of work done on oceanography in rela- of wintc.r when they have resumed a bottom-living tion to cod has likewise been large. Much of it has existence. been planned around the belief that cod are particn- The rcyort of the first two years of this program larly selective about temperature, which arose in part provides interrstitig rcatling for those intrrcsted in through the work of Thompson (1943) at Newfound- Europemi fi4ieries oceanography from the point of land. The fact that the cod extended its range north- view of outlook, organization, facilities, aiid mothods wards, from Iceland well into Greenland waters, after (Graham et al. 1954). the middle nineteen-twenties when the maters of this There have been many other investigations 011 the region were warming, gave further support to the effects of oceanographic variables 011 Enropcan fish, general idea and encouraged the expansion of cod some conclnsiw and some not. Much of the work has research generally. Biological and hydrographic obser- been done ill the " Transitioii Area" at the eiitrance vations regarding this extension of the cod's range of the Baltic (Jensen 1!)52), on niatiy species. Other were given by Hanseii (1949). Fishery records of the references to such work, and to oceanographic work early nineteenth century suggest previous invasions 1ec;s intimately connected with fisheries at the present and retreats of cod in Greenband waters. Sea tempera- time, may be found in papers in the Tnternational ture data do not go back quite so far, only to 1876; Council's "Rapport Jubilaire" (19.52) , and in re- they show the recent warming very well, of course, views by F. S. Russell (1952) and Fleinitig and and also the preceding cold period about 1920 to IJaevastu (1936). It is proper also to mention the which reference has been made (Smed 1949). iinniense contribution made over about 20 years by There was ail unexpected increase in the abund- Britaiii, to the fisheries oceanography and general ance of cod in the Baltic Sea in the nineteen-forties oceanography of the Antarctic Ocean, through the which Alander (1952) has explained in terms of a "Discovery Committee ". The fisheries in this case series of rich year-classes associated with increased were for whales, whose relationships to environment salinity of Baltic water. have been sumniarized by IIartlp and Giinther (1935). Returning to Arctic cod, we note an important in- The efYort continued as a general oceanographic sur- vestigation by England, commencing in 1949, of the vey of the A4ntarcticand adjacent seas, including the distribution of cod in relation to the environment Peru and Benguela Currents, and the ships and per- around Rear Island, between Norway and Spitz- ~oiinelfinally became the nucleus of the National In- bergen. Bear Island lies in the region where the warm stitute of Oceanography (Mackintosh 1950). Norwegian Current divides and cold water penetrates We may return to the matter of whether or not it from the north-east (Fig. 1). The intention was to is iinportaiit to understand the causes of year-class study relationships of cod to temperature, not so fluctuations. It is now generally held that they should much to predict abundance changes of the whole pop- be known, if oiily because it is not always possible ulation through year-class strengths as to predict the to measure the changes properly by sampling juve- location of any payable concentrations of the fish (i.e., nile fish. Carruthers has for some years approached similar in motive to some of the herring studies). this problem ex hypothesi; he has assumed that the The results of this work are still appearing. The differences are niaiiily differences in mortality of results available are somewhat inconclusive on the young and that they occur in the North Sea because value of temperature Observations to fishermen in the of the vagaries of currents that may carry the young area as a whole, but they point to some situations in into favorable or unfavorable situatioiis ; and he has which measurement of bottom temperatures would examined the relationships between year-class sizes facilitate location of fish (Lee 1952). It was shown and measurements of the winds held responsible for that paying quantities are rarely caught in water the currents, and has found some satisfactory agree- colder, than l.75"C7 except in summer when the fish ments. The idea is shown in figure 7 which is from are feeding heavily to the east of Bear Island and one .of his earlier papers (1%8), showing the cor- may be found down to -0.5"C. Another finding was r*espondence over a series of years between the rela- that on grouiids west of Bear Island, in early sam- tive size of haddock year-class and sum of east compo- mer and in autumn, Atlantic water touching the Bear nents in the wind, in the North Sea. Since the war Island bank can give good catches with bottom tem- he and his colleagues have produced similar arrays perature between 3" and 5°C. For the area as a whole the range of temperature associated with cod is wide -about -0.5" to 5°C-in the summer feeding season, and narrow-about 2" to 4"C-in the winter non- feeding season. Even in the winter there may be waters of "suitable " temperature without fish. Another member of the Bear Island research team has attempted to explain the distribution of cod in the area in a different and more comprehensive way (Trout 1957). The hypothesis, derived from tagging experiments and other biological observations on the cod, is that cod are carried along with currents dur- FIGURE 7. Comparison of annual changes in haddock year-class size, ing summer when they leave the bottom in response pressure gradient, and sum of east components in wind; North Sea to light, and work back against currents in the dark data (from Carruthers 1938). 38 Ch1,IFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS of data, regardhg year-class strength and wind, for 1951. An attitude on “fishery hydrography.” J. Mar. Res. IO: other species including herring, plaice, cod, and sprat 101-18. 1953. Review of Tait, J. B., 1952 (q.v.). J. Cons. Perm. (Carr,uthers, T,awford, aiid Veley 1951 ; other refer- Znternat. Esplor. Mer, 19: 211-3. ences in Carruthers 1954). 1954. Some inter-rrl;itioiiships of oceanography and fisheries. As a result of this work Carruthers (1951) has -4 rch. Jfeteorol. Geophys. Bioklintatol. B., 6: 167-89. taken what he calls “An attitude on fishery hydrog- Carruthers, J. S., A. 1,. I,awford, and V. F. C. Veley, 1951. raphy” which amounts to this : oceanographers who Fishery hydrography : brood-strength fluctuations in various are not obliged to produce practical results may study Sorth Sea fish, with suggested methods of prediction. Kid. Xeeresforsch. 5-li. what they like, but fisheries oceanographers should 8: 1953. Winds ant1 fish fortunes. J. (‘ons. Perin. Znternat. Ezplor. study winds. Wind data, he emphasizes, are cheap. Zer, 18: 364-8. Warming to this theine in other papers (1953, with Conseil Permanent Internationnl pour I’Exploration de la Mer. Lawford and lTc1ey; 1954) Carruthers asks whether 1952. Rapport Jubilaire. Rapp. Cons. Perm. Znternat. Erplor. the complex kind of work now thought of as fisheries Jler, 132: 85. oceanography, the kind we hare been considering, can Cooper, 1,. H. X., 1955. Hypotheses connecting fluctuations in do the fisheries any good within reasonable time, or Arctic climate with hiological productivity of the English Channel. Pnp. Mar. Hiol. Ocennog., Deep-sea Research, 3, ever. IIe says (1953) of Tait’s work: suppl. : 212-23. ‘‘Tait is well known of course for the meticulous I)evold, F., 1952. A contribution to the study of the migrations detail of his work carried out orer many years on of the Atlanto-Scantlian herring. Rapp. Cons. Perm. Intertiat. water nioyements in the sea areas to which attention Erplor. Jler, 131: 103-7. has to be paid by those whose task it is to study the Dietrich, G., 1950. Die natiirlichen Regionen von Nard- und Ostsee auf hydrogr:il)liischer Grundlage. Riel. dleeresforsch. fish of the North Sea. Unless the reviewer has got 36-69. quite a wrong impression, Tait wants more detail 7: Fleming, R. H., and T. Laevastu, 1956. The influence of hydro- still . . , If so, we must wait a goodly time yet be- graphic conditions on the behavior of fish. F. A. 0. Fish. fore ‘ Fisheries Hydrography’ can pay the dividends Rull. 9: 181-96. which the biologists seem to expect. There is at least Glover, R. S., 19Xi. Science and the herring fishery. Advance- the chance, howerer, that fruitful associations of the ment of Science, 44: 426-34. kind needed may be worked out between fish fortunes Graham, AI., 0. C. Trout, R. .J. H. Reverton, J. Corlett, A. .J. Lee, and R. W. Blacker, 1934. Report on researrh from the aiid environniei~talconditions in a more simple way Ernest Holt into the fishery near Bear Island, 19.4‘3 and 1‘360. iiivolviiig less industry and less cost.” Ministry dgric. Fish., Londott, Fish. Znvest. II, 18(3), 85. However, the pattern of fisheries oceanography iii Hansen, 1’. Jl., 1939. Studies on the biology of the rod in Europe has not yet changed much. The work of the Greenland waters. Kapp. Cons. Perm. Isterntrt. JSsplor. Jfer, last decade has r,eflected the following recommenda- 123: 1-57, tion made by the Consultative Committee of the Hardy, A. C‘., 1933. “The open sea-its natural history : the In- world plankton.” IIoughtoii Aiifflin Co., Boston : 335. ternational couiicil in 1951, after a special scientific of Hardy, A. C., ant1 E. R. Gunther, 1935. The plankton of the meeting on “Fisheries Hydrography ” which fea- South Georgia whaling grounds and adjacent w:iters, 1926- tured many of the above-mentioned papers. W25. “Discovery” Rep. 11 : 1-456. “As the proceedings of this meeting have shown Hardy, A. C., C. E. Lncas, G. T. I). Henderson, and J. €1. that a combination of biological, hydrographical, me- Fraser, 1936. The ecological relntions between the herring teorological, physiological or statistical researches has and the plankton investigated with the plankton indicator. J. Mar. Hiol. Assn.. l-. : 147-291. furthered our knowledge of biological productivity, R. 21 Helland-Hillisen, I%., arid F. Nailsen, 190‘3. The Norwegin11 Sea. of the habits of the fish stocks, and also the fisheries Its physical oceanography basetl npon the Norwegian re- themselves, it is recommended that researches along searches 1000-1904. Rep. Xortr. Fish. Mar. Invest. 2(2) : 390. these lines should be continued.” Hjort, .J., 1914. Flnctuations in the great fisheries of northern Enrope viewed ill the light of biological research. Rapp. Cons. REFERENCES I’erm. Znternat. Bspror. .Ifel+, 20: 1-228. Hodgson, W. C., 1932. The forecasting of the East Anglian hl;inder, II., 1922. Fisheries hydrography in the Baltic. Xapp. herring fishery. 6.-4 TliIJl. >:col. 1 : 108-18. (’ojis. Perm. Iiaternnt. E’rplor. Mer, 131 : 61-2. Jensen, A. J. C., 1952. The influence of hydrographical factors hleksec~,A. I>.,and 13. V. Istoshin, 1966. Chart of constnnt 011 fish stocks and fisheries in the transition area, especially cnrrents in the Sor\vegian and Greenland Seas. Trudy Polar on their fluctuations froin year to year. Rapp. Cons. Perm. h’ci. Res. Inst. Sea, Fish. Oceanog. Jlirrntansk, 9, in Russian : Internat. h’xplor. Uer, 131 : 51-60. I<:nglish transl. in C.S. E’ish 1ViZdI. New., Spec. Sci. Rept., Lee, A. J., 1952. The influelice of hydrography on the Rear Fish. 327 (1!159) : 69-76, Island cod fishery. Kapp. Cons. Perm. Internat. E’cplor. Mer, I3erge, G., 1958. The prinmry production in the Norwegian Sea 131 : 74-102. in June 1974, nie:isured by an adapted l‘C Technique. Rapp. JIackintosh, N. A., 1950. The work of the Discovery Committee. (‘ons. Iieriii. Znternat. E’rplor. Jler, 144: 85-91. I’roc. 12oy. Roc. London, E, 137: 137-52. Ilrill, H. O., 1952. An evaluation of our knowledge of fish be- li;1ntellfel, R. I)., 1941. I’latrkton and herring in the Barents htivioiir in relation to 1iytlrogral)hy. Kapp. Cons. Perm. Sett. Y’rans. Knipovich Polar Sci. Inst. Sea Fish. Oceunog. Intertint. 13splor. Iller, 131 : 8-23. .jlz~r?tinnsk,7: 12.5-218, in Russim. (’:wrnthrrs, .J. S., 1938. Fluctuations in the herrings of the Jlnrty, .J. J., 1’356. The fundamental stages of the life-cycle of East Ang1i;in :iutunin fisher>-, the yield of the Ostend spent the Atlantic-Scandinaviarl herring. Trudy Polar Sci. Res. herring fishery, and the had(1oc.k of the North Sea-in the Inst. Sea j’iuh. Oceanog. Mnrmansk, 9, in Russian : English light of relevant wind conditions. Rapp. Cons. Perm. Internat. tr;rnsl. in l’.R. Fish Wildl. Serz:., Spec. Sei. Rept., Fish. 327 Eqlor. Jfer, 107: 10-15. (1959) : 5-68a. REPORTS VOLUJIE~VIII, 1 JULY 1969 TO 30 JUNE 1960 39

Orton, J. H., 1920. Sea-temperature, breeding and distribution Tait, .J. n., 1937. The surface water drift in the northern and of marine animals. b. Jlar. Bioi. dssn. U. K. 12: 339-66. middle areas of the Xorth Sea and in the Faroe-Shetland Russell, E. S., 1937. Fish migrations. H~ol.Rev., Cnnib. Phil. Channel. Fish. Bd., Scotland, Rci. Znvest. 1937 (1) : 1-60. Soc. 12: 320-37. 195%.“HJ drography in relation to fisheries.” Edward Arnold, 1912. “The overfishing prohletn.” IJniversity I’ress, Camhriclge : 1,ontlon : 106. 130. 1975. Long-term trends and changes in the hydrographj of the Russell, E’. S., 1939. Hydrogrnphical and hiological conditions Faroe-Shetland Channel region. Pap. Xar. Hiol. Oceanog., in the Sorth Sea as indicated by plankton organisms. d. Deep-sea Research, 3, suppl. : 4S2-98. Cons. Perm. Internat. Esplor. Ver, 14: 171-92. Thompson, H., 1913. A biological and economic study of cod 1952. The relation of plankton rehearch to fisheries h>drogrii- ( G0du.p callarias, I,.) in the Sewfoundland area, including phy. Rapp. Cons. Perm. Internat. Explor. Mer, 131 : 28-34. Labrador. h-ezcfoundld. Res. Bull., Fish. 14: 160. Smed, J., 1949. The increase in the sea temperature in northern Trout, G. C‘., 1937. The Rear Island cod : migrations and move- waters during recent years. 12app. Cons. I’erui. Internat. ments. Ministry dgric. Fish., London, Fish. Znvest., 11, 21 Explar. Mer, 125: 21-3. (6): 51.

TUNA OCEANOGRAPHY PROGRAMS IN THE TROPICAL CENTRAL AND EASTERN PACIFIC

MILNER B. SCHAEFER Inter-American Tropical Tuna Cornmission

As recently as a decade ago, knowledge of the beginning that such investigations could not be very biology arid ecology of the tropical tunas was almost fruitful unless we obtained some understanding of the entirely lacking. In the subsequent period substantial ecology of the tunas, in consequence of which investi- research programs have been initiated, in the Central gations of the physical, chemical, and biological ocean- and Eastern Pacific, which have gone a long way to- ography of the Eastern Pacific have constituted an im- ward elucidating the ways of life of these very impor- portant part of the Commission’s continuing research tant pelagic fishes. The major features of their life program and have been indispensible in understand- histories have been described in a general way, and ing the spatial and temporal variations in the occur- fair progress has been made toward understanding rctice of the tunas. The Commission, fortunately, their population structure and population dynamics, established its headquarters at the Scripps Institution in some regions at least. At the same time fairly broad of 0c.eanography (810) which has made possible very studies in physical, chemical and biological ocean- fruitful cooperation with the staff of that institution ography have led to some understanding of features of in stndyitig the owanography of the Eastern Pacific their oceanic environment which control, in part, their in relation to the ecology of the tropical tunas. distribution and abundance, or, at least, have brought By 1957, the researches had progressed sufficiently out some relationships between the occurrence of the to indic*ateto the tuna fishing industry that a more de- tunas and features of the eiiviroriment which appear tailed iuiderstanding of some featiires of the ocean- to be of predictive value. ography of the Eastern Pacific conld be expected to be In speaking of the “tropical” tuuas, I refer to the helpful ill directing the operations of the fishing fleet yellowfin tuna, Tliuizizirs (X~otlzic)iw~rs) niacroptcrirs, to locate aiid vatch tuna more rapidly. The IT.S. Fish and the skipjack, Katsuir~o)itrspclancis, both of which and Wildlife Service, therefore, two ycwrs ago pro- ocwn right across the equatorial Pacific b(.tween, ap- vidd funds for a group of scientists of the Scripps proxiniately, the isotherms of 30°C. In these tropical Institution (headrcl by Dr. AZauriee Blackburn) to waters then. also ocviir bigeye tuna, TIiiotni~s(Para- coniiii~~nceworking intensivrly 011 this particiilar ave- tkir~irirs)sibi, and soinrtimes, iti deeper, cwolrr siib- iiiie of investigations. snrface layers, thr temperate-water albacore, 7huw M~~aitwhile,the government of Peru, off the shores )ius gcrmo. Sonw prelintinary field observations on the tropical of w11ic.h occurs a very rich fish fauna due to the bio- tiinas were made by scientists of the 17.S. Fish and logical effects of the Peru Current, and whose fishcries Wildlife Servicr in waters off Central America, and in have been growing very rapidly, also established in the Ccxntral l’acific south of Hawaii and westward 1957 a research organization, the Consejo de Investi- through the Trust Territories, iii 1947 and 1918, gaciones Ilidrobiolhgicw, first under the direction of aboard fishing \wsels of the Pacific Exploration Com- Dr. Warren Wooster, and now under Dr. Zacarias pany IIoa.ever, teniatic., largmeale research was l’opovi(i, to study the fisheries oceanography of that not initiated until 1!)49 when the Parific Oceanic region. Fishwy Investigations (1’OE’I) were established by From thv rrsults of thtw various invrstigatioiis we the I-. S. Fish and \TTildlife Srrvic*eat IIonoluln. This are beginning to understand some of the relationships laboratory has, since then, (Zondiict(x1extensive studies of the tropiral tunas to their environment, but there (luring s(mwi1 years of the tropical tunas ili the are yet many intc~r~stingi~nsolved probleiiis. One in- eqiiatorial rrgioir to the south of Hawaii, and detailed portant e~ivironnirntalfactor appears to be food, since stiidi(’s iii the vicinity of the Hawaiian Islands, as there has been showii, both in the equatorial Central \vel1 as research in the regioii about the Marqiiesas and l’wific aid in th(>tropicd Eastc.rii Pacific, a good (‘or- Society Islands, south of the equator. respoiidvnce b~tweenthcl abundancc~of the tiinas and In 1!151, the I titer-Anicricaii Tropical Tuna Com- the productioii of organisin\ lower in the food chain, inissioii ~~oniin~~~~cediiir~dgations of the tuna re- related to the etirichni~ntof the euphotic zone by sonrces of thr tropid Eastern Pacific in the region of nutrient-rivh water from deeper layers. At the ex- the large. coiiinierckl fishery off the wcst coast of the tremes of their ranges, in the Eastrrii l’acific at least, Americas. These investigations are direcdterl toward the thcre also appcw-s to be a direcat effcct of tcmpc’rature classic “ eoiirervation ” problcin, the elucidation of the on the tropicd tunas. Other variations in the owur- effwts of the fishery and of natural, fishery-independ- renw of thc tropicd tunas se(~nito be related to owan- ent fwtors 011 the tuna populations, as a basis for ographic pheiiornrna of other kinds, but no good nianagenicnt of the fishery. It was realized from the understailding of these is yet available. 42 CALIFORSIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS TUNA ABUNDANCE IN RELATION TO island. Data of this sort, however, have not yet been FOOD SUPPLY published. In the Eastern Pacific, near the eastern boundary, The tropical ocean, in general, tends to be a two- it has been found that the yellowfin and skipjack layer system of persistent stability, with the upper tuna are encountered in greatest abundance in regions mixed layer separated by a strong pycnocline from the where there are high standing crops of zooplankton, lower layer, and lacking the seasonal renewal of the and Dr. Blackburn’s group has also recently found upper layer which occurs in high latitudes due to the that where there are high zooplankton volumes there annual thermal cycle. In these circumstances, much of are high volumes of tuna-forage organisms. It has, the upper layer in the tropical seas becomes depleted furthermore, been shown that the zooplankton crop of nutrients and is, therefore, a biological desert where is quite well correlated with phytoplankton (meas- a few poor organisms barely survive. In some places, ured by chlorophyll) and basic productivity (meas- however, the winds and currents cause upwelling or ured by C14 uptake). The regions of high basic pro- upward mixing of deeper water into the upper, sun-lit duction, and consequent increased crops of organisms zone, and here there is high biological production, higher in the food chain, culminating in the tunas, leading to large crops of phytoplankton, zooplankton, are places where the euphotic zone is enriched by the forage orgaiiisnis, and finally the large carnivores such upward admixture of nutrient-rich deeper water. The as the tunas. physical mechanisms of this enrichment are several, One such situation has been beautifully documented however. Along the coast of Baja California and by the Pacific Oceanic Fishery Investigations in the along the coast of Peru the mechanism appears to be equatorial Central Pacific. The stress of the southeast coastal upwelling induced by winds blowing more or trade winds, plus the Coriolis force north and south less parallel to the coast toward the equator. Another of the equator, causes a divergence of the South zone of high production is off the Gulf of Guaya- Equatorial Current along the equator, bringing to quil, which is at the boundary between the water of the surface nutrient-rich water which moves west- the Peru Current and the warmer, less saline water ward and away from the equator. North of the equa- to the north. Whether the enrichment is due to mix- tor there appears to be a zone of convergence near ing along this boundary, to nutrients brought down the southern boundary of the Equatorial Counter the Guayas River, or by other means not now known, Current. Due to unequal wind stress at different is not understood. longitudes along the equator, the upwelling is most Coastal upwelling may also occur in the Gulf of intense near the 120th meridian. The standing crops Tehuantepec during the winter months when strong of zooplankton have been found to be maximal some winds from the north blow there. Recent studies, how- distance downstream from this meridian, their abun- ever, indicate that this may not be simple upwelling, dance being greatest at about 150” to 160”W longi- but vertical mixing associated with the development tude (nearly due south from Hawaii), and displaced of a cyclonic circulation and a thermal dome. to the north of the equator. The yellowfin tuna, as The high biological production off the coast of shown by longline catches, are most abundant near Central America is apparently due to upward mixing the area of zooplankton maximum, but perhaps dis- processes (not well understood) associated with the placed somewhat further to the north. These observa- thermal dome located there. This dome, over which tions are consistent with the hypothesis that the tuna the mixed layer is characteristically less than 10 are most abundant where their forage is best, and that meters thick, is present at all times of year, though this occurs as a consequence of the equatorial up- it may shift somewhat in position and extent, and at welling, but “ downstream’’ thereof, due to the water’s times extend right to the surface. Other areas with drift during the time between fertilization by up- persistent or regular seasonal dewlopment of thermal welling and the produetion of tuna forage. domes occur off the coast of Colombia and off Cape Corrientes in Mexico. The hydro-dynamics of these It further appears that this system is not entirely features are not well understood, but their effects on steady, but that the upwelling varies in extent both biological productivity and on the abundance of tropi- within years and between years. Fishing trials and cal tunas in their vicinities are quite evident. temperature observations near Christmas Island Smaller scale oceanographic features also appear (which is located a bit north of the equator in the to be of importance in causing local tuna aggrega- rich fishing zone) over several years indicated that tions. Both fishermen and scientists have oPten ob- the abundance of yellowfin tuna at this locality was served tunas, and other fishes as well, associated with directly correlated with surface water temperature. surface temperature discontinuities, ‘‘tide rips”, and This was interpreted to mean that when the water of was “older”, that is a longer time had elapsed since slicks. Investigations these features by scientists both at POFI and here at SI0 have shown that these enrichment by upwelling, allowing more time for the development of tuna forage, it proyided a better feed- are surface indications of “fronts”, marking the boundary between water masses of different tempera- ing area for the tuna. If this interpretation is correct, ture characteristics, and that along such fronts there one would expect associated changes in biological is convergence, which results in accumulation of factors, such as zooplankton volumes, and also one zooplankton organisms. It is believed that these, in would expect increased tuna catches north and west turn, attract the predatory fishes. It is likely that cer- of Christmas Island during cold periods near the tain oceanic areas, such as the region of the boundary REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 43 between the Peru Current and the more tropical fore, it appears that an active migration of the tunas waters to the north,-or the boundaries of the Equa- is involved. torial Counter Current, are especially likely places The relationship of tuna and temperature is even for the formation of fronts, and that such areas may, more striking off Peru and northern Chile, although therefore, be favorable for the surface-feeding tunas. the hydrography has been less well investigated. Along The mechanisms of fronts, their causes, and their geo- the Peruvian and Chilean coast occurs cold, upwelled graphic distribution deserve much more study. water associated with the Peru Current, extending to It has been demonstrated that the tunas are much the vicinity of Cape Blanco, where the Peru Current more abundant in the vicinity of islands and sea- turns west to become the South Equatorial Current. mounts than elsewhere in the open sea. It has been North of this is warmer, less saline water. Seasonally, hypothesized that these geological features modify the each year in the early months, the warmer water local circulation so as to result in higher biological moves further south, the extent and persistence of its production, but this hypothesis cannot be regarded southern movement being variable from year to year. as yet confirmed. Studies which we have made near It also appears that this warm water south of Cape Clarion Island and Shimada Bank do show some mod- Blaiico is characteristically a fairly thin layer over- ification of the distribution of physical and chemical lying colder water of the Peru Current. There is, properties, and some increase in productivity and furthermore, offshore from the Peru Current at more chlorophyll very near the island, but zooplankton southerly latitudes again warm water, but of high volumes are not much, if any, higher than in the salinity, in contrast to the low-salinity water to the offshore waters near Clarion Island, although some north of about Cape Blanco. For reasons not yet increase was found over Shimada Bank. There is a understood, there frequently occur at certain fairly possibility that benthic forms feeding directly on well defined locations, such as off Chimbote, Peru and detritus or benthic plants may constitute an import- off Iquique, Chile, tongues of this warm, high salinity ant part of the tunas’ diet in such places and thus water, extending in toward the coast, which at their provide more forage with little or no increase in basic inshore ends, at least, occur as thin layers over the production (by utilizing a shorter food chain), but cold Peru Current water. Tropical tunas are found in this has not been conclusively demonstrated. There is abundance only in the warm water; that from the evidence of increased standing crops of zooplankton north contains both yellowfin and skipjack, while the near Cocos and Clipperton Islands, but these locations “tongues” off Chimbote and Iquique contain almost have not yet been studied in detail. exclusively skipjack. Some years the warm water from the north, and POFI researchers have demonstrated the existence perhaps the warm water from offshore, are especially of a complex system of eddies on the downstream sick widespread and persistent in inshore areas. In such of the Hawaiian Island chain, and that concentrations years, the so-called El Niiio years, the tropical tunas of skipjack seem to be associated therewith. Whether are apparently much more widely scattered than in the association is because of effects on the tunas’ food non-El Nifio years, and also occur in abundance fur- is, however, not clear. ther to the south than normal. The physical and biological effects of islands and Although it seems clear that the tunas in this region seamounts is a very fruitful subject for further study. are directly influenced by temperature, and that the variability of the oceanographic circulation is a caus- TUNA AND TEMPERATURE ative factor of the variations in the distribution of the tunas, we have no clear understanding of how and At the northern and southern extremes of their why the oceanographic changes occur. It is hoped ranges, at least, the tropical tunas almost certainly that the continuing hydrographic studies of the Con- respond directly to temperature changes. Off Baja Cali- sejo de Investigaciones Hidrobiol6gicas, and other fornia and California there are large crops of food agencies, will elucidate them. organisms at all times of the year, yet the yellowfin tuna are found in commercial quantities only in OTHER PHENOMENA waters of about 19°C and warmer (skipjack occur in somewhat cooler water, down to about 16°C). The sea- The tropical tuna research programs have also sonal appearance and disappearance of these tunas on brought to light other variations in the occurence of the local “banks” off Baja California follows the the tunas which are evidently related to variations in march of the isotherms. In years, such 1957 and 1958, the oceanic circulation, but the way in which they when the warm water extends further up the coast, operate is not yet understood. the tropical tunas are likewise taken further north, In the near vicinity of the Hawaiian Islands the and the persistence of warm water on the banks off skipjack tuna appear high in abundance seasonally, Baja California beyond the normal season corre- from about May to October, and are in low abundance sponds to a similar persistence in the occurrence of during the rest of the year, although there is little or tuna catches. Analyses by G. Roden of SI0 have indi- no seasonal variation in food supply and the temper- cated that the seasonal advance and retreat of these ature is at all times of the year above the minimum isotherms is due almost entirely to the balance of for commercial abundance of this species in other incoming and outgoing heat from the sea surface, ad- regions. Furthermore, the abundance during the ‘‘sea- vection of warm water being small or absent. There- son.” varies markedly from year to year. 44 CALIFORXIA CQOPER.ITIVE OCBASIC FISHERIES INVESTIGATIOSS POFI investigators first found a rather good inverse chemical, and biological properties and of the tuna correlation between the percelltage of northeast trade distributions are fairly adequate. A start has been winds during February to April and the skipjack made on more detailed studies of some smaller-scale catches of the following sumnier “season”, over the features which appear to be of importance to tuiia years 1951 to 1956. Subsequently, it has been shown ecology, but very much more needs to be done. that there is a close relatioilship between skipjack Likewise, teiiiporal variations on both the large scale catch in the IIamaiian fishery (on a weekly basis) and aiid small scale have only begun to be studied. salinity, the skipjack occurring most abulldantly when Perhaps the greatest need for making progress both the water of lower salinity is present in the region. It on the study of particular features, such as the hydro- has also been shown that the time of occurrence of the dynamics of thermal donies or island effects, aiid for seasonal sharp temperature increase (which occurs in studies of temporal variations, is for better means of February and March each gear) is well correlated data collecting. Collecting observations from ships is with the tuna catch of the following sunimer “sea- expensive and not very satisfactory, because the obser- son” ; the earlier the warming occurs the better the vations lack both syiiopticity and continuity. In those catch, over the years 1951 through 1958. cases where semicontinuous observatioiis have been From these observations, POFI scieiitists have de- possible, such as the IIawaiian temperature and salin- veloped the following hypothesis : Duriiig the winter ity series, or sea-surface temperature series which it months the Hawaiian Islands are bathed by the waters has been possible to piece together froni merchant ves- of an exteiision of the Kuroshio Current, which are of sel and fishing ressel observations, a great deal has relatively low temperature and high salinity. In the been leariietl that could not have been obtained with spring, there is a northerly movemeilt of these maters any reasonably small number of research ships. Unfor- and they are replaced by water, of lower salinity and tunately, we now are able to get time series of quasi- higher temperature, of the California Currellt exten- synoptic data for only a few parameters at tlie sea sion. Coiiicident with the movemellt of the boundary surface from merchant ressel observations. Likewise, bet\\Teen these waters through the Islands, the skip- data froni shore and island stations are of limited jack appear in abundance. 111 years when the bound- utility, because they are seldom located where we ary does not move through the area, but merely ap- waut them, and, in any case, need to be supplemented proaches it, there is a poor sumnier fishery. by data froni further offshore. The crying need is for Why the skipjack are associated with the water of observing stations which can be operated, where we the California Current extension is not at all under- want them, at modest expense, for both short and stood, but the observed relationship does seem to have long time periods, and which can malie continuous fairly good predictive value. It is also possible that observations both at the surface and to depths of at this might point the way to an offshore fishery to tlie least a fern score meters. southeast of the islancls during the winter months. Fortunately, the same need has become evident to This stndy is, of course, being further pursued. people in other branches of oceanography. Conse- POFI researchers have also conducted, cluring the quently, both the tuna programs (particularly the past three years, studies of the tunas (primarily Tuna Oceanography Program at SIO) as me11 as skipjack) of the Xaryuesas Islands area and simul- other programs related to military oceanography, ma- taneous oceanographic investigations. These investi- rine meteorology, and basic research in physical gatioiis have shown a niarlred seasonal cycle in skip- oceanography, are supportiiig effort to develop un- jack abunclance, which is highest in the southern iiiaiiiied stations which can be put where required suniiner (January-February) , but which bears no and operated for both short aiid long time periods. currently uiidrrstaiidable relationship to the hyclro- The successful development of such unnianiied data- graph.. In a recent progress report, it has been collecting devices should make possible a great ad- written “ Although the data do reveal a pronomiced vance in the study of the sort of problems which seasonal variation in the apparent abundance of skip- are of special pertinence to tmia ecology. jack in the AIarquesan waters, we have been unable as yet to pin down the reasons for these variations. CONCLUSIONS In geiieral, indices of productivity are higher during The researches accomplished by tuna oceanography the months when fish are least abundant. There are no progranis in the Eastern and Central Tropical Pa- evident seasonal variations in type or abunciance of cific during the past decade have made great progress forage. As yet, the oceanographic data have not re- toward elucidating the effects of the ocean circulation vealed any significmit seasoiia,l variations in circnla- 011 the geographical and teniporal variations in the tioii features, such as those described for Hawaiian abundance of the tunas, and have been indispensable waters ”. to the understanding of the ways of life of these coin- pletely pelagic, high-seas fishes. Of special importance, FUTURE DEVELOPMENTS to my mind, is the demonstration of how much can I wonlcl slim up the present status of our tuna be accomplished by physical, chemical and biological oceanography programs about as follows : Broad sur- oceanographers working in close cooperation toward veys of the circulation and distribution of physical, joint objectives. OCEANOGRAPHY AND NORTH PACIFIC ALBACORE

., JAMES W. McGARY' Oceonographer and JOSEPH J. GRAHAM and TAMlO OTSU Fishery Research Biologists, Honolulu Biological Laboratory U.S. Bureau of Commercial Fisheries, Honolulu, Hawaii

The purpose ot' this preseiitatiou is to summarize to the west coast of Xorth America during each sea- briefly the Hoiiol~l~iBiological Laboratory 's (IIBL) son of the year. oceanographic studies in the North Pacific aiid the Both the oceanographic features and the summer oceanographic features with which are associated alba- distribution of albacore indicate that the area can core distribution and abnnclance. These studies were be divided at about 150°-135"TV. longitude, and we begun in 1954 following two rather significant de- nil1 follow this division in our discussion. velopments. On the negative side, the Washington Figure 1 shows the extent of our oceanographic and Oregon landing had fallen froni a high of 34 coverage to the west of 140"W. All the cruises were million pounds in 1944 to less than a inillion pounds made in summer or winter with the exception of one in 19.53. 011 the positive side, at least wieiitifically, made tliwing the 1954 September-November period. recoveries of albacore tagged with the "spaghetti" tag (Wilson 19<5R) had clearly demonstrated that the albacore could and did migrate froni the Anierican to the Japanese fi4ier.v (Gaiissle and Cleniens 1953), thus giving eyiclence that the oceanwide stocks are iaterrelated. The IIBL oceanographic and biological surveys north of Hawaii were planned to determine (1) the geographical limits and relative abundance of albacore and (2) the relationship, if any, of these to the en- vironmmit, as described by oceanographic data. These studies were coordiiiated with those of the other research agencies through an informal Albacore Xteer- iiig Committee composed of representatives of the Rurean of Commercial Fisheries aiid the States of Washington, Oregon, and California. Three years were alloted for exploratory cruises to determine the distribution of albacore in thr North FIGURE 1. Tracks of four HBL oceanographic cruises to the central North Pacific. Central Pacific, and three €or the study of their abundance, migrations, and seasonal fluctuations in Figure 1 is froin an earlier paper (McGary et al., abundance. The latter effort was based on the preiiiire that coiicentrationr of commercial poteiitial mould be 1958) in which the oceanographic features were reviewed in a discussion of the possible eiirichment pat- found. The rewlts of a conimercial charter in 1938 term of the area. that report the term "Transi- were disappointing, primarily because the locations of In commercial coriceiltratiolls of albacore are subject to tion Zone" was used to refer to the area in which the water characteristics changed from those typical large annual variations. of the two central North Pacific Water masses to The major portion of both the fishing and ocean- those of the Subarctic Water niass. It was used be- ographic efforts have been to the west of 140" W. cause the authors preferred to avoid the use of the longitude. Cruises have. however, been made to the term "Polar Front ", which implies a single, abrupt coastal areas in cooperation with other agencies and change. Oceanographic data showed, instead, a series as part of our studies of the geographical and sea- of small but frequently abrupt changes. Further- sonal limits of albacore and their migration routes. more, the large seasonal migration attributed to the There has been at lcast one romplete systematic fish- Polar Front was only apparent in the surface teni- ing survey, with limited oceanographic observations, perattire field. The zone of most abrupt change ob- of the'area north of the IIawaiiaii Islands from 180" - served in winter at about 35ON. is not comparable in 1 Present address : Office of Saval Research, Washington 25, D.C. structure or origin to the similar zone located farther 9 Formerly known as the Pacific Oceanic Fishery Investigations (POFI). north during summer at about 43"-46"N.

( 43 ) 46 CALIFORNIA COOPERATIVE OCEANIC FISHERIES ISVESTIGATIOSS SALINITY IN PARTS PER MILLE 33.0 34.0 35.0 33.0 34.0 35.0 30 IIIIJIIII~III

a ...... WESTERN NORTH PACIFIC CENTRAL WATER

FIGURE 2. LEFT PANEL-superimposed temperature-solinity curves for the meridional station series along 160'W. longitude Hugh M. Smith cruise 25, January 1954. RIGHT PANEL-same along 157"W. longitude Hugh M. Smith cruise 30, August 1955 (McGory et 01. 1958).

The temperature-salinity curves (fig. 2) froiii 111:T~ shift iii the siirftive portioiis of the c*iir~w.'l'li(~ iiiost winter and summer cruises, in the vicinity of' 160q\V., ahriipt shift oc~ii~~cvlkwt\vwtI :V'.X'S. aiicl i34' 28's. show the features used in a preliminary tldiiiitioii I+rtti(~r.to the iiortli aiiotticbr siic.h shift ovc,ii1~c~lI)+ of boundaries, as well as the effect of SeaSOtiiil ten- t\vrt.ti 41 "56's.~ii(1 4:3' 2:l'S. Sort11 of ttiiit Iiititiiclci. perature changes iii the surface portio11 of thc. riirws. tht. shifts were Iargv bilt \vvr(~iiiiifortii latit~~~liiiiill~-. Those from Hugh $1. Sniith cruise 25, January 1954, At 43',23'S. aiitl tiorth\v;irtl, il siiliiiity ~iiiiiiiiiiiiii \viis (McGary and Stroup 1956), show ail abrupt shift in prcwiit at thc. siirfave, wtiic~h is itr kwpiiig with tliv the T-S characteristics of the surface waters between tlefiiiitioii of Silbitr(.tic. \\'iit

J" lllllIIIII11I1IlII 24O 26O 28" 30° 32- 34O 36" 38O 40' NORTH LATITUDE 16OoW

FIGURE 40. Temperature section from BT casts along 16OoW., Hugh M. Smifh 25, January 1954. Brokedine in upper panel shows surface air temperature.

shows the eflect of seasonal heating and cwoliiig. Diir- ,om 27 24- XD 28. 3JD W W Ha 40. 420 44- 46- 4e Y NORTH LATITUDE ing both vruises the subsurface isotherms row 157O3O'W abruptly, e.g. 60"F., in the virinity of the southern limit of the Trailsition Zoite. I II winter. the iiorthrrii FIGURE 3. Salinity cross section 157°30'W., Hugh M. Smith cruise 30, zont' August 1955. limit of this ascent marked thc b(yinning of the with little or no distinct surfare layer ; in summer. it marked the beginiiing of a very shallow siirfare layer sufficiently with more saline water to tlie south to with ~IIextremely sharp thernioc+liiie. At the northern ('aiise it to sink even diiring winter. The limits of the limit of the Transition Zone, tlie .50" P. isotherin Transition Zone shown in figure 1 werf' drawn on marked the boiiiidary of a distinct difYerc.irc.e in the this basis. The differences ill the sections for the vari- subsurface structure ; to the north it niarketl the> bc- oils seasotis showed, as expected, that the southern ginning of aii area with frequent small inversions bo- limit was 1"-2" of latitnde farther south it1 winter lo~the thermocline. The siirfaw temperatures of both than in siiiiinier. Oiily slimmer data were available seasons showed a striking coincidence in the trnipera- to tlie north bnt no doubt the iiortherii limit shifts to ture range at which the greatest horizoiital gradient the north iii winter,. since cooling TTOIII~break np the owurred : on hoth sections inspecdtioii showed it was shallow sonthrrii limit of the lens. betneeii 56" and 68°F. with the inaxiiiinm iisnally ('oinparisoii of the loiigitudiiial sections from Hiiglr owurring betweeii 58" and 62°F. Inspwtioii of other -11. SwifA vriiise 25 (January-March 1951) showed transects shows that this is triie of all s(visoIis of thtl that boundaries and sharp chaiiges in the T-8 ciirves year. hoth in tht. wtitral atid eastern I'acdific (Shomiira became less distinct from west to east (Stronp and atid Otsn. 19,X ; (irahain, l!le5i),In spriiig. thc. zotte of Ilv(+ar>-.1956). surface teniperatiire tliscviitiiiiiity is assoc.iatcd with R(3ports by Japanese oceanographers suggest that the development of the shallow thermovlint3 aiid its the Transition Zone extends n.estwar,d to the area northward advancar as siinini('r approavhes. It also ininiediately off Japan and that it tiarrows to the cwiiic.icles with the southwly niov(wieiit of thc breakup west. l'tla (1!~3) described a double froiltal of the shallot? warm siirfaw 1;iyt.r i.ti fall and winter. in the Knroshio, and Illasuzawa (1957) disrnss&l the llitia (1957), it1 a stiidy of the tlistribution of chae- striwtnre in more detail. The latter reported the tognaths and ptclropods ill thr Sorth I'arifiv, foniicl width of the "Frontal Zone" betweeii the warm and that t hew owa11 og rap h i t 1i v i s i o 11 s (2 o i 11(2 idetl with (.old water to be 200-300 miles within about t500 miles faiuial divisions and, ill fa(+, the trrni "Trailsition of the cwast. 111 grneral. he stated that the "Polar Zonr " in tho SPIIS(' used ht1r.c. \vas introtlricwl iii his Froiital Zoiie " shows particularly predoniiiiaiit dis- report. ,Jones (JIv(hry et al., l!j-X?), in H stlidy of continuities along both the riortherii atid southern copepods, found that the foriiiatioll of th(. warm SIIY- edges. He also foiiiid that the northrrti edge was iiot faw layer appareiitly offered favorable cwtitlitioiis for as nt1ll defined as thc soiitherii edge. phytoplankton blooms. followctl hy ail i nvrc;is(J iii The siirfaw aiid vertical ttmpwature distributions zooplankton statidiiig c~op,Ilowc~wr, thc. Iattc>t*was typical of the Transition Zone are shown in figure 4. not c*haravterized by H Iiicritliotial advatt~~of tropical Tht. voiitrast betwerii the siininitlr atid winter profiles spe(*ics,biit rather by the tlevelo1)tiit~titof '' blooms '' of' 48 CALIFORNIA COOPERATIVE OCEANIC FISHERIES ISYESTIGATIONS eurythermal forms such as Calanws helgolandiczis and been included. The curve shows that the central Paci- C. tonsus. fic catches were made within a narrow temperature It cannot be ascertained from the data at hand band, with a dominant mode between 58" and 60'F. whether (1) the distribution of albacore is controlled The overall range may be narrowed to about 55°F- directly by surface temperature (presumably through 66"F., since almost all of the catches at temperatures physiological mechanisms) or whether (2) the distri- below 55°F. were made in the fall immediately after bution of albacore is related only indirectly to surface periods having winds of Forre 6 or greater arid it was temperature through the seasonal march of events evident that the area had experienced a recent sharp leading to a large standing crop of forage organisms. temperature decline becausp of wind-induced overturn In rebuttal to the hypothesis of direct relationship, (Shomura aiid Otsu 1956, Graham MS'). The Cob11 data from other areas do not ahrays show the same al- data are typical of the fishery which freyuently de- bacore-temperature relationship, and within the velops off the coast of British Columbia, Washington, survey area uniform distribution with temperature and Oregon during the late slimmer (Powell and Hil- did not occur from area to area or season to season. In debrand, 1950) and indicate that the albacore o(+ciir rebuttal to the forage hypothesis, the seasonal move- in the same narrow temperatiire range as in the cen- ment of the biological frontier coincided with the tral Pacific. The Jini Muru plot was vonstructed from movemeiit of the isotherms, but it did not indicate the data collected in the spring live-bait fishery off Japan presence of albacore in all seasons nor in all parts of during a cvoperative tagging program conducted ill the survey area. the spring of 1956 by the ,Japanese Fishery Agency In figure 5, the total catch of albacore is shown and IIEL (Van Campen and Murphy, 1957). These as the percentage of the catch made at different sur- data are representative of this fishery. Here, the tem- face temperatures. The total catch includes all of the peratixre range aiid mode are ereti tnore sharply de- albacore taken by HBL in the central Pacific by troll fined than in the other two plots. These roii(1itioiis and gill net during all seasons. For comparative pur- .I .I Gtdbnm, ' Jldcroecolog! 01 the dlt~dc(~iP,Thunnus germo (Ldcepedt.), In the poses, similar examples from the east and west have centrnl Noi th Pacific" MS

300 IIIIIIII I IIIIII(I lllllIllll.llll 200 25O 30" 350 40' 450 50° NORTH LATITUDE 157O30'W

FIGURE 4b. Same along 157'30W., Hugh M. Smith 30, August 1955 (McGary et 01. 1958). REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 49 perature as an indicator of the areas in which albacore might be expected to occur. Beginning with winter, a comparison of the catches o€ John R. Manning 60'O acruise 39 along 160'W. (fig. 6) during January 1954 with the temperature profile from Hugh M. Smith cruise 25 (see also fig. 4a) shows that the deep swimming albacore captured on longline gear occurred in an area having surface temperatures of 56'-66°F. and just south of the point where the largest horizontal temperature gradients occurred. Figure 7 shows the winter distribution of albacore in the eastern North Pacific as indicated by the surveys of 1955 and 1956 and the Japanese winter fish-

170. 160' 150 140' 130. 120- TEYPERATURE OF

FIGURE 5. Percentage of albacore catch versus temperature of catch. arise in part from the fact that the Jini Maru was en- 40' gaged in commercial fishing and an effort was made to stay with the maximum concentration of fish. How- rvrr, in spite of the fact that the general oceanographic features of the area indicate that the Jini Maru data were taken in the western extremity of the Transition Zone, the temperature range does not even overlap that of the central and eastern Pacific. A discussion of the seasonal distribution of albacore will serve to illustrate the use and limitation of tem- I I I I I I II 160' 170. 160. 150. 140. 130' 120- 80-1 5 80-1 30" 35" 40" I JANUARY - FEBRUARY RANGE OF ALBACORE I

FIGURE 7. Winter distribution of albacore in the eastern North Pa- a. 60- SURFACE_1 cific. Dashed lines indicate position of Subtropical Convergence from W Hugh M. Smith cruises 25 (January-March 1954) and 27 (January- 1111 February 1955). 0 '0 h cn CI ery. It shows quite clearly that the latitudinal limits 02 are almost coincident with the southern boundary of W 200 W the Transition Zone (fig. 1) and the temperature L 100 LL front (fig. 4 and 6) and are not associated with the H 400 subtropical convergence as originally hypothesized. v I I- The dynamic topographies from the 1954 and 1955 I winter cruises indicate that the subtropical converg- I- f~600 n 200 ence was in the vicinity of 24"-27"N. during both W n years (McGary and Stroup, 1956 and 1958). Suda n 800 (1958) also described the subtropical convergence as an area of minimal albacore catch iii the western cn Pacific. Y6 0 Our spring cruises (1955 and 1956) showed a coli- 0 spicuous band of phytoplankton and zooplankton I abundance in the 55'65°F. surface temperature 04 range between 180" and 140"W. There was an abund- 0- ance of forage organisms and a rich phytoplankton \ bloom was indicated by the Fore1 color of the water, I but few albacore were caught. Only one was taken o2 east of 17OoW., and only a few scattered fish to the G west of 170"W. on all three types of gear (troll, 0 longline, and gill net). ~llllllllllllllllllll During the fall of 1954, 1955, and 1956 albacore 25" 30" 3 5" 40" were taken in considerable quantities from offshore NORTH LATITUDE of San Francisco to 170"E. longitude, the western limit of the survey area. They were taken in waters FIGURE 6. Vertical temperature profile, Hugh M. Smith cruise 25, with surface temperatures between 55°F. and 65"F., Januory 1954 and longline catch/100 hooks, John R. Manning cruise 19, January 1954, along 16OoW. longitude. with the peak catches ocurring in the 58"-60°F. range. 50 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

FIGURE 8. location of albacore taken in the North Pacific by exploratory vessels, July-September 1955. 55.F 27 Again, this temperature appeared to coincide quite well with the biological frontier. During the summer of 1955, because of KORPAC _--- and the activity of the North Pacific Salmon Investi- ______------gations, there was excellent coverage between 175"E. and the west coast of North America. The results showed two interesting features. Firstly, the albacore GILLNET CATCH were divided into two groups, one in mid-ocean ex- \ tending as far east as 155"W. (fig. 8) and the other extending northwest from the California fishery. The existence of the area of albacore scarcity has been FIGURE 9. Summer 1956 albacore survey, John R. Manning cruise 32. at least partially verified in all subsequent summer surveys. Secondly, comparison of figure 4 and figure small amount of latitudinal change necessary to pro- 8 shows that the mid-ocean albacore were in a shallow duce the 10"-12"F. temperature anomalies. surface layer in Subarctic Water (figure 2) and at The existence of a "toiigue" of albacore extending about the same surface temperature range as during northwest from the California fishery in 1955 (fig. 8) the other seasons. The J. R. Manning returned to the and the work of Powell and Hildebraiid (1950), area in 1956 and found a similar situation (fig. 9) Powell et al. (1952), Shaefers (1951), and Partlo with an even more pronouiiced biological frontier (1950) suggest coiiditioiis under which the Pacific indicated by the plankton abundance and light pene- Northwest fishery develops. First, albacore may be tration readings. In 1958 the 1955 and 1956 summer present in the summer off the northwest coast during surveys were followed up by a commercially uiisuc- all years but move into and concentrate in the coastal cessful gill net effort from the chartered vessel Para- waters of British Columbia, Oregon, and Washington gon. Again the fish were found in the warm layer in only when oceanographic conditions (exact nature the southern limits of Subarctic Water, but the .cen- unspecified) are favorable. Second, the extent in time ter was 3"-4" farther south, and the staiiding crops and space of the coastal band of upswelling may also of plankton and forage were much less than observed influence the development of the fishery and limit its at correspoiidirig temperatures for the previous years. shoreward extent. The hydroptic charts for the mid-August period for During the last 10 days of July 1957, nine char- the past four years illustrate the amount the tempera- tered commercial trollers and two HBL vessels made ture' field can shift in mid-ocean aiid the radical a synoptic survey of the distribution and abundance changes in local temperature that result. The lower of the albacore in the coastal area between 35"N. atid panel, figure 10, shows the variation in location of 47"N. The results of the survey (fig. 11) showed that the 55" to 66°F. surface temperature band that ap- these fish were present in a wide band along the coast. proximates the limits of albacore distribution. As a In the northern part of the survey area, albacore were result of such variations, annual differences of up to most abundant offshore of the area of largest surface 10"-12"F. can occur at a given locality. This could temperature gradient which indicated the outer limit make considerable difference in the development of of the band of upwelling (fig. la). However, their the seasoiial plankton cycle in the area, particularly distribution was far from continuous and was iiot when one considers that the average seasonal range completely associated with any feature of the bound- in temperature is only 15°F. The upper panel shows ary, such as the tongues of relatively warm water the latitudinal changes in the temperature raiige that which penetrated shoreward. corresponds to the peak of the albacore abundance in The radical annual temperature differences shown the central and northeastern Pacific. It illustrates the in hydroptic charts of the coastal areas (fig. 10) might FIGURE 12. NEPAS surface temperatures and best albacore catch rotes.

I I I I \ I* IW IY 1p 1p I AUCUST 11-20 SURFACE ISOTHERMS

FIGURE 10. LOWER PANEL-mid-August meon and 1955 through 1958 positions of the 55’ and 66’F. isotherms; UPPER PANEL-mid-August mean and center of the 58’-6O0F. isotherms for 1955 through 1958.

FIGURE 11. NEPAS (Northeastern Pacific Albacore Survey) troll catches, July 1957. 52 CALIFORNIA COOPER.iTIVE OCE.\NIC FISHERIES INVESTIGA\TIONR

ILOCT 4, 1954-NOV 28, 1955-420 2-M 5.1954-dML 19.1956-411 3-M 9. 1955-dW 24,1956-259

5LJUL.31. 1956p-dUL. 23,1957-357 6LWC. 1. 1957-5EP. 11, 1951- 47 1LdUL. 22, 195l-.OCl. 1. 1957 77 SLOCT. 16. 19M-NOK 23. 1951-169 9LMV. 17, 1956-NOK 17. 1951-365 10LdUL. 23,1957--_.MU/ll: 261958-287 IILdUL. 22.1957 pJUN. IO, 19Y-323 RLdUL 16.1957- JUL. 11. 1958-360 13LdUL. 16. I957 -AK 22. 1958-401 14LMY. 14. 1956p-..A1K 23. 1958-.-647 r- 1LRV 21. 1956-JUL 21. 1958 _.607 I 16 NOV 16. I956 --Affi 12.1958- -634

*- @1 1 I I 3" 160" 150" 140" 130° 1200 I I I ALBACORE TAGGING

FIGURE 13. Chart of points of release and return of tagged albacore.

in thg North Equatorial Current and is based on stndies of albacore landed in the IIawaiiaii loiigline fishery as well as on HIRL's exploratory cruises and data from the Japanese loiigline fishery in the west.

REFERENCES (hissle, I)., nntl H. I<. Clemens, l!XX C;iliforni:i-txFge(l alh- core recovered off J~ipti.California Fish cind Vccnze 39 (4) : 44-13. Gr;ih:im, .J. .J., 10.57. Central Sorth I'ncific :illincore surveys, Ma>-to Sovemlier l!)Xi. V.S. Fish and Wildlife Serrice, Ape- cicil Scieiztific 12eport-Fisheries To. 21.2, 38 1). Hida, T. S., 19-57. The ch:ietoifnaths ant1 pteroI)otls 11s hiologi- ea1 intlittitors in the Sorth I'ncific. V.S. E'ish nnd Il-ldlife Service, Special *Scientific h'eport-E'isherias A-0. 215, 33 1). JIusiisuwa, J., 1957. A contribution to the knowledge of the Knrostiio east of .T:ip;in. The Oceanographic JlMngcrzine 9 (1) : 21-31. AIcGary, .J. TV., :ind E. 1). Stronp, 19X. Mitl-P:rc~ific ocvmiog- FIGURE 14. Albacore migration. raphy , 1':m t V I 11, middle latit utle w:iters, J a niiary -Ma rch 1!)34, U.S. E'ish and Wildlife Rerrice, Special Scientific Re- 195%. Study of age determination by hard parts of ~illiwore port-E'isheries 3-0.1S0, 17:3 p. from central North I'acific and I3aw:iii:in waters, r7.S. Fish 1958. 0ce:inoifraphic ohserwtions in the cer1tr;il R'orth I':icific, cind Wildlife Service, E'ishtry Riilletin 39 (150) : 353-363. Sei)tenil)rr l!%fi-.iiijiust 1!)55, T-.N. Fish and ICildlife Service, Partlo, J. M., 1S0. A report on the 1049 till):ic.ore fishery Special Scientific Report-Fisheries 3-0. A??, 230 1). (l'hunnus alnlunga). Fisheries research Board of Canada, Jld>:~ry,.J. IT., E. C. .Jones, ant1 J. J. Graham, lK8. $hirich- (,"ii-cidarSo. 20. nwnt in the Trmsition Zone between the Sul):irctic zinc1 ('e~i- Powell, I). E., :ind H. A. Hildebrxnd, 1950. Albacore tuna ex- tral Water masses of the central North l'acific. Proceedings ~)lor:iti(in in Ailtisktin and :idj:~centwaters, 1949. U.R. Fish of ths 9th Pocific Science Conyress, 1.927, 16: 82-8!).. e, Fishery Leaflet 576, May. Otsu, T.,and It. N. Uchidn, 1!)59:i. Sexnil1 maturity and spalvn- Powell, I). E., D. L. Alversen, and R. Livingstone, Jr., 1!1,32. inif of albacore in the Patitic Ocean, U.S. Fish alld ICildZife Sorth l'acific allincore tuna exploration, 1950, 0.S. Fish and Service, Fishery Hutletin 69 (148) : 287-305. Wildlife Service, Fishery Leaflet 402, April. REPORTS VOLUME 1‘111, 1 JULY 1959 TO 30 JUNE l0GO 53 Shaefers, E. A,, 19.73. North Pacific albacore tuna exploration, Srerdrup, H. W.,h1. W. Johnson, and R. H. Fleming, 1042. 1032, U.S. Fish and l17ildZife Service, Coni mercinl Fisheries “The Oceans, their Physics, Chemistry, and General Biol- Reviezu 15(9) : 1-6. ogy.” I’rentice Hall, Inc., 10%’.

ITda9 11‘*lg4” On the structure Of the bonndarJ- ‘‘’le Of ‘Iiffer- Shomura, R. s., all,jT. Otsi1, 19.76. Celitr;ll xorth PaLcific exit water masses. Journal of the Oceunogrophic Society of core survej s, January 19W-Fel)ruary 103.5, C.8. E’tsh and Jopnn TVildlife Service, Special Scaentific Report-Fisheries 3-0. 2(4). 173, SO. Van Campen, W. G., and G. I. Murphy, 3037. Tagging allm core on a Jai)anese bait boat. Pacific Fishwinan 55(R) : 3i. Suda, A, 1958. Itecrnitment and dispersion of the Sorth I’acific 39, 41-43. albacore. Repot t of Sankai Regionnl Fisheries Research Lab- Wilson, R. C., 1953. Tuna marking, a progress report. Cult- oratory, Xo. 9. (Translation hy G. T. Beard) fornia Fish and Game 37(4).

OCEANOGRAPHY AND VARIATIONS IN THE PACIFIC SARDINE POPULATION

GARTH I. MURPHY California Academy of Sciences, Scripps institution of Oceanography

The Pacific sardine fishery has produced catches defined as the absolute magnitude of the sardine popu- ranging from 789 thousand toils in the 1936-37 sea- lation or some segment of the population such as a son (Schaefer et al, 1951) to 10 thousand tons in the year-class. 1953-54 season (Felin et al, 1954). Fishing pressure By aiid large, availability is a reflection of the has remained sufficiently constant so that most of momentary response of the population to thc eiiviroir- this fluctuation can be safely interpreted as the result mmt. This response takes the form of variatioiis in of variations in abundance and availability of the distribution and behavior. Oceanographic studies rv- sardines themselves. lated to availability generally include estiiiiatiiig the With respect to abundance the basic variable is appropriate parameters of the cnviroiiinriit of thtr year class size, and the most readily available esti- fish population in question. As the stiiditls progress, mates of year class size are the virtual populations, attempts are made to predict whtAre and whcii (~iivit~)ii- that is the summation of those individuals actually mental patterns associated with high avaihibility will taken by the fishery. These have ranged from over occur. Examples of this typ of study are dtwribcd seven billion fish for the 1939 year class (Clark and by Sette (19.55) who disrnsscld several t~iivirotiiiic~iital Marr, 1955) to a low of .1 billion fish for the 1949 nievhaiiisnis oc~iirriiigili ini(l-l’a(~ifi(~Owan that arc year class (estimated from published aiid uiipub- associated with high c.oi~c~~titrittiotisof tiiiia, and Ciish- lished catch statistics). This is a range of 70 times iiig (1!)55), who was able to rrlatcx thv tlistribiitioii but the virtual population can be severely distorted of lirrriiip to vhangw in the food slipply. A11 siwh by variations in availability, particularly during the studies siiggest a niore or lrss iiistaiitau(wiis respotis(& second, third arid fourth years of life. Variations in of the fish populatioii to a paranic~trriti the ocean. availability (in the sense of Marr, 1951) are difficult In contrast to availability studics, iiivt~stigations to document precisely. The results of oiie study of abiindai~cc~Inlist cwiisidrr the integratctl rtqoiisc? (Widrig, 1954) suggest that availability might have of tlir. popiilatiott to thv (~ti~iroiini~~tit~ilpatterii owr had a range of five times during the period 1941-.X, N period of tiin~.Tlicx i~ilrercwtdifficw1tic.s of this ap- and the inclusion of mow recriit years would iindoubt- piwwli arc’ siigpc.stcd ill figure 1. Four curves are edlp increase this range. sliowii, wvh rep~.rsc~iitiiigthe rrlativr change in iiistaii- In terms of the fishery the, obrioiis iiwd, is to unrl(~r- taliroiis survival rate (Z) iiwdrd to rc4iic.c a parti(+ii- stand and predict thew flii(*tiiatioiisiii abiiiitlance atid lar initial survival rat(. (8) over a fixed period of availability. This may also b(>vim ctl as a problcin iii tinic.. Thr ahrissa is the ratio of thct initial siirvival ecology; that is, the r(+poiise of a popiilatioii to its rate (iiitlic-atcd oii the (~irv~)to any give11 1owc.r sur- changing environment. The C‘aliforiiia C’ooperativr vival rat(.. 011 thri ordiiiatr is a scale of the t*atio of Oceanic Fishery Investigations is inquiritig into tliescb fluctuations, and attempting to tlditic1 the rclative role of nature aiid mail ia iiifluencitlg the. size of the population. The objectives and general re\iilts of the prograuls have been documented in a series of six pro;nress reports (Marine Research Committee ; 1950, 1952, 1933, 1955, 1956, 195S), which include bibliographies of completed scientific papers so there is no iieed to report general results here. Rather, I wish to discuss certain aspects of the oceanographic material that seem to bear most directly 011 the problem of abundance.

AVAl LA BI LlTY AND ABU NDANCE Before examining some of the facts it is well to consider the theoretical basis underlyi1ig the application of oceanography to fisheries problems. The fishery problem may be coiivc~iiieiitlyseparated into avail- FIGURE 1. Diagramatic representation of the relation of changes in ability, defined as the acwssability of such sardines survivol rate (S) to changes in the instantaneous mortality rate (Z), as exist to the cornmc.rc.ia1 fishery, aid abundance, that is S./SI e- (’2 -‘I), the instaiitaneoiis mortality rate associated with the areas where spawning has been heavy. For instance at initial survival rate to the instantaneous rate associ- a moderately high density station (117.50, Cr. 5602, ated with the lower survival rate. For example, taking Ahlstrorn, 1958) there were about 800 eggs, with a the curve repr(2wiitiiig S = 0.8 at initial survival, a volume of about 2.5 nil. per 1000 1n3 of water. At this change to AS’ = 0.08, a reduction to g,,, would be associ- station the total catch of small plaiilrton in the tow ated with an 11.5 times increase iii the instantaneons net was about 100ml. and total plankton was 425 ml. mortality rate. (Thrailkill, 1957). Eggs were therefore about 1/40 of 1;nfortunately when \tidying the California bar- the biomass of small plankton. If the water added to dine we must deal with survival rates in the order of the perivitelline space after spawning is deducted ,001 during the first 45 days of life and of vourse from the egg mass the fraction is reduced to about much lower if a longer tinie-span is considered. At l/l20, and, of course, if all the plankton were caught this initial level of survival (.001) an increase in the and considered, the fraction would be much smaller. instantaneous rate of 1.6 times results in a hundred- Finally, random dispersal of the suspended eggs fold decrease in the number of survivors. slioulcl serve to rapidly reduce such high concen- In ternis of an investigation we can hope to nieas- trations. ure survival at some point (s) in the life of the sar- Food is frequently considered the key to larval dine (the result of integration) but we cannot survival but in the instance of the California sardine measure an integrated eiivironmeiital effect ; all that this does not seem likely. Arthur (1956) showed that it is possible to measlire is an instantaneous value of the factor (s) contributing to the mortality rate. In ABUNDANCE OF NAUPLII IN THE CALIFORNIA REGION prartical terms a large change in the integrated re- WATER AND THEIR AVERAGE DISTANCE APART sults, i.e., a spectacular change in survival, may be associated with very small changes in the instantaneous measure of the environment as suggested above. Thus the problem of understanding the relation be- iooro-o~~90 tween changes in the enviroiimeiit and changes in survival of the California sardine niay be much more difficult than an investigation of availability. It would appear that it is iiecessary to identify precisely the operating factors because if small changes I NT in the operator(s) result in large changes in survival, it will be nearly impossible to achieve success by meas- uring factors merely associated with the true operators ; we must know the precise maimer iii which the operators affect the population in order to permit some form of integration either formal or conceptual, aid there must be precise inventories of the fish populations at critical stages. The alternative is to content ourselves with a rather general approach to changes iii the ocean and changes in the populations without 20 serious consideration of cause-effect relations. This is an analogue of the water mass approach of explaining 10 the geographical distribntion of aiiimals with time 01 I I I I I I I substituted for space. For certain purposes this is not 1 3 10 32 100 316 1000 3162 10,000 without profit but it falls short of being fully satis- fying. C 0 N CEN TR AT IO N S IIistorically, many attempts to relate the environ- (NO./M~ ment to survival have centered around the larval period because the major reduction in the numbers of FIGURE 2. Cumulative summary of the density of the nauplii of micro- a given year class occurs then. Moreover, it is at this copepods in the California Current, and the computed mean distance between nauplii associated with each density assuming they are ran- stage of the life of the sardine, prior to its acquiring domly distributed in the water column, (after Arthur, 1956). the ability and motivation to aggregate, that the factors responsible for mortality are most likely to operate in a density-independent manner ; this should the standing crop of potential food organisms was simplify the analysis. 1,000 nr3or greater in 7070 of his stations and 3000 The fundamental support for the assuniption of m-3 or greater in 50% of his stations in the general density independeiice is that sardine eggs and/or area of sardine spawning (figure 2). Referring to the larvae are a very small fraction of the biomass. For high density station discussed above, the ratio of instance, sardine larvae comprind only four percent larvae (if all eggs survive to hatching and there is no of the total fish larvae during 1935 and 1956 (Ahl- dispersal, arid if all eggs were assumed to be in the top strom and Bramer, 1957, Ahlstrom, 1958) and, of 50 meters) to food organisms ~~~oulclhave been 1 to 500 course, a much smaller portion of the plankton in or 1 to 1,500 or greater in 70 percent or 50 percent of general. This statement seems to hold true even in the station respectively. REPORTS VOLr31E VIIL 1 dr’LY 19~79TO 30 JUNE 1960 57

Arthur also shows that the mean distance between rate of predation will be considered in the next sec- food organisms was 10 ceiitinieters or smaller at 70% tion. of his stations and 7 centimeters or smaller at 50% of his stations. The ineaii distaiwe between a larva and a OCEANOGRAPHY AND LARVAL SURVIVAL food organism would be one half this. Because of mo- tion of food and larvae, oiily a very short time need The most conspicuous variable likely to affect animals directly or indirectly in the California Current pass before a larva has a food orgaiiisni within easy Systeni has been temperature. Reid, Roden, and reach. This informatioil siiggests tlliit there is little or \T‘>-llie haw sunirnarized certain aspects of no competition among sardine larvatc for food, aiicl (1958) these vhangrs. The years prior to 1944 were charac- that there is always or nearly always sufficient food trrized by more or less alteriiatiiig positive and nega- within easy reach of the larvae. The qualifications iii tive aiionialirs froin the mean. From 1944-1956 the this statement stein from the fact that occasional tows aiionialies were predominantly negative, and from reveal concentrations of eggs about 10 times as great 195i to the present (1960), they were mainly positive. as the high density station under consideratioil. Maxi- The variations in teiiiperaturc ar~essentially the re- niiim observcd concentrations of larvae in a given sult of the iiiteractioii of the ytreiigth of tlip California year, however, are about the same as the above de- Current, upwelling along the roast, and thc strength Variations in food sup- scribed concentration of eg and character of the countercurrent. 13ec~aiiwthe his- ply may not then be a sign aiit factor in larval sur- torical record suggests that \varni years tend to be vival, except in rare instances. a5sociated vith good year-classes, and cold years n ith Tlie arguinents above are based 011 what amounts to poor oiies (see Clark and Maw, 1933, K(.icl et al. 1958, an instantaneous view of the numerical and spatial and Marr, in press), I wish to exaniiiie the possibility relations of the larvae and their food. Their validity tliat cold temperatures caii act to increase larval mor- turns on whether or not it is possible to regard this tality through predation and/or that cold tempera- iiistaiitancous picture as a sample from a system in a tures are associated with other phenomena that ad- rc.asoiiably stcady state. That is, the iiilplicit assuinp- versely affect the sardine population. tioii is iiiad~that thc obsc.rved food supply at a given Ahlstroiii (1954) has presented evidence indicating illstillit is thr iiitegrated stead)- statc result of its that survival during the first 45 days of life is ap- r(~iiewa1rate, its “ dmth ” rate from predation, and its proximately .OO1 (figure 3). The mean instantaneous growth to a size too large to wrve as food for the niortality rate associated with this is 6.9. Graham Iai~av.Though it is difficult to prove the assimption (1936, p. 244) suggrstcd, but did not elaborate, that for~liall,~-,the fact that ai1 appawiitly favorable supply colder teniperatures might prolong the larval phase, of footl S(YWS geii(~a11yto prevail I~iitlsstrong intni- leiig theniiig this period of high mortality, and there- tiye support. fore decrease survival. Ricker and Foerster ( 1948) l’re(latioii rrinaiiis for c.oiisitlr~ratioii. It swiiis rea- eo11clnde that mortality of soclreye fry is a function soliabl(~tliat fish larvae will b(, siibjwtcd to roiiglil?, of growth rate. In their data, growth is accelerated tlic sanit’ rcitrs of ~>~~~li~ti~iiiis art’ otllclr O~~~LII~SII~S in or rctarcled as a function of population density, but tlic. 1)laiiktoli possc~s~irigtlic.ir ~illl~(’g(~li(~riil (liiii(An- the operator is predatioii, mortality being greater as sioiis aiid hrliaviorwl c~liiii.actc~ristic.s.I a111 IIII~IV~IYof thr length of time spent at small, wry vulnerablr prvcise (lata on this 1)oiiit but tlrc. ratw al)p+ir to l)(> size5 is prolonged. Certain aspects of this problem are (~strc~iii(~lyhigh. For iiistaiiw. (‘ushiiig (l!l.X) (‘on- alw treated in Reverton and Holt (1957, p. 53), arid ~111tlrtlthat ov(’r 90 p(~rwiitof tliv tliatoiiis prodiiwtl liickcr (1938, p. 263). Ahlstroni (1954) finds little tliIl.ing tlir periott ,\pril 10 to lliiy 16 iii his stutl?7 evidence of srrious mortality in the egg stage, which a rpi~ jvere coiisuiiit~d. I~(~iiii~(~thr larvar are t>.p- might well be associated with the extreme transpar- ic.;i11y a sinal1 f ractioii of’ tlicl biomass it sw1ns unlikely ency of the eggs, and I hare already noted that eggs tll;it v;iriation ill thcir iiiiirrhcrs rcsirlts in variations in are not aggrc.gatecl densely enough to attract filter tli(2 popiihitioii of predators. I’i~~sninahl~-the arrival feeders. Thus the mortality (99.9% ) under considera- tlll.o\lgll Ilat(*hi1iga~itl (l(~l>iil~il11.(~ (tlirollgli g1.o\\.tll) of tion occurs after hatching and before the stages at it rpliItiv(~l~-f(>n. sartliiica Iiii~vw(.cllivits littl(1 or 110 spc- which the larvae are no longer sampled iii the routine cia1 iy’slmi~se fronr tlit. pol,~~liitio~~of I)rdators, ~11g- planktoii hauls (a period of about 45 days). grstillg tlliit I,rctIatioii is tlmisity iiidclwndeilt. If these high mortalities after hatching are simply rrilearpnnlrilts i~l)ovc,iiiclicsatr that eiiviro~niir~~tal a reflection of the larva’s belonging temporarily to (lata s11ol1lcl br esainiil(vl iii tlic light of thrrr work- the relatirely helpless plankton community, mortality iilg :lssillllptions. (1) The sardilic rgg-larva is a ininor will be sharply reduced or enhanced if the period of (~IpIiIpiitiii tlw pvlagic faiuia ai~dthereforr thr factors this association is altered. Tlie only way a sardine aff(>Vtiiigthe. slirvival ratr operate in an esseiitially can escape this community is by growth and this is a tIc~isit~.-iiidcpc.lldrlltmaliner. (2) Vood supply is not fuiiction of food and temperature, and larval food a11 iiiiportaiit survival factor, btit ixther (3) the in- seeiiis to be adequate for survival though the supply tellse piyylatioii typical of the 1,laiikton conin~uiiity may not always be adequate for maximum yro~vtli. is r(,spo1isible for the rapid tlccliiie iii iiiimbers of Thus, if predation is held to be the maiii source of the. siirtlinc. laryae, aiitl variations in this rate of pre- larval mortality its effect can vary with the density (latioil arc1 primarily rcspoiisiblr for variatioiis iii the of predators, and the duration of the vulnerable rat(>of liii.\-al siirvival. 110~obsc,rvetl vhaliges in the stages. The latter can be a function of food, tempera- (‘iilifortiia Currc.lit System can operate to Yary the ture, or both. There is some indication (Keid et al. 58 CALIFORNIA COOPERATIVE OCEANIC FISHERIES ISVESTIGATIONS

LARVAL LENGTHS IN MILLIMETERS EGG 3.25 4.75 5.75 675 875 10.75 1275 14.15 l7.25 21.25 I I , I I I I 11111111 I I

1,000.000

ioaooo FIGURE 4. Hatching time of sardine eggs as a function of temper- 1 ature. The numbers along the curve represent the relative hatching time taking 15OC. as a base, (adapted from Ahlstrom, 1954).

we can estimate the relative larval survival from the equation : 8 = e-zt where S denotes survival, 2 is the mortality rate and t denotes time. Because the survival to 42 days or 24 mm (Ahlstrom, 1954) was based 011 years in which the average temperature was about 15"C, I use this t 1" temperature as the base setting 42 days as t = 1. Sirn- 01 1 I I I I I I 1 5 10 15 20 25 30 35 40 ple computation then gives the following relative sur- ESTIMATE0 AGE IN DAYS vivals to 24 mm. 'I'eirrpern t tire (C) Relative Survival FIGURE 3. Apparent mortality curve of sardine larvae, (after Ahl- 14.0 0.43 strom, 1954). 15.0 1 .00 16.0 2.12 1958) that food was more abundant during the cold 17.0 4.14 years associated with poor year classes so it seems Thus a three degree range of temperature, approxi- logical (though not necessarily correct) to discount mately the difference between warm and cold years food, and examine the associations among tempera- in the California Current diiriiig the spring months ture, larval growth, and the density of predators. (Reid and Roden and Wyllie, 1958), might result in Temperature, through its control of the metabolic a 10 fold variation in survival. The effect of growth processes, can alter the growth and development rate. on survival is clearly shown with respect to one pre- Data from Rhlstrom (1954) show that eggs hatch in dator (a plankton net) in figure 5. 54 hours at l'i"C, 60 hours at 16"C, 68 hours at Other effects of lower temperature on the survival 15°C and 77 hours at 14" C (figure 4). If these rates of the sardine larva can be postulated. For instance, are projected into larval developmeiit which, at least Brett, et a1 (1958) have shown that the swimming in part, is an extension of embryonic development, speed of small fish varies with temperature. Tempera- REPORTS VOLUME VIII, 1 JULY 1969 TO 30 JUNE 1960 59 A third effect associated with temperatures arises from the fact that in general lower temperatures are associated with a more vigorous California Current and more vigorous upwelling. If there is any loss of larvae due to the drift into unfavorable locations offshore and/or to the south, this loss would tend to be greater during years of cold water. At the present, however, it is impossible to state whether this is a major e 12.0- or minor source of mortality. The fact that the larva 9. develops the ability to elude the plankton net at an 2 10.0 - I early age argues against it. However, the cooler temperatures attendant on vigorous coastal upwelling 5 8.0 - would teiid to induce the sardine to spawn further e- offshore, seemingly, increasing the likelihood that ? 6.0- larvae will be swept to unfavorable areas. A full dis- %- cussion of this problem is included in Sette (In press). 2 4.0- a- Finally, of course, cooler temperatures would tend J - to favor northern fishes, some of which might be com- & 2.0 petitors or predators of the sardine. With respect to - the larval state during which sardine mortality is ~o~~-~'~'~'~'~'~'1 I I I 4.75 6.75 0.75 10.75 12.75 1475 1225 19.25 21.25 assumed to be density independent, the larvae of a 2 SIZE OF LARVAE more northern form such as the anchovy, which pre- IN MILLIMETERS sumably is better adapted to cool temperatures, might grow relatively faster, and as a byproduct might prey FIGURE 5. Relative esciency of the one meter net in capturing sardine larvae between day and night as a function of larval size, (adapted on the sardine larvae. At the moment this is con- from Ahlrtrom, 1954). jecture, and in any event it would simply be a special case of the general predation by the plankton com- tures below the optimum for sardines would place munity. A hypothesis involving the anchovy (some of them at a disadvantage, particularly in the face of cool these ideas together with documentation) has also been water predators. advanced by Marr (In press), but his working as- Possibly even more important, in the California sumption is that the anchovy larvae affect sardine Current region a decrease in temperature has been ac- survival by competing for food in a density dependent companied by an increase in the standing crop of model. zooplankton (Reid et al, 1958). For example, off southern California the temperature during February JUVENILES AND ADULTS to August raiiged from 14.3" to 15.4" Centigrade Almost nothing is known concerning the sardine among the years 1949 to 1956. The zooplankton vol- between age 45 days and age six months. During umes varied from about 50 to 800 ml. per thousand this period they develop the schooling habit, and ap- cubic meters, larger volumes being associated with parently move inshore (Phillips and Radovich, 1952). cooler temperatures. This must be accompanied by an Because of this they will then be a significant element increased rate of predation, depending, of course, on of the fauna at their points of aggregation, and the the qualitative composition of the net zooplankton. As- sizes of the past catches (nearly a million tons) sug- suming the type of organism to be constant, the instan- gest they are a significant element in the general taneous mortality rate from this source might vary as biomass of their habitat. Mortality must be now as- much as 20 times judging by the variations in plankton sumed to be at least partially density dependent. abundance. Thus, lower temperature, in addition to Direct competition may exist among young sardines prolonging the duration of the larval stages, and hence and between the young sardines and the young of the vulnerable period with respect to predation, seems other fishes. It also suggests that predation might op- to be accompanied by an increase in predators. erate in a density dependent manner. The only posi- Because of the relationships shown in figure 1 only a tive indication of a density dependent effect on record fraction of the combined effects could easily produce is a study by MacGregor (1959) which showed that more than the total observed range in sardine year- condition factor of the catches of adults varies in- class strength. The increased food that must aceom- versely with population size suggesting competition pany the lower temperatures and higher plankton for food. volumes might operate in the opposite direction, Because of the scarcity of data, reliance must be though probably not strongly enough to overcome the placed 011 the general principle that as the range of adverse effects. Definitive examination of the problem an animal is decreased for a protracted period during awaits, among other things, point-by-point examina- density dependent stages of its life, its numbers will tion of the in situ environment of the eggs and larvae. decrease. The sardine is an inshore, pelagic, south- 60 CA.LIFORi\’IA COOPERATIYE OCEANIC FISHER7”f TSVESTIGATIONS temperate form. The southernmost end of its range Here, dealL,, competition with other species is ab- abuts on tropical water and this boundary is relatively sent in the seiise of the fisheries problem. steady geographically because of the nature of the Taking each of the four points in turn, in the light eastern Pacific circulation. The northern end of its of the earlier discussion, it is clear that reducing the range is near Vancouver Island but this can vary iiunibers of eggs reduces the numbers of sardines pass- greatly. Between these limits flow is generally parallel ing through the deiisity iiidependerit mortality stages, to the coast and temperature isotherms and other during which mortality is postulated to be largely physical attributes teiid to parallel the coast. If we a function of predation. The problem then narrows assume that the population spreads over its available dowii to whether the combinatioii of egg number times range, with the limits set by its own inherent responses a density iiidependent survival rate provides enough to the physical enviroi~nieiit,and by competition, the individuals to occupy efficiently the available environ- stage is set for large fluctuations in the population. iiieiit during later stages when survival is expected For instance, a change in the temperature regime will to be density dependent. The term “efficieiitly” re- be transmitted dowii the coast, as far as the relatively quires definition. For this discussion it nieaiis to oc- coiistant southern limit of the California Curreiit, cupy the environment to the extent that the addition thus producing a large change in the habitat avail- of more individuals to a year-class at the beginning able to the sardine. of the density dependent stage will not significantly Kegulation of the population during a shift from increase the year-class when it enters the fishery. warm to cool coiiditions can then be postulated as fol- Reduction of the size of the population by the fish- lows: A regime that cools the waters “pushes” the ery also interacts with the environment, because the adults south. Their southern boundary remains rela- amount of eriviroiinzcnt available to an organism is a tively fixed thus temporarily increasing the density of functioii of the absolute amount of environment avail- large fish in the now restricted habitat, a phenomenon able and the numbers of individuals competing for that must result in increased mortality until the that environment. Thus a fishery should enhance the population adjusts to a iiew appropriate size. Follow- survival of those individuals remaining and those in- ing or paralleling this, recruitment is reduced by the dividuals in a pre-recruit stage. Or paraphased the several effects associated with lower temperatures on classical formula : a (aniiual mortality) = ni (fishing larval survival, because of the reduced extent of the mortality) + n (natural mortality) - nz n will not ap- inshore nursery area and because of competition from ply in a predictive seiise to the extent that a rediic- species better adapted to the altered environment. tion in population decreases n. It can be concluded that a fishery exer’ts a positive effect on year-class size by relieving density dependent environmental pres- EFFECT OF THE FISHERY’ sures. Possibly the size of the 1939 year-class (the Though it is not appropriate to consider the fishery largest on record) was in part a function of this ef- in detail, for the sake of continuity it is necessary to fect, as it was produced by a population that had review briefly how the fishery might affect the nat- been heavily fished for several years. ural ‘ ‘ enviroiimental ” regulation of the population Reduction iii the average age of the population as a size. The effects of a fishery can be expressed qualita- result of a fishery has the general effect of making the tively with reasonable exactness j the difficulties size of the fishable population less stable, simply be- emerge in attempting to quantify these effects. Helice, cause the size of the adult or fishable population terids this discussion will be in the main qualitative, and to become a function of the size of only the most re- for this reason will do little more than attempt to cently recruited age class, and thus the population identify the problems. becomes more sensitive to short term changes in the The fundamental effect of a fishery on adult sar- environment. So long as enviroiimeiitally induced dines is to increase the mortality rate of the fished fluctuations in year-class size coupled with the effect stocks. This results in four secondary effects : (1) the of the fishery do not reduce the population below numbers of eggs spawned each year are reduced; (2) the level at which it can provide sufficient recruits to the total biomass of the population is reduced; (3) the density dependent phases of sardine life, and so the average age of the population is reduced; and (4) long as the time scale of the environmental fluctua- the size of the population relative to the sizes of the tions is short, the effect of a decreased average age populations of competitors is reduced. will be to increase the amplitude of the fluctuations Effect number 4 is not necessarily characteristic of in population size, but it will not necessarily reduce all fisheries, and is perhaps one of the fuiidanieiital the average size of year-classes. They may, in fact, reasons why the changes in the sardine population be increased. If the time axis of the eiivironmeiital have been intractable to the usual fisheries theory. changes is long compared to the average age of the Conventional theory was largely established on stocks adult population, the relative amplitude of the fluc- of fish that seem to lack serious competitors such as tuatioiis will teiid to be the same in the presence or the Pacific halibut, or on relatively nonselective fish- absence of a fishery, the effect of the fishery induced eries such as North Sea trawling. An extreme, of niortality being coilfined to affecting the shape of the course, is a mono-specific fish populatioii in a pond. curves, particularly by steepening a decline. The fourth effect of a selective fishery such as that 1This and subsequent sections of this discussion neere not pre- on the sardine is to reduce the numbers of the popu- sented at the symposium because of time limitations. They have been freely revised since the symposium. lation iii relation to its competitors and predators. In REPORTS VOLUhlE 1’111, 1 JULY 1959 TO 30 JUNE 1060 61 the absence of competitors or significant predators It seems doubtful that this would ever occur in this effect will, of course, be zero. This does not seem nature. More likely is the possibility that the selec- to be true in the instance of the sardine which has tive pressure on the population by man might “sensi- predators and probable competitors. Projection of the tize” the Pacific sardine so that slight changes in effects is difficult. There is little insight to be gained the environment in unfavorable directions would tip by reference to agriculture or pond culture, or even the scale in favor of some other species or combina- forestry, for here man is able and does intervene to tion of species at the same trophic level, and thus counteract some of the adverse effects of the heavy precipitate a decline much greater than would be ex- mortality he induces. It would appear more appropri- pected from the environmental change alone. ate to refer to the principles of evolution and survival for guidance. APPLICATION TO DECLINE OF In nature, most of the attributes acquired by a THE PACIFIC SARDINE population are probably directed at one goal, sur- The first section of this discussion sought to show vival of the species at a maximum commensurate with specifically how recent changes in the California Cur- the environment. In nature there is no survival signif- rent System, particularly cooler temperatures and icance in maximizing annual production of protein associated changes, could adversely affect the Pacific at the species level, though there may be at the com- sardine. In the second section an attempt was made munity level. What we interpret as potential produc- to determine qualitatively how a fishery might mod- tion in a virgin stock may be the energy utilized by ify the effect of the environment on the population, the species to maintain itself in the ecosystem. If the and it was shown that the minimum effect would be particular population in question either by accident to nialre the population more responsive to en1won- ’ or design has no significant predators or Competitors ment changes, particularly because the sardine fishery man may then utilize this potentially available sur- acted selectively within the trophic level. By way of plus production, e.g., a fish farm or a wheat field. If, a summary, this section will review the history of again by accident or design, the population has pred- the sardine population and fishery to see if the ap- ators, but no competitors, i.e., is the sole occupant of proach fits the facts. Temperature will be used in the its trophic level, man may still safely utilize the PO- discussion, not in the sense that it is the sole opera- tentially surplus produetion though to a lesser extent tor, but rather as a convenient index of the oceano- for he is simply competing with another predator, graphic information. and the population of that predator will probably shrink to a size commensurate with the now smaller Very briefly, cooler temperatures along the Califor- prey population. nia Current are associated with accelerated southward transport, and more vigorous upwelling with its In the instance of the Pacific sardine it appears attendant offshore movement of water. Plankton that there need not be great concern over species that densities are increased. Warmer temperatures are as- simply prey on the adults. If man reduces the popu- sociated with the converse of the above, as well as lation of sardines these predators will eventually re- greater influence of southern water and/or the duce their predation, either by becoming less numer- warmer water to the west. ous, or by utilizing alternate foods. Competitors are another matter, particularly in the instance of a se- Schaefer, Sette, and Marr (1951) described the lective fishery. The sardine’s range greatly overlaps growth of the fishery from 1916 to 1942. They found that of species that share its diet in large part, share that the growth pattern could be fitted to a logistic its living space, and prep on the sardine at some curve with an upper limit of about 600,000 pounds, stages in its life, e.g., jack mackerel, and anchovies. Though they attributed this growth primarily to eco- Such species as these, though they obviously do not nomic factors, they also concluded, from biological occupy precisely the same niche as the sardine, are information available, that this limit was close to the potentially its serious competitors. To the extent that productive potential of the then existing population. a proportion of the productive potential of the sar- In addition they suggest that, “Indeed, the limit at- dine which was “wasted” was in fact necessary for tained, about 600,000 tons, may have been the result its survival in the presence of its competitors, re- of a series of years during the late 1930’s, which were moval or reduction of this “buffer” by a selective exceptionally favorable for reproduction and survival fishery will be detrimental to the species. of the pilchard, in which event the average maximum stabilized yield may be expected to be lower To some extent, then, the increased opportunity than this value.” In point of fact, the largest year for surviral of younger stages that man provides by class by a factor close to 2 resulted from the 1939 thinning down the adult sardine population may be spawning, during the period of heaviest exploitation. more than countered by increases of other species 111 the light of present knowledge oceanographic which occupy the space made available. Carried to a conditions during the late 1930’s and very early logical extreme, in the face of increasing pressure by 1940’s were generally warm (Reid et a1 1958) and man the selected species (in this instance the sar- judging from the year classes produced appear to dine) would ultimately be restricted to an ecological have been favorable for sardine spawning, and the range and population size such that its ecological adult population was certainly large and well dis- niche iiivolved little or no overlap with other species tributed over the spawning grounds. Finally, heavy at its trophic level. cropping by the fishery must have served to some extent to increase the habitat for the pre-recruitment aiid Iiadovich (in press) show that the relative sig- sardines. Thus it appears that the fishery, by making nificance of a year-class to the fishery, especially the the population more responsive to the favorable en- Monterey-San Francisco fishery, is a function of the vironment, was a positive factor with respect to the apparent origin of the year class as judged by the production of large year-cla4ses during the period of relative catches when they are one year olds; those favorable enrironmeiit, and because a young popula- originating ill Central California weighing more tion is more efficient with respect to food conversion, heavily in the catch, in particular the Central Cali- growth, and natural mortality, the stancling crop of fornia catch during subsequent years. Thus it seems adult sardines may well have been larger than (luring that sardines spawned to the north have a predilection earlier years. to occupy the northern part of the range, and vice Beginning in 1944-45 the northern fisheries (British versa. The last northern year-class (by Felin’s and Columbia, Oregon, and Washington) began to decline Radovich ‘s dcfiuition) m7as that of 1947, apparently (Clark and Marr, 1955). These were somewhat cooler the progeny of a fairly small stock of spawners judg- years, and years when the salinity at La Jolla sug- ing by the distribution of the fishery that year. gested little influence of tropical water, so this decline, The Monterey fishery collapsed in 1950 atid 19511, which initially resembled earlier fluctuations, may three and four years later, that is when these 1947 have been caused at first by inhibition of the seasonal fish were three aiid four year olds. Since they con- northward migration of the sardines by cooler water tributed heavily to the southern California catch and accelerated flow. those years (in fact, the 1947 year-class made its The decline, unlike earlier dips, continued ; the far greatest contribution in southern California as three northern fisheries virtually ceased to exist after 1946 ; vear olds during 1930-51, strongly suggesting that its the Monterey-Sa11 Francisco fishery virtually ended in distribution changed), it must be concluded that the’ 1950 ; and the southern Califoriiia fishery declined in environment altered their normal northerly oriented 1953 to about one percent of its peak, later rising to distribution during those fishing seasons. It thus ap- fluctuate around 50,000 tons a year until 1958. pears that contraction of range was brought about by Coinciderit with this, the cooler than normal tem- failure of the adults to migrate as far north as usual peratures in 1943 and 1944 strengthened and persisted during the spawning season which of course would in through 1956 (Reid et al, fig. 18) instead of oscillat- itself preclude a northern year-class, and also was ing between cold and warm anornalies as in the past, caused by failure of spawn to survive in the north, a though there was a slight warming in 1946 and 1947. phenomenon that feeds back into the first operator. In 1944 and 1945 quite poor year-classes were pro- After 1949 the sardine population was essentially duced and the catch fell dramatically during 1946-48. confined to the area from Point Conception south. During 1946-48 moderate sized year-classes were pro- Substantial catches (over 300,000 tons) were made duced and the catches rose to over 300,000 tons in during the fall of 1919 and 1950, so there must have 1949 and 1950, though nearly all the catches were been substantial spawning in the spring. Yet, the re- made off Southern California. Among other things, siilting year-classes were apparent failures. A further this series of events illustrates the sensitivity of a major effect of the environment with respect to these young, heavily fished population to fluctuations in year-classes is suggested by their catch curves. The year-class size which may have bren caused by en- 1919 j-ear-class shows pronounced negative mortality vironineiital changes. between ages 1 and 5, the catch of five year olds being During earlier years the Pacific sardine ranged six tiines as large as the catch of four year olds, and from lower 13aja California to British Columbia. By almost as large as the catch as of two year olds. I’re- 1950 the population was essentially confined between vious to this year-class the typical pattern was for lower Baja California and Point Conception, roughlp maximum contribution a5 two year olds or three year a 50 percent reduction in range. Most generally, a olds. Similarly, the 1950 year-class made its maximum reduction in the range of an animal must be accompa- coiitribution at age four. From this record it is obvi- nied by a reduction in numbers. One critical question ous that the virtual population method markedly is which came first. A reduction in nuiiibers does not underestimates their size relative to other year-classes necessarily involve a dramatic change in range ; the though there is no evidence they were large year- population may simply become more thinly dispersed. classes. Perhaps more important the catch curves In the present instance the coincidence of the contrac- clearly show that the time-space distribution of these tion in range with cooling of the ocean climate lends year-classes with respect to the fishery clearly de- support to the thesis that a change in the environ- parted from the iiorinal, presumably in response to the ment restricted the habitat of the sardine, and that same kinds of environmental changes that altered the the population adjusted to a new level commensurate distribution of the 194’7 year-class after its second with its new range. year of life. The reduction in range could have come about The ocean climate remained cold until 1957, alld simply because the adults were unable, in the face of the now reduced population remained centered in swifter currents and/or keener competition, to occupy the south. During these years, the souther11 Califorilia their range. This probably was one of the factors, but fishery, instead of being located in the center of the more significant is the probability that the contraction range of the species, was stationed at the northern in the range was a result of failure of northern edge of the range, and since the fishery has the essen- spawning because of the environment. Felin (1954) tial character of a day fishery it was exceedingly sensi- REPORTS VOLUME VIII, 1 JUTJT 19Xl TO 30 JCNE l9GO 63 tive to minor changes in the north-south distribution where in the world, Proceedings of the World Sardine of the sardine, the landings ranging from 3 to 76 Conference, in press), and for which it is possible thousand tons. From 1957 through 1959 the ocean to assign a logically significant effect on the Pacific climate warmed and the sardine population ranged sardine, suggests that the environment has played the farther north in 1958 arid 1959 (as far as Sa11 Fran- major role in the regulation of the population. In a cisco), though the catches have not suggested a sig- purely negative sense, the impossibility to date of nificantly larger population. being able to satisfactorily apply any of the existing The recent warming of the California Current and and in some instances highly successful mathematical slackening of its speed is the reverse of the events models of fish populations to the excellent series of that were associated with the decline, so presumably catch statistics on the sardine is evidence that the a recovery should ensue if the oceanic climate con- environment is playing a major and possibly dominant tinues to be favorable. If the recovery follows a course role, and that the fishery’s chief but not necessarily which is the reverse of the decline it should proceed exclusive effect is to make the population more re- as follows: (1) more northerly distribution of exist- sponsive to changing environmental pressures. ing adults and more northerly spawning (already underway but involving very small numbers of sar- REFERENCES dines), (2) maturity of northerly spawned fish, and Ahlstrom, E. H., 1954. 1)istrihution and abundance of egg and a spawning by them, and (3) growth of these north- larval populations of the I’acific sardine. Fish. Bul. 93, U.S. erly spawned fish to commercial age (2 year olds). Fish & Wild. Ser.: 83-140. Obviously this will take time, and a real recovery Ahlstrom, E. H.,1958. Sardine rggs and larvae and other fish could not be expected before 1961 at the earliest. In larvae, 1’:icific Coast 1956. Spec. Sci. Rept.-Fisheries, KO. the meantime the population of competitors such as 251, U.S. Fish & Wild. Ser.: 87. Ahlstrom, IC. H., and D. Kramer, 1967. Sardine eggs and larvae the anchovy should decline as a result of the shift ant1 other fish larvae, I’adfic Coast, 1953. Spec. Sci. Rept.- in ocean climate and increased competition froin the Fisheries, KO. 224, U.S. Piah (e Wild. Aer.: 90. sardine. Arthur, D. K., 19SG. The particulate food and the food re- In conclusion, it is possible to divide the natural sources of the larvae of three pelagic fishes, especially the I’acific sardine, Sardinops caerulea. l’hD. Thesis, University pressures that regulate the Pacific sardine into three of California, Scripps Institution of Oceanography, : 231 groups. The first is the physical-chemical environment (Typewritten). and trophic levels below the sardine. These seems to lieverton, R. .J. H.,and S. .J. Holt, 1937. On the dynamics of affect the sardine directly in a density independent rsploitetl fish populations. 0. IC. Nin. Agr. and Fish. Fish. manner arid indirectly in a density dependent man- Itzrest., Ber. 11, 19: 533. ner by contracting the available habitat. The second Brett, J. R., 31. Hollands, and r). F. Ahlderdice, 1958. The is the within-trophic level competition. This involves effect of temprrntore on the cruising speed of young sockeye the sardines themselves and such species as anchovies. and coho salmon. .I. Fish. Res. Rd. Canada, 15(4) : 687-605. These would seem to exert their effect in a density (’l:irli, F. N., ant1 .J. C. Marr, 1956. Population dynamics of dependent manner either by competition for food the I’acific sardine. Part ZI of Cadifornia Cooneratiwe Oceanic and space or by specialized predation, e.g., adults E’isheries Investigtrtions Progress Report, 1 July 1953 to 31 March 193.5, : 11-48. eating larvae, large larvae eating small larvae, etc. The third is predation of various sorts from higher Coshing, I). H., lK7. Production and a pelagic fibher>. Xin. Ag. Fish. and Food, Or. Br. Fishery Inc. Ser. 11, 18(7) : trophic levels. The rate of this source of pressure I-VI, 1-104. may well be density dependent and self-adjusting, Felin, I!’. E., 1954. I’opulation heterogeneity in the Pacific pil- and perhaps this regulatory mechanism is of second- c*h:lrtl.Fishery Bnl. 86, U.S. Fish & Wild. Aer.: 201-225. ary importance. (Man, of course, preys heavily on Felin, F. E., .J. JIacGregor, A. E. Lhughertj, and L). J. Miller, many of the potential predators tending to produce - 1954. Age and length wmposition of the sardine catch off the a compensatory effect.) Pacific Coast of the rnited States and Jlesico in 1953-54. Finally, there is the role of the fishery. It seems Calzf. Fzsh and Game, 40(4) : 423-431. clear that the minimum effect of the fishery is to JIacGregor, .J. S., 1959. Relation hetwern fish condition and sensitize the population to environmental changes. population size in the sardine (Wardinops caerulea). Fish. It Bull. 16G, [-.E. Fish. and lt‘ald. Ser. is equally clear that the growth-mortality regime of the sardine leaves little to be gained by manipulating Marine Research Committee, 1950. California Cooperative Sar- din0 Research I’rograai, Progress Report l%fi: 34. a year-class after it enters a fishery (Clark and Marr, 1955). This leaves the question as to whether or not 1952. C”a1zf. Cooperative Sordane Research Program, Progress Report, 1 Jan. 1951 to June 1!)52 : 51. the fishery has or will reduce the pre-spawning popu- 1933. Calif. Cooperatice Oceanic Fisheries Investigations, lation to a point that there are not enough survivors Progress Report, 1 Julj 1932-30 June 1953: 44. through the density independent mortality stages to 1955. Calzf. Coopet ative Oceanic E’ishei zes Investigations, efficiently utilize the environment during density de- Progress Repoi t, 1 July 1053-31 March 19X : 52. pendent stages. Even a positive answer to this question leaves unresolved the problem of the practicality 1!)30. Calif. (‘ooperative Ocennic Fisheries Znuestigations, 1’1 ogresx Report, 1 April 1955-30 June 1956 : 44. of population manipulation through regulation of 19.58. Calif. (’oopercilioe Oceanic Fisheries Investigations, the catch. Progress Repoi t, 1 .Julj 19.76-1 Jan. 1958 : 67. The fact that at this writing it is possible to identify hlarr, J. (’., 1951. On the use of terms nbuntlanre, a~ailability, environmental changes that parallel changes in the and apparent ;~buntlaiicrin fishery biology. Copeia, (2) : 163- sardine population along the Pacific Coast (and else- 160. 64 CA1,IFORSIA COOPERATITE OCEASIC FISHERIES ISVESTIGATIONS

Jlnrr, J. C. In press. The causes of major variations in the Rchaefer, ill. B., 0. E. Sette, and J. C. illarr, 1951. Growth catch of the Pacific sardine, (Sardinops caerulea, Girard). of the Pacific Coast Pilchard Fishery to 1942. Res. Rept. 3-0. Proceedings of the IVorld Sardine Conference, FAO, Rome, 29, V.S.Fish adWild. Ser.: 1-31. Italy. Sette, 0. E., 1953. Consideration of mid-ocean fish production Phillips, J. E., and J. Radovich, 1952. Surveys through 1931 as related to oceanic circulatory s~stenis.Jour. Mar. Res. 14 of the distribution and al)untlance of young snrtlines, (Sardi- (4) : 398-414. ,tops caerulea) Fish Hul. 3-0. 87, Calif. Dept. of E’iuh and Game: 1-63, Sette, 0.E., 1960. The long term historical record of meteorological, oceanographic and hiological data. Califorilia Cooper- Railovicli, J. In press. Some causes of fluctuntioiis in catches utive Oceatric Fishery Investigations Reports 7: 181-104. of the l’acific sardine (Sardinops caerulea, Girard) , l’roc. of the World Sardiiie Conference, FAO, Rome, Italy. Thrailkill, J. R., 1937. Zooplaukton volumes off the Pacific Reid, J. I,., Jr., G. I. Roden, arid J. Wyllie, 1958. Studies of Coast, 1936. Spec. Sci. Kept. Fisheries, No. 232, U.S. Fish & the California Current System. Progress Report, California Tl-zld. Ser.: 52. Pooperatire Fisheries Iiivestigations, 1 July 1936-1 Jan. Widrig, T. 1934. Methods of estimating fish populations, 1958 : 27-56. AI., with application to Pacific sardine. Fish Hut. 94, U.S. Fzsh & Ricltrr, E., and R. E. Foerster, 1948. Computation of fish IT. Wtld. Ser. : 141-166. protlaction. Bull. Hing. Ocean. Coll., 11, Art. 4 : 173-211. Kicker, W.’E., 1958. Handbook for computations for biological World Sardine Conference. In press. Proceedings of the World stntistics of fish populations. Fish. Res. Bd. Canada Bull., Sardine Conference held in September 1939 in Rome, Italy 119: 1-300. under FA0 auspices. RELATIONSHIP OF VARIABLE OCEANOGRAPHIC FACTORS TO MIGRATION AND SURVIVAL OF FRASER RIVER SALMON

1. A. ROYAL International Pacific Salmon Fisheries Commission and J. P. TULLY Fisheries Research Board of Canada

INTRODUCTION north through Queen Charlotte and Johnstone Straits. In 1957 it was noted that a larger share of Management of the various Fraser River salmon the population approached Vancouver Island from fisheries requires an intimate knowledge of the abund- the north, a larger percentage (approximately 16 ance and movements of salmon runs, which is drawn per cent) diverted through Queen Charlotte Strait largely from a detailed study of commercial catches. and it was also noted that the fish were slightly de- Abnormalities in abundance, in routes of migration layed, and migrated over a longer period of time. or times of occurrence, unless forecasted, may negate the beneficial effect of fishing regulations formulated In 1958, the sockeye run of approximately 19,000,- in advance of the fishing season. A flexible method of 000 fish failed to appear off the west coast of Van- regulatory adjustment during the course of fishing couver Island in its usual initial landfall and arrived may help rectify the adverse effects of vagaries in the northerly of Vancouver Island in the Queen Char- salmon population but it cannot be accepted as a lotte Sound area. The fish then moved in a southerly substitute for advance knowledge and planning. direction with an estimated thirty-five to forty per cent of the total population entering Queen Char- lotte Strait, thus approaching the Fraser River, from OCEANOGRAPHIC FACTORS AFFECTING the north instead of the usual route through the Juan SOCKEYE SALMON MIGRATIONS IN de Fuca Strait. In addition to the vagary in their THE NORTHEAST PACIFIC route of migration the fish appeared in the fishery ten The formulation of fishing regulations in advance days later, and oyer a longer period of time than of the fishing season requires knowledge of abund- was anticipated on the basis of previous catch rec- ance, timing and migration routes of the Fraser River ords. sockeye runs. It is known that adult sockeye abund- The effects of the radical departure from normal ance has been affected considerably by marine sur- in the route and timing of the 1958 Fraser River vival rates which have ranged from 4.22 to 18.54 per sockeye populations were serious. A surplus escape- cent in recent years. However, since there is some ment of 1,500,000 fish occurred. These could have reason to believe that freshwater and estuarial fac- been harvested had the fish followed the usual migra- tors as well as the physiological condition of seaward tion route at the usual time. The delay in the arrival migrants may influence subsequent marine survival, of the escapement on the spawning ground disturbed an oceanographic or ecological study of marine con- the normal relationship of the spawning population ditions, as related to sockeye survival, appears rather to its reproductive environment, and map have seri- impractical until these factors are more fully under- ously jeopardized its reproductive potential. The stood. While variations in survival are important, it physiological development of the fish was so disturbed is relatively simple to make compensatory adjust- in the latest arriving section of the population that ments in fishing regulations for unexpected changes an estimated ten per cent of the total escapement in abundance, provided that the timing of the runs is failed to reach their spawning grounds, some 350 inherently consistent, as it is in most years, and the miles upstream. routes of inshore migration follows a consistent pat- 111 analyzing all possible causes of the vagaries of tern. If, however, the measurement of abundance the route and timing of the 1957 and 1958 runs of from commercial catches in initial fisheries is con- sockeye to the Fraser River it is logical to conclude fused by a delay in arrival, or a change in the route that the physical marine environment, at least dur- of migration, or both, then the problem of properly ing the maturing year, must have been responsible. managing the fishery becomes most complex. Oceanographic observations over recent years show I Normally the bulk of the maturing Fraser River that when the fish went to sea the normal currents sockeye arrive off the west coast of Vancouver Island and approach the Fraser River around the south end and water masses were present off the approach to of the Island through Juan de Fuca Strait (figure 1). Juan de Fuca Strait, and that anomalous oceano- A smaller part of the population, less than ten per graphic conditions first developed off the southern cent, passes around the northern tip of Vancouver British Columbia coast in 1957 and were intensified Island and approaches the Fraser River from the in 1958. 66 CATJFORNIA COOPERATIVE OCEANIC FISHERIES ISVESTIG.\TIOXS

FIGURE 1. Vancouver Island and Fraser River approaches.

The trans-Pacific drift current normally flows east- sion showed that the distribution of feeding Fraser ward into the Canadian approaches. In the spring of River sockeye apparently tended to shift iiortlirrard 1957 this flow began to veer towards the north. By coincident with the warm water intrusion, and that mid-summer 1958 the flow was due north in the region no concentrations of these fish were found in the area within 600 miles of the Canadian coast. Associated westward of Vancouver Island as in precediiig years. with this change of current direction, warm water It niay be postulated that the more northerly landfall intruded northward. Tt was evident as far as the of the 1!157 and 1958 sockeye runs was the direct northern end of the Queen Charlotte Tslantls. It result of temperature preferellee on the part of the reached its maximum extent in August 1958, and has feeding sockeye in the ocean. The latenr\s in arrival degenerated somewhat since then. In the affected re- might be ascribed to the conseqnent displacenieiit of gion the seawater temperatures were 2" to 3" centi- the sockeye to more distant feeding grounds or to grade warmer than is considered normal at the sur- a circuitous migration path to avoid the warm water face. The temperature anomaly decreased with depth, iiitrusion. This concept is tenable because in l!)M arid vanished at about 500 meters. The occurrence, there was a narrow band of cooler, near normal waler, duration, and extent of the intrusion was determined close along the west coast of the Queen Charlotte by observing the temperature increase at 180 meters Islands. This would provide a coast-wise migration depth, well below the influence of seasonal heating route into Queen Charlotte Sound. A modification oB and cooling. The increase in surface temperature is this concept would be that strengthened ocenii CIII*- significant because the normal seawater temperatures rents slowed the migration rate of the fish. It can in this area, at the time of the summer shoreward be further postulated that the more northern distri- migration of sockeye salmon, are near the upper limit bution of the fish brought the sockeye into higher lati- of preference for sockeye. tudes where the days mere longer dnring the late Sampling by investigators in the northeast Pacific spring and sumnier months. This may have thc (Ift'cct during 1956, 1957 and 1958 under the sponsorship of retarding maturation and thus delaying the cor- of the International North Pacific Fisheries Commis- related migration. A study of the degree of matnra- REPORTS VOI*UAII’: 1‘111, 1 .TIJJ,Y 1959 TO 30 JUNE 1960 67

tion and energy reserves of sockeye arriving at the allow for equality in the total catch of each of the mouth of the Fraser River commencing in 1956 re- two national groups involved and €or adequate escape- veals that maturation and energy reserves were ap- ment. The pink and sockeye runs entering Juan de proximately the same in 1958 as were those of sorkeye Fuva Strait are highly vulnerable to a Canadian fish- in the preceding years, in spite of the ten day differen- ery opwating on the north side of the Strait. A U.S. tial in arrival. These findings teiid to support the lat- fishery on the south side of the Strait has never been ter of the above hypotheses. The increased temporal consistentl,~effective and its lack of siicwss has been dispersion of the run can also be explained on the attributed in a large part to the vertical distribution basis of light response. If the fish were spread over of the fish in the southern area. If the U.S. fishermen a range of latitudes wider than normal, the resultant were to develop a successful fishery in the southern greater variation of day-length encountered would side of the Strait the whole scheme of regulation account for the increased interval over which fish ~vouldhave to be revised with a major displacement arrived in the fishery. of the existing over-all fishery. Whatever the pattern of forces that created the study of the currents in Juan de Fuca Strait re- vagaries in 1957 and 1958 Fraser River sockeye runs, veals that in the northern half of the Strait the ebb a serious derangement in the management of the 1938 trilnsport of mixed salilie and land drainage water fishery resulted. The economic importance of predict- greatly exewtls the flood transport of highly saline ing these vagaries in advance has been established. ocean water. The reverse situation occurs on the southern side. Since the fish are apparently seeking THE RELATION OF OCEANOGRAPHIC waters of reduced salinity as they approach the CONDITIONS IN GEORGIA STRAIT E’raser River they probably prefer the lower salinities TO PINK SALMON SURVIVAL of the northern shore and for this reason it is highly doubtful if salmon are available at any depth or in’ Great variability in the abundance of pink salmon any great numbers on the southern side. populations has been well established in North Amer- ica and the runs of this species to the Fraser River The eastern end of the Juan de Fuca Strait lying are no exception. Annual catches of pink salmon in principally in \Vashington State consists of a zolle of the approaches to the Frasrr Hiver have varied from mixed water made up from the flood transport of sea- 950,000 to over 11,000,000 fish during the last twelve water and land drainage. In this area the Fraser biennial runs. Predictions of ultimate adult survival River salmon separate from other stocks destined for based on the success of fry emergence have been southern I’uget Souiid streams and apparently pro- notable for their inaccuracy, hence any reliable cerd northerly to a major extent through Rosario method of predicting adult survival must bc based Strait (east of San Juan Islands). The alternate chan- on something more than the number of pink salmon nels represented by ITaro Strait (west of San Juan fry migrating to their marine feeding grounds. Islands) and San duan Channel are available to them Studies of the relationship of various fresh water but based on availability indices these channels are and estuarial environmental factors to ultimate adult not utilized to any great extent except by fish migrat- survival have resulted in what appears to be a fairly ing during the late summer and early fall period. reliable method of predicting the approximate abnnd- Occasionally, however, the northerly nlovement of fish ance of adnlt piilk salmon populations destined for will shift suddenly froiu Rosario Strait to Haro the Fraser River. Surface seawater trinperature for Strait, and very occasionally the principal movement the period April to August in Georgia Strait, which will shift into Canadian waters. Thev shifts in niigra- lies adjacent to the mouth of the Fraser River, has tion have a drastic effect on the management of the shown a close inverse correlation with the total pink fishmy and niust be related to changes in the physical salmon catch of the following year (r = -0.8.i9t5). environment of the available channels. Studies of The period April to August includes the time during salinity and temperature changes as related to varia- which young pink salmon may be expected to be tioris in the flow of the Fraser River arid to wind di- residing in Georgia Strait in advance of their emigra- rection in the approach channels to the Fravr River tion to the Pacific Ocean. Whether the water tempera- wonltl probably provide reamIs for sudden and ture of Georgia Strait has a direct or indirect rela- periodic shifts of the fish from their normal approach tionship to survival is not yet known and further channels, to those which are not normally used by knowledge regarding the relationship can be obtained the majority. It is well established that the major flood only by a detailed study of the young pink salmon tide transport is through Rosario Strait and the major during their estuarial existence. ebb transport is through Haro Strait. Local and tem- porary fluctuations in temperature and salinity or in OCEANOGRAPHIC FACTORS AFFECTING THE daily net transport through the various channels MIGRATION OF SALMON THROUGH from Georgia Strait and the Fraser estuary have not COASTAL CHANNELS ADJACENT been deterniined. An understanding of the causes of TO THE FRASER RIVER the shift in migration and an ability to anticipate the The Fraser River salmon fishery is composed of change even twenty-four hours in advance, would several individual units each of which is highly effi- eliminate a very serious problem in the proper man- cient and subject to severe restricfions in order to agement of the fishery. 68 CALIFORNIA COOPERATIVE OCEANIC FISHERIES ISVESTIGATIONS SUMMARY viva1 of Fraser River sorkepe salmon. Oceanographic It appears that offshore variations in influences ill inshore waters largely coiitrol the uki- oceanographic features have an effect on the feeding mate survival rate of the piilk sahnon aid the ap- distribution, path and time of inshore migration, tem- proach characteristics of both Fraser River sockeye poral dispersal of the inshore migration and the sur- and pink salmon migrations. OBSERVATIONS ON THE ECOLOGY OF THE PACIFIC COD (GADUS MACROCEPHALUS) IN CANADIAN WATERS

KEITH S. KETCHEN Fisheries Research Board of Canada, Nanaimo (Abstract) The Pacific cod (Gaclzis nzacrocephalzls) occurs in warmer water than in the apparently more produc- economic concentrations around the great arc of the tive regions of western Alaska to Kamchatka. Possibly Sorth Pacific Ocean from Korea to Oregon. Near the as a consequence of these warmer conditions, growth southern limit of its range in the eastern Pacific rate is much more rapid; maturity is reached at a (British Cdlumbia waters) the cod inhabits much much earlier age; and lifespan is relatively short. ~- There are more pronounced fluctuations in the year- 1 This paper will be published in the Journal of the FiSheTieS Research Board of Canada. to-year average catch per unit of effort.

FISHERIES OCEANOGRAPHY

ELBERT H. AHLSTROM U. S. Bureau of Commercial Fisheries La Jolla, California

This symposium has emphasized the importance of Rollefson noted that there was a higher percentage of two concepts in fisheries biology: 1) that of popula- abnormal eggs among youiig stage eggs, in which the tion abundance including the fluctuations to which embryo had not yet completely encircled the yolk populations are subject, and 2) that of availability in- mass. He found that by dropping eggs from a height cluding those factors which result in the aggregation he could produce this condition in eggs that were nor- or dispersal of a species. By interest and preference mal before dropping. He was led to the coriclusion I am going to speak principally about the first of that the injury was mechanical and that it could be these-population abundance. pr*oduced in the sea. Dr. Sette, in the introductory speech at the sym- Temperature has often been suggested as an im- posium, outlined very lucidly the problems of fluctua- portant factor in year class fluctuations. Garth Mur- tions in marine fish populations. He gave instances phy has pointed out the effects of warmer water on of the magnitude of such fluctuations. You will recall success of sardine survival, and in this hypothesis he that in the Atlantic mackerel the observed fluctua- is following in the footsteps of various other workers. tions in year class size were of the order of 15,000 Dr. Sette prepared a paper some 20 years ago which to 1. Instances of such fluctuations could be drawn stressed the relation between warmer temperatures from most fish populations that have been studied. In and good year broods of sardines. Unfortunately this the Pacific sardine, with which I am better acquainted paper was iiever published, but it has been commented than other kinds of fish, the most successful year class upoii fairly frequently. Although successful year on record was several hundred times as lar*ge as the classes are more frequent during warm periods, yet least successful. The ody exception that has been ad- there have been middling or poor ones during such vanced is the population of yellowfin tuna, a tropical periods also. 1941, for example, was one of the warmer species. Dr. Schaefer indicated that there was no evi- years on record-yet the 1941 class was only middling dence of marked fluctuations in success of year classes large. There is no evidence as yet that the 1958 year- of yellowfin. Differences between year classes were of class of sardines was at all successful, yet 1958 was the magnitude of 4 or 5 to 1. a year of uiiusnally high water temperatures. Certainly in temperate latitudrs, large fluctuations Walford postulated a correlation between salinity in the success of individual year classes is the rule for and successful year-classes of sardines. His model any species that has been investigated over a period was actually an attempt to correlate good year broods of years. with high prodnctivity. High salinity was assumed by There have been a number of attempts to account Walford to have resulted from intense upwelling for fluctuations in fish populations. Among the causal with the attendant enrichment and increased produc- factors that have been suggested are the following tivity. There are quite positive ways of identifying seven : upwelled water, but a combination of several obser- winds vations are needed : temperature, salinity, oxygen, currents and nutrient colicelitrations such as phosphate. Up- temperature welled water is characterized by its high salinity and salinity phosphate conteiit and its low temperature and oxy- general productivity gen content. IYalford obtained an excellent correla- competition, either interspecific or intraspecific tion over the period of years he was using-which predation were the years between 1934 and 1941. The correlation breaks down entirely in more recent years. Dr. Blackburn discussed the emphasis placed by If any hypothesis can be laid at rest it is that Carruthers on “winds”. Winds are important indi- which associates successful year broods of sardines rectly in effecting transport of larvae into favorable with high prodnctivity. Productivity, as measured or unfavor’able conditions. Iiollefsoii postulated an- by plankton volumes, has been high during the past other effect of winds on the success of year classes- ten years ; success of sardine survival has been low. that of direct injury to eggs through tnrbulence resulting from strong winds. Rollefson was led to this In fact, if productivity is a major consideration in conclusion by observing the high proportioil of “ab- the siwcess of a year-class of fishes, oiie would ex- normal” appearing eggs that are taken in most plank- pect that the increased food should result in better ton hauls. Rollefson was working with cod eggs, but than average survival of all species which are pri- this observation applies to pelagic fish eggs in general. marily plankton feeders. This has not taken place. It is one of our vexing problems with sardine eggs. In fact the northern anchovy, a direct competitor with the sardine, has prospered during a period when physiological differences between the two species. The the sardine has had unusually poor survival. This threshold temperature for sardine spawning is ap- contrast is so marked that one is justified in posing proximately 13", that of the anchovy is approxi- the problem in another way: What is there in the mately 1" lower. Is the adjustment of an organism environment or the response of these two species to to its environment so critical that a decrease of say thp environment that has favored successful year 1" C. in average water temperatures can mean the broods of anchovies at a time when the success of tlifterence between successful and unsuccessful year- sardine year-classes has been unusually low? The classes? I am only going to pose this question-not diff ereiices in survival may result from inherent attempt to answer it. FISHERIES OCEANOGRAPHY WARREN S. WOOSTER ’ Scripps Institution of Oceanography University of California, La Jolla, California In considering how to summarize the oceanographic stirring aiid cmistal np~~~lling)or secontlary effects aspects of the papers we have just heard, I found of thr wind-driven circulation (for exaniple, doming my thoughts falling into several distinct rategories- or ridging). the definition and goals of fisheries oceanography, cur- The near surface circulation may aff~ctdirectly rent practice in this field, adthe success of this the distribution or abundance of organisms at all approach. trophic lcvc~ls. In addition, changes in the intensity Some may feel that fisheries oceanography is noth- of the refertilizing procc’sst’s are reflected in the time ing more spwific than the broad assemblage of prob- mid spaw distribution of primary produc:tioii. This lems being studied by oceanographers working for in turii affects production iii thr n~xthig1ic.r trophic fishery laboratories. The interests of marine fisheries level, aiid so on. with assorted time and spac:e lags, scieiitists and those of oceanographers overlap in SO to the desircd fish. At each step ill this line of rwson- maiiy arras that aliiiost any marine rc>search can be iiig, rrfinmnieiits and eoniplications are involvrtl, but, inclnded in a suitably broad definition. But unfor- the basic thciiie of the model reinaiiis: tniiately \uch broad definitions haw little opera- ii (’haiiges in the wind firld lcacl eve~ltually to tional value. changes in the succ(~ssof fishing. ’’ One might speak, however, of a somewhat restricted Th(3 goal of investigatioii of this niotlel is often roil- area of marine research which could be calld ecologi- siderrd to be prdictioii. Crrtainly if one could fore- (.a1 oceanography. This deals with the relationship cast awurately the c~hang(~siii ahiiiidaiwe, tlistribrition between the ocean as a physico-chemical environment atid availability of fish, this \wnld be of great eco- and the populations of organisms inhabiting it. A nomic significance. Riit, concc~ivably, a useful prp- special case exists in which the populations consist of diction coiild result from the blind statistical treat- species of commercial intmest. This point of view meiit of a large Iinniber of variables, rather than from lrads to tlie following definition : a fundanirntal iintl(.r.staiitliiig of the iiitclrplay of the Fishcries Orea7iograpliy-the study of oceanic proc- pertinent atiiiospheric, oceanic, aiicl biospheric processes affecting the abundance and availability of esses. It is the acquisition of this fiuitlanieiital knowl- commercial fishes. edge which is the scieutific goal of fisheries oceanog- Obviously populations of comniercial fishes are af- raphy. fected by other than oceaiiic processes. In particular C‘iirrc.iit prartice in fisheries owanography has coii- they are subject to the pressure of an aggressive noii- centrated on clociuneiitiiig and trying to intcrprcJt the eiiviroiiriieiital factor, the fisherman. Much of fisheries changes in the iiiariiie environment. One of tlic lcad(~rs research has resulted from the fear that this pressure in this field in the 1Tiiitcd States was the late Town- was excessive and ~vouldsoon lead to decimation of send Croniw.11, who originall>- proposed this syin- the stock ; thus it has concentrated primarily on the posium. His first iniportaiit work, iii Honolulii, JYBS a dyiianiics of the populations involved. The working study of wiiid-induc.ec1 upwelling aloiig the quator. hypothesis of the fisheries oceanographer, oil the other Subseyiwntly, the IIonolnlii group has examined the hand, is that variations in apparent abundance are mechanisms of surface enrichnicwt near the Hawaiian due primarily to changes in the cnvironnient. These Tslands aiid to the north. At Scripps Institiition and changes niiist bc described aiid understood before the the Tnna Cornmission, thcre have been st,iitlies of role of man can be propcrly evaluated. cwastal upwelling, oceanic2 froiits, ant1 of processw The relationship is traced out in a model which such as those Crom\wll labelled “doming” or iiridg- starts with the observation that sigiiific~ant c1iangc.s ing ”. Dr. Sette and his c*olleagiies have bwn exaniin- in the atmospheric pressure field ocmr from place iiig past weather and the riiariiie climate, looking for to place and from time to time. These ehanges lead long-term changes related to those in the fisheries. 1111 to variations in thc stress applied to the sea surface of these iiivestigations have been facilitated by the by the wind. It is now generally believed that the presence of certaiii cdoiispiciions featurcs or disconti- major near-surface circnlation of the oc~aiiis wind nuities. As Henry Stoinniel has suggested, studying drivrn, so that the changing \tilid stresy caiises the oceans resembles diswcting a lobster-if is easier changes in the velocity, depth, breadth, transport or to do at the joints. other characteristics of tlie surfacr cwrrciits. Further- That has been the success of fishery oceanography more, tlie processes whereby the surface layw is refer- in recent years? Certainly in the Pacific it has had no tilized with nutrient elements from below appear to dramatic impact 011 the cornniercial fish(4es. L\iitl if be either directly mind-prodnced (for example, wind the goal be considered prediction, very little SII(T~SS

1 Contribution from the Scripps Institution of Oceanography can be recorded. Yet the fund of basic knowledge of 74 CALIFORSIB COOPERATIVE OCEXKIC FISHERIES ISVESTIGATIOSS the ocean has increased tremendously. The near sur- other important surface-enriching mechanisms are face circulation and the distributions of properties known. In short, the general scheme by which atmos- such as temperature, salt aiid oxygen have been much phere, ocean and biosphere are interrelated is taking more adequately described. The theory of wind-driven form, and we are ready to formulate aiid test hypoth- circulatioii is well established. The variations in time eses having bearing on important and specific prob- and space of coastal upwelling are recognized, and lems in fisheries science. EVIDENCE OF A NORTHWARD MOVEMENT OF STOCKS OF THE PACIFIC SARDINE BASED ON THE NUMBER OF VERTEBRAE

ROBERT 1. WISNER’

INTRODUCTION within the 0 and 1 age groups. No adults (2-year fish 01’ older) were available from south of Sail Diego with Evidence on the mo\Tenients of stocks of sardines the exreption of a small wnple of 31 fish from the along the l’acific Coast of the I‘nited States and Mex- Gulf of California. The areas treated ranged from ico has been derived from tagging, from counts of ,Ilaska to the Gulf of California alld the years vertebrae, and from related data. The present study sampled from 1921 to 19-11. represents a second major attempt using numbers of vertebrae. Previous studies, reviewed below, have From tiic combined ag’e-group data Clark (1947) demonstrated an unknown but presumably significant reached a teiitativc conclusion that, “sardines found amount of hrtcxogeneity within the range of the in southern Tlomer California and the Gulf of Cali- sardine, as ~wlla considerable intermingling of stocks fornia constitnte a separate population which rarely inter~iiiiigleswith the iii6rc~~iorthern population, but between the reg ns of l’unta Eugenia (ill central Kaja California) , California, and the Pacific North- that a cotisi(lerabIe, aiitl perhaps variable, amount west. of intc~~haiigetakes placc throiighoiit the range of the northern population from*.Ilaska to I’ta. Eugenia in Tugging stutlies: An extensive tagging program, central I~oiwrCalifornia.” Tlie result\ of the tagging initiated in 1936: aiicl carried 011 coiitilluously until in the Magdalena I3ay area, though somewhat incon- 19-22, clemonstrated that the sardine stocks inter- clusivr, rc3countcd above, augmeiited Clark’s interpre- mingle considerably within the area from Pta. Eu- tatioii of vertebral data from thtl area and strength- geiiia to Ccntral (’alifornia and from Southern Cali- enrd tlir liypotli(& that the stocks of fiqh in the area fornia to British Columbia (Hart, 19-23; Clark ancl may well coiistitiite a separate popnlation that inter- Janssen, 1945). Some fish tagged iii Southern Califor- niiiigles little if at all with the more northern stocks. nia were recovered along the west coast of Hritisli Columbia and some tagged in the latter region were Growth SIudios: Felin (1954) attempted to assay recoycwd in Southern California. The tag-ibecoveries the hetcrogcnc~ityof the sardine popnlations from the denionstrated a rather rapid migration : sardines tag- stantlpoint of gron th characteristics, using age ancl ged ofl Southern California in February ant1 March length data gl~itil(~dfrom sii1111)1(>\of the commercial were retaken off British Columbia in the following landings in Southern and Central California during July ; others, tagged off British Columbia in July and the years 1942-1950. Tlie following excerpts froin her August, were caught ofT Sail Pedro, California, in the conclnsionr appear to be pertinent to the present sucreediiig December and January (Hart, 19433). As- study : sociated with this rapidity of migration is a size re- Complete intcrniisture and homogeneity in pop- lationship. Clark and Janssen (op. cit.) stated that, ulation of adult fish as sampled by the fishery “The largest sardines may be expec4ecl to move in different regions is not eridenced from data 011 quickly to distant fishing grounds whereas the smaller mean calculated lwigths. The apparent cline in fish tend to remain longer in the locality where tagged the growth characteristic. . . appears indicative of or to make short rnigratioiis. These smaller fish reach intraspecific populations in which there is limited distant grounds in later years when they haw grown intermingling, and suggests a series of over- to larger sizes.” lapping coastal migrations of more than one The one lot of sardines that was tagged south of stock. I’ta. Eugenia, in April, 1938, comprised 963 fish in \Volf and Daugherty (1960) have reported that Magdalena Bay. So recoveries were made from this t\vo-year-old fish comprised the biilk of the Southern lot, possibly because the number was too small to in- sure a return, or, as the authors suggested, because California catch in the 1958-59 scasoil (to be (Xis- sardine\ from the area do not intermingle with the cussed later). These fish were uniqne in that they more northern fish. averaged 191 mm. (7.7 inches) in standard length, 12 mm. (0.5 inches) smaller than the previous recent TTwtcDrnl Sftrdics: Clark (1947) compiled the re- iiiiliimum (1934-55), but not much smaller (3 mm. or sults of vertebral counts made by previous inrestipa- 0.1 inch) than the 1939 year-class as two-years-olds, ail tors in the 1Tiiited States and Canada-Hubbs (1925), outstanding year-cla It has fnrther been reported Thompson (1926). Hart (1983), and Clark (1936). ( Ibid., 1960) that the fish in Baja California averaged The 1947 data, when grouped by latitude, showed a even wiallcr (162 mni., 6.4 inches)-so much smaller rathcr definite trend from north to south, particularly as ___ to suggest that the fishery there was not utilizing 1 Contributions from the Scripps Instltutlon of Oceanography. the same stocks of fish as was the Southern California 76 C,\T,IFORSI,Z C001’F:RATIVE OCEI:hSI(‘ FISHERIES ISPERTIGATIOSS fishery. illso, the two-year-olds landed at Moiiterey samples studied are characterized by a marked reduc- were snialler than the Sail Pedro fiyh, indicating tion in average number of vertebrae. somewhat more population heterogeneity than usual (no samples from the Montercy landings were avail- METHODS able for the present study). These previous studies have all pointed either to a As age determinations were not available for Sam- mixture of stocks, of phenotypic or genotypic origin, ples from most of the localities treated, all age-groups that interniiiyle more or less completely within the are combined. Only the 1957-58 Eiisenada and Sail area from north and central Raja California to South- Pedro samples were aged (by CalCOFI personnel, ern California and from the latter area to British employing a subsample amounting to approximately Columbia, or to more or less sporadic influxes of non- 20 percent of each sample studied). In the other areas intermingling fish from other areas, presumably the studied, aging would have been possible only by arbi- south. trarily assigning an age to any given length fre- The present study attempts to cover essentially the quency. This was not attempted as age-length compo- same areas (Fig. 1) reported on by Clark, with the sition studies have shown that adult sardines (2 years exception of Ceiitral aiid Northern California, aiid or older) vary in length from 1.5 to 2.5 inches in each provides a body of data wel1,separated in time from age-group. Thus, an arbitrarily assigned age may re- any previous similar data. KO samples were available sult in an error of two or three years. from thwe northern areas, as it was not until the All fish, except one sample of 500 postlarvae from 19.58-59 season that the sardines reappeared there in the Gulf of California, were X-rayed for determination of vertebral numbers. The postlarvae were stained with alizarin and cleared in glycerin before counting. All vertebrae were counted by the author. The urostyle was iiicluded in the count. The nature of the data does not warrant detailed statistical analyses, aid the study is not intended as a purely racial one. The real significance lies in the cliff ereiices in average values, both in individual and in grouped samples, between the present body of data and that presented by Clark (1947) for the region north of I’ta. Eugenia, Baja California. Because of the rather small range in total numbers of vertebrae the standard deviations about the mean do not difler markedly. For the region north of Pta. Eugenia the range of deviations, including both Clark’s aiid the present data, is 0.5980 to 0.6719 (ave., 0.6191). The range of standard errors of the means is 0.0006 to 0.0225 (ave., 0.0149). This large range in standard error values is due to the quite large variation in number$ of specinieiis used (Table 1). An indication of the significance of the difference between means of Clark’s and the present data, for the northern region, is the average difference of 0.23, nearly one-fourth of a vertebra. In comparing this with the almost three-fourths of a vertebra that separates the Gulf of California fish (average mean, 51.00) froin those reported by Clark for northern .. Baja California aiid northward (average mean, CAPE SAN LUCAS I t 51.70), elementary statistical treatment will provide odds of but one in hundreds of thousands of chances that the two sets of data comprise a single population FIGURE 1. Chart of areas discussed, showing the localities from which samples were secured in the present investigation and in that of or stock of fish. Such orders of significance do not, Clark (1947). of course, pertain to the southern areas-from Pta. Eugenia southward and into the Gulf of California- sufficient numbers, siiice about 1950, to warrant a because of the striking similarity of the two sets of fishery or a sampling program (Marine Research data in this southern region. Committee, 1960). KO samples were available for study from the 1958-59 central California fishery, although RESULTS about 25,000 tons were landed at hlonterey and San Figure 2 is a plot of all samples embodied in the Francisco (Ibid., 1960). In general, the present data present data, arranged by area. Wherever the areas agree with Clark’s findings only for the areas south are closely comparable, Clark’s 1947 data are also of Pta. Eugenia. From there northward to San Pedro, plotted. Each solid dot represents an individual sam- California, the data show that a major portion of the ple of the present data, aiid each open circle repre- REPORTS VOLUME VIII, 1 JULY 1929 TO 30 JUNE 1960 7 '7

TABLE I

VERTEBRAL NUMBERS-ALL AGE GROUPS

Number of Vertabrae __- Number VertebraeAverage Locality Year, Month 49 50 51 52 53 54 of Fish ____ San Pedro ...... 1951: I 1 33 40 3 77 51.58 X 4 52 39 3 98 51.42 XI 2 31 28 3 64 51.50 XI1 1 11 10 __~. 22 51.41 1954: X 1 18 17 1 37 51.49 1957: X 7 78 69 4 158 51.44 XI1 10 225 228 20 483 51.53 L958: I 7 88 101 8 204 51.54 I1 5 87 74 12 179 51.51 V 2 39 49 1 91 51.54 VI1 4 101 99 7 211 51.52 VI11 .... 40 35 5 80 51.56 IX 7 68 78 6 160 51.51

San Pedro total ...... 2 51 871 867 73 1,864 51.51

Clark total ...... 4 119 3,039 5,802 677 9,652 51.73 ~- ~-___ ~__ --~~ Sari Diego ...... 1950: VI1 .... 2 89 96 5 193 51.55 19.51: I ._.. 1 10 13 1 25 51.56 1952: VI1 .... 1 11 22 2 36 51. 69 1957: I11 __.. 2 4 7 2 15 51.60 V ._.. 8 16 1 25 51.72 1958: VI 1 26 26 1 55 51.45 VI1 _._. 32 21 4 68 51.48 X .... 46 41 2 91 51.47 XI1 .... 26 25 1 53 51.49 1959: I ..~. 17 19 2 39 51.56 XI1 .... 37 34 1 76 51.42

San Diego total ...... 1 16 306 320 22 1 666 51.52 _____ Clark total ...... 2 2,085 3,923 452 6,553 51.73 ____ -~ ______-___ ~- Ensenada ...... 1952: IV .... 8 18 .... 26 51.69 VI11 .... 1 19 13 1 34 51.41 XI .... 1 16 23 40 51.55 XI1 .... 15 21 2 38 51.66 1956: IX .... 1 33 26 .... 60 51.42

Enscnada total prior to 1957...... 3 91 101 3 198 51.53

1957: IX .... 5 83 102 8 .... 198 51.57 X 17 315 434 33 .~~.799 51.60 XI .... 10 265 230 8 .... 513 51.46 1 562 24 .... 1,116 51.45 XI1 35 ____ 494 1957 total.-..- 1 67 1,225 1,260 73 2,626 51.51 ~- 1958: I .... 2 22 23 2 .... 49 51.51 I1 1 2 112 158 14 .... 287 51.63 111 .... 10 383 435 42 .... 876 51.57 IV .... 12 197 240 17 .... 466 51.56 V ._._ 9 181 2oc 13 .~~.403 51.54 VI .... 19 290 26E 5 .... 588 51.44 VI1 .... 20 362 340 12 _.._ 734 51.47 VI11 1 27 346 29s 18 .._. 691 51.44 IX .... 24 497 421 17 1 961 51.45 X .... 18 358 362 18 .... 757 51.50 XI ._.. 31 600 581 45 .... 1,257 51.51 XI1 .... 32 390 335 16 .... 777 ___51.44 1958 total. 2 212 3,714 3,66E 219 1 7,846 51.50

Ensenada total ...... 282 5.0~ 5,02S 295 10,67C 51.50 3 ______-- Cedros and San Benito Islands ...... 1946: VI11 ._._ 9 86 71 3 171 51.40 1952: XI .... 1 1C 11 1 21 51.52 1955 ...... 31 95 13 14: 51.87 1958: VI11 .... 2 21 2; __.. 5c 51.50 IX ..~. 1 107 8: 5 191 51.47 XI .... I 44 45 4 9; 51.54 XI1 .... 4: 4- 6 lo( 51.51

34; 32 .... 78: 51.55 Cedros and San Benito total-- ...... _.._ 1s -__ CALIFORNIA COOPERATIVE OCEANIC FISHERIES IKVESTIGATIONS

TABLE I-Continued

VERTEBRAL NUMBERS-ALL AGE GROUPS

Number of Vertabrae

Number Average Locality Year, Month 49 50 51 52 53 of Fish Jertebrae

Sebastidn Viscaho Bay. _.-...... _ -.. ... - 1953: IV 2i 20 54 51.30 1958: VI11 75 24 104 51.18 VI11 1C 24 5c 51.34 IX 11 17 ... 31 51.42 XI1 7c 58 132 51.43 1959: I 3c 35 73 51.45 IV 54 53 11: 51.39

Sebastidn Viscafno Bay total __..._..._...... _ 35 28s 231 558 51.36

Clark total ...... 22 42c 596 1,093 51.63 -__ ___ Pta. Eugenia to Cape San LBzaro __.._..._...._1951: V 18 14C 141 30E 51.40 IX 14 105 130 256 51.51 1952: I1 .... 7 11 18 51.61 IX 8 39 25 74 51.18 1953: I1 2 5 5 12 51.25 V 3 63 54 123 51.46 VI11 .... 15 18 34 51.59 XI 21 178 117 32 1 51.33 1956: VI1 6 73 55 137 51.40 VI11 3 51 87 144 51.63 IX 3 22 23 50 51.50 X 2 9 23 34 51.62 1958: VI11 9 103 55 168 51.29 IX 20 190 78 290 51.22

Pta. Eugenia Cape San Ldzaro total ....___.__.__ 3 109 1,006 822 24 3 1,987 51.39 Clark: total Pta. Eugenia to Pt. San Juanico--.-. 32 ___417 248 __710 51.34 Santa Maria Bay 1951: VI1 _... 27 17 45 51.42 X 3 31 43 78 51.54 XI _.__ 4 7 13 51.85 1952: IV 1 8 3 13 51.31 VI1 7 57 6 70 50.99 VI11 1 7 10 19 51.58 1953: VI1 5 68 51 124 51.37 VI11 5 51 41 103 51.47 1955: VI11 3 20 25 51 51.55 IX 1 14 18 33 51.52 X 1 11 9 21 51.38 XI 2 23 24 51 51.51 1956: VI-VI1 9 10 _... 19 50.53 XI 33 22 55 51.40 1959: I1 19 107 21 147 51.01 Santa Maria Bay total ._..__..__.__.___..______57 471 297 17 842 ___51.33 Magdalena Bay .--..-..-...... -..--..--..--..-1949: I1 1 27 17 1 46 51.39 1951: VI 1 8 10 3 22 51.68 1952: VI1 4 38 9 __.. 52 51 .06 VI1 8 43 39 _.._ 90 51.34 VI11 5 41 13 .... 59 51.14 IX 32 35 1 68 51.54 1953: VI 1 23 20 2 46 51.50 1958: VI 7 105 27 ..._ 139 51.14 VI 18 82 20 .... 120 51.02 small VI 2 23 12 .___ 37 51.27 larger

Magdalena Bay total ...... 47 422 202 7 679 51.25

Clark total ...... 500 243 795 51.27 __- Pt. Marquis ...... 1952: X 38 18 65 51.28 ~- Gulf of California ...... 1953: VI11 71 25 111 51.09 1956: XI1 315 97 500 51.01 rpostlarvae)

Gulf of California total ...___._.....__..._..... 386 122 611 51.02

Clark total _.._.._-._.._.____..--.-..--.--.... 91 521 118 3 735 51.04 REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JCNE 1960 79 NUMBER OF VERTEBRAE 50.9 51.0 51.1 51.2 51.3 51.4 51.5 51.6 51.7 51.8 51.9 I I I I I I I I I UIUS SOUTff Of SA1 P€&!!O SAN PEDRO 0 * SAN DlEGO BO A 4- ENSENADA (1952-1957) 145

ENSENADA ( 1958 1 145

CEDROS Q SAN BENITO IS. 370 t. .

SEBASTI~NVISCAI~O BAY 400

BOUNDARY TO PT. EUGENIA 100-400 L

PT. EUGENIA ,TO CAPE SAN LAZAR0 400- 640

PT. EUGENIA TO ,. -- - 400 -600 0 0 PT. SAN JUANICO

MAGDALENA BAY 700 !-~ - .-

CAPE SAN LUCAS h 850-1000 GULF OF CALIFORNIA OA

50.9 51.0 51.1 51.2 51.3 51.4 51.5 51.6 51.7 51.8 51.9

FIGURE 2. Frequency distribution of individual sample means and the mean of each area. Each solid dot (present data) or open circle (Clark’s data) refers to an individual sample. The acute triangles represent the area means. sents a sample listed by Clark (1947). Table 1 sum- l1standard” for sardines spawned under what has marizes the present data in Fig. 2 by year and month. been thought to be ‘(normal” oceanographic condi- Clark’s 1947 data, with all age-groups combined, tions. For Southern California alone Clark’s average indicate that prior to 1942 the mean vertebral values for all age groups combined was 51.73. She reported changed abruptly in the vicinity of Pta. Eugenia, on only five samples with a combined mean value of Baja California : average values south of this point 51.62, sampled from 1938 to 1940 from the Boundary were markedly lower than .those to the northward. to Pta. Eugenia-excluding Sebastibn Viscaino Bay The higher values to the northward formed the basis samples, considered separately. for Clark’s statement that “The average number of Whereas Clark’s data from north of Pta. Eugenia vertebrae is approximately 51.7 for all sardines north comprised samples taken between 1921 and 1941, the of southern Lower California and about 51.2 for sar- present data stenis from samples taken between 1950 dines from southern Lower California. ” These nu- and 1959. A distiiict and significant change has taken merical values have since been considered to be place in the decade that separates the sampling efforts (Pig. 2). The plotted ~n(xiiivalues of all sainples cantly. Whereas MclIugli’s high value of 51.54 for from each arcii am1 the area il~~aiisfor the prestxlit postlarva(> spawned in the spring in southern Raja data sho\\r a rather marlied reduction iii arcal mean California falls well below the average figure of ap- values as wrll as a very grclat number of sample nieaii proximately 51.71 for the older fish under considera- values that fall well below the lowest rrportecl by tion froin these southern areas, it does indicate the Clark. It rill be noted that this r’educ4ioii pcmists possibility that eiiviroiinientally induced iiirristie dif- from Southern California to l’ta. Eugeiiia. Both sets ferences may result in abnormally high incan verte- of data are in excellrut agrcwnciit from l’ta. Ihgeiiia bral numbers in at least some samples from these sonthward and into the Gulf of California. areas, and, possibly, for all areas throughout the It is unfortiiiiate that very few snniples from tlie ratige. Of coiirs~the reverse situation may also occur region, ‘ ‘ Bouiidary to l’t. Engetiio, ’ ’ were available as is sho\v~iby two saniples takeii near the Xan neiiito to Clark, as compared with tlie much larger iininber Islands, aboiit 15 iiiilcs ,west of Ccdros Island. In from Enseiiada that are used in the present study. l!M, 172 0-yrar fish fnriiishrtl a iii~aiivertebral value Clark’s five sample meaiis from this region all fall of 51.40 atid in 195.5, 14:3 fish (estimatc.rl at nearly within the raiige of those from Ensenatla, though two years of age) providd a iiieaii of 51.85 vertebrae. they are distiiictly on the high side. Clark’s (lata from This is ari area tliat was qiiite “iiortherii” in the Sebastihi Viscaitio Bay aloiic correspond very closely data prcwiitetl by Clark, hiit, c~o~isiderablyless so in with those for tlir broad areal groupitlg from the the present (lata: the means are abont 51.64 and 51.50, lloundary to l’ta. Eugenia. The higher avcragcs ap- rcspvct ively. pear to have persisted further south in 1!)21-41 tliaii in Ruth ?vlagdaleiia atid Saiita Maria Ilays also show 1950-59. a tendency to prodnce some samples with relatively It, is of coiisitlerable interest that so niaiiy saniple high ni(Jaii vrrtcbral iimnbers. Ex~liidiiigthe two low mealis fro111 the northern region should fall well below ni~aiivalues, 50.99 and 51.01 (Fig. 2, Table l), raises those reported by Clark. For areas sonth of l’ta. Eu- thc Saiita Naria Thy av(mg(1to 51.44, a fignrc notably genia, exc11i(liiig those of Cape Sal1 li~(:tisatid the higher than the combined average for the Bay and (iulf of California, very few saniplcs iii the prcwnt oiie that \~ouldappear to fit th(x plottccl distribntioii hotly of (lata (olily 22 p(>i-cent)havr lo\wr vertebral of saniplc niraiis (Fig. 2). It is reasonable to assiime rneali values tliau do those from Elisenatla and South- that as Sairta Maria Bay is etitirvl?- open to the s~a ern California (Fig. 2). Clark’s (lata aloiie S~OW82 atiy niigratiiig northern stock may rcwlily enter. In perceiit of the samples from thwe soiithc~iiareas to coiitrast, t tie rrlativrly ciic.loscil Magdalena Bay is hay(: appreciably lower va11ic.s tha~clo thc samples presuniably less acc.essible to coastal niigrat>iou and froni Southern California reported 011 ill 1947. This may be expected to harbor a more iiidigenous popnla- differeiice of 61 percriit is strikiiig cviilc1iic.e of the tioii. The tiumber of saiiiples froiii each bay is hardly reduction in mean iiuiiibers of vertebrae betwemi the adeyuatr to resolvc thr qiiestion. Actiially the highest two time periods in the samples from Soiitherii Cali- iiiean valne rrcordccl for Xagtlalciia Ilay in each set fornia and northcrn Llaja C’aliforiiia. The populations, of (lata rscwcls any of thc iii(wis for Saiita Maria or stocks, of fish froin areas soiith of l’ta. Eugenia Ray. IIo\wver, thi. high valne of 51.68 listed in the may intermingle to at least a small t1cgrc.r with those presetit data is cl(~rivvtlfrom a sample of oiilj- 22 from the C4ulf of California : of all samples from specinic~iisaiid niay not bo of groat sigiiificaaiice. The south of 1%. Eugenia, excluding those froill the Gnlf, iiext Iiigliest value, .>1..54, is tlwivetl from 68 fish. It abont 23 percent have vertcbral nieaii valiies similar tliiis appears 1)robablr that higher Tallies may occiir to those reported from the Gulf (Fig. 2). in ttir. Ihy. Clark’s l!N5 data iiic.1iitlc.s a sample of 127 In contrast to the c.xprcted lowr m(wi values for I-j-ear fish \zhirli fiirtiishetl the mow sigiiificwit m(~i the areas south of Pta. Eugenia are the sc~rrtilrrla- valiir of 51.66!). Thus it is apparc.nt tliat either migra- tively high values. The value of 51.5!15 for 58 fish tions froiii the iiorth niay e1itc.r these bays or that (0 age-group), listed by Clark for the. ‘I I’t. Eugeiiio local sp\vtiing coiiclitions may occasioiially be suc~l-i to Pt. Saii Juaiiico” area, and the valiie of 51.66!) that highw avmage vertebral iiiiiiibrrs inay rt+xlt. It for 12i fish (I ape-gronp) from filag(1aleiia Ray, is also appareiit that the same sitnation pertailis to are generally considered as typically “ iiorthcrn ” or tliosc sanrplrs liavitig lower vrtebral iii(~a11valiies “high ” means. The present data inclniles one sample wmparablr to those 1istc.d for the Gnlf of California. of limited sigiiificaiice (comprising only 22 fish) from I4’roiii the. stantlpoilit of migration these bays may the same area with a high mean of 51.68. The rrasoiis be cwiisitleretl to be more withiu the raiige of Giilf for these higher mean values are difficult to assess. stock thaii those, from mor(’ northern waters, particq- McHiigh (1950) offered a possible explanation of larly tliose fi-om iiortli of 1%. Engenia. these high values, which average 51.71. Postlarval As noted by Clark (1947) the Gulf of California sardines from southern Raja California, spawned in may well harbor a separate population of sardines spring and summer, were found to have sipiificaiitly that iiiteriiiiiigles little if at all with the northern different mean vertebral numbers of 51.54 and 51.21 populations, but that niay mix to a degree with stocks respectively (as McHuph did not count the urostyle, occiirriiig below I’ta. Engeiiia. The prrseiit data has his counts are increased by one to conform to those of strengthened Clark ’s coiiclusioii, for the two sets the present study). The vahies of 51.79 aiid 5l.li for of data arc’ remarkably similar, consideriiip both the spring- and summer-spawned fish in samples from time interval between the two sets of data and northern Baja California differ even inore signifi- the total niiiiiber of years iiivolved. Wliereas Gulf samples coiisist of 0 fish or postlarvae, except for the iiito Southeiw California contributed to the lowering 31-fish sample of adults rcported by Clark, affording of the awrage iiuniber of veitebrae within the area in a mean of 50.968, all samples have provided ineaii 1951, though they are suggestive. Oiily one sample of value? of about 51.00. 3i fish, from Saii Pedro, was available for study. This Ycar CLuss Ihta am1 the “Jt’urrri ” I’ear-s: The oc- sample afforded a nieaii of 51.4!1 (Table l), quite currence of abnormally high ocean teiiiperatures in characteristic of the receiit period in the Southern 1957 and 1958, aiid the observed reduction in average California fishery, whereas the average for the earlier iiunibers of vertebrae in saiiiples from iiortherri Baja period reported on by Clark was 51 73. California and Southern California, requires that That these 1954 fish, aiid those of subseqiwiit aid some consideratioil be given to year clasqes coriiprisiiig earlier saniples to 1950, were probably not spawned in the present data. 11s a major portion of the samples the area of capture tluriiig any wariii period is evi- were takeii ill 195748, particularly iii these two areas, denced by the follov iiig stateriieiit by Reid, liodeii, the age data are considered for these areas oiily. and Wyllie (1958), “the period froiii 3949 to 1956 is Daugherty and Wolf (1960) show that, for the tlistingnished from the previous 15 or 20 years by 1957-58 season in Southern Califoriiia, thc 1957 year substantially colder waters iii the first few nioiiths of class coiitributed only 1.3 percent to the conimercial the year. ” catch. During the same period of the iiorthern I3aja The foregoing cvideiiccl of the pcbriod of 1949-1956 California season (iiiclucliiig Eiisenada and Cedros beiiig coldcr than tlic prrvious 15 or 20 years adds Island laiicliiips) these authors shon~that the 1957 fnrther interest to the c.oniparison of the two sets of year classes coiitributed only 1.4 percent. The 1938-.59 data 011 vertebral numbers. If niost specimens in the catvh data for Souther-ii California showed that 1957 presriit study are prcsumecl to have beon spawiied in fish comprised oiily 10.2 percent of the mtch ant1 waters colder thaii those in which Clark’s must have in Baja California oiily 11.6 percriit (Kolf and beeii (all spawrietl prior to 1941), it could also be I)aughert?-, 1961). presnnied that the avrrage ~iiiiiibers of vertebrae Thus it is clear that the 19.57 and 1938 year classe5, would be higher than obscwed. This obviously is not presnniably hatched in warnier water, were a minor the casc iii those sampks studied fmiii iiorthrrii Baja coiistitneiit of the catches and the vertebral count California and Southern California ; thus by every samples. Therefore, the recent ‘ warm w ater ’ ’ can be test the cahaiige iii vertebral niimbers is opposite that almost completely cliscouiited as contributing to the suggested by the eiivironiiieiital changes. There must, low vertebral numbers off Southern Califoriiia and then, hare been major changes in the stocks on the norther11 Baja California dnriiig the 1957-38 and fishiiig grounds. 1958-59 fishing seasons. One qnestioii posed by the present body of data is, what happened to the “iiortherii ” type of sardine f DISCUSSION Temperature data indicate that sufficicwtly cold wa- If the effect of the warm years may be cliscouiited ter was available to have produced high vertebral as being a major cause for the observed reduction in iiumbers. The number of samples studied would ap- average vertebral numbers iii the area of iiortherii pear to aford a11 adequate representation of the fish Baja California aiid Southern California, other hy- available to thr sanipling techniques. It may be that potheses may be erected : (1) The eiiviroiimeiit had the northern types were “somewhere else” arid iiot begun to change as early as 1930 and the change was available to the fisheniia~ior the sampliiip techniques. reflected in the reduced average number of vertebrae It may also be that the influx of sontherii fish in 1954 (but there is 110 oceanographic evicleiice of such a coiitiiiued and that interbreeding with the northern change) ; (2) hlicro-eiivironmeiital changes afYected types has rcsulted in lower ayerage numbers of verte- the vertebral numbers and morphometry of the in- brae, or that the northern types were still present dividual spam nings that were later sampled (such an but in coiisiderably reduced numbers. event is, of course, always possible ; however, there arc’ 111 review, south of Pta. Eugeiiia there was no great odds against the chance that such spawiiiiigs aiitl chaiige iii vertebral nnmbers between the earlier and samples make up the bulk of the preseiit data) ; (3) later period, all counts being low. North of I’ta. Eu- Stocks of fish migrated iiorthwartl from the southern genia, vertebral counts during the recent decade are areas, which are rharacterizecl by lower averag(’e iium lower thaii the earlier period, the average being in- bers of vertebrae, accompanied by a reduction iii the termediate between the stable southern average and nuiiibers of iiortherii-type fish. This is the most attrac- the earlier northern average. Further, the direction tive hypothesis. of the change, Le., a drcrease, has been contrary to Ahlstrorii (1939) has reported that such a iiorth- that expected to have rewilted from the cooler envi- ward iiiooenieiit did occur iii early 1934. Associated ronment chiring the recent period. Thus, there is good with this iiorthward nioreiiieiit was ail appreciable cvicleiice that there has been a change in the “racial” rise in water temperatures, particularly in the early coiiiposition of the stocks taken by the fishery in spring. Clothier adGreeiihood (1936) reported that Southern California and northern Raja California. 67,258 tons of sardines were landed during the 1954- The precise iiatiire of thc change is not so clear. 55 season in contrast to the slightly more than 3,000 There has been a, severe decline in the stocks available tons in 19X-54 aiid some 4,500 tons in 1952-33. to those fisheries compared to the early period (Clark Viifortuiiately, the present data are iiot adequate to and Marr, 1935). This alniost rules out the possibility test whether or iiot this observed niigratioii of sardines that the lowering of the vertebral numbers was caused only by an influx of southern fish. Rather, the tory, Terminal Island, and to Franklin G. Alverson, change in vertebral numbers must have been associ- of the Inter-American Tropical Tuiia Commission, for ated with a reductiou of the population of the north- making available samples respectively from the com- ern, high vertebral number, “race, ” possibly, but not mercial landings in Southern California and from the necessarily, acconipaiiied by additional influx from bait samples takeii by the tuna fleet in Mexican waters. the south. Especial thanks are clue to A. M. Vroornan, J. S. McQrepor, and R. S. Wolf of the 1,a Jolla liaboratory SUMMARY of the C.S. Bureau of Coinmerc*ial Fisheries for iiiakiiig available the saiiiples froin the 1957-58 Ense- 1. The results of the suinniarized three widely dif- nada coinniercial fishery ant1 to E. IT. Ahlstrom, Carl ferentiated earlier approaches to the problem of het- 1,. IInbbs, and (iarth I. Murphy for valuable criticism cx-ogeneitg of sardine stocks throughout the range in- of the niaiiiiscript. dirate the presence of more than one population, or stock, of sardines inhabiting the total range. REFERENCES 2. The present study demonstrates that a marked *\hlstroiii, E. H., l!):!). 1)istril)ntioii nntl ;ihuntl:ince of eggs of reduction in average numbers of vertebrae, for sar- the 1’:icific. s;irtlinc~,1!).72-19>6. l..N. lkpt. of Int., Fish alrd (lilies sampled in northern Baja California and ll’ildl. S’o.i.., E’ish Biill. So. 165, 60 : 185-213. Southern California, has occurred throughout the Chrk, 14’. S., I!):%, Yariations in the nnmher of vertel)r;ie of period froin about 1950 through 1958. This reduction the s:irdinr, Su~.dinop.sccc.errclecr ((:irartl). Coprin, (3) : 147- results in a much lower average value than that re- 150. ported by Clark (1947), the respective yalues being Chrk, I“. S., 1!)47. A\ii:ilysis of pol)nl:itions of tlie P:ic.ific sap- 51.50 and 51.73. (line 011 the basis of verteI)r:il counts. f’cilifornia Dioision of 3. The plot of average numbers of vertebrae for I’iah rind Urtine, Fish. 11ii11. Yo. Ki: 26. each sample within a given area, and the area aver- Vl:irk, I<’. S., atid J. F. .J:iiisse~i, Jr., I!K. JIovernents and ages (Fig. a), indicates that inany of the samples ~il~n1111:111ceof the sardine ns n1e:isarrtl by tag returns. Cnli- taken in northern Baja California aid Southern Cali- fornici 1)irisio)r of Fish nnd Girtttr, Fish Ilull. 3’0.61 : 7-42. fornia, in the period from 1950 to 1959, have siiffi- Clothier, C. R., and E. C. Gre.c~iihoot1,1956. Jack mackcrel nntl s:irtlinr yield per nre~from C‘alifornia waters, 11146-47 cirntly 101%values to have originated in the maters off through 1!).74-& In the marine fish c:itch of Cnlifornin for Central or Southern California. the years 1!)5:2 nntl 1954. fy(t?ife>rnia11 4. Year-class data for a major portion of these fish (:(tine, E’ish Hicll. KO. 102 : 99. with low values indicate that only a small percentage I):inglierty, A. IC., and It. S. Wolf, 1960. Age ant1 length com- of the total catch could have brei1 spawned in the area pnsitinn of the s;irtline c;itcli off the L’acific coast of the 1.iiitc.d St:itr.s :inti Mexico in l!).ji-.78. Califoritin. Fish and of capture during the abnormally high ocean temper- Grinir. 46 (2): 18!)-104. atures that begaii in 1957. Felin, 11’. I.;,, l!lX. I’opalntion hi~tc~rogeneityof tlie I’ncific pil- 5. For the areas of southern Raja California (Pta. ch;irtl. L.S. Fish uiid T17ildl. Nrrr., Fish IZnll. A-o. 86 : 201- Engenia into the Gulf of California) the present data -_.‘>.,7, is in excellcnt agreement with that presented by Hart. .J. I,., l!M. Stntisticnl studic~son the Iiritish Colninbia Clark (1947). The two sanipliiig efforts corer a period pilcliartl : Vertebra counts. T’vaiis. Roy. Koc. (‘tinadel, Sect. V : of time from 1926 to 1958 for this broad area. This 79-8.7. fact indicates that these arras may harbor indige- Hart, .J. I,., 1943. ‘Rigging rsperinients on I4ritish (’oliinil~ia nous populations in that no apparent change in the pilcliartls. Joitrn. Fish. Res. ljd., Conrida, 6(%) : 104-1S2. average vertebral numbers has occurred in the 32- Hiiblhs, C. I,.,1925. 1t:ici:iI :ind sr:ison;il vari:ttion in the l’acific year period. htwiiix, (‘;iliforni:i sardine, ;mtl Chliforni:~ ;~nrIi~)vy.Cnlif. I’ish aird (;nnie f‘ornn~.,Fish Ritll. KO.8 : 23. 6. It is concluded that the sardines caught in north- Marine 1tese:irch Committee, 1!)55. Progress Report, Cnlifornin. ern Baja California between the years 1950 and 1959 (’oopemtire Ocecinic Fisheries Jni%estigations,1 July, 1953 to were comprised of a different “racial ” mixture than 31 .Ilorch, 19Z : 52. those taken prior to 1941, because the recent reduction B1:irine Resrnrc.h (’ominittee, IN;(). Ireports of the Cnlifornia in vertebral numbers is contrary to the phenotypic (’oopei-ntii.r Ocetinic E’isheriss Ini~estigntion.~,Vol. VII, 1 increases usually associated with cooler water. Con- Jannnry, 1.058 to 30 Jnne, 19,JS : 217. currently, abundance has declined, so the change in JlcFugh, .l. L., 19.70. V:iri;itions and populntions in the clu- yertebral number is most easily explained by postulat- peoid fishes of the Sorth I’acific. A Doctoral dissertation on fire in the Scripps Institution of Oceanogrnphy, La doiln, ing reduction in a northern “racp,” possibly accom- California :116. panied by an influx of additional southern fish. Reid. .J. L.,Jr., G. Rotlen, and .J. Wyllie, l!Xi8. Studies of the Va1iforni:i Current System. J’royrsss Report, California Co- operative Oceanic Fisheries Ithcestigations, 1 July, 1956 to 1 ACKNOWLEDGMENTS cJaiiuci~~y,1958 57. Knmerous members of the various agencies compris- Thompson, W.F., 1926. The California sardine and the study ing the California Cooperative Oceanic Fisheries In- of the av:lil:tl)le supply. California Fish and Game Comm., restigatioiis ( CalCOFI) have coiitributed time and Fish. Bull. KO.11 : 8-17. Wolf, R. S., and A. IC. 1)aagherty. In press. Age and length effort toward collecting and preserving the samples cwmposition of the sardine catch off the Pacific coast of the used in this study. Particular thanks are due Anita B. 1-nited States and nIesico in 1958-59. Catifornia Fish and Daugherty, of the California State Fisheries Labora- Gn in e. THE WATER MASSES OF SEBASTIAN VlZCAlNO BAY

JOHN G. WYLLIE’

INTRODUCTION This report is a descriptive analysis of the water masses and certain processes, which combine to pro- ~UCPthe major changes in the physical environment observed throughout the year in Sebastian Vizcaino Hay. No effort is made to explain the forces involved or mechanics of the changes described. A discussion of the currents of the bay has been avoided since it is felt sufficient measurements for such a discussion have not yet been obtained.

GENERAL AREAL DESCRIPTION Sebastian Vizcaino lhy is located 350 miles (all distances are expressed in nautical iiiiles) south- southwest of Sail Diego on the Pacific Coast of Baja California (Fig. 1).The bay has the shape of a fish- hook complctely exposed to the sea toward the northwest. The distance from the iiorthern end of Cedros Is. due north to l’unta Canoas on the mainland measures 63 niiles; the distalice from Puiita Canoas to the southernmost shorc of the bay is about 115 miles. The iiarrowest diinrnsion of the bay measures 48 miles from the northwii ciid of Cedros Is. to Puiita Maria. The inset in Figurci 1 show Moliterey Bay of central Califorilia for a size. comparison. Two ehannels, separated by Natividad Is., opeti to the sea in the south- wwtmii portion of tlic. bay : thcy are Kellett Channel, which is 8 niiles widv ant1 10 to 4.5 iiietrrs dwp, and T>cwy,- Chaiiiicl, which is 1 iiiiIc’4 widr aiitl 25 to 30 iiicitc.rs tleep. ‘l‘hrw very \lliillo~v lagoOtis are foulid along tho soi~tJi(vtst~r~~honiitlary : S(~aiiii1io11,I3lack IViLrrior iilitl ?rIiinttc~la.‘l’h(~ SWII 12viiitos Islallds lie ~ii)i)ro~iiiiiit(~l~1.5 iiiil(s wrst of‘ (‘c~lr.04 Is. Ranger I~IIIIC,nk)orit 100 itiet(>rs(Ivei), lies 1.5 niilw uorthwest of (’c~lro\Is. ‘I’lrc~rc~is littlo ritiIlfall iii this arm arid 110 riwrs (wipty iiito the bay.

DESCRIPTION OF WATER MASSES

FIGURE 1. Geography of Sebartian Vizcaino Bay.

( SS ) 84 CALIFORSIA COOPERATI\'E O('EA\X1(' FISIIERIES ISVESTIGATIONS

I15O II I I .. e e *. e*. .. UPWELLING

29 '9O

.e. CALIFORNIA CURRENT \ e (subsurface) . . ..e .e ... e I NTERM EDlATE .e. e...... e* e... e .. .e.*.

28 !8O

FIGURE 2. Water masses of Sebastian Vizcaino Boy. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 85

FIGURE 3. Section 1, Cruise 5208. Vertical section from the surface to 110 meters showing distribution of salinity (isohaline lines), thermosteric anomaly (lines of equal A’ or 8~)with oxygen values (ml/L) entered. in this region, results in the coldest water because the 3. Central Bay water is also the result of a local vertical temperature gradient is normally negative. condition and cannot be described as having definite Whether or not this water will have a high or low characteristic values. There appears to be a permanent salinity value depends upon what water is being up- or semi-permanent clockwise rotating eddy in the bay. welled. In the Punta Canoas area higher-salinity Its position and size vary considerably. It is definable water is most frequently upwelled. because it is characteristically delineated from the 86

t , , ,c 145.

FIGURE 4. Cruise 5208. Block Diagram showing general char acterirtics and locations of water masses in August 1952 (5208). surrounding water by a very sharp thermocline and A general descriptive analysis of cwtain spwial oftcntinies by “ fronts” (Croiiiwell aiid lharacteristics forins thpre. about 13°C n(xe observctl in the core of the tongue 5. 1Tpn.elletl water from the I’nnta Eugenia area at a depth of 50 to .53 Iiiclters iinmediately west of the (wi be described in the sanie nianiier as the water bay riitrai~cc(Fig. 3). ,I 1)atc.h of similar uater is from the I’iiiita Canoas region. The riiairi cliffhence is foniitl aloiig the soutlierii short. at a depth of 20 ineters in the iniportaiiw of the two areas in relation to the or inore It niay be a continuation of the present in- bay. l’iuita Etigenia water is probably the least iin- I wsioii or thc reniiiants of an earlier invasion. portant of any of the \niter-mass sources described. I’iinta Calioas upwelled Tvater has a temperature Some of this xater may eiiter through the southwest- raiigr of 12 to 14°C aiid a salinity range from 33.60 erii chaiinels aiitl affect soine of the southern coast of to X3.6e5 I),’,,,, in the surface layers. It is found as a the bay, as cvidciiced by wrtaiii cold water algae cre\criit sliaprd ridge opening to tlie west, exteiidiiig foniitl there (Dawson 1932). south from Puiita Canoas, reaching to withiii 10 miles 6. Xorithcrii or(>iiliic siirfac~.water is characterized of Cedros Is. (Fig. 4 and 5). This water is found by high tt~nipcratiirc~saiid saliiiity. This water is found at the surface iii the iiortherii one-third of this dis- aloiig the coast to thr south of I’niita Eugenia tilid taiice only. It apparently upwells from depths of less iiifl iieiic t’s t h e bay 111 ai 111y (111r i I 1g t hc. late siiin ni t’r a ntl than 100 meters. A r~lativelystrong “front” is fouiicl fall lnollths. hcfwwii the California Cnrrcnt water and this up- 5. Vatrr froiii tlir clwp coastal salinity iiiaxiiiiuiii welled water. The entrance of upwelled water into is a subsirrfaw iiiass ( rrswdly foiincl a.t 200-300 iric,ter the bay circulation probably accounts for the depres- tlepth) w1iic.h is cliarac*terizc~dby rc~lativclyhigh tc.111- sioii of the bay temperatures as compared to the peratrirt~saiitl stiliiiitic~s. ti verj- low osygeii content iic6gliboring snrface oceaiiic waters (Fig. 4). aiitl a rrlatiwly high phosphattl coiitc.iit. Jt is ;Ictnwlly (‘eiitral bay water is restricted to a proinineiit a 11 (1st (’1 1 sion of $2 (1 r iat o r i ii I l’tw ific IV a t e r 1 ior t Ii ~va rd clocokn ise rotatiug eddy occupying the main part of aloiig the coast. It cJiit(br’sthci bay cirviilatioii froni thr bay east of Cedros Island (See the isobaths and below by iiim-iiig 111) o~~r‘ill(> c.oiitiiieiita1 slo1)c iIIl(I isotlirrnirs of the A’z 300 surface on Figure 6). The sh(.lf. A’= 300 surface has been chosen as a reference sur- In atldition to tlie tylws of \viit(.r jtist descril~etl faw brcanse it coiiicides closely with the core of the other water is soiiietiiiirs foiiiicl which tloes not fit ;illy Ion saliiiit> California current water in this region of these classificatioiis. ‘l’his is. of roiirs~,to be es- airtl irsiially delineates the bottom of the central bay pwtecl whrii two or nior(’ of tlrc. above witcr t:-pes e.tltly. iI sharp thermocline separates this water from mix. We will call this intrriiicvliate watcr. the cdol(ler leis saline water below. The central bay REPORTS POLTJIE I’III, 1 JULY 1939 TO 30 JUNE 1960 87

115’ 114’ I I I I

FIGURE 5. Cruise 5208. Surface temperature and salinity distribution (Isotherms in ‘C, isohalines in “/oo). water is well mixed froin the surface to the tlreriiio- Soiitlimw surface oceanic water is found southwest cline, which has a inaxinium depth of 60 to 70 nietcrs. of thc hay bcyoiid the I’uiita Engenia upwelled water The surface teniperatures range from 18 to 20°C aiid with temperatures, at the surface, from 19 to 21°C or the salinities from 33.50 to 33.60 o/oo. 11 stroiig grc.atc,r (Fig. 4). None of this water seems to have “front” is foniid along the northwe\tern boiiiidary bccii in the bay at this time. of the eddy. The southern eiid of the cold Punta Iiiteriiiediate water was found as a yery shallow Canoas upwelled water is fouiid adjacent to the cell- 1ayc.r (less thaii 20 meters thick) imiiiediately west tral water and helps to intensify the “front” (Fig. of Punto Coiio (Fig. 4 and 5). It exteiicls along the 5). Another weaker “front” can be traced between ea\tcrn aiicl southrasterii shoreline of the bay where the central bay and the iiiterniediate water fouiid to it blriitls with the central water. Teiiiperatures range the east aiid northeast. Along the sontheaitern edge froni 1.5 to 15°C and salinities are near 33.50 O/()o. of the eddy the iiitermetliate and central water be- This water appears to be a mixture of California Car- conies homogeneous aiid the “ front ” disappears. rent, I’iiiita Canoas upwelled aiid central bay waters. Lagoon water is found immediately offshore of the The thermocliiie separating it from the cooler water lagooiis along the southeastern edge of the bay. We bclow eonies to the surface just southeast of Punta have little data but we do know the water is very Caiioas foriiiiiig another front. This front betweeii the warm and saline. upwelled and intermediate water becomes diffuse to tJpwelled water from the Puiita Eugeiiia area had the south. surface temperatures from 16 to 18°C and salinities Underlying most of the water in the bay is water of 33.50 o/oo or greater (Fig. 4). Apparently some of from the deep coastal salinity maximum. This watcr this water is carried into the bay on the iiicomiiig tide may slide up the coiitiiieiital shelf south and \vest of and gives evidence of being fouiid along the northcm Kcllett aiid I~Me)- Cliaiiiiels mil continue oyer tlic shore of I’uiita Eugenia (Dawson, 1932). sill between Cedros Islaiid and the San Beiiitos Is- 115‘ 114” I I

FIGURE 6. Cruise 5208. The depth of, and temperature distribution FIGURE 7. Cruise 5209. Surface temperoture and salinity distribution. on, the thermosteric anomaly surface 1’= 300. (Isobaths in meters, (Isotherms in “C, isoholines in ” ’g ....) isotherms in “C.) REPORTS TOLUJIE TIII, 1 JULY 1050 TO 30 JUXE 1960 89

clearly seen as a warmer, more saline zone on the A'= 300 surface. This water is about 1°C warmer than during August. The general meteorological conditions over the Pa- cific should have resulted iii a relatively weak flow of air froni the northwest. This could account for the absence of coa5tal upw>lliiig at this time. Rcfereiice has been iiiacle to surface oceanic water originating to tlie sonth aiid southwest as playing a part in the water-mass seyuciice in the bay. Neither cruise in 1952 offered any definite evidence of this. We were fortunate, however, in obtaining direct evidence during a special bay cruise in September, 1953,

FIGURE 8. Cruise 5309. The depth of, and temperature distribution on, the thermosteric anomaly surface I' = 300. (Isobaths in meters, isotherms in "C.)

115" II n I I

CRUISE 5209 ISOHALINES (TI ..... ISOBATHS (ml

I I I 115"

FIGURE 10. Cruise 5309. Surface temperature and salinity distribution. (Isotherms in OC, isohalines in '/oo.)

(CR 5309, see Fig. 10). At this time coiiditions were found to be similar to those in September, 1952, (CR 3209). KO upwelling was in evidence at either Purita Canoas or Puiita Eugenia. California Current water dominated the scene, being fouiid along the ciitire northeastern shore of the bay to a depth of 50 meters. It was also found beneath the central bay water as a layer 10 iiietrrs or more thick. It could be traced as a subsurface layer 20 to 30 meters thick seaward from the northern half of the bay and as a thin slid to the northern end of Cedros Island. The central bay water mas diluted with California Current water. It had uniform saliiiity values of

I I I 33.35 to 33.10 o/oo, aiid surface temperatures near 115' I 20°C. The eddy was displaced to the southern portion of tlie bay (Figure 10) as coiitrasted to September, FIGURE 9. Cruise 5309. The depth of, and salinity distribution on, 1!152, ~hcnit was fomid to the north (Figure 7). A the thermosteric anomaly surface A' = 300. (Isobaths in meters, isotherms in "C.) weak front caii be fouiicl along its iiortliern edge. 90 CAT,IFORXIA COOPERATIT'E OCEASI(' I'ISHERIISS ISTI~~STIGATIOSS

115" 11. As in the 19,i2 criiises, lagooil water was found along n I I I the southern shore line. Tcniperatures of at least 22°C CRUISE 5309 and salinities greater than 34.00 o/oo are found. -ISOTHERMS ('C) ...... ISOBATHS (ml The most interesting featur'c, howrver, is the south- %... ON A'=300 SURFACE ern surface oceanic water found inirriediately west 30...... of the channels and Cedros Island in the upper 30 to . '.. 50 meters. As can be clearly seen on Figures 11 and 12, this warm saliiie water extends into the bay around the northern end of Cedros Island. Invasion of the bay by water of this type and in this amount could certainly account for any sudden increase in the temperature and salinity ralues, especially in the surface layers.

...... 3o ......

.. i ......

FIGURE 11. Cruise 5309. The depth of, and temperature distribution on, the thermosteric anomaly surface 4' = 300. (Isobaths in meters, isotherms in "C.)

I I I L 115' II4

FIGURE 13. Cruise 5809. The depth of, and temperature distribution on, the thermosteric anomaly surface A' = 300. (Isobaths in meters, isotherms in "C.)

In Srptember, 1958, (CR 5809), a good c.s;tlul)l(i of the influx of deep coastal saliiiity iiiaxiiiililii 11 ;itw into the bay eirculation is found (h'igiir(>s I:{, 11 aiitl lj). This is of special interrst since it sho~\\tliis deeper water being injected into the upper (sli;illo\\ ) levels of the bay water aiitl therefore being ~iiit(l(> available for rapid inclusion in the siirface circulation. As i\ e\ idcnt froin the figures, this watw is relativvly ~va1'11iand saline with ail exceptiolially low osygc'll colltent. Otlier spec.ial criiiscs have been made into the bay t I I I biit the above mentioned cruises give cxamplcs and 115O I1 support tlir hypothesis that the sequencc of events of the phj-sical enviroiiincnt, both seasolially and yrar FIGURE 12. Cruise 5309. The depth of, and salinity distribution on, the thermosteric anomoly surface 4' = 300. (Isobaths in meters, iso- to >ear changes can, in part, be cxplaiiietl ant1 de- halines in "/oo.) scribed in terms of these ~atermasse\ and processes. REPORTS VOLUME VIII. 1 JULY 1959 TO 30 JUNE 1960 91 SEASONAL VARIATION I I

CRUISE 5809 In discussing seasonal patterns at any particular point, we wish to delineate between variations which are easily predictable, such as the effect of incoming radiation from the sun and sky, and those which are not. The less predictable events which cause variations throughout the year are the replacement of one water mass by another, pliis all the degrees of mixing which attend these changes. The changes we observe are a result of a combination of all these phenomena. In the upper layers each water mass undergoes its own seasonal changes which can vary to some extent from year to year. These year by year variations can change the characteristics of a water mass, but these changes are not usually so great that the water mass loses its identity. Therefore, even though water mass characteristics may vary, they can still be distinguished from other water masses. An examination of a monthly plot of surface temperatures for Station 120.30 in the bay, shows the coldest months to be March and April and the warmest, September and October (Fig. 16). This follows closely the 10 year averages for neighboring inshore oceanic stations not in the bay. These stations also I I I 115- I show March-April as the coldest period and Septem- ber-October, the warmest. The bay tends to be 1" or FIGURE 14. Cruise 5809. The depth of, salinity distribution and oxy 2°C cooler than thr oceanic water outside the bay gen values on the thermosteric anomaly surface A' = 300. (Isobaths in meters, isohalines in "/m, oxygen values in ml/L.) most of the year. This is probably due to its proximity

Station 113.35 I15,35 117.35 118535 11933 120.30 I I I 0 m- -

20- -

40- -

60- -

80- -

IOO- -

120- -

140- -

160- -

180- -

200-

FIGURE 15. Section 2, Cruise 5809. Vertical section from the surface to.210 meters showing distribution of salinity (isohaline lines), thermosteric anomaly (lines of equal A' or ST) with oxygen values (ml/L) entered. 92 CALIFORSIA COOPERATIVE OCEANIC FISHERIES ISVESTIGATIONS

1959 ' IF!-34.30%, 22- -34.20

21- -34, IO -34.00 ,, -33.90 -33.80 17- d5 (-33.70 16- -33.60 :d' \, p 15- -33.50

14- -33.40

1311 IIIII lIIlIlII1Il I r33.30

FIGURE 16. Plot of surface temperature and salinity values for each month observed from May 1953 to February 1959 for station 120.30. to the coastal upwelling area near Punta Canoas. The lar time. Considering the lower points on this graph annual temperature range for the surface layers in to be relatively cold, the upper points relatively warm, the bay is normally 5" to 7°C. the points farthest left as having relatively low sal- A plot of the salinities for each month does not inity and the ones to the right relatively high salin- show any such simple pattern on the average (Fig. ity, we can regroup the data into a table showing the 16). If we keep in mind the relative temperature relative values, as defined above, from month to and salinity conditions at any particular location and month (Fig. 18). This table also includes results within a period of time of a few months, at most, so from station 120.30. A pattern emerges which can as to eliminate from the discussion longer treads, the be explained in terms of the water masses and pro- presence of a more or less predictable cycle is found. cesses defined as available for the bay. In general, the bay waters are relatively cool from January through June. At the same time the salinity can be either high or low. The water is cold at this time because of two factors. The first is the predictable normal temperature pattern for all waters of similar latitude due to the incoming radiation. The second factor is the presence of coastal upwelling which tends to keep the water cold. Upwelling can and does occur at any time of the year but seem to be most persistent during this period. The salinity variation can be explained by the presence or absence of upwelling 01' California Current water, or both. With upwelling in progress, a source of higher salinity water is available. Although all upwelled water is not necessarily of high salinity, it usually appears to be in this area. With California Current water present, a readily available source of low-salinity water is found. Therefore these are the dominating factors controlling conditions in the bay from .January through June. The months of July and August tend to be somewhat warmer with low-salinity values continuing. De- tailed weather data throughout the year would clarify the picture, but it is proposed that the upwelling FIGURE 17. Monthly values of the surface temperature and salinity at process usually weakens at this time, owing to lighter station 119.33 plotted on a special chart used in the processing of winds. This, in addition to the normal increasing solar hydrographic data (Klein, MS). radiation, would allow more rapid warming of the mixed layer. The main source of water available to Figure 17 gives some examples of the surface tem- the bay wonld most likely be low-salinity California perature and salinity values plotted on a special Current water. This would account for the relatively chart developed by Klein (Klein, MS). Each point warm lo~v-salinitywater usually found at this time. represents the observed value at station 119.33 for a During the months of September through December given month. This station was chosen for the illus- the bay waters tend to be warmest and most saline, tration since it is centrally located and is usually un- although the warm water is normally in the process der the influence of the central bay eddy which in of cooling later during this period, especially during turn seems to reflect the characteristics of the water December. The higher salinities are most likely the masses which are dominating the bay at any particu- result of advection of water from the coastal regions 93

Re la t ively AMJJASON FMAMJJASONDI LOW SALINITY HIGH SALINITY LOW TEMPERATURE HIGH TEMPERATURE mAAAAA FIGURE 18. Tabulation showing the relative seasonal values obtained from Fig 17. The data are repeated through a two-year period to show the pattern more clearly. south of the bay. ,111 (>xaiiipleof this type of iiira4oii ACKNOWLEDGMENT was desc'ribed earlier in this report (Septeiiiber 1933, The antlior is inclrbted to Rlr. 11. S. Shirley for the CB 3309). preparation of all the fig.ures. This n orlr was supported The cycle begins again with the oiiset of more wind by the &Iariiie Life Eescarch Progiwin of the Scripps collseqaellt upwelling ill late or early Tiistitiition of Oceaiiographp, a part of the California alorlf with cooler weather. This se- Cooperative Oceanic Fisheries Investigations. quewe is probably oversimplified and may be inter- REFERENCES rupted but the available data indicate the presence 1. Cronrwell, T., and J. 1,. Reid, Jr., 1956. A stud3 of oceanic of a basic pattern. If this is the case, a combination of fronts. Tellus, 8 (1) : 94-101. monitoring certain water niases aiid a good coverage 2. I>tiwsoii, E. T.,7982. Circulation within Bahia Tizcaino, Iiaja California, and its effects on marine 1 egetation. Jour. of wind and weather for the bay may be sufficient to Bot. 39 (7) : 423-432. folloW satisfactorily the challges in the physical en- 3. Klein, H. T. A new technique for processing physical ocean- viroiiineiit . ographic data. (MS.) C.ZLIFORSIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS ON NONSEASONAL TEMPERATURE AND SALINITY VARIATIONS ALONG THE WEST COAST OF THE UNITED STATES AND CANADA

GUNNAR I. RODEN'

ABSTRACT and bays and oiily few of them are well exposed. Blunts Reef represents a light-ship, anchored a few Power spectra of sea surface temperature and salin- miles off Cape Meiidocino, in the heart of the up- ity aiiomalies are investigated for the frequency range welling region. For meteorological records the nearest between zero and six cycles per year. It is found that meteorological station was chosen. In most cases the the power of these anomalies is concentrated at low tide gage ail meteorological stations are located within frequencies aiid that high frequencies contribute al- a radius of 20 km, but in a few instances stations fur- most nothing to it. The power spectra of temperature ther apart had to be combined. Where the nearest anomalies are similar for all 20 stations investigated meteorological station was more than 50 km away, here, but large regional differences exist in the salinity no comparisons between oceanographic and meteor- spectra. The coherence between sea and air tempera- ological variables were made. ture anomalies is moderate to good for well exposed stations, and there exists a direct relationship between DATA them. The coherence between salinity and precipita- Sea surface temperatures and salinities were ob- tion anomalies is good only for stations along the open tainod from the U.S. Coast and Geodetic Survey coast where there is little river discharge, the anom- (1956), the Fisheries Research Board of Canada (1958) alies being related inversely to each other. The eo- herence between salinity and river dischar,{Te anoma- and the Scripps Institution of Oceanography of the University of California (unpublished). At the Ca- lies is moderate to good for stations at the boundary nadian stations the monthly means were computed between river and oceanic water aiid poor elsewhere. from daily observations made oiie hour before the day- The coherence between surface temperature and salin- time high water. The individual temperature readings ity aiiomalies is moderate for light ship and island are estimated to be accurate within 0.2 C and the indi- stations and poor along the continent. The relation vidual salinities, determined by a modified Mohr titra- between them is mostly an inverse one, as is characteristic of upwelling regions. tion, are reported to have an accuracy of % 0.06°/oo. At the Uiiited States stations the observations are made at a random hour during each meek day, and in INTRODUCTION niany cases the monthly means are computed from less For many years careful records have been kept of thaii 20 daily values. The accuracy of individual tem- sea surface temperature and salinity at various sta- perature readings is estimated to be within 0.2 C tions along the west coast of the United States aiid except for stations in Southern Califoriiia where the Canada. If these records are plotted against time, sea- accuracy is somewhat greater, because better ther- sonal as well as nonseasonal variations are obtained. mometers were used. The accuracy there is estimated The seasonal variations have been studied by McEwen to be within 0.05 C. The surface salinities up to July (1912, 1919, 1929, 1938), and Pickard (1953) and the 1954 were obtained from density observations by hy- nonseasoiial variations by Hubbs (1948), Reid (1960) grometer at all stations. The accuracy of a single and Rodeii (1958). The particular relation between hygrometer reading is not very great, and must very temperatures and tides was investigated by Herlin- with the individual skill of the observer. In some cases veaux (195'7), and between salinity and river dis- the salinities so determined are probably accurate only charge by Tully (1952). within 0.3°/00. Since July 1954 the paliiiities at La In this paper the regular seasonal variation is elimi- Jolla, Balboa, Hueneme, Pacific Grove, SE Farallon nated by taking differences between the monthly mean Island, and Blunts Reef Lightship are determined by and the long term mean of the same month. The re- a Rnudseii titration, the accuracy of which is about sulting anomalies are then used to compute the aver- % 0.050/(),,. age power contained in each record and also the power Air Temperature and precipitation records were ob- spectral density, coherence and phase as defined below. tained from the U.S. Weather Bureau (18'72-1949, In performing the calculations it has been assumed 1950-1960, and 1916-1960) and the Canadian Depart- that the anomaly records are stationary, at least in the ment of Transport (personal communicatioii) . The wide sense, so that the functions derived depend oiily nioiithly arerage3 of air temperatures are coinpnted upon the sampling interval, and are independent of from four eqnidistant daily readings at 0000, 0600, any origin in time. 1200 and 1800 GRIT. The accuracy of an individual The location of tide gage and lighthouse stations is reading is estimated to be within 0.1 C. The accuracy shown on figure 1. Most of them are located in harbors of rainfall is estimated to bc 1% of the total or about

1 Contribhtlon from the Scripps Institution of Oceanography. 0.3 cin, whichever is the greater. 50N

40 -

30 -

20N-

I40 W 1: I20 110 IbO w

FIGURE 1. Location of stations where temperature and salinity observations ore made. 97

I BALBOA

'2 ++- ++- -2 0 -2 V

HUENEME

+2 +Z 0 0

-2 -2

PACIFIC GROVE

S E FARALLON ISLAND

+2 -2"'- -2

WN FRANCISCO Y +:E 0 -2 -2

BLUNTS REEF

+E[+--2

1916 1920 1925 1930 1935 1940 I IIIIIIII I1 IIIIII I/,II 11 [111,1I945 11,11/1/111(1,1950 1955 1959

FIGURE 2. Sea surface temperature anomalies at California stations. I CRESCENT CITY +2 0 -2

ASTORIA

NEAH BAY

+2 +2r 0 -2"1 -2 v, 2

KETCH I KAN

+2 +2

0 0

-2 -2

1916 1920 1925 19% 1935 1940 1945 1950 1955 1959 1111111111111111111111111111I11111111111111l

FIGURE 3. Sea surface temperature anomalies at Oregon and Washington stations ond at Ketchikan, Alaska. Ill11111.1 IfIIIIIIIIIIIIIIIIIIII I930 1935 1940 1945 1950 1955 1954

0 W.&+A.rn hm r\cu n -0

--2

AMPHITRITE POINT

+2 - 0-

PINE ISLAND

CAPE ST JAMES

+2 - n 1 - I%. 0 V" w .4

LANGARA ISLAND

1930 1935 1940 1945 1950 1955 1959 IIIIIIIII I 11 I I I II I 1 I I IIIIIII

FIGURE 4. Sea surface temperature anomalies at British Columbia stations. 100

111l1~1111~~111)(1111(1(111( 19% 1935 19401 1945 I950 1955 1959 ' ' ' SITKA -+2

0. n-n, 0 -2- - -2

JUNEAU

2 cno I Y u 0 _J

YAK UTAT

-+2 0 h.4 0 -2 -

1930 1935 1940 1945 111111111111111111~~~~~~,1~11950 1355 1959

FIGURE 5. Sea surface temperature anomalies at Alaskan stations. BALBOA

ta5 0 -0.5

HUENEME

+a5 0 0 - a5 -a5

1 PACIFIC GROVE z +a5 2 r0.5 3 0 00 I -a5 -0.5 w[La rn S E FARALLON ISLAND

-a5OF

BLUNTS REEF

t0.5 0 -a5

FIGURE 6. Salinity onornolies ot California stations. The records for Lo Jollo, Balboa, Hueneme and Pacific Grove ore questionable between 1944 and 1955. 102

ASTORl A

z0 NEAH BAY

t2 0 -z

KETCHIKAN +4 t4 t2 +2 0 0 -2 -e -4 -4 -6 -6

1916 1320 1925 1930 (935 1945 1111111"~,~~~~11 11111111111,~111111953 1955 1959

FIGURE 7. Salinity anomalies at Oregon and Washington stations and at Ketchikan, Alaska. The Seattle record is questionable. AMPHITRITE POINT 1

- 1.0

CAPE ST JAMES +0.5 +a5 0 0 - 0.5 - 0.5

LANGARA ISLAND

t0.5 +0.5

0 0 -0.5 -05 1

1930 1935 1940 1945 1950 1955 1959 1111111l11111111111111111111

FIGURE 8. Salinity anomalies at British Columbia stations. IIIIII I 111 1 I IlIl(lIIlII Ill 193 1935 194d I ' I945 I950 1955' 1959' S ITKA ti 0 -2

t6 +6 +4 +4 +2 +2 0 0 -2 -2 -4 -4 -6 -6 YAKUTAT

*2 +2 \ 0 0 -2 -2

1930 1935 1940 I 1111111111~1~~,11~,11111I945 1950 1955 1959

FIGURE 9. Salinity anomalies at Alaskan stations. REPORTS VOT2UJIE VIII, 1 JI'LT 1959 TO 30 JI-NE 1960 103

Runoff records wrre obtaiiied from the U.S. Geolog- TABLE 1 ical Survcly (1!)58) aiid the Snrface Water Branches Standard Deviations (u) and Extreme Ranges (r) of Monthly in Mrnlo l'ark, California, and Sacramento, Califor- Sea Surface Temperature (O), Salinity (S), Air Temperature (A) nia. They are considered to be accurate within 5% of and Precipitation (p) Anomalies ~ _. - ~ ~ ~ the iiidividual flow. - The sea surface tempwature records are shown in us UA UP rA rp figures 2 to 5. Although the details of the records vary, STATION* O/OO "C Clll "C cin

1: 1: ~-~ -- a few prelimiiiary results can be obtained from a ~- -_ visual iiispevtion of them ; (a) the niagnitudc of the La .Jolla-San Diego--- .. ... 0.9 0. I 1.1 2.8 7 24 anoriialies is about the same for all stations, (b) the Balboa-Los Anreles.. .~ ~ ~ ~ 1.0 0.1 1.3 ,$ . 1 7 24 appears Huenenre-Sta. Barbara-. ~ ~ 1.1 0.2 1.2 4.1 7 34 rnaxirriurri amplitude of the aiiornalies to be Pacific (;rorr-Sta. Crirz ~ 0.9 0.2 1.2 ii . 0 7 315 4C, (3) there are no obvious perioclicities diwernible, SE. I'arallon Id.-Pt. Kcyes. 0.9 0 . 3 2.!1 24

San Francisco ~ ~ ~.~.~. ~ . ~ 0.8 2.3 1 .0 3 . 0 7 33 (d) the aiioinalirs are similar over distances of several Blunts Rccf-Eure%a------0.!J 0.3 1.2 5.fi 8 41 hundred kilonicters and (e) for the majority of sta- Crescent City- ~ ~ ~ ~.~ ~~ .~.1.1 1.7 1.2 1 . .i 8 GI tions the meaii of the record dops not change appre- Astoria-North Head ....- ~ 1.0 1.1 1.2 !I Neaii Ray-Tatoos11 Id-. -. I.0 1 .o 12 7.8 IO 1.1 ciably, iiidicating stationary conditions. Seattle------~~~ ~~~ ~ 0.8 1.9 1.4 4.0 10 21i The surfacr salinity rccortls are show11 in fi, ; stations were all done by the same observer, and it is possible that the discontinuity is due to a change in +AA(T) = ; instrument, or to some other reason. Since the cause &A(.) = is not known, the records between 1944 and 1955 should be regarded with suspicion. Similarly the Seattle provided the records are stationary. 1Ier.c. < > tic- record suggests some radical change, and should be notes the time average over the entire WWI tl lixiigtli, questioned. aiid t is the time lag in months. The probability distribution of sea surface tempera- The power spectral densities and cross po~wrs1)w- ture and salinity anomalies is shown in figure 10. It is tral densities are obtained from a Fourier traiisforiii seen that the temperature anomalies are normally dis- of the correlation functions tributed, and that their probabilities vary very little from one station to another. The salinity anomalies (2) Eoe(W) = - &o(T) COS WTdT; are generally not normally distributed, and tend to 7r270 be negatively skew. This is particularly the case where there is considerable influence of runoff (San Fraii- 2" cisco). EAA(W) = -S+AA(T) COS WTdT; ?r The standard deviations and extreme ranges for 0 each of the stations are given in Table 1. For sea surface temperature anomalies there is no dependence EOA(O) = ;J&,(T)e-LW'dT1" upon latitude ; the anomalies appear, however, to be larger for stations situated in shallow bays and harbors --m than for those situated in straits and soulids, suggest- l'rac.tic.al ~iic~thodsfor evaluating the power spwtra ing that tidal mixing may be important in reducing are gircw by Miink, Suodgrass and Tucker (1951)). the amplitude of the temperature variation. For air The coherc.nc.e between 0 (t)and A (t)is defiiicd by temperature anomalies there is a slight dependence upon latitude, the anomalies bring somewhat larger at high than at low latitudes. The salinity and precipitation anomalies are dependent upon the particular environment of the station and do not suggest a and iiiiist lie bct\rwii the values zero and one. If relation to latitude. R = 1 the cohereuce is perfect, if 12 = 0 there is no 106 CALIFORSIA COOI’ERATITT OCE.iSI(’ I’ISHERIES ISVICSTIGATIOSS coherence between the two variables. The phase lag POWER SPECTRA OF TEMPERATURE of 8 (t)over A (t)is defined by AND SALINITY ANOMALIES The power spectra of sea surface temperature anomalies are shown in figure 11 for the frequency range between zero atid G cycles per year (c.p.y.). It is seen that most of the power is concentrated at low provided that for a positive numerator 0 4 T L 180 frequencies, mainly between 0 and 1.5 c.p.y. and that aud for a negative numerator 180 A T 360. If T = 0 the high frequency end of the spectrum contains rela- degrees, the two variables are related directly to each tively little power. This suggcsts that there is very other, if T = 180 degrees there exists an inverse rela- little contamination from freqiiencies beyoiid 6. c.p.y. tionship between them. in the spectra shown here. Most of the spectra indicate Similar relations can be established for any other an exponential decrease of the power with frequency, pair of variables. The length of the records used varies suggesting a Markov type process, for which predic- from 163 months at Crescent City to 444 months at tion is not very effective (Koden and Groves 1960). Ketchikan. The length of the record at each station, The spectra for Neah Bay, Washington, atid Seattle, n, the number of lags used, m, and the resulting de- Washington, show peaks near 1 c.p.y., which, when grees of freedon, v, defined by real, would indicate an annual periodicity. The peaks are, however, barely significant (upon applying the 2n confidence limits shown in Table 2), and the peri- (5) = - v m odicity is therefore not well established. The power spectra of surfave salinity anomalies are are given in Table 2. Also shown in this table are the shown in figures 12 and 13. Most of the power is con- 95% confidence limits for the power spectra, provided centrated between 0 and 2 c.p.y., and for the majority the variables are normally distributed. of stations the spectra do not contain any power for frequencies higher than about 4 c.p.y. There is a large variation from station to station of the power con- TABLE 2 tained in each spectrum ; the Canadian stations lo- Total length of Record (n), Number of Spectral Estimates Used cated on small islands and the Californian island (m), and Resulting Degrees of Freedom (u). Also Given Are stations contain a hundred times less power than those the 95 Percent Confidence limits for the Power Spectra (P) along the mainlaiid coast, excepting the Southern California stations. There are no significant peaks, or P periodicities, contained in spectra of the salinity STATION anomalies.

La Jolla-San Diego...... 504 24 42 0.67-1.62 COHERENCE AND PHASE BETWEEN SEA Balboa-Los Angeles...... 348 24 29 0.63-1.81

Hueneme-Sta. Barbara. - -.-. ~ -. . 420 24 35 0.66-1.69 SURFACE AND AIR TEMPERATURE Pacific Grove-Sta. Cruz...... 480 24 40 0.67-1.64 SE. Farallon Id.-Point Reyes>..-. 204 24 17 0.56-2.25 ANOMALIES San Francisco ...... 336 24 28 0.63-1.83 The coherence (Fig 14) between sea and air tem- Blunts Reef-Eureka...... 228 24 19 0.58-2.13 Crescent City...... 163 24 14 0.53-2.54 perature anomalies is moderate to good for stations Astoria-North Head...... 372 24 31 0.64-1.78 along the open coast and for island stations, and is Neah Bay-Tatoos11 Is1 ...... 276 84 23 0.60-1.97 Seattle ...... 408 24 34 0.65-1.71 relatively poor for stations located considerable dis- Race Rocks-Victoria..-...... 192 24 16 0.56-2.73 tances from the open coast along inland sounds. Thus Amphitrite Point ...... 276 24 23 0.60-1.19 Pine Island-Port Hardy ...... 174 24 15 0.55-2.40 for Blunts Reef, Neah Bay and Yakutat the coherence Cape St. James...... 192 24 16 0.56-2.31 varies around 0.7 whereas for Seattle and Juneau it is Langara Island-Masset...... 228 24 19 0.58-2.13 Ketchikan ...... 444 24 37 0.66-1.67 only 0.4 or less. The maximum values of coherence Sit~a...... 192 24 16 0.56-2.31 vary around 0.9, indicating that in the most favorable Juneau ...... 204 I 24 17 0.56-2.25 Yakutat ...... 216 24 18 0.57-2.19 cases about 80% of the anomalies can be related to each other. The phase (Fig. 15) between sea and air temperature anomalies varies around zero degrees and is at most stations independent of frequency, which It should be pointed out that the power spectral suggests a direct relation between sea and air tempera- densities depend upon the amount of averaging done ture anomalies, for all frequencies between 0 and 6 upon the original data (Munk 1960) ; if the monthly means were obtained from hourly readings, the power c.p.y. The considerable scatter at Juneau together with the low coherence indicates the absence of any spectrum will be different from one where the monthly means were obtained from daily readings. In case of significant relationship there. The ratio of the power of sea surface to air tem- a simple Markov process the power is decreased by an perature anomalies is shown in Table 3. For the Cali- increased amount of averaging. fornian stations and Yakutat the ratio is somewhat The computations of the power spectra, coherence larger at low than at high frequencies suggesting that and phase were performed on a IBM 709 electronic a certain time is required for the anomalies to respond calculator at the Western Data Processing Center in to each other. For Neah Bay and Langara Island there Los Angeles. is no simple dependence upon frequency. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 107

I I I 50 80 95 99 99.9

FIGURE 10. Probability distribution of sea surface temperature and salinity anomalies. 108 CALIFORNIA COOPERATIVE OCEASIC FISHERIES IKVESTIGATIONS

LA JOLLb SIN FRANCISCO SEATTLE KETCHIKAN 0.4 . a2 *. 0.6 ...... 0.4 00 ~ d ; : :cw 00 I ; : "4' LP. E E 0.2 E 0.8 .. BALBOA BLUNTS REEF 0 I 2 3 4 5 ~CPI SITKA

RACE ROCKS 0.2 ,0.6 ... '0 I 2 3 4 5 6CW E 0.2 CRESCENT CITY AMPHITRITE POINT 0 I 2 3 4 5 ~CPY 0bEee JUNEAU 0.4 0.2 0.20.4L ...... 0 ?...I...,. 0 I 2 3 4 5 6cW 0 I 2 3 4 5 6w1

0.4 42 42 ...... E 00 ; h ; : : ECP" 2 3 4 5 ~CPI

YAKUTAT

E,* , 1.0 PACIFIC GROVE ASTORIA CAPE ST JAMES OB a6 0.6 . 0.4 0.4 0.4 a2 0.2 0.2 ...I O012345 6 CPI '0 I 2 3 4 5 ~CPI 0 I 2 3 4 5 ~CPI E,, E,, FREQUENCY 1.2 1.0 SE FARALLON ISLAND LANGARA ISLAND

0.4 0.4 ...... 0.2

FREQUENCY FRFOULNCY

FIGURE 11. Power spectra of sea surface temperature anomalies. REPORTS VOLUME VIII, 1 JULY 1939 TO 30 JUNE 1960 109

- c L %s ss 0.04 r LA JOLLA 0.20 r RACE ROCKS 0.02'C0

Oo~~oo~nmm~~~n~~~

0 I 2 BALBOA 8 0.02 -0 0 Ess Oboa~booo~rnab~~~~~~~~0.30

0.04 000

0.02

000 I 2 3 4 5 6 CPY 0 I 2 3 4 5 6 CPY 0

11 0.02 -O 0.0 2 000 ~OOO~Ooo~ooo~nao~~aa~0

Ess SE FARALLON ISLAND CAPE ST. JAMES

8 0.04

0.02 0.0 2 00 000 2 3 4 5 6 CPY I 2 3 4 5 6 CPY

00 0 0.02 ooo oooo 0.02 - IO 0 1 I ~ooopooooo I ~o~o~~p*oopoo*p~oo~

FREQUENCY FREQUENCY

FIGURE 12. Power spectra of salinity anomalies. 110 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

4.0 SAN FRANCISCO Ess 4.0 r KETCHIKAN

ESS NEAH BAY 2.0 YAKUTAT 0 2.0 0 0; 0; G a04000600.~.e,~..,~oo~~ 0 I 2 3 4 5 CPY 6 0 1 2 3 4 5 6 CPY E 5 FREOUENCY

SEATTLE 6.08*oL

FIGURE 13. Power spectra of salinity anomalies. REPORTS VOLCDIE VIII, 1 JULY 1959 TO 30 JUNE 1960 111

R %A I.O?A LA JOLLA - SAN OIEGO 1.0 r BLUNTS REEF - EUREKA PINE ISLAND - PORT HARDY

I I I I I I O: 6 I L0' I I I I 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY

1.0% '.RF: ~ CANGARA ISLAND - MASSE1

I I I L I P I I I I I I 01 I I I I I 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY

0.2 0 00 0.21 I I , I , 00 0 I I I I I 0 O20 h 0 1 2 3 4 5 6cpy 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 6cPY R '#A ''Or 1.0 PACIFIC GROVE - SANTA CRUZ SEATTLE o.B 1 JUNEAU 0.8 0.8

0 I I I I I I I I ! I I 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY R R R 1.0 SAN FRANCISCO RACE ROCKS -VICTORIA YAKUTAT DO 0.8 t 0 0 00 eo*

0 OD 00 0.4 0.4 0.2 eo 0 I I I I 01 I I I I I 0 I 2 3 4 5 6CPY 0 I 2 3 4 5 6cp1 0 I 2 3 4 5 ~CPY FREQUENCY FREQUENCY FREQUENCY

FIGURE 14. Coherence between sea surface temperature and air temperature anomalies. 112 CALIFORSIA COOPERATIVE OCEASIC FISHERIES INVESTIGATIONS

"e4 BALBOA - LOS ANGELES "e4 ASTORIA -NORTH HEAD "e4 LANGARA ISLAND -MASSET v) 90 90 90 ...I...*..., a*.. ...eo. me. ..e-, 'DP...*..O*.a , "E 0 ' 0 .-.' 0 -90 I 2 3 4 ~*..~cPY I 2 3 4 5 6 CPY I 2 3 4 5 6 CPY

=e4 HUENEME - SANTA BARBARA "A NEAH BAY -TATOOSH ISLANO 7Te4 SITKA 901 90 90 ~ D .P"':''.e..., .*., ..;...*.~': I..& 0 -?-A: ... I 2 3 4 5 ~CPY I 2 3 4 5 ~CPY -90 -90

neA PACIFIC GROVE- SANTA CRUZ d 90

...I.. If. ' 5 0 *-.:.**'--.A -90 I 2 3 4 5 ~CPY

nB4 SAN FRANCISCO ne A RACE ROCKS-VICTORIA YAKUTAT 90 90 90 0...... '.. a. .?a* ,_^ * 7 *, 0 **-kr-*d--*Y ...... ,.. 0 ..'1"'9 , I 2 3 4 5 6CPY -9 I 2 3 4 5 GCPY I 2 3 4 5 ~CPY

FREQUENCY FREQUENCY FREQUENCY

FIGURE 15. Phase between sea surface temperature and air temperature anomalies. REPORTS VOLUME 1'111, 1 .JTLT 1939 TO 30 JUNE 1960 113

R R l.tr LA JOLLA - SAN OlEGO I.osp BLUNTS REEF - EUREKA 1.0 sp PINE ISLAND - PORT HARDY 0.8 0.8[

0.21

01 I I I 1 I 6 01 I I I I * 0 I 2 3 4 5 6cPY 0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 6crv R R R B.LBOA;LOS A"GEL:i LANGARA ISLAND- MASSET

0.6 0.6 eo0 0.4 oooe 00

0.2 ~ o*oo 0.2 00 0 , 0 I I I I I I :Ip0 I 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY R R R HUENEME - SANTA BARBARA ''orNEAH BAY - TATOOSH ISLAND KETCHIKAN a8 0.8laor 0.6 0.4

00 I I I I I I I I 1 I 0 I 2 3 4 5 6CPY 0 I 2 3 4 5 6cPY 0 I 2 3 4 5 ~CPY R R R 1.0 sp 1.0 i' PACIFIC GROVE - SANTA CRUZ SEATTLE SITKA 0.81

0.4

0.2 0L------I 2 3 4 5 ~CPY '0 I 2 3 4 5 ~CPY r h h b 6CPY "Z W R R SP 1.0 sp SE FARALLON ISLANO - POINT REYES RACE ROCKS -VICTORIA JUNEAU 0.8

0.61 o O 0 Qoeo 0.4 DO 0.21 "

I I I 101 I *I I 0' I I I 0 I 2 3 4 5 6cPY 0 ! 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY

RSP '.Or AMPHITRITE POINT -TATOOSH ISLAND YAKUTAT 0.8'*'r

Oa4I0.2 00 Q2 t 0' I I I t I 01 , I I e 0 I 2 3 4 5 6cP1 0 I' 2 3 4 5 ~CPY 0 I 2 3 4 5 ~CPY FREOUENCY FREQUENCY FREOUENCY

FIGURE 16. Coherence between salinity and precipitation anomalies. 114 CALIFORNIA COOPERATITE OCEANIC FISHERIES INVESTIGATIONS

%P TTP LA JOLLA- SAN DIEGO BLUNTS REEF -EUREKA PINE ISLAND - PORT HARDY IBO-...... I I 180 -90- I 2 3 4 5 ~CPY '2. ' * 3. * -90 . .. 0. . 6 CPY -0

%P sp i 0 '1 BALBOA -LOS ANGELES CRESCENT CITY LANGARA ISLAND -MASSET 90 90- .. .. I I I *I - ' 18o-f ...... 180-t -. .... -90- . - I 3 4'5 ~CPY -90- I..,: 3 4 *'5 ~CPY -0J -0-

HUENEME-SANTA BARBARA NEAH BAY -TATOOSH ISLAND

...... 6 CPY -90 ~CPY -90 3".i 6 CPY -90 3.. 4'. -0 -0 -0 YI Yw

PACIFIC GROVE - SANTA CRUZ 180 . . -90 6 CPY 6 CPY 6 CPY -0 -0 -0

SE FARALLON ISLAND-POINT REYES RACE ROCKS-VICTORIA JUNEAU 90- 90- 90- .

~ IB0- ' I .I '80. I I I I I ,.. , .L . :. ".- ...... 1€30-- ' . ..e.. .. w" TSP xo 0- SAN FRANCISCO AMPHITRITE POINT-TATOOSH ISLAND YAKUTAT 90- I I I I I I .L 180.. * ..L- I ..La I 180-t * . . I . ' IE0-,. . .: -*.I...... *.. -90- ...... -90- I 2 5' 4."6 ~CPY I 2 3 4 5 ~CPY I 2 3 4 5 .~CPY

FREOUENCY FREQUENCY FREOUENCY

FIGURE 17. Phase between salinity and precipitation anomalies. REPORTS VOLUhlE VIII, 1 JULY 1959 TO 30 JUNE 1960 115

RSR TTS R

Oo00 ago. 90 - 0 . 000 0 0°00 1 1 I ..I 0.6 O.0 180 ;;oofo-o: o..ooOOOOOaa- 6 CPY 0.4 -90 1 0.2 1 I I I I I I -0 -1 0 I 2 3 4 5 6 CPY

0- 0 90 -

10001 I 0 l-ool 0.4 I80 I 02 3 4 '4 6 CPY 000 0.2 -00 -90 00 0 0 00 00 -0 - 0

FREQUENCY FREQUENCY

FIGURE 18. Coherence and phase between salinity and river discharge anomalies. 116 CALIFORNIA COOPERATIVE OCEANIC FISHERIES IKVESTIGATIONS

R R LA JOLLA SAN FRANCISCO 0.8 e9 SEATTLE LANGARA ISLAND 0.61 . . . 0.4 0.20.41. e...... e ...... 0.2 * 01""" 01 1 " " ' 0 I 2 3 4 5 6sPI 0 I 2 3 4 5 ~CPI 0 I 2 3 4 5 6CPI R KETCHIKAN 0.8 es BALBOb RACE ROCKS 0.8 r"' 0.61 0.6 .... 0.4 0.4 -. . .

0 I 2 3 4 5 ~CPI 0 I 2 3 4 5 ~CPI HUENEME CRESCENT CITY AMPHITRITE.. .. POINT SITKA . .* .... 0.4c .. 0.4 ...... 0.2 ...... :.. 0 0 I 2 3 4 5 ~CPI 0 1 2 3 4 5 ~CPI R R PACIFIC GROVE ASTORIA PINE ISLAND 0.8 r"" JUNEAU ...... :* ; . ..:e 0.4 :::I0.4 ..I ...... 0.2 . . 0.2 . . . .. 0 II1A 0 0 I 2 3 4 5 6cw 0 I 2 3 4 5 ~CPY 0123 4 5 ~CPI 0 I 2 3 4 5 6CP1

Res SE. FARALLON ISLAND NEAH BAY 0.8 r ;.. ;*.-,..,.., .; 0.4 -. 0.4 ..*. e. 0.2 .. 0 0 I 2 3 4 5 6CPI 0 I 2 3 4 5 ~SPI FREDUEWC" FREQUENCY

FIGURE 19. Coherence between sea surface temperature and salinity anomalies. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 3 17

TI TI PACIFIC GROVE O$? . . ASTORIA PINE ISLAND oy@9 . . JUNEAU

-90 ... .. -0

SE FARALLON ISLAND NEAH BAY CAPE .ST JAMES YAKUTAT

I80 'I.*., .. .*. 97..;-..;*..;.**, ~CPY-90 6 CPI -90 4' 6 CIY -0 -0 FREOUENC" FREOENC" FRLOUENC" FREOUENCY

FIGURE 20. Phase between sea surface temperature and salinity anomalies. 118 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

TABLE 3 charge aiiomalies. The coherence between the Cres- The Ratio E&AA As a Function of Frequency cent City salinity anomalies and the discharge of the Smith River near Crescent City is not particularly f San Blunts Langara good. The phase between salinity and runoff anomalies c.p.y. 1 La Jolla I Francisco 1 Reef I Neah Bay Id. I Yakutat is 180 degrees where the coherence is high. -- I- I- I- I- 1-1- 0.70 0.73 I 1.01 0.45 0.36 0.65 COHERENCE AND PHASE BETWEEN SEA 0.84 0.69 0.67 1.39 0.68 0.30 0.67 0.64 0.57 0.54 0.25 0.10 SURFACE TEMPERATURE AND SALINITY 0.44 0.70 0.42 0.28 0.11 0.14 A NOMALI ES 0.40 0.50 0.47 0.39 0.15 0.16 0.50 0.50 0.31 0.83 0.13 0.10 The coherence between surface temperature and 0.56 0.50 0.37 0.50 0.45 0.06 salinity anomalies (Fig. 19) is moderate for stations on islands and on well exposed capes and points, and relatively poor elsewhere. The best coherence is found COHERENCE AND PHASE BETWEEN SALINITY for Blunts Reef Lightship, where for frequencies be- AND PRECIPITATION ANOMALIES tween 1.5 and 6 c.p.y. about 50% of the variations can be related to each other. The phase (Fig. 20) is The coherence (Fig. 16) between salinity and pre- 180 degrees indicating that an inverse relation ex- cipitation anomalies varies widely from station to ists between the anomalies. This is not surprising, station, because of the very local character of rain- since in summer low temperatures are associated with fall. For La Jolla, Crescent City, Neah Bay, Amphi- high salinities, owing to upwelling, and in winter trite Point and Sitka the coherence is relatively high relatively high temperatures are connected with low and does not vary much with frequency. About 40% salinities due to abundant precipitation. A similar but to 60% of the salinity anomalies at these stations can poorer relation is also found at SE Farallon Island. be related to anomalies in precipitation. At Sail Fran- Several northern stations, particularly Race Rocks, cisco and SE Farallon Island the coherence is high B.C., and Seattle, Washington, show an oscillation of (0 to 3 c.p.y.) and low at high at low frequencies the coherence with frequency; the cause for this re- frequencies, which suggests that short period salinity markable feature is not known. The phase between variations are related to some other factor than rain- temperature and salinity anomalies scatters about 180 fall. At Blunts Reef the coherence scatters widely, degrees at many stations which suggests that high which is probably due to the influence of upwelling, trmperatures are associated with low salinities and in addition to precipitation, upon the surface salinity. vice versa; this is not an unexpected result in view The phase (Fig. 17) between the anomalies is gen- of the extensive upwelling occuring during summer erally 180 degrees, which shows that salinity and months. precipitation anomalies are related inversely to each other. At Sail Francisco and a few other places there CONCLUSION is a phase drift from 180 degrees at low frequencies The following results were obtained from a sta- to 0 degrees at high frequencies; this is, however, not tistical analysis of the various oceanographic and significant in view of the low values of coherence at meteorological records : higher frequencies. (1) At most well exposed stations air temperature anomalies can be used as an indicator of sea surface COHERENCE AND PHASE BETWEEN SALINITY temperature anomalies. The relation between these AND RUNOFF ANOMALIES anomalies is direct and instantaneous. At stations located near rivers, the variation in the (2) At most stations not affected by river discharge discharge of the river can be expected to have con- local precipitation anomalies can be used as an indi- siderable influence upon the surface salinity. This is cator of salinity anomalies. The relation between these particularly true of stations located near the boundary aiiomalies is an inverse one. of oceanic and river water. In figure 18 are shown the coherence and phase between the salinity anom- (3) The relation between river discharge and sa- alies at Fort Point, San Francisco and the discharge linity anomalies is good at stations located at the of the Sacramento River at Verona, and other cases. boundary between oceanic and river water. It is not It is seen that about 70% of the San Francisco salin- good upstream. The anomalies are related inversely ity anomalies can be accounted for in terms of anom- to each other. alies of river discharge. The relatively low coherence (4) Temperaturrs and salinity anomalies at most between the salinity anomalies at Astoria and the dis- coastal stations are not related to each other. At island charge of the Columbia River at The Dalles is not stations and on lightships there is a moderate and surprising in view of the fact that Astoria is located inverse relation between these anomalies. The inverse right at the river and is seldom influenced by oceanic relation between temperature and salinity anomalies water. The coherence between the Blunts Reef salin- is characteristic of upwelling regions off the west coast ity and the Eel River discharge at Scotia is moderate, of the United States. and is better at certain frequencies than at others, (5) Sea surface and air temperature anomalies which is probably due to the additional influence of occur more or less simultaneously over very large upwelling upon the salinity. In the best cases about areas. Salinity and precipitation anomalies are very ,50% of the salinity anomaly can be related to dis- local phenomena. REPORTS VOLUME VIII. 1 JULY 1959 TO 30 JUNE 1960 119

1929. A mathematical theory of the vertical distribution of ACKNOWLEDGMENTS temperature and salinity of water under the action of radi- The author is iiidebted for valuable discussions and ation, conduction, evaporation, and mixing due to the resulting convection. Bull. Scripps Inst. Ocean, Univ. Calif. advice to J. D. Isaacs, G. R. Miller, W.H. Munk, and Tech. Ser. 2(6) : 107-306. J. 1,. Rc>id,Jr. The research reported herein was partly 1938. Some energy relations between the sea surface tmlpera- fiiiaiiced by the Marine Life Research Program of the ture ;ind the atmosphere. J. Mar. Res. 1(3) : 217-238. Scripps Iristitntioii of Oceanography of the University Jlunk, W.H., F. E. Snodgrass, and Af. J. Tucker, 1959. Spectra of California, La Jolla. of low frequency ocean waves. Bull. Scripps Inst. Oceanogr. The computations were carried out at the West- Cniv. Calif. 7(4) : 283-362. ern Data Processing Center of the University of Cali- AIunk, W.H., 1‘360. Smoothing and persistence. J. Xet. 17(1) : fornia in Los Angeles. !Y%-‘33. I’icltard, G. L., and D. C. hlcleod, 1953. Seasonal variation of temperature and salinity of surface waters of the British REFERENCES Columbia Coast. J. Fish.’lZes. Bd. Canada 10 (3) : 126-146. Fisheries Itesearch lioard, C:~n;itl:i, l!)X. 0l)servations of &:I- Reid, J. I>., Jr., 1‘360. Oceanography of the northeastern Pacific xiter Temperature ;ind Sulinity on the Pacific Coast of during the last ten years. Calif. Coop Oceanic Fish. Rep. (::in:ida. Ilauuscript ltelwrt Series 23. Pacific Oceano- 7: 77-‘30. graphic Group, K:iiiniiuo, U.C. 100 pp. Roden, G. I., 1958. Spectral analysis of a sea surface tempera- Herlinveaux, It. II., 1!)>7. On tit1:il rrirreiits and properties of ture and atmospheric pressure record off Southern Cali- the sea water along the I

THE USE OF A COMMON REFERENCE PERIOD FOR EVALUATING CLIMATIC COHERENCE IN TEMPERATURE AND SALINITY RECORDS FROM ALASKA TO CALIFORNIA MARGARET K. ROBINSON INTRODUCTION One of the major aims of the California Coopera- tive Oceanic Fisheries Investigations has been to cor- Yakutat relate fluctuations in fish catches with fluctuations in the oceanic environment. Toward this end, it has been necessary to evaluate the scale of variability of ocean temperature and salinity for periods of a month, season, year, decade and longer. The area of interest has expanded from Southern California waters to include the area from Cape San Lucas to the Gulf of Alaska. When the widescale Island warming in the eastern North Pacific took place between 1957 and 1959, our horizons were again expanded to include the western North Pacific in search of cause and effect relationships which might explain these large-scale changes (Sette and Isaacs, 1960). 50"h The ultimate aim in studying climatic records is to make prediction of climatic events possible. Entrance Island One of the most useful groups of data which have Race Rocks shed light on the climatic changes of the ocean is that of the daily temperature and salinity observations at shore stations between La Jolla, California, and Yakutat, Alaska. (U.S. Coast and Geodetic Survey, 1956, 1958; SI0 Ref. 60-30, 1960; Fish. Res. Bd. of Canada, 1947-1960.) There are two important reasons why these shore station data are so valuable: First, there is a very Crescent City high positive correlation between air temperatures at 40'1 adjacent meterological stations and sea-surface tem- 1 peratures at shore stations (I-Iubbs, 1948 ; Tully, 1938) ; Second, it has been possible to correlate sea- San Francisco surface temperature anomalies up to 150 miles offshore with sea-surface temperature anomalies at Avila nearby shore stations, and thus indirectly with air temperatures along the Pacific coast (Reid, Roden, Wyllie, 1958 ; Robinson, 1957). La Jolla Therefore it has been possible to extrapolate physical conditions in the California Current region for 309 periods where observations in the open ocean were unavailable, and to use these probable values to extend FIGURE 1. Shore Station Locations. our understanding of the relation between the occurrence and behavior of marine organisms with changes Temperature and salinity anomaly charts were in their physical environment. therefore prepared, using means from the period There are 24 shore stations along the Pacific coast 1949-1958 as a common reference base, for twenty-four with almost continuous data. The periods over which Canadian and American shore stations whose locations the data have been taken, however, vary from 13 years are shown in figure 1. Before the publication of the at Santa Monica to 44 years at La Jolla. While it may charts, Japanese, Canadian and American oceanog- be preferable to use the entire length of record when raphers agreed to use the period 1950-1959 as a com- studying variability at individual stations, it is im- mon base of reference, not only for shore stations' possible to properly evaluate differences among sta- data but also for open ocean data in the North Pacific. tions without using a common base of reference. Therefore, in order to make the charts which had Contribution from the Scripps Institution of Oceanography. already been completed more useful, and to relate 122 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

Period of LongTerm oc Mean A J FMAMJ JASON D A JFMAMJ JASOND Yakutat “5r Son Francisco (Fort Point) 1” 1941-1958 9”Fv1922-1958 0.5 u-q---Jit 0.5 Pbcific Grove 1937-1958 ‘0 1921- 1958

0.5

19444958-

Ketchikan + I924-l958&--l I924-l958&--l Sonta Monica 1946-1958 Langaro +

05- Los Angeles Triple1gm4=?LF+- Island 1924-1958

Pine Island05 Balboa I937-1958& 1924-195Ei I I ~ - 05-

U Scripps Pier (Surface)

ScrippsPier(5Meter Bottom) 1926-1956 - Jo. OC

Entrance ISGINI 1949-1958 MEANSand 1950-1959 MEANS 1937-1958 6 MINUS LONG TERM MEANS

0.5 - I949 - 58 Ond 1950-59 MEANS SAME Departure Bby 1936-1968; AMT 1950-59 MEAN HIGHER THAN 1949.58 MEAN

AMT 1950-59 MEAN LOWER THAN Race Rocks 1949-58 MEAN

0.h Oc

FIGURE 2. Relation of 10-year temperdture means to long period means. REPORTS VOLUME 1'111, 1 JULY 1069 TO 30 JUNE 1060 123

Period of Long Term Mean %e %e A J FYAM J JASO ND A J F MA YJ J A 0 o,2rA J FYAMJJASONDn Lanaam .:: .::

OND/ Tripb Wand 19404958

+ -0 I- 0.2 a4 I 10.4 Koins ISlaM Juneau + 1935-1958

Amphitrite

1935-1958B 0.6

0.0

... Pacific Grove 1921-1958 0.4 Ifi .. 0.1 .. 0.2 J0.3 Pwt Hum

j0.4 0.6 Santa Mmico 1946-1958 U !? 0.4 0.1 0.2 Ketchikon Deportura Bay 1924-1958fi 1935-1958

02 0.2 1 I 0.4

0.6 Scripps Pier(Surfoce) 196-1958 0.0 -+; Race Rocks l.ok 1942 -1958D Scripps Rer (5Meter Bottom) I.2 0.1

- 1949-58 and 1950-59 MEINS SAME 4MT 1950-59 MEAN HIGHER THAN 1949-1958 MEANS md19504959 MEANS MINUS WNG TERM MEANS 1949- 50 MEAN

MMT 1950-59 MEAN LOWER THAN 1949- 58 MEAN

FIGURE 3. Relation of 10-year salinity means to long period means. 124 CALIFORSIA COOPERATIYE OCEASIC FISHERIES ISYESTIGL\TIOSS

+20 -+1.0 -0 --1.o -2.0 - +2.o -+l.O -0 --LO ~ -2.0 -+LO -0 --IO - -2.0 -+z.o

FIGURE 4. Chronological Temperature Anomalies based on 1950-1959 reference period. them to the 1950-19.59 reference period, changes and lower-than avtarage teniperatiwes, and the relation of additions were macle in the charts prior to publication, persistenrcl to coherence. 5. Salinity coherence as permitting an evaluation of the effect of altering the shown in SI0 Ref. 60-30 (1960). 6. The relation of base period, as well as meeting the original objectives. temperature aiiomalics to salinity anomalies. Th(. relation of the 1949-1958 and thr 1930-1959 ten- year monthly aiid annual meails to each other and to DISCUSS10 N the loug-term means for each of twcnty stations is LYampZixq error iii shore In order to shown for temperature in figure 2, and for salinity in statzon data: evaluate the reliability of the nioiitlily temperature figure 3. Four stations were omitted becanse data aiid salinity means and the anomalies derired from were incomplete for the period of reference. them at the shore stations, it is necessary to examine The chroiiological temperature anonialy charts for variabilit>- and the staiiclarcl error of the iiieaiis (om). all 20 stations were ]lot redraw11 bawd 011 the 1930- The stantlard deviations of temperature at shore sta- 1959 reference period. Instead, only four key stations, tions T\ ithin single months vary froni 0.2"C. in winter each typical of the nearby area, were redrawn. Sta- to 2.0"C. in summer and from 0.04"C. to 0.37"C. tioiis selected wre Ketchikan, Alaska ; Dcpartnre Ray, IS", Vancouver Island, B.C. ; Pacific Grove and La Jolla, For the Southern California stations the staiidard California. Figure 4 presents the temyeratnrc anoma- deviations of salinity within indivitliial months are lies based on the 1950-1939 reference period for these very low, because there is little rainfall and runoff. stations. IIere om is iisimlly less than O.O1°/o(,. However, at the The chronological temperature and salinity charts Alasliaii and British Colimibian stations there is excess based on the 1949-1958 reference base and anomaly precipitatioii and rimoff with resulting extreme ranges charts for individual stations plotted by months for in salinity. Sail Francisco falls iii this class also, be- all years are included iii SI0 Ref. 60-30 (1960). Eoth cause of the treniendons river runofl which enipties 1949-1938 and 1950-19c5!3 reference perio(1 means are into Sail Francisco Bay. Sample standard deviations given on the inclividiial station charts. of salinity for individual months at these northern The following points will be discussed in reference stations range from 0.16~)/Ooto 5.000/oo. The standard to the arioinal3- charts published in SI0 Ref. 60-30 error of the mealis, based on these values, ranges from (1960) and to figurcs 2, 3 and 1. Station-to-station eo- 0.03°/oo to 0.910/,,,l. At both northern and southern herence, for the purpose of this discnssion, is defined stations the aiioinalies from nioiith to month and from as the agreement in sign of tempcratnre (or salinity) year to year exceed the possible sampling error. aiiomalies when conipnted from the same base of ref- The relation between the 1949-19.58 and the 1950- erence. 1. Sampling error for nioiithly temperature 1959 referenre periods to each other and to the aiid salinity means and their respective anomalies. long-term iiieans at each station is shown in figures 2 2. The relation of the 1949-1958 aiJd the 1950-1959 and 3. For both figures base period anomalies were refmelice periods to each other and to the long-term computed at each station by siibtracting the long- means at each station. 3. The extent of station-to-sta- period means from the 1949-1958 monthly and annual tion temperature coherence ; visiblr evidence of eo- means, and also froin the 19t50-1959 means. herence ; statistical evidence of coherence ; the relation Iz cl a t ion 1) etu'c e n t c ti -Jjeor n n tl lon {j-t e r nx n7 cans : of iiicidence of coherence to time of year; comparison The 1949-1958 temperature means (Fig. 2) were lower of coherelice computed for 20 stations with that corn- than the long-term meails at three of the northern putecl for 4 key stations ; prediction of coherelice ; and stations in all months At other northern stations the relation of magnitude of anomalies to incidence of period from July to December inclnded several posi- coherence. 4. Evidence of persistence of higher- or tive anomalies. The negative anomalies for the same months were smaller than in the first six months. At Bay, Pacific Grove and La Jolla, is similar to the three southern stations-Pacific Grove, Port llueneme charts for all 20 stations based on the 1949-1958 refer- and TAX Aiigeles-the anomalies were predoniinantly eiwe period in all major features. Inspection reveals positive, while at the remaining southern stations the the strikingly widespread north-to-south and station- anomalies were pretlominantly negative. There are to-station agreement in the well-known warm years marked differences between anomalies at Port IIue- (1926, 1931, 1941, 1937-1959) and in the cold years neme, Santa Alonica, Iios Angeles and I3alboa. The (1933, 1950, 1955, 1956). Another visible climatic fea- long-period record at the latter two stations covers the ture which emerges from both the temperature charts same years, so it wo~~ldappear that the marked differ- of SI0 Ref. 60-30 (1960) and figure 4, is that changes ences at these two nearby stations are due to local from positive to negative anomalies, or vice versa, do effects, which may be the coiistrnctioii of Los Angeles not occur at the saine time from north to south. Note, harbor and/or the modifications to the harbor at New- for example, that the warm period of 1957-1959 ended port. On the other hand, there may have been real earlier in the north and persisted through 1959 in the differences in the c>ircwldtion at these two localities. south. The 1950-1959 means are, in general, higher than A quantitative expression of the station-to-station the 19-19-1958 means, but they, too, are lower than the coherence of temperature anomalies was obtained by long-term means. These differences are greater at the determining the percentage of the total number of southern stations than at tlie northern stations. With stations having temperature anomalies of the same the exception of Port IIuenerne aid Tlos Angeles, the sign, either positive or negative. This was done for 1950-1959 means agree inore closely with the long-term the 20 stations used in S10 -Ref. 60-30 (1960), (Table means than did the 1919-3958 means. 1) and for the four stations in figure 4 (Table 2). It is important to note that the differences between Note iii Table 1 that the number of stations with the ten-year teniperature iii~aiisand the long-term temperaturr data ranges from 10 in 1935 to 20 in means never exceed 0.5"C." Thus, the change in the 1946 and subsequently, and in Table 2 that data were common reference period principally effects those not rollected at Pacific Grove in 1940. Percentages months and years with small deviatioiis. The spectacu- for 1935-1959 appear in Table 1 only where there is larly different years (e.g.-1931, 1933, 1911, 1958) a 95% probability that the voherence could not have woiild still stand out regardless of the base of refer- happened by chaiwe. This requirm that the sign of ence. the anomalies must be tlie same at 80% of the stations IJnlike the relatively systematic temperature rela- whm based on 10 stations, and at 70% of the stations tion discussed above, the salinity relation between the when based on 20. The prrcentages in Table 2 are ten-year reference period means and the long-period included for comparative purposes even though agree- means is rancloni aiid confused. Figure 3 shows that ment at 3 out of 4 stations (75% ) is not significant tlie patteriis of the anoinalies, even for stations where at the 9.5:4 probability level. Table 1, based on the long-term records are of the same length, seein to have larger number of stations, is, consequently, the best little relation to each other. Port llneneme, generally estimate of coherence. Bold face percentages in both little affected by runoff, had such low salinities follow- tables indicate that at the majority of the stations the ing a very heavy rainfall in April, 1958, that the 1949- anonlalies were positive. A -, with no percentage 1958 April mean salinity was Yery much lower than listed above, indicates that coherence was not statis- the long-term mean. The ten-yrar mean is so low, in tically significant, and the monthly anomaly was posi- fact, that the April values in the Port Hueneme salin- tive at the majority of the stations. Blank spaces indi- ity anomaly charts are all positive except for 1958. cate the majority of monthly anomalies were negative There is also a random and confused relation be- and coherence was not statistically significant. tn7c~~lithe salinity means of the two ten-year periods The effect of the shift of base reference period on at all stations north of and inclitding Sail Fraiicisco. the percentages of positive and negative anomalies For exaniple, the 1950-1939 nieaii salinities are lower occurs when the anomalies are small, for statistical than the 1949-1958 means 60% of the time; however, evidence shows that 50',i of the changes in the two the (lifterelices occur irregularly in time and randomly ten-year means were 0.1"C. or less and never exceeded from Yakutat to Sail Francisco. Differences south of 0.3"C. In a previous reference to figure 2, it was rioted Sail Fraiicisco are so small that no trend can be that means for both ten-year periods were generally detected. lower than the long-term means at most stations, Tenzperatzire Coherence: Station-to-statio11 coher- although those of the 1950-1959 period were closer to ence as defined on page 124 is visibly evident in the the loiig-term mean The use of either ten-ycar base chronological temperature anonialy charts. The most reference period not only produces the expected re- complete evidence of station-to-station coherciice is sult of nearly equal numbers of po\itive and negative published in SI0 Ref. 60-30 (1960), where both tem- anomalies within the iiidividual ten-year reference pe- perature and salinity chroiiolopical anomaly charts riods, but also rewlts in larger niinibers of positive for 20 stations are available. Figure 4, based on the anomalies in the years 1933-3949. It is worth noting 1950-1959 reference period at Ketchikan, Departure that there is a snrprisiiig similarity in percentages of positive monthly and aniiual temperature anomalies. 2 75 % of the differences between the means of the two ten-year periods at the northern stations were 0.1"C. or less and [Jsing the 194!)-1958 base reference period, Table 1 exceeded 0.25"C. only twice. At the southern stations, only shows that 64% of the monthly anomalies and 68v 27% of the differences Tvere as small as 0.1"C. or less, and 26% exceeded 0.25"C. of the annual anomalies were positive at the majority 126

TABLE 1

SIGNIFICANT PERCENTAGES OF TEMPERATURE COHERENCE BASED ON 10-20 STATIONS _____ Month

~ -- __ 4nnual inoma- No. of Stations Year 1 2 3 4 5 7 10 11 12 lies 1 ______-__ _- - - 90 - 80 100 80 82 - - 82 82 __ __ 91 82 __ 82 73 87 80 - 80 - 73 86 73 - 86 - 86 - - - - - 86 - - - - 86 100 - 100 100 94 94 94 75 __ 94 - 94 83 94 100 94 100 94 94 82 82 100 94 100 95 95 95 95 72 83 89 83 96 - 72 72 83 -- 95 95 95 78 95 90 90 - 74 - 84 89 - 84 95 90 74 - - - 90 79 __ 75 - 70 - 85 95 75 90 90 70 70 - 100 85 90 90 70 80 90 96 85 - 70 75 - 90 100 100 100 70 80 - 80 70 85 95 100 100 70 70 100 75 - 80 - 85 - 80 - - 80 75 95 80 90 - 70 85 90 75 - - 80 -_ 70 - 70 100 85 80 70 75 80 - - 70 70 - 85 90 95 95 100 100 100 100 80 90 100 90 70 90 80 75 100 80 100 - 75 85 80 95 95 90 95 100 100 100 100 100 90 70 - - 85 100 90 95 95 100 85 85 90 75 75 95 __ ~ ______20 15 17 1 14 16 13 13 I5 18 18 17 __ -- Rolrl t,pe-I’ositive mimalies at majority of statioiis-pr~cent;i~c signific;int. IkiYli-Pndtive iiiiiiiniilies ;it majority of stations-~peirr.nt;lgenot significant. Light tyrie-Negarive anomalies at majority of st;ttilins-l!~irent~l~esianifieant Rltinli-Xegativr anomitlies at majority of stations-percentage not significant.

TABLE 2

PERCENTAGE OF TEMPERATURE COHERENCE BASED ON FOUR KEY STATIONS _____ Month 4nnual __ ~ inoma- No. of Stations I Year 1 9 3 4 5 6 7 8 9 10 11 12 lies ~ ______~ ___~__- 100 75 100 100 __ 75 75 100 75 75 75 75 75 75 75 75 100 100 75 75 100 75 75 100 75 100 100 75 75 75 100 75 75 75 100 - 75 7 5 75 75 __ _- _- 75 100 - - 75 __ 75 75 75 75 75 100 100 75 100 100 100 100 100 - 100 100 _- 100 100 100 100 100 100 100 100 75 75 75 75 100 too 100 100 100 75 100 76 75 - 75 75 75 75 100 75 75 __ 75 100 100 75 100 75 75 __ - 75 75 - 75 75 75 75 100 100 - 75 - 75 75 100 75 75 75 ___ - - - 75 100 75 - - 75 100 100 75 75 75 100 100 100 100 - 75 _- 75 75 - 100 - __ 75 75 - - 75 100 100 100 75 100 100 75 75 75 75 75 75 100 __ 100 100 100 100 100 75 100 - 75 - 100 100 100 100 100 75 - 75 75 100 7.i 75 75 100 - 100 75 75 100 - __ - 75 75 - - - 75 75 - 75 75 -_ _- _- _- 75 75 75 75 75 75 75 - - 25 __ - 75 __ - -- 75 100 100 100 75 100 100 100 100 100 100 I00 75 100 75 - 100 75 100 75 73 75 75 75 - 75 100 100 75 100 100 100 100 100 100 100 75 ‘76 75 100 100 100 100 100 100 100 100 - 75 75 100 75 75 100 __ -- ______14 11 11 8 7 0 2 10 12 9 10 20 20 18 18 17 11 10 19 16 21 23 ______Bold t>-pe-Positiv(x anomalies at m;ijorits of st;itions-prt centage significant. Dasli-I’oWve anomalies at maji)rity of st;itilins-percent;isc lint significnnt. Light type-Negative anomalies at m;ijority of st;itions-iiercent;ig~ signifieant. Blank-Negative anomalies at majority of station-percentage not significant, REPORTS VOLU;\IF, VIII. 1 JULY 1959 TO 30 JUNE 1960 127 of the stations. Table 2, using the 1950-1959 base ref- the probability of any coherence reduced to chance, erence periocl, shows 57% positive monthly anomalies there were 90 instances where no coastwise coherence arid 5654 positive aiiiiual anomalies. would have been predicted; yet, in 28 of these months The bottom row of numbers in Table 1 represents significant coherence did occur among the 20 stations. the total number of years in which significant coher- Two questions arise concerriilig the absence of any ence o(uirs iii a given month ; hence, coherence is best significant coastwise coherence during about half of in Decembcr and January, and poorest in August and the months of the past 25 years. Are the northern Srptrmber. Turning to Table 2, the row of figures next and southern parts of our coasts in different climatic to the bottom represents tlie iiuniber of years in which regimes? Or, recaognizing that there are local effects there is wniplete agreement in the sign of the anoma- which differentiate stations near the more typical lies at all statioiis. The results are too conservative, four key stations, is the coiispicuous absence of co- so the second row of figures is given for 75-100c/c coherence related to the size of anomalies during those hercwAe. Xow the results are too liberal an estimate periods when lod rf-F'ects might be masking wide- of coherence, poiiitiiig up the dificulty in dealing sta- spread climatic effects? tistically with only four stations. One of the major purposch in studying long-period TABLE 4 records is to develop relations which will make tlie AGREEMENT IN SIGN OF ANOMALIES AT predirtion of climatic events possible. In Table 1, FOUR KEY STATIONS 1935-19.5!3 station-to-statioii caoherriice was significant No. No. at the !).5V probability le~lin 170 of the 300 months Ionthly Annual (37% ). Hornever, analysis of data from 20 stations Coherence Inorna- Sub- horns- is tiiiie cwiisimiiiig and data are not received immedi- 4 Stations lies Total lies atc.ly from all stations. Therefore, a detailed compari- so11 was made of the results using all 20 stations arid 100% 10 those bawd 011 four key stations in order to determine 75% La Jolla differed ...... ~__~_ -.. .. . 3 how good ail estimate of coherence could be made on 75% Ketcliikan differed._.- ...... ~~~~ 8 75% Departure Bay differed-...-. . . . .~ 1 the basis of the four stations alone (Table 3). 75% Pacific Grove differed.- ..._...... 1

Table 3's ronsidcration of four key station coher- 50% Iietclrikan-1)eparture Bay alike eiiw at loo%, 75>4,and 50'/(, shows the extent to Pacific Grove-La Jolla alike .... 2 50% Ketchikan-Pacific Grove alike TABLE 3 Departure Bay-La Jolla alike.. . 0

COMPARISON OF COHERENCE COMPUTED FOR 10-20 STATIONS 50% Ketchikan-La Jolla alike AND THAT COMPUTED FOR 4 STATIONS Departure Bay-Pacific Grove alike. 0

I I I I No. Totals.. -.-. -...... -... 25 Colicrenrc Coherence -4nnual 4 Stations All Stations Anomalies -1 -1 1-1-1-- For purposes of examining the first question, a Significant, same sign.. ~..- 82 tabulation of the precise may in which coherence oc- Not significant, same sign-. 11 Not significant, diff. sign-.. 1 curred among the four stations was constructed (Ta- 94 ble 4). In 31% of the months, all four stations had

Significant, same sign.. ~ ~.. 58 Not significant, same sign..- 43 the same anomaly sign. In 39% of the months, one

Significant, diff. sign-. ~ ~... 1 of the four had a different sign. Though La Jolla Not significant, diff. sign.-. 14 116 differed most frequently, the differences, percentage- Significant.- ~.~.. ~. .._ ~... 29 wise, were not significant among the four stations.

Not significant-. .. ~ - ~...~.61 90 In the remaining 30% of the months, there is co- ___-__ herence between different pairs of stations. The two Totals-...... _~ ...-. 300 300 25 northern stations (Ketchikan and Departure Bay) have similar but opposite anomaly signs to those of which accurate coastwise prediction would have failed. the two southern stations (Pacific Grove and La With four-out-of-four ( 100% ) coherence, prediction Jolla), twice as frequently as either of the other coastwise would have been right in 82 out of 94 two possible combinations. (Note: Table 7 gives ad- months ; however, such a conservative prediction ditional evidence of somewhat higher frequency of wonld have failed to anticipate 88 additional cases coherence when we separate the complete group of of coa\twise cohercnce which actually occurred. With 20 stations into northern and southern components.) three-out-of-four (73% ) coherence, predictions for It is of utmost importance to note, however, that c.oastwise cohrrence would have been correct for only coherence among all four stations results twice again half of the time, or 58 months. Fifteen of the re- as frequently as when the northern pair is not co- maining months would have been assigned incorrect herent with the southern pair. The previously noted signs for their rrspective anomalies : during these north to south time lag in climatic events partly months anomaly signs at the majority of the stations accounts for those periods which show no coherence. were actually opposite those at three of the four key Relation of magnitude of anomalies to coherence: stations. With two-out-of-four (50% ) coherence and In order to investigate the second question, the mean 128 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

TABLE 6 RELATIONSHIP OF MAGNITUDE OF MONTHLY TEMPERATURE ANOMALIES AT DEPARTURE BAY TO COHERENCE AT FOUR KEY STATIONS

(Departure Bay Monthly Anomalies)

No. Months % I""I""I" ""I""I""I""I" ""I""I""I"" Im- XI 0 0 x. 0 m -too Size of No. monthly Coherence Sub- Anomalies Anomalies was 100% Total - -0 0 I 1 1 1 /o 0 90 - -sa I I I I ox 0.0"C-1.O"C.._____. 1 225 I 225 I 66 19 1.6O -2.0°._____._.. 2.1' -2.5"...... I :I 2.6' -3.0°._..____.. 4 1 3.1' -3.5" ...... 1 1 1 ,5 1 1 1 28

Totals...... 300 300

0 - Raltln Anmalhs . N.prtlWAkTmlIn tions. Unfortunately, however, there was nearly as FIGURE 5. Correlogram. Absolute Magnitude of large a percentage (29%) of the months with 100% Anamolies vs. Coherence. coherence at all four stations when the La Jolla and Departure Bay anomalies were 1.O"C. or less.3 ered in Table 1. Regretfully, the correlograms failed Though about one-fourth of the 300 months in the to demonstrate any linear relationship between the past 25 years La Jolla had anomalies greater than anomalies and the coherence. It was of interest to l.O°C., in only 9% of those 300 months did anamolies note, however, that when the magnitude of the mean greater than 1.0"C. concur with 100% coherence at anomaly was greater than l.O°C., the coherence was the four key stations. The same was approximately statistically significant. Unfortunately, this occurred true at Departure Bay. Therefore, if the magnitude only twice in the past 25 years for annual anomalies, of anomalies at either station is useful in prediction five times for February anomalies, and not at all at all, it is only that they might serve to alert us to for August, the largest mean absolute anomaly for possible widespread coherence. August being only 0.85"C. Persistence: The relationship of persistence to co- Because two of the four key stations are located herence, though suspected, has never been so con- at major scientific institutions and would be first to clusively evident as that revealed in Tables 1 and 2, receive notice of major climatic changes, La Jolla and figure 4, and SI0 Ref. 60-30 (1960). Using auto-cor- Departure Bay records wete examined further for relation computations, Roden and Groves (1960) had a possible relationship of the magnitude of their re- reported a significant tendency toward the persistence spective monthly anomalies to the incidence of 100% of temperature anomaly signs over periods of five coherence at all four stations (Tables 5 and 6). In- months in an ocean area just off the Washington vestigation disclosed that anomalies at La Jolla and coast. But, Tables 1 and 2 and figure 4 show that Departure Bay were greater than l.O°C. during 37% when coherence is highly significant along the en- of those months with 10070 coherence at all four sta- tire coast, persistence in the sign of the anomalies is apparent over even longer periods. For example, TABLE 5 note that during the 1955-1956 period when signifi- RELATIONSHIP OF MAGNITUDE OF MONTHLY TEMPERATURE cant coherence occurred coastwise, negative anoma- ANOMALIES AT LAJOLLA TO COHERENCE lies persisted for a period of 16 months without a AT FOUR KEY STATIONS break ; and during 1957-1958, positive anomalies persisted for a period of 17 months. In fact, during (La Jolla Monthly Anomalies) the winter of 1958 there was a five-month period when positive anomalies persisted at all 20 stations. Vo. Months Size of No. monthly Sub- Coherence Sub- Because present records are too short, auto-correla- Anomalies Anomalies Total was 100% Total tion computation methods would fail to reveal the probabilities of reoccurrence of persistence over peri- O.O0C-1.O0C...... 2 15 62 ods as long as 16 or 17 months. 215 62 1.0" -1.5O ...... 53 18 Salinity Coherence: Coastwise station-to-station co- 1.6' -2.0°_...... 24 10 2.1' -2.5°_.....__.. I7 3 herence in salinity anomalies was shown to be very 2.6" -3.0"...... 1 poor in the chronological salinity charts published 85 32 sThe fact that these percentages apply to both stations appears Totals ...... 300 300 94 1 94 to be fortuitous. In less than 1% of the 300 months did both stations have anomalies greater than 1.0" C. with the same signs at the same time. REPORTS VOLUME VIII, 1 JULY 1959 TO 30 JUNE 1960 129

them during the exceptionally warm years of 1931, 1941 and 1957-1958, or during the exceptionally cold years of 1933 and 1955-1956. It was found that the warm years of 1941 and 1957-1958 were accompanied by low salinities (high precipitation) at most of the stations from north to south. This was not true, however, at the southern stations during 1931. Ketchikan, the only northern station with salinity observations at that time, showed very low salinities in the winter of 1930-1931 accompanying its higher than average temperatures. In the cold years, there was no consistency in the temperature-salinity anomaly relationship from station to station, nor from one cold year to another.

CONCLUSIONS The use of a common reference period has given us considerable new information about coastwise coherence in the northeastern Pacific. We can now accept station-to-station agreement or local station disagree- ment with more confidence. Large-scale climatic events and associated coherence were already well known, but now there is evidence that significant coastwise coherence in monthly temperature anomalies occurred during 5770 of the months for the past 25 years and that coherence was highest in winter and lowest in August and September. Furthermore, in 17 of the 25 years there was significant coherence in annual temperature anomalies. In comparison with the possibilities for coastwise Salinity Temperature prediction based on 20 stations, prediction based on four key stations would be either too conservative ALL STATIONS (10-20) using the criteria of 100% coherence, or too liberal Annual Anomalies ...... 7 (28%) 17 (68%) using the criteria of 75% coherence. February Anomalies ...... 6 (24%) 15 (60%) August Anomalies ...... 9 (36%) 6 (24%) Coastwise, no linear relation between coherence and absolute magnitude of mean temperature anomalies NORTH STATIONS (4-13) Annual Anomalies ...... 9 (36%) 15 (60%) exists, except €or the occasional instances when mean February Anomalies ...... 11 (44%) 17 (68%) anomalies exceeded 1.0" C. Respective mean anomalies August Anomalies ...... 12 (48%) 8 (32%) for La Jolla and Departure Bay did exceed 1.O"C. SOUTH STATIONS (5-7) more frequently, but these occasions were rarely con- Annual Anomalies ...... 5 (20%) 14 (56%) current with signific'ant coastwise coherence. Thus, February Anomalies ...... 6 (24%) 19 (76%) August Anomalies ...... 9 (36%) 8 (32%) prediction of coherence on the basis of magnitude of anomalies at one of the major oceanographic institutions would have little probability of success. and southern stations. Coherence for the correspond- There is evidence of a positive correlation between ing temperature anomalies is included in this table station-to-station coherence and persistence in time of for contrast. positive and negative anomalies. The table shows that while temperature coherence Coastwise coherence for salinity anomalies is gen- in August is least frequent, salinity coherence in Au- erally poor. However, it is somewhat better in summer gust, even though it occurs only slightly oftener than than in winter and also in the north if the sample August temperature coherence, is most frequent. Sa- stations are divided into north and south groups. linity coherence is also more frequent at the northern No clearcut relation exists between the signs of stations than at the southern stations or for the two temperature anomalies and the signs of salinity combined. In the north, though coherence is almost as anomalies. frequent in February as in August and though in Therefore, predictions of climatic events have low both months it occurs more frequently than in the probability of success at present if based on statistics annual anomalies, even here salinity coherence occurs from past records of shore station temperatures and less than 50% of the time. salinities. Namias (1960) and others are currently Relation Between Temperature and Salinity Anom- working toward clarification of the interaction of the alies : The temperature and salinity anomalies were atmosphere and the ocean. Perhaps when a better examined to see if there was any relation between understanding of these physical inter-relationships is 130 CALIFORSIA COOPERATITI? OCEASIC FISHERIES ISTESTIGATIONS gained and when a major break-through toward suc- Hubbs, C. L., 194% Changes in the fish fauna of western Sorth A.nierica correliited with changes in ocean temperature. J. cessful long-range weather forecasting is made, we Mar. Res. 7: 459-482. may be able to use these results in successfully pre- Samias, J., 1969. Recent seasonal interactions between north dicting ocean temperatures. Pacific waters and the overlying atmospheric circultltion. J. Geophys. Res. 64, (6) : (31-646. Reid, J. L., Jr., Q. I. Roden, and J. G. Wyllie, 1938. Studies ACKNOWLEDGMENTS of the Californiii curreiit system. Calif. Poop. Oceanic Fish. Ziinest. Progress Rept. 1 Jitly 195G to 1 Jaiiuary 1958: 29-55. The author wishes to thank Marguerette Hchultz, Robinson, M. K., 1957. Sea Temperature in the Gulf of Alaska dean Gilley and Dawn Eddy for their careful assist- ant1 in the northeast Pacific Ocean, 1941-1932, R rill. Scripps ance in the tedious task of computing and plotting Znst. Oceanog. 7 (1) : 1-98. anonlalies and drafting final results. This paper rep- Rotlen. 0. I., and Q. W. Groves, 1960. On the statistical pre- resents rmnlts of research aarried out by the Univer- diction of ocean temperatures. d. Geophys. Res. 65 (1): 2.19- 263. sit- of Califoriiia, under contract with the Office of Sette, 0. E., and J. I). Isaacs, 1960. Symposium on changing Sara1 Research. Reproduction in whole or in part is years, 1957-1!)58; edited by 0. E. Sette and .J. I). Isaacs. permitted for aiiy purpose of the U.S. Government. Calif. C‘oop. Oceniiic B’iuh. Incest. Rept. 7. This research also received support from the Cali- Tully, J. I’., 1938. Some relations between meteorology and foimia Cooperative Oceaiiic Fisheries Investigations, a coast gradient-currents off the 1’:icific coast of Sorth America. l‘raiis. -4ncer. Ceophys. Cnion, 19th .4iitiitul Meeting, 1: 176- project sponsored by the Marine Research Committee 183. of the State of California. C.S. Coast and Geodetic Survey, 1936. Surface water temperatures at tide stations, I’acific coast. Special Publication 280. U.S. Coast ant1 Geodetic Survey, 19%. 1)rnsity of sea water at REFERENCES tide stations, I’nrific coast. CdC8 I’ubliccctiotc 31-4. Fishoriw Rescwrch lioard of Cnnada, Pacific Oceanographic University of California, Scripps Institution of Oceanography, (;roup. S:inainin, 13. C., 1947-1960. 0bservtitions.of sea water, 1960. Temperature and stiliiiity :inonialy charts for 24 Ca- t+>tiip(,ratiire9salinity and density on the Pacific coast of nadian and Amrrican shore stations based on a common ref- C’ait:i(l:i. (niimeo.), 1-18. erence period. SI0 Hef. 60-30.

printed in CALIFORNIA STATE PRlNllNG OFFICE A34722 1-61 4M */ OREGON

1 .\ \

These maps are designed to show essential details of the area most intensively studied by the California Cooperative Oceanic Fisheries Investigations. This is approximately the same area as is shown in red on the front cover. Geographical place names are those most commonly used in the various publications emerging from the research. The cardinal station lines extending southwest- ward from the coast are shown. They are 120 miles apart. Addi- tional lines are utilized as needed and can be as closely spaced as 12 miles apart and still have individual numbers. The stations along the lines are numbered with respect to the station 60 line, / the numbers increasing to the west and decreasing to the east. Most 2s. of them are 40 miles apart, and are numbered in groups of 10. This .Clarion permits adding stations as close as 4 miles apart as needed. An . example of the usual identification is 120.65. This station is on line ,e-* vG 120, 20 nautical miles southwest of station 60. The projection of the front cover is Lambert's Azimuthal Equal Area Projection. The detail maps are a Mercator projection. Art work by George Mattson, U. R. Bureau of Commercial Fisheries.

4. \