Methods of Oabservation at Sea Sea-Surface Temperature

Home , Jerome Namias, Meteorology, Sea surface temperature, Weather forecasting, Weather ship

WORLD METEOROLOGICAL ORGANIZATION WORLD METEOROLOGICAL ORGANIZATION

TECHNICAL NOTE No. 2 TECHNICAL NOTE No. 103

METHODS OF OABSERVATION AT SEA SEA-SURFACE TEMPERATURE

Lecture presented during the scientific discussions at the fifth session of the PARTCommission I – SEA SURFACE for Maritime TEMPERATURE Meteorology

WMO-No.WMO-No. 247. 26. TP. TP. 135 8

Secretariat of the World Meteorological Organization – Geneva – Switzerland WORLD METEOROLOGICAL ORGANIZATION

TECHNICAL NOTE No. 103

SEA-SURFACE TEMPERATURE

Lectures presented during the scientific discussions at the fifth session of the Commission for Maritime Meteorology

un C 55I.526.6

IWMO - No. 247. TP. 1351

NOTE

The designations employed and the presentation of the material in this publication do not imply the expressiun of any opinion whatsoever on the part of the Secretariat of the World Meteorological Organization concerning the legal status of any country or territory or of its authorities, or concerning the delimitation of its frontiers. III

TABIE OF CONTENTS

Foreword •••••••••••••••.•••..•.••••.•••••••• "...... V

Summary (Englisp, French, Russian, Spanish). •.. .• ..•••••••.•.. •• ••••.•..••••..••..• •• VII

Use of sea-surface temperature in long-range prediction (by Jerome Namias (U.S.A.» 1

Collection and dissemination of sea-surface temperature data for the north-west Pacific and their utilization for fisheries (by Iehiro Imai (Japan» •••••.•••••.••.• 19

Sea-temperature structure and its relation to the United States tuna fisheries in the eastern Pacific Ocean (by Glenn A. Flittner (U.S.A.» 37

Sea-surface temperature patterns in the north-east Atlantic (by J. D. Booth (U.K.». 77

Variations de la temperature de la mer au voisinage de la surface (by J. Romer (France» 97

Sea-surface temperature; some instruments, methods and comparisons (by A. B. Crawford (South Africa» .. 117

CpaBHHTeJIbHble H3MepeHHR TeMl1epaTypbl IioBepxHocTH BO,llbl (TITB) B CCCP (r.M. TayCiep (CCCP)) 131

The comparative measurements of sea-surface temperature in the U.S.S.R. (by G. M. Tauber (U.S.S.R.» 141

FOREWORD

At its Fifth Session, in 1968, the Commission for Maritime Meteorology conducted a series of scientific discussions based on a series of invited lectures on the measurement of sea-surface temperature and the uses of such measurements. The lecturers were selected in the light of their specialized knowledge in particular aspects of this question.

The table of contents of the present publication in itself indicates the wide and interesting range of topics covered. The discussions i~volved the many ways of measuring sea-surface temperature, including their respective advantages and disadvantages, the ways in which observations of sea temperature, both surface and sub-surface, may be used to in crease s~gn1ficantly the harvest from the sea, the persistency of ocean temperature patterns, and the uses of sea temperature measurements in long-range weather prediction.

In view of the great interest expressed in the various papers that were presented and to enable them to be read by scientists who were unable to attend the session, it has been decided to include these papers in the present Technical Note.

It should be mentioned that two of the lectures presented at the session have not been reproduced here since their substance has been, in most part, presented elsewhere. These lectures concerned "Numerical Synoptic Analysis of Sea-surface Temperaturell and the "Barbados Oceanographic and Meteorological Experimentl1 by P. M. Wolff and F. Ostapoff respectively.

I should like to take this opportunity of expressing my sincere appreciation to all of the experts who presented papers and especially to Dr. M. Rodewald (Federal Republic of Germany) whose untiring efforts and devotion in organizing the scientific discussions provided the main inspiration of this most successful programme. t:::>.~c. ••-

(D. A. DaVies) Secretary-General

VII

SUMMARY

Use of sea-surface temperature in long_range prediction The problems associated with complexly"coupled systems such as the oce~ and- the atmosphere are so difficult to solve that it is not yet possible to lay down a set of reliable forecasting rules, especially with regard to long-range forecasting, where the response time of one medium to the other is not known. In spite of these gaps in knowledge there seems to be sufficient order in large-scale air-sea interaction over periods of months and seasons, so that some qualitative use of this information can be helpful in prediction. Dr. Namias, in his paper, suggests some large-scale and cumulative air-sea interactions which seem to have influenced wind and wave patterns over months and seasons during the 1960s. While some of these concepts ~re copsid~red in the routine preparation of five-day~ monthly and season~ forec~sts, Dr. Namiasshows that, they are used mainly to support or negate other indications derived purely from meteorological data.

Collection and dissemination of sea-surface temperature data for the north-west Pacific, and their utirization for fisheries Dr. !mai devotes some time to an explanation of the sources, quantity andqua4ity of marine env~ronmental information available to Japan. From this information, a ten-,day surmnary of the sea-surface temperature for one-degree squares is prepared. In additiop, charts of surface currents, underwater temperatures and other elements are prepared on a monthly basis and disseminated by facsimile. Dr. Imai shows that an evaluation of.,this in~ formation has greatly helped t'o explain the cool swnm'ers which occasionally frequent Japan~' Dr. Imai also introduces some examples to show how the" Japanese fishermen utilize this in formation for their own special purposes."

Sea-temperature structure and its relation to the United States tuna fisheries 'in" the' eastern Pacific Ocean The problem that has plagued the United States tuna-fishing industry has been locating the fish. Fluctuations in distribution have varied greatly; and because of the large areas involved, locating fish concentrations has become a major problem. The nature and degree of movement of fish are highly variable from year to year, and Dr. Flittner relates these variations to changes in oceanography conditions; namely, the cycle in the changing pattern of sea-surface temperature." Dr. Flittner also stresses the other features of which the fisherman must have knowledge: the vertical structure of the upper layer of the sea. When the upper layers are relatively cold, the fish will not dive and are easily caught. However, when the thermocline is deep, the fish have an environment conducive to eScape from the nets used in pursuing operations. Dr. Flittner then emphasizes the point that there is a close relation between the actual water temperature and where the fish are found, and the fishing industry would benefit greatly from having this kind of sea informa tion placed at its disposal, both observed and forecast. Considerable attention is then devoted to the problems related to forecasting for the fishing industry and the challenges that lay ahead.

Sea-surface temperature patterns in the north-east Atlantic In his presentation, Captain Booth outlines some of the problems which the Royal Navy has encountered in the routine synoptic analysis of sea-surface temperature over parts of the north-east Atlantic. He mentions some of the characteristics and peculiarities of SST patterns which have been noticed; both with regard to persistency and with regard to change. VIII SUMMARY

Temperature variation at the near surface of the sea I Precise measurements of temperatures in the near surface layer of the sea were carried out at open sea where no disturbing effect from coast, ships or platforms is ob served; such measurements should permit to assess the value and representativeness in a synoptic sense of an observation and to define what is known as "sea-surface temperature". The temperature profile in the first five metres -- i.e. at" 0.1, 0.5, 1, 2, 3, 4 and 5 metres _ was investigated using a thermometer with an accuracy of O.02% .03°C. The difference Cd) in the temperatures between TO.I-T5 - Le. between 0.1 m and 5 m - obtained from 3,614 observations was as follows: 1,540 observations d $ O.l°e 253 observations 0.1 < d $ 0.2°e 428 observations 0.2 < d $ O.5°e 1,393 observations 0.5 < d Further, variation of vertical thermal structure with time (diurnal change) and the results of comparison of measurements using bucket and thermometer are discussed.

Sea~surface temperature; some instruments, methods and comparisons Mr. Crawford examines the meaning of sea-surface temperature, the ways in which it may be measured and the advantages and disadvantages of the various methods used. Mr. Crawford acknowledges the complexity of the problems involved, and admits that the choice ofa suitable instrument is dependent upon a number of factors, some of which are completely lUlrelated.

The comparative measurements of sea-surface temperature in the U.S.S.R. Prof. Dr. Tauber also considers the technical definition involved in measuring sea-surface temperatures. His paper deals with measurements made in different climate areas (Pacific Ocean and Indian Ocean) as well as the various conditions of the ship (drifting or lUlder way). A great variety of instruments were tested, and the paper deals in some great detail with the comparison of the readings obtained. IX

RESUME

Utilisation de 10 temperature de 10 mer en surface pour 10 prevision a longue echeance

Les problemes que posent des systemes interdependants complexes teis que l'ocean et l'atmosphere sont si difficiles a resoudre qu'il nlo pas encore ete pos sible d'etablir un ensemble de regles dignes de fo~ pour 10 prevision, et plus parti culierement pour 10 prevision a longue echeance, lorsqu'on ne connoit pas Ie temps de reaction dlun milieu par rapport a l'autre. Malgre ces lacunes dans nos connais sances, Ie schema de l'interaction air-mer qui S6 produit a grande echelle au cours de plusieurs mois et de saisons semble etre suffisamment regulier pour que lion puisse utiliser cette information qualitativement aux fins de prevision. Dans sa communica tion, M. Namias mentionne certaines grandes interactions cumulees de llair et de la mer qui, durant les annees 60, semblent avoir influe pendant des mois et des saisons sur Ie regime des vents et des vagues. Bien que certains de ces concepts soient pris en consideration lors de la preparation cour~nte des previsions pentadaires, mensuelles et saisonnieres, M. Namias demontre qulils servent surtout a confirmer au infirmer d'autres indications resultant uniquement de donnees meteorologiques.

Rassemblement at diffusion de donnees relatives a la temperature de la mer en surface dans Ie nord-ouest du Pacifique et utilisation de ces donnees pour la peche

.M. Imai explique assez longuement l'origine, la quantite et 10 qualite des renseignements dont on dispose au Japan sur Ie milieu marin. Sur la base de ces ren seignements, on prepare un etat recapitulatif decadaire de la temperature de la mer en surface pour des carres de un degre. En outre, on prepare et on transmet par fac simile des cartes mensuelles des courants superficiels, des temperatures en profondeur et d'autres elements. M. Imai indique que l'evaluation de ces donnees a largement contribue a expliquer l·origine des etes frais que connalt de temps a autre Ie Jopon. M. Imai mantre egalement, a l'aide de quelques exemples, comment les pecheurs japonais utilisent ces renseignements pour leurs propres activites.

Interet que presente la structure thermigue de la mer dans Ie Pacifique oriental pour les pecheurs de thon des Etats-Unis

Le principal -probleme auquel se heurtent les thoniers des Etats-Unis est Ie reperage des bancs de thons. En effet, leur distribution est tres variable et l'aire de peche tres voste. La nature et llampleur des deplacements de paissons varient considerablement dlune annee a llautre et, selon M. Flittner, ces variations sont fonction des modifications survenont dans les conditions oceanographiques, clest-a-dire dans Ie cycle des fluctuations de 10 temperature de 10 mer en surface. M. Flittner met egalement l'accent sur un autre phenomene que Ie pecheur doit connaitre : la struc ture verticale des couches superieures de la mer. Lorsque les couches superieures sont relativement froicles, Ie poisson ne plonge pas plus bas et peut done etre faci lement capture. Mais quand 10 thermocline se trouve plus bas, Ie thon en cherchant un milieu qui lui convient sera amene a descendre a des profondeurs qui Ie mettent x RESUME hors de portee des filets. M. Flittner fait voloir qu'il existe un rapport etroit entre Ie temperature de l'eau et l'emplacement des banes de poissons, et qulen conse quence I'industrie de 10 peche pourrait tirer un grand profit des donnees d'observa ticn et des previsions qui lui sont fournies a ce sujet. M. Flittner developpe ensuite longuement 10 question des previsions effectuees a I'intention des pecheurs et des difficultes qu'elles suscitent.

Regimes de 10 temperature de 10 mer en surface dans Ie nord-est de l'Atlantigue

Dans son introduction, Ie capitaine Booth traite de certains problemes auxquels slest heurtee 10 Royal Navy a propos de l'analyse synoptique CQurante de 10 temperature de 10 mer en surface dans certains secteurs du nord-est de IJAtlantique. II mentionne quelques-unes des caracteristiques et particularites observees dans les regimes de 10 temperature de 10 mer en surface, tant pour ce qui est de leur persis tance que de leur variabilite.

Variations de 10 temperatu.re dans 10 couche superficielle de 10 mer

On a mesure avec precision 10 temperature de l'eau dans 10 couche super ficielle de 10 mer au grand large, ou lion n~ constate aucun effet perturbateur dO a 10 cote, a des bateaux ou a des plates-formes; ces mesures devraient permettre de determiner 10 valeur et 10 representativite d1une observation, au sens synoptique du terme, et de definir ce que lion croit etre 10 IItemperature de 10 mer en surface". On a etabli Ie profil des temperatures dans 10 couche superficielle jusqu1a une pro fondeur de cinq metres - soit a 0,1, 0,5, 1,2, 3,4 et 5 metres - en effectuont les mesures a llaide d'un thermometre d1une precision de 0,02% ,030 C. La difference (d) qui existe entre les temperatures allant de TO,l a T5, soit entre 0,1 m et 5 fi, etablie d1apres 3 614 observations, est 10 suivante

1 540 observations d 5 O,loC 253 observations 0,1 < d .; 0,2°C 428 observations 0,2 '" d 5 0,5OC 1 393 observations 0,5 <: d

On examine ensuite 10 variation de 10 structure thermique verticale dans Ie temps (variation diurne) et les rssultats de comparaisons de mesures effectuees selon la methode du thermometre et selon celIe du seau.

Temperature de 10 mer en surface - Quelques instruments, methodes et comparaisons

--M. Crawfard etudie la signi fication de la temperature de la mer en surface, les diverses methodes de mesure possibles, ainsi que leurs avantages et inconvenients respectifs. II reconna!t la grande complexite des problemes en jeu et admet que Ie choix dlun instrument approprie depend de plusieurs facteurs, dont certains sont sans aucun rapport avec les autres. RESUME XI

Mesures camporees de 10 temperature de 10 mer en surface en U.R.S.S.

Le professeur Tauber etudie cussi 10 definition technique de 10 temperature de 10 mer en surface. Son rapport traite des mesures effectuees dans les zones cli matiques differentes (ocean Pacifique et ocean Indien) ainsi que des diverses con ditions dons lesquelles se trouve Ie navire (derivant au faisant route). Une grande variate d'instruments ant ete essoyes et Ie rapport compare de maniere tres detaillee les resultats des differentes lectures obtenues. XII

PE3IDME

McrrO~D30BaHHe TeMrrepaTYP~ rrOBepXHOCTH MOP~ B ~o~rOCpOqHOM ITpOrH03HpOBaH~B

rrpOo~eMH, CEHS8HHR8 C CHCT8MaMM, HaXO~H~HMHCH B C~O~HOM BSaHMO ~efiCTBHH , rro~o6Ho DKeaRy H aTMocwepe ,HaCTO~bKO TpY~Hopa3pemMMM, QTO B HaCToamee BpeME H8B03MomHO orrpe~enMTb Ha~e~H~e rrpaBHna rrporH03HpOBa HRR, ocooeHHO B QTROmeHHH ~onrOCpOqHOrO rrporH03MpOBaHMR, r~e H8R3B8CTHa ~nHTenbHOCTb BnHHHHff O~HOH cpe~H Ha ~pyryro. HeCMOTpa Ha 3TH rrp~6e~H- B 3HaHHHx,nO-BH~HMOM~BHHBneHa ~OCTaTO~HaH saKOHOMepHOCTh B KpyuHoMaclliTao HOM B3aHMo~eHCTBMH aTMoc~epll R Mopa B TeqeHHe M8CHQHliX M C8S0HHllX rrepHO ~OB, B CER3H C q8M orrpe~eneHHoe KaQeCTB8HHoe MCTIOnb30BaHM8 STOM HH~OP Ma~HH BrrpOPH03HpOBaHHHMomeT OMTb rrone3HliM. ~-p HaMHac B CBoell CTaTbe B@~BHraeT rrpe~xo~eHHe 0 Ha~H~HH orrpeAe~eHHoro KpyrrHoMacillTa6Horo H KyMy llHTHBHoro B3aHMo~eHCTBHH aTMoc¢ep@ H MOpH, KOTOPO~ rrO-BH~MMOMY, oE3SHBa llO BKHHHH8 Ha xapaKTep HeTpa H BOllR B TeqeHHe MHorHX MeCHQeB H C8S0ROB B rneCTH~eCHT@e ro~@. B TO BpeMH KaK H8KOTop@e H3 STHX KOHQerr~HH pac CMaTpHBaroTcH rrpE 06@qHOH rro~rOTOBKe rrHTH~HeBH@X ,MeCflQHRX H ce30HH@X nporH030B, ~~P HaMMac YKa3@BaeT, qTO OHM HcrroKb3yIDTCH rRaBR@M 06pa30M C ~e~bID rrO~TBep~AeHHH HRH OTpH~aHHH APyrHx rrOKa3aT8ReH, KOTop@e orrpeAe KHIDTCR Ha OCROBe qHCTO MeTeopOAOrHqeCKHX AaHH@X.

C60p H ~acrrpOCTpaHeHMe AaHH@X 0 TeMnepaType nOB8pXHOCTH MOpE. B C8Bepo 3anaAHOH ~aCTH Tlixoro oK8aHa H HX HcrroRb30BaHHe ARH P@DOROBCTBa

~~p HMaH paCCMaTpHBaeT HCTOqHHKH,KORHqeCTBO H KaqeCTBO HM8romeH CH B RrrOHHH HHwopMa~HH 0 MOpCKOE cpeAe. Ha OCHOBe 3TOH HHwopMaQHH CO CTaBRaIDTcH 10-AH8BH@e CBOAKH TeMrrepaTyp@ rrOBepXHOCTM MOpH ARB KBaApa TOB co CTOPOHOH B 1°. KpoMe Toro, e~eMeCHqHO COCTaBRHIDTCH H pacrrpocTpa HHIDTCR C rrOMO~bID waKcHMHRe KapT@ ITOBepXHocTH@X TeqeHHM, T8MnepaTYP@ BO~@ ITOA nOBepxHocxbID H APyrHx 8R8MeHTOB. ~-p MMaH nOKa3@BaeT, qTO OQeH Ka 8TOtt HHwopMa~HH B 3HaqHT8KbHOM CTeneHH nOMorRa 06~HCHHTb,no~eMY Ke TO B RnoHRR HHorAa 6@BaeT XOKO~H@M. ~-p MMaH TaK~e npMBB~ H8CKORbKO npH MepOB Toro,KaK HITOHCKHe p@6aKH HcrrO~h3YIDT STy HHWOPM8QHID B CBOHX crre~H WHqeCKHX ~enHX.

CTpYKTypa TeMrrepaTyp@ MOpH M ee BRHHHHe Ha KOBnID TYH~a B Coe~HHeHH@X illTa Tax B BOCTOqHOM ~aCTH TMxoro oKeaHa

ITp06n8MOH, KOTopaH rrpeACTaB~HeT TPY~HOCTH ARB p@6HOH npOM@rnneH HOCTH,3aRHMaro~eHcH KOBneM TYH~a 'B Coe~HHeHHMX lliTaTax, HBnReTcR o6Hapy ~eHHe P@O@. B CBR3H C BeCbM8 3Ha~HTe~bHMMH Ko~e6aHHHMH B pacrrpOCTpaHeHHH p@o@ H B CBR3H C 06mHpHOCTbID p8HOHa Bonpoc oOHapy~eHHff KOH~eHTpaQHtt P@D@ CTax OCHOBHOH npo6xeMott. XapaKTep H HHT8HCHBHOCTb rrepe~BH~eHHR p@_ DB CHXbHO Kone6~eTCH B pa3XHQH@e ro~@, H ~-p ~nHTTHep CBH3@BaeT 8TH KO ReoaHHH C H3MeHeHHffMH B OKeaHOrpaWHqeCKHX ycnoBHRX, 8 HMeHHO C ~HKXHQ HOCTbID H3MeHeHHH TeMnepaTypti ITOBepXHOCTH Mopa. ~-p ~XHTTHep TaK~e rrOA qepKHBaeT ,qTO pM6aK AOR~eH HMeTb CBeAeHHH 0 BepTHKaKhHOH CTpyKType Bepx Hero CXoH MOpH. RorAa TeMnepaTypa BepXHRX cnoeB cpaBHHTeXhHO HH3Kaff , pHoa He orrycKaeTcH BHR3 IT ee cpaBHHTeXbHo xerKo ROBHTh. O~HaKO KorA8 TepMoKXHH HaxoAHTCff Ha 60XbillOH rnyOHH8, oKpy~aID~aH cpeAa crroco6cTByeT yxo~y pti6ti OT ceTeh, ncrro~b3yeMtlX rrpn pti6HOH ~OB~e. ~a~ee ~-p ~~nTTHep rrOAqepKHBaeT TOT W8KT,qTO HMeeTCH TeCHaH CBR3b Me~~y T8Mrr8paTypoH BOA@ H MeCToHaX.O~AeHH8M p@6@ H qTO p@60nOBHaH rrpOM@m~eHHOCTb nonyqHT ooxbmyID PE3lOME XIII rrOHbSY, eC~R OHa 6y~eT ffM8Tb B CBoeM pacrropH~eHHH 3TQT BH~ HH~opMaQHH o Mope K8K B BM~e Ha6EID~eHHMX ~aHHHX, TaK H B BH~e rrporH030B. 3Ha~H Te~bHoe BHHMaHH8 y~enHeTCH npo6neM8M, CBHS8HHMM C rrpO,HOS8MH ~nH pM6o nOBcTB8, M TPY~HOCTHMH , KOTOpM8 MoryT 6MTb BCTpe~eHM.

X08 T8MITepaTypM rrOBepxHocTH Mopa B caBBpo BOCTO~HOH ~aCTH ATnaHTHQeCKOrO GRe8Ha

B CEDeR CT8Tb8 K8ITRT8H BYT paCCM8TpHBaeT H8KOTOpM8 li3 rrpo6neM, C KOTOpMMH cTanKHyncH 6pHT8HCKHH B08HHO-MOpCKOH WnOT rrpR rrOBce~HeBHOM CHHOITTHQeCKOM 8HanHS8 T8MrrepaTypM ITOBepXHOCTH MOpE B paHOH8X C8Bepo BOCTOqHOH qaCTH ATnaHTHQeCKOrO CReaB8. OH OTMeQa8T H8KOTOpM8 xapaKTep RN8 qepTM H oco6eHHOCTH XOA8 T8MrrepaTypM rrOBepXHoCTH MOpH K8K B OTHom8HHH ee rrOCTOHHCTBa,TaK H B OTHoIDeHHH ee H3MeHqRBOCTR.

Ko~eoaHRff TeMnepaTypH \rro~rrOBepXHOCTHoro C~OH MOpH

TOQHR8 R3MepeHHfl TeMrrep8TYP B rro~rrOBepXHOCTHOM c~oe MOpB OH~R rrpoBeAeHM B OTKpHTOM Mope B MeCT8X, r~e He HaO~ID~aeTCH OTK~OHHID~ee B~H BRae rrooepe~bH ,Kopaoxefi H~R rr~aT¢opM; TaRRe H3MepeHRfl ~O~~HH rr03Bo~HTb rrpOM3BeCTH oQeHKY 3H8qeHHH H perrpe3eHTaTHBHOCTH,B CHHOrrTHQeCKOM CMNCRe, rrpOBe,n;eHHoro Haomo,n;eHI1H H onpe~eJIHTb "TeMrrepaTypy IIOBepXHOCTH MOpJI 11 • TeMrrepaTypH~H ITPOWH~b Ha r~YOHHe rrepBHX IIHTH MeTpOB, T.e. Ha r~yoHHe 0,1; 0,5; 1; 2; 4 H 5 MeTpoB Hcc~e~oBa~c~ C HcrrO~h30B8HHeM TepMoMeTpa 8; 0 C TO"lHOCTbIO 0,02 /0,03 e. PaSHHlja (d) B TeM paTypax MelK,I1y TO.1-75, T.e. MelK,n:y 0,1 M H 5 M, IIOJIYQeHH8B H3 HaOJIID,n;eHHL 3,614, OHJIa CJIe,n:YIOrn;eH : o 1,540 HaOJIID,n;eHHH d $ O.l e 0 253 HaOJIIO,1J,eHHH 0.1 < d ~ 0.2 e .< 0 428 HaOJIID~eHHH 0.2 d $ 0.5 e 1,393 H80~IO,1J,eHHfl 0.5 <: d

KpOMe Toro,paccM8TpHBaIOTcH KOJIeOaHHH BepTHK8JIhHOH TepMHQeCKOH CTpyKTypu BO BpeM8HH (CYTOQHNe H3MeHeHHH) H pe3YJIhT8TH cpaBH8HHH H3M8 P8HHH C HCITOJIh30B8HHeM Be~pa H TepMoMeTpa.

TeMITepaTypa rroBepxHocTH MOpH, rrpHooPH, M8TO,n;H H cpaBHeHHR

r-H Kpoyvop,n; paccMaTpHBaeT nOHHTffe TeMnepaTypH rrOBepxHoCTH Mopa, MeTo,1J,ll ee H3MepeHHH, 8 TaK~e rrpe~My~ecTB8 M He~OCTaTKM pa3~HQH~X Hcrro~h3yeMEx MeTO~OB. f-H Kpoy¢op~ rrpH3HaeT C~O~HOCTh npo6~eM M YKa3HB8 eT,QTO BHOOP rrO,1J,XO~H~HX rrplioopOB 3aBHCHT OT pH,n;a ¢aKTopOB, HeKOTopHe li3 KOTOpHX COBepmeHHO He CBH38HH ,n;pyr C ~pyrOM.

CpaBHHTe~hHHe H3Mep8HHH TeMrrep8TypN rrOBepXHOCTH MOpE B COOP

ITpoW. A-P Tay6ep TaK~e paCCMaTpHBaeT TeXHH~eCKoe orrpe,n;eJIeHHe H3MepeHHH TeMnepaTypH rrOBepXHOCTR MOpH. Ero ~oKllaA KacaeTCH H3MepeHHH, rrpoH3Be~eHHHx B pa3JIR~HHX KJIMMaTHQeCKHX paHoHax (THXMH H HHAUHCKUH OKea HH), a TaK~e rrpR P83~HQHNX COCTOJIHRHX CYAHa (B ,n;pefi¢e H~M B ABH~eHHH). EHllO HcrrHTaHO MHoro pa3~H~HNX rrpwoopOB,H B CTaTbe IIOAPOOHO paCCMaTpliBa IDTCH B cpaBHeHHH no~y~eHHNe rroKa3aHHH. XIV

RESUMEN

Utilizaci6n de las medidns de In temperatura del mar en 18 8uperficie para las predicciones a largo plazQ

L08 problemas que 8e plantean con respecto a los sistemas de interrelaci6n compleja, tales como los que existen entre e1 oceano y Ie atm6sfera, son tan diffciles de resolver que no ss pOBible establecer todavia un conjunto de reglaa de predicci6n seguraa, especialmente Quando se tratn de predicciones a largo plaza, en las que se deacDnoee e1 tiempo que tarda en reaccionar uno de 108 medias a In acci6n provocada par e1 otre. A pesar de eaa faita de informaci6n en la materia, pareee existir una continuidad suficiente en las interaccionesde gran esasia entre e1 aire y e1 mar, observadas a 10 largo de periodos mensuales y estacionales, para que se pueda uti lizar, cualitativa y eficazmente , parte de esa informaci6n para las predicciones. En au informe, e1 Dr. Namias meneiona algunas interaccionesacumulativas de gran enver gadura entre el aire y el mar que, durante los aBos 60, parecen haber ejercido, a 10 largo de meses e incluso de estaciones enteras, una influencia sabre el regimen de los vientos y de las olas. Aun cuando en la preparaci6n corriente de las predioeiones para per!odos de cinco dias, periodos menauales y perlodos estacionales se tienen en cuenta algunos de eaos fen6menos, el Dr. Namias demuestra que sirven sabrA todo para confirmar 0 invalidar otras indieaciones procedentes, pura y simplemente, de datos meteoro16gicos.

Com ilaci6n difusi6n de datos relativos a 1a tern eratura del mar en la su erficie 81 noreste del Pac fico y su utilizaci6n para la pesca

El Dr. Imai explica, de forma bastante extensa, el origen, la cantidad y la calidad de la inLormaci6n de que se dispone en el Jap6n sabre e1 media marino. Basado en esa informaci6n, se prepara un resumen~ que abarca un periodo de diez dias sobre la temperatura del mar en la superficie para cuadros de un grado. Ademas, se prepa ran mapas mensuales de las corrientes en superficie, asi comc de las temperaturas submarinas y de otros ele~entos, y esta informaci6n se difunde por medio de facsimiles. El Dr. Imai sefta1a que la evaluaci6n de esos datos ha contribuido extenssmente a ex plicar e1 origen de los estios frescos que algunas veces conoce e1 Jap6n. El Dr. 1mai cita, asimismo, algunos ejemplos para explicar de que forma los pescadores japoneses utilizan esa informaci6n para sus propias actividades.

Estructura termica del mar en 81 Pacifico oriental y su relaci6n con lOB pescadores de atun de los Estados Unidos

La industria de la pesce de atdn en los Estados Unidos de America tiene grandee dificultades para localizar los bancos de atun, debido a la distribuci6n muy variable de estos y a la amplitud de las zonas consideradas. La naturaleza y la ampli tud de los desplazamientos de los peces varian considerablemen"te de un afta para otro y, segUn e1 Dr. Flittner,estas variaciones son funoi9n de las modificsciones que in tervienen en las condiciones oceanograficas, es decir, el cicIo que se produce en los regimenes variables de 1a temperatura del mar en la superficie. Segdn el Dr. F1ittner, el pescador debe conocer tambien la estructura vertical de Is capa superior del mar. Cuando las capas superiores son relativamente frias, los peces no se inmergen profun damente y pueden ser, por ende, facilmente capturados. Pero cuando la termoclina es profunda, e1 atun se hal1ara en un media situado fuera del alcance de las redes. El Dr. Flittner hace observar que existe una relaci6n estrecha entre la temperatura del agua y e1 emplazamiento de los bancos de peces y que, por consiguiente, la industri~ RESUMEN

pesquera tendria gran interes en util~zar eBe tipo de datos, tanto 108 observados co mo los previstos. El Dr. Flittner explica, seguidamente y de forma exteoss, la cues ti6n de las predicciones destinadas a Is industria pesquera y las dificultades que dichas predicciones plantean. Regimenes de 1& temperatura del mar en 1& superficis, &1 noreste del Atlantica

En su exposioi6n, el oapitan Booth esboza algunos de los problemas oon los ouales tropieza la Royal Navy, en 10 que respeota al amHisis sin6ptioo oorriente de 1& temperatura del mar en superfioie en ciertos sectores del noreste del Atlantica. Manelona, Bsimismo, algunas de las caracter!stic&s y particularidades observadas en los regimenes de la temperatura del mar en la Buperficie, tanto en 10 que respects a eu persistencia como a au variabilidad.

Variaciones de la temperatura del mar en la capa superficial

Se ha medido con precisi6n Is temperatura de la capa superficial del oc~ano en alta mar, donde no se observa ningdn efecto perturbador debido a Is costa, a 108 buques 0 a las plata10rmasj esas medidaa deberlan permitir determinar, desde el puntc de vista sin6ptico, el valor y la representatividad de una observaci6n y definir 10 que se pianaa que puede ser la "temperatura del mar en la superficie". Se ha obtenido e1 perfil de las temperaturas a 10 largo de los 5 primeroB metros, as decir a 0,1, 0,5, 1, 2, 3, 4 Y 5 metros, efectuando las medidas par medio de un term6metro de una preoisi6n de 0,0200/0,03°0. La diferenoia(d) existente entre las temperaturas oom prendidas entre TO,1-T5, es deoir entre 0,1 m y 5 m,estableoida despues de 3.614 ob servaciones, fue la siguiente: o 1.540 observaciones d t" O,l C 253 observaciones 0,1 < d k 0,20 0 428 observaciones 0,2 < d t" 0,50 0 1.393 observaeiones 0,5 < d A continuaci6n, se examina Is variaei6n de Is estructura termieR vertical en el tiempo (variaci6n diurna) asl como los resultados de las comparaciones de las medidas efectuadas segdn el met OdD del term6metro y segUn el metodo del oubo.

Temperatura del mar en~la Buperficie Instrumentos, metodos Y eomparaciones

El Sr. Crawford estudia Is significaci6n de la temperatura del mar en la Buperficie, asi como los diversos m~todos de medida posibles y sus ventajas e ineon venientes respectivos. ~eeonoce que eS08 metodo8 plantean grandee problemas y que la seleeci6n de in8trument~s adecuado8 es funei6n de multip~es factores, ~lgunos de los cuales no guardan relaci6n alguna entre st.

Medidas comparadas de la temperatura del mar en la superficial en la URSS

El Profesor Tauber tambien estudia la definiei6n teeniea de la temperatura del mar en la superfieie. Su informe trata de las medidas efeetuadas en zonas elima tieas diferentes (oceano PacIfico y oc~ano Indieo), as! como de las condiciones diver sas en que se halla e1 buque (derivando 0 nav~gando). Se ha ensayado una gran varie dad de instrumentos y e1 informe compara, de forma muy detallada, los resultados de las diferentes leeturas 0 indicaciones obtenidas.

USE OF SEA-SURFACE TEMPERATURE IN LONG-RANGE PREDICTION

Jerome Namias ESSA - Weather Bureau

1. Introduction

The problems associated with complexly-coupled systems like ocean and atmosphere are so difficult to solve that it is not yet possible to lay down a set of reliable forecasting rules. This is particularly true in extended and long-range forecasting where the response time of one medium to the other is not known. In spite of these gaps in knowledge there seems to be sufficient order in large- scale air- sea interactions over periods of months and seasons, so that some qualitative use of this information can be helpful in prediction.

While the work to which I shall refer is mainly connected with the long-range forecast program of the U. S. Weather Bureau, it draws upon pioneering work by many other workers, past and present, who have carried on basic studies in air-sea interaction. Among the most notable of these are Helland-Hansen and Nansen, Jacobs, Rodewald, Bjerknes, and more recently scientists at the Scripps and Woods Hole Institutes of Oceanography, the Bureau of Commercial Fisheries, and the U. S.

Naval Fleet Weather Facility at Monterey, California. 2 SEA-SURFACE TEMPERATURE

Some of the difficulty for application of air-sea interactions to long- range forecasting sterns from the fact that the interactions may differ in kind from season to season and from area to area. Expressed another way, the stress placed on physical parameters may vary between seasons and between areas. When the physics of these factors is properly understood so as to be taken into account by numerical prognosis, the need for the climatological stratification as a partial basis for prediction, as dis cus sed herein, will be eliminated. In my talk I shall suggest some large- scale and cumulative air- sea interactions which seem to have influenced wind and weather patterns over months and seasons during the 1960 1 s. While some of these concepts are considered in the routine preparation of 5-day, monthly and seasonal forecasts, it must be stressed that they are used mainly to support or negate other indications derived purely from meteorological data.

II. The Joint Interactions which Appear to Produce Sea-Surface Temperature Variations

As is well known, the cyclones and anticyclones of the weather map are embedded in upper-level tropospheric wind systems of a wavy character whose wavelength is of the order of 3,000 to 5,000 miles. These long waves are interactive with the cyclone waves along the polar front.

Even if one averages the upper-level or sea-level pressure and wind distributions over periods of weeks, months, or seasons, the long waves do not disappear. They assume positions and amplitudes which vary USE IN LONG-RANGE PREDICTION 3 fro:m one period to another, and thus the prevailing air :masses, stor:m tracks, and associated weather conditions :may be inferred frOlTI the

:mean configuration of the long waves.

The :mean pressure distribution -- let us say at sea level -- gives a

good approxi:mation of the resultant wind strea:ms for the period. Further

:more, the :mean wind can be considered as being co:mposed of a nor:mal

co:mponent and an ano:malous co:mponent. The latter co:mponent can be

calculated fro:m isopleths of :mean pressure departures fro:m the long ter:m nor:mal. Ano:malous co:mponents of the' wind produce an ano:malous drag

on the surface water and force an ano:malous Ek:man drift. The southerly

co:mponents usually i:mply transport of war:m air northward and less than nor:mal latent and sensible heat transfer fro:m the cooler underlying water, while the reverse is true for the northerly co:mponents. It is not surprising, therefore, that in the observed :monthly :mean sea- surface te:mperature

distributions, war:mer-than-nor:mal surface water is found east of low

pressure syste:ms and cooler-than-nor:mal water to their west. Cooler

than-nor:mal water is also associated in part with upwelling near the center

of the negative :mean pressure ano:maly where horizontal divergence of

surface water takes place. The preceding concepts have been incorporated

into a :methodology for nu:merically e sti:mating what the sea- surface te:m

perature distribution is likely to be with a given :mean pressure distribu

tion whose effects are superi:mposed upon an initially abnor:mal sea-surface

te:mperature pattern (Na:mias, 1959, 1965; Arthur, 1966; Jacob, 1967). 11 SEA-SURFACE TEMPERATURE

To predict the sea- surface temperatures a month ahead with the help of these concepts, some· first approximations are now being made by using predicted mean sea-level pressure distributions. The accuracy of these predictions leaves much to be desired because it depends not only upon the atmospheric pressure prediction, but also upon the unsolved problems associated with air-sea interaction including the response time of water masses to wind drag and other factors. Monthly estimates of sea-surface temperatures centered on forecast day are now made routinely, and attempts are being made to improve the method by incorporation of terms not previously considered.

It might appear from examples showing the specification of sea- surface temperature by atmospheric pressure patterns (not reproduced) that the sea is simply a slave to the atmosphere. However, since anomalous water of the same sign temperatures/cover very large areas often as much as one~third of·the

North Pacific ocean, and frequently extend to depths of 200 meters or more, these large reservoirs of anomalous heat (or cold) frequently last so long that they may force anomalous atmospheric wind and weather patterns in subsequent months after their formation. Thus the difference in time constant of the ocean and the atmosphere, and the response time of one medium to another become highly germane to the problem of long-range weather forecasting.

Some specific cases where the sea appears to have had a distinct and strong influence on atmospheric circulation will be described below. USE IN LONG-RANGE PREDICTION 5

Only a few illustrations will be provided with this report since more detailed papers have been, or are to be published. The reader will be referred to these.

III. General Concepts Useful in Long-Range Weather Forecasting

Perhaps the most obvious use of sea- surface temperatures in long range forecasting involves prediction for litoral areas. If the surface winds flow from sea to land more frequently than normal, the overlying air masses will naturally be affected by sea- surface temperature anomalies.

This influence is especially important along the west coast of the U. S. , where the prevailing winds are onshore, and occasionally along the east coast. Not only will the mean temperature be affected at coastal points but also the frequency of stratus and amounts of rainfall. The frequency of fog and stratus along the west coast of the Ui::tited States during the warm season is frequently associated with the coldness of the water which in turn is related to the extent of coastal upwelling. The coastal upwelling is in turn related to the position and strength of the Pacific anticyclone.

Precipitation along the west coast - - particularly along its southern half

is also related to the water temperatures in the sense that warm offshore waters. given the proper vertical motion-producing systems in the atmosphere, are apt to produce substantial rains. This situation occurred

during the Falls of 1965 and 1967. An example is shown in Fig.!. Simi

larly, during the warm season off the east coast of the United States, warm water may augment the supply of water vapor in southerly air 6 SEA-SURFACE TEMPERATURE streams and increase the rainfall. Or it may greatly enhance the precipitation from tropical storms as was the case in hurricanes Connie and in 1955. . Diane / On the other hand, cold water along the Atlantic shelf is apt to be associated with drought along the Atlantic seaboard, as in the case of the springs and summers of the 1962-66 period. In this case the cold shelf water may have assisted in developing a baroclinic atmospheric zone east of its normal position and thus steer cyclones sufficiently off the coast so as to reduce coastal rainfall and induce northwesterly sub- siding flow over the land areas. Similar relationships of precipitation to sea- surface temperatures have been found off southern Australia by Priestley (1966).

Surface temperatures in the open ocean areas are also important both for generating or suppressing storm systems, and otherwise affecting the general circulation. Perhaps the largest-scale influence of this type has been described by J. Bjerknes (1968), who related sea-surface temperatures in tropical waters of the Pacific to the variation in the extent and the intensity of the subtropical Pacific anticyclonic belt and the westerlies of temperate latitudes. Bjerknes presented evidence of these linkages by showing that equatorial rainfall is positively correlated with the underlying sea- surface temperature. He further reasoned that the resulting variable latent heat of condensation produces variations in the strength of the Hadley cell which then transmits its influence to the temperate latitudes through angular momentum transport. Assuming that USE IN LONG-RANGE PREDICTION 7 there is SOlne lag atnong these processes the forecasting itnplication of

Bjerknes ' work is that high equatorial sea-surface tetnperatures in late fall in the central Pacific are clues to strong subtropical highs and strong tetnperate latitude westerlies in the subsequent winter. An exatnple, taken frotn Bjerknes' work is shown in Figs. 2 and 3.

Of course, the use of this concept requires adequate sea-surface tetnperature data in equatorial regions -- data which, itnportant as they tnay be, are not presently abundant nor easy to assetnble. Hopefully, the

World Weather Watch progratn will tnake it possible to collect and transtnit

such infortnation in titne to be of practical use.

Large-scale sea-surface tetnperature abnortnalities in tetnperate

latitudes of the North Atlantic and North Pacific tnay stabilize long-tertn

circulation patterns by leading to preferential areas for cyclogenesis and

anticyclogenesis. Perhaps the tnost striking exatnples of this kind are

the explosive cyclogeneses which have occurred frequently since 1960

over anotnalous extensive wartn pools of water in the central North

Pacific. The enhanced cyclogenesis over these pools appears to be

especially pronounced during winter, for at this season several re-enforcing

processes operate -- rapid diabatic heating of cold continental air tnasses

flowing over the wartn pools, increased release of latent heat and increased

vertical destabilization of the overflowing air tnasses. Such recurring

cyclogenesis during the winters of the 1960's has resulted in pronounced

southward displacetnent of the Aleutian low. The upper troughs associated 8 SEA-SURFACE TEMPERATURE

with these disturbances frequently established a downstream wave pattern

favoring a ridge along the west coast of North America and a trough along

the Continental Divide. This pattern of general circulation has indeed

been sufficiently recurrent during the 1960's so as to cause a short period

climatic flucuation in the United States whereby the winters have been

cold east of the Continental Divide but warm west thereof. The lag effect

between sea and atmosphere appears to be sufficiently long (at least of

the order of a season) that a clue to winter circulation may often be obtained

from the sea-surface temperature pattern during fall. An example is

shown in Fig. 4.

Sea- surface temperature anomalies generated in summer can also

assist in predicting North Pacific wind and weather systems along the path

that cyclones gene.raUy take in the subsequent fall. Thus, the presence or

absence of Gulf of Alaska lows and their intensity in fall seem to be related

to the sea-surface temperatures developed during the antecedent summer.

If these sea temperatures are much warmer than normal, the cyclones will

. develop very rapidly and generate strong east Aleutian or Gulf of Alaska

cyclones. Cold water along the same route can result in weak lows or,

in special case s, anticyclone development. The relationship is illustrated

in Fig. 5. Summertime sea-surface temperatures in the North Pacific

depend upon the atmospheric circulation and the prevailing air mass

characteristics in a manner briefly discussed. Here again, the long

range forecaster recognizes characteristic wind and weather patterns USE IN LONG-RANGE PREDICTION 9

over North America which usually accompany prescribed circulations in the Pacific. These "teleconnections" are now reasonably well-defined by

objective tools.

A special type of interaction may be involved in phenomena involving

the extent to which the subtropical highs become abnormally strong or

abnormally weak during Autumn over both the North Pacific and North

Atlantic. Cold pools of water generated during summer and early fall

have a tendency to be followed by anticyclogenesis over them in fall,

while antecedent warm pools lead to weak anticyclones. This response

may be associated with variable wind drag. For example, a stable friction

layer in the atmosphere over a cold water pool inhibits frictional outflow

from the anticyclones and they tend to develop more strongly than otherwise.

This process seems to occur most frequently during Autumn, for it is at

this time when the subpolar lows vie for domination with the subtropical

highs. The buffer zone between subpolar lows and subtropical highs is

usually the area subject to anticyclogenesis due to underlying cold water.

As with other air- sea interactions this anticyclogenetic effect will not

occur if dynamical influences in the atmosphere from remote areas are

strongly unfavorable for anticyclogenesis. An example is shown in Figs.

6 and 7.

There are many other examples where sea- surface temperatures may

be of help in long-range forecasting. Increased baroclinicity is generally

established rapidly in the atmosphere over regions where strong oceanic 10 SEA-SURFACE TEMPERATURE

temperature gradients exist, and this baroclinicity usually leads to

increased cyclogenesis. Then again, the frequency and development of

tropical storms and hurricanes seems to be partly a function of the over

lying sea-surface temperatures.

It seems to be axiomatic that sea- surface variations will be found

to be highly necessary input for successful long-range numerical

predictions. Until these complex interactions and processes are sufficiently

understood, long-range forecasters would be well-advised to keep up-to

date analyses of sea-surface temperatures over as large portion of the

world as pos sible, and to consider these in the preparation of their

forecasts.

Meanwhile, a beginning is being made at several research centers to incorporate some air-sea interactions into numerical models.

REFERENCES

Arthur, Robert S., 1966: Estimation of mean monthly anomalies of

sea-surface temperature. J. Geophys. Res., l!.' 2689-2690.

Bjerknes, J., 1968: Further studies on large-scale interaction of

the atmosphere and the oceans. (To be published in Mon. Wea. Rev. )

Jacob, W. J., 1967: Numerical semiprediction of monthly mean

surface temperature. J. Geophys. Res., I!:.., 1681-1689. USE IN LONG-RANGE PREDICTION 11

Namias, J., 1959: Recent seasonal interactions between North Pacific

waters and the overlying atmospheric circulation. J. Geophys. Res.,

64, 631-646.

Namias, J., 1965: Macroscopic as sociation between mean monthly

sea-surface temperature and the overlying Winds. J. Geophys. Res.,

70, 2307-2318.

Namias, J., 1968a: The labile Gulf of Alaska cyclone - - Key to large-

scale weather modification elsewhere. Proc. International Conf.

on Cloud Phys., Aug. 26-30, 1968, Toronto, Canada, 735-743.

Namias, J., 1968b: Long-range forecasting of the atmosphere and its

oceanic boundary - - an interdisciplinary problem. (To be published

in CalCOFI Rept. V. XII. )

Namias, J., 1968c: Seasonal interactions between the North Pacific

Ocean and the atmosphere during the 1960's. (To be published in

Mon. Wea. Rev.)

Namias, J., 1968d: Autumnal variations in the North Pacific and North

Atlantic anticyclones as manifestations of air-sea interactions. (To

be published in Deep Sea Research. )

Priestley, C.H. B., 1966: Droughts and wet periods and their association

with sea-surface temperature. Aust. Jour. Sci., 29, 56-57. f-' [\)

I W'" ! ~ ;g I

Figure 1 _ Mean mid-tropospheric and sea-surface conditions preceding unusually heavy and early rains in South ern California in late November and early December, 1967. Shading indicates antecedent (October) sea surface temperature anomalies, solid lines are 700-mb. contours, and broken lines are IO-day height tendencies of the 3D-day means centered at the beginning of November, (5 days of tendency are based on numerical prognoses). Heavy arrow shows that upper-level trough is progressing eastward to coast over abnormally warm water surface. (From Namias, 1968b). USE IN LONG-RANGE PREDICTION 13

FI---2='!L--+---..!i~-+--'-"-'!b.--!---'-~L-+---'-"-"-='--1C1950 1951 1952 1953 1954 86 30~ ," I~ CANTON tSL " ,. I \i \ ,.... . 2"48'5 17I°43'W / V V I r".. J B4·'f------,,\ j ~="""""v. 2B9' ... : \ " "'I~ l \. ) ..... -+ --1-=-....:..-_" B2' /'" -- 1-1---'---- - 278'"

m 2 o'&I-----.J.-----.J.-----f--rl---f----, 200'

100 --+------< 100'

".. ~\\.- ../··l'~\:;.l.~\·c..., ...... '/ I--'-"-+--'-----+-----j---"I /-----1 27.B·

mm ------j----+------1 400

3001------+----+---1 ----+------1 300

200f----+----+--~ h------1------1 200

100 1-----1fn-c=----lloo

1961 1962 1963 1964 1965 1966 F C B6 30'

Alit ,. .. \ B4 " -I '\--"... 7~KL~/\-..'7-, \ ~/~~. .(L\~~ J ~J \\: \..... "'.....r f \f " .,.,f\1'1 B2.... -V ~.r. 27.B' mm mm 500 500

400 400

300 300

200 200

1001--- 1- 100

h..J L- 0 fh ILnn .rJth UThn Ku o S08Qmm 4016 7125 519.4 I 14328 Figure 2 - Time series of monthly air and sea temperatures and of monthly precipitation at Canton Island from 1950 to 1966 (from Bjerknes, 1968). Note heavier precipitation when sea is warm, at which time air temperature is below sea temperature. f-' 100' 120· 140· 180" 1"'0· 120· 100' OJ' "'. 40' ''''' ''''' -l="

~ I i g.J ~ ;g ::0 j

I~- ,~u ,~ ,~ ~ ~- I~ IW ~ ~ ~ ~ Figure 3 - January 1964 distribution of pressure (mb) at sea level. Change from January 1963

in dashed isallobars (after Bjerknes J 1968). This strong Pacific High was asso ciated with warm equatorial water temperatures (see Figure 2). USE IN LONG-RANGE PREDICTION 15

I~-+'-+~'O I" +3G , \ , '--' i-->-r--'k,-3o\ ..... __ ... I 1170 I \ I \ / MEAN SST ON/ WINTER 1967.68

Figure 4 - Top: Mean sea-surface temperature anomalies in fall 1967 (dashed 180 pleths, centers labelled in OF) and subsequent winter's sea-level pressure

anomalies (solid isopleths J centers in mbs.). Bottom: Sea-surface temperature anomalies in winter 1967. Note large negative pressure anomalies over warm central Pacific pool which existed in fall and was only slightly modified during subsequent winter. (From Namias, 1968c). I-' 0\

2.5- -10 2.5- -10 2.0- ISummer SST 2.0--:- -8 J8 Fall SST 1.5- 1/\ /-6 1.5- in GUlf,\ -6

1.0, -4 1.0 \ -4 II ·w \ » '"I of 0.5), -2 mbs of 0.5-\ mbs w \ :i1 / ~ ..... fi---- 0- \ I 0- \ I ~ \ -0 \ """\-0 \ I I \ \ \ ~ \ I \ . -0.5- \--2 \ I \':--2 .'U \ I \ -0.5- . IoJ \ I \ ::d ,. l \ » \ \ '~Fall -1.0- ,'Fall SLP ...)-4 \ SLP \ \ -1.0- I --4 ~ I in Gulf \I \I in Gulf -1.5- " --6 -1.5- \/ --6 IIIII I, III ! I II '61 '62 '63 '64 '65 '66 '67 '61 '62 '63 '64 '65 '66 '67 YEAR YEAR

Figure 5 - Showing the dependence of fall pressure in the Gulf of Alaska on the antecedent summer's sea surface temperature. The index for the intensity of the Gulf Low is the average pressure anomaly between l600 wand l300 W along 45°N. The sea-surface temperature index is the average anomaly from 45°N to 50 0 N and from l600 E to l30oW. (From Namias, 1968a). C t:J:l t c:, '" '\ "0 ~ OJ \ til r::t:J /.\ f-3 d \ , z \ ,,

Figure 6 - Isopleths of sea-surface temperature anomaly for October 1967. Shaded areas show below-normal temperatures. Centers labelled in of. (From Namias, 1968d). ~ ~

~ I ~o ! ~ ~ j

/ . / ,/ "VI .. ISOBARS FOR NOV 1967 1ll:: AND CHANGE FROM OCT

Figure 7 - Sea-level isobars for November 1967 (solid lines labelled in mb) and change from October 1967 (dashed, centers in mb). Note marked positive change over October major cold-water pools. (From Namias, 1968d). COLLECTION AND DISSEMINATION OF SEA-SURFACE TEMPERATURE DATA FOR THE NORTH-WEST PACIFIC, AND THEIR UTILIZATION FOR FISHERIES

(Scientific lecture at CMM-V)

Dr. Ichiro Imai

Marine Division Japan Meteorological Agency

1. Sea-Surface Temperature Observation

Sea-surface temperature data in the north-west Pacific are collected by the Japan Meteorological Agency (JMA) through reports from research vessels, merchant ships and fishing boats.

Among them, the data from research vessels are generally the most reliable. An average of ten research vessels daily cover this area for various purposes, such as oceanographic surveys, meteorological obser vations and research on fisheries. However, the sea-surface temperature data from these sources are limited in number.

The number of sea-surface temperature data available from fishing boats are more than one hundred per day. Their accuracy is relatively high, but their coverage is usually limited to areas near or within the fishing grounds.

Most of the sea-surface temperature data are collected through weather reports transmitted from merchant ships. About 500 to 600 reports are sent daily from various areas of the north-west Pacific.

The sea-surface temperature observations aboard research vessels are usually made by using an electric resistance thermometer or by the bucket method with mercury thermometer. The accuracy of observation is usually within O.loC. 20 SEA-SURFACE TEMPERATURE

Merchant ships and fishing boats use one of the following methods:

(1) The bucket method with alcohol or mercury thermometer;

(2) The condenser intake method;

(3) The electric resistance thermometer.

Although alcohol thermometers are less reliable (with maximum error about O.SOC), they are still used widely. JMA recommends the use of mercury thermometers with certification instead of alcohol thermometers. Recently, the condenser-intake method has been adopted by many merchant ships, but the observations by this method are not always satisfactory because the sensor is set at a certain depth beneath the sea surface, where the sea-water temperature differs in most cases from that at the surface. The intake valve is often set several metres beneath the sea surface.

Figure 1 shows the distribution by area of sea-surface temperature observations during the first ten days of June 1968. Numerals in the figure indicate the number of observations in each one-degree mesh. The data are concentrated around two ocean weather stations, Victor (34Drr, 164°E) and Tango (29°N, 13S 0 E), along regular routes of vessels, and near fishing areas. The sparse-data area is still. fairly large, and our urgent problem is how to cover the area with reliable data.

2. Collection and Dissemination of Data

The Marine Division of JMA collects the sea-surface temperature data from the ships' weather reports from various kinds of ships. They are summarized every ten days for everyone-degree mesh in the north- west Pacific west of the 1800 meridian. The sea-surface temperature chart is thus prepared every ten days. The charts of surface currents, under sea water temperatures (SO m depth, 100 m depth, etc.) and other elements are also prepared monthly by use of the data from research vessels. Such charts are combined into a single sheet and disseminated by facsimile broadcast six times a month: the ten-day mean sea-surface temperature COLLECTION OF DATA FOR N.W. PACIFIC 21

chart for the north-west Pacific is broadcast on the 4th, 14th and 24th of every month, and the hydrographic conditions in the seas surrounding Japan,in a chart, including the most recent data, on the 9th, 19th and 29th of every month. They are widely utilized by the fishermen, meteor ologists and otherg.

Figure 2 shows an example of the original copy for facsimile broad cast of the ten-day mean sea-surface temperature chart for the period of 1 - 10 June 1967, and Figure 3 shows that of the detailed patterns of sea-surface temperature as well as the surface current in the seas sur rounding Japan for the same period. These charts are also printed three times a month under the title of "The Ten-Day Marine Report", and are distributed to users in Japan and in other countries. This report was first issued in 1948, and tbe facsimile broadcast began in 1958. The statistics of sea-surface temperature for this purpose were first compiled manually, until the computer was introduced in 1966.

On the other hand, the monthly statistics of sea-surface tempera ture are prepared, covering almost the whole area of the north Pacific 00 0 .0) . ( o - 52 N, 110 E- 110 W, based upon the formal meteorologlcal logs from selected ships. The mesh size of two degrees in latitude' and five degrees in longitude is used for the statistics. These marine climatological sta tistics in Japan were first initiated by the Kobe Marine Observatory in 191~and the sea-surface temperature tables were published annually by the observatory until 1941, when the Marine Division of JMA took over the work. Since then, the tables have been published as a part of a JMA publi cation, "The Marine Climatological Tables of the North Pacific Ocean".

Besides, the Nagasaki Marine Observatory has been publishing "The Ten-Day Marine Report of the Seas West of Japan" since 1948. This report includes the sea-surface temperature charts, the pattern of water masses and other information concerning oceanographic conditions in the East China Sea. An example of this report is shown in Figure 4. A similar report for the Japan Sea has been published by the Maizuru Marine Obser vatory since 1954. These two kinds of report are issued for the purpose of local oceanographic services. 22 SEA-SURFACE TEMPERATURE

Figure 5 shows another special type of sea-surface temperature infor mation rendered by JMA. Since 1958, the agency has prepared the charts of detailed sea-surface temperature patterns around the fishing area of the Japanese salmon-fishing fleets, in the north Pacific and in the Bering Sea, every five days during the summer fishing season, and disseminated them by facsimile broadcast for those fleets.

3. Sea-Surface Temperature Forecasting Oceanographic studies in Japan were first prompted by the necessity of forecasting cool summers in the northern part of Japan that cause sev ere damage to the rice harvest. Such summer coolness was generally thought to be related to the low sea-surface temperature off the east coast of northern Japan. The relation was outstanding in the case of the abnormal ly cool summer in 1934 (Figure 6), and the theory had been accepted until recently.

In 1935, the Ministry of Agriculture and Forestry began to encourage the oceanographic survey and forecasting in the Pacific Ocean adjacent to northern Japan. They supposed that the_oceanographic survey from winter to spring would enable us to forecast sea-surface temperature for the fol lowing summer season, and that the forecasting of cool summers would be realized. The oceanographic survey in Japan started under these circum stances.

Of course, long-range weather forecasting has been one of the im portant services of JMA, so the oceanographic work in the agency has been carried out also in this connexion.

At present, however, our knowledge has been greatly improved, and it is now generally believed that cool summers are related to the large-scale patterns of the oceanic as well as atmospheric conditions rather than local variations of sea-surface temperature patterns cff the coast of northern Japan. It is further considered that the accuracy of long- range forecast, either of oceanic or atmospheric conditions, will surely be improved by taking large-scale air-sea interaction into account. The patterns of heat, mcisture and momentum transfer through the air-sea interface must play an important role. COLLECTION OF DATA FOR N.W. PACIFIC 23

Since 1951, JMA has been making forecasts of the sea-surface temperature for the coming summer season in March every year for the seas ad jacent to Japan. Under present conditions, the forecasting is merely based on statistical and synoptic methods. Analogies, periodicities and correlations among various meteorological and oceanographic variables are still useful tools.

The observational network is still too poor for a grasp of oceanic conditions; also, the physical rules of their variations are quite un known. So we must aim at improving the forecast by reinforcing the obser vational network and by developing more objective and physical methods of analysis.

With the aid of computers, we are now trying a multiple correlation method in forecasting the sea-surface temperature of specified areas by screening the correlated quantities objectively from many possible pre dictors. Figure 7 shows an example of the sea-surface temperature fore cast by this method using five predictors. The dashed lines indicate the observed isotherms, and the solid lines those predicted six months before. The agreement seems generally good in this case. Figure 8 shows the prediction areas used in the method, and the mean sea-surface temper ature is predicted for each of these fourteen areas. Of course, the final forecast is determined by comparing and adjusting many possible results of analysis made by various methods. The specialists of the Marine Division of JMA and its four marine observatories meet in March every year for this purpose.

Besides, forecasts of oceanic conditions for the fishing grounds are made by six regional fisheries research laborat'ories of the Ministry of Agriculture and Forestry. They issue short-range forecasts (about one a month or so) every two weeks, and also long-range forecasts. These forecasts are mainly based on statistical methods by the analysis of actual conditions. 24 SEA-SURFACE TEMPERATURE

4. Utilization of Sea-Surface Temperature Data for Fisheries As is well known, there are close relationships between the oceanic and fishing conditions. The Tohoku Regional Fisheries Research Laboratory disseminates charts showing the distribution of observed sea-surface tem~ perature and of catch of fish every five days. Figure 9 is an example of the composite map showing the sea-surface temperature distribution and the catch of fish, and Figure 10 shows the successive variations of five-day mean sea-surface temperature and of the catch of fish. These maps clearly show that sea-surface temperature variations are well followed by trans fer of fishing grounds and variation of catches of bonitoes and albacores.

Thus, the predictive location of fishing grounds seems to follow the forecast of cceanic conditions. Some organizations concerning fisheries disseminate the fishing-ground forecast based on the actual pattern of sea-surface temperature and.of fishing conditions (catch of fish, etc.).

The Ocean Research Institute of the Tokai University issues the "Albacore Fishing-Ground Forecast" every ten days. This forecast is based on the current data of sea-surface temperature and current received from JMA and the Japan Maritime Safety Agency, and also on the data of fishing conditions from the Fisheries Experimental Stations and others. The fore cast is also disseminated by facsimile broadcast through the Japan Fisher ies Radio Station. An example is shown in Figure 11 (translated into English) • In Japan, about 1,000 fishing boats and 1,500 merchant ships are now equipped with facsimile receivers. As well as daily weather maps and maps of actual oceanic conditions, the fishing boats can receive various fore casts at every issue to check their fishing plan.

We have another case of sea-surface temperature data utilization. Summertime salmon-fishing fleets are organized every year for the north wes~ Pacific grounds. Before sailing, they try to locate the fishing areas by the analysis of oceanographic data including sea-surface temper ature, and then send a research ship to the located areas for test fishing and sea-surface temperature observation. During the fishing activities, each boat reports ·sea-surface temperature data to its mother ship, COLLECTION OF DATA FOR N.W. PACIFIC 25 which checks the fishing plan from time to time by compiling these reports and through reception of the facsimile- brcadcast of ooeanic conditions sent by JMA.

These fishing areas in the north-west Pacific are famous for the dense sea fog in summer, so the weather forecast is oertainly necessary. A couple of forecasters are usually on board the mother ship and provide her fleet with on-the-spot weather forecasts. Of course, sea-surface temperature data are most important for sea-fog forecasts.

In winter, on the other hand, JMA disseminates sea-surface temper ature oharts for the Okhotsk Sea, the north Pacific and the Bering Sea. They are mainly based on the data from the Khabarovsk broadcast (RDW). The sea-ice data are also included in the same charts, and these charts are very useful for the fisheries in these cold areas.

In the foregoing, some examples have been introduced to show the extent of the utilization of sea-surface temperature data for fisheries. JMA plays an important role in the dissemination of sea-surface temper ature data and other oceanographic as well as maritime meteorological data for fisheries purposes, and those data are widely used by fishermen in Japan and in other countries. OJ 0\ DENSITY OF OBSERVATIONS OF THE I (II ~ '\~ ~/ / ~·'·'·'·'1 • SEA SURFACE TEMPERATURE Mc')\ I 7 /oY . 7 7'Q",!",0'l .•

First Ten-day of June, 1967

0 40 • Ul I>J I '"Ul ~ '"&1 01 I ~ 0 30 til 01 I ~ ~ 0 0

1 200

I JI· 1/'9 I 0 '-1 0 I 0 I I .I let 0 0 0 0 0 ,,>,..,0 130 140 150 160 170 18o"E Figure 1 - Distribution by area of the number of ships' sea-surface temperature observations for the first 10 days of June 1967. T: Tango, V: Victor 6 CSPA JMH I ~ Ml:f-\I'l ~LA ::,ut"'\r,L.j~~ ~cr~t:r-::IUI"

I. N. TEMP i"NOM. I. N. TEMPI<\NOM I. N. EMP. NOM' '112 26:'1 426118 +2.6755 20.9 +0.9 '118 :26:7 0.61418 8.7 +10 Boo 2(;9 +-1.9 '1:27 268 + 00 420 838 222 '45254 +04 436 Q4 -0.7 9.29 26:1 4OS!to 0.3 ~ '136255 +-02402 '10 '1 243 + 0.1 4-01 !7.4- + 0 s 83625.5 +121,(21 !40 +0.'1 655 :2S(; 4-28 !4-q +06 (:12 678232 +09 S~r~A 18.2 +22 673 220 +.9602 !90 +0.'1 598 !94 + .(; 6001'17 +09 74-0 585 !3.3 +;.3 (;5 +!2 >-3 H ~ fjj ')~~~~a~ --" '1 ______~ "'25 i;j .",.~ 25 I tj '-----.. 26 28 o 20 ~-- --- CSPA JMH "--.--- 29 " I I 1400 JAPAN METEOROLOGICAL AGENCY _1 TO.1Q.. 1!.ill. 1967 Figure 2 - Ten-day mean sea-surface temperature for the period of 1-10 June 1967. (Original copy for radio facsimile broadcasting, CSPA, JMH. Temperature: DC) f\) --.J 26 '-...--J 05 \./ -J I-I -- ...... 0~./ ,.o~ "- . 0 ! 03 0.3 1 .10"'#\'~~'OL_" .2 Xu. 0.9 09 0: 0·~0.6 O~ 0.4 9.5 0.6 9 03 0.6 ~O . ' --," / - ... 14 , ~OO \ N if \}~I'JI· /I Ys;: 6'0 1< I 14t S'~A' l??\:0\ $~ 1< ~~ -~ ~~ ~'V 21 /,Q /' <5'/ gj \\I~ ~15° ;J> '/f'J )I Co: III ;, ~ " / /i C 21 SEA SURFACE 22 ~ TEMPERATURE 23 0 .-,---" 1 TO 8 JUNE 2 29 I 1967 ~ 4 I L:~ -. \." r------1 ~ ---r --i 125 ...... (J)

w 26 ~Z o...::::J 27 (J) ...... U I 127 0 133 0 1390 1450 15n

Figure 3 - Composite SST chart for the period of 1-8 June 1967 and surface current chart for the period of 2-9 June 1967 (upper left). (Original copy for radio facsimile broadcasting, CSPA 1, JMH. Current: knots; Temperature: Oc) COLLECTION OF DATA FOR N.W. PACIFIC 29

tt J~,~i iM ~f. Sci '* fi \-\f,\S."f<:1 .\111<1.'\1- IJlhEi(\'II'1 TE~-DAY MARINE REPORT OF THE EAST CHINA S~\ li'\ 736 'if June . 10 19 68 ilil ;$ 0) II ;l! If Jj H #; & ) 4-3 6 12

~ ..,"

. ~-61 27 ._'-"-'-~~

~Jt!l (/) in ~ *-;:J. il'j , iJ.:. liil 1{, "I' , " "I llif ffJ / ...... _. ~ 11j( I*8. 1 r I,D - 0.3" 1\ ,---.-.~'-'-' I' 1 1.4 T 0.8 t- ['.3 ~" ;* '1-1.9 -{I.U I 9,1 + 0.9 l' -r 2•J ,- 1,& 1,\ I'; f~, 1 1.0 't' , !..\ 1 1.1 t- .0.4 - 2.2 -t- 3,2 1- I,J 1: r, ,10 1 J.8 r 1.0 - 0.4 'I- 0.8 1- 1.0 IJl~ ;;1 1 4.9 + 0,4 - 0,11 'I- 0.8 3,0 '1- , ~~ x.~ 4.4 - 0·2 - 2,4 11 1'" 1,4 - 3.0 1 "1:( ,o, J.!, 1 4.9 - 1.3 - 2.4 ; -t- 2.0 - 1.6 I, J'; ", 11 1- -fJ - 0.5 i-o.o 'J I~; II) ,_~ I 6,5 + 1,5 ID r'll j'. -0.• i-o.8 Ih},

Figure 4 - Ten-day marine report for the East China Sea. (Issued by the Nagasaki Marine Observatory, JMA.) \.>Jo CSPA2 JMH 21T025 MAY68 3.0 I 4.0 Kamchatka P.

.~w . 4.0 4.0 3.0.~3.53.~.: 404.~ttu Is. ) .;-4.0· 'i::> 4 ..( w /. ,tY:L/ ),•., \~ ~4.0

.- o &J / - - C :» I l .- (f.l 3.0.,_/ c 3.5__ '.-' .N ! • V, 60 '" /( rr:t+-t-+=!::::J, 6 / / I ~.. COLLECTION OF DATA FOR N.W. PACIFIC 31

140° 1600 E

-2

-----._-~ -4

-4

-2

+2 ----0

30°

o +2 +2

o Om 68

Figure 6 - Sea-surface temperature difference (66°C) between August 1934 and August 1933 )2 SEA-SURFACE TEMPERATURE

~_-+--'--I'-<----+----.1fL-1-----t------i45.N

\---+---~="""'-od------j 40' - ...... , ------20

•• 1..:.-/lJ?f'=------'Oj~,-,---b-L--"'....-+----l35· ----

f-'°..:'---+----f-----+----I------+------i30·

0 130' 135' 140' 145' 150" 155' 160 E Figure 7 - Comparison between the predicted isotherms for 15°, 20° and 25°C (solid curves) obtained by the screening method and those determined by observations (dashed curv~B)

~ Vi ;-:pP J '.r'\. 45'N )~'-'1'/7 };. D ~ 1 .~0 ~ / 2 3 40'

~\ 9 . 4 5 6 cI . I 1/ 35' VI / VA0 v, , k/:5 7 8 9 <:7 . • 19 ~O 11 12 13 0' 3O'

A 125• 130..135 140. 145..150 155.E Figure 8 - Prediction areas of the screening method SST and Catch of Fi sh 140 0 145°E 21 T6 25 MAY 1968 I Bonito Albacore 3t ::c:- O "" 0 0 3-10t 0 &. ~~"\ 20 1O-30t @ ~ £. \ co 30t <::::: 1-'1 A..",", ..:' _ 17 20 H ?", 0 • z 35°NI- A_ nI ·~l ~~~ 35°N ~ ~ ):; ";I @ z 20 ~ I >E: ;p 0'" 23, ~ ~)) ( -0 ~ }O~ H "" I 0 ... -- ... , , ,, ,2~1/-, \20 , , ,/ 140 0 145°E

Figure 9 - Composite map showing the SST distribution and the catch of fish

(;j 34 SEA-SURFACE TEMPERATURE

11 TO 15 May 35°N

20 1 21

23 16 TO 20 May 23 35°N 22 35°N C;1

35°N

",8 o•

o 23 24~O 140 0 E

Figure 10 - Consecutive changes of the SST distribution and the corresponding transfer of fishing grounds COLj~CTION OF DATA FOR N.W. PACIFIC 35

ALBACORE ~~~ISI , ~ ~ )?fi;). FISHING GROUND FORECAST ~".' ." '" 16~.l (.<:',~i' NO.193 APR. 2, 1968 09:$.L.!~o.~H/ .:: ~::::},::18 OCEAN.INSl, TOKAI UNIV.; \ . '(~';J :..... KYUS1~1~)i~;/JFt ,~H)l~ (-OS! l (-I) A .. ··, .. I ..' -:: . >/ /' C ' , (-1) • (!-1) 19 .• •. (1)"- +-- Prom 151njj- o" ...... <\.- of - '. -. ,.' I (-25) 17 ~"""--"'-",. ":"l'" I -----I----y<-19-· 'N ~i"'--'I ~.k ~5) 130 :::i;(·:······ -- jw .:-T.... 19 I~CURRENT (-15) "" (0) . ~ ':: 8', ' ",' RIP~~ I "" .' .~-, 1 " Lo.t"~e S ,~ r'sr,...... ie, - BONITO , / J\: ,,~_ ~ \' ",WVv\<. t 4 6k 1. 5 _2K} . :PROMISINq-AREA-, '" ~ - 8 ~.' .. dc) BI~ FlS I, 19 \ ,/-,' MIDDLE APRlif\:'. ' /.'j.S- (-05) ~-rokl 4-1 0 ",1.., (-~~\O) : ~ ,!,-20 4-6k~ (-15) ", ~ t'" 1(-05) 'I- s_j2kl ",-." Low T... m~ w...hY' (-3) ", V CURRENT Rern'"Ml~,.,\. ~·6-:·r~2K}}./ 21 '--;;-;------h-.,.,------~_/....-- (-2 ~\·.~W:·:·~,\~~ / 22 ., //.//;';';'';: '••; WARM Cu,". (0) (-2) (-I) 21 j..:.:< BRANC W 25' 130' 135' 140' 145'E 'Y V Birds 1,000 'V\ANW Predicted Emerging Line of Fish ~ ~ Predicted Cruising Fish in Surface Water 7' Predicted Cruising Fish in Subsurface Water ( ) SST Departure from Preceding Year

Figure 11 - Albacore fishing ground forecast chart issued on 2 April 1968. Oceanogr, Inst., Tokai Univ. (Original in Japanese)

SEA-TEMPERATURE STRUCTURE AND ITS RELATION TO THE UNITED STATES TUNA FISHERIES IN THE EASTERN PACIFIC OCEAN

Glenn A. Flittner*

The Fisheries

Four species of tuna support the United States tuna fishery in the eastern

Pacific Ocean. The California-based industry fishes from Vancouver Island on

the north, at about 50° N. latitude, all the way south to Antofagasta, Chile, about

25° S. latitude. The fishery operates in both the tropical equatorial region and the

temperate zones at both the northern and southern limits of the range. The species harvested are the albacore, the bluefin, yellowfin and the skipjack'J'* The yellowfin

and skipjack are the tropical species comprising the bulk of the harvest, which is

taken in the light-shaded district from upper Baja California down to Peru (fig. 1).

*Fishery Biologist, Bureau of Commercial Fisheries Fishery-Oceanography Center,

La Jolla, California 92037.

**Scientific names for the species are as follows: albacore, Thunnus alalunga

(Bonnaterre); bluefin, Thunnus thynnus orientalis (Temminck and Schlegel); yellowfin Thunnus albaeares (Bonnaterre); and skipjack, Katsuwonus pelamis (Linnaeus). SEA-SURFACE TEMPERATURE

The yellowfin is the major species in value and tonnage, and the skipjack is the secondary species. The albacore tuna is the only species accepted for the fancy white-meat pack, and commands the highest price at the dockside, whereas the blue fin is a lesser-valued species, but commands a higher price than skipjack and is harvested in the smallest quantities of the four. Albacore and bluefin are found along the North American coastal region in the dark-shaded areas, respectively.

Historically, the tropical tunas have produced from 70 to 80 percent of the total production in the eastern Pacific, and in recent years, the temperate species have contributed the remainder.

One of the problems that has persistently plagued the tuna fishing industry has been availability. Of the four species, yellowfin tuna appear to be most uniformly available to fishing, whereas albacore and skipjack appear to be most unpredictable from year to year and season to season. Bluefin appear to be intermediate in character.

Our best record of fluctuations in distribution has been observed for the albacore tuna. Since 1940, landings have varied widely, and California, Oregon, and washington have experienced large fluctuations which are reflected in the fol lowing graph (fig. 2)--in 1953-55, California waters produced almost all of the albacore, whereas in 1966-68, Oregon and Washington landed up to 75 percent of the total production, reflecting major dislocations of as much as 1,000 miles in the range of the species. ITS RELATION TO U.S. TUNA FISHERlES 39

The tropical tuna fishery characteristically moves northward into Baja

California waters during the Northern Hemisphere summer, and exhibits similar but less noticeable movements to southern latitudes during the Southern Hemisphere summer. Fish are found intermittently all through the year in the region of the meteorological equator, about 5° N. latitude, but one can readily observe that the coastal and offshore region visited by the tropical tuna fleet exceeds a distance of

5,000 nautical miles. Finding fishable concentrations in this vast region becomes a major problem at times in the prosecution of this fishery.

These two resources, the tropical and the temperate tunas, are harvested by three major classes of vessels. These are shown (fig. 3) because they signifi cantly affect our concern with oceanic conditions and the weather that is encountered in this large region. The troller primarily produces albacore; it has limited range, limited carrying capacity, and limited sea-keeping capability.

ApprOXimately 1,500-2,000 of these little boats participated in the 1968 albacore fishery. The baitboat works on yellowfin and skipjack during the Northern Hemisphere winter season, and in the past 2 years has entered the albacore fishery during summer and early fall. It has a more efficient refrigeration system, larger hold capacity, and somewhat better sea-keeping capability. An estimated 45-50 of these boats are presently active in the U. S. west coast tuna fishery. The purse seiner, the major vessel in the California high seas fleet, now carries from 150 to about

1,000 tons of fish. It concentrates primarily on the tropical species in the distant water fisheries; it has superior sea-keeping characteristics, the greatest range, carrying capacity, and fishing efficiency, and it can generally stay with the fish 40 SEA-SURFACE TEMPERATURE

wherever they are found. The high-seas baitboat fleet converted to purse seines

in:L959-60, initially fishing only yellowfin and skipjack tuna, and began fishing

bluefin and albacore tuna in 1961-62; the bulk of all domestic bluefin landings are

now taken by the big purse seiners. A few seiners have fished for albacore but

production has been generally insignificant. The failure of the purse seiner to

operate efficiently on the northern albacare grounds is probably related to 1)

generally small-sized schools of fish, and 2) weather and sea conditions which make

the setting and hauling of the purse seine and recovery of the seine skiff generally

too hazardous to the ship's crew. Approximately 105 tuna purse seiners having

an estimated 40,000 capacity tons operate out of Pacific Coast and Puerto Rican

ports.

The Ocean Environment

All the major tuna fisheries move rapidly from place to place during a

given season. For example, in the temperate tuna fisheries, albacore may be

found near Guadalupe Island, about 30° N. latitude, in June when the fishery usually

commences, but may occur as far north as Cape Flattery, about 49° N., by early

September (fig. 4). Bluefin usually appear just north of Cape San Lucas, Baja

California, about 22° N., in June and later appear off southern California, about

33° N., by early September (fig. 5). Yellowfin and skipjack may be found at 20° N.

off the coast of Mexico in January but occur off Baja California at 30° N. by mid

August (fig. 6). In each instance, the nature and degree of movement is highly variable

from year to year and is related to changes in oceanographic conditions. ITS RELATION TO U. S. TUNA FISHERIES 41

The general coastal migratory patterns thus follow a poleward course from winter to summer, depending upon whether one is in the Northern or

Southern Hemisphere. One of the factors which dictate this cyclical pattern of movement is sea temperature. Here (fig. 7), one can see that the 70 0 F. (21 0 C.) limit has a significant correlation with the summer appearance of both yellowfin and skipjack off Baja California; note the variations of northward movement from year to year: the farther north the limiting isotherm moves, the farther north the tropical species are found (Blackburn, in press). The same situation prevails for the temperate species, albacore and bluefi'l.

One of the other features we need to know a great deal about is the vertical temperature structure of the upper layer of the sea. This is a characteristic bathy thermograph trace taken in the tropical tuna fishing region (fig. 8). Having the information as shown in this illustration, one can appreciate how the purse seine's efficiency is affected by the depth of the thermocline (the zone of sharp temperature decrease with depth)--fig. 9. In the first case, one will observe that when the thermocline is shallow and the waters underneath are cold, the yellowfin and skip jack tend to stay up in the net. As the rings are closed and the net is pursed, the fish do not dive down, and are captured. In the second example when the thermocline is deep, the fish have an environment conducive to their escape during pursing opera tions. Thus, good fishing strategy in the tropical tuna fishery is to look for areas where the first set of conditions prevails; when schools are found and the net is set, the probability of catching the tuna is significantly enhanced (Green, 1967). 42 SEA-SURFACE TEMPERATURE

The albacore tuna, the most boreal of the four species, exhibits a similar

kind of temperature relationship (fig. 10). This is a 5-year tabulation of albacore

logbook data for approximately 7,700 tons of troll-caught fish landed in California

where the fishermen recorded sea temperature at the time of catch (data from

Clemens, 1961). The modal temperature is about 64° F. (17.9° C.), and the upper

and lower one-third limits are about 62° F. (16.8° C.) and 66° F. (19.2° C.). Very

few are taken below 60° F. (15.5° C.) and above 68° F. (20.0° C.).

The same situation prevails with bluefin tuna. This is a purse seine fishery,

but again, excellent logbooks maintained by the fleet have provided a similar picture

(fig. 11). This illustration is based on a 3-year record including approximately

22,000 tons of bluefin: the bulk of the catch is taken near the modal temperature of

67° F. (19.5° C.), and the upper and lower one-third limits are 65° F. (18.2° C.)

and 69° F. (20.5° C.).

Now, if one has this information, how does it relate with depth, and where

might we expect to find albacore and bluefin tuna? The next illustration (fig. 12)

shows three different examples of upper mixed layer temperature structure com

monly observed for both albacore and bluefin tuna. One can readily see that when

there is no vertical dimension to the critical lower temperature limit, no fish are

found. As soon as the temperature falls within the optimum zone (shaded from the

previous illustrations) and the surface layer assumes permanent vertical dimension,

the fish may be found. Later on, when surface temperature conditions exceed the

upper limit, but a cool layer exists below, then the albacore run de-ep in the layer ITS RELATION TO U. S. TUNA FISHERIES suiting their preference. In the third example for bluefin, a typical double-layer situation is depicted: bluefin are often found on top of albacore late in the season off southern California.

It should be apparent at this point that the tuna fishing industry would benefit considerably by having the right kind of sea temperature information placed at its disposal. Furthermore, not only analyses of present temperature conditions, but temperature forecasts, would be extremely useful. The same also applies to other oceanographic parameters, such as salinity.

Forecasting for the Tuna Fisheries

One of our most difficult problems in forecasting sea temperature structure for the tuna fisheries is that we have insufficient hydroclimatological data for the world oceans on which to base forecast models. The thermal cycle in the upper layer of the sea is only generally described and poorly understood. In the tropical seas, too few useful data are available even to describe existing conditions, especially in the Southern Hemisphere. In the Northern Hemisphere, sufficient data are now at hand, particularly in the Pacific Ocean, but forecast models are largely empirical and have little statistical or physical basis.

Up to now, our only successful applications have been for albacore and bluefin tuna. Until more observational data become available, forecasts of thermal conditions in the tropical seas inhabited by yellowfin and skipjack tuna cannot be attempted. 44 SEA-SURFACE TEMPERATURE

The first step in our program was to analyze and publish charts of average sea temperature conditions for the eastern Pacific Ocean on a monthly basis (figs.

13 and 14). This publication is now entering its 10th consecutive year; it is also the medium for current articles and predictions. The second step was to commence publication of average sea temperature conditions for the U. S. west coast (fig. 15) on 15-day intervals each year starting on April 15 and ending on October 31. These charts have been most useful to the albacore and bluefin tuna fleets who operate in districts where temperature conditions vary considerably over short intervals. The third step was to recognize that these publications most often failed to reach the primary producer who most needs to see them--but is at sea--so in 1966 we com menced a series of daily albacore fishing information broadcasts from BCF-licensed radio station WWD located on the ScrippsInstitution of Oceanography campus at

La Jolla. These broadcasts commence on June 1 and terminate on October 31, and include the latest albacore information obtained from research vessels, fishing vessels, and unloading station operators. They also contain sea temperature reports, regional weather information, and current market prices and unloading information whenever it is pertinent.

The fourth step is still in progress. In the past 3 years, we have heard a common reply: " ... don't tell us what it was like during the last interval, but give us forecasts for the next period, so that we can plan where to operate next.•.. "

In 1966, we first attempted to forecast sea-surface temperatures off the

Pacific Coast (fig. 16). Using our own 5-year data base, we determined the average ITS RELATION TO U. S. TUNA FISHERIES temperatures for each 15-day chart, and the average temperature differences as well for the season. Then, taking the existing analysis for the initial period we added the average temperature change of record and graphically constructed a "prognostic" chart which served as a first approximation to the new sea surface temperature field to be realized 2 weeks later. At the close of the calendar interval, we then compared the actual analysis with the prognostic chart graphically, and the third frame of the figure shows the error field--as much as 6° F. too warm in the coastal upwelling field and as much as 2° F. too cold offshore in the north.

Our assumptions in this forecasting approach were that 1) no changes occurred in the mixed layer depth during the forecast interval, 2) that there is no change in the anomalous temperature patterns which preceded the forecast period, and 3) that no changes in the prevailing climatology occurred during the forecast interval.

ObViously, none of these conditions were met. Our experience subsequent to this first experiment has shown that accurate climatological forecasts up to 15 days in advance are required before meaningful progress can be made in sea surface temperature forecasting.

Meanwhile, we realized that a considerable amount of useful information resided in our long-term data base: the 15-day sea surface temperature normals, the average liT between successive 15-day charts, and the anomalous liT from one period to the next. From this information, we have substituted interpretive bulletins instead (fig. 17). These bulletins are intended to provide the commercial fishing community with our best estimates and judgments; these are issued.in lieu of the 46 SEA-SURFACE TEMPERATURE present prognostic models for sea temperature and climatological parameters which have very poor skill over intervals as long as 15 days.

Another factor that we have had to consider is that the annual thermal cycles for various regions of the oceans are not the same. To develop information of this kind, we have selected several key fixed points in the eastern Pacific Ocean which report meteorological and oceanographic observations regularly. We have assembled a 5'-year data base for ---in situ heat flux estimates derived from the meteorological data, as is shown in figure 18. One can readily observe that none of the 5-year average curves are identical, and that several differ from what one might expect. In the open ocean, one might expect to see the heat flux and sea temperature curves to reach a maximum in the middle or late summer, as at Ocean Station Vessels Victor, Papa, and November, for example. But when one approaches the U.S. west coast where upwelling dominates the scene, one observes virtual reversal in the annual cycle-- so much so, that at Blunt's Reef in May-July, the sea temperatures are at a minimum, even though the meteorological-climatological minimum occurs in December-January.

The other coastal stations also demonstrate similar characteristics.

The 5-year record for Weather Ship Stations Papa and November also exhibit other features (fig. 19). The annual cycle for Station Papa appears pre- dictable in that heat flux and sea temperature move smoothly, with a maximum on the heat flux curve preceding sea temperature by 4-6 weeks about August of each year. However, at Station November, located in a tradewind zone, the annual cycle is disjunctive and complicated. Here, quickening of the tradewinds apparently ITS RELATION TO U.S. TUNA FISHERIES 47 results in excessive evaporative cooling, causing sharp fluctuations in the thermal properties in the upper mixed layer of the sea. Consequently, accurate forecasting of sea temperatures in this district must take into account the short-term effects of relatively minor perturbations in the tradewind weather.

From the above discussion, it should be clear that our fishery forecasting challenge is manifold: to give meaningful sea surface temperature prognostic charts to the fishermen, we must first have accurate climatological forecasts for periods up to 15 days. In the absence of workable models which can predict weather so far in advance, it appears most practical to shorten the forecast interval and to increase the day-to-day reporting of synoptic information.

To serve the needs of the tuna industry the present forecasting activities carried out at the La Jolla Laboratory are being addressed to a complex set of problems; these are readily divisible into two groupings--short-term and long-term.

Short-term. --The primary producer--the fisherman (but also the processor)- benefits from constant updating of present conditions and projections of trends of several variables. EnVironmental conditions such as sea temperature and especially weather seriously alter the course of daily events. Fishing operations and strategy are planned over short periods of from 1-5 days or more, depending upon boat size, and boats may move 300-1,000 miles during the interval. Processors, by having constantly-updated information, can plan for surges in landings caused by boats seeking refuge from the weather, and can make better decisions affecting day-to-day 48 SEA-SURFACE TEMPERATURE unloading, storage and canning operations. Exchange of such information through the medium of daily radio broadcasts and weekly or biweekly bulletins appears to be practical.

Long-term. --Processors and governmental agencies--but also fishermen and boat owners--have an urgent need for long--term predictions and advisory information. The nature of the fishing season--good or bad, early or late, big fish or small fish, shortage or abundance of preferred species--seriously affects the entire fish business. Anticipating oceanographic events has proven to be an easier problem to deal with than the biological events; but in both areas, considerably more information is needed before accurate predictions having substantial reliability can be made. The ultimate objective of such long-term predictions would be to enable better business planning and operations, and to permit improved management and regulatory measures to be taken.

To date, we at La Jolla have determined that we are equipped to operate most effectively on the short-term problems as opposed to the long-term ones.

Until such time as we can assemble and analyze the large body of information reqUired to make long-term predictions, we can only observe the urgency of the need. In the meantime, we are concentrating on more effective means of acquiring, processing, and disseminating current information on a real-time basis. By careful monitoring of large masses of environmental data, we have learned that it is pos sible to observe and to recognize changing trends quickly, and to collate and to report our findings to the tuna fishing community promptly. ITS RELATION TO U. S. TUNA FISHERIES 49

The experience of the 1966-68 albacore tuna fishing seasons seems to indicate

some measure of success. In 1966, just as we were observing the poorest July

start in over 20 years in southern California, Oregon was experiencing the earliest

opening since about 1950. Our first bulletin was broadcast by radio on July 15,

noting that weather and sea conditions had favored early movement of albacore into

Oregon, bypassing California waters. In about 10 days, a large part of the California

fleet moved north on the basis of our daily broadcasts from WWD and arrived on the

northern grounds at least 2 weeks earlier than previously. The result was an esti

mated 1,500-2,000 short tons of production valued at $585,000-780,000 which was

accumulated at the beginning of a season which, even though fishermen operated at

peak capacity the remainder of the year, fell short of the 1945-66 average production

by almost 13 percent. The early advantage was thus carried through the entire

season, preventing an even poorer catch record.

In 1967, the fishing community appeared to accept our advisory information

very early, and consequently it was difficult to tell just how much our activities might have contributed in value to the catch. Nevertheless, we do know that our monitoring of the unusual oceanographic warming off Oregon and Washington success fully foretold another unusual northern year. We issued our first bulletin of the

season on July 14 and published a followup bulletin on July 31 which alerted the

industry to the rapid changes which were taking place. Despite the latest start in several years, virtually 90-95 percent of the albacore production was taken off

Oregon and Washington subsequently in the 70-day interval between July 21 and

September 28. All-time record catches were logged by southern California bait- 50 SEA-SURFACE TEMPERATURE boots operating far north of their usual range, and even though the entire albacore season lasted between 80 and 90 days, almost 42, 000,000 pounds were produced, placing 1967 right on the 1945-66 production average.

Early in 1968, we were encouraged by the highly favorable response to our questionnaire which was circulated in November 1967. Last year, in the summary which was presented in our January sea temperature publication, we observed that about 80 percent of the individual respondents approved in principle every phase of our information reporting services as presently rendered. Criticisms and suggestions were offered in addition to questions answered on fully one-third of all replies. After the albacore season began in late June we learned that fishermen and processors were reading our temperature charts, advisories and reports more closely than ever, and continued to do so throughout the entire season. The fishery was centered in Pacific Northwest waters again, which we first reported in our Bulletin Number

68-4 on July 15. Subsequently, despite persistent bad weather, an early-season price dispute lasting more than 2 weeks after albacore appeared off southern California, and unloading delays in northern ports, a near-record production year emerged.

Oregon set an all-time recordsince albacore were first landed there in 1935-36, and the combined U. S. -Canadian landings totals placed 1968 in the third highest position for all years since the fishery first started in California in 1915.

The Challenge for the Future

Up to now our tuna forecasting staff has concentrated most of its effort on ITS RELATION TO U.S. TUNA FISHERIES 51 the albacore and bluefin tuna, and the least on yellowfin and skipjack, despite the fact that the latter species form the bulk of the annual production.

The choice of the albacore was not an accident--we know much more about the ocean environment off North America than in the tropical and southern latitude districts. The excellent data base accumulated by the California Cooperative Oceanic

Fisheries Investigations since 1949 has provided us with a wealth of information, from which we have been able to assemble some meaningful statistics. In addition, the excellent catch records and logbook data maintained by the California Department of Fish and Game since the fishery began, have given us the initial biological base on which the present system rests.

We already know that the tropical area poses a different set of problems-- the relatively narrow thermal preferenda exhibited by albacore and bluefin are not as specific for yellowfin and skipjack. We know very little about the oceanography of the vast tropical ocean, and so about 2 years ago at this time, the BCF-sponsored multi-agency, international EASTROPAC surveys got underway. Thepreliminaryfindings of the EASTROPAC surveys already indicate that forecasting for the tropical fleet is possible,--but that it must wait until some kind of regular measurement and monitoring schemes are devised to prOVide us with the data base we require. For example, it was not until we had published 5 years' worth of sea temperature charts for the Pacific coast that we could understand what was taking place on our tempera ture differential charts, and just what an average temperature change from one interval to the next looked like. 52 SEA-SURFACE TEMPERATURE

One immediate benefit to the tropical fleet can be derived from our albacore experience: day-to-day observations and monitoring of fishing conditions and weather will provide the basis for daily broadcasts which could be directed to this fleet all year long. Another benefit should result from our closer working experience in the tropical oceans: the unpredictable nature of the skipjack fishery along the

Central America coastal region may become better understood. 'An additional benefit would be to remove some of the hit-or-miss variation from the yearly skipjack landings, and perhaps to increase the average annual production slightly. And, with better oceanographic data at hand, it may be pcssible to push farther offshore in quest of both skipjack and yellowfin tuna.

In the future, as the maritime nations push to the sea with ever growing needs for animal protein, fishing competition for the common property resources of the sea will become extremely keen. At that point, U. S. fishermen, in their search to maintain competitive advantage, may have the need for special kinds of advisory materials which will help them to use their remote sensing devices, such as sonar, to better advantage. One immediate result is shown in figure 20. Here the effects of various gradients of temperature with depth are depicted; the fisher man will have to obtain this information to interpret his sonar traces properly.

Later on, as his experience increases in the use of such equipment, he may resort to variable-depth sonar, for example (fig. 21), to maximize his search for fish.

When this day comes, we plan to provide him regularly with the information on depths, sound velocity profiles, ranges, sound ray envelopes, etc. The technology of these problems has already been explored by U. S. Naval authorities, and it remains ITS RELATION TO U. S. TUNA FISHERIES 53 a relatively simple task for us to adapt these techniques to fishery needs.

Other new problems will arise as we push farther along on the road to better fishery forecasts. Other major fisheries besides tuna could well benefit from forecasting and monitoring activities performed by our Bureau in much the same manner as I have described above. Each regional fishery will present a somewhat different set of problems--but as our understanding of the relation between meteorological; oceanographic, and biological events grows, so also will our capability to provide meaningful information to the fishing industry.

Clearly, our environmental forecasting challenge is multifold: we must take significant advances in several areas. To give meaningful sea surface temperature prognostic charts, we must first have meaningful and accurate climatological prognoses for equivalent intervals so as to compute the flux of heat from the atmosphere to the sea and vice ~ Secondly, we must develop practical methods for measuring and reporting heat flux processes, so that we can place sea temperature prognoses on a more physical and less empirical basis than at present. Thirdly, we must develop more useful weather and sea forecast materials for the fisheries users because this kind of information had direct daily impact on fishing tactics and operations. And lastly, we need to develop suitable forecast models of the lapse rate of sea tempera ture with depth so that fishermen employing acoustical ranging devices (e. g. , sonar) can properly evaluate sound ranging characteristics under specific operating situations adversely affecting fishing efficiency.

The problem of "•.. what to measure...II is a subject of the present SEA-SURFACE TEMPERATURE

CMM-V Meeting. For fisheries purposes, Airborne Radiation Thermometer (ART)

measurements may not always prove suitable in representing conditions at the

average depth that a surface school of tuna occupies (about 10 meters). Questions

of accuracy are related to need: scientific research demands are more stringent

than fisheries needs; accuracies to the nearest 0.5° C. appear adequate for pelagic

fisheries purposes. Time and spatial scales are dependent upon the nature of the

problem: coastal processes frequently require data from a microscale observational

grid, whereas the high seas (oceanic) conditions may be satisfactorily described on

either the mesoscale or macroscale.

Thermal processes in the sea cannot be adequately modeled without a com prehensive understanding of the air-sea coupling mechanism and the feedback processes associated with it. To confine one's approach to the problem by limiting concern to only the upper layer of the sea above the thermocline and the gradient wind (friction) layer of the atmosphere is to ignore coupling processes on a much grander scale. The time well may come when the fishery oceanographer may have to look at the meteorologist's upper atmosphere charts routinely to describe what will happen at some future time in the upper layer of the sea.

Literature Cited

Blackburn, M. B.

(In press) Conditions related to upwelling which determine distribution of

tropical tunas off western Baja California. U. S. Fish and Wildlife

Service, Fishery Bulletin. ITS RELATION TO U. S. TUNA FISHERIES 55

Clemens, Harold B.

1961. The migration, age, and growth of Pacific albacore (Thunnus germo),

1951-1958. California Department of Fish and Game, Fish Bulletin

115, 128 pp.

Flittner, Glenn A.

1964. Review of the movement of albacore tuna off the Pacific coast in 1964.

U. S. Fish and Wildlife Service, Commercial Fisheries Review 26(2):

13-19.

Green, Roger E.

1967. Relationship of the thermocline to success of purse seining for tuna.

Transactions American Fisheries Society, 96(2): 126-130.

Howard, Gerald V.

1961. The Pacific coast tuna fisheries. Western Fisheries 62(2): 1-7.

Lynn, Ronald J.

1967. Seasonal variation of temperature and salinity at 10 meters in the

California Current. California Cooperative Oceanic Fisheries Investigations

Reports, 1 July 1963 to 30 June 1966, 11: 157-186. SEA-SURFACE TEMPERATURE

Figure 1 - The four major tuna species and their ranges off the Pacific Coast of the Americas (After Howard, 1961) ITS RELATION TO U. S. TUNA FISHERIES 57

LANDINGS (millions of pounds) o 20 40 60 80

1940 • WASHINGTON 41 42 ~ OREGON 43 44 o CALIFORNIA 45 46 47 48 49 y 50 E 51 A 52 R 53-.------,------' 54-1- ----,_-----'-, 55 56 57 58 59 60 61 62 63 64 65 66 67 PRELIMINARY 1968

Figure 2 _ U.S. Pacific Co~st albacore landings since 1940 VI (l)

o ~ ~I B CJ B -- 7 t:l " -- ~- -~ . ------=------~...... - TROLLER 53' 30 ton BAIT BOAT 73' 65 ton fiJ :r> II ~

:,;til I

o I 10 I 10,,01 1.. 0'/ I I 0

CJ CJ

PURSE· SE INER 149' 500 ton

Figure 3 - Classes of vessels participating in the U.S. Pacific Coast tuna fisheries

._--_.~ ALBACORE MIGRATION MODEL

~ ~ ~ C5 c CJ] ~ ~ "J ~t- i ~

Figure 4 - Schematic model of albacore tuna movement off the Pacific Coast (From Flittner, 1963)

V1 \0 60 SEA-SURFACE TEMPERATURE

P,T CONCEPTION

120 0 115 0

Figure 5 - Schemat1c model 0 f bluefin tuna movement 0 ff the Pacl.'he Coast ITS RELATION TO U.S.· TUNA FISHERJES 61

30"

25"

20"t~-!:;- -;;\;O-==~~:=~~:~IO~5;o._ liD" 120" 115" 1 of yellowfin una movement t" mode and skipjack t Figure 6 - Schemaoff the10Paclflc.. Coast Ri

IIIII 34°~ NORTHERN LIMITS 33° I-- 21°C noon AT SURFACE -- :: SKIPJACK TUNA ---- : I' ". 32° ~ YELLOWFIN TUNA II ,I _ I II If ~ I I:i I II en~ 310 ,.""'" I ~ I :1 II :z: II ~ I;~ I 1: 1 ce ~ I· I I: I I ..., 300 ii,' ,'I . , " - • " ,.. I •. CI:: Il II :;' I:I : J ~: I ~ ...... _ I :: I:' 3 t: I I ::E II " , ,I t I "" II I il ';' ,'I ""'290 , ., .I ., ...... - I II II :I c:a I I : , .: I i ~ :::> I l I ;'1 ii" ~W .' ..,.. ,I I, , '. , 11 I "'".....lI I .f· E I"£ ~ : ".....I :. M'I ~ :z: ° II I 1: . i , ~. ' S :;;;; 27 ~ V f , \ ,I' " I liJ ~~ ~ :~::": ~ = I f .: J: . \ . I. I :z: 26° .I I I ,.,, I' I . 1 I ,I'" II I I' I I.~ f "I ~ If' .II'' I I·: . k I ..: , I 250 I I I ! I , I I : f f I Il f i 'I' ,if ,I I fl' : J J ..: ~ I: •• I! II , ,I ! t Ii" I 240'' : : I . ," I J ::1 'V I I , ,rr I !? • "1 ' ':"~ ... • .f' ° ~l ~ :. .I iil : I I ~ I ! II II 23 1"1' 'I , "I' ."1' 1""1 'I 'I '1' 'I' "I' "1"1"1' "1 'I '1'1'1 Tl"I"I' "1"1 'I" 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965

Figure 7 - Changes in latitudinal position of the 21°C (70°F) surface isotherm and the northern limits of the commercial yellowfin and skipjack fisheries off the Pacific Coast, 1971-65 (After Blackburn, in press) n ,j:> 40. - 45. - 50 55 60 65 70. - 75 80 0 3 ------85 9~

50 50 -- "t- --. ~ ..... )0 00 ~ - - -0 -+ _ ...... - ---;-. )0 50 ~ Ho 2 00 .Z - o>-3 , C 250 2.50 (j) , i:'! 3 300 ~ --f- + ..... f- ~ ,- I:t' OJ 350 350 ---0-4-0-~ --_.- ""t i ~~ t;j 4 •400 (j) . ,--;- - i- T , ~~~ . + -. t -f - - - 4 5"- 40- 45.------65 - 450 30 35 70 75 80 85 90 Figure 8 - Typical mechanical bathythermograph trace from the tropical tuna fishing district off Central America

0\ '-" 0'\ ~

SUCCESSFUL SET UNSUCCESSFUL SET

/PURSE SEINE NET~

&J :J> I ~ ~ @ 1 ""'-.;~_:'" D :irf,..... - "fl. ,/:!-,~~ ~~ E - .....-J,o. P ... " :... ; T - ...... H .. ~

I DEEP THERMOCLlNE------

FISH CAPTURED FISH ESCAPE

Figure 9 - Sea temperature structure and its relation to purse seine efficiency ITS RELATION TO U.S. TUNA FISHERIES

250,000

N= 1,159,461 FISH SOURCE: CLEMENS '(1961)

200,000

I.J.J a:: 0 u « 150,000 CD «....J I.L. 0 a:: I.J.J 100,000 CD ~ ::::> z

50,000

15 20 SEA TEMPERATURE (OC) AT TIME OF CATCH

Figure 10 - Relation of albacore tuna catch to sea surface temperature 66 SEA-SURFACE TEMPERATURE

200

en I- N=777 SETS SOURCE: lATTC W en w 160 z -w en w en a:: => 120 a. --' => l.L. en en w u 80 u => en l.L. 0 a:: 40 w co ~ => z 0 15 20 25 SEA TEMPERATURE (OCl AT TIME OF CATCH

Figure 11 - Relation of bluefin tuna catch to sea-su~face temperature ALBACORE TUNA OF OF OF ~ 35 40 45 50 55 50 55 70 75 80 85 35 4,0 4,5 5,0 5,5 60 6,5 7,0 7,5 8,0 8,5 35 4,0 4,5 5p 5,5 6,5 .7p 7,5 8,0 8,5 • h:~::::"i:::::,::;::'\ 0 1 .-':"';","":':",::,:1 Ii 0,1 ',,, 1"""":4" ,,, o I i 1 2 i 2 2 t: 4 t: 4 t: 4 ~ ~ ~. ~6 S5 S6 ~ ...:>: i= i= :1;8 1>.8 :1;8 CI ~ CI til 10 NO FISH 10 FISH 10 FISH DEEP ~ H 121 ' 121 ( 121 ' @ C5 OF OF 'F c:: 35 4,0 4,5 5,0 5,5 60 6,5 7,0 7,5 8,0 8,5 35 40 45 50 55 60 55 70 75 80 85 35 4,0 4,5 5,0 5,5 6p 6,5 7,0 7,5 81l 8,5 . oI } !:~::::,::::,i::,::::;:,;:j ) I ',, ! 'b>.::::::I:;:,Jd ' ,, oI k:,,:,::,j,:,;::::;:j (JJ 2 2 2 ~ t: 4 t: 4 t: 4 ~ ~ ~ 0; S6 Ss S6 i= :>: :>: :1;8 li: 8 li: 8 SURFACE I w w CI CI CI [;:J BLUEFIN (JJ 10 NO FISH 10 FISH 10 DEEP ALBACORE 121 I 121 ' 121 '

BLUEFIN TUNA

~ Figure 12 - Relation of albacore and bluefin tuna to vertical temperature structure off Soutbsrn California

~ 68 SEA-SURFACE TEMPERATURE

UNITED STATES DEPARTMENT OF THE INTERIOR BUREAU OF COMMERCIAL FISHERIES

CALIFORNIA FISHERY MARKET NEWS MONTHLY SUMMARY MAY 1968 PART II-Fishilla Information·

Sea-surface temperatures. fishina and research information of interest to the West Cosst tuna-fishing industry and marine scientists

CONTENTS Page Temperate Tuna Forecast for 1968 ,...... 1

Albacore Season--1968 OperatlollB ...... ••...... •.. 6

Basking Shark Fishery Needs Help...... 10

Sea Surface Temperature Charts, Eastern Pacific Ocean, May 1968 ...... •.. 12

U. S. Department of the Interior Bureau of Commercial Fisheries Fishery-Oceanography Center P. O. 80x 171 La Jolla, California 92037

-Port J• "Fishery Products Production ond Market 0010" IS Issued by the BUft!!OU 01 Commercial Fisheries, Morkel News Service,U,S.Cuslom House, Terminal Island, Colilornio

Figure :3 - Sample title page of the monthly publication i8sued to the fishing and scientific communities 57 SEPTEMBER 1-30, /967 FIG. i-MEAN SEA-SURFACE TEMPERATURE {GF.I 52

59 /!$"

59 ~ iil t-< ;., >-3 H 0 Z >-3 66 I 0 c: .(JJ ~ ~

'o;l tiJ tE ::0 t;j (JJ

"76.

0 1800 1700 1600 150° 140" 1300 120

Figure 14 - Sample mean sea-surface temperature chart 'from the monthly publication (Units in OF)

& 70 SEA-SURFACE TEMPERATURE

september 16-30, 1968 Part II, Supplement 2

Sea-surface Temperature of. I

CALIFORNIA FISHERY MARKET NEWS - MONTHLY SUMMARY -SEPTEMBER 58 U.S, nEPAHTMENT OF THE INTERIOR Fish and Wildlife Service Bureau of Commercial Fisheries Fishery-Oceanography Center P.O. Box 271 La Jolla, California 92037

v-~t--_62

53 66

:~ 68 'L .74

72 72·

liS'"

Figure 15 - Sample IS-day mean sea-surface temperature chart published between April 15 and October 31 each year (Units in OF) ITS RELATION TO U. S. TUNA FISHERIES 71

300

7 70 2S' 13S' 130' 12S' 120' liS'

SOO JULY 16-31, 1966 3 PROGNOSTIC SEA - SURFACE 4S' TEMPERATURE .~ COLD

WARM .. 40' .2 SEA -SURFACE TEMPERATURE ANALYSIS

3S' 3 PROGNOSTIC ERROR

2S' 13S' 130' 12S' 120' liS'

Figure 16 - Sample of prognostic 15-day mean sea-surface temperature chart and prognostic error field (Units in of) 72 SEA-SURFACE TEMPERATURE

UNITED EITATES DEPARTMENT OF THE INfERIOR U.S. FISH AND WILDLIFE SERVICE BUREAU OF COMMERCIAL FISHERIES Fishery-Oceanography Center La Jolla, California 92037

BULLETIN N8. 68-8

ALBACORE PRODUCTION LEVELS HOLD AS SEASON NEARS END

Albacore production levels continue to increase the all-time landings record in Oregon as the 1968 fishing season nears termination. At the present time, Oregon landings are estimated to total between 34 and 35 million pounds. California porb3 have registered slight improvement over what has been a dismal seaBon since the start, and the production to date amounb3 to only 5 million pounds. California's poorest year previously was in 1941. when only 4 million pounds were produced.

Fishing action has held up well, even though annual cooling associated with the first fall storms haa begun. The centers of fishing activity remained in two areaB: one between Columbia River and Gray's Harbor from 40 to 60 mUes offshore, and the other in the Point Arguello-Monterey region off Central California. Halibut schooners have done consistently better than moat jigboab3 trolling in rough weather In the northern region; baltboats logged a few good days with scores to 2500 fish per day off Oregon around September 23. Jigboat fiBbing off California has been spotty and subject to frequent changes in wind flow patterns; best fishing occurred when Santa Ana-type weather prevailed over Central and Southern CaUfornia.

Weatherwtse, much depends on how the fall storm systems affect Pacific Northwest coastal waters. The downward sea surface temperature trend was checked briefly during the past week when relatively calm weather prevailed over much of the west coast. Stormy, windy weather Boon will remove heat from the already-thin warm upper mixed layer tn Oregon and Washtngton, precipitating the annual otfshore migration. We have 80me. evidence that albacore which were tn the southernmost extension of their range this year off Southern California are com menctng to migrate northweBtward and offshore in the 60_66° F optimum temperature zone; theBe fish have yielded a few good catches over brief periods 35 to 50 miles off Potnt Arguello and Morro Bay in the past week.

The outlook for the next 2 weeks looks good unless bad weather commences to close in on the Pacific North west. Oregon should expeot to receive another 4 to 5 million pounds in October, thus bringing the state total to 38 to 40 million pounds for the year. By the time boats returning to California porU! unload fares caught in Pacific Northwest waters, adding to that being taken locally, landings should fall between 8 and 10 mUlion pounds. Thus,

total Pacific Coast production is expected to fall between 46 and 50 mtllion pounds I plactng 1968 well above average despite major geographic dislocations in the centers of avaUabUtty. Our previous estimate of 40 to 42 mUlton pounds was predicated on earBer termination of the Pactfic NorthweBt fishery and hence lower September landings figures.

September 30, 1968 Glenn A. FIUtner Leader, Fishery-oceanography Program

Figure 17 - Sample of bulletin issued at irregular intervals during the albacore tuna fishing season 300 COLUMBIA 16° 300~ PAPA 0 200 W 200 ~..\ 1'5 12° 100 12° 100 "'. . , 0 ...... 10° 0 9° -100 So -100 '"'" -~ -200 6° -200 ------"'"'" "1J -300 -300 3° ...... J -. '" . }J H E f-3 U 500 300 NOVEMBER 27° - (f.l ':-- 400 15° 200 li1 - 240 w ~ 0 300 100 0=:: f-3 U 200 .. 11° 0 21 ° ::::> H .... _------...... ~ - 100 -- -100 ISO: f-3 >< 0 7° -200 0 ::::> w c:: ..... Q. -100 -300 15° (f.l IL ~ . .... J J ....w ~ «w 400 FARALlON ISLANDS ISo 200 27" ::I: 300 16· 100 OJ ,'" 24· b:i 200 ..,...... --...... 14° 0 , fij \ ~ 100 ",," ' .... 12" -100 \ 21° ------", '" \ [;j 0 10° -200 '" (f.l ...... ,'"'" \ ISo -100 S· -300 ---""'" -200 6° -400 "-:l15°

~TOTAl HEAT FLUX --- SEA -"SURFACE TEMPERATURE

Figure 18 - Average in situ heat flux and sea-surface temperature curves for selected stations in the Pacific Ocean (1961-1965 base)" -..J 'vJ ....;j oF SHIP STATION PAPA (50 0 N,145°W)

\ ('\ 1\ ( il \ ; l \ /x \ ('/.. I \/\ I +200-1 \ \ f \ , \ I\ 1-12.0 I' I I ->. I c I , , "0 ...... 0 8.0 '"E -u (,) 0 ...... - - (j) c lLJ I>J (,) -200 4.0 \I a:: I - V ~ "(j) x " t- §i ~ 1 1 1 I 1 1 « ";j -J 1961 1962 1963 1964 1965 1966 a:: @ I...L w " 1 I I 1 I 1 a... t- , :::!: « w w +200 24.0 I::>:l :c t- « i'3 t- LU ~ W en z 0 20.0 I I I \ , . I' I I I V -200, I 1-16.0 'v I \ '. SHIP STATION NOVEMBER (300 N,140 0 W)

Figure 19 - Monthly mean in situ heat flux and sea-surface temperature curves for Ocean Station Vessels Papa and November, 1961-1965 ITS RELATION TO U. S. TUNA FISHERIES 75

NEGATIVE GRADIENT ""'-~- ~ .,, .' " -A.1 TEl/PERATURE SOUND VELOCITY

''''''''''""1111111111111111111 -

111111111111111111111.

POSI TI V E GRADU:rliT TEl/PERATURE SOUND VELOCITY

....:z: Q" ...o ~~j ~I" ,'. SLIGHT NEGATIVE GRADIENT TEl/PERATURE SOUND VELOCITY

- ''''''''''''111111111111111111111111111 ....:z: Q" ...o

DEEP WATER HI/PERATURE SOUND VELOCITY

....:z: Q" SOUND CHANNEL ...o ~"'IIIIIIIIIIIIIII"'IIIIIIIIIII"""IIIIIIIIIIIIIIII1111111111 1 '

Figure 20 - Relation of sonar sound velocity and sound ray paths to vertical temperature profiles in the upper layer of the sea --'J 0\ VARIABLE DEPTH SO;\iAR

.------.\ ... TEMPERATURE SOUND VELOCITY , I i

.• "1111 11111 1111111 gj :r

% ~ Q. lU I o ~~ I~' """111111111111111 I

Figure 21 - Comparison bfftween hull-mounted and variable-depth sonar efficiency under identical seactemperature profile conditions SST PATTERNS IN THE NORTH-EAST ATLANTIC

Captain J. D. Booth, R.N.

In this short paper, I would like to outline some of the problems whioh we in the Royal Navy have found in tackling the routine synoptic analysis of sea-surface temperature over parts of the north-east Atlantio, and to mention some of the characteristics and peculiarities of SST pat terns which have been noticed.

The chief reason for attempting such analyses was because of their relevance to oceanography and sub-surface thermal structure. In order to try to arrive at something approaching a true picture of SST distri bution, and since SST is undoubtedly affected by changes in atmospherio processes, it was necessary to consider it on a similar time soale to that of the weather - i.e. synoptically. The synoptic treatment of SST, however, poses many problems which are not easily solved. In the first place, a considerl',ble number of data were needed. These were not, and still are no~forthcoming over large areas of the north-east Atlantic which are little frequented by ships - such as the Norwegian Sea. Even where ships' reports appear plentiful, there are seldom enough of them for a detailed synoptic analysis. An example of the number of data which may be expected on a typical day is as shown in Figure 1. Many more data are needed for SST analysis than for sur face-weather analysis, since the former is much more complex and far less well organized. Tongues of, warm or cold water may be of the order of 30 miles or so across, whereas weather features are mere often meas ured in hundreds or even thousands of miles in extent. In order,there fore, to make up for the sparsity of data, the intervals for collection have to be lengthened, and here a compromise must be reached between a sufficiency of data and a significant change of ssT during this period of 78 SEA-SURFACE TEMPERATURE

collection. Data collected over several days result in a composite chart which attempts to show a reasonably current picture, but which really represents more of·a short-term average state of affairs. In areas where data are fairly plentiful, about three days would normally be needed to yield the required amount, but in sparsely covered areas this has to be extended to five or ten days or longer. Typical coverage of SST reports in the north-east Atlantic aver a three-and-a-half-day interval is as shown in Figure 2. The other major problem in making an SST analysis lies with the quality of the data. Whether one favours bucket or intake, and these methods supply almost all SST data, both are likely to produce errors .in unskilled hands. This has already been discussed at length and I do not wish to labour the point. However, to give one illustration, a recent careful comparison of bucket and intake temperatures made in a naval ship in an area where SST was changing rapidly showed differences between the readings averaginglOC with a maximum difference of 20 C. The bucket was carefully used by a meteorological unit, whereas the intake temperatures were as obtained from the engine-room staff. The intake was located at a depth of 28 feet. The intake thermometer was then changed for an ac curate meteorological thermometer, and it was located properly. in the oil bath provided. The previous thermometer had only just penetrated the top of the oil, which was six inches deep, and in which there was - at the time - astable temperature gradient of nearly 20 C. After this had been done, a. long series of bucket and intake temperatures gave an extremely good correlation between the two, with differences always within 0.30 C, and usually within O.loC. There was also a negligible time lag between the bucket reading and the reaction of the intake reading. There are, of course, other difficulties involved with intake readings, but neverthe less it would seem reasonably simple to eliminate or at least reduce the observational error in such readings,providing some simple precautions were taken.

Another problem in the synoptic treatment of SST concerns its local variability. How accurately does the sea know its own temperature, PATTERNS IN N.E. ATLANTIC 79 or what is the ambient thermal noise? This would seem to depend upon the place and the weather, and Captain Wolff has suggested that it might be as much as 0.20 C in a quarter of a nautical mile. Ambient thermal noise apart, however, the quality and quantity of data clearly set limits to the accuracy and degree of detail which can be shown cn an SST chart. If the grid is toc large, as it so often is, fine detail just cannot be determined.

Our early charts were based on the philosophy that SST patterns were dominated by interlacing tongues of cold and warm water which were persistent, this persistence being maintained in a general way for long periods, and to some extent even throughout the year. It was therefore necessary to establish a basic pattern and to use this to some extent as a ~emplate for analyses. Our practice was to draw charts every five days based on ten daYff worth cf data. There was thus a five-day overlap with each observation appearing en two consecutive charts. A typical SST pattern was a9 shcwn in Figure 3. Becaus e of this philosophy and as a result of unreliable data, the resulting analyses usually exaggerated the horizontal temperature gradient. Of course, the absolute values of isc therms varied but' apart from minor perturbaticns the pcsitions of warm and cold tongues appeared tQ change little.

In 1964, we carried cut a comparison of actual SSTs and SST charts over a large area cf the north-east Atlantic using data gathered for a particular month in 1964, and for the period exactly one year earlier. It was some reassurance to us when it was possible to show, by a statis tical analysis, that not only were there conservative properties of actual SST over this period, but that there was also a marked measure of persis tence in the patterns themselves.

However, notwithstanding this reassurance, these complex charts, in which long "ccls" of S,ST isotherms were drawn in order to maintain tongues, became more difficult to believe. The exaggerated horizontal gradients often failed to materialize when the temperature profiles from ships fitted with continuous SST recorders were studied. Such profiles usually showed gentler gradients. 80 SEA-SURFACE 'IEMPERATURE

Oomplexit~ nevertheless,is only one of degree. Figure 4 shows an example of an SST analysis off the west coast of Scotland drawn from data obtained from an airborne radiation thermometer. The time was April 1966, and the aircraft was tracking over the area on legs 30 miles apart. Apart from the generally colder water near the Scottish coast, the pertur bations further to seaward showed warm and cold centres typically about 30 miles apart. I must, of course, stress the subjectivity in drawing such a pattern, and this clearly becomes all the greater as the obser vational grid increases. Subsequent traverses of the area by the air craft and by a survey vessel, however, showed that the main features of the pattern were identifiable for several days.

A few years ago, a careful study of SST was made by naval ships taking bucket temperatures in the vicinity of OWS (Ocean Weather Station) Juliet. The grid of observations was dense, about 5 x 12 miles, and every precaution was taken to ensure observational and navigational ac curacy. Isotherms drawn from the observations, perhaps optimistically though with some justification at 0.10 0 spacing, showed patterns typic ally of the form shown in Figure 5. Again, there was no doubt that con siderable subjectivity was involved in drawing such patterns, and firm conclusions were difficult to draw. Nevertheless, some tentative oon clusions seemed possible. In the first place, coherent patterns could be drawn in fine detail from a dense grid of accurate observations. The patterns showed warm and cold patches with distances between centres of the order of 30 miles or so. A tem;erature.di£ference of up to 1°0 between adjacent warm and cold patches was typica~ and the patches were sometimes intense. Warm or cold patches were often grouped in families of the same kind, which may have indicated that they belonged to warm or cold tongue. And, finally, there was good evidence of the persistence of these patches over several days. A ship which traversed some four to six days later - a track identical to one covered previously - rec orded SSTs which were in very close agreement •. The two. temperature pro files are shown in Figure 6. Throughout, the wind was generally between Beaufort force 3 and 6, and from various directions. PATTERNS IN N.E. ATLANTIC 81

It is clearly not possible to draw SST charts routinely over large areaB of the north-east Atlantic with anything approaching the detail of the analysis shown in Figures 4 and 5. There just are not enough data.

The SST analyses drawn at present in the U.K. by naval analysts are typically of the form shown in Figure 7. The earlier tortuous com plexities have to a fair extent been eliminated, resulting in gentler and, we hope, more realistic horizontal gradients; but emphasis on tongues at the expense of closed isotherms remains. The method of drawing, which is subjective, remains essentially the same, though data are usually col lected over periods of three and a half days. Virtually, all the data for these charts come from ships, since there are no routine ART (Air borne Radiation Thermometer) data available. It is of interest that when these analyses are extended towards the western Atlantic, although they matoh in a general way the ones produced in North America for that area, there are usually considerable differences in the detailed con volutions of warm and oold tongues. Much the same thing occurs, of course, with weather charts, even though these are much simpler, and it is no doubt due to differing techniques used and to subjectivity.

Mean sea-surface temperature charts are also drawn by the U·.K. Meteorological Office. The charts are plotte~ on a one-degree grid from computer output, which gives for each one-degree "square" the number of observations, maximum and minimum temperatures reported, and the computed 0 mean. Mean isotherms are drawn at 1 or4°C intervals for five- and ten-day periods respectively. All observations lying outside the long 0 term extremes for each 1 "square" are rejected, but this quality con trol is not considered satisfactory, and other possibilities are being considered including the use of continuously updated means and extremes, though this poses computer storage problems. A typical chart for the waters near the British Isles is as shown in Figure B.

Whatever the manner in which an SST analysis ·is drawn, however, the notable characteristic of some persistence of pattern is evident; nor is this persistence always confined to the short term. 8-2 SEA-SURFACE TEMPERATURE

In O-ctober 1963, a naval ship sailed from the south coast of England on a voyage which tcok her past Ushant and Fini~terre. Herdistant~reading sea thermograph consisted_ of an external sensor connected by a mercury in-steel tube to a recorder. Nearly a year later, in September 1964, she took a very similar track which was at all times within 20 or 30 miles of the 1963 track and usually much closer. The profile of SST recorded in 1963 was as shown in Figure 9. In 1964, at a time about three weeks earlier than the corresponding period in October 1963, the sea was gener 0 ally about 2 C warmer, but the profile was remarkably similar (Figure 9). Although this voyage was repeated in 1965 and 1966 by the same ship, for one reason or another, a continuous SST record was not kept in those years. Early in October 1967, however, the same ship again sailed south wards on a nearly identical track-to those taken in 1963 and 1964, and the SST profile was again recorded (Figure 9).. The 1967 profile can be seen to be generally about mid-way between those measured in the earlier years and, in fact, the time of sailing that year was also mid-way be tween the corresponding times sailed in 1963 and 1964. Once again, apart from a small _displacement (and this, might be partly attributable to small differences of track) the profile is in extremely close agreement with those recorded three and four years earlier. If we 100F closely at the bathymetry along the track taken in these instances, it is clear that some of the features of the SST profiles are associated with bottom topo graphy. A cross-section along the track is as shown in Figure 10. The continental shelf extends to about 47.50N before giving way to deep ocean water in the Bay of Biscay and then shoaling considerably, though only temporarily, near the coast at Cape Finisterre. The associated SST pro file shcwn for 1964 (Figure 10), when superimposed, indicates a steady, fall of temperature which reaches a minimum in the mouth of the Eng~ish Channel just north of Ushant, but still well on the continental shelf. Then comes an abrupt rise of temperature of over 20 C in a very short distance. Pa~sing from the continental shelf, a small fall occurs, fol lowed by an abrupt rise in temperature. Across the Bay of Biscay, tem- , 0 perature rises with decreasing latitude', when at about 44 N an abrupt fall _followed to th~ south by an equally abrupt ~ise- and then further fall appear to reflect bottom contours. PATIERNS IN N.E. ATLANTIC

The SST profiles illustrated do not necessarily indicate that similar ones persist throughout the year, but the extremely high correla tion between them suggests some quasi-permanent control of SST distribu tion in this area, and the evidence points to bottom topography and, or in association with, currents.

Another example of bottom topography influence on SST is to be found much farther out in the Atlantic. The British weather ships oper ating out of the Clyde and proceeding to OWS India and Alpha normally follow identical tracks as far as India, which take them directly over the Rockall Bank. This feature rises from an ocean floor of over 1,200 fathoms to a bout 80 fathoms. A s tud.y of the SST profiles recorded by these ships shows that in just about every case a drop in SST of 1 to 0 1.5 C takes place on the eastern flank of the Rockall Bank as the ship ·proceeds north-westwards. Sometimes, the fall is abrupt and the lower temperature persists for 60 to 90 miles or so along the track to OWS India, after which distance some recovery takes place. At other times, the fall may be more gradual and persist over a longer distance.

A typical SST profile and simplified bottom profile along this track are as shown in Figure 11. This colder patch over Rockall appears to occur in both winter and summer.

A similar though less marked effect often also occurs when ships pass near or over the Porcupine Bank (about 85 fathoms), which is almost en route for weather ships travelling from the Clyde to OWS Juliet •. No such effect, however, has been noticed as weather ships cross the Reykjanes Ridge to the south-west of Iceland, en route to station Alpha. It may be too deep. In the Iceland Faeroes Ridge region, the cold,. deep water from the Norwegian Sea overflows across the ridge into the Atlantic Ocean. For the most part, it is overlain at the surface by Atlantic water. There is considerable evidence to suggest that bottom topography in the area is responsible for the complex distribution of water temper ature at depth, and that this control is reflected, albeit to a much smaller extent, in SST patterns near the ridge. 84~ SEA-SURFACE TEMPERATURE

There are doubtless many other examples of bottom topography influ ence on SST distributicn in the north-east Atlantic.

In summary, it can be said that despit~ the gen~ral absence of the great water mass ccntrasts which exist in parts of the western Atlantic, some fairly complex patterns of SST distribution do occur in the north-east Atlantic. Where the contrasts of SST are most marked in the north-east Atlantic ~hat is, in high latitudes) synoptic data are,however, normally inadequate. Detailed analyses which can be drawn only when data grids are small show coherent patterns with identifiable features commonly separated by about 30 miles or so. Patterns on this scale have a strong tendency to persist for several days, and on a grosser scale, some persistence of pattern, or at least pattern repetition, may be considerably longer - at times of the order of a year or more. Despite the modification of SST patterns undoubtedly induced by weather changes, the characteristic of persistence indicates steadier controls on SST distribution, such as drift currents, and in some cases bottom topography in association with deeper currents. The form and degree of complexity of SST analyses depend upon the data available, the method of handling the data, and often on a good deal of subjectivity., The two great handicaps of the analyst are, as ,already stated, the quality and the quantity of data. So far as the -latter is concerned, unless some means of measuring SST other than from ships becomes routinely available, on an extensive scale, there seem likely to remain large areas where reliable synoptic analysis will be extremely diffioult, even in an otherwise well-frequented north-east Atlantic. (A modification of the abbreviated ships-reporting codes, however, so as to include the parameter SST, might do something to im prove the position.)

And finally, SO far as the quality of data is concerned, I think the following quotation is relevant:

"Accordingly, everyone who uses the sea is invited to make certain observations; or, in other words, propound certain queries to Nature, PATTERNS IN N.E. ATLANTIC 85 and to give a faithful statement of the reply she may make." So wrote Maury in his report of the first International Meteorological Conference in Brussels in 1853. His report goes on to say that: "Unless we have accurate instruments, instruments that will themselves telr the truth, it is evident that we cannot get at the real meaning of the answers that Nature may give us. An incorrect observation is not only useless in itself, but when it passes undetected amongst others that are correct, it becomes mischievous • •.•" It is my belief, gentlemen, that we can do a great deal to improve the quality of SST data. 86 SEA-SURFACE TEMPERATURE p,r-.:;.·;o;~:-:-L:;-~:-:-... :::.~:::.~-::-'.··.:;:'~~: :-:.::..::'.. r:-::;;=:;;:;=:;::;:--::;:-'=;;;::~:;:;::;;:;;::-:;:;;;;:;:;;:;-::;;;;:;;;;::-:--;-=-:--l;i'~" : ./., l ~\ ". .." ..... >.~i",~~::~·::\·:: ~\~-~--t ~ -" \-J....-----.-!>- - ;0'

...i!,Y'\.. \ '.. .; PATTERNS IN N.E. ATLANTIC 87

F /

\ \ .< 88 SEA-SURFACE 1EMPERATURE -.. ; 1 I ,'! I PATrERNS IN N.E, ATLANTIC

~58' --..]...7

• ~ p",,;' •~ . _:II!?' , ~5r o

......

- 56'

10 I II" 10" .,.,:.:.: ., :::: FIGURE 4

SST chart Isotherms °C' 90 SEA-SURFACE TEMPERATURE

""", \

FIGURE 5

Isotherms °C PATTERNS IN N.E. ATLANTIC 91

.0 ~~ _.. 4t l' III.. 2 ~ ~ ~ .,) t +' .... '.,. m ...... > 'rl t •• '"oS ,/ \;.. <.... ~"' • T "ro m "'"-, ~. ".- ~ • f- .!4 "ro +'" .!I ';;j " ... ."+' 0 ~ ~ <; '0." ?I f/. I "" 11/ "' ..ct. .~ "".. '"m t:. ";s ~i bO ·rl i-~( ~ "" Ii:

..,: --.,ir------"'I.--...;..--~·~ ...... -."I'. ..Jl.o ~ oV ~_ °tlO . I::- V)- 92 SEA-SURFACE TEMPERATURE

. .: \ \ 1-\~--~~ '-...- . '. - 1, \ \ \ . - _.\ .. \ I -" \ ~ \ . \ ~ -.--l-...

Figure 7 - SST analysis, 221200Z - 260001Z, July 1968 PATTERNS IN N.E. ATLANTIC 93

's S"OA"( MeAN eA ISonteRJ\4$·C I'~· ,...;rllj,.. - ~ 19"& .

FIGURE 8 \0 oF

+ ..oN 4ioN D 4-4-°N UON "C SOaN 4fN "7°N 4S N I I I I I I I I 19

.:fer-I. I'U.. ( ~ l>.~ _ JJ"<' ~ 17 -l """'-£ - \ /I l I f;'.J :<> I (J] 16 -I "- OCT. 1967 , II/ 1/1la ./VV I ./ "- c::: ~ :<> ~ 15 I - =ta c{ \. t • r" { J'IA('.../ " ~ 14 -l J\ V ~

13 -l OCT.!1963 i I 12

Figure 9 - SST profile, English Channel-Ushant-F1nisterre + ·C 11

11 SST PROFILE SEPT, 1964 + 17

.." 15 :t> ~ ::d OEPTH ~ 14 FATHOMS 1ii f'~~~'~~~~~~~~~~~~~~~'\----- 13 !z! Z !lDO I.'l

1000 :t>

BOTTOM PROFILE .~ 1500 H ENGLISH CHANNEL- USHANT- 0

FINISTERRE -2510

\\\\\\,-,,,,, .

'--~~--'--'------~-Ti-~--~---'i~--~~-' -I I i SOoN ..,ON ..oN ..7 oN ...oN 4SoN 44°N 43°N.

Figure 10 - Latitude

\D VI 96 SEA-SURFACE TEMPERATURE

~ \ I I I \ I ~ .... I o <:: \ Q) \ ! ...... , Q) .0 '( UJ Ql I ,.-{on 'H o H ) 0. I S ",,'" o , +' +' o I .0 \ - . ~ Ii ".", "'" . 8 . ' '. CfJ , CfJ ~ 1\ I I 1\ I I - VARIATIONS DE LA TEMPERATURE DE LA MER AU VOISINAGE DE LA SURFACE

par

J. Romer

Meteorologie Nationale, France

Introduction

Parallelement aux mesures comparatives de la temperature de 1a mer en surface par les differents procedes de mesure (seaux de types divers, prise d'eau de circulation, et~ •••)t nous avons pense qu'il serait utile de se faire une idee plus exacte)R I'aide de mesures fines, de la strudUre thermique des couches super ficielles et de sa variation dans l'espace et dans Ie temps, ces mesures ~tant effectuees au large sans atre perturbees par Ie voisinage de la cote ou par la proximite d'un navire oU d'uneplate-forme. Une tel1e connaissance devrait per mettre d'evaluer la valeur et la representativite d'une observation effectuee en un point a un instant donne en vue de son~ilisation a l'echelle synoptique et egalement de definir ce qu'on entend par temperature de la mer en surface.

Dans ce but, DOUS Rvons utilise las mesures effectuees p~r une equipe dirigee par M. GONELLA du Laboratoire du professeur LACOMBE a bard d'une bouee-satellite de la bouee-laboratoire mouillee en Mediterranee dans Ie cadre d'une convention entre la Delegation Generale a la Recherche Scientifique et Technique (COMEXO) et Ie Museum National dtHistoire Naturalle. Cas mesures sont utilisees par ce laboratoire pour l'etude a l'echelle fine des interactions ocean-atmosphere.

La bouee-laboratoire mouillee a mi-distance entre NICE et la Corse par 42°47'N, et 07°29'E, par fond de 2680 matresJconstitue un point d'observation extremement interessant surtout en ce qui concerne les interactions ocean-atMosphere.

On peut schematiquement decrire la bouee comme une perche de Froude de 60 m de longueur dont 50 sont immerges. L'ensemble pase 250 tonnes, Ie lest, pour sa SEA-SURFACE TEMPERATURE part, attei~nant 115 tonnes. La partie superieure emerge a 10 m du niveau de la mer et comporte 3plate-formes delimitant d'une part le local machine at d'autra part le local laboratoire - carre - couchettes. L'equipage permanent est de 2 personnes et 4 passagers scientifiques peuvent etre embarques.

A la bouee-laboratoire est associee une petite bouee-satellite a une distance de 200 m environ portant divers capteurs de temperature at de vent.

Les capteurs de temperature, au Dambra de 10, sont des resistances de platine. Sept sont immerges aux profondeurs 0,1 m; 0,5 mj 1 m; 2 Mi 3 m; 4 m et 5 metres; trois sont dans l'air aux altitudes de 0,5 m; 1,5 m et 3 metres.

Les capteurs sont relies a una centrale de mesures situee sur la bouee laboratoire. Les capteurs de temperature presentent un temps de reponse de 15 a 30 secondes dans l'eau. Un pointe est effectue a chaque niveau toutes les 15 minutese

La precision des mesures de temperature est evaluee a 0,02° / O,030 C dans l'eBu. La sonde a 0,1 m dans l'eau n'est pas protegee durayonnement ; Ie cou rant de surface est toujours suffisant pour dissiper eventuellement les calories absorbees par rayonnement (20 a 100 cm/s).

Les mesures interrompues en 1967 seront reprises debut 1969 et 1es resul tats qui suivent seront completes lorsque DOUS disposerons d'une serie de mesures plUS importante.

Les mesures dont nous disposons Bctuellement, repartees sur cartes perforees, sont au nombre de 3614 et correspondent a 1a periode du 16 mai au 10 aoUt 1966.

Structure thermique vertica1e dans les 5 premiers ~etres

Nous avons evalue l'ecart de temperature TO,l T5 entre le niveau de 0,1 m et Ie niveau de 5 m. Sur un total de 3614 mesures 1540 font apparaitre un ecart de TO,l - T5 inferiau~ ou ega1 aI/10°, 253 un ecart superieur a 1/10° et inferieur ou egal a 2/10°. 428 un ecart sup~rieur a 2/10° et inferieur ou egal a 5/10°, - 1393 un ecart superieur a 5/10°. VARIATIONS AU VOISINAGE DE LA SURFACE 99

II Y a lieu de noter des maintenant que ces mesures correspondent a Une saison favorable a l'etablissement de gradientseleves dans les couches super ficielles. II est tres probable que Ie pourcentage de cas d'isotheraie sera beaucoup plus eleve si l'on effectue une statistique sur une annee complete.

Variations de la structure thermique verticals dans Ie temps - Evolution diurne

Les courbes tracees a l'aide du calculateur Cdc 6400 de la Meteorologie Nationale sont de deux types : d'une part, les courbes obtenues en portant pour chaque niveau en ordonnees les temperatures mesurees toutes les 15 minu tes de Oh a 24h, d'autre part les courbes correspondant aces mames mesures et lissees sur une periode de Ih30 de fa~on a reduire les fluctuations de periodes comprises entre une heure et deux heures qui semblent se produire frequemment sur l.s graphiques en question.

Nous avons trace ces courbes pour 30 journees completes de Oh a 24h, entre mai et aoOt 1966.

Les courbes font ressortir la grande complexite des phenomenes qui viennent perturber la stabilite de la structure thermique (mouvements verticaux et horizontaux, ondes internes) et il ne sera raisonnable d'en entreprendre une analyse detaillee que lorsque nous serons en possession d'un nombre suffi sant de series de mesurese Ces analyses devront tenir compte des divers para metres meteorologiques et oceanographiques.

Dans Ie tableau ci-joint nous avons porte pour chacune des journees etudiees

colonne 1 la date.

colonne 2 lesconditions meteorologiquea moyannes au cours de la journee nebulosite et vent,

- colonne 3 : l'amplitude de la variation diurne de la temperature au niveau 0.1 m, c'est-a-dire la difference entre la temperature la plus haute et la temperature la plus basse observee. avec l'heure du minimum et l'heure du maximum. 100 SEA-SURFACE TEMPERATURE

Las amplitudes se repartissent co..e suit pour les 30 journees etudiees

AO.I'IO 18 jours 1°

Dana Ie cas d'une evolution non perturbee (absence de perturbations meteo rologiques, absence d'ondes internes importantes), Ie minimum se situe Ie plus generalement entre 4h et 7h (15 jours sur 30) et. Ie maximum au cours de Itapr~s midi et plus particuli~reaent entre 15h et IBn (Tu+lh) (14 jours sur 30).

- colonne 4: les m'mes elements que ci-dessus pour Ie niveau 5 m.

Les amplitUdes se repartissent comme suit pour les 30 journees etudiees

A5,"O,5° 12 0,5° <'A5 "-1° 9 1° <"A5'-1,5° 5 1,50 <-A5 '2· 3 2· .( A5 ~2,5° 1

Les minima se situent generalement entre Oh et 12h et les maxima entre 12h et 24h, mais leur repartition au cours de la journee est beaucoup plus irreguliere qutau niveau 0,1 m. II faut sans doute voir la lteffet de perturbation apporte j entre autres, par des ondes internes.

Dans Ie cas dtune evolution non perturbee, l'amplitude au niveau 5 m est environ 2 a 3 fois plus faible que ltamplitude au niveau 0,1 m.

colonne 5 les valeurs max~male Gmax et minimale Gmin de l'ecart de temperature entre les niveaux 0,1 m et 5 m au cours de la journee ainsi que ltbeure a laquelle s'estproduit Ie maximum. Ltecart est generalement faible et mOme souvent nulla nuit.

Dans Ie cas dtune evolution non perturbee, il croft au cours de la journee, atteignant sa valeur maximaIe au moment du maximum de temperature du niveau 0,1 m pour decroltre ensuiteo VARIATIONS AU VOISINAGE DE LA SURFACE 101

Les valeurs maximales se repartissent comme suit pour les 30 journees etudiees

/ I" Gmax ... 18 jours I" .(Gmax (2" 5 jours 2" ~GmRX L3"- 4 jours 3° (Gmax (,.4" 2 jours 4" .(G max t..5"~ 1 jour

- colonne 6 : les valeurs minimales et maximales de ltepaisseur de la couche isotherme superficielle.

NallS :entendonspar epaisseur de la couche isotherme superficielle la profondeur du niveau Ie plus bas pour lequel la temperature differe de moins de 0,1" de celIe du niveau O,lm.

L'evolution diurne de l'epaisseur de cette couche est analogue a celIe de l'ecart des temperatures a 0,1 m et a 5 m (colonne precedente). Elle atteint et depasse souvent 5 m 18 quit et ne devient inferieure a 0,5 m au cours de Is

journee que pendant des periodes generalement courteso

o o 0 <." Nous avans reproduit (fig.l a 9) les courbes 11ssees correspondant a une selection de 8 journees particulierement typiqueS, ainsi que, a titre d'exemple, les courbes non lissees pour la journeedu 16 juin.

Chaque graph1que comporte les 7 courbes correspondant aux niveaux 0,1 m, 0,5 rn, 1 ro, 2 m, 3 ro, 4 m at 5 m.

Periocte du 18 au 20 mai (figures I, 2 et 3) ( Situation meteorologique non perturbee, Meteo ( Vent generalement faible, ( Ciel peu nuageux. Evolutipn diurne aSGez reguliere, perturbee dans les couches inferieures, notamment les 19 et 20 mai, vraisemblablement par des phenom~nes d'ondes internes. 102 SEA-SURFACE TEMPERATURE

Le 8 juin (figure 4) ( Situation meteorologique non perturbee, Meteo (_ Vent faible, ( -Ciel peu nuageux. On notera ici l'amplitude elevee de la variation diurne dans les couches superieures et Is valeur elevee du gradient vertical dans la journee. On notera egalement l'evolution analogue dans les niveaux superieurs avec dephasage du maximum dans Ie temps (preponderance des phenomenes de conduction) et l'existence ctJun niveau limite a temperature constanta a 3 m.

Periode du 16 au 20 juin (figures 5, 6, 7 et 8) 16 juin ~ - Situation meteorologique non perturbee, Meteo - Vent faible, ( Ciel peu nuageux. Forte variation diurne et fort gradient. On notera Ie decalac_ dans Ie temps du maximum jusqu'a 3 m. Le minimum entre 10h30 et l4h au niveau 5 m est vraisem~ab~tdQ a une onde interne d'une periode voisine de 24h; on Ie retrouve en effet au cours de la journee precedenU\ puis, plus amorti, au cours des 2 jours sUivants.

18 juin ( - Un vent de SW modere a sssez fort s'est leve dans la journee du 17 et Meteo ( persiste Ie 18. Gradient et amplitude&sont nettement moins marques que dans la journee du 16 jUin. 19 juin 1 - Un front froid, en provenance de l'W passe vers Oh accompacne d'un Meteo ( f vent de SW 25/40 nds soulevant une mer forte, Ie ciel restant degage ( avant et apres Ie passage frontal.

La temperature est alors la meme a tous les niveaux et decrott lineairem.nt de 2,5" entre Oh et 24h.

20 juin Le vent et la mer restant assez forts, les courbes restent confondues a tous les niveaux avec faible amplitude de la variation diurne (0,.-).

Nota Pour Is journee du 16 juin, nous avons reproduit les courbes obtenues VARIATIONS AU VOISINAGE DE LA SURFACE 103 directeaent Sans lissage (tigure 9). On pourra compar"r ces courbea avec c"lles de la figure 6 (courbes lisse"s pour la .eae journee).

Les courbes non lissees font apparaitre les fluctuations de periode de une a deux.heures dont nous avons parle au debut de ce paragraphe. On peut evaluer 1. valeur moyenne de l'ecart entre 1. courbe lissee et la courbe non lissee a 0,.2- environ.

Comparaisons des temperatures au seau et des temperatures aux sondes de platine Des mesures au seau Macabolier ont ete effectuees a toutes les heures synoptiques a la bouee-laboratoire. Nous avons compare les valeurs obtenues a partir de ces .esures au seau avec celles obtenues a la bouee_satellite ~ prises coame reference.

Nous avons admis que la structure thermique de la couche superficielle etait identique a la bouee-laboratoire et a la bouee-satellite sous les conditions suivantes : situations meteorologiques et oceanographiques stables, variations lentes de la temperature et du gradient sans trace de perturbation bru&que pouvant reveler Itexistence dtondes internes dans la cQuche des deux premiers metres. 212 mesures au seau ont ainsi ete comparees et se repartissent comme suit

a) cas d'une structure isotherme de la couche superficielle : 155 cas ----_...------_.------Nous avans pris comme critere d'une structure isotherme un ecart de temperature inferieur ou egal a 0t03° entre les niveaux 0.1 m et 2 m.

Ces 155 cas ont eteeux-m&mes divises en deux groupes 1) periode mai - juin : 99 cas 2) periode juillet - aout : 56 cas.

En effet, Ie thermometre a Mercure du seau Mecabolier avait ete remplace dans l'intervalle.

pour la premiere periode mai - juin, on a trouve un ecart moyen de 0,20° avec une frequence maximum de 0,25°. Pour la seconde periode juillet aout, on a trouve un ecart moyen de 0,04° avec une frequence maximum de 0,00°.

La difference entre les deux histogrammes peut provenir d'un mauvais etalonnage. 104 SEA-SURFACE TEMPERATURE

b) Cas------d'une structure non isotherme de la couche superficielle : 57 cas Sur ces 57 cas, 6 ont ,He .Hi mines en raison des anomalies qU'ils presentaient.

Les 51 mesures restantes f·ont apparaitre que les temperatures mesurees au seau se repartissent d'une maniere aleatoire entre les valeurs des tempe ratures mesurees a la sonde de platine pour les niveaux de 0,1 m et 1 m. Cependant, il semble que pour les gradients importants, la temperature mesuree au seau soit plus proche de la temperature de la couche de 0,1 m.

Conclusion Comme nous l'avons dit plus haut, l'etude (purement descriptive) entre prise a pour but de contribuer a se faire une idee plus precise de la repre sentativite a l'echelle synoptique d'une mesure isolee faite en un point donne, a un instant donne; et egalement de fournir une definition de l'ex pression "temperature de la mer en surface"o Les resultats presentes ci-dessus sont encore tras fragmentaires puisqu' ils ne portent que sur une periode de 3 mois. lIs devront ~tre completes par d'autres series de mesures permettant egalement la determination de la variation de la structure thermique verticale dans l'espace (gradients horizon taux) et comportant egalement la mesure de la tsaperature de la pellicule .superficielle (par radiometre infrarouge) et par une mesure simultanee de divers parametres meteorologiques et oceanographiques (vent, nebulosite, rayonnement, temperature de l'air, courants, etc••••).

Les correlations entre la structure thermique (et ses variations) et ceS divers parametres devront etre recherchees.

Grace aux possibilites offertes par la bouee-laboratoire associee a un reseau de bouees-satellites, nous esperons, en ce qui nous concerne, poursuivre ces travaux en Mediterranee. Pour parvenir a une connaissance plus complete du sUjet, il faudrait que des mesures analogues soient entreprises dans les differentes mers du globe ou les phenomenes peuvent certainement ~tre tres differents.

Ce travai~comme tous ceux concernant Ie domaine des interactions entre l'ocean et l'atmosphere, doit faire l'objet d'une cooperation etroite entre les meteorologistes et les oceanographes. TABLEAU DES VARIATIONS

FIGURES 1 A9 I-'o 0\

:I A 5 ! GRADIENT 1·~1.¥l Jl..Q..1. . : "valeur :: --- Prafandeur : nebulaeite (oct.... )" valeur - couche DATE :vent (naeuds) heure/heure heure/heure :G max : G min min~maxi mini!maxi : heure: -- j.eatherme • 2 ! : , ;; 5 oo,q 00,'> : 10 : 0 17 mai NE 5/10 1Oh/1 7h Oh-5h/16h30-1gh: 311 : (0 ,2 0,5 a).5 m "-::-::-i'.-~'-_.-:-- 2 0 0 ,8 . 2°,8 18 mai 2°,9 0°,4 var. 1/4 4h30/15h15 4h/13h30 ;1flh15 O"3m 3 2° : 0'0 5 3° 19 mai . , 0° ,8 0,5 a m : NNE 5/8 : 7h15/17h ;10h/2hI 5-13n15 17h a

o ,°,1 . 1 r ? : 3°,9 QO,7 20 rnai var. 1/8 5h15/17h30 ;lh,0/13h45et22h30;18h30 ° a 2 m ~ I 21. 1 a 8 1°,5 1°,3: 2°,2 :__n:.~__.21L.2 __ 4h30/18h15 Oh-20hI5/8h :20h15 ::° Oa)5m : 22 . 1 1 ° , 9 0° ,8 : 1° , 9 I ma~ . _: E_ SE 5 : 51145/15h30 : 12h/15h30 : 15h30 0°,6 0,5 a 3 m til "2 : 00 6 : 1° 9 : 1° 7 • 23 mai ~ S\{ 8/15 : 61145/12W; : 4h/18h : 12h : ° : 0,5 ")5 m 1 • 20 4 • 1° 2 "30' ") 24 mai ;_~'l.1_2-W4/1° ; 6h30/16hl <; ; 15h/Oh ; -.12.h!.L; 0 ; ° a 5 m__ • • 1 : 00 7 : 00 7 : 1 0: : 7 juin : 810/12 :6h30/12h30,,-17h15: 18h/21h :1F30:-Il!J: 0 : 0,5 a?~~_: I ~uin 4 : 4°,4 : 00" :4°,8 8 81/8 : 7h/15h15 : Igh30/31145 ..:15hI5 o ° R)5 m ... 3 a 8 : 0° , g : 1° , 2 : 1° , 4 o o a) 5", :_~"~_:9 E ~l:!L2_h .:_~l.1!.45.i.!.11l1..'L 4/12 _ : __: 11h30 : -: - ., : 10,? : 0°,11 : 10,4 14 juin o ~ 0,5 ,,)<; m NE 8/15,:!! ~__: 7h/15h : 7h/21 h30 _: 141145 .__.__._------.. • 3 : 1 0 : 0 0 ,4 : 0 0 ,1=\ 0°,2 15 )U1.n : _IT".2/10-1~JO/16 : 7h/1I)h45 : 6h/l gh30 :14'>45=1h%" 2 a)5 m : 0 : 30 4 : 1° 1 : 3° 5 16 jl)in ., • , • t .f " NE O/10-S\{ 10 • 5hl<;/13h45 : 12h30/24h :1,h45 ° 0"-) 5 m "0 ; 0°, g ; 1°,2 ; 0°,6 17 ~uin 4 a)5 m S\{ 10/15 : 3h/17h30 : lh30/18h15 plP.I2h45: ° 2 3 " 4 ..__• __ 5 6 . 0 QO,7 0°,4: 0 0 ,7 18 juin : 8W 10/14 6h/13hl~ o 0,5 a) 5 m --.--.-. - _ ..- 2 6h/17 h .:13h15 2°,5 2°,0;: 0 o >5m _____19 ~uin , __ SW 20;/4:.12... , ,----:-----: ) 5 ~ ,0 ~uin ? " 8 0°,4 0°,4 o 20 SW -S 20; 1 0 • 1 0 10' • 24 .iuin , , 2,h/1 1h4o; : _Q..W,,,,6,,,h,.,30,,-__ :11h15 ,0 : 0,5 a>5 m , .. . . . 0 0 ,7 :. 0 0 ,7 ~ 29 .iuillet: o 0°,2, 0 3 a) 5 m SW 10/20 3h30/15h15 : 3h30/15h15 ,11 h1 5 ~ : ----_.__ .,'----: :---_.._--; H o 0 0 ,4 0°,4 0°,2: 0 ,1 .iuillet: m Wl0-8W20/28 24h/14h30 24h/14h30 '9h30 )5 ~ ,------.: --: 0 0 ,6 OO,s 0°,6 :--0---:---0-,5a)5 m ~ 1 er aoll.t o W 6/14 5h/llh 5h/19h 11 h ----_.: : :----- ;3 0 2 aoll.t 2 0°,4 0 ,4 0°,2 o tiJ 8li' 1~/20 3h45/18h30 3h45/18h30 13h30 >5. ~ 3 aoll.t 2 a 6 0°,5 0 0 ,5 . 0°,2 ~ SI< 10;/20 Oh/17h30 Oh/17h30 ; 17h30 o 3)~m :-._------: ------,- -, til 4 aoat .J a 5 1 o. , 1° ,1°,3' 0 0a.)5m $: SEI~-Sl 5 m I< 10/30 7h/l1 h 7h/llh 11 h ,------,---6 ----"; 0°,15 ,; 00,6 ; 0°,4",,------• 7 aoat I< - SI< 8/15 : 6 h/12h30 , Oh/18M5 '17h o 2 a) 5 .; 8 aoat :=~~/~O -~ 7h?O/~;h~~ ~~1-5h~;i~h-3h15~l ;01~i, . . . 0. : '0,5 a)5 m 9 aout 2 '1°,6 , 1°,6 '0°,2: 0 : >5 m 8W-,{ 8/25 ,9h/17h30 , 9h/17h30 : 12h..iQC-..!..' .!..: --'-

VARIATIONS AU VOISINAGE DE LA SURFACE 107

, 6'

ECUEE LABc;AATaRE 'i'...2··P'N G.Or29' E 2 LE 10 MAL 1966

lI.b\l\.... I~., Zilf.J/6

1/.",'r: VA .. ~<10 .... e~/o~ \01" E:tq~ G'" Ie. "'..-, -i

23"

22'

21'

20'

",' IFIGURE 1

17'

15'

II 17 '\' 19, 20 T' ,f .. r i y",1p, '1 I' r T 108 SEA-SURFACE TIlMPERATURE

26'

2 '

BOUEE LABORATOIRE 'f.42'47'N G 07'29' E 22' LE19 MAl 1966

~ ..I>..\..;t.. , ...t1S'1' Va"t, 1'I"',".!,'l.T/il. S"'~IU 1:"1'..'

2 '

J FIGURE 21 Hr

15'

!l 17 13 ,1f 17 ., I of. ,,1,0, '1 '~ VARIATIONS AU VOISINAGE DE LA SURFACE 109

25'

23"

21'

20'

IFIGURE 3 I "''1 j i ; 17-:1,

1,0~~3~~7 10 13 i~2~2,,3 19

26'

BOOEE..l.AI3CRA"TdAE "..... 2·... 7N G.07"29'E 25' LE OB-.JUIN 1966

N....~""..,."', 2. ~ v. 'l.."h 5 GA/o...... (Tnt ~.. \ ...... , AI&.

23"

22'

21"

20'

19'

10' FIGURE 11

16'

15' VARIATIONS AU VOISINAGE DE LA SURFACE 111

I3CXJEE.LABORATaRE q>.42"4 7 N _G.O~29't. LE 16JUIN 1q66

N... b"\o,,,t":: 0 ~.vs

~o:~~: NI! c/~o'>t< ll.. SYII ..... 11'<

LTC,Td... "''''... : .'lIz,

23'

20"

19"

",' I FIGURE 5 I

17' 112 SEA-SURFACE TEMPERATURE

2 '

BOJEE.LAl3CW\TORE 'f.42·47'N 6.07"29'E 25' LE 1BJUIN 1966

r/"to ..\<>:,.ri". 0 &. 2/8 ~ .. nr, 5w .,lo(Alo-l'I"T E:t~, Ul<>. "'..... , 2. {ll 24'

23'

22'1- -'--_-~~;:;;;;;~;;;;:;;;===c:::::;

21'

20'

",' IFIGURE 6 ,

17'

15' VARIATIONS AU VQISINAGE DE LA SURFACE 113

2 •

BOl..EE.LABORATOiRE 4'.42",dN G.07"29'E 25' LE 19JUIN 1966

I/e't\~· SW z.S/~o\("r [t,,~ ~.. I. ,.,,_r: ~!S'

23'

21'

20'

19'

w' I FIGURE 7 I

17'

15' 114 SEA-SURFACE TEMPERATURE

EOUEE.LABffiATOIRE LE 20 JUIN 1966. .;2~47·N 25' 'f... G.07"29'E N,;b,,',,~,h!; It 0. SJS

'( .. nt: 50# 15 ....' J,. S toK'

Et"t &L I...... u, I,./S

23'

20'

10f~------I 1 11l'~ I FIGURE 8 J

, 17~ j

15'

7 n 1(,. , 21 I'i i f . 1p '1 lp i' I i VARIATlONS AU VOISINAGE DE LA SURFACE 115

BOJEE.LABc:::RATOIRE Cf'.42-,-,7N G.07·29E 25· U:: 16JUIN "1966

" ~: /1....

21·

20·

18" I FIGURE 91

SEA-SURFACE TEMPERATURES SOME INSTRUMENTS, METHODS AND COMPARISONS

Allan B. Crawford Republio of South Africa

There are many different methods of measuring sea-surface temper atures, and in the interest of uniformity, the Commission for Maritime Meteorology has been asked to study the measurement of this parameter so that meteorologists, oceanographers and related scientists can be guided in the selection of the various instruments and methods best suited to their needs.

Before listing some of the different instruments and methods, it should be quite clear what is the parameter it is desired to measure, because opinion seems to be divided on that important point. And until this matter has been settled, there could be a lack of uniformity in the data collected.

Let us briefly examine this question before describing the instru mentation, because the type of instrument used i~ of course, directly related to the parameter to be measured.

Temperatures which are measured near the surface can be classified in three different groups, viz:

(a) "Skin" temperature, which is the temperature of the thin film of water in direct contact with the atmosphere, about 1 mm or less thick. 118 SEA-SURFACE TEMPERATURE

(b) "Surface-layer" temperature, which is generally accepted as being the temperature of the surface layer 10-20 cm thick, and is the water which is normally sampled by buckets and similar "surface" sampling instruments.

(c) "Injection" or "Intake" temperature, which is the temperature of the water at various depths below the surface, which is taken into a ship through a piping, for the purpose of engine cooling, con denser circulation, sanitation, etc. Nine methods of measuring water temperatures are as follows: (1) Bucket method; (2) Distant-reading thermometer; (3) Engine-room injection temperature; ,(4) Trailing thermistor; (5) Hull contact method; (6) Drogue; (7) Radiation thermometer; (8) Weather satellite with infra-red radiation measuring ability. } ( 9) Weather buoy, relayed by radio either direct, through a ship, or through a communications satellite, etc.

ACCURACY The degr~e of accuracy is important; and in order to be able to produce the type of ohart required, data to O.loC accuracy are in fact essential. RADIATION TEMPERATURE Before studying the more commonly used, and the cheaper methods which concern our Commission primarily, let us first consider the radia tion (infra-red) thermometer (see Figure 1). This instrument, which measures the "skin" temperature, consists of a sensor which can be mounted in the bow of a ship, on a beam extending over the water from the wing of the bridge, or even in an airoraft or weather satellite. The sensor of this instrument is sensitive to variations of wavelength of the out going radiation, which is directly related to the temperature of the surface water. It is not the object of this paper to deal with this INSTRUMENTS, METHODS AND CCMPARISONS 119 method in any detail, because, generally speaking, the equipment is ex~ pensive and fairly complicated.

Nevertheless, valuable data are collected on a monthly basis by aircraft in various parts of the world using this type of instrumentation, the ground control of which is dependent upon accurate spot-temperatures taken by bucket (or some other method) by ships at sea.

The ability of earth-orbiting satellites to measure water temper atures and relay the data to receiving centres is also of interest, for this method offers great possibilities of expansion in the future.

Having now considered the surface film or "skin", let us have a look at the other extreme - Le. the "injection" or "intake" temperature which is one of the easiest temperatures to measure from a practical point of view (see Figure 2).

ENGINE-ROOM INJECTION TEMPERATURES All ships have inlets below the surface, where water is taken into the ship for various purposes. This intake water is used frequently for engine cooling; in order to control the machinery, the engineer records the temperature of this water by means of thermometers which are in close oontact with the water. He is not, however, a meteorologist, and the thermometers he uses are frequently installed in rather "unscientific" positions in the ship. Sometimes, they are situated several metres from the intake valve or flange, because the engineer is concerned only with the temperature of the water as it actually enters the machine that he is controlling.

Again, the type of thermometer installed by the ship-bUilder can be of many different types, and the scale divisions vary from single degrees to as much as five-degree graduations. Accurate measurements to O.loC are therefore quit~ unrealistic. Additionally, the officer on the bridge can never be sure that the reading he is given on the teleph6ne from the engine-room is in fact a "synoptic" observation. Evidence has shown that sometimes the reading given is a record from the previous watch! 120 SEA-SURFACE TEJVlPERATURE

Some of these thermometers are so badly installed that they can only be approached with difficulty, and often have to be read from an angle which gives rise to considerable parallax errors.

Nevertheless, when no alternative instrument is available, it is useful that these ships should ccntinue to supply engine-room temper atures. Many countries at present use this method as routine practice. It may be advisable, however, to be able to recognize suoh temperatures by an indication in the code form.

DISTANT-READING THERMOMETERS These thermometers can be divided into two types: (a) Mercury or other liquid in tube; (b) Electrical, ·with temperature-sensitive resistance sensors.

Distant-reading thermometers can be installed either as engine room equipment, with a dial in the engine-room itself; or alternatively with the dial cn the bridge. The former is more economical for the ship owner, whereas the latter is infinitely more satisfactory from a meteoro logical point of view. This problem is usually resolved by the ship cwner, who is the cne that pays the cost of installation.

The position cf the sensor in a distant-reading installation is a vital factor. Various systems can be classified according as to whether the sensor is situated in the engine-room intake pipe, ina special sea tank or in direct contact with the ship's hull.

The sensor is usually either a platinum resistor or a thermistor, which is placed in close contact with the water. The lag time of the former is usually greater than that of the latter, but generally accept able.

An interesting distant-reading thermometer is the hull-contact type, which was recently tested by the British Meteorological Offioe. Known as the "Limpet", the sensor unit is secured to the inside of the hull of the ship close to the waterline. A dial in the chart-room regis ters the temperatures of the hull. This type of instrument, however, is not in direct contact with the surface· water. INSTRUMENTS, METHODS AND CavIPARISONS 121

The sea-chest or tank in the bottom of a ship is a specially designed cavity into which intake pipes from the main and auxiliary engines terminate. This is a favourite position for a distant-reading thermometer probe •

. DROGUES Little information is available concerning drogues. The United Kingdom reported that they have recently carried out experiments by using a drogue, but the trials had to be abandoned because the drogue was lost at sea. Many 'oceanographers, however, will not accept the measuring of the temperature of the water through which the ship has passed as a sub stitute for the true surface temperature.

An interesting drogue-type of bucket designed in Poland works on the paravane principle. Controlled by 2 ropes, 50 metres long, the bucket measures the temperature of the water several metres away from the ship. There are, however, difficulties in operating a device of this nature.

BUCKETS One of the most common means of obtaining the temperature of the water is by means of a sampling bucket, and here there are many different designs.

Originally, a canvas bucket was lowered over the side of the ship, a sample of water hauled on board, and a thermometer used to register the temperature.

It was recommended by CMM-V, however, that existing designs of buckets should be modified because of the many errors which can arise, although these errors are not as great as those obtained by using engine room methods already described.

A NEW BUCKET In view of Recommendation 1 (CMM-IV), which called for Members to develop a simple, cheap and reliable instrument for measuring sea-surface temperatures, a considerable amount of time was spent by the South African Weather Bureau in developing a bucket which would not have the disad vantages of existing instruments; the result was the bucket which was 122 SEA-SURFACE TEMPERATURE

known by the Working Group on Technical Problems as the Crawford Bucket (see Figures 3 and 4). This bucket was quite a departure from standard models, especially in.the method of suspension. A precision thermometer is installed in the centre, and a pipe near the bottom provides through circulation of the sampling water. A flap in the side of the bucket enables the thermometer to be read "in situ" and the angle of suspension 0 (45 ) ensures that the greater part of the thermometer is always covered by water.

An additional feature is the non-return valve in the top, to pre vent evaporation and insolation errors. The weight of the instrument is adequate to provide stability when sampling at high speeds. A tail which trails in the water before it is lowered to its full extent orients the 0 bucket fore and aft. The bridle and stay hold it at an angle of 45 , and this ensures that the bucket samples only the surface layer, for this is the layer which influences the atmosphere. Diagrams showing the design of this special bucket are available on request.

Another interesting new bucket is one recently desigued by the Danish Meteorological Institute (see Figure 5). During sampling, water circulates round the thermometer bulb, which is protected by a large rubber fender.

HELICOPTERS Another means of measuring sea-surface temperatures is by lowering a bucket from a helicopter hovering over the water. This is a practice common near naval air stations. It is not known, however, to what extent the down-draught of air caused by the rotating blades of the helicopter affects the true surface temperature of the water.

TRAILING THERMISTOR Successful experiments have recently been carried out in South Atlantic waters with a new technique for measuring sea-surface temper atures, know!?' as the "trailing thermistor". Basically, this consists of a thermistor in the end of a rope which is trailed from the bridge of the ship. An electronic thermometer or recording device is situated within easy reach of the officer-on-watch. INSTRUMENTS, METHODS AND COMPARISONS 123

The name "trailing thermistor" can sometimes give a false impres sion of the position of the water sampled - one imagines a temperature sensor trailed astern of a ship. Such a practice would measure the tem perature of a mixed mass of water churned up by the ships' propellers. These temperatures would by no means always give a true representation / of the surface temperature. /

A correctly sited "trailing thermistor" samples the water in exactly the same position as a bucket, but with a great deal less trouble, though equal accuracy.

Although this type of instrument has only recently been in opera tion, over 200 comparative observations have already been made at ship speeds up to 25 knots, the vast majority of the readings being practi cally identical with those obtained by precision bucket close by. Further details will be available in published form shortly.

COMPARISONS The chairman of the CMM Panel formed to compare sea-surface measur ing instruments and methods has informed the Commission that no actual comparison tests have been carried out by the panel among the different instruments and methods, though some countries have made their own experi ments.

One bucket, however, has been used by several different countries for comparison tests, and the outcome of these tests is awaited with interest.

Methods and instruments can be compared in other ways; for instance, cost. We notice that the cost to the meteorologist by measuring temper atures through the engine-room is zero, whereas to install a radiation thermometer may cost several thousand dollars.

In bad weather, buckets are not easy to handle, whereas. distant reading thermometers are reasonably reliable because the surface water layers are well mixed by the wind and waves under such conditions. 124 SEA-SURFACE TEMPERATURE

In the case cf a vessel which is stationary (for example, an ocean weather ship), the bucket is more reliable than engine-room methods, when the flow of intake water drops considerably. The water in the "tank" may even become stagnant.

Reading from the engine-room as well as distant-reading methods are usually easy, whereas the use of a bucket requires some effort, a fact which sometimes places this more accurate method at a disadvantage.

We may also wish to consider which instrument is most suitable for a certain type of ship, and the choice may well be related to the speed of the vessel through the water. Some form of trailing thermistor may be the only way to obtain temperatures from hovercraft and other fast-moving vessels; but the more scientific bucket with precision thermometer is vital for ships engaged in research work. The frequent control of distant reading recording thermometers by means of accurate mercury thermometers seems to be essential. Model PRT·5 Optical Head \ ~ 5'l Electronic Control Unit ~.~C:-" ;~;.;),." J~~:'~" ~;"'~$"'~with Carrying Case .~~i / __. _". ,';r ..", I ~ ~ ~ Q ~ ~ td i

Figure 1 - Infra-red radiation thermometer for the measurement of "skin" sea-s.urface temperatures (U.S.A.)

f-' ~ \J1 126 SEA-SURFACE TEMPERATURE

l ;-- --.;.--=:' Spiral spring >- - I - '~r ~_J '"./-; L~,~I I Top cap I , Thermometer

I r: I l' I",'i, I • I, I : i l " !,' I 'I :! i I

:1 " , 400 ii, I'i __ Upper part of the case I, 'I !, .' i.i . ' '1 ,-, I :1

U!I" ,I

Nut with flange Packing Wart Intake tube

Lower part of the case Level of mercury

Spiral spring

Figure 2 _ Specially designed "intake" thermometer (Norwegian Meteorological Institute) INSTRUMENTS, METHODS AND CCMPARISONS 127

Figure 3 - South African bucket. The thermometer is read by opening a flap in the side of the "Crawford" bucket

Figure 4 - Part-section through the "Crawfordll bucket showing thermometer mounting and circulation pipe f-' [lb

:.:">"."" I '"i:';I I ::::1 I '" '"~ __ .. :·1 1 . 0 '"i:';I ':-:1 .1 ~ liJ ::0 '"~ Figure 5 - An instrument for the measurement of sea-surface temperature developed at the Danish Meteorological Institute INSTRUMENTS, METHODS AND CCMPARISONS 129

Figure 6 - British remote-reading'electronic thermometer with 6-sensor connexions

Figure 7 - Trailing thermistor

OPABHMTE]bHHE M3MEPEHMH TEMITEPATYPH ITOBEPXHOOTM BO~H (TITB) B OOOP ~OK~a~ rrpoweccopa r.M. Tay6epa Ha rrHTOH ceCCHH KMM BMO

B HaCTOHl~ee BpeMfI Ha MffpOBOH ceTll CY,D,OBbIX [,H,D,p:)lileTeopo~orH qeCKHX CTaH~HH rrp~MeHHIDTCH caM~e pa3Hoo6p33H~e MeTO,D,~ II rrpH60p~ ~~H llSMepeHffH TITB - OT pa,D,lla~HOHH~X TepMOMeTpOB IT 3~eKTpllqeCKllX ~HCTan~HOHH~K IIpn6opOB ,n;o CTapllHHoro M2TO)I,a "Be,n;pa" II MeTO)I,a II condencer intake " Oc06eHHO 50m:'lIIoe pacrrpOCTpaH(~ffe llMeeT TIO- c~e,n;HllH crroc05, SaKRIDqaIDillHHCH B orrpe,n;eReHffll TITB Q:) ll3MepeHffHM B KHHrcTose oXRa,n;llTeRhHoro Tpy6arrpoBo,n;Horo M3lIIllHHoro OT,D,e~eHllH CY,lI,Ha.

ITo YKasaHH~M rrpnqAHaM cy,n;OB~e ,n;aHH~e TITB xapaKTepllSYIDTCfI paS~HqH~Mn IIorpewHocTHlilll ff HeO,D,HOp0,I\H~ IIO CBQeMY KaqeCTBY. 3TO CH~bHO saTpY,D,HHeT CHHOITTllqeCKllH aHa~llS pesY~hTaToB Ha6~~,n;eHllH TITB II HX llcrrO~bSOBaHlie ,n;~H rrpaKTHqeCKllX qe~eH (p~6H~H IIpoM~ce~ II rrporHOS IIOrO,D,H) II B HayqH~X Hcc~e,n;oBaHliHx (BsallMO,I\eHCTBMe OKeana H aTMocwepH, 06illaH ~HPKY~H~HH aTMocwepH, TeII,ITOBOH 6a~aHC OKeaHa H MOpCKOH K~llMaT).

YY:HThrBaH llSRO)l(eHHcie, qeTBepTaH ceCCllH KMM B pesomO~llll 3 II peKOMeH,D,a~Hll 1 IIpH3Ba~a CBOdX q~eHOB Y,I\e~HTh cephe3Hoe BHHMaHHe np06~eMe llsMepeBHH TITB C ~e~hro BHHB~2HHH HaH60~ee npoCToro, Ha ,D,e~HOrO IT ,I\emeBoro MeTO,D,a, ,I\aID~ero o,I\llHaKoB~e pesY~hTaTcl Ha cY,D,ax pasHoro THIIa II TOBHa~a.

Terrepb Ha IIOBeCTKe ,D,HH rrO~B~HeTCH HOBclH BOUpOC 06 Y~OqHeHllff nOHH'nU[ TOB B CBHSH C TeM, qTO HapHlI,Y co cTapHMll MeTO,I\aMH HaqaJUl rrpHMeHHThcH HOB@e (palI,Ha~ITOHHHH TepMOMeTp), 6RarO,I\apH qeMY HsMe penTIe TITB y~e rrpollsBo,D,llTCH Ha pasHHx ropH30HTax nOBepxHocTll C~OH MOpH - OT Bel'XHeH IT~eHKH TO~~llHOH B HeCKO~h!W MliKPOH ,Il,0 r~y6ilHbI 2-3 M. B~ecTe C TeM ASBecTHo, qTO TIpH OIIpe,D,e~eHHclx YC~OBHHX cTpaTHcpHKa::J;HH Ha P3SHclX rJIy6llHax sToro C~OH TeM.rrepaTypa MO)l(eT CY ~eCTBeHHO paS~HqaTbCH, qTO TaK~e HB~HeTCH rrpHqllHOH HeO,I\HopO,I\HO CTH rro~yqaeMOrO MaTepHa~a.

B COOTBeTCTBHH C peso~ID::J;lleH 3 II peKOMeH,I\a~lleH 1 KMM-IY B OOBeTcKoM OOIDse B 1967 H 1968 rr. npOBe,I\eHcl cpaBHllTeJIbHcle llcc~e ,D,OBaHHH paS~ilqHHK MeTo,I\OB HSMepeHHH TilB. 3TH llSMepeHAH ocy~e~T BJIenH He llCC~e,D,OBaTeJIhCKllX CY,I\3X "AKa,I\eMffK illapmoB" (BO,I\OHSMe~e Hue 6800 TOHH), "A.IT.BosaKGB" H "ID.M.llioKanhcKllH" (3600 TOHH) B paSJIHQHhrX K.1J"lMaTliqeCKHX ~,OHax TllXO,O 11 HH,I\llHCIW,O oReaHoB, a TaKKe Ha MaJIHK cYlI,ax THIIa CPT (400 TOHH) B qepHoM IT ASOBCKOM MO pHX. Ha6~ro,I\eHHH rrpOSe,I\eHcl IIpll pa3HhrX rrr,I\pOMeTeopOROrllqeCKllX YC ~OBHHX B rreBllO~ C anpene 1967 P. no weBpaJIb 1968 r. B TIpe,Il,e~ax m~poT OT 40 c.m.,I\o 60 ro.ill. OHff BHuonHHnrrCb Ha XOlI,Y cY,I\Ha II B lI,peHwe, rrprr CIIOKOHHOM Mope, IIPll yMepeHHOM II CH~bHOM BonHeHall. 132 TEMTIEPATYPA ITOBEPXHOCTM BO~H

o 3a STOT rrepHO~ MHHHMallbHaE TeMrrepaTypa BO~H oHlla -0,5 C, MaK- CHMallbHaE 30,50C; COOTBeTCTBeHHO TeMrrepaTypH Bos~yxa OHllH B rrpe~eJIax oT-O,40C ~o 29,50C. CpaBHHTeJIbHHe HSMepeHHE rrpOBO~HJIllCb rro e~HHoH rrporpaMMe, paspaooTaHHoH B focy~apcTBeHHoM oKeaHorpaqmCIecKoM HHcTHTyTe, H ~HKCHpOBaJIHCb B TaollH~ax e~HHoH ~OPMH sarrHCH. ilporpaMMa H ~opMa sarrHCH BeCbMa CXO~HH C TeM,CITo MH rrbs~Hee rrOJIyCIHJIH HS ElOPO rro HSYCIeHHlO rrpHoopoB H MeTO~OB llsMepeHHH TilB, cos~aHHoro B cocTaBe Pf KMM rro TeXHHCIeCKHM rrpoolleMaM.

~JIH corrOCTaBJIeHHH HCrrOJIbSOBallHCb CHHxpoHHHe OTCCIeTH rro CJIe~YlOm;HM rrpnoopaM : PTyTHHH TepMoMeTp B cTaH~apTHoH orrpaBe (THrra orrpaBH llirrHH~ JIepa) sa OOpTOM H B Be~pe; 3JIeKTpOTepMOMeTp corrpoTilBlleHilH CY~OBOll ~HcTaH~iloHHOH cTaH ~HH (C~C), BMOHTHpoBaHHHH H CrrelIHaJIbHYlO HHIDy B Koprryce cy~ Ha Ha rJIyoHHe 1,8 M; 3JIeKTpOTepMOMeTp corrpOTHBJIeHHH H TepMilcTop,crrycKaeMHe sa OOpT, C perilcTpa~HeH Ha rroTeH~iloMeTp;

PTyTHHH TepMoMeTp B MamHHHOM OT~eJIeHilil.

KpOMe Toro, Ha ilCCJIe~OBaTeJIbCKOM cY~He "AKa~eMHK lliHpmoB" HcrrHTHBaJICH rrpoToTilrr KOBma ABC,HsroToBJIeHHHH rro CIepTe~aM, rrpe~ CTaBJIeHHHM HaM r-HOM KpOY~OP~OM. MSMepeHHH rrpoilsBo~HllHCb C rrpaBoro il lleBoro OOpTOB, Ha oaKe, Ha KopMe H B cpe~HeH CIaCTH cY~Ha. O~HOBpeMeHHo OTMeCIaJIHCb TeMrrepa Typa Bos~yxa, COCTOHHile OOJIaCIHOrO rroKpoBa il oaJIJI BOJIHeHilH.

Bcero Ha yKasaHHHx cy~ax BHrrOJIHeHO 420 rrOJIHHX cepHH Ha OJIlO~eHHH. OopaooTKa H aHaJIilS HaOJIlO~eHHil

~~H BHHBlleHHH rrorpemHocTeil, BOSHHKaIOmHX rrpH ilcrrOllbSOBaHHH yKasaHHHx BHme crrocoooB HSMepeHHH TilB, rrpoBe~eHa CTaTliCTil~eCKaH oopaooTKa Bcex cepHil cpaBHHTeJIbHHX HaOJIlO~eHilll. OHa SaKJIIO~aJIaCb B orrpe~eJIeHHH paSHHIIH (6 T) Me~~y rJIYOOKOBO~HHMH orrpOKli~HBaIO mHMHCR TepMOMeTpaMH (TlIrra HaHceH-ITeTepCOH), rrpHHHTHMli sa STaJIOH, H pesyJIbTaTaMH liSMepeHliH rro Ka~~OMY liS rrepe~HCJIeHHHX rrpliOopOB. TaKHM 06paSOM OHJI BHHBJIeH rrpHOOp,RBJIRIOmilllCR HaliOOJIee TO~HHM rrOCJIe rJIYOOKOBO~Horo TepMOMeTpa, KOTOpHH Mar OH OHTb ilCrrOJIbSOBaH B Ka ~eCTBe cyosTaJIOHa ~llR oopaOOTKli Bcex rrOJIy~eHHHx ~aHHHx. TaKaR Me pa OKaSaJIaCb HeOOXO~HMoil rrOCKOJIbKY rJIYOOKOBO~HHe TepMOMeTpH HC rrOJIbSOBaJIHCb TOJIbKO B ~peil~e.

KaK BH~HO liS TaOJI. 1 H 2 cyosTaJIOHOM Ha HMC "AKa~eMliK lliHp mOBil OKaSaJICE KOBID ABC, Ha ~pyrHx cy~ax - SJIeKTpOTepMoMeTpH M3MEPBHHH B CCCP 133 corrpOTHB~eHHH ClC. la~bHeMillaH o6pa6oTKa saKHroqaHaCb B cOrrOCTaB ~eHHH cy6sTaHoHOB C rrOKaSaHHHMH BCeX ~pyrHx HCrrHTHBaeMHX rrpH60 pOB. PeSY~bTaTH B BH~e rrOBTOpHBMOCTH paSHHqHHX SHaqeHHH rrpe~ CTaB~eHbI B Ta6H. 1 H 2. KpoMe Toro, rrpoBe~eH aHaHHS ~aHHHx Ha6 ~ro~eHHH C ~eHbro orrpe~eHeHHH CTerreHH BHHHHHH paSHHqHHX~aKTopoB (MopcKoro BOHHeHHH, RPHHqHH pasHocTH TeMrrepaTypH BO~H-BOS~YX, HH COHH~HH), a TaKme BH60pa MeCTa Ha6Hro~eHHH Ha TOqHOCTb rroKasaHHH Tex H~H HHHX rrpH60poB.

PeSY~bTaTH cpaBHeHHH

AHa~Hs MaTepHa~oB cpaBHHTeHbHHx HccHe~oBaHHH rrOSBoHHeT c~e~aTb CHe~yromHe BHBO~H : 1. HBH60Hee perrpeseHTaTHBHHM MeCTOM HSMepeHHH TITB HBHHeTCH 6aK cY~Ha rro 60pTy, r~e oTcyTcTByeT SHaqHTeHbHBM CTOK cy~o BbIX BO~. YCTaHoBHeHO,qTO HSMepeHHH y 60pTa ,HMeromero CHHBHlle OT BepcTHH, rrpHBo~HT K saBHilleHHro ~eHcTBHTeHbHoH TeMrrepaTypll BO~ll B cpe~HeM Ha 0,20 C, a HSMepeHHH Ha KopMe cY~Ha - Ha 0,3 - 0,40 C.

2. HaH60Hee TOqHllM H y~06HllM B sKcrrHyaTa~HH HBHHeTCH ~HCTaH ~HOHHllM sHeKTpoTepMoMeTp, yKperrHeHHHM B Koprryce HOSOBOH HHH cpe~HffiqaCTH cY~Ha. lHH rroHyqCHHH ~aHHllX, TOQHee XapaKTepHSyromHx TeMrrepaTypy rroBepxHocTHoro CHOH, SHeKTpo TepMOMeTp ~eHeco06paSHO yCTaHBBHHBBTb He HHme O~Horo MeTpa OT BaTepHHHHH. illKaHa ~HH CHHTNH rrOKasaHHM~OH~Ha 6llTb pas MemeHa B illTypMBHCKOM py6Ke, qTO CYilleCTBeHHO 06HerqHT pa60 Ty BaXTeHHoro illTypMaHa rro HSMepeHHro TITB H,BMeCTe C TeM, 06eCrreqHT rroHyqeHHe Ha~e~HHX ~aHHHX. .

YqHTllBaH HSHo~eHHoe,npe~CTaBHHeTCH~eHeco06paSHllM rrpHHHTb peKOMeH~a~Hro KMM, a~peCOBaHHYro MOpCKHM CTpaHaM, 0 TOM, qT06u rrpH rrpoeKTHpOBaHHH HOBllX CY~OB H KarrHTaHbHOM peMOHTe cTapllx 06HsaTeHb HO rrpe~YCMaTpHBaHrrCb yCTaHoBKa yKasaHHoro rrpH60pa B KaqeCTBe ~eTaHH illTypMaHCKoro 060pY~OBaHHH Kopa6HH.

3. ITPH OTCyTCTBHH ~HCTaH~HOHHllX TepMoMeTpoB HSMepeHHe TITB Ha cy~ax ~OH~HO rrpoHsBo~HTbCH rryTeM BSHTHH rrp06 BO~H sa 60p TOM. lJIH STOM ~eJIli HaH60Hee TOqHbIM 1I y~06HllM liS Bcex lIcrru TaHHllX HaMli rrpH60poB OKaSaJICH rrpH60p THrra KOBilla AEC.

ITo pesyHbTaTaM lIcrrbITaHHH Ha IlliC "AKa~eMHK illlIpillOB" STOT rrpH 60p HMeeT cHe~yroilllIe KBqeCTBa :

IToKasaHHH rrpH60pa BeCbMa 6HlISKli K ~aHHHM rJIy60KOBo~HllX OrrPOKlI~HBaromHXCH TepMoMeTpoB. CorrocTaBHeHlIe rro HailllIM ~aH HllM rroKasllBaeT, qTO paSJIlIqHe B rroKasaHHHX 0601IX rrp1I60poB B 99% cHyqaeB B cpe~HeM COCTaBHHeT O,lo C. O~HaKo TaKHX ~aHHHX eille He~OCTaTOqHO ~HH Toro, qT06ll OKOHqaTeHbHO yCTaHo BHTb TOQHOCTb rrpH60pa; TEMITEPATYPA ITOBEPXHOCTH BO~H

TaoJIu:u;a 1 IToBTop1I8MoeTh (%) pasHH~ ~T) M8lK,n;y rrOKasaHHlIMH GJI8KTPOT8PMoM8Tpa C~C H ,n;pyrHx rrpHoopoB

T T - T ..!J' T - T(5 Mm 6,T Te,n;c-Tr Te,n;o-T 3T (O )T e,n;e 31(K) To,n;e- B o,n;e e,n;o

0,0 28 37 - 30 34 - 0,1 51 43 7 46 44 - 0,2 19 17 40 18 16 - 0,3 2 3 30 5 4 - 0,5 - - 13 1 1 2 , 0,6 - - 10 - 1 - 0,7 ------1,0-1,2 - - - - - 10 1,3-1,5 - - - - - 37 1,6-2,0 - - - - 37 -2,1 H ,n;aJIee Beero 100% 100% 100% 100% 100% 100%

To,n;e GJI8KTPOT8PMOM8TP C~C T Te T GT CO,) GJI8KTPOT8PMOM8TP Ha oaK8 T GT (K) GJIeKTpOT8pMOM8Tp Ha KOpM8 To T8pMOM8Tp B orrpaBe sa OOpTOM TMill T8pMOM8Tp B MaillHHHOM OT,n;8JI8HHH. ll3MEPEHMH B OOOP 135

Ta6JUII\a 2

IToBTopHeMOCTb (%) paSHllI\ (~T) Me~~y rrOKasaHllHMll KOBilla AEO II ~PYI'llX rrpll6opoB

T T T T - T T K-T TI' K-T c,n;c K -T B Ti" T6 (M) TK-T6 (rr) K Mill

0,1 99 65 55 58 10 - 0,2 1 22 18 8 7. - 0,3 - 8 9 25 7 2 0,4 - 5 9 6 6 - 0,5 - -- 3 - 8 0,6 - - - - 12 13 0,7 - - 9 - 6 7 0,8, - - - - - 11 0,9 - - - - 10 11 1,0..1,2 - - - - 13 16 1-3-1,5 - - - - 18 9 1,6-2,0 - - - - 8 ' 14 2,1 II ,n;a- JIee - liToI'o 100% 100% 100% 100% 100% 100%

T KOBill ABO K TTI' TepMoMeTp I'JIy6oKoBo~HHH orrpoKll~HBaID~llHcH Tc~c . TepMoMeTp corrpOTllBJIeHllH O~O T TepMoMeTp B orrpaBe (B Be~pe) B T (M) TepMOMeTp B orrpaBe (MeTaJIJIllq.cTaKaHqllK)sa 60pTOM 6 T (rr) 6 TepMoMeTp B orrpaBe (rrOJIll3TllJIeH.cTaKaHqllK)sa 60pTOM T OT~eJIeHllll. Mill TepMoMeTp B MaillllHHOM 136 TEMITEPATYPA ITOBEPXHOCTH BO~H

KOHCTpyKTllBHHe OCOOeHHOCTll nplloopa YMeHbillaIDT rrorpeillHoCTll, CBRsaHHHe C llcrrapeHlleM II B~llHHlleM BHeillHllX ~aKTopOB (Harpe BaHlle II ox~a~AeHlle) Ha TeMrrepaTypy llSO~llpOBaHHOR BOAH.

HcrrHTaHlle Ha ox~a~AeHlle, BHrrO~HeHHoe Ha RHC "AKaAeMllK IIIllp illOBII, rrOKasa~o, ~TO rrpll paSHll~e Me~AY TeMrrepaTypoR BOAH II BosAyxa B 3_40 TeMrrepaTypa BOAH B KOBrne Ha TpeTbeR MllHyTe rrOHllSll~aCb Ha 0,20 C, a Ha 20-R MllHyTe - Ha 1,20 C.

COBeprneHHO aHa~orll~HHR pesY~bTaT OH~ rro~y~eH r-HOM ~~. KeMrroe~~oM rrpll llcrrHTaHllll Ha HarpeBaHlle B YC~OBllRX oo~ee sHa~llTe~bHoR pasHll~H TeMrrepaTypH BOAH II BosAyxa, AocTllrrneR 100.

KOHCTPYK~llR rrplloopa ooecrre~llBaeT CBoooAHyID ~llPKY~R~llID BOAH B HeM BO BpeMR saoopa rrpoOH. Cy~ecTBeHHoe SHa~eHlle llMeeT II TO,~TO TepMOMeTp II ero orrpaBa BO BpeMR BSRTllR rrpOOH Haxo ARTCR B TOR cpeAe, TeMrrepaTypa KOTOpOR llsMepReTcR. B CBRSll C 3TllM c~eAyeT oTMeTllTb,~TO rrpll llcrrO~bSOBaHllll MeToAa "BeA pall norpeillHocTllllsMepeHHR ~aCTO BOSHllKaIDT rro TOR rrpll~llHe, ~TO onpaBa TepMoMeTpa Mo~eT OHTb rrpeABapllTe~bHo HarpeTa B nOMe~eHllll ll~ll Ha cO~H~e, ll~ll ~e,HaooopoT, ox~a~AeHa rrpll HllSKllX TeMrrepaTypax BosAyxa.

Kperr~eHlle rrplloopa K ~llHID crrocoooM "opllAe~b" II Ha~ll~lle AByX MeTpoBoro XBoCTa ooecrre~llBaeT ropllsoHTa~bHoe ero rro~o~eHlle . II CKO~b~eHlle rro rroBepxHocTll BOAH Des cy~ecTBeHHoro yr~yo~e~ HHH.

ITPllOOP Mo~eT OHTb llcrrO~bSOBaH rrpll OO~billllX CKOpOCTRX CYAHa, ~TO o~eHb Ba~HO, rrocKo~bKY APyrlle MeTOAH BSHTllH rrpoo BOAH rrpaKTll~ecKll orpaHll~eHH CKOpOCTbID He oo~ee 8-10 YS~OB.

ITPllOOP rrpocT B llsrOTOB~eHllll II YAooeH A~R llcrrO~bSOBaHllH rrpll CYAOBHX rllAPoMeTeopo~orll~ecKllx HaO:JIIDAeHllHx.

4. B ApeR~e II rrpll Ma~Hx CKOpOCTRX cYAHa (AO 6-8 YS~OB) A~H llsMepeHllR TITB MoryT OHTb llcrrO~bSOBaH MeToA "BeApa" II HenocpeAcTBeHHoro llsMepeHllR sa OOpTOM c rrOMO~bID TepMoMeTpa B onpaBe. CpaBHeHlle c KOBillOM AEC II C~C nOKasHBaeT, ~TO norpemHOCTll 3TllX cnocoooB B 70-80% c~y~aeB He npeBHillaIDT 0,20 c. 0AHaKo rrpll oO~brnllx paSHOCTRX TeMrrepaTypH BOAH II BosAyxa, a TaK~e rrpll Cll~bHOM BO~HeHllll II BeTpe (BHsHBaID~llx YCll~eHHoe llCTIRpe Hlle) llSMepeHllR TepMoMeTpoM B orrpaBe rrpllBoAllT K oO~brnllM ornllo KaM II rrosToMy ero rrpllMeHeHlle B AaHHHX c~y~aRx HaAO rrpllsHaTb HeAorryCTllMHM. OcooeHHo 3TO OTHOCllTCH K TepMoMeTpaM B onpaBe C ~erKllM nO~ll3Tll~eHOBHM CTaKaH~llKOM. M3MEPEHMH B OOOP 137

5. OpaBHHTe~bHHe Ha6~ro~eHHH rrOKasHBaroT,~To ~aHHHe 0 TilB, rro ~y~eHHHe HS MamHHHoro oT~e~BHHH CY~Ha oT~H~aroTcH 60~bmHMH TIOrpemHocTHMH H TI03TOMY He MoryT xapaKTepHSOBaTb waKTH~e CKyro TeMTIepaTypy TIOBepXHOCTH BO~H, Ha HMO "AKa,I\eMHK I1IHpmoB" OillH6KH 60~ee 0,50 0 HMeroT TIOBTOpHeMOCTb 98%, HaOCTa~bHHX cy ,I\ax 3TH HSMep8HHH B 83% c~y~aeB ~a~H saBHmeHHHe TIOKaSaHHH Ha 1,2 - 2,30 0. BaEHo saMeTHTb,~TO Be~H~HHa omH6KH HB HM8eT KaKoH-~H60 BHpaEeHHOn saKoHoMepHocTH H saBHcHT, r~aBHHM 06pasoM , OT peEHMa pa60TH ox~a~HTe~bHoH CHCTeMH B MamHH HOM OT,I\e~eHHH,

B ,I\peHwe CY,I\Ha TIpH oCTaHOBKe MamHHH TIpHTOK BO,I\H B ox~a~H Te~bHyro cHcTeMy oc~a6eBaeT H~H COBceM rrpeKpa~aeTcH, B CBHSH C ~eM Be~H~HHa omH60K peSKO BospaCTaeT H ~OCTHra8T 8_100 ,

Be~H~HHa rrorpemHocTeH HSMepeHHH B MamHHHOMOT,I\e~eHHH saBH CHT OT !CHrra H paSMepa cY,I\Ha" TaK KaK OT 3Toro saBHc'HT, r~y6HHa OT-' BepcTHH BO,I\OrrpHeMHHKa H ~~HHa Tpy6H BHyTpH CY~Ha MeB~y BrrycKHHM oTBepcTHeM H TepMoMeTpoM.HeMa~oBaBHoe SHa~eHHe HMeeT Ka~eCTBO rrpHMeHHeMoro TepMoMeTpa H MacmTa6 ero mKa~H, MeCTO yCTaHoBKH TepMOMeTpa B Tpy60rrpoBo~e H CTerreHb BOSMOBHoro B~HHHHH Ha ero rro KasaHHH ropH~HX H XO~O~HHX MarHcTpa~eH, pacrro~oBeHHHx rro coce,I\CT By. Bce HS~OBeHHoe CBH,I\eTe~bcTByeT 0 HeHa~eBHOCTH ,I\aHHHX HSMepe HHH TilB B MamHHHOM oT~e~eHHH H He~orrycTHMoCTH HcrrO~bSOBaHHH MeTo ,I\a " condenser intake " B rrpaKTHKe CY,I\OBHX rH,I\pOMeTeopo~orH~ecKHx Ha6~ro~eHHH. '

Orrpe8e~eHHe TepMHHa TilB ilpu paccMoTpeHHH Borrpoca 06 y~y~meHHH MeTO~OB HSMe'peHHH TeMrrepaTypH rroBepxHocTW BO,I\H BOSHHKaeT He06xo,I\HMOCTb 'YTo~HeHHH caMoro TepMHHa TilB. MHeHHe Eropo rro Hsy~eHHro rrpH60poB H MeTo~oB HSMepeHHH TilB Pf rro TeXHH~eCKHM rrp06~eMaM 0 He06xo,I\HMOCTH rrpHHH THH KMM peso~ro~HH rro CTaH,I\apTHoMy orrpe~e~eHHro TepMHHa TilB c~e ,I\yeT rrpHsHaTb rrpaBH~bHHM, TaK KaK 6asHpYRcb Ha pesY~bTaTax HSy ~eHHH TepMH~ecKoH CTpyKTypH rroBepxHocTHHX BO~, MH Terrepb MOBeM 60~ee KpHTH~ecKH OCMHC~HTb SHa~eHHe rro~y~aeMHx ~aHHHX HSMepeHHH B saBHCHMOCTH OT rrpHMeHReMHx MeTO~OB.

OOBpeMeHHHe TeopeTH~ecKHe H 3KcrrepHMeHTa~bHHe HCC~e,I\OBaHHH rrOKa3HBaroT, ~TO rrpo~eccH Terr~O-H B~aro06MeHa MOPCKOll rrosepxHocTH C aTMocwepOH BHsHBaroT 06pasoBaHHe TOHKoro ox~aB~eHHoro C~OH Ha caMOH rrOBepXHOCTH, CB060,I\HO rrporrYCKaro~ero rrOTOKH CO~He~HOH pa~Ha ~HH B HHBe~eBa~He C~OH, r~e H rrpOHCXO~HT HX rror~o~eHHe.

Ha~H~He XO~O,I\HOro rrOBepXHOCTHoro C~OH rrpORB~ReTCH B BOS HHKHOBeHHH OTpH~aTe~bHHX rpa,I\HeHTOB TeMrrepaTypH B 3TOM CHoe TO~~HHOH OKO~O 1-2 CM. il03TOMy TeMrrepaTypa rrOBepXHOCTH BO~R MOBeT c~eeTBeHHO OT~H~aTbCH OT TeMrrepaTypbI c~oeB, ~eEa~HX HHEe.

ilo. Ha6~ro~eHHHM Ha HMO "AHc6epr" B OeBepHOH AT~aHTHKe"o'rpH ~aTe~bHRe SHaKH rpa~HeHTa TeMrrepaTypH B yKasaHHOM c~oe HMe~H 138 TEMilEPATYPA ilOBEPXHOCTli BO~~

Me8TO B $:;l,%' GJIycrasBH TO.JIbKO B S%.CJIycraeB OHH QKaSaJIHCb rrOJIOlKH TeJIbHIllMH. Cpe,n:Hflfl BeJIHcrHHa rpa,n:HeHTOB B ,n:HeBHoe BpeMfl COCTaB 0 JIflJIa _0,43 C, a HOcrbID -0,340 C. B OT,n:eJIbHIllX CJIycraHx BeJIHcrHHa rpa,n:HeHTa ,n:OCTHraJIa -2,00C.

AHaJIOrHcrHIlle BeJIHcrHHll rpa,n:HeHTOB rrOJIycreHIll paccreTHIllM rrYTeM o ,n:JIfl paSHllX illHpOT THxoro OKeaHa. B cBe,n:HeM rpa,n:HeHT paBeH -0,5 C, HSMeHflflCb B rrpe,l\eJIax OT -0,1°,n:0 -O,S C.

SKcrrepHMeHTaJIbHO yCTaHOBJIeHO, crTO rroBepxHocTHIllR OXJIalK,n:eH HIllR CJIOR, HJIH, KaK ero rrpilHflTo HasIllBaTb, "XOJIo,n:Hafl rrJIeHKa", BeclJMa yCTOHcrHB B P8SJIH~HllX YCJIOBHflX H coxpaHfleTcfl ,n:alKe rrpil illeCTHOaJIb HOM BOJIHeHHH.

lis SToro KpaTKoro oosopa BH,n:HO, crTO rrpO~HO YCTaHOBHBillHRCH B rrpaKTHKe MOpCKHX HaOJIID,n:eHHll H HCCJIe,n:OBaHilR TepMHH TilB WaKTil crecKH MOlKeT OTHOCHTbCfl TOJIbKO K ocreHb TOHKOMy rrOBepXHocTHOMy. CJIOID BO,l\Ill, TeMrrepaTypa KOToporo B KaKOR-TO CTerreHH rrpHOJIillKeHHfl xapaKTepHsyeTcfl rrOKasaHHHMil' pa,l\Ha~HOHHoro TepM8MeTpa.

STOT T8pMHH He rrpaBOMepeH rro OTHomeHilID K pesyJIbTaTaM HS MepeHHHc~TaH,n:apTHIllMH Cy,n:OBllMH rrpHoopaMH, TaK KaK OHil il3MepHIDT TeMrrepaTypy He Ha rrOBepXHOCTH BO,n:Ill, a Ha KaKOR-TO ~JIYOHHe rroA rroBepxHocTHIllM CJIOeM. rJIyoilHa H3Me.peHHR saBilCHT OT CTerreHil rro·-" rpYlK8Hilfl rrpiloopa H rrpil cYll\eCTBYIDll\ilX B HaCTOflll\ee BpeMfl MeTo,n:ax KOJIeOJIeTCfl B 3HacrilTeJI~HIllX rrpe,n:eJIax - OT 0,5 M ,n:o 2-3 M. ECJIil ycrecTb rrepeMeillHBaHHe BO,n: rrpH. i,l\BHlKeHilil 'KOpaOJIfl, a TaKlKe HHep~llIo pTyTHGrO TepMoMeTpa, rrorpy~aeMoro B BO,n:y, TO OOJIee rrpaBHJIbHO rOBopilTb He 0 TeMrrepaType Ha KaKOR-TO orrpe,n:eJIeHHOR rJIYOHHe, a 0 KBaSHHHTerpaJIbHOR xapaKTepHCTHKe TeMrrepaTypll rro,n:rrOBepXHoCTHoro ElJIO,fI BO,n:bI.· ECTec'rBeHHO, eITO ,l\JIfl ,ll;OCTHlKeHHH HailOO.JIbmel1 O.!(HOpO,r;- HooTH-Ma'iepHaJIOBHaOJIID,n:eHHR'Ha,n:o CTpeMHTbCfl K orpaHHcreHHID TOJI ll\HHIll STOrO CJIOfl, rrpHHHB ee. B' HaCTOflll\ee BpeMH sa 1 M. TaKafl TO.JIll\HHa CJIOfl rrpe,n:JIaraeTCfl, HCXO,n:H HsrrpaKTHcrecKHx BOSMolKHocTeR HCrrOJIbSOBaHHfl cYll\eCTByroll\HX rrpHoopOB, a TaKlKe Ha OCHOBaHHH ~a~ THcreoKHx )(aHHIllX 00 HSMeHeHHflX B HeM TeMrrepaTypIll BO,n:bI, JIelKall\HX B rrpe,n:eJIax TpeoyeMoR TOcrHOCTH Hao~ro,n:eHHR.

TaKHM oopaSOM, BOSHHKaeT Heooxo,n:HMOCTb BO BBe,l\eHHH HOBO ro TepMHHa TilCB (TeMrrepaTypa B rrOBepXHOCTHOM CJIoe BO,n:ll) rrpHMeHH TeJIbHO K pesyJIbTaTaM HSMepeHHR TeMrrepaTypH BO,n:Ill C rrOMOll\bID CTaH ,n:apTHIllX CY,l\OBIllX rrpHoopOB H MeTo,n:OB.

PesIDMHpyfl CKasaHHoe, MOlKHO rrpe,n:JIO:KHTb CJIG,ll;YIDll\ile TepMHHO' JIOrHcreCKHe.orrpe,n:eJIeHHH:

TilB TeMrrepaTypa rrOBepXHOCTH BO,n:Ill, xapaKTepHsyIDll\afl TOHKHR rro· BepXHOCTHIllR CJIOR BO,n:ll (OT HeCKOJIbKHX MHKpOH ,n:o 1-2 CM), H3MepfleMaH Cy,n:OBIllM HJI!'!' CaMOJIeTHIllM pa,n:Ha~HOHHIllM TepMOMeT pOM. H3MEPEHHH B CCCP 139

TITCB ~ TeMuepaTypa B UOBepXHOCTHOM c~oe BO~H TO~~HHO~ He 6o~ee 1 M, H3MepKeMaK UpH UOMO~H cTaH~apTHHx CY~OBHX upH6opOB H MeTO~OB.

ITpHMeHeHHe STHX TepMHHoB Mo~eT 6HTD oupaB~aHo TeM, ~TO OHH ~aIDT KCHoe upe~cTaB~eHHe 0 ~H3H~ecKoM 3Ha~eHHH uo~yqaeMHx ~aHHHx 0 TeMuepaType BO~H.

B HacToKlI\eM ~OK~a~e He o6cy~~aeTCJI Borrpoc 0 B03MO~HO~ H Heo6xo~HMOH TO~HOCTH H3MepeHHK TeMuepaTypH Ha UOBepXHOCTH MOpK ~~K Hcc~e~OBaTe~DCKHX H upaKTH~eCKHX ~e~eH. AUpHOpH MO~HO upe~ UO~O~HTD Ha~H~He cy~ecTBeHHoro B~HKHHK upOCTpaHCTBeHHO·-BpeMeH.,. HO~ H3MeH~HBOCTH TeMuepaTypH, UOCKO~DRY H3MepeHHK B OCHOBROM UPOBO~KTOK He B TO~Ke H He MrHOBeHHO, a Ha upOTK~eHHH OTpe3Ka UyTH, 3aBHCK~ero OT CKOpOCTH CY~Ha H HHep~HH upH6opa.

B CBK3H C STHM ~e~ecoo6pa3HO peROMeR~OBaTD upoBe~eHHe Ha MOpKX H OReaRax Hoc~e~OBaRHH B~HKHHK SToro ~aKTopa Ha TO~HOCTD H3MepeHHK TeMuepaTypH BO~H (TilB H TilCB). 140

~nTEPATYPA

1.: AllseHmTaT E.A. Pa~Ha~HOHH@ll MeTO~ HSMepeHHa TeMITepaTYP@ ~eaTe~bHoll ITOBepXHocTH. Tp. rro, B@IT. 107, 1961.

2. raeBcKHll B.~. nCc~e~OBaHHe ~~HHHOBo~HOBoro Hs~y~eHHa aTMoccpep@. Tp. rro, B@IT. 100, 19600

3. Eoryc~aBcKHI c.r. llor~o~eHHe cO~He~Hol pa~Ha~HH B Mope H ero HeITOCpe~CTBeHHoe B~HaHHe Ha HSMepe HHa T8MITepaTyp@ Mopa. Tp. MopCKoro rH~pocpHSH~eCKoro HHcTHTyTa AH CCCP, B@IT. 8, 1956.

4. EcpHMoBa H.A. PaCITpe~e~eHHe 3cpcpeKTHBHoro HS~Y1J:eHHa Ha CTpoKHHa ~.A. ITOBepXHOCTH seMHoro mapa. Tp. rro, [email protected], 1963. 5. llHBoBapoBa A.A. ° B~HaHHH rrpoHHKaIDmell B Mope CO~He1J:HOll pa~Ha~HH Ha CPOpMHpOBaHHe TeMITepaTYP@ EO,1I;hI •

JKypHa~ "OKeaHo~orHa" No 2, 1963.

6. Ma~eBcKHll 06 BsMepeHHH TeMITepaTYP~ BO~HOI ITOBepX Ma~eBH1J: C.ll. HOCTH. Tp. rro, B@IT. 95, 1963.

7. THMocpeeB M.ll., 3aKOHOMepHOCTH TepMH1J:eCKOrO pe~HMa rroBepx Ma~eBcKHll HoCTHoro c~oa BO~@. Ma~eBH1J: Collo JKypHa~ "MeTeopo~orHa H rH~pO~OrHa" No 2, 1967.

8. Ball F. Sea surface temperatures.Austr. Jorn. Phys. No.7, 1954 9. Swing G., McAlister E. On the thermal boundary layer of the Sea. Science, vcl. 131, Nc. 3410, 1960. TEE cmrgARATIVE MEASUREMENTS OF SEA-SURFACE TEMPERATURE IN THE U.S.S.R.

Report to the fifth session of CMM, WMO

ProfessorG. M. Tauber

At present, a great variety of methods and devioes for SST measure ments from radiation thermometers and remote electrical instruments to the old "method of bucket" and "the condenser intake method" - is used in the world network of ships' hydrometeorological stations. The latter method, which consists of SST determination from measurements in the kingston of the cooling pipe of the vessel's engine-room especially is in wide use.

For the reasons mentioned above, the SST vessel data have different errors and are qualitativelyunhomogeneous. It makes the synoptic analyses of the results of SST observation difficult, and also their use for prac tical purposes (fisheries and weather forecasting) and for scientific investigations (ocean-atmosphere interaction, general circulation of the atmosphere, heat balance of ocean and maritime climate).

Having this in mind, the fourth session of CMM, in Resolution 3 and Recommendation 1, called upon its members to pay serious attention to the problem of SST measurements and to develop the most simple, re liable and cheap method that will give identical results on vessels of various types and tonnages.

Now we have the new problem of arriving at a more correct definition of SST, because, concurrently with the old methods, the new methods (radiation thermometer) are used, which results in the SST measurements being made in different sea-surface layers-- from an upper-surface film, a few microns in thickness, to a depth of 2-3 metres. At the same time, 142 SEA-SURFACE TEMPERATURE it is known that in oertain stratifioations at different depths the temper ature oan differ greatly, which results in heterogeneity of the informa tion obtained. Aooording to Resolution 3 and Reoommendation 1 (CMM~IV), oompara tive investigations of different methods of SST measurements were made by the Soviet Union in 1967 and 1968. These measurements were made on research vessels "Academician Shirshov" (displacement of 6,800 tons), "A. J. Woeikof" and "J. M. Shokalsky" (each 3,600 tons) in different clima tic areas of the Pacific and the Indian oceans and also on small vessels of the "trawler" type (400 tons) in the Black Sea and in the Sea of Asov. The observations were made under different hydrometeorological conditions o o· from April 1967 to February 1968 between latitudes 40 Nand 60 S. They were made with the vessel in motion and drifting in a calm sea and in moderate and high seas. During this period, the minimum temperature was 0 -0.50C, maximum temperature 30.5 0; the corresponding air temperatures were _0.40 0 and 29.50 0, respectively. The comparative measurements were made under the unified programme developed in the state Oceanographic Institution, and they were tabulated in a unified form. The programme and the form of recording are very similar to those that we obtained later from the Panel for the Study of Methods and Instruments for Measuring Sea-Surface Temperature established within the OMM Working Group on Technical Problems.

Simultaneous readings of the following instruments were used for comparative analysis:

Mercury thermometers in standard mounting (the type of Spindler's mounting) overboard and in a bucket;

Electrical resistance thermometer (VRS) placed in a special recess in the hull of the vessel at a depth of 1.8 metres;

Electrical resistanoe thermometers and thermistor lowered overboard with recording at a potentiometer;

Mercury thermometer in the engine-room. MEASUREMENTS IN THE U.S.S.R.

On the research vessel "Academician Shirshov" the prototype of bucket ABC, constructed according to the working drawings presented to us by Mr. Crowford, was also used.

The measurements were carried out on the port and starboard sides of the vessel, .on the foredeck, on the stern, and· amidships. The air temperature, cloud amount and state of sea were recorded at the same time.

Altogether, 420 complete observations were carried out on these vessels.

Processing and analysis of observations In order to detect errors arising from the above methods of SST measurement, all series of comparative observations were SUbjected to statistical processing. This consists of determining the differenoe (~T) between readings of reverse deep-sea (Nansen - Peterson) thermometers, taken as standard, and the results of measurements of all the instruments. Thus, the most accurate instrument, after the deep-sea thermometer, was found., and this could then be used as a subsidiary standard for processing all data obtained. This step was found to be necessary as the deep-sea thermometers were used while drifting only.

As can be .seen from Tables 1 and 2, on r/v "Academician Shirshov" the bucket ABC was found to be suitable as a subsidiary standard, while on other vessels the electrical resistance thermometer VRS was found to be suitable. SUbsequent processing consisted in comparing subsidiary standards with readings of all other tested instruments. The results showing different values of ~T are set out in Tables 1 and 2. Moreover, an analysis of observational data was made to determine the effect of different factors (sea waves, the values of water-air temperature differ ence, insolation) and also location of observations on the accuracy of the readings of specific instruments.

The results of comparison Analysis of the data of comparative investigations enables us to come to the following conclusions: 144 SEA-SURFACE TEMPERATURE

Table 1

Frequency (%) of differences (~T) between the readings of the electrical thermometer VRS and other instruments

T T -T T -T T -T T T T Tvrs-Tdst vrs etf vrs ets vrs b vrs- m vrs-TeI

0.0 28 37 - 30 34 - 0.1 51 43 7 46 44 - 0.2 19 17 40 18 16 - 0.3 2 3 30 5 4 - 0.5 - - 13 1 1 2 0.6 - - 10 - 1 - 0.7 ------1.0-1~2 - - - - - 10 1.3-1.5 - - - - - 37 1.6-2.0 - - -- - 37 2.1 and more Total 100% 100% 100% 100% 100% 100%

T ' .,-': Electrical thermomet'er VRS. vrs Tdst - Deep-sea reverse thermometer. ~etf, - Electrical thermometer on foredeck. T ets - Electrical thermometer on stern. Tb Thermometer in mounting in the bucket,. T Thermometer in mounting m overboard. ,- Thermometer in engine-room. MEASUREMENTS IN THE U.S.S.R.

Table 2 Frequency (%) of differences ( T) between the readings of the bucket ABC and other instruments

TT -T T -T T -T ~:r abc dst abc vrs Tabc-Tb abc m(m) Tabc-TM(p) T.aDc-Te

0.1 99 65 55 58 10 - 0.2 1 22 18 8 7 - 0.3 - 8 9 25 7 2 0.4 - 5 9 6 6 - 0.5 - - - 3 - 8 0.6 - - - - 12 13 0.7 - - 9 - 6 7 0•.8 - - - - - 11 0.9 - - - - 10 11 1.0-1.2 - - - - 13 16 1.3-1.5 - - - - 18 9 1.6-2.0 - - - - 8 14 2.1 and - - - - 3 9 more

Total 100% 100% 100% 100% 100% 100%

T - The bucket abc ABC. Tdst - Deep-sea reverse thermometer. T- Resistance thermometer vrs VRS. T - Thermometer in mounting (in bucket). b Tm(m)- Thermometer in mounting (metal glass) overboard. Tm(p)- Thermometer in mounting (polyethlflene glass) overboard. T Thermometer in engine-room. er 146 SEA-SURFACE TEMPERATURE

1. The most representative place for SST measurements is the foredeck of a vessel, on the side where the ship's exhaust-water is not substantial. It was found that measuring on a deck near an exhaust-water outlet results in an apparent increase in the actual water temperature amounting on average to 0.20 C, and measurements on the stern result in an increase ofO.3-0.4°C. 2. The remote electrothermometer mounted in the bows or amidships is the most accurate and convenient to use. To obtain data representa tive of the temperature of the surface layer, it is advisable to mount this thermometer not lower than 1 metre below the draft line. The reading scale should be placed in the chart-house; and this will reduce the work of the navigator on watch in making SST measurements, and will provide the most reliable data. Therefore, .it seems desirable to adopt the recommendation of CMM addressed to maritime countries, and when designing new vessels or over~ hauling old ones, it is necessary to allow for the mounting of the above mentioned instruments as a part of the navigational equipment of the vessel. 3. When remote thermometers are not available, the SST measurements on vessels should be made by water sampling overboard. For this purpose, the ABC bucket-type instrument is the most accurate and convenient of all the instruments we have tested.

According to ~he results of tests on r/v "Academician Shirshov" the ABC bucket-type instrument has the following characteristios:

Readings of the instrument are extremely close to those of reversing thermometers. Comparison of our data shows that differences between the readings of the two instruments are, on average, O.loC in 99% of the cases. However, these data are not sufficient for the final evaluation of the accuracy of the instrument. MEASUREMENTS IN THE U.S.S.R. 147

Constructional features of the instrument decreases the errors due to evaporation and the effects of external factors (heating and cooling) on the temperature of isolated water samples. The test of cooling made on r/v "Academician Shirshov" showed that with a difference of 3-4°C between water and air temperatures, the o water temperature in the bucket decreased by 0.2 C in three minutes, o and by 1.2 C in 20 minutes. The same results have been obtained by Mr. G. Campbell in the testing of heating under conditions of greater differences between . 0 water and a~r temperature up to 10 C. The construction of the instrument permits water to circulate freely during the sampling. It is very important that the thermometer and its mounting, while sampling,should be in the environment, the temperature of which it is intended to measure. In this connexion, it should be noted that with the bucket method, these are often errors of measurement because the thermometer mounting may have been previously heated inside or by the sun or, conversely, it may have been cooled, due to low air temperatures.

Attaching the instrument to a rope by the "bridle" method with a line 2 metres long to ensure a horizontal position and sliding over the water surface without sinking appreciably.

The instrument can be used on vessels moving at high speed, which is very important since other methods of water sampling are prac tically. limited to speeds of not more than 8-10 knots. The instrument is easy to manufacture and is convenient to use during severe 'hydrometeorological conditions.

4. While drifting and at low speeds of the vessel (up to 6-8 knots) the bucket method and method of direct measurement overboard with thermometer in mounting can be used for SST measurements. 148 SEA-SURFACE TEMPERATURE

Comparison with bucket ABC and VRS shows that using these methods errors are less than 0.2oC in 70-80% of cases. With greater differences between water andai:r temperature9,however, and with high seas and winds (which result in increased evaporation) mea surements with a thermometer in a mounting give·rise to greater errors and, therefore, its use in these cases cannot be justified. This refers especially to thermometers in mountings of light poly ethylene glass.

5. Comparative observations show that the SST data obtained from the engine-room of a vessel contain larger errors and are not, there fore, representative of the actual water-surface temperature. On r/v "Academician Shirshov" errors of more than 0.5°C have occurred in 98% of the observations, while on the other vessels the measure 0 ments have shown readings which were too high by 1.2 to 2.3 C in 83% of cases. It is important to note that the values of the error do not follow any marked pattern and mainly depend on the operating conditions of the cooling system in the engine-room.

While drifting wi th the· engines stopped, the inflow of water into the cooling system decreases or ceases, which results in a sharp increase in the error of up to 8_100 C.

The values of errors in measurements in engine-rooms depend on the type and size of the vessel, which determine the depth of water intake and the length of pipeline inside the vessel between the intake and the thermometer. The quality of the thermometer, its scale, its place of mounting in the pipeline, and the degree of the possible influence of hot and cold neighbouring water-mains on its readings are of great importance.

All this shows that data relating to SST measurements from the engine-room are unreliable and that the use of a condenser intake method is unreliable and impracticable for a ship making hydrometeorological observations. MEASUREMENTS IN THE U.S.S.R.

Definition-of the term SST While considering the problem of methods for improving water-surface temperature measurements, it is necessary to define SST more precisely. We must recognize as correct the opinion of, the Panel for the Study of Methods and Instruments for Measuring SST (Working Group on Technical Problems) concerning the necessity for a definition of the term SST, since, being based on the results of the study cf the thermal structure of the surface cf the water, we can now more critically interpret the sig nificance of the data obtained, depending on the methods used.

Present theoretical and experimental investigations show that pro cesses of heat and water exchange between sea surface and atmosphere result in the formation of a thin cooled surface layer which freely transmits solar radiation to the lower layers, where it is absorbed. Presence of a cold surfaoe layer results in negative gradients of temperature in this layer having a thickness of about 1-2 cm. There fore, the' sea-surface temperature can differ considerably from the temper ature of lower layers.

According to the observations on r/v "Iceberg" in the North Atlantic,' negative temperature gradients in this layer occurred in 83% of the cases, and only in 8% of cases were there positive gradients. The mean value of gradients in the daytime was -0.43°C and at night -0.3400C. In some cases, the gradient reached a value of -2.0 C. Similar gradient values were obtained for different latitudes of c 0 the Pacific. On the average, the gradient is _0.5 C, varying from _0.1 o to -0.8 C.

The cooled surface layer called '~a cold film" was experimentally found to be extremely stable for different conditions; and the layer persists even in sea conditions up to force 6 on the Beaufort scale.

From this short review, one can see that the term SST, which is well established in the practice of marine observations and investigations, can actually be related only to a very thin surface-layer, the tempera ture of which is to some degree of approximation given by radiation thermometer readings. SEA-SURFACE TEMPERATURE

This definition is not applied to the results of measurements with standard ship instruments, as they measure the temperature, not of the water surface, but at some depth under the surface layer•. The depth of measurement depends on the extent to which the instrument is sub merged, and with methods available at present this varies between wide limits, frcm 0.5 m to 2-3 m. Considering mixing of water due to the ship's motion and the inertia of mercury thermometers, it will be more exact to speak not of the temperature at a certain depth, but of the quasi-integral· temperature characteristic of a subsurface water layer. Naturally, to obtain the most homogeneous observational data it is necessary to endeavour to limit the thickness of this layer, taking it now as 1 metre. The value is proposed on the basis of practical pos sibilities of using existing instruments, and on the basis of actual .da1a of water temperature changes in this layer, which do not exceed the required limits of accuracy of measurement.

Thus, it becomes necessary to introduce a new concept - that of TSSL·(temperature in surface-sea~water layer) as applied to the results of water temperature measurements with standard ships' instruments and methods.

Summarizing, it is possible to suggest the following terminolo gical definitions:

SST - Sea-water surface temperature, representative of a thin surface layer of water (from. a few microns to 1-2 em) being measured with a ship's or airborne radiation thermometer.

TSSL Temperature in surface-sea-water layer having" a thickness of not more than I metre, being measured with standard ships' instruments and methods.

Application of these concepts can be justified in that they give a clear idea of the physical meaning of the sea-water temperature data which is obtained.

In this report, there is no· discussion of the question of possible and required accuracy of t·emperature measurements at the surface of the MEASUREMENTS IN THE U.S.S.R. 151 sea for research and practical purposes. One can suppose a priori that ~~ is cconsiderable effect due to the space-time variability of temperature as the measurements are not generally made at a point, nor are they made instantaneously but over a certain distance, depending on the speed of the vessel and inertia of the instrument.

In this connexion, it is desirable to recommend investigations into the effect of this factor on accuracy of water-temperature measure ments (SST and TSSL) in seas and in oceans •

. REFERENCES

See page 140 of this volume.

WMO TECHNICAL NOTES

No. 10 {The forecasting from weather data of potato blight and other plant diseases and pests. P. M. Austin Bourke. No. 11 The standardization of the meaSlUement of evaporation as a climatic factor. G. W. Robertson. No. 13 Artificial control of elonds and hydrometeors. L. Dnfonr - Fergnson Hall- F. H. Ludlam - E. J. Smith. No. 17 Notes on the problems of cargo ventilation. W. F. McDonald (reprinted 1968). No. 18 Aviation aspects of mountain waves. M. A. Alaka (reprinted 1967). No. 20 {The climatological investigation of soil temperature. Milton L. Blanc. No. 21 Measurement of evaporation, humidity in the biosphere and soil moisture. N. E. Rider. No. 24 Tnrbulent diffusion in the atmosphere. C. H. B. Priestley - R. A. McCormick - F. Pasquill. No. 25 {Design of hydrological networks. Max A. Kohler (reprinted 1967). No. 26 Techniques for surveying surface~water resources. Ray K. Linsley (reprinted 1967). No. 28 Seasonal peculiarities of the temperature and atmospheric circulation regimes in the Arctic and Antarctic. Professor H. P. Pogosjan. UpperMair network requirements for nUmerical weather prediction. A. Eliassen - J. S. Sawyer - J. Smagorinsky. No. 29 Rapport pr6liminaire du Groupe de travail de la Commission de meteorologie synoptique sur les reseaux. J. Besse No. 30 { moulin, president - H. M. De Jong - W. J. A. Kuipers - O. Lonnqvist - A. Megenine - R. Pone - P. D. Thompson - J. D. Torrance. No. 32 Meteorological service for aircraft employed in agriculture and fOl'estry. P. M. Austin Bourke - H. T. Ashton IVI. A. Huberman - O. B. Lean - W. J. Maan - A. H. Nagle (reprinted 1969). No. 33 Meteorological aspects of the peaceful uses of atomic energy. Part I- Meteorological aspects of the safety and location of reactor plants. P. J. Meade (reprinted 1968). No. 34 The airflow over mountains. P. Queney- G. A. Corby - N. Gerbier - H. Koschmieder - J. Zierep (reprinted 1967). No. 35 Techniques d'analyseet de prevision des champs de vent et de temperature a haute altitude (edition £raul/aise). No. 36 Ozone observations and their meteorological applications. H. Taba. No. 371 Aviation hail problem. Donald S. Foster. No. 38 Turbulence in clear air and in cloud. Joseph Clodman. No. 39 Ice formation on aircraft. R. F. Jones (reprinted 1968). No. 40 Occurrence and forecasting of Cirrostratus clouds. Herbert S. Appleman. No. 41 { Climatic aspects ofthe possible establishment ofthe Japanese beetle in Europe. P. Austin Bourke (reprinted 1968). No. 42 Forecasting for forest fire services. J. A. Turner - J. W. Lillywhite - Z. Pieslak. No. 43 Meteorological factors influencing the transport and removal of radioactive dehris. Edited by Dr. W. Bleeker. No. 44 Numerical methods of weather analysis and forecasting. B. Bolin - E. M. Dohrishman - K. Hinkelmann - K. Knighting - P. D. Thompson (reprinted 1969). No. 45 Performance requirements of aerological instruments. J. S. Sawyer (reprinted 1969). Methods offorecasting the state of sea on the hasis of meteorological data. J. J. Schule - K. Terada - H. Walden No. 46 - G. Verploegh. No. 47 { Precipitation measurements at sea. Review of the present state of the problem, prepared hy a working group of the Commission for Maritime Meteorology. No. 48 The present status of long~range forecasting in the world. J. M. Craddock - H. Flohn - J. Namias. No. 49 Reduction and use of data obtained by TIROS meteorological satellites. (Prepared hy the National Weather Satellite Center of the U.S. Weather Bureau). No. 50 The problem ofthe professional training of meteorological personnel of all grades in the less-developed countries. J. Van Mieghem (reprinted 1967). No. 50 Le probleme de la formation professionnelle du personnel meteorologique de tout grade dans les pays insuffi- samment developpes. J. Van Mieghem. No. 51 Pl'otection against frost damage. M. L. Blanc - H. Geslin - 1. A. Holzberg - B. Mason (reprinted 1969). No. 52 Automatic weather stations. H. Treussart - C. A. Kettering - M. Sanuki - S. P. Venkiteshwaran - A. Mani. No. 52 Stations meteorologiques automatiques. H. Treussart - C. A. Kettering - M. Sanuki - S. P. Venkiteshwaran - A. ManL No. 53 The effect of weather and climate upon the keeping quality offruit. No. 54 Meteorology and the migration of Desert Locusts. R. C. Rainey. No. 55 The influence of weather conditions on the occurrence of apple scab. J. J. Post - C. C. Allison - H. Burckhardt - T. F. Preece. No. 56 A study of agroclimatology in semi-arid and arid zones of the Near East. G. Penin de Brichamhaut and C. C. Wallen (reprinted 1968).

Note: Publications in the "Technical Note" series not appearing in this list are out of print, and will not be reprinted. No. 56 Une etude d'agroclimatologie dans les zones aridas et semi-arides du Proche~Orient. G. Perrin de BI"ichambaut et C. C. Wallen. No. 58 Tidal phenomena in the upper atmosphere. B. Haurwitz. No. 59 Windbreaks and shelterhelts. J. van Eimern - R. Karschon - L. A. Razumova - G. W. Robertson. No. 60 MetcOl'ological soundings in the upper atmosphere. W. W. Kellogg. No. 61 Note on the standardization of pressure I'eduction methods in the international network of synoptic stations. M. Schiiepp - F. W. Burnett - K. N. Rao - A. Ronand. No. 62 Problems of tropical meteorology. M. A. Alaka. No. 63 Sites for wind-power installations. B. Davidson - N. Gerbier - S. D. Papagianakis - P. J. Rijkoort. No. 64 High-level forecasting for turbine-engined aircraft operations over Africa and the Middle East. Proceedings of the Joint ICAOfWMO Seminar, Cairo-Nicosia, 1961. No. 65 A survey of human biometeorology. Edited by Frederick Sargent, II, and Soloo W. Tromp. No. 66 WMO-IUGG symposium on l'esearoh and development aspects of long~rangeforecasting. Boulder, Colorado, 1964. No. 67 The present situation with regard to the application of numerical methods for routine weather prediction and prospects for the future. Bo R. DBos - E. M. Dobrishman - A. Eliassen - K. H. Hinkelmann - H, Ito - F. G. Shuman. No. 68 Meteorological aspects of atmospheric radioactivity. Edited by W. Bleeker. No. 69 Meteorology and the Desert Locust. Proceedings of the WMOjFAO Seminar on Meteorology and the- Desert Locust. Tehran, 25 November-II December 1963. No. 70 The circulation in the stratosphere, mesosphere and lower thermosphere. R. J. Murgatroyd - F. K. Hare B. W. Boville - S. Teweles - A. Kochanski. No. n Statistical analysis and prognosis in meteorology. Proceedings of the WMO inter-regional Seminar on Statistical Analysis and Prognosis in Meteorology. Paris, 8-20 October 1962. No. 72 The preparation and use of weather maps by mariners. No. 73 Data processing in meteorology. Proceedings ofthe WMOjlUGG Symposium on Meteorological Data Processing. Brussels, 1965. No. 74 Data-processing by machine methods (Report ofthe eCI Working Group on Data-Processing by Machine Methods prepared by J. F. Bosen, chairman - P. E. Kamenskaja - K. N. Rao - E. J. Sumner - T. Werner Johannessen) No. 75 The use of satellite pictures in weather analysis and forecasting. R. K. Anderson - E. W. Ferguson - V. J. Oliver (Applications Group, National Environmental Satellite Center of the Environmcntal Science Services Admi nistration). No. 76 Instruments and measurements in hydrometeOl'ology. Lectures given at the second session of the Commission for Hydrometeorology, Warsaw, 29 September - 15 October 1964. No. 77 Lower troposphere soundings (Report of a working group of the Commission for InstJ.'uments and Methods of Observation, prepared by D. H. Pack, chairman - G. Cena - A. Valentin - M. F. E. Hinzpeter - P. Vockeroth and P. A. Vorontsov). No. 78 (Revised version of Technical Note No. 27,) Use of ground-based radar in meteorology (excluding upper-wind measurements) (Report by two working groups of the Commission for Instruments and Methods of Observation, prepared by R. F. Jones, chairman - J. P. Henderson - R. Lhermitte - H. Mitra - A. Perlat - V. D. Rockney - N. P. Sellick and revised by S. G. Bigler, chairman - H. N. Brann - K. L. S. Gunu - 1. Imai - R. F. Jones L. S. Mathur - H. Treussart) (replinted 1968). No. 79 Climatic change (Report of a working group of the Commission for Climatology, prepared by J. M. Mitchell, Jr., chairman - B. Dzerdzeevskii -H Flohu - W. L. Hofmeyr - H. H. Lamb - K. N. Rao - C. C. Wallen). No. 80 Utilization of aircraft meteorological reports (A revised edition of Technical Note No. 57, published under the same title) (Report of a working group of the Commission for Aeronautical Meteorology, pI'epared by S. Sim plicio, chau'man, and V. Hoem). No. 81 Some methods of climatological analysis. H. C. S. Thorn. No. 82 Automatic weather stations (Proceedings of the WMO Technical Conference on Automatic Weather Stations, Geneva, 1966). No. 83 Measurement and estimation of evapOl'ation and evapotranspiration (Report of the CIMO Woddng Group on Evaporation Measurement, prepared by M. Gango:radhyaya, chairman - G. Earl Harbeck, Jr. - Tor J. Nordenson - M. H. Omar - V. A. Uryvaev) (reprinted 1968). No. 84 A note on climatological normals. RepOl't of a working group of the Commission for Climatology, prepared by P. J agannathan, chairman - R. ArIery - H. ten Kate - M. V. Zavarina. No. 85 Precisions des mesures pyrheliometriques. Communications et discussions presentees au COUTS de la troisieme session du Gl'oupe de travail du rayonnement de l'Association regionale VI qui s'est tenue a l'lnstitut Royal Meteorologique de Belgique it Bruxelles, 23-27 mai 1966. No. 86 An agroclimatology survey of a semiarid area in Africa south of the Sahara. J. Cocheme and P. Franquin. No. 86 Etude agroclimatologique dans une zone semi-al.'ide en Afrique au sud du Sahara. J. Cocheme et P. Franquin. No. 87 Polar meteorology. Proceedings of the WMOjSCARjICPM Symposinm on Polar Meteorology, Geneva, 5-9 Sep' tember 1966.