A STUDY OF SURFACE, 50 METER AND 200 METER TEMPERATURE AND SALINITY FLUCTUATIONS AT OCEAN WEATHER STATION NOVEMBER, 1968-1970 Donne 1 El don Hansen Library Naval Postgraduate School Monterey, California 93940 VAL POSTGRADUATE SCHOOL Monterey, California THESIS A STUDY OF SURFACE, .- 50 METER AND 200 METER TEMPERATURE AND SALINITY FLUCTUATIONS AT OCEAN WEATHER STATION NOVEMBER, 1968-1970 by Donnel Eldon Hansen Thesis Advisor R. H. Bourke June 1973 Approved {oi pubtic izZzjOAz; dli&uhutLon unlimited. 7155155 A Study of Surface, 50 Meter and 200 Meter Temperature and Salinity Fluctuations at Ocean Weather Station November, 1968-1970 by Donnel Eldon Hansen Lieutenant, United States Navy B.S., Oregon State University, 1966 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN OCEANOGRAPHY from the NAVAL POSTGRADUATE SCHOOL June 1973 - Library Naval Postgrad*. Monterey, Calif. ABSTRACT Long and short-term surface, 50 and 200 meter temperature and salinity fluctuations at Ocean Weather Station November during 1968 through 1970 were examined using several statistical techniques. One technique, a unique monthly bi-variate modal analysis of surface temp- erature and salinity, proved to be a valuable research tool in that the resulting monthly modal cells, when plotted on a T-S diagram, provided a simplified view of the annual T-S relationships. The average range of mean monthly temperatures at the surface was found to be 5. 6 C and 4.0 C at 50 meters. Salinity at both the surface and 50 meters exhibited a semi-annual periodicity. Range of temperature at 200 meters was 1.5 C. The results of a seasonal linear regression anal- ysis show that surface and 50 meter temperatures were correlated during periods of increasing or decreasing surface temperatures. Surface temperature and salinity were correlated only during the apparent advection of modified subarctic water into the region around OWS November. TABLE OF CONTENTS I. INTRODUCTION 9 A. REVIEW OF THE LITERATURE 13 1.- The Subtropic Region 13 2. Surface Circulation __ 16 3. Surface Transition Zone 21 4. Average Annual Heat Balance 22 5. Surface and Subsurface Temperature Anomalies 22 II. DATA ANALYSIS 25 A. THE DATA 25 B. DEPTH SELECTION 27 C. ANALYSIS TECHNIQUES __ 27 III. RESULTS 31 A. TEMPERATURE FLUCTUATIONS 33 1. Surface 33 2. 50 Meters 42 3. 200 Meters 52 B. SALINITY FLUCTUATIONS 53 1. Surface 53 2. 50 Meters 56 3. 200 Meters 57 C. CORRELATIONS 61 1. 50 Meter Temperatures 61 2. 200 Meter Temperatures 64 3 . Su rf a c e Sa linity 66 IV. CONCLUSIONS AND RECOMMENDATIONS 67 BIBLIOGRAPHY 70 INITIAL DISTRIBUTION LIST 72 DD FORM 1473 74 LIST OF FIGURES Figure Page 1 Oceanographic Climatic Regions of the North Pacific Ocean Showing the Location of Ocean Weather Station November at 30 North, 140° West (After Tully, 1964). 12 2 Schematic Representation of the Seasonal Growth and Decay of the Thermocline at OWS November (after Beland, 1971). 15 3 Winter Surface Circulation Patterns for the North Pacific Ocean (Department of Commerce, 1961). 1' 4 Summer Surface Circulation Patterns for the North Pacific Ocean (Department of Commerce, 1961). 18 5 Total Monthly Meridional Transport Between 135° and 145° West at 30° North during 1969 and 1970. Positive Northward, Negative Southward, (after Wickett and Thomson, 1970, 1971). 20 6 Seasonal Vertical Temperature Profiles at OWS November, (a) Late Winter (2_19 May, 1967), (b) Summer-to-Winter (15 Jun- 1 Jul, 1967), (c) Summer (3-24 Sep, 1967), and (d) Summer- to-Winter (26 Nov- 17 Dec, 1967). (After Husby, 1969). . 28 7 Monthly Temperature and Salinity Modal Points Resulting from the Bi-Variate Analysis of the Data, 1968 through 1970. 32 8 Monthly Means of the Observed Surface Temp- eratures, 1968 through 1970 and Robinson's (1971) Long-term Monthly Means. 34 9 Annual Variations in the Mixed Layer Depth at 30° North, 140° West (after Bathen, 1972). 37 Figure Page 10 Variations in Monthly Mean Stability as Indicated by the Difference Between ^50 and *"sfc' 41 11 Monthly Means of Observed Temperatures at 50 Meters, 1968 through 1970. 44 12 Surface, 50 and 200 Meter Average (Daily) Year Temperatures computed from the Data, 1968 through 1970. • 46 13 Monthly Means of Observed Temperatures at 200 Meters, 1968 through 1970. 51 14 Monthly Means of Observed Surface Salinity, 1968 through 1970. 54 15 Monthly Means of Observed 50 Meter Salinity, 1968 through 1970. 58 16 Monthly Means of Observed 200 Meter Salinity, 1968 through 1970. 60 LIST OF TABLES Table Page I Surface Temperature Anomalies at OWS November, Based on Robinson's (1971) 20-year Long-term Mean, 1968 through 1970. 39 II Monthly Standard Deviations of Surface Temperature Calculated for the Years 1968 through 1970, in Degrees Celcius. 43 III Month of Occurrence and Range Between Minimum and Maximum Temperature Means at the Surface and 50 Meters, 1968 through 1970. 48 IV Monthly Standard Deviations of 50 Meter Temperature Calculated for the Years 1968 through 1970, in Degrees Celcius. 50 V Month of Occurrence and Range Between Minimum and Maximum Salinity Means at the Surface and 50 Meters, 1968 through 1970. 59 VI Correlation Coefficients, by Season, Resulting from a Linear Regression Analysis of Surface Temperature, Surface Salinity, and 50 Meter Temperature Data, 1968 through 1970. 62 VII Seasonal Correlations Between Surface and 200 Meter Temperatures, 1 968 through 1970. 65 ACKNOWLEDGEMENTS The author wishes to express his sincerest appreciation to Dr. R. H. Bourke, for his patience, guidance and advice throughout this study. Appreciation is also due to Professor Noel E. J. Boston for his review of the thesis and for his constructive comments. To my wife, Sharon, I offer my special thanks for her encourage- ment and patience, without which this project would have been impossible to complete. 8 I. INTRODUCTION While temporal variations of sea surface temperatures have received much attention by researchers studying the interactions between the sjea and atmosphere, few such studies have been made of the simultaneous variations occurring below the surface. Even fewer studies have been made of the concurrent variations in salinity at both the surface and within the upper layers of the ocean. Yet both of these observed quantities are of great significance in that they characterize such oceanographic features as the thermo-haline structure and heat content of a water column. Sea surface temperature is one of the parameters used to cal- culate energy transfers across the air-sea interface [Wyrtki, 1965], Generally it has been the only synoptic oceanographic information available on the thermal conditions of the oceans. For this purpose, sea surface temperature variations have been studied intensely in the past. Sea surface temperature anomalies, the differences between observed and long-term mean temperatures, have been used as an important source of information on the thermal content of the ocean. However, more recent work has shown that subsurface temperature anomalies, which often extend below 100 meters, represent a sizable amount of heat which can affect the energy feedback system [Beland, 1971]. Surface and subsurface temperature variations are important to other oceanographic parameters as well. Changes in the thermal structure are often the dominant factor in the vertical stability of the water column. This structure is also of great consequence to the propagation of sound in the ocean. It is apparent that an understanding of the thermal response of the subsurface layers to fluctuations in sea surface temperature would be of immense value. Similarly, fluctuations in salinity are of interest since many oceanographic parameters are a function of this quantity. Again, vertical stability and sound propagation are examples of two para- meters that respond to variations in salinity, although to a lesser extent than temperature. There have been surprisingly few published reports resulting from the study of long-term temporal variations of salinity. This most likely arises from the fact that few long-term salinity observa- tions at any one station have been conducted due to the great expense of maintaining such a data collection program. However, this is not to deny that such studies are of great importance to the understanding of oceanographic processes within the upper layers of the oceans. In the past due to instrument limitations, the ocean has been considered to be slowly changing or quiescent. Scientists were inclined 10 to rely on monthly and annual means to describe the state of the ocean. In recent years small-scale fluctuations in synoptic conditions have been used to describe significant short-term changes of state in the ocean. The scale of short-term changes in the upper layers of the ocean correspond to the scale of the meteorological driving forces. Ignoring advection, sea surface temperature fluctuates on a day-to-day basis depending on the temperature of the overlying air mass, the amount of insolation, mixing due to wind waves and the amount of cloud-cover. The magnitude of these daily fluctuations of sea surface temperature may be as large as the fluctuations in monthly means. Large-scale changes in oceanic properties, such as year-to-year temperature anomalies, correspond to the scale of changes in the general circulation pattern. Such macro- scale changes are of con- siderable magnitude and often persist in both time and space. An examination of short-term (day-to-day) and long-term (yearly) responses of temperature and salinity in the upper layers of the north- eastern Pacific Ocean is made. Synoptic data collected by the United States Coast Guard at Ocean Weather Station (OWS) November, located o o at 30 North and 140 West (Figure 1) was used. Temporal variations of temperature and salinity observed at the surface, within the seasonal mixed layer and within the upper portion of the permanent thermocline were studied in an effort to gain a more complete understanding of the upper layers of the ocean and variations in sea surface temperature. 11 12 Co-incidentally long-term temporal variations of surface and subsurface salinity were studied since there is an apparent gap in the literature with respect to this subject. Indications of the response of the upper layers of the ocean to changes in heat input will be examined in two ways. First, the correlations existing between temperature at the surface and at selected depths of interest will be calculated. These correlations may be used as an indication of the relation between sea surface temperature and the temperature at a particular depth of interest. Second, the time lag between surface temperature maximum and minimum and those occurring at depth will be determined. Such a study will yield an indication of the response time involved. No attempt will be made to examine the synoptic atmospheric mechanisms responsible for the fluctuations of sea surface temperature. The primary concern of this thesis will be the relationship between the mixed layer and the upper portion of the main thermocline to observed sea surface temperature variations. A. REVIEW OF THE LITERATURE 1. The Subtropic Region OWS November lies within the subtropic region as defined by Tully [1964]. This region is characterized by (1) an excess of evaporation over precipitation throughout the year, (2) convective 13 mixing and (3) seasonal heating and cooling. It is the largest of the oceanographic regions and includes the North Pacific Gyre. The excess of evaporation over precipitation results from the maintenance of high temperature levels at the sea surface through- out the year. Tully [1964] estimated an annual mean difference between evaporation and precipitation of about 80 centimeters per year o o at 30 North and 140 West. Removal of freshwater from the sea surface by evaporation results in evaporation-driven convective mixing. The denser surface water sinks until it reaches an equilibrium level, forming a mixed layer deeper than that formed by simple wind mixing. Seasonal heating occurs from spring to the summer equinox when daytime heating exceeds nighttime cooling. Heat accumulates at the surface until September when temperatures are at their maximum. Cooling of the sea surface dominates during the rest of the year, with the coldest temperatures occurring in March. The seasonal growth and decay of the temperature structure of the upper layers of the subtropic region are schematically illustrated in Figure 2. The upper layer is isothermal to about 150 meters at the end of the cooling season. Warming of the near surface waters results in the formation of a seasonal thermocline between 40 and 60 meters. Continued warming results in an increase of the thermocline gradient while its depth remains nearly constant. Cooling of the surface waters 14 o 18C 19 20 21 22 23 18 C 19 20 21 22 23 -100. Figure 2. Schematic Representation of the Seasonal Growth and Decay of the Thermocline at OWS November (after Beland, 1971). 15 results in accelerated convective mixing which erodes the seasonal thermocline until isothermal conditions extending to the depth of the main thermocline are re-established [Tully, 1964], 2. Surface Circulation The Pacific Gyre is the dominating feature of the surface circulation in the mid-latitude region of the eastern North Pacific Ocean. This gyre is wind driven and responds to changes in the general atmospheric circulation pattern. Generally, during summer the axis of the gyre is displaced more to the north than during winter. A chart showing the mean surface currents in the North Pacific during winter and summer is reproduced in Figures 3 and 4 [Department of Commerce, 1961]. From these figures it can be seen that the mean current speed ranges from 0. 1 to 0. 3 knots throughout the year. The apparent intensification of the North Pacific Gyre in summer centered to the northwest of station November has several important influences on the oceanography at this location. A hypo- thesis is developed in a later part of this thesis that occasionally large volumes of cold subarctic water are advected into the area around OWS November. The result is significant decreases in temp- erature and salinity in the upper layers of the ocean. Also, there is evidence that this circulation pattern was altered during the early months of each of the years 1968 through 1970. 16 *£_\i 1 1" ,4 \ 1 H I" I j'lM^i-L Uil# 1= IP'.I /J I IV^^VhlV ^fr «'fh!' z _ o z O 0 o u o a 1 5 |S ! * ^ z 5 * 55 !P ?lU.iT^\-^fr^,^\ v ufn" MfeAO : i : J • J W A ^± — I • i-i o n) Oh o W i— I d § 1 .H O *^ 2 ° U g w Q 0) w £ O 0) 00 h 17 t&br '.'j; iitpw TrV1* Ti^ni &1S0 iU.K v/V, i pr Q z a <3 o s « 2 3 5=2 a II |i I!13 *&5Ss\g si li I Mi ^ ^ u J-l +-( U w u-< c O Vh A — s •j-> ■— i £ ^ g 2 -H to «*< 0> *h d b£ h 18 The low speed of the currents associated with the Pacific Gyre in the area of OWS November allows the surface water to adjust to climatic conditions [Tully, 1964]. However, to some degree, the waters retain their properties. The persistence of anomalous sea surface temperatures in the North Pacific has been studied intensively by Namias [1970]. From his studies he has made inferences regarding the transport of surface water between the Subarctic and Pacific Gyres. He found that anomalous water generated in the Subarctic Gyre frequently enters the subtropics and persists for indefinite periods of time. He based this conclusion on seasonal sea surface temperature lag correlation studies. The magnitude of the indicated transport (0.2 to 0.3 knots) agrees well with the mean surface current charts (Figures 3 and 4). Further, he has determined that anomalously cold or warm water generated in the north in any season usually does not show up in the southern area until the fall of the same year. Mean monthly transport computations for the North Pacific Ocean at 100 meters, based on mean monthly values of sea level pressure, are routinely computed and published annually by the Fisheries Research Board of Canada in their Technical Report series. The computed meridional component of total mass transport between longitudes 135 to 145 at 30 North for 1969 and 1970 is shown in Figure 5 [Wickett and Thomson, 1970, 1971]. Before discussing these 19 o lo ° ^ ^) t- CU r-i O -4-» I— 1 ^ V T3 D o * * T3 £ c^ h £ -0 a PQ cr^ £ i— » A . -4-> o I— ( CO 5-1 in o c d cu i— i rt T3 > hi ■i-i -H ^ o «j s V i— i c Z z .2o d T3 O CU ■'-, J-i o co £ a z ,d 0 .d thly Wes ■4-> U o H d Gr, CtJ o° cu 6 5 > •r-l t— 1 4-> ^ 0j ITJ CO 0 4J r< ° M 0 £ in cu ^ 3 b0 - -I h o o o o o 1 1 1 CO 1 o 1 (Z01X) jajauioiDj/pucoos/suox 20 transport figures, it must be pointed out that the monthly calculations are carried out independently of each other. The monthly transports can be interpreted as the transport that would result if the pressure distribution for that month were to persist indefinitely. No account of inertia of the ocean is taken in going from one monthly mean to another. Therefore, the computed transport values must be inter- preted as indications of tendencies of the ocean circulation and are necessarily conservative in magnitude. It can be seen that, in general, considerable north to south mass transport occurs at various times of each year. Apparently during 1969 and 1970 most transport occurred between April and October in 1969 and between April and June during 1970. The com- paratively large amount of transport towards the south noted in 1969 is noteworthy and will be discussed in more detail in a following portion of this thesis. 3. Surface Transition Zone One of the most prominent features of the northeastern Pacific is the transition zone between the subarctic and subtropic water masses. This zone occupies a belt between 32 and 42 North in the western part of the Pacific and turns southward as it approaches the North American continent. To the north is found the subarctic water mass distinguished by its lower surface temperatures and salinities. Southward can be found water of subtropical origin, which 21 is higher in temperature and salinity. Between these two water masses lies the subarctic-subtropical transition zone marked by strong hor- izontal gradients [Roden, 1970]. The southern boundary of the transition zone was found to o o exist between 30 and 31 North, about the same latitude as OWS November. The principal feature of this boundary was a rather sharp transition toward higher salinities and higher temperatures charac- teristic of the subtropical water mass found to the south. Salinities o in excess of 35 /oo characterize the subtropic water mass in this region. Seasonal and year-to-year variations in the position of the southern boundary have been noted in the literature but have not been documented in detail. 4. Average Annual Heat Balance Wyrtki [1965] examined the average annual heat balance of the North Pacific Ocean and its relation to circulation. Interpolating from Wyrtki's figures, it is apparent that the ocean gains a relatively small amount of heat during an average year in the vicinity of OWS November. In this area the upper layers of the sea absorb an annual average of 25 calories per square centimeter per day. 5. Surface and Subsurface Temperature Anomalies Sea surface temperature and its anomalies have received considerable emphasis to date in both oceanographic and weather fore- casting. However, only recently have subsurface temperature 22 anomalies received attention. Beland [1971] has indicated the con- siderable importance of large amounts of anomalously warm or cold subsurface waters on the flow of thermal energy between the sea and atmosphere. The presence of anomalously warm water may act as a source of thermal energy for the atmosphere. A negative anomaly may indicate the presence of a thermal energy sink for the atmosphere. An observed anomalous warming of the northeastern Pacific from May to June, 1968, was found by Namias [1971] to be due largely to increased insolation and horizontal convergence of the surface layers of the sea and associated downwelling, with some contribution from reduced latent and sensible heat losses. The warming was associated with the development and maintenance of a Pacific anti- cyclone in June. This anticyclone was persistently regenerated by o an unusually strong mean jet around 40 North. The Namias study of the anomaly at Ocean Station November indicated that the warm pool of water extended to a depth of 122 meters. Bulk computations of heating, ignoring advection and mixing, accounted for one-third of the heating while the rest apparently was due to down- ward advection of heated water. The warm pool was demolished during the fall and early part of 1969. Cyclonic activity was greatly amplified during this winter peiod as cold air masses, fronts and cyclones siphoned off the excess heat and moisture of the warm pool. 23 Clark [1972] has concluded that both advective and non- advective processes are important in determining the nature of temperature anomalies in the surface layers of the ocean. Further- more, he states it is apparent that advective heat transfer is a more significant process in winter and spring, while non-advective proc- esses have a more significant effect during summer and fall. Beland [1971] found that about 50 percent of the observed surface anomalies extended to at least 100 meters. Of these, most o were negative anomalies with magnitudes greater than 1.0 C. Negative anomalies were most common during March and April. He hypothesized that the greater occurrences of negative anomalies were due to the distribution of heat throughout the upper layer by convective mixing. Positive anomalies were more common during late winter or early spring (December through February) and generally penetrated only to 30 meters or less. The magnitude of the positive anomalies were generally less than 1.0 C. Beland went on to suggest that the seasonal thermocline inhibited the conduction of heat downward result- ing in relatively shallow positive temperature anomalies. 24 II. DATA ANALYSIS A. THE DATA Data used in this study were collected by the United States Coast o Guard at Ocean Weather Station (OWS) November located at 30 North, 140 West in the northeastern Pacific Ocean. The original ocean- ographic sampling program started in 1966 consisted of daily Nansen bottle casts to 1500 meters on alternate 21_day patrols. Since 7 January 1968, however, the program has been expanded to daily hydrocasts on every patrol. Examination of the data revealed that all data prior to 7 January 1968 contained significant time segments of missing data, some as long as several months. Therefore, the time series for this study was limited by the available data to the 3 year period from 7 January 1968 to 6 March 1971. The data record from 7 January 1968 to 6 March 1971 contained several data discontinuities. The mean length of the missing data segments was 5.6 days. The longest break in the data was 25 days and the shortest was 2 days. All of the discontinuities were evenly distributed throughout the entire record length. Discontinuities were eliminated by interpolation. The data at each depth of interest was hand plotted and a best fit curve drawn. 25 Missing data were interpolated from the curve and inserted into the original data record. All off-station, duplicate, and nonrepresentative data were replaced by interpolated values. There were a total of 1, 155 data points in the time series used in this study, 266 of these were interpolated (22%). This is not con- sidered an excessive amount since the mean interval was relatively small and the intervals were well distributed throughout the data record. Inductive salinometers and deep-sea reversing thermometers were used to measure salinities and temperatures. Duplicate water samples were also obtained for comparison with the surface and 1500 meter Nansen casts for quality control. It was found that on the average 89 percent of the salinity samples differed from the ship's values by less than 0.01 /oo [Husby, 1969]. Accuracy of the temp- erature measurements was not discussed in the literature. An o accuracy of plus or minus 0.01 C is assumed. The hydrocasts were normally conducted at 1100 hours local time. When two or more casts were made on the same day, the one conducted nearest to 1100 hours was retained and the remainder eliminated from the data record. This effectively eliminated diurnal effects. 26 B. DEPTH SELECTION Temperature and salinity values measured at 50 and 200 meters were chosen for comparison with surface values. The decision to use these levels was based on an examination of typical OWS November vertical temperature profiles [Husby, 1968], previously mentioned schematic temperature profiles presented by Beland [1971] and solar radiation penetration data presented by Sverdrup, et al [1942]. The goal in selecting these depths was to compare and contrast fluctuations of temperature and salinity at the surface, within the mixed layer and within the main thermocline. Station November vertical temperature profiles after Husby [1969] are shown in Figure 6. From these profiles it can be seen that 50 meters lies within the mixed layer approximately 6 months out of a year. The main thermocline is found to lie permanently above 200 meters. C. ANALYSIS TECHNIQUES Several analysis techniques were used during the preparation of this thesis. These techniques included calculation of annual and monthly means, monthly standard deviations, bi-variate analysis and a seasonal regression analysis. Monthly means, of temperature and salinity were calculated for the years 1968 through 1971. The means were then plotted against 27 Celsius 2 4 6 8 10 12 14 16 18 20 22 24 26 2000 Figure 6. Seasonal Vertical Temperature Profiles at OWS November, (a) Late Winter (2_19 May, 1967), (b) Summer-to-Winter (15 Jun- 1 Jul, 1967), (c) Summer (3-24 Sep, 1967), and (d) Summer- to-Winter (26 Nov- 17 Dec, 1967). (after Husby, 1969). 28 time in order that the behavior of each property with time throughout the year could be analyzed. These plots also made it possible to compare the behavior between temperature and salinity throughout the three year period. Standard deviations were computed in order to measure the variation of each property at each depth of interest. The unbiased standard deviation was used throughout this study. The standard deviations calculated were also used as a measure of the relative variations of temperature and salinity. An average year was calculated by computing the daily mean of temperature and salinity for the 3 year record, 1968-1971. A bi-variate analysis was performed to establish the seasonal characteristics of temperature and salinity over the three year period. The simultaneous modal analysis of both temperature and salinity simplified the relation between the variables so that the variation of each parameter could be readily visualized. A brief description of the technique used to perform this analysis follows. The modes of temperature and salinity for each month of the average year were determined by summing the daily occurrence of categorized pairs of salinity and temperature. The increments used o to establish the categories of temperature and salinity were 0.5 C and 0. 10 /oo, respectively. The frequency of occurrence of each T-S cell was converted to a percentage of the total and plotted as a 29 scalar field on the T_S diagram. Isopleths were drawn at 2, 4 and 6 percent occurrences. The resulting monthly modes of each pair were also indicated. A multiple linear regression analysis was performed for each season using the statistics library program, REGRE. The procedure used was to divide the three year and two month data record into seasons and perform the regression analysis on each season. Sea surface temperature was held as the dependent variable and all other variables at the surface and at the other depths were held as indepen- dent variables. A mean seasonal correlation coefficient was computed from the results. 30 III. RESULTS The discussion of the results is based on the recurring sea surface temperature cycle suggested by the bi-variate analysis of surface temperatures from 1968 through 1970. The results of the bi-variate, or combined modal analysis of sea surface temperature and salinity is presented in Figure 7. From this figure it is possible to readily discern four phases of the yearly temperature cycle. The first of these phases is the period of time when surface temperatures are the lowest. This condition persists through the months of March and April when the mode of the surface temperature o is about 17. 8 C. This phase of the temperature cycle will be referred to as the winter period. During the months May through July the surface temperature increased markedly. The total change in modal temperature during o this transition is about 5.0 C. This period of surface warming will be referred to as the winter -to -summer transition period. The summer period, characterized by high surface temperatures o of about 22. 8 C exists from August through October. During this period the monthly temperature mode is unchanged. Following the summer period surface temperatures decrease. This cooling period will be referred to as the summer-to-winter transition period. This period was observed to occur from November through February. 31 23 22 21 20 Z> V) a> i CU bfi a O H h! • r4 (L) X> UJ O o (11 rj J3 rt m o O t^- o • UJ i—i en C C cj X3 rt > t— ( Ai •— 1 X 4-> X *J 4-> c uu c o o CO ojO •I-l CM fN CM CM o CM oo -jo*»jnjDJ»duj*| 34 Dietrich [1957] and Sverdrup, et al, [1942] have attributed this annual fluctuation of temperature primarily to the annual periodicity of in- coming solar radiation. The calculated monthly means of observed surface temp- erature for the period 1968-1970 are plotted in Figure 8. Also shown in this figure are the 20-year long-term monthly mean values compiled by Robinson [1971]. From this figure it is possible to visualize the dominate features of the annual fluctuations in sea surface temp- erature and their deviations from the long-term mean during the period of interest. The most pronounced feature observed is the marked annual cycle of the sea surface temperature. Each year is charac- terized by a distinct minimum and maximum in monthly mean temperature. Surface temperatures exhibited an average annual range of o 5.3 C during this period. The greatest range in temperature, about o 6.0 C, occurred in 1968. The shape of each monthly mean surface temperature curve is typical for the latitude under consideration as described by Isaacs [1969] except for short-term perturbations. He has indicated that the cycloidal character of the temperature curve is. due to the differences in anisotropic heat flow resulting from fluctuating mixed layer depths. Annual variations in the depth of the mixed layer, 35 after Bathen [1972], at OWS November is shown in Figure 9. The resulting differences between the processes of shallow heating, i. e. , the production of a shallow stable layer during the winter-to-summer transition, and deep cooling (with associated instability and mixing) during the late summer-to-winter transition and the winter periods is believed to account for the shape of the temperature curves at the surface. Comparison of the calculated monthly means with Robinson's [1971] long-term mean reveals several apparent long-term and short- term deviations during the period of interest. The most obvious deviation is the consistently large difference between the long-term mean during the winter months of March and April. All three years o were significantly (about 1 C) cooler during these months than the long-term mean. Beland [1971] noticed this recurring difference but did not elaborate on it. It is interesting to hypothesize as to the cause of this recurring difference. There are four possible explanations: 1. ) the advection of cold water into this region each year suggesting a recent change in the winter surface circulation pattern in the subtropic zone, 2. ) a greater than normal loss of heat to the atmosphere in winter, 3. ) recurring seasonal reductions in the amount of incoming solar radiation due to increased cloud cover, or, and most unlikely, 4. ) an error in the long-term mean. 36 a •c 4-> a, r\i > c id Jh S 4-> <+H 4-J c (0 • r-4 0) (0 £ CI o 0 ■ 1-1 o -t-> Tf rt t— 1 • r-l S-i rt X > 4-> *H f— i o z o o *A CO 0) U 4-> !h CD Oh s cu H£ cu •— i > u V CO O O 0 oo cu >s 0) < ^ 0) c o o .1-1 ci ^ CD a, a CD H J-l rt D >h , o >s r- »— 1 o • H •—i nJ Q A CO) d (U n CJJ u rt j3 >n -UJ CI) > CO <; u ■—i v +j * C) rt s 4-> Q o o cu ro J3 -u> TJ a C nJ o O in T3 * CI) 0) -t-> CJ 3 0) O, a 2 o W o CM i—i CD U ex o CO CN -j0'»jn|DJ»duj»| 46 Temperatures warmed again after September reaching a o maximum of 21.7 C in October, followed by decreasing temperatures until the following year. The range of temperature during 1969 was approximately 3.8 C. As at the surface mean monthly temperatures observed in 1970 were significantly lower than those of the preceding two years. o The minimum mean temperature of about 17. 0 C was observed in March. Temperatures then increased until October when the max- imum of only 20. 6 C was reached. The range in temperatures was o approximately 3.6 C during this year. Comparing Figures 8 and 11 and examining Figure 12 it can be seen that the winter minimum at 50 meters occurs simultan- eously with that at the surface. Referring to Figure 9 which relates the yearly fluctuations in the depth of the mixed layer as compiled by Bathen [1972], it can be seen that during this period the water column is isothermal well below 50 meters. Therefore, one would expect the simultaneous occurrence of the temperature minimum. Due to the thermal stratification of the water column resulting from the formation of the seasonal thermocline, the ex- change of heat from the surface to 50 meters is limited to the process of thermal conductivity. However, in the absence of a seasonal thermocline, heat may be exchanged by a combination of convective mixing, mechanical mixing and thermal conductivity. A net result 47 TABLE III. Month of Occurrence and Range Between Mean Monthly Minimum and Mean Monthly Maximum Temperature at the Surface and 50 Meters, 1968 through 1970. ** 1968 1969 1970 Sur face 50 Meter Lag direction Month Range Month Range and magnitude Min APR 6.0 MAR/APR 4. 0 0 Max AUG 6. 1 NOV 4.2 +3 Min APR 4.4 APR 3.8 0 Max OCT 5.4 OCT 4.6 0 Min MAR 6.0 MAR 3.6 0 Max SEP OCT + 1 In column 7 "+" means surface leads 50 meters, magnitude in months . 48 of thermal stratification is a restriction in the flow of heat from the surface to the 50 meter depth. This restriction in the flow of heat is indicated by the delay in the occurrence of the maximum monthly mean temperature at 50 meters relative to the surface. In 1968 the surface maximum occurred in August and the 50 meter maximum was delayed until November, a difference of two months. In 1969 the surface and 50 meter maximum temperatures occurred simultaneously in October, a possible reflection of the abnormal mid-summer cooling discussed earlier. During 1970 a one month delay between maximum tem- peratures (September and October) was observed. It is apparent that thermal stratification, unless reduced by abnormal cooling, may result in a one to two month delay in reaching the maximum tem- perature at 50 meters. Differences between the annual range of temperatures and the occurrence of temperature extremes (maximum and minimum) at the surface and 50 meters are presented in Table 3. From this table it can be readily seen that all winter mean temperature minima occurred (with the exception of 1969) one to three months earlier at the surface, and the range of mean temperatures decreases by o approximately 1.3 C at 50 meters compared with the temperature range at the surface. 49 TABLE IV. Monthly Standard Deviations of 50 Meter Temperature Calculated for the Years 1968 through 1970, in Degrees Celsius. 1968 1969 1970 JAN 0.05 0.09 0.12 FEB 0.13 0.07 0.19 MAR 0.01 0.00 0.05 APR 0.05 0.07 0.18 MAY 0.11 0.15 0.12 JUN 0.11 0.20 0.06 JUL 0.03 0.01 0.05 AUG 0.10 0.18 0.05 SEP 0.24 0.05 0.10 OCT 0.07 0.15 0.03 NOV 0.07 0.11 0.06 DEC 0.08 0.19 0.18 50 • a in n 00 o o xo so h- o o^ o^ i I + rt CO U 0) a, u> ■■ a H > u 0) CO O o r- o^ ■—I OX) o -4-> 00 NO ■— « CI o o o ro ** * »dlU9| bX) 51 Monthly standard deviations of temperature at 50 meters were generally smaller than those at the surface (Table 4). During 1968 and 1969 the larger standard deviations occurred during the months of the winter-to-summer transitional period (as they did at the surface). In 1970 several larger standard deviations occurred throughout the year indicating more variations occurring this year than during the previous two years. 3. 200 Meters Temperature variations at 200 meters are, in contrast to those in the upper layers, much reduced in range and show little annual periodicity. Mean monthly temperatures are shown in Figure 13. A power spectrum analysis, employing a fast Fourier transform of questionable reliability due to the presence of large low frequency components and a relatively short record length for such an analysis, indicates that most of the energy is centered in the component with a period of approximately 1. 5 years. Temperatures appear to center around a mean of 13. 3 C. The generalized behavior of temperature at 200 meters is perhaps best shown using the average (daily) year plot (Figure 12). This figure shows a slight seasonal cooling trend in temperature during January and February, followed by a slow increase through June. Then another cooling period follows and lasts through August after which temperatures very slowly increase until December or early January. 52 Husby [1969] indicated that slightly higher temperatures at 200 meters noted during 196 8 resulted from convective mixing that occurs during this period of time at the surface due to the erosion of the seasonal thermocline and continued surface cooling. This trend was also noted in the plot of the computed average (daily) year. If this were the general case, both increases in temperature and high negative correlations with surface temperature during this phase of the temperature cycle would be noted to occur annually. The correla- tion between the surface and 200 meters during January and February of 1968 were found to be relatively high (0. 67) and positive. However, as shall be discussed in a later section, low correlations were cal- culated during the other three cooling and winter phases of annual temperature cycles. This indicates that generally there is no correla- tion between the surface temperatures and temperatures at these depths during periods of convective mixing. B. SALINITY FLUCTUATIONS .1. Surface As mentioned earlier, the lack of long-term mean values precluded the determination of salinity anomalies. However, an examination of mean monthly salinities computed over the period 1968 through 1970 reveals several interesting features in the annual fluctuation of surface salinity. Monthly surface salinity means are presented in Figure 14. 53 • r-l c .1-1 ■—I uy 9) o U in CD > U v CO o o o 2 •^ O ■g co o ^£ ■—I 9) U 0 CN ^" O o 00 r~ o Ift ui IT) Tf ^r *r Ti- ro ro m m o ro ro oo /O ' A J J u I | D 5 54 It is apparent that there is a semi-annual periodicity in the salinity cycle at this station. A salinity minimum occurred during either March (1970) or April (1968 and 1969). In 1968 this o mean minimum was about 34. 9 /oo while in 1968 the minimum was about 34. 8 /oo. The following year, 1970, apparently witnessed a significant reduction in the surface salinity with the mean being only 34.5 /oo. This significant decrease in the mean salinity noted in 1970 may be related to the presence of an abnormally cold water mass in this area. It appears that a large mass of cool, fresh subarctic water from the north was advected into this region (Figure 5) during April and May of 1970. Due to the inherent errors of lag time in the calculations introduced by the assumption to ignore the inertia of the surface transport system, it is feasible that this persistent and sig- nificant amount of transport might well account for the depressed mean temperatures and salinities noted in 1970. The occurrence of such periodic advections of norther water amounts to a north- south displacement of the transition zone between subarctic and subtropical water masses [Roden, 1970]. During 1968 and 1969 following the first minimum the salinity continued to increase until a maximum occurred in July. For both years the range of salinity between these extrema was a o little over 0. 3 /oo. The maximum in 1970 occurred in May two 55 months earlier than in the previous years, the range in salinity being o 0.4 /oo. Following these maxima, salinities decreased rapidly in both 1968 and 1969 until September when the second minimum occurred. The decrease in salinity in 1969 was considerably greater than in 1968 o (greater than 0. 2 /oo). The second minimum in 1970 occurred in July, again two months earlier than in 1968 or 1969- After this second minimum salinity increased until October in 1968 (with an increase in salinity of 0.3 /oo over the minimum) and December in 1969 (with o an increase of 0.4 /oo). In 1970 the second maximum was reached in September and had a range of approximately 0. 5 /oo. Afterward salinity decreased until a third minimum occurred in November. A range in the monthly mean values of approximately 0. 4 /oo was observed. It is interesting to note that the successive minimum and maximum values increased at each occurrence, thereby indicating an overall increase in salinity during the entire year of of 1970. This is an apparent indication of the re- establishment of the normal salinity condition superimposed over fluctuations caused by forces resulting in the occurrence of periodic maximum and minimum. 2. 50 Meters The fluctuations in salinity at 50 meters appears to be very similar to that at the surface. The monthly mean values during 56 the period of interest are shown in Figure 15. There are some dif- ferences, however. These differences are most readily evident from Table 5 which summarizes the range in salinity between maximum and minimum extremes and the time of occurrence of these extremes at both the surface and 50 meters. The range in salinity is generally less at 50 meters (by about 0. 11 /oo) than at the surface. This is also the case with temperatures as previously indicated. From the data presented in Table 5 it is nearly impossible to draw conclusions regarding possible time lags between the occurrence of salinity extremes at the surface and 50 meters. It seems that surface salinity led 50 meter salinities by one month three times, lagged by one month five times and there were no time differences five times. However, note that the occurrence or non-occurrence of time leads or lags appear to be grouped accord- ing to direction, indicating a possible common connecting relationship. 3. 200 Meters Salinity variations at 200 meters differ markedly from those in the surface layers in both range and periodicity. Figure 16 is a plot of the monthly means of the observed salinity values measured at 200 meters, 1968 through 1970. The overall mean of all observed values is 34.2 /oo. The range of salinity during the period of interest was small. The mean minimum was about 34. 03 °/oo while the maximum 57 CM T- O o 00 t^ o IT) in m ^r •* *■ *r ro ro ro ro ro ro ro in o m > u V CO ^> O o CO d 2 >» o r- 1—1 o U o *° in u oo/o ' A | I u I | o 5 58 TABLE V. Month of Occurrence and Range Between Mean Monthly and Mean Monthly Maximum Salinity Means at the Surface and 50 Meters, 1968 through 1970** Sui ■face 50- meters Lag direction Year Min Month Range Month Range and Magnitude 1968 APR MAR -1 0.34 0. 32 Max JUL 0.27 JUN 0.24 -1 Min SEP 0.24 SEP 0.27 0 Max OCT 0.25 NOV 0.34 + 1 1969 Min APR 0.38 APR 0. 31 0 Max JUL 0.56 JUN 0.26 -1 Min SEP 0.44 AUG 0.21 -1 Max DEC 0.54 NOV 0.47 -1 1970 Min MAR 0.46 MAR 0.26 0 Max MAY 0.35 MAY 0.21 0 Min JUL 0.48 AUG 0.26 +1 Max SEP 0.33 OCT 0. 17 +1 Min NOV NOV 0 ** In column 7 " + " means surface leads 50 meters, "-" means surface lags 50 meters, magnitude in months. 59 in ^r ro CN T- O OO/O 'A | I U I | D £ 60 was 34.40 /oo, the difference between the extreme means being o approximately 0. 4 /oo. Results of a power spectrum analysis, again using a fast Fourier transform, indicates most of the energy is contained in that component with a period of 1. 5 years, as was the temperature fluctuations at this depth. This does not necessarily imply that the fluctuation of these two parameters are caused by the same forcing processes. An overall correlation coefficient calculated between 200 meter temperature and salinity values was found to be only 0. 23. This indicates that there is no significant correlation between these parameters at 200 meters or that each fluctuates independently of the other. C. CORRELATIONS Correlation coefficients from the linear regression analysis are summarized in Table 6. Each coefficient was calculated with surface temperature as the dependent variable. Each year was separated into winter, winter-to-summer, summer and summer-to- winter segments based on surface temperatures and an analysis run on each segment. 1. 50 Meter Temperatures Temperatures observed at 50 meters are well correlated to surface temperatures during the transition periods. However, 61 TABLE VI. Correlation Coefficients, by Season, Resulting from a Linear Regression Analysis of Surface Temperature, Surface Salinity, and 50 Meter Temperature Data, 1968 through 1970. 1968 1969 1970 Mean Sfc 50 m Sfc 50 m Sfc 50 m Sfc 50 m Sal Temp Sal Temp Sal Temp Sal Temp Winter 0.61 0.58 0.61 0.56 0.77 0.42 0.66 0.52 Warming 0.59 0.90 0.77 0.70 -0.11 0.67 0.42 0.75 Summer -0.28 0.31 0.16 0.40 -0.05 -0.29 -0.05 0.14 Cooling 0.57 0.88 0.84 0.98 -0.01 0.96 0.47 0.94 62 they are poorly correlated during both the summer and winter phases of the annual temperature cycle. The lowest correlations occurred during summer periods. Referring to Figures 3 and 4 it can be seen that during the summer period, surface temperatures are nearly constant at the annual maxi- mum while temperatures at the 50 meter level are still increasing due to the time delay discussed in an earlier section of this thesis. The mean correlation for the winter periods was about 0. 52. This indicates no significant correlations occurred during the winter periods. It is hypothesized that this is due to the short- term response of surface temperature to atmospheric influences. These influences are not of sufficient length in time to be transmitted to 50 meters. The highest correlations occurred during the transition periods. Correlations were somewhat variable during the winter-to- summer transition, having an average of 0.75 over the three years. The warming period during 1968 had a high correlation coefficient of 0. 90 and the other two years had a coefficient of approximately 0.70. Higher correlations during the warming transition period are a result of the simultaneous increase in temperatures at both the surface and at 50 meters. From Figures 8 and 1 1 it can be seen that, initially, the two temperatures increase concurrently. However, 63 after a short period of time, the formation of the seasonal thermocline occurs resulting in the construction of an effective barrier to the flux of heat between the surface and lower depths. This would suggest a positive weak correlation between the two depths with respect to temperature. Significant correlations between surface and 50 meter temperatures were found to occur during the summer-to-winter temperature transition period. The mean correlation for the three years was 0. 94, with the highest correlation (0. 98) occurring in 1969. It is during this time period that the seasonal thermocline is eroded away and the barrier to the flux of heat between the surface and the 50 meter level is removed. Furthermore, cooling of the surface waters results in convective mixing resulting in almost simultaneous changes in temperature throughout the mixed layer. It is most likely this latter process that causes the correlation coefficient to be so high. 2. 200 Meter Temperatures There appears to be little correlation between surface and 200 meter temperatures except during the warming phase of the surface temperature cycle. Seasonal correlation coefficients are shown in Table 7. A mean correlation coefficient of 0. 63 was com- puted for the winter-to- summer warming transition. This coefficient is not indicative of a high correlation during this period of time, but 64 TABLE VII. Seasonal Correlations Between Surface and 200 Meter Temperatures, 1968 through 1970. 1968 1969 1970 Mean Winter -0.26 -0. 04 0. 38 0.03 Warming 0. 71 0. 63 0. 55 0. 63 Summer -0.26 0. 37 -0.24 -0. 04 Cooling -0.41 0.51 -0.27 -0.06 65 it has possible significance when compared to the extremely low coefficients computed for the other seasons. 3. Surface Salinity It is apparent that surface salinity in general is not well correlated with surface temperature. This is true during each of the four yearly phases of the annual temperature cycle. The highest correlation, averaged over the three years of interest, occurred during the winter. The lowest correlation occurred during the summer. During the summer-to-winter transition period of 1969/ 1970 a relatively high correlation of 0. 84 was computed. Interestingly, this is the period during which surface temperatures and salinities were decreasing, resulting in the abnormally low temperatures a.nd salinities observed during the whole of 1970. Apparently this high correlation is a result of the simultaneously changing temperatures and salinities brought about by the influx of subarctic water. 66 IV. CONCLUSIONS AND RECOMMENDATIONS This study was involved with the description of short- and long-term fluctuations of concurrent temperature and salinity observa. tions taken at the surface, within the mixed layer, and within the upper portion of the permanent thermocline at Ocean Weather Station November during the years 1%8 through 1970. The results of this research show that: 1. Surface temperatures had an annual periodicity with a o mean range of approximately 5.6 C. Comparison of the data with Robinson's long-term means reveals that there is great variability in year-to-year mean temperature curves. It was observed that anomalous decreases in temperature were often accompanied by decreases in salinity. 2. Temperature maximums at 50 meters were found to lag surface temperatures by one to three months. This lag is probably caused by thermal stratification resulting from the formation of the seasonal thermocline. Temperatures at both levels became nearly identical when the seasonal thermocline was eroded away during the summer-to-winter phase of the surface temperature cycle. It was also found that the annual range in temperature at 50 meters was o only about 3.8 C. 67 3. Salinity at both the surface and 50 meters has a semi- annual periodicity. No satisfactory conclusion was drawn as to what causes this periodicity. Possibly periodic advection of low- salinity subarctic water into this region occurs. 4. Surface temperature appears to be highly correlated with 50 meter temperatures only during the surface temperature transitional periods. The highest correlation occurred during the summer-to- winter transition period. 5. There appears to be little correlation of surface tempera- ture with temperatures at 200 meters. Temperature at 200 meters show only slight annual periodicity, as does salinity at this level. 6. Salinity appears to be, in general, poorly correlated with surface temperature. A single high correlation coefficient was calculated during an anomalous cooling trend that occurred during the summer-to-winter transitional period of 1969. This study revealed completely unexpected and unexplained semi-annual variations in salinity. A further study of the fluctuations in salinity should be undertaken in order that the cause of this feature be found. Such an undertaking is, however, an extremely expensive proposition involving a tremendous investment in oceanographic research time. Recently the decision was made to discontinue two Ocean Weather Stations, including November. This will make future research even more difficult. 68 It is further recommended that this study be used as a basis for the study of the correlation between fluctuations of metereorologic factors and ocean surface temperature and salinity fluctuations. Perhaps such a study would shed more light on the causes of the observed oceanographic fluctuations noted during the course of this research. 69 BIBLIOGRAPHY 1. Bathen, K. H. , "On the Seasonal Changes in Depth of the Mixed Layer in the North Pacific Ocean", J. Geo. Res. , Vol. 77, no. 36, p. 7138-7150, December, 1972. 2. Beland, C. L. , Sea Surface and Related Subsurface Temperature Anomalies at Several Positions in the Northeast Pacific Ocean, Master's Thesis, U. S. Naval Postgraduate School, 148 pp. , March, 1971. 3. Clark, N. E. , "Specification of Sea Surface Temperature Anomaly Patterns in the Eastern North Pacific, " J. Phy. Ocean, Vol. 2, no. 3, p. 391-404, October, 19lY. 4. Dietrich, G. , "General Oceanography, " Wiley, 1963, 588 pp. 5. Husby, D. M. , "Oceanographic Observations North Pacific Ocean Station November, " CG 373-18, February, 1968. 6. Husby, D. M. , "Oceanographic Observations North Pacific Ocean Station November, " CG 373-26, September, 1969. 7. Isaacs, J. D. , "The North Pacific Study, " J. of Hydronautics, v. 3, No. 2, p. 65-72, April, 1969. 8. Namias, J. , "Macroscale Variations in Sea-Surface Temperatures in the North Pacific, " J. Geo. Res. , Vol. 75, no. 3, p. 565-582, January, 1970. 9. Namias, J. and Born, R. M. , "Temporal Coherence in North Pacific Sea-Surface Temperature Patterns, " J. Geo. Res., Vol. 75, no. 30, p. 5952-5955, October, 1970! 10. Namias, J., "The 1968- 1969 Winter as an Outgrowth of Sea and Air Coupling During Antecedent Seasons, " J. Phy. Ocean, Vol. 1, no. 2, p. 65-81, April, 1971. 11. National Oceanographic Data Center, Ocean Weather Station November Data, Washington, D. C. , 1972. 12. Robinson, M. K. , "Atlas of Monthly Mean Sea Surface and Subsurface Temperature and Depth of the Top of the Thermocline North Pacific Ocean, " FNWC, May 1971. 13. Roden, G. I. , "Aspects of the Mid-Pacific Transition Zone, " J. Geo. Res. , Vol. 75, no. 6, p. 1097-1109, February, 1970. 70 14. Roden, G. I. , "Aspects of the Transition Zone in the North- eastern Pacific, " J. Geo. Res. , Vol. 76, no. 15, p. 3462-3475, May, 1971. 15. Sverdrup, H. U. , Johnson, M. W. and Fleming, R. H. , The Oceans, Prentice-Hall, 1942, pp. 1087. 16. Tully, J. P., "Oceanographic Regions and Assessment of Temperature Structure in the Seasonal Zone of the North Pacific Ocean, " J. Fish. Res. Bd. Canada, V. 21, no. 5, p. 941-970, 1964. 17. U. S. Department of Commerce, "Climatological and Oceano- graphic Atlas for Mariners, North Pacific Ocean, " U. S. Department of the Navy, Office of Climatology and Oceanographic Analysis Division, 1961. 18. Wickett, W. P., and Thomson, J. A., "Transport Computations for the North Pacific Ocean, 1970, " Fish Res. Bd. Canada, Tech Rep. No. 238, February, 1971. 19. Wickett, W. P., and Thomson, J. A., "Transport Computations for the North Pacific Ocean, 1969, " Fish. Res. Bd. Canada, Tech. Rep. No. 239, February, 1971. 20. Wyrtki, K. , "The Average Heat Balance of the North Pacific Ocean and Its Relation to Ocean Circulation, " J. Geo. Res. , V. 70, no. 18, p. 4547-4559, September, 1965^ 71 INITIAL DISTRIBUTION LIST No. Copies 1. Defense Documentation Center 2 Cameron Station Alexandria, Virginia 22314 2. Library, Code 0212 2 Naval Postgraduate School Monterey, California 93 940 3. Lieutenant Donnel E. Hansen, USN 2 3985 Marcola Road Springfield, Oregon 97477 4. Oceanographer of the Navy 1 The Madison Building 732 N. Washington Street Alexandria, Virginia 22314 5. Department of Oceanography 3 Code 58 Naval Postgraduate School Monterey, California 93 940 6. Office of Naval Research 1 Department of the Navy Washington, D. C. 20360 7. Dr. R. H. Bourke, Code 58 3 Department of Oceanography Naval Postgraduate School Monterey, California 93940 8. Dr. Noel Boston, Code 58 1 Department of Oceanography Naval Postgraduate School Monterey, California 93 940 72 9. Commanding Officer TJ. S. Coast Guard Oceanographic Unit Bldg. 159-E, Navy Yard Annex Washington, D. C. 20390 10. Mr. Kevin Rabe Environmental Prediction Research Facility Bldg. 14 Naval Postgraduate School Monterey, California 93940 11. Lieutenant Commander W. A. Caster, USCG Chief, Oceanography Branch Coast Guard Western Area 630 Sansome Street San Francisco, California 94126 73 Unclassified Security Classification z DOCUMENT CGMTKOL DATA -R&D (Security clrt si lie rt ion ol title, body o( abttrcct and Indexing annotation mutt be entered whan tha ovarall tapott It elattlllad I ORIGINATING ACTIVITY (Corpotata author) Naval Postgraduate School Monterey, California 93940 2*. REPORT SECURITY CLASSIFICATION Unclassified 26. GROUP 3 REPORT TITLE A Study of Surface, 50 Meter and 200 Meter Temperature and Salinity Fluctuations at Ocean Weather Station November, 1968-1970 4. DESCRIPTIVE NO T CS (Type ol report and. Inclusive deft) Master's Thesis; June 1973 S. au TMORISI (l-'trtt nama, middla Initial, la at name) Donnel E. Hansen 6. REPOR T DATE June 1973 7a. TOTAL NO. OF PAGES 75 7b. NO. OF REFS 20 •«. CONTRACT OR GRANT NO- b. PROJEC T NO. Us. ORIGINATOR'S REPORT NUMBKRISI 9b. OTHER REPORT NOISI (Any othar number* that may ba aetlgnad thla report) 10. DISTRIBUTION STATEMENT -Approved for public release; distribution unlimited. II. SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY Naval Postgraduate School Monterey, California 9 3940 S. ABSTRACT Long and short-term surface, 50 and 200 meter temperature and salinity fluctuations at Ocean Weather Station November during 1968 through 1970 were examined using several statistical techniques. One technique, a unique monthly bi-variate modal analysis of surface temperature and salinity, proved to be a valuable research tool in that the resulting monthly modal cells, when plotted on a T-S diagram, provided a simplified view of the annual T-S relationships. The average range of mean monthly temperatures at the surface was found to be 5.6 C and 4. 0 C at 50 meters. Salinity at both the surface and 50 meters exhibited a semi-annual periodicity. Range of temperature at 200 meters was 1.5 C. The results of a seasonal linear regression analysis show that surface and 50 meter temperatures were correlated during periods of increasing or decreasing surface temperatures. Surface temperature and salinity were correlated only during the apparent advection of modified subarctic water into the region around OWS November. DD ,F°™..1473 FORM I NOV S/N 0101 -807-681 1 (PAGE 1) 74 Unclassified Security Classification A-S14O0 Unclassified Security Classification key wo ROI Surface Temperature 50 Meter Temperature 200 Meter Temperature Surface Salinity 50 Meter Salinity 200 Meter Salinity Temperature Salinity Correlations DD,F.r..1473 i back) S/N 01 01 -807-682 1 75 U nclas s ified Security Classification A-31 *09 Th Thesis H2 H20124 Hansen 145265 Thesis H20124 c.l A study of surface, 50 meter and 200 meter temperature and salinity fluctuations at ocean 7 JUNM ^$*mm 145265 Hansen A study of surface, ^0 meter and 200 meter temperature and salinity fluctuations at ocean weather station November, 1968-1970.