EVIDENCE OF SUBARCTIC WATER MASS INTRUSIONS AT OCEAN WEATHER STATION NOVEMBER John Francis Pfeiffer 9k <*. NPS-68BF76091 NAVAL POSTGRADUATE SCHOOL Monterey, California THESIS EVIDENCE OF SUBARCTIC WATER MASS INTRUSIONS AT OCEAN WEATHER STATION NOVEMBER by John Francis Pfeiffer September 1976 Th es is Advisor: R. H. Bourke Approved for public release; distribution unlimited. r i - Unclassified SECURITY CLASSIFICATION OF THIS PACE {Wt>t* Dmlm Bnfrmd) REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM 1 REPORT NUMBER NPS 68BF76091 2. GOVT ACCESSION NO 3. RECIPIENT'S CATALOG NUMBER 4. TITLE (mnd Subtttlt) Evidence of Subarctic Water Mass Intrusions at Ocean Weather Station NOVEMBER S. TYRE OF REPORT * PERIOD COVERED Final Report 1 July 1975 - 30 June It « PERFORMING ORG. REPORT NUMBER 7. AUTHOR^ John F. Pfeiffer, in conjunction with Robert H. Bourke I. CONTRACT OR GRANT NUMBERS; 9. PERFORMING ORGANIZATION NAME AND ADDRESS Naval Postgraduate School Monterey, California 93940 Code 68BF 10. PROGRAM ELEMENT. PROJECT, TASK AREA 4 WORK UNIT NUMBERS N63134-75-PO-50005 II. CONTROLLING OFFICE NAME AND ADDRESS Commanding Officer Fleet Numerical Weather Central Monterey , California 93940 12. REPORT OATE September 1976 13. NUMBER OF PAGES U. MONITORING AGENCY NAME * AOORESSf// dltloront from Controlling Ofllem) Naval Postgraduate School Monterey, California 93940 IS. SECURITY CLASS, (ol thlm riport) Unclassified ISa. OECLASSIFI CATION/ DOWN GRADING SCHEDULE 16. DISTRIBUTION STATEMENT (o< thia Report) Approved for public release; distribution unlimited 17. DISTRIBUTION STATEMENT (of fna obttrmct ontmrmd In Block 30, II dlllotont from Roport) IS. SUPPLEMENTARY NOTES l». KEY WORDS (Contlmim on rororao tldo II nacaaaarr mnd Identity by block nuaioar) Thermal advection Baroclinic Rossby wave Ocean Weather Station NOVEMBER Heat budget Subtropic front Subarctic water mass L 20. ABSTRACT (Contltnto on rmvormo tide II nacaaaarr anal Identity by block nvmmm*) A divergent heat budget equation which included the effects of surface heat flux, horizontal and vertical advection, and horizontal divergence on the near-surface heat content was used to examine the role of thermal advection in the upper 250m of the water column at Ocean Weather Station NOVEMBER. This station is located on the southern boundary of the transition zone separating the Subarctic water mass from the Subtropic VU 1 JAN 73 (Page 1) 1473 EDITION OF I NOV »(« Enfrmd water mass. Values for horizontal thermal advection changes were computed over the period 1962-1970. This term was correlated with salinity fluctuations over the period 1968- 1970. Pulse-like periods of cool advection were associated with periods of reduced salinities suggesting these were intrusions of Subarctic water. Over the nine-year period of analysis, these intrusions had a periodicity of 7 to 8 months with a duration of 3 to 4.5 months. It is suggested these wave-like intrusions along the Subtropic front are the result of the passage of non-dispersive baroclinic Rossby waves . DD Form 1473 TT . . - . 1 Jan 73 Unclass if led b/ N 0102-014-6601 7 SECURITY CLASSIFICATION OF THIS P*OEfW.r Dmtm Cnitrmd) Evidence of Subarctic Water Mass Intrusions at Ocean Weather Station NOVEMBER by John Francis Pfeiffer Lieutenant Commander, united States Navy B.S., United States Naval Academy, 1967 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN OCEANOGRAPHY from the NAVAL POSTGRADUATE SCHOOL September 1976 NAVAL POSTGRADUATE SCHOOL Monterey, California Rear Admiral Isham Linder Jack R. Borsting Superintendent Provost This thesis is prepared in conjunction with research supported in part by Commanding Officer, Fleet Numerical Weather Central, Monterey, California 93940, under Project Number N63134-75-P0-50005. Reproduction of all or part of this report is authorized. Released as a Technical Report by: <^y X : ABSTRACT A divergent heat budget equation which included the effects of surface heat flux, horizontal and vertical ad- vection, and horizontal divergence on the near-surface heat content was used to examine the role of thermal advection in the upper 250m of the water column at Ocean Weather Station NOVEMBER. This station is located on the southern boundary of the transition zone separating the Subarctic water mass from the Subtropic water mass. Values for horizontal thermal advection changes were computed over the period 1968-1970. Pulse-like periods of cool advection were associated with periods of reduced salinities suggesting these were intrusions of Subarctic water. Over the nine-year period of analysis, these intrusions had a periodicity of 7 to 8 months with a duration of 3 to 4.5 months. It is suggested these wave-like intrusions along the Subtropic front are the result of the passage of non-dispersive baroclinic Rossby waves. TABLE OF CONTENTS I . INTRODUCTION 10 A. GENERAL 10 ' B. THE HEAT BALANCE-- 10 C. OBJECTIVES 12 D. METHODOLOGY 12 II. THEORY 14 A. GENERAL 14 B. SIGN CONVENTION 16 1. Net Surface Heat Exchange-- 16 2. Divergent Heat Change 18 3. Horizontal Advection 18 III. ANALYSIS OF DATA 20 A. SOURCE OF DATA 2 0 B. ANALYSIS TECHNIQUE 21 1. Determination of Local Heat Change 21 2. Net Surface Heat Exchange 22 3. Heat Divergence 23 4. Example of Computation Process 23 IV. RESULTS " 2 8 A. GENERAL 2 8 B. MEAN ANNUAL CYCLE 5 2 C. HORIZONTAL ADVECTION ANOMALIES 38 D. BAROCLINIC ROSSBY WAVES 41 E. SUMMARY 45 5 V. CONCLUSIONS 45 APPENDIX A. HORIZONTAL ADVECTION EVENTS 47 BIBLIOGRAPHY 50 INITIAL DISTRIBUTION LIST 52 LIST OF TABLES I. Extract from mean monthly temperature profiles by Ballis, 1973 24 II. Salinity changes (°/oo) during horizontal advection anomalies - - 49 LIST OF FIGURES 1. Direction of positive (heating) values for the DHB terms 17 2. Nine-year plot of the monthly change in heat content (3H/3t) in the upper 250m at OWS N 29 3. Nine-year plot of the major contributors to the monthly heat change (3H/3t) at OWS N 30 4. Mean annual trend in the monthly change in heat content (3H/3t) at OWS N. Bars indicate one standard deviation (±lcr) 34 5. Mean annual trend in the horizontal advection contribution (-V^ • VftH) to local heat change (3H/3t) at OWS N. Bars indicate one standard deviation (± 1 a) 35 6. Mean annual trend in horizontal divergence plus vertical advection contribution (H^iv) to the local heat change (3H/3t) at OWS N. Bars indicate one standard deviation (± 1 a) 36 7. Mean annual trend of net surface heat exchange contribution (Qn) to local heat change (3H/3t) at OWS N. Bars indicate one standard deviation (tiff) 37 -*- 8. Major events in horizontal advection (~vh • vh^) for 19 68-1970 40 9. Typical temperature, salinity, and density structure in the Subarctic Pacific Ocean (after Tully, 1963) 42 10. Major events in horizontal advection (-V^ • V^H) for 1968-1970 48 ACKNOWLEDGEMENT The author wishes to express his sincerest appreciation to Dr. Robert H. Bourke for his excellent advice, encourage- ment, and constructive criticism throughout the entire study Professors Robert L. Haney and Jacob B. Wickham, and, in particular, LCDR Norman T. Camp provided valuable insight and were instrumental in making this research a worthwhile educational experience. I. INTRODUCTION A. GENERAL The heat content of the upper layers of the ocean has recently been of interest, particularly with respect to the factors or forces which alter the heat state. Changes in the heat content result in changes in the temperature struc- ture and, with that , changes in the density distribution and oceanic layering. A knowledge of these factors and an attempt to predict and quantify their contributions is of importance to the United States Navy, especially with regard to anti-submarine warfare and its strong dependence on sound speed profiles which readily respond to alterations of the thermohaline structure. B. THE HEAT BALANCE An influential process in the alteration of the thermal structure begins at the air-sea interface with insolation, the rate at which short-wave energy from the sun enters the sea. This factor, combined with the effects of back radia- tion, evaporation, condensation, and conduction, produces a net surface heat exchange across the air-sea interface. In shallow water the effect of this net heat exchange is dis- proportionately large; in deep water its role as a significant process can vary with respect to other forces. 10 Redistribution of heat can occur via convective and mechanical mixing, the former primarily a seasonal process responding to the air-sea temperature differences, the latter responding to the effect of wind and waves. Both are significant in the layer above the main thermocline. Advection, vertical or horizontal, is a process of transport of an oceanic property (e.g., temperature) solely by the mass motion of the ocean. Its importance as a major heat contributor has been uncertain. Studies as recently as Thorne [1974] have computed horizontal thermal advection as the difference between the surface heat exchange (Qn) and the local change in heat content (aH/8t). However, no cor- relations between the changes in horizontal advection and those of net surface heat flux, atmospheric pressure, or salinity were undertaken to support the validity of the model It appears from this work and others [Tabata, 1961; Clark, 1967; Bathen, 1970] that horizontal advection does not com- pletely account for the difference between local change in heat and net surface exchange; additional processes must be considered to account for this difference. Emery [1976] assumed that, in addition to Qn and hori- zontal advection (-Vn" vhH) » vertical advection and the compensatory action of horizontal divergence/convergence were major heat processes in the upper layer of the ocean. Correlation studies which included these additional terms suggested that they played a significant role in the heat budget of the upper ocean. Emery developed a divergent heat 11 budget (DHB) equation to model the correlation of these terms with the heat content of the upper layers. However, the contribution of each of the DHB terms to the heat balance was not specified nor were they correlated with changes in physical properties, e.g., salinity or oxygen. C. OBJECTIVES The purpose of this study was to complete and extend the findings of these earlier investigations. Specifically, the objectives of this research were: • to quantify each expression in the DHB equation • to determine the following trends - mean annual - yearly by month - anomalies • and in particular, to examine the significance of horizontal advection in changing the heat content of the upper ocean and to determine the validity of the results by correlation with other oceanic parameters. D. METHODOLOGY In order to measure .horizontal thermal advection (-Vh« VnH) , one must either possess a large horizontal data field of temperature and velocity to calculate the advective term directly, or deal with temperature and salinity measure- ments from a point source which necessitates accounting for all the expressions in the heat balance and attributing the residual to horizontal advection. In either case, it is important to validate the horizontal advective heat process 12 by comparing changes in the thermal structure with changes in a physical property such as salinity. Such a thermohaline re-structuring can be the result of water mass intrusions. Therefore, in order to realize the objectives of this research, a long-time data record is needed as well as a location that is readily influenced by water mass intrusions. Ocean Weather Station NOVEMBER (OWS N) affords such an oppor- tunity, since it offers both a long period of data and is strategically positioned at the transition zone between the Subarctic and Subtropical water masses. The Subarctic water mass is a cool, low-salinity body of water occupying the upper 300m of the Pacific Ocean, generally north of 40°N. Subtropical water is characterized by a shallow, warm, salty layer in the vicinity of 32°N. The transition zone between these water masses is characterized by sharp horizontal heat and salinity gradients [Roden, 1976] . 13 II. THEORY A. GENERAL In his examination of the role of vertical motion in the heat budget of the upper 250m of the ocean, Emery [1976] found changes in the heat content (3H/8t) at OWS N could only be moderately explained by changes in heat across the air-sea interface (Qn) . However, heat content changes could be explained far more accurately when two additional pro- cesses, horizontal divergence and vertical advection, were incorporated with the surface heat flux. To establish this correlation, he developed a divergent heat budget (DHB) equation which included these processes. The major assumptions in the development of the DHB equation are: • Molecular diffusion and radiative heat flux out of the column are neglected. • The field of motion is comprised of a steady mean flow upon which are superimposed internal Rossby waves with periods from months to years. • Temperature changes at the bottom of the water column can be attributed mainly to vertical advection of the mean thermal gradient at that depth. Since the base of the column is always beneath the main thermocline at OWS N, turbulent energy at this depth is small; therefore, temperature changes due to turbulent mixing can be considered negligible. Starting with the equation for the conservation of heat and by incorporating the above assumptions along with a Reynolds decomposition to the temperature and velocity 14 expressions, Emery arrived at the divergent heat budget equation, 3H/3t - -^ [H - pC . DTD] * % ■ VhH = Qn (1) where term 1 is the local change in heat content of a column of water of depth D; term 2 is the heat change due to hori- zontal divergence (VLH/D) and vertical advection (pC WUTn) through the bottom of the column; term 3, the change due to horizontal advection; and term 4, the resultant gain or loss of heat across the air-sea interface. Term 2 may be combined as shown and expressed as H,. , the net change in heat content r div ° due to near-surface divergence/convergence and the compensatory upwelling/downwelling mechanism. Q is comprised of the following major terms, -n where % - %-%-%- % w Q = the rate at which short-wave energy from the sun enters the sea. Q, = the net long-wave back radiation from the sea surface. 0 = the net latent heat transfer (includes xe evaporation and condensation) . Q, = the net sensible heat transfer (conduction) . 15 H represents the heat content per unit surface area, evaluated for the upper 250m, and is computed by 250 H = Q/pCpT dz (3) where p is the density and C is the specific heat at constant pressure. W~, the mean vertical velocity, is computed by D 91D/8z which assumes that temperature changes at D are due mainly to vertical advection of the mean thermal gradient at D. BT^/ represents the monthly temperature change at 250m and BTywj. is the mean vertical temperature gradient at 250m. B. SIGN CONVENTION The following sign convention is explained to aid in later physical interpretation of the results (refer Figure 1) Taking z positive downward and rearranging equation (1) to solve for the change in heat content in the water column, 3H/3t = Qn + WD/D [H - pC DTD] - \ • VhH. (5) 1. Net Surface Heat Exchange, Qn A positive value for Qn indicates that the column of water gains heat as the net result of the various heat 16 + y (North) Surface (-Vh-VhH) 250 m ► +x (East) WD^.CPTD Figure 1: Direction of positive (heating) values for the DHB terms. 17 exchange processes acting at the air-ocean boundary [equation (2)]; negative values indicate that the ocean releases heat to the atmosphere via these processes. 2. Divergent Heat Change, Hdiy = WD/D [ H - pC DT ] a. [H - p C DT 1 is always positive since it represents pC (T - T^™ ) where T is the average temperature throughout the column. b. From equation (4), W~ can be either positive or negative. Examination of the records shows that at OWS N 3TD/3z is always negative at D = 250m. If 3Tn/3t were positive, i.e., a heating process, then a positive W~ results (downward vertical velocity) . A negative W„ (upwelling) represents cooling. It follows then that term 2 of equation (5) responds directly to the sign of W^. Therefore, the cooling process is characterized by an upward vertical velocity (W~ negative) , accompanied by an outward flow across the vertical sides of the column (divergence) in accordance with the law of continuity. Downward vertical velocity and inward flow are associated with heating. 3. Horizontal Advection (-V, • V,H) Expanding the horizontal advection term into its component parts yields -V, • V,H = -(ui + vj) • (9H/3x i + 9H/3y ]) . (6) In the vicinity of OWS N isotherms are generally zonally distributed, hence horizontal advection reduces to(-v9H/9y). The north-south temperature gradient, 9T/9y is generally negative and can be appreciable when the southern boundary of the Subarctic/Subtropic transition zone migrates to the position of OWS N [Roden, 1976]. Positive horizontal advec- tion indicates that heat is carried into the column generally under conditions of a northward flowing current. A cooling process is generally in response to a flow from the north. When horizontal advection is due to the passage of an eddy, the flow direction cannot be readily inferred. Whether heat is gained or lost is dependent on the sign of the horizontal temperature gradient, i.e., whether a warm core or a cold core eddy is involved in the advection. 19 III. ANALYSIS OF DATA A. SOURCE OF DATA Mechanical bathythermograph (MBT) data have been taken two to four times daily at OWS N since 1947. Monthly means of temperature at 5m intervals have been produced by Ballis [1973] from this data file. Significant gaps exist in these data for the earlier years, limiting a detailed analysis to the nine years, 1962-1970. In addition, MBT casts were limited to 135m prior to 1962, a depth too shallow for use in this study as the mixed layer is often slightly deeper than this depth, especially in winter. Within the 1962-1970 MBT time series, minor data gaps were eliminated by incorporating temperature data from the daily hydrocasts taken at OWS N since 1964 and on file at the National Oceanographic Data Center (NODC) . Although the monthly means calculated from the hydrocast data are based on fewer observations than those from the MBT data, no sig- nificant difference was detected between either data set. Standard deviations were similar and for those months when both sets of data existed, little difference in mean values was noted. A salinity time series was created from the NODC hydro- cast data. Data gaps limited this series to the period 1968' 1970 since only a few short records existed prior to 1968. 20 Monthly means and anomalies of the net surface heat exchange (Q ) were taken from the report by Rabe and Bourke [1975] which examined the various surface heat exchange pro- cesses at OWS N. These data are based on three-hourly observations from whence mean daily values for each month were calculated from standard bulk flux formulae. The dominant heat flux term, net solar insolation (Q ) , was calculated from the improved version of the Q equation developed by Seckel [1973] . B. ANALYSIS TECHNIQUE Of the terms in the divergent heat exchange equation (1) , the primary focus of this report was directed towards the horizontal advection term (-V, • V, H) . This is a residual term computed from the difference between the observed change in heat content (3H/8t) and the computed values of surface heat flux (Q ) and divergent heat flux (H , - ). This type of analysis affords the possibility of error accumulation in the residual term. However, it will later be shown that, in spite of the possibility of error accumulation, the large anomalies in horizontal advection are significantly greater than the potential error contribution. 1 . Determination of Local Heat Change Changes in the heat content of the water column are evaluated in this study on a seasonal scale by analyzing the temperature profiles from the MBT data of Ballis [1973] and the hydrocast data of Rabe and Bourke [1975] . These monthly 21 observations can be considered, in most cases, to be centered about the 15th of the month, since most monthly observations were taken over the entire 30-day period. Some monthly means were skewed to either side of the 15th in response to the biased nature of the data collection, e.g., temperature data for May 1968 were collected between the 12th and the 31st, shifting the effective mean date for this month to 21 May. It then follows that the average monthly heat content of the upper 250m, H, can be determined from these mean temperature profiles and assigned the same effective date. The mean heat content of the column is found from H = /25° PC T dz (7) OP where p"C is the product of the mean water density and - 3 - 1 specific heat, assumed constant at 0.978 g-cal cm °C , and T is the mean temperature profile over the specified time interval . The average monthly heat content was computed for the period 1962-1970; 3H/3t was subsequently determined for the interval between successive H's. The interval, 3t, ranged from 23 to 52 days. Values are computed in langleys per day (ly day"1). 2 . Net Surface Heat Exchange The monthly mean daily values of Q , Q , Q, , and Q, were combined to determine the monthly mean daily contribution of net surface heat exchange, Q . 22 3. Heat Divergence (H, . ) As previously discussed, H , . is the expression that div r includes heat exchange due to horizontal divergence and vertical advection, where WQ = -3TD/3t [ a^D/azl" • Similar to the procedure used to calculate 3H/3t, 3Tn/3t was computed from the mean temperature at 250m, i.e., Tn = T?c;n over the specified time interval, the interval between effective dates. The temperature gradient (9TD/9z) at 250m was computed from the mean temperature profiles between 150 and 250m. This value ranged from -0.043 to -0.069 °C m . 4 . Example of Computation Process The following is an example of the analysis technique applied to May and June 1968 (see Table I) . a. Effective Date As shown in the listing of the daily distribution of MBT observations for each month [Ballis, 1973], MBT casts for May 1968 were made daily between the 12th and the 31st; no observations were taken prior to the 12th. Therefore, the effective data for the mean monthly temp, T., , is 21 May. Examination of the following month's record indicates at least one lowering per day was taken every day of the month (the most common situation). Hence the effective date, Tj , is 15 June. The time interval (3t) between L and TJune is established as 25 days. 23 X E- a, w P co ■a H < Q a s-s •-3 co v— ' co CD i— 1 u •H 4H E- Q r* co 5-. < D c3 >-• >^ CD § a o CO v— ' & = CD +-> CJ E- to "3- ^3" OUIOiflO to o lo o lo P*. 00 00 CT> C"n OlflOWO OOHHN cm cm cm cm cm iflomoi/i CM to fO rf *-r cm cm cm cm cm o lo o lo o cm cm cm cm cm 0^ to i—t lo o r-. *rf t-» o r-- r-* oo o m m n (Onoooq co i-h to o o vOOLO LO LO x-f rj- tO tO tO tO tO to K) M (O K1 Kl tO tO tO CM CM N Kl fl rO tO t*» h 't o ^ vO cr> to o o cm o lo n^omooo Hootvoon Lncorv^-to pH^-iH 0^0(0000 \0 fO r- 1 CT> vO ^HmvOLO IOOOOOU1 ^ H O C) CO r^voo lo lo lo lo t-j- rf nr «* ro w o ro n n n nnhh h r-( i— i r— I O O OOO OOOOO OOOOO OOOOO OOOOO O O O O C"> cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm n nnn h m tv rr t i/inoo^ r-- oo r^ vO o lo lo to to to to to cm cm cm cm cm cm cm to cm cm LO LO cm o O 00 r— * CM LO co to CO C-- i—i LO a> in r^ o O o r«» rf r-i 00 •*r cm a\ VO to r-t CO vO ^f cm ct> r-. in to rH c-n oo ■o rr to CM rH o r-» \o o vo ifl ui ui t rr ^r ^ to to to to cm cm cm n n hh h c o X E- s CU 3 • a CD to rom m , 197 CO < Q <+H CO •H UA +J r-t U r-t X a o ai aj »~3 CO s— ' ?-• M +-> X >> W .0 u E- i— t co CD i— 1 >• < Q as E- .>- >/-> < a o «£. CO W CJ o lo o lo o lo o lo o lo o lo o lo o lo o lo o lo o lo o lo o lo o H H CN CM tO tO Tj- Tf WLOifl^S NCOCOOlCTl O O i— I i— IN CM tO CTA C7» CTA 0"> 0"» CTft CTT> cm O C> 0*> C7> CTft CJ* CTT> Ol CT> G*» d C7) Cl Ol CTl Cl O C> cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm cm oio onco n to t r>. r^ i— ir>-s»OLn «g- to to cm «rr c-- rH lo o o loovO-OvO go o> Q r> io lo Tr •<-j- rr *rr fr t ^ ^* *3" ti/im in o LO "5f r-t vO oo r-t rr i—t <=r o *r rj- r>» i— t r^ rg r~> "* o Lit rH rr CT> to to i— 1 LT) <3 vO vO LO ^ to ~ O • r-» TT CN N CN N CN CMr-lr-tOO O C^ OMTl Ol 0%Cnc7^COCO CO OO 00 CO N r^r»~ rMCSJCNICMfM fsjrMrMrMCM Nr IHHr- 1 Hi- 1 i— I r-t rH i— I r- 1 i— I r- If- 1 i— I i— I OOOOO CM N CM N N OOOOO CM fs] CM CM CM OOOOO CM CM CM CM CM OOOOO CM CM CM CM CM OOOOO o o CM CM CM CM CM CM CM t^ <* to t^ to rra>o lolo oooooo a^o^ocr>o ooi—ttor^ cocm MOO N W^t tO N CM CM N CMCMCMtOtO CMCMCMCMtO tOtOtOtOtO tO »* o r-» r^- <2 vO r^- r- vO LT) LO rf t~» CM Cl LO CNJ 00 LO CM 00 to O r-«. LO _ o — < o o oo vO ^r CVJ o 00 r~^ o LO "3- '^r to to to CM CM CM i— t r-t o CTl CO r-- LO to oqoc^Ci q ctico co oo cocooococo x oo coco oo aocor-.r-.r-. r>~r-- CM i— t i— t i— I i— I rH rH r-H rH i— I rHrHrHi—ti—t rHrHrHr-li— I r-t — t r-t rH ri i— t — t 24 b. Heat Content The mean monthly heat content was computed by- integrating the mean monthly MBT profile [Ballis, 1973] .to 250m at 5m intervals. From equation (7), H = 401,000 gm-cal cm"2 (langleys = ly) "june = 418>600 ^ It follows that the 3H/3t for this 25-day period is a heat gain of 17,600 ly or an average daily heat gain of 700 ly day"1. c. Vertical Velocity Recall that 9T~ ,_n/3t Wn = - ^^ ; D 8T250/8z (1) For May !I250 = [11.49- 16. 10]°C = _0>0461oc m-l 33 250 - 150m (2) For June ^250 = [11.35-15. 90j°C = _0<0455oc .-1 83 250 - 150m Therefore, the effective vertical temperature gradient during the 25-day interval separating the average profiles is -0.0458°C m"1. 25 (3) Change in Temperature at 250m 3T250 [11.35 - 11.49]°C n nnz,Qr A -1 — ~ - - * = -0.0056 C day dt 25 days Therefore, D rr _.[-0.0056] n n oo A "1 Wn = - = -0.122 m day [-0.0458] d. Average Net Surface Heat Exchange, Q Average daily Q May : -2 24 ly day"1 June: 22 7 ly day The average net surface heat exchange for the period 21 May- 15 June is: Qn . [224] [10 days] * 227 [15 days] , 225_8 ly day-l 2 5 days e. H, . div The H, • term is dependent upon the mean heat div r r content in the column between the effective dates (H) and the ean temperature at 250m over this same period (T2;-q) : W "div = -^ [ K ~ CpDpT250 ] ; aiv 250m p Zi3U D = 250m From 4.b. and Table I, Tf= 418,600 + 401,000 = 409j85Q 2 -=- 11.49 + 11.55 -- .9or 450 2 ' 26 m £. Horizontal Advection ( -vT • V FT ) Solve the remaining terms in equation (1) where D = 250m : — W ~R 3H TV D r TT vh,vhH = 3?-^ = ftH- PCpDTD ] [409,850 - CO. 978) (2. 5 x 104 cm) (11 . 42°C) ] 538 ly day ( a heating process ) . This is the average daily contribution to the heat column over the period from 21 May to 15 June 1968. 27 IV. RESULTS A. GENERAL Emery [1976] showed that the mean annual cycles of heat content within the mixed layer and within the upper 250m of the water column were significantly out-of -phase. This indicates that below the main thermocline, where turbulent kinetic energy is minimal, there are additional factors involved in the heat transfer process other than horizontal advection. Noting the fluctuations of the 14°C isotherm at both the short and long time scales, Emery assumed that vertical motion at 250m contributed significantly to the variability in the heat content and should be included in any conservation of heat equation. As illustrated in the preceding section, the magnitude of these vertical terms and the net surface heat exchange were computed from the monthly mean temperature profiles. The remaining contribution to the observed change in the heat content was attributed to horizontal advection, a term not computed specifically, but determined as a residual from the DHB equation. The mean local change in heat content (3H/3t) is plotted for the period 1962-1970 (Figure 2). A plot of the factors which contribute to the local heat change ( Q , H,. , -V, • V, H ) are plotted for the same period in Figure 5. div* h h ' r 28 Figure 2: Nine-Year plot of the monthly change in heat content (3H/3t) in the upper 250m at OWS N. 29 — — HOmzxrm. aovecroH i-vuvumi NTT SUtnct X4T EXOUNOC (01 M OV ^(M^COTo) ■iud NO o»r» Figure 3: Nine-year plot of the major contributors to the monthly heat change (3H/3t) at OWS N. 50 One immediately notices in Figure 2 the pulse-like occur- rences of large anomalies in the local heat change, especially the numerous events of large heat withdrawal, e.g., February 1968 and April 1969. Comparing this time series with the corresponding values of the DHB terms of Figure 3, one notices that the pulse-like nature of the large anomalies in 8H/9t, which is physically measured, cannot be satisfactorily accounted for by Q or H,. , since the former exhibits no pulse-like anomalies and those of the latter are small. It is horizontal advection, in fact, which must account for the large fluctua- tions in the 8H/at time series. Emery showed that upwelling, which is reflected in the H , . term, operates to change the relationship between Q and 3H/3t. However, this does not mean that H, automatically acts as the major influence in div } J the balance of heat. What Emery has done, in fact, in his examination of the role of vertical motion in the heat budget, is to make more reliable the approximation of the horizontal advection contribution. Therefore, of the DHB terms, the primary interest of this study lies in the role that horizontal thermal advection plays in altering the heat content of the upper layers of the water column from year to year. Since -tf, . y,H is a residuallyydetermined expression which is potentially contaminated by measurement and modeling inaccu- racies, it is necessary to evaluate whether the magnitude and fluctuations of the term bear any resemblance to physical reality. 31 One possible approach is through water mass analysis, since the transport of heat is usually accompanied by the transport of other properties such as salinity or dissolved oxygen. Water mass analysis through temperature and salinity correlations is particularly useful at OWS N since it lies on the southern edge of the transition zone between the Sub- arctic and the Subtropical water masses, a frontal boundary distinguished by its sharp horizontal and vertical gradients. The position of OWS N is fortuitous since any significant in- trusion of cool, low-saline, Subarctic water should strongly influence the thermohaline structure of the upper 250m. Correlation of salinity changes with thermal advection changes at OWS N is necessarily limited to the three years, 1968 through 1970, due to a lack of salinity data for the previous years. However, because of the good correlation between salinity and thermal advection during 1968-1970 (shown later in this section) , one can surmise that thermal "events" of horizontal advection in the earlier years can be related to water mass intrusions. First, the data will be analyzed by examining the mean annual contribution from each of the terms in the DHB equa- tion, followed by a discussion of the major horizontal advective anomalies. B. MEAN ANNUAL CYCLE The mean annual cycle for each term in the DHB equation was calculated by averaging the mean monthly values from 32 1962-1970 (Figures 4 to 7) . One must interpret the results with caution as the mean annual curves are based on a maximum of nine years of monthly averages, far too few to draw conclusive results. The mean annual cycle of 3H/3t (Figure 4) exhibits a broad general tendency for heating from April to September, followed by a cooling trend for the remainder of the annual cycle. The actual curve is a series of oscillations fluctu- ating about this trend by 200 to 300 ly day . However, of the component terms in the DHB equation, only Q displays an annual cycle and is therefore the term which creates the cyclic trend in 3H/3t. The other terms appear to fluctuate as pulses and serve to enhance or to suppress the trend in surface heat flux. The pulse-like nature of the horizontal advection term appears non-seasonal and is noticeably larger in most cases than Hi. in its contribution to 3H/3t. If, as anticipated, its oscillatory nature is due to a water mass intrusion, the thermohaline structure of the environment should be significantly altered. Hence, salinity changes during any large advective cycle should also exhibit a pulse-like nature This hypothesis is supported by Hansen [1973] who observed that surface salinity changes at OWS N were pulse- like occur- rences displaying neither a seasonal trend nor a correlation with any seasonal process, e.g., evaporation minus precipita- tion. 33 LANGLEYS PER DAY JAN | FEB | MAR | APR | MAY | JUN | JUL | AUG | SEP | OCt| NOV | DEC Figure 4: Mean annual trend in the monthly changes in heat content (3H/3t) at OWS N. Bars indicate one standard deviation (±la) 54 LANGLEYS PER DAY | JAN | FEB | MAR | APR | MAY | JUN | JUL | AUG | SEP | OCT | NOV | DEC Figure 5: Mean annual trend in the horizontal advection contribution (-V^ * ^H) to local heat change (3H/St) at OWS N. Bars indicate one standard deviation (±la) . 35 LANGLEYS PER DAY JAN JFEB j MAR | APR | MAY | JUN |JUL | AUG | SEP j OCT | NOV | DEC Figure 6: Mean annual trend in horizontal divergence plus vertical advection contribution (H,. J to the local heat change aiv(3H/9t) at OWS N. Bars indicate one standard deviation (±1*^— . _»■ '"^ fc~\ ,^" ^SS\'-- ^ Ml 1 - ^ 1 •' .«"'" 7 ^ X o Jr.* JZ > i±j \ > i X • z X _ 8 -° r^i~__^ o UJ Q * ^^^N 5 UJ J*- X X "\, ^^y UJ > a '■■,/>T < _j S|o )\ > 0. 2 .■■'\s / z o M 3 2 ■' ^*^ ^*^' « > < /^- £ I- 5 iV o UJ o \ ^\ X Z X z \, ;\ X ^-^s^ X I X X ^^v ' '" "■—■«« r X 1 X 'si. \ ^••- — t \ c _ ■ ) 1 — 1 1 1 *\ 1 • 1 1 1 1 1 03 < 03 J- < O o> rH I 00 vO G\ u o 03 00 < b>~ o •f-l u > 03 O CM ■H S- O ^= V) c ~ 03 < o 00 L CM j-l <0 0) o 03 :d V 3 CO — o o S 9 AVQ H3d SA312NV1 40 salinity in the upper layers are lowered. During coincident upwelling periods (Hdiy negative) , the salinity decrease due to advection is suppressed by the introduction of higher salinity water from the lower portion of the Subarctic halo- cline (Figure 9) . However, during Event 4 the downwelling action enhances the salinity reduction due to advection by mixing downward the lower salinity values of the upper halocline. D. BAROCLINIC ROSSBY WAVES These thermohaline changes suggest the intrusion and subsequent recession of the Subarctic water mass at OWS N in a pulse-like, non-seasonal manner. This translation of the boundary may be explained only partially by examination of surface wind stress since the resultant Ekman dynamics cannot fully account for the temperature and salinity fluctuations between the main thermocline and 250m. The non-seasonal nature of the horizontal advection anomalies indicates that perhaps the driving mechanism itself is non-seasonal. An examination of atmospheric data from 1968-1970 reveals no unusual atmospheric pressure anomalies or significant weather deviations during the time of the large advective anomalies. What is apparent, however, about these pulse-like cooling/heating trends is their 7- to 8-month periodicity -- an indication perhaps that the southern boundary of the Subarctic/Subtropic transition zone might be responding to the passage of non-dispersive baroclinic Rossby waves, as suggested by Bernstein and White [1974] . 41 0) u 3 *-» o • 3 ^^ J-. ^ +J vO co O 1— I >s ■M * •H >N t/> l—i e f-t 0) 3 ^3 H T3 u C (U 3 4-) U-t * X >s >— - •M •.-( g C C3 •H O i— 1 U rt o V) U «* •i-( N xs U 3 =0 (S&3-LjW) Hld30 42 In their study of the time and length scales of baroclinic eddies in the Central North Pacific Ocean, Bernstein and White examined the subsurface temperature fluctuations at OWS N using the MBT data compiled by Ballis [1973]. They determined that the large-amplitude fluctuations of the average monthly temperature at 200m, which was below any seasonal influence, could not be due to effects of internal waves or variations in the geostrophic flow. They noted a long-period fluctuation of six years upon which smaller variations occurred with periods of two to six months. They attributed these scales of motion to baroclinic, non-dispersive, Rossby waves. How- ever, they were unable to localize the dominant period of maximum energy in the spectral density functions for the sub- surface temperature profiles as their data record length (the same as used in this report) limited the maximum resolution to about six months. Much energy is undoubtedly contained in longer period waves. From his study of spectra from tempera- ture records from the ocean weather ships in the Atlantic, Gill [1975] showed that the spectra are different at each station, but that large energy levels existed at periods greater than six months, especially for stations near frontal boundaries. It is suggested, then, that the 7- to 8-month occurrences of large advective cooling/heating, evident in this analysis, are the results of baroclinic Rossby waves. E . SUMMARY It is apparent that the effect of these wave-like eddies on the temperature and salinity structure near water mass 45 boundaries can be substantial. Each of the large advective anomalies was characterized by an initial cooling trend, an indication that the southern edge of the transition zone at OWS N was being pushed southward. The thermal and salinity structure of the water column reacted accordingly to the intrusion of fresh, cool water from the north. As the eddy progressed westward, it allowed the transition zone boundary to "rebound" towards the north wherein the thermal and salin ity structure responded to the return of the warm, saline subtropical water mass. 44 V. CONCLUSIONS This study investigated the effects of horizontal thermal advection in the upper 250m at OWS N from 1962-1970. In accordance with the divergent heat budget (DHB) equa- tion, heat changes that were not due to the combined effects of net surface exchange, vertical advection, and horizontal divergence were attributed to horizontal thermal advection in maintaining the balance of heat in a column of water. The investigation concentrated on the period 1968-1970 and centered on water mass analysis and its relation to the large anomalies in horizontal advection. The results of this research show: • The divergent heat budget equation is a valid approach to the determination of horizontal advection. • Qn, the net surface heat exchange, exhibits an annual cycle and is therefore the term which creates the cyclic trend in 3H/3t. The other DHB expressions fluctuate as pulses and serve to enhance or suppress the trend in surface heat flux. • The large pulses in 3H/3t are the result of large pulses of horizontal thermal advection and, to a small degree, are influenced by the pulses of H^^y. • During the years 1968-1970, there were five significant horizontal advective anomalies, each distinguished by an initial cooling trend in conjunction with a sharp salinity decrease, followed by an advective warming period and salinity increase throughout the entire 250m column. These large advective cycles coincided with the largest of the changes in mean monthly salinity. • The change in salinity is in response to translation of the Subtropic/Subarctic water mass boundary zone. Cold horizontal advection is a result of the intrusion of Subarctic water while advective warming is indicative of a return of the Subtropical water mass. 45 • Horizontal advection and H^jy do not necessarily fluctuate in phase. A period of upwelling (H^iv negative) serves to suppress the salinity change at 200m due to horizontal advection by mixing upward the higher salinities from the lower portion of the intruding Subarctic halocline. Conversely, down- welling (H^v positive) enhances the salinity decrease due to horizontal advection. • The pulse-like nature of these five events is suggested to be in response to a non-dispersive baroclinic Rossby disturbance whose wave-like characteristics are respon- sible for the depression of the southern edge of the Subtropic/Subarctic transition zone into OWS N. Intrusions of Subarctic water occur approximately every 7 to 8 months. The duration of each intrusion is about 3 to 4.5 months . • Other horizontal advective anomalies occurred in the period 1962-1967, but remain unsubstantiated by any changes in oceanic physical properties. In particular, a distinct lack of documented salinity data for this period prevented the examination of any thermohaline response to horizontal advection. However, these anomalies are, as in 1968-1970, non-seasonal in occur- rence and exhibit the same characteristic period and duration. 46 APPENDIX A HORIZONTAL ADVECTION EVENTS The details of each horizontal advection event are pre- sented in Figure 10. Each event is divided into a cooling trend (A) and a heating trend (B) . Each of these trends is associated with a salinity change throughout the column, values for which are incremented at the surface, 50m, 150m, and 200m and are listed in Table II. Note that cool advec tion is always associated with a reduction in salinity and that salinity increases during periods of advective warming The large change in salinity at 200m during Event 4 occurs because H -, . is out of phase with horizontal advection. div r 47 r ** * • < < L ui • O o —1 f". 1 2> oo o\ r-t J- o (4-1 CQ /-~\ J- !c > < • J- g i C o ■H 4-» U ed r-H CQ cd 00 c " o < N en • — » o C/> ♦J C (U > - OQ o CN ft < o _ CM *^1 a o> 2 OQ O i- =0 o o o o o o o o o 6 o ? o o o o o ■ '■0 AVQ H3d SA319NV1 48 Table II: Salinity changes (°/00) during horizontal advection anomalies Salinity Change Even t Surface 50m 150m 200m 1A -0.26 -0.19 -0.20 -0.18 2A 3A Coo •V, H h ling -0.14 -0.11 -0.18 -0.13 -0.11 -0.20 -0.15 -0.10 4A -0.14 -0.15 -0.10 -0.34 5A -0.28 -0.17 -0.12 -0.06 IB + 0.25 + 0.19 + 0.14 + 0.23 2B 3B -\ •VhH 0.24 0.26 0.13 0.20 0.05 0.30 0.01 0.22 4B heating 0.30 0.27 0.11 0.08 5B 0.10 0.12 0.10 0.05 49 BIBLIOGRAPHY Ballis, D. J., 1973. Monthly mean bathythermograph data from Ocean Weather Station NOVEMBER, Scripps Institution of Oceanography, La Jolla, SIO Ref. Ser. 73-7, p. 121. Bathen, K. H. , 1971. Heat storage and advection in the North Pacific Ocean, Journal Geophysical Research, v. 76: 676-687. Bernstein, R. L. and White, W. B., 1974. Time and length scales of baroclinic eddies in the Central North Pacific Ocean, Journal of Physical Oceanography, v. 4: 613-624. Clark, N. E. , 1967. Report on an investigation of large scale heat transfer processes and fluctuations of sea- surface temperature in the North Pacific Ocean, Ph.D. disser. , M.I.T., Cambridge, p. 148. Emery, W. J., 1976. The role of vertical motion in the heat budget of the upper Northeastern Pacific Ocean, Journal of Physical Oceanography, v. 6: 299-305. Gill, A. E., 1975. Evidence for mid-ocean eddies in weather ship records, Deep-Sea Research, v. 22: 647-652. Hansen, D. E., 1973. A study of surface, 50 meter, and 200 meter temperature and salinity fluctuations at Ocean Weather Station NOVEMBER, 1968-1970, Master's Thesis, Naval Postgraduate School, Monterey, p. 75. Rabe, K. and Bourke, R. H. , 1975. A statistical and spectral analysis of the air-sea interactions at OWS NOVEMBER, ENVPREDRSCHFAC Tech. paper 9 - 75, Monterey, p. 89. Roden, G. I., 1970. Aspects of the Mid-Pacific transition zone, Journal Geophysical Research, v. 75 (6): 1097-1109. Roden, G. I., 1971. Aspects of the transition zone in the Northeastern Pacific, Journal Geophysical Research, v. 76 (15) : 3462-3475. Roden, G. I., 1976. On the structure and prediction of oceanic fronts, Naval Research Reviews, v. XXIX (3): 18-35. Seckel, G. R. and Beaudry, F. H., 1973. The radiation from sun and sky over the North Pacific Ocean, Trans., American Geophysical Union, v. 54 (11): 1114. 50 Tabata, S., 1961. Temporal changes of salinity, temperature, and dissolved oxygen content of the water at Station "P" in the Northeast Pacific Ocean, and some of their determining factors, Canada Fisheries Research Board Journal, v. 18 (6): 1073-1124. Thorne, L. M. , 1974. The effects of heat exchange and thermal advection on the rate of change of temperature at Ocean Weather Station NOVEMBER, Master's Thesis, Naval Post- graduate School, Monterey, p. 90. Tully, J. P., 1963. Oceanographic domains and assessment of structure in the North Pacific Ocean, Fisheries Research Board of Canada, Pacific Oceanographic Group, file N 6- 13 (4) : p. 16. 51 INITIAL DISTRIBUTION LIST No. Copies 1. Department of Oceanography, Code 68 3 Naval Postgraduate School Monterey, California 93940 2. Oceanographer of the Navy 1 Hoffman Building No. 2 200 Stovall Street Alexandria, Virginia 22332 3. Office of Naval Research 1 Code 480 Arlington, Virginia 22217 4. Dr. Robert E. Stevenson 1 Scientific Liaison Office, ONR Scripps Institution of Oceanography La Jolla, California 92037 5. Library, Code 3330 1 Naval Oceanographic Office Washington, D.C. 20373 6. SIO Library 1 University of California, San Diego P.O. Box 2367 La Jolla, California 92037 7. Department of Oceanography Library 1 University of Washington Seattle, Washington 98105 8. Department of Oceanography Library 1 Oregon State University Corvallis, Oregon 97331 9. Commanding Officer 4 Fleet Numerical Weather Central Monterey, California 93940 10. Commanding Officer 1 Environmental Prediction Research Facility Monterey, California 93940 11. Director, Naval Oceanography and Meteorology Bay St. Louis, Mississippi 39520 52 12. Department of the Navy Commander, Oceanographic System Pacific Box 1390 FPO San Francisco 96610 13. Defense Documentation Center Cameron Station Alexandria, Virginia 22314 14. Library (Code 0212) Naval Postgraduate School Monterey, California 93940 15. Lieutenant Larry M. Thorne, USN Box 11 Dutton, Montana 59433 16. Dr. R. H. Bourke Department of Oceanography Naval Postgraduate School Monterey, California 93940 17. Dr. G. H. Jung Department of Oceanography Naval Postgraduate School Monterey, California 93940 18. Mr. Kevin Rabe Environmental Prediction Research Facility Bldg. 14 Naval Postgraduate School Monterey, California 93940 19. Lieutenant Donnel E. Hansen, USN USS FULTON (ASH) FPO New York 09501 20. Dr. C. E. Dorman Department of Geology San Diego State University San Diego, California 92115 21. Dr. L. Magaard Hawaii Institute of Geophysics and Department of Oceanography University of Hawaii Honolulu, Hawaii 96822 22. Mr. G. R. Seckel National Marine Fisheries Service Monterey, California 93940 53 23. Dr. D. McLain National Marine Fisheries Service Monterey, California 93940 24. Lieutenant Commander N. Camp, USN Department of Oceanography Naval Postgraduate School Monterey, California 93940 25. Dr. S. Tabata Fisheries Research Board of Canada Pacific Oceanographic Group Nanaimo, British Columbia 26. Dr. K. Kenyon NORPAC Scripps Institution of Oceanography University of California La Jolla, California 92037 27. Dr. K. Wyrtki University of Hawaii Honolulu, Hawaii 96822 28. Dr. R. L. Bernstein Scripps Institution of Oceanography P.O. Box 1529 La Jolla, California 92037 29. Dr. W. B. White Scripps Institution of Oceanography University of California La Jolla, California 92037 30. Dr. W. J. Emery Department of Oceanography Texas A§M University College Station, Texas 31. Dr. A. E. Gill Department of Applied Mathematics and Theoretical Physics University of Cambridge Cambridge, England, United Kingdom 32. Dr. R. L. Haney Department of Meteorology Naval Postgraduate School Monterey, California 93940 54 33. Dr. M. Laurs National Marine Fisheries Service La Jolla, California 92037 34. LCDR J. F. Pfeiffer 77 Adams Place, Apt. 407 Quincy, Mass. 02170 35. Mr. Huseyin Yuce DZ YZB Deniz Harbokulu Heybe 1 iada/ I s t anbul Turkey 36. Professor J. Wickham Department of Oceanography Naval Postgraduate School Monterey, California 93940 55 I I I APR79 £$¥ 3"^> Thesis 1GG723 P46165 Pfeiffer c#l Evidence of sub- arctic water mass intrusions at Ocean Weather Station NOVEM- BER. ! I APR79 <2> V33 Thesis 166723 P4616S Pfeiffer c.l Evidence of sub- arctic water mass intrusions at Ocean Weather Station NOVEM- BER. thesP46165 Evidence of subarctic water mass jntrusi 3 2768 001 97857 0 DUDLEY KNOX LIBRARY