STATISTICAL STUDIES OF WORLD-WIDE SECCHI DATA Gerald Lee York DUDLEY KNOX «-«BR^ NAVAL POSTGRADUATE S-HOO. S££r. ca"~rn,a 9394° NPS-58TX7lt031 rUolbK Monterey, Californi L "" — STATISTICAL STUDIFS OF WORLD-WIDE SFCCHI DATA by Gerald Lee York March 19 lh Thesis Advisor: S. P. Tucker Prepared for: Office of Naval Research Code U80D Arlington, Virginia 22217 T 161502 Appiovzd ^on. public A.c£eoie; dli>t/uhixtion antimiX.zd. Statistical Studies of World-Wide Secchi Data by Gerald Lee ,York Lieutenant, United States Navy B.S., in Animal Husbandry, Southern Illinois University, 1967 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIFNCE IN OCFANOGPAPHY from the NAVAL POSTGRADUATE SGHOOL March 1974 YS2 NAVAL POSTGRADUATE SCHOOL Monterey, California Rear Admiral Mason Freeman Jack R. Borsting Superintendent Provost This thesis is prepared in conjunction with research supported in part by the Office of Naval Research under Project Order No. POh-0121. Reproduction of all or part of this report is authorized. Released as a Technical Report by: ABSTRACT An investigation was made to determine possible correla- tions between Secchi depths and other simultaneously measured oceanographic parameters which were on file at the National Oceanographic Data Center as of March 1972. Sixty-three one- degree sub-squares occurring in Japanese and Korean waters and eleven Atlantic and Pacific open ocean areas were chosen for linear correlation analysis using both sea surface data and mean values of some fourteen different oceanographic parameters averaged over the Secchi depth. In particular, oxygen measurements exhibited trends toward an inverse pro- portionality with Secchi depth while temperature indicated a possible direct proportionality. Time series analyses of Secchi depths were performed and compared with upwelling indices computed for the Oregon' coast and near Monterey Bay, California. An inverse proportionality and possible phase lag of mean Secchi depth compared to monthly upwelling index was observed. Multiple regression equations relating Secchi depth and upwelling index were calculated for both locations. TABLE OF CONTENTS I. INTRODUCTION - 11 A. GENERAL 11 B. BACKGROUND ON THE SECCHI DISC 13 C. PURPOSE OF INVESTIGATION 19 II. METHODS OF INVESTIGATION 21 A. DEVELOPMENT AND DESCRIPTION OF THE CORRELATION COEFFICIENT --- 21 1. Variance and Covariance 21 2. Correlation Coefficient 22 B. METHOD USED IN OBTAINING CORRELATION COEFFICIENTS 22 1. Biomedical Computer System Program (BIOMED) 22 2. BIOMED 02D (Correlation with Transgeneration) 23 III. ANALYSIS OF DATA 24 A. GENERAL 24 B. LINEAR CORRELATION ANALYSIS ' 25 1. Japanese and Korean Waters 25 a. Correlations Using Sea Surface Chemistry Values 25 b. Correlations Using Mean Chemistry Values 26 2. Open Ocean Areas 26 C. TIME SERIES ANALYSIS .----_ _-- 27 IV. DISCUSSION OF RESULTS 29 A. LINEAR CORRELATION COEFFICIENTS USING SEA SURFACE CHEMISTRY VALUES 2 9 1. Japanese and Korean Waters -- 2P a. Color 30 b. Bottom Depth 31 c. Temperature 31 d. Salinity 32 e. Sigma-t 32 £. Oxygen 3 3 g. Silicate 33 2. Open Ocean Areas , 33 a. Atlantic Ocean 34 b. Pacific Ocean 34 B. LINEAR CORRELATION COEFFICIENTS USING MEAN CHEMISTRY VALUES 35 C. TIME SERIES ANALYSIS 35 1. Oregon Coast 36 2. Monterey Bay a. Relationship Between Secchi Depth and Upwelling Index - 40 40 Relationship Between Phytoplankton Wet Volume and Upwelling Index 42 V. SUMMARY AND CONCLUSIONS 44 47 VI. PROPOSED FUTURE RESEARCH AVERAGING PROGRAM SAMPLE BIOMED02D OUTPUT TIME SERIES ANALYSIS -- APPENDIX A APPENDIX B APPENDIX C 136 138 139 BIBLIOGRAPHY 141 INITIAL DISTRIBUTION LIST - 144 FORM DD 1473 148 LIST OF TABLES I. Surface Data Distribution by Marsden Sub-Square -- 49 II. Parameter Means by Marsden Sub-Square 54 III. Linear Correlation Coefficients by Marsden Sub-Square 61 IV. Data Density Code Used in Figures 5-19 65 V. Open Ocean Area Delineations 66 VI. Surface Data Distribution by Open Ocean Area 67 VII. Parameter Means by Open Ocean Area 68 VIII. Linear Correlation Coefficients by Open Ocean Area 70 IX. Parameter Means by Marsden Sub-Square Using Values Averaged to the Secchi Depth '1 X. Linear Correlation Coefficients by Marsden Sub- Square Using Values Averaged to the Secchi Depth 73 XI. Parameter Means by Open Ocean Area Using Values Averaged to the Secchi Depth '^ XII. Linear Correlation Coefficients by Open Ocean Area Using Values Averaged to the Secchi Depth 75 XIII. Regression Analysis Results (Oregon Coast) '° 7 7 XIV. Regression Analysis Results (Monterey Bay) ' LIST OF FIGURES 1A Marsden Square Chart Showing Open Ocean Areas Studied in the Atlantic - 78 IB Marsden Square Chart Showing Open Ocean Areas Studied in the Pacific 79 2 One Degree Sub-square Numbering System 80 3A One Degree Sub-square Delineation Chart for Korean Waters 81 3B One Degree Sub-Square Delineation Chart for Japanese Waters 82 4A-4N Correlation Coefficient Graphs - Western Pacific 83-96 5-7 Color Plotted as a Function of Secchi Depth 97-99 8-9 Bottom Depth Plotted as a Function of Secchi Depth 100-in 10-12 Surface Temperature Plotted as a Function of Secchi Depth 102-104 13-14 Surface Salinity Plotted as a Function of Secchi Depth 105-106 15-16 Surface Sigma-t Plotted as a Function of Secchi Depth 107-108 17-18 Surface Oxygen Plotted as a Function of Secchi Depth 109-110 19 Surface Silicate Plotted as a Function of Secchi Depth ll1 20A Correlation Coefficient Graph - Atlantic Ocean -- ^' 20B-20C Correlation Coefficient Graphs - Pacific Ocean 112-114 21 Points for Which Upwelling Indices were Computed by Bakun (1973) - 115 22 Secchi Depth and Upwelling Index vs. Month of Year for the Oregon Coast - 1961 - 116 7 23 Secchi Depth and Upwelling Index vs. Month of Year for the Oregon Coast - 1962 117 24 Secchi Depth vs. Upwelling Index for the Oregon Coast - 1961 118 25 Secchi Depth vs. Upwelling Index for the Oregon Coast - 1962 119 26 Secchi Depth vs. Upwelling Index for the Oregon Coast 1961 - 1962 120 2 7 Monterey Bay, Showing Locations of CalCOFI Stations Occupied by Hopkins Marine Station of Stanford University 121 28 Secchi Depth and Upwelling Index vs. Month of Year for Monterey Bay Station 3 - 1970 122 29 Secchi Depth and Upwelling Index vs. Month of Year for Monterey Bay Station 3 - 1971 123 30 Secchi Depth and Upwelling Index vs. Month of Year for Monterey Bay Station 3 - 1972 124 31 Secchi Depth and Upwelling Index vs. Month of Year for Monterey Bay Statoin 3 - 1973 125 32 Secchi Depth and Upwelling Index vs. Month of Year for Monterey Bay Station 4 - 1971 126 33 Secchi Depth and Upwelling Index vs. Quarter of Year for Monterey Bay Station 3 1970-1972 127 34 Secchi Depth vs. Upwelling Index for Monterey Bay Station 3 - 1970 128 35 Secchi Depth vs. Upwelling Index for Monterey Bay Station 3 - 1971 129 36 Secchi Depth vs. Upwelling Index for Monterey Bay Station 3 - 1972 130 37 Secchi Depth vs. Upwelling Index for Monterey Bay Station 3 - 1973 131 38 Secchi Depth vs. Upwelling Index for Monterey Bay Station 4 - 1971 132 39 Secchi Depth vs. Upwelling Index for Monterey Bay Stations 3 and 4 1970-1973 133 40 Secchi Depth vs. Upwelling Index for the Oregon Coast 1961-1962 and Monterey- Bay Stations 3 and 4 1970-1973 -- -- 134 41 Phytoplankton Wet Volume vs. Upwelling Index for Monterey Bay 1956-1967 135 ACKNOWLEDGEMENTS I would like to express my appreciation to my thesis advisor, Professor Stevens P. Tucker, Department of Ocean- ography, without whose dedicated interest, patience, and enthusiasm, this project could not have come to successful completion. I would also like to thank Professor Robert S. Andrews for suggestions during the final stages of this study. I am also indebted to Henry Odum of the National Oceanographic Data Center for providing the data tapes which were funded by the Office of Naval Research (Arlington, Va , ) and to David Norman of the Postgraduate School Computer Center for his assistance in adapting the data tapes for use on the computer. Lastly, I would like to thank David Bracher of Hopkins Marine Station and Andrew Bakun of the National Marine Fishery Services of Monterey for supplying additional data used in this study. 10 I. INTRODUCTION A. GENERAL Optics, considered as a special branch of oceanography, has been the subject of renewed interest among oceanographers during the past few years. Solar radiation serves as the source of energy for the oceans, supplying them with heat and supporting their ecology through photosynthesis. Light is important for nekton and zooplankton of the ocean in finding their food and evading attack. Daylight and arti- ficial lighting are also important for underwater viewing. And light may be used on occasion as an effective probe to resolve otherwise ambiguous measurements in physical oceanography. Several applications of light to the study of the oceans have been noted. Tyler and Preisendorf er (1963) have class- ified these under three broad areas, including, (1) Descriptive oceanography and other geophysical applications ; (2) Photosynthesis and other biological phenomena; and (3) Image-recording equipment. Duntley (1965) speculated on the possibility of conduc- ting oceanographic studies by human observers in a Manned Orbital Research Laboratory (MORL) . Among the potentialities discussed were the determination of sea state and surface wind velocity by means of visible light. He explained that the shape and size of the glitter pattern due to the 11 reflection of the sun by the surface of the sea is inter- pretable in terms of surface wind velocity, and that spatial- ly averaged "inherent" radiance* of the ocean varies in a known way with sea state. The above potentialities have been achieved to a limited extent in recent years by the use of satellites such as Sky- lab and ERTS-1. Petri and Starry (1973) have also estab- lished the feasibility of remotely measuring wind magnitude and direction in a real environment by the use of pulsed laser systems. Growing attention has been attracted to the possibility of characterizing water masses by means of their optical properties (Jerlov, 1968). For example, Pak and Zaneveld (1973) traced the Cromwell Current to the east of the Gal- apagos Archipelago using optical techniques. Among important applications of optical oceanography, one of the most important is in the field of marine biology. The physics of radiant energy is of direct importance for evaluating the photosynthetic activity in the sea. Optical measurements have served as a valuable aid in locating areas of high biological production and potential fishing grounds. Duntley (1965) has pointed out that multi-spectral photo- graphy conducted from an MORL should enable a quantitative assay of chlorophyll in sea water, and that other biological features of the ocean, for example, the occurrence and Radiance is flux per unit projected area per unit solid angle in a specified direction. 12 distribution of red tide, should be observable under clear ■weather conditions. Clarke, et al. (1970) have shown that spectral measurements of backscattered light can be used to determine the abundance of chlorophyll as well as to trace currents, pollutants, or other significant materials in the water . The work of Duntley (1952) emphasized the importance of underwater lighting for vision, television, and photography. He explained the importance of quantitative prediction of the irradiation produced at the object, on its background and throughout the observer's path of sight by incondescent lamps or flash tubes. This can enable optimum lighting arrangements and camera positions to be planned in advance and exposure to be predicted with sufficient accuracy to permit high-contrast photographic techniques to be employed effectively. Duntley (1971) explained that the greatest hope for truly long range underwater imagery is by means of pulsed lasers and gated electro-optical cameras. Scatterance and beam transmittance meters are commonly used in the field of pollution research. Another frequently employed measurement scheme involves the use of fluorescent dyes as tracers in order to study diffusion in the sea. B. BACKGROUND ON THE SECCHI DISC The Secchi disc is one of the most widely used devices for measuring ocean water transparency. The disc was first mentioned in a published report by Commander Cialdi in 1865 and recently translated into English by Collier (1968) . 13 Cialdi's report contained a scientific diary by Professor Secchi in which the factors affecting the visibility of a disc when lowered vertically in the sea were examined. These factors included disc color, solar altitude, sea surface reflections and refractions, ship's shadow, sky clearness, water color, disc diameter, and the height of the viewer above the water surface. Secchi observed an increase in depth at which the disc disappeared from sight associated with increased disc whiteness, solar altitude, sky clearness, and disc diameter. He noted that image dissection by surface refraction caused the visibility of the disc to decrease, and that the ship's underwater shadow also influenced its visibility. He also demonstrated the detrimental effect of surface reflections on the measurement and recommended a wide shadow over the place where the observations were being made. Secchi' s work established the experimental procedure for obtaining transparency with a Secchi disc, and in the years following his work, the Secchi disc became a widely used oceanographic tool. However, Tyler (1968) noted that it was never really standardized. That is to say, it was used widely because of its simplicity, but its physical properties were never fully specified. Holmes (1970) also noted that both disc diameter and reflectance have never been standardized or specified. Postma (1961) observed the 14 X following limitations of Secchi disc measurements compared to measurements carried out by submersible K_-meters: (1) Secchi measurements can only give information on the extinction in surface waters and they can only be carried out in daylight of sufficient brightness, whereas irradiance measurements can be performed to somewhat greater depths. (2) When using a Secchi disc no continuous registration is possible, nor are determinations at various wavelengths, whereas the use of appropriate filters in a K_-meter allows recording continuously with depth. (3) Finally, the result of a measurement with a Secchi disc depends upon the visual acuity of the observer and on the daylight illumination and reflection from the sea's surface, which is not the case with a K_-meter. Because of these limitations and difficulties it might appear that Secchi measurements are of no great importance. On the contrary, they can give valuable results as will be shown below. The Secchi disc has been widely used because of its low cost and convenience, and considerable research has been devoted to its utility as a practical instrument for measuring water transparency. Secchi depth measurements have been especially useful to marine biologists, who have established practical relation- * K_ is the diffuse attenuation coefficient, a measure of the exponential attenuation of downwelling irradience in the sea. Biologists often use the term "vertical extinction coefficient" to denote K . It is not to be confused with the beam attenuation coefficient ("c" or "«*") , a measure of the total attenuation of a collimated light beam through a fixed path length. 15 ships between Secchi depths and vertical extinction coef- ficients. Holmes (1970) mentioned that it is common prac- tice for biologists interested in primary production to consider the bottom depth of the euphotic zone to be equal to three times the Secchi depth. An inverse relation between the amount of phytoplankton and the visual range of the Secchi disc has been observed by Atkins, Jenkins, and Warren (1954), Arsen'yev and Voytov (1968), Voytov and Dement'yeva (1970), and others. From data collected in the English Channel, Poole and Atkins (1929) developed a widely used empirical formula for approximating extinction coefficients: K_ = 1.7/ZS where K is the vertical extinction coefficient and Z is the — s Secchi depth in meters. Murphy (1959) established a positive correlation between albacore troll catches and water clarity. He asserted that the Poole-Atkins relation can be used to approximate closely the horizontal visual range of albacore. The Poole-Atkins relation has also served as an aid in the investigation of primary organic productivity as demonstrated by Ryther and Yentsch (1957) . Holmes (1970) investigated transparencies in Goleta Bay and suggested that for turbid water 1.44 is probably a more appropriate factor than 1.7 in the equation above in estimating extinction coefficients from Secchi depths. He also suggested that the relation between Roughly the depth at which the downwelling irradiance (K ) has decreased to 1% of its value at the surface. 16 Secchi depth and the \% optical depth merits additional study to incorporate a wide range of Secchi depths. Visser (1967) examined Secchi and seawater color obser- vations from the North Atlantic Ocean and developed the following empirical relation relating Secchi depth and yellow content of seawater: iy^- = 0.26Y + 1.9 where Z is Secchi depth in meters and Y is the percentage yellow calculated from the Forel color scale. However, he cautioned that the relation was valid only for the particular ocean area investigated. Frederick (1970) examined possible similar relations between Secchi disc observations and color codes for other ocean areas based on Visser's findings. Much variability was found to exist, and no simple empirical relation could be determined. Brown (1973) observed a similar pattern in relating Secchi depth and Forel color code as reported by Visser. Although he observed the same trend, no universal numerical relationship valid for all oceans was found. Graham (1966) determined relationships between diffuse attenuation coefficients (K_) , reciprocals of Secchi disc readings, and color observations from data collected in the central and eastern North Pacific Ocean. He concluded that the Secchi disc is a useful tool, but that caution should be observed when extrapolating the relationship between Secchi 17 disc measurements and extinction coefficients from one oceanic environment to another. Postma (1961) investigated the relation between Secchi depth measurements and suspended matter both experimentally in the laboratory and in the coastal waters of the Nether- lands. He concluded that Secchi disc measurements are a valuable source for additional information concerning prop- erties of suspended matter. Estimates based on the empirical relationships between diffuse attenuation coefficients (K__) and amount of suspended matter per unit volume of sea water and Secchi depth discussed above are usually strictly valid only in one particular oceanic region and are not generally useful elsewhere. Although these estimates may have rela- tively large standard errors associated with them, they may be acceptable for certain types of work, such as in some areas of marine biology, where a high degree of precision and accuracy is not always required, or in marine geology, where gross measures of sediment transport are desired. In developing practical relationships between Secchi depth and other oceanographic parameters, correlation coefficient analysis is considered to be a useful starting point. Brown (1973) conducted such an analysis on a world- wide basis using sea surface data. Because mid-oceanic data were sparse, nearly all areas analyzed were coastal areas subject to localized effects such as fresh water run- off and upwelling. With these limitations, no simple and consistent relations between Secchi depth and other 18 parameters were evident; however, several trends were noted. He found that oxygen measurements exhibited trends toward an inverse proportionality with Secchi depth, while bottom depth data indicated a possible direct proportionality. He also observed that lower salinity water and high amounts of sili- cate were associated with decreased transparency in coastal areas subject to fresh water runoff. C. PURPOSE OF INVESTIGATION In view of the studies discussed above it was proposed to continue the search begun by Brown (1973) for possible correlations between Secchi depths and other simultaneously measured oceanographic parameters in areas of high data density. Areas as small as one degree latitude by one degree longitude were chosen to avoid unnecessary averaging of data from varying water types and differing coastal influences but at the same time to maintain a high data density, insur- ing a fairly representative analysis. Open ocean areas having no coastal-type influences, such as from fresh water runoff and upwelling, were also to be examined, since correlations determined for such areas might yield results which could be simply and accurately extrapo- lated to similar ocean areas. Resulting correlations from coastal and open ocean areas were then to be compared and consistent trends were to be noted. Furthermore, it was proposed to compare correlations between 'Secchi depths and sea surface data and those be- tween Secchi depth and mean values of oceanographic parameters 19 averaged over the Secchi depth. This was to determine the validity of the use of sea surface measurements conducted in past correlation studies of this nature (Brown, 1973). In addition, monthly and yearly time series analyses of Secchi depths were to be performed and compared with upwellinj indices computed from historical meteorological data by Bakun (1973) for the west coast of North America. 20 • II. METHODS OF INVESTIGATION A. DEVELOPMENT AND DESCRIPTION OF THE CORRELATION COEFFICIENT 1 . Variance and Covariance The variance and covariance are necessary in the development and formulation of the correlation coefficient. A brief description and a summary of these statistical measures are provided in this section (Dixon and Massey, 1957). The variance, a , is defined as: 1=1 where N is the number of observations X. and u is the mean 1 N " of the Xif y = ^ I Xj[. The variance is concerned with a single measured variable. The object of statistical analysis is often directed at discovering relationships among two or more variables. The simplest way of determining a relationship between two variables is to compute their covariance, a measure of the common variance between two variables. This measure is hard to use directly but is very important in the development of more advanced analysis. The covariance be- tween X and Y, with arithmetic means u and u , respectively, x y is given as : 21 2 . Correlation Coefficient To put the variances of two individual variables and their covariance into a meaningful measure, the corre- lation coefficient is used. This statistic ranges from -1 to +1, where +1 correlation indicates that two variables are exactly alike, i.e., the rate of change in both is propor- tional. Zero correlation implies statistical independence or the absence of any association. Negative correlation implies opposite association with one another. That is to say, as one variable increases the other consistently decreases. The correlation coefficient is defined: 2 , , 2 2 ,h Pij = a±. I (o±±o..) where p.. is the correlation between the i and j vari- able, a-- is the covariance between the i and j variable, and a-- and a., are the respective variances. 11 33 F B. METHOD USED IN OBTAINING CORRELATION COEFFICIENTS 1 . Biomedical Computer System Program (BIOMED) The Biomedical computer system programs were developed at the University of California at Los Angeles (Dixon, 1973). The programs were initially developed to handle extensive analyses of large amounts of data in medical research. However, they are written in such a way that a wide variety of problems may be handled by each pro- gram by specifying the appropriate parameters of the problem. 22 2 . BIOMED 02D (Correlation with Transgeneration) This program is designed to provide basic descrip- tion and tabulation on raw data. The output consists of the sums, means, and sta-ndard deviations of all variables. In addition three matrices are provided. All three are square and symmetric with dimensions equal to the number of variables. The first and second matrices are the cross- product deviations matrix and the variance-covariance matrix respectively. The third matrix is the correlation matrix. The diagonal elements show the correlation of variables with themselves and, by definition, they should correlate perfectly. Hence, as a check of validity of the correlation matrix, the diagonal elements should all be 1.0. A sample output of this program is provided in Appendix B. The two most significant features of this program are the Boolean selection of cases on input and the cross- plotting of variables on output. The Boolean selection enables the screening of cases in order to omit those of no interest . The cross-plotting feature enables the user to iden- tify a base variable and plot other variables against it on individual graphs. Transgeneration options are also avail- able for use in this program. 23 III. ANALYSIS OF DATA A. GENERAL The primary oceanographic data used in this study were on magnetic tapes obtained from the National Oceanographic Data Center (NODC) . The information included a global coverage to March 1972 of all NODC Secchi data plus all the other station data collected at the same time Secchi measure ments were made, including all chemistry from 86,258 sta- tions. Screening of data to simplify computer handling was conducted for a former study (Brown, 1973) and preserved on tape. The data used in the present study consisted of the following : Secchi depth Day Year Latitude Longitude Marsden square Water depth Forel color Cloud cover Month Water temperature Salinity Sigma-t Oxygen Phosphate Phosphorus Nitrite Nitrate Silicate Only those chemistry measurements obtained at depths above or at the same level as the Secchi depth at each station were employed. Sample pH, although available, was not used in this study, 24 The data were stored on disc at the Naval Postgraduate Computer Center for analysis. A previous inventory of the data indicated a sparsity of open ocean data and an abun- dance of data in some coastal waters, especially off Japan and Korea. In referring to the geographical areas studied, a ten- degree latitude by ten-degree longitude Harsden square numbering system was used. Figures 1A and IB show the global Marsden square coverage. In high data density areas Marsden squares were further broken down into one-degree sub-squares. Figure 2 shows the one-degree division number- ing system used. B. LINEAR CORRELATION ANALYSIS 1 . Japanese and Korean Waters a. Correlations Using Sea Surface Chemistry Values Due to the great relative abundance of data in Japanese and Korean waters, they were selected for initial analysis. They fall within Marsden squares 130, 131, and 132, which were broken down into one-degree subsquares. After preparation of a data distribution inventory, the 63 subsquares indicated in Figures 3A and 3B were chosen for linear correlation analysis. Coefficients and cross-plots were then obtained using Secchi depth measurements as a base variable against latitude, longitude, water depth, Forel color, cloud cover, month, and all sea surface chemistry measurements . 25 b. Correlations Using Mean Chemistry Values Upon completion of the initial correlation analysis, 21 previously chosen subsquares were selected for further analysis. Correlation coefficients and cross-plots were again obtained using mean values of parameters averaged over the Secchi depth at each station. Temperature, salinity, sigma-t, and oxygen were selected on the 'basis of consistency of correlations and data density. A sample program used for averaging the parameters is provided in Appendix A. Screen- ing was necessary in both analyses to eliminate stations with erroneous or questionable data. 2 . Open Ocean Areas Although open ocean data were limited, 11 areas were selected for correlation analysis. These included six areas in the Pacific Ocean and five areas in the Atlantic Ocean and are shown in Figures 1A and IB, designated by an area number. As can be seen from the figures, each of the areas selected in the Atlantic Ocean contains several Marsden squares. It was necessary to use more than one square in order to provide enough data for a reasonably representative analysis. All areas were selected to provide a sufficient data base and to minimize coastal influences such as fresh water runoff and upwelling. Correlation coefficient analyses were accomplished using both surface values of oceanographic parameters and the mean values of parameters averaged over the Secchi depth. The boundaries of these areas are given in Table V. 26 C. TIME SERIES ANALYSIS The purpose of a time series analysis was to group Secchi depths by month and year and to compute average Secchi depths by month. The resulting average Secchi depths were then compared to previously computed upwelling indices along the west coast of North America. Appendix C is an example of the type of Fortran program utilized in the time series analysis. Coastal upwelling indices were obtained from Bakun (1973) for years 1946 through 1971. In addition up- welling indices were obtained for years 1972 and 1973 (Bakun, 1974). Figure 21 shows the data grid and intersections at which his upwelling indices were computed. Bakun 's monthly indices were based on offshore Eckman transport calculated from daily mean surface atmospheric pressure data. Summaries by quarter and by year were also included. In generating the indices Bakun estimated the daily mean wind stress on the sea surface at points near the coast, from this computed the Eckman transport, and finally resolved the component of Eckman transport perpendicular to the coast. The resulting "upwelling indices" have units of cubic meters per second per 100 meters of coastline. The magnitude of the offshore component is considered an indica- tion of the amount of water upwelled to replace that driven offshore. Negative index values indicate onshore transport or convergence at the coast (downwelling) . The time series program was utilized for two areas for which upwelling indices were available to allow a comparison 27 between the indices and monthly Secchi depth averages. In addition to the NODC data, Secchi depth data were obtained from the Hopkins Marine Station of Stanford University. Since 1951 Hopkins Marine Station has carried on a continu- ous hydrobiological survey for the California Cooperative Oceanic Fisheries Investigations (CalCOFI) with cruises at approximately two-week intervals on Monterey Bay and includ- ing stations at the six locations shown in Figure 27. 28 IV. DISCUSSION OF RESULTS A. LINEAR CORRELATION COEFFICIENTS USING SEA SURFACE CHEMISTRY VALUES 1 . Japanese and Korean Waters Summaries of data distribution, parameter mean values, and resulting correlation coefficients are tabulated in Tables I, II, and III respectively. They are listed according to Marsden square and Marsden subsquare numbers . Linear correlation coefficient graphs for most of the sub- squares are plotted in Figures 4A through 4N, and several samples of cross-plots are illustrated in Figures 5 through 19. The data density codes used in the cross-plots are translated in Table IV. As was expected (Brown, 1973) no consistencies in correlation coefficients between Secchi depth and latitude, longitude, cloud cover, and month of year were apparent. However, cross-plots of latitude and longitude served as a valuable aid in determining erroneous station positions. This was true where coastal boundaries were within the sub- square boundaries. Stations with inland position locations were then screened and discarded during analyses. Cross - plots of month of year were also valuable in determining if station densities were representative throughout the year. Although one might expect a dependency of Secchi depth on month of year due to varying sun altitude, it appeared that 29 other parameters and factors such as upwelling and fresh water runoff had a dominating influence on transparency. Some parameters were not included in the correlation coefficeint graphs and cross-plot figures, although they may occur in the summary tables. This was due to a limited amount of data available for analyses and/or strong inconsistencies resulting from correlation analysis. The parameters excluded were phosphate, phosphorus, nitrite and nitrate. Correla- tions between Secchi depth and remaining parameters will be discussed separately. It should be noted that cross-plots of all paramater pairs were made for each ocean area studied. Figures 5-19, discussed below, were selected as representa- tive or typical. a. Color Forel color as was to be expected (Brown, 1973) correlated more consistently than any other parameter, with negative coefficients resulting in all cases but one. The exception occurred in Marsden square 132, subsquare 38, and resulted in a slightly positive coefficient. Typical examples of Forel color plotted against Secchi depth are shown in Figures 5 through 7. Cross-plots of these two variables appeared to range from a nearly linear to a nearly exponential trend. The same ranges occurred in subsquares directly within coastal influences and subsquares having little or no coastal influence. 30 b. Bottom Depth As was expected, bottom depth correlated posi- tively. However, in a few cases negative coefficients resulted. Positive coefficients were especially pronounced in subsquares with shallow mean depths and well within the range of coastal influences. One would expect this result, considering the high amount of annual rainfall and runoff that occur in Japan and Korea. Large quantities of suspended and dissolved materials would be expected, resulting in decreased transparency in shallo\\r coastal waters. Such a trend can be seen in Figures 8 and 9. Subsquares including waters with greater mean depths and with little coastal influence did not exhibit a pronounced trend. Coefficients for these subsquares varied from strongly negative to strongly positive. These observations appear to indicate that bottom depth has little or no influence on Secchi depth measurements in mid-ocean areas. c. Temperature In all but four subsquares, temperature exhibited a positive correlation. The four exceptions occurred in sub- squares situated near or within bays and resulted in slightly negative coefficients. Cross-plot examples of temperature against Secchi depth are given in Figures 10 through 12. Figure 10 is for a shallow water, coastal subsquare, while Figures 11 and 12 are for deep water subsquares separated from coastal influences. The coefficients and cross-plots resulting from the temperature analyses indicate a strong 31 dependence between Secchi depth and sea surface temperatures. This is not surprising, especially in areas where upwelling results in lower temperatures and increased amounts of nutrients near the sea surface. This in turn would tend to enhance phytoplankton blooms and thus lead to lower Secchi depths. d. Salinity A consistent correlation or trend between Secchi depth and salinity was not apparent except in subsquares sub- ject to high amounts of fresh water runoff. In these sub- squares positive coefficients resulted and the cross-plots have an exponential-like character. This pattern is illus- trated in Figure 13 and was not unexpected considering the high amounts of terrigenous suspensions that can result with fresh water runoff. However, in deep water subsquares away from coastal influence no consistent pattern or correla- tion was apparent. An example of a cross -plot for a deep water subsquare is provided in Figure 14. e. Sigma-t Due to inconsistencies in salinity patterns and coefficients no single general correlation was noted between sigma-t and Secchi depth measurements. However, the same exponential-like pattern exists for subsquares subject to fresh water runoff. Examples of patterns resulting from fresh water runoff and deep water subsquares are given in Figures 15 and 16 respectively. 32 £. Oxygen In all but five subsquares oxygen exhibited negative correlation. The five exceptions resulted in slightly positive correlations and occurred in both deep and shallow water subsquares. Negative coefficients were expected due to effects of fresh water runoff and photo - synthetic activity. Examples of both shallow and deep water subsquares are illustrated in Figures 17 and 18 respectively, g. Silicate No consistent patterns were noted between sili- cate and Secchi depth except in subsquares subject to fresh water runoff. ' An exponential -like pattern resulted in these subsquares and is illustrated in Figure 19. Brown (1973) also found a similar pattern existing in the vicinity of the Columbia River discharge along the Northwestern coast of the United States. 2 . Open Ocean Areas Open ocean areas were selected for further analysis to determine if the trends noticed in the Japanese and Korean waters held elsewhere. The areas are shown in Figures 1A and IB and their boundaries are given in Table V. Special attention was given to trends resulting for deep water subsquares. Summaries of data distribution, parameter mean values, and resulting linear correlation coefficients are tabulated in Tables VI, VII, and VIII respectively. Graphs of correlation coefficients for most of the selected areas are plotted in Figures 20A through 20C. Results for the Atlantic and Pacific Oceans will be discussed separately 33 a. Atlantic Ocean For all areas selected in the Atlantic Ocean bottom depth and temperature resulted in weak positive cor- relation coefficients. A strong dependence of Secchi depth on temperature is again displayed for the open Atlantic waters. Forel color and oxygen also exhibited the negative correlations "observed for the Japanese and Korean waters. However, salinity and sigma-t exhibited strong positive and negative correlations, respectively, in areas located north of 20 degrees south latitude, whereas in the Japanese and Korean waters much variability was found. The correlation coefficient graph (Figure 20A) for areas located between 60 degrees north and 20 degrees south latitude was of particular interest. Because of the consistencies in correlation coefficients for several para- meters it was felt that a fairly reliable relationship between Secchi depth and other simultaneously measured parameters might result from further analysis of this area. b. Pacific Ocean Unfortunately, data for analysis in the mid- Pacific Ocean were very limited in number. However, five areas in the western Pacific and one area in the eastern Pacific (Figure IB) were selected for study. Resulting correlation coefficients were highly variable for the areas analyzed, as can be seen from the correlation coefficient graphs in Figures 20B and 20C. Forel color again exhibited negative correlations except for 34 Area 9. The exception resulted in no correlation due to a standard deviation of zero in color. Bottom depth and tem- perature normally led to positive correlations. However, bottom depth correlated -negatively in Area 6. This is believed to be the result of data from stations located in the vicinity of the Mariana Trench. Although great depths do . occur at this location, low transparency may have resulted due to runoff from the nearby islands. Eastward of the trench, shallower waters and higher transparencies could be expected. B. LINEAR CORRELATION COEFFICIENTS USING MEAN CHEMISTRY VALUES Summaries of parameter mean values for seawater chemistry and resulting linear correlation coefficients are tabulated in Tables IX and X for the Japanese and Korean waters shown in Figures 3A and 3B. Similar summaries are also tabulated in Tables XI and XII for the selected open ocean areas indicated in Figures 1A and IB. The use of mean values of parameters averaged over the Secchi depth accord- ing to the procedures previously outlined did not result in significant improvements in correlation coefficients over those based on surface values only. C. TIME SERIES ANALYSIS Marine pollution has resulted in long term changes in certain chemical parameters obtained along polluted coastal waters and in shallow seas. For example, dissolved oxygen content below the halocline has decreased during 35 recent decades while phosphate concentrations have been steadily increasing during the past six decades in the Baltic Sea (Fonselius, 1970). Time series analyses were attempted for the Baltic and Red Seas to find possible long term trends in average monthly and yearly Secchi measurements as a result of increased pollution. Secchi data from N0F)C were compared to data col- lected in the Red Sea by Luksch (1901) . Long term trends were not apparent for either of these areas based on the available data. Time series analyses were further utilized to study the relationship between Secchi depth and upwelling index near the Oregon coast and for Monterey Bay. Results from these analyses are discussed in the following sections. 1 . Oregon Coast Unfortunately, insufficient NODC data were available for time series analysis for most of the areas for which upwelling indices were available. The only exception was in the vicinity of 45°N x 125°W near the Oregon coast (Figure 21) . Sufficient Secchi data were available for analysis at this location for the years 1961 and 1962. Three-month running means of Secchi depth and monthly upwelling indices are plotted in Figures 22 and 23 for years 1961 and 1962 respectively. An inverse relation between the two parameters was evident with a possible phase lag of mean Secchi depth compared to monthly upwelling indices. In Figure 22 the latter are seen to peak more than two months 36 before a minimum in the Secchi curve is reached, while in Figure 23 for the following year such a phase lag is not evident. The results were expected since upwelling intro- duces large quantities of nutrients to the euphotic zone and is thus conducive to high organic production, which in turn leads to decreased Secchi depths. Bakun's (1973) calculations show that near the Oregon coast -upwelling is both less per- sistent and less intense than off the California coast to the south, where offshore Ekman transport is present throughout most of the year. In Oregon waters summer upwelling is seen to accompany the change in wind pattern from southwesterly in winter to northerly in summer. During 1961 and 1962 upwel- ling was strongest in July with values of 51 and 107, respectively, while yearly averages were -34 and -6. Upwelling values remained nearly constant throughout the summer of 1961. On the other hand, a rapid increase in up- welling was observed for July 1962, with a rapid decrease in the following months. Anderson (1964) studied the seasonal and geographic distribution of primary productivity of the Washington and Oregon coasts as evidenced by data collected on 14 cruises conducted from January 1961 to June 1962. He observed a spring bloom of phytoplankton during May and a smaller autumn bloom in August 1961. However, a close inspection of seasonal and horizontal contours of primary productivity * 3 The units associated with the upwelling 'index are m /s/ 100m. See p. 27 above. 37 revealed a steady increase from May through August near 45°N x 125°W off the Oregon coast. Anderson also found stimula- tion of production by coastal upwelling to be especially evident in August. Figures 22 shows a rapid decrease in mean Secchi depth occurring from February through May and a smaller decrease from June through October. The rapid decrease is attributed to the spring bloom of phytoplankton observed by Anderson. However, a minimum in mean Secchi depth did not occur until October. Anderson noted that difficulty was encountered with the productivity measurements during his September-October cruise with too few values to contour adequately. Nevertheless, he did observe that the influence of coastal upwelling appeared to advance westward as the summer progressed with a maximum westward extent occurring in October. A minimum value of mean Secchi depth was observed for May 1962 (Figure 23) , again corresponding to a spring bloom of phytoplankton observed by Anderson. The increase in Secchi depth observed for the remainder of the year possibly may be attributed to the rapid decrease in upwelling indices after July. Figures 24 and 25 are plots of Secchi depth versus upwelling index for years 1961 and 1962 respectively. Figure 26 gives a combined plot using data from both years. Regression equations and corresponding curves are also provided in the figures for each year, with a dashed 38 regression curve shown in Figures 22 through 25 and based on the two-year combined data. Tahle XIII is a tabulation of multiple regression equations of the form Z = Z (U ,U,U ,U ) , where Z represents Secchi depth and U represents upwelling index which resulted from the Oregon coast study. These were generated by using a stepwise regression subroutine available in the B I OMFD program (Dixon, 1973). Several transgenera- tions of upwelling index were performed in constructing the equations, and at each step the transgenerated. form which made the greatest reduction in the error sum-of -squares was added to the regression equation. At each step in the pro- cedure the multiple correlation coefficients served as an indication of how well the regression equations fit the data. The higher Secchi values occurring during 1961 compared with 1962 were probably the result of a somewhat lower yearly average in upwelling index. The regression curve for 1961 shown in Figure 24 approached a maximum Secchi value with decreasing upwelling indices. However, upwelling did not appear to be strong enough to result in a normal seasonal maximum productivity and a resulting minimum Secchi value as low as is usually found. Both a maximum and a minimum were approached by the regression curve for 1962 shown in Figure 25. This was also apparent in the regression curve for the combined data (Figure 26) with maximum and minimum calculated values of 19.3 and 8.3 meters respectively. 39 2 . Monterey Bay a. Relationship Between Secchi Depth and Upwelling Index Secchi data were available for years 1968 through 1973 from cruises conducted by Hopkins Marine Station of Stanford University on Monterey Bay. A preliminary inves- tigation revealed an abundance of Secchi data obtained from Hopkins CalCOFI station 3 for the four years 1970-1973 and from CalCOFI station 4 for year 1971. For this reason, and due to the distance separating the stations and the surrounding coast, stations 3 and 4 and years 1970-1973 were selected for analysis. The locations of the stations are shown in Figure 27. Unfortunately upwelling indices were not available for Monterey Bay. The indices used were calculated for the point 36°N x 122°W (Figure 21) which is approximately 52 nautical miles south of CalCOFI station 4. In contrast to the Oregon coast Monterey Bay is a region of strong upwelling during much of the year. Peak upwelling values ranged from a low of 221 for June 1971 to a high of 297 for April 1970 with intermediate peak values during June 1972 and July 1973. Yearly averages were also high, the average index for the 4 -year period being 116. Mean monthly Secchi depths for CalCOFI station 3 and monthly upwelling indices are plotted in Figures 28 through 31, corresponding to years 1970 through 1973, respec- tively. Figure 32 is a similar plot for CalCOFI station 4 for 1971, and Figure 33 provides a quarterly plot for CalCOFI station 3 for 1970 through 1972. An inverse relation between 40 mean Secchi depth and upwelling index was observed in all cases. A phase lag of from one to two months in mean monthly Secchi depth was observed at station 3 for years 1970 through 1972. However, such a phase lag between minimum Secchi depth and maximum upwelling index was not observed for 1973 at CalCOFI station 3 or at CalCOFI station 4 for 1971. Plots of mean Secchi depth versus upwelling index were again constructed for the Monterey Bay study. These appear in Figures 34 through 37 for CalCOFI station 3 corresponding to years 1970 through 1973 respectively. Figure 38 is a plot of 1971 data for CalCOFI station 4 and Figure 39 gives a combined plot of all data used in the Monterey Bay study. Regression equations and corresponding curves are again provided in the figures with a dashed curve (Figures 34-38) representing a best fit to the combined Monterey Bay data. A dashed curve representing a best fit to the combined Oregon and Monterey data is given in a quarterly plot (Figure 33). A tabulation of multiple regres- sion equations and multiple correlation coefficients is provided in Table XIV. The regression curve for 1970 approached a minimum in Secchi depth with increased values of upwelling. Downwelling apparently was not sufficient for a maximum in Secchi depth to be approached by the regression curves for both 1970 and 1971 (stations 3 and 4) . Higher Secchi values occurred in 1972 as a result of a lower yearly average in upwelling index, and a maximum Secchi depth was approached by the 1972 regression curve. However, more scattering of 41 data points and a significantly lower multiple correlation coefficient resulted in the 1972 analysis compared to pre- vious years. Scattering was even more pronounced in 1973, resulting in lower multiple correlation coefficients. The regression curve resulting from the combined Monterey Bay data is shown in Figure 39. A minimum in Secchi depth was approached by the curve beginning at an up- welling value of approximately 200. On the other band, down- welling was not sufficient for the Secchi curve to approach a maximum Secchi value for the overall Monterey Bay study. A plot of mean Secchi depth versus upwelling index for all data used in both the Oregon coast and Monterey Bay studies is given in Figure 40. Both a maximum Secchi value of 18.7 and a minimum of 9.3 meters resulted from the overall regression curve, corresponding to low and high upwelling values of approximately -200 and 200 respectively. b. Relationship Between Phytoplankton Wet Volume and Upwelling Index Plankton hauls were conducted by Hopkins Marine Station during cruises made on Monterey Bay from years 1956 through 1967. Wet settled volumes of net phytoplankton were then measured, and monthly averages were tabulated using data taken at the six standard stations illustrated in Figure 27. The data were obtained from Hopkins, and a plot of monthly phytoplankton wet volume in milliliters against upwelling index was constructed and is shown in Figure 41. Unfortunately, plankton volumes and Secchi data were not 42 measured simultaneously, and a direct comparison between the measurements could not be performed. Although there Avas considerable scatter, especial- ly for high values in wet volumes, in general a direct pro- portionality existed. It is speculated that the scatter may be due to errors resulting from the technique used in the wet volume measurement. When the measurement is performed, complete settling does not always occur to produce a distinct boundary between the plankton and the liquid above. The incomplete settling may be the result of electrical charges existing in the plankton and lead to values in the measurement higher than would otherwise be obtained. Because of the scatter no attempt was made to establish a regression equa- tion between the wet volume and upwelling index. 43 V. SUMMARY AND CONCLUSIONS Secchi depth reading's are influenced by many sea water parameters. Although linear correlation coefficients cannot determine the exact nature of these relationships, they do provide an indication of general trends. Forel color, oxy- gen, and water temperature appear to be the most consistent in their linear correlations with Secchi depth in both coastal and open ocean waters. Forel color exhibited trends tov/ard an inverse proportionality with Secchi depth as previously indicated by Visser (1967) . Oxygen measurements also exhibited trends toward an inverse proportionality with Secchi depth while temperature data indicated a possible direct proportionality. However, much variability was encountered in correlation coefficient values for coastal waters. In shallow coastal water areas subject to high amounts of fresh water runoff bottom depth data indicated a direct proportionality with Secchi depth, and salinity and sigma-t exhibited positive correlations with exponential-like patterns when plotted against Secchi depth. Silicate cor- related negatively and also resulted in plots of an exponen- tial-like character. Such special trends were not apparent in deep water not subject to coastal influences. No consistencies in linear correlation coefficients between Secchi depth and latitude, longitude, cloud cover, 4 4 and month of year were found, and the scatter resulting from cross-plots of these parameters did not indicate possible consistencies in correlations of higher order. Nitrate, nitrite, phosphate, and phosphorus data were too sparse to allow representative analyses. Linear correlation results from the Atlantic open ocean .waters, including the region between 20 degrees south and 60 degrees north latitude, indicate a high degree of consistency in coefficients. Sea surface chemistry values appear to be valid in correlation analysis based on the present study. The use of mean chemistry values of parameters averaged over the Secchi depth did not indicate significant differences beyond the use of sea surface values in linear correlations. Although marine pollution in the Baltic Sea has resulted in long term trends of certain chemical parameters, especially evident below the halocline, Secchi values there do not appear to be significantly altered by such effects. Nor were long term trends evident for the Red Sea, an area for which Secchi data also span about seventy years. An inverse variation between mean monthly Secchi depth and upwelling index was observed for coastal waters along the Oregon coast and for Monterey Bay. A "phase lag" was observed between the time at which minimum Secchi depth occurred and that at which maximum upwelling occurred. Upwelling indices may be a valuable aid in the predic- tion of transparencies in coastal waters and in locating 45 areas of high biological production and potential fishing grounds. Although the regression equations resulting from the Oregon coast and Monterey Bay studies will not provide absolute values of Secchi measurements, a fairly reliable estimate should result when used in those areas studied, as indicated by high correlation coefficients. The regression -7 3 equation, Z =13.85 - . 04 U + 3.13 x 10 U , with a mul- tiple correlation coefficient of .80 resulted from the Oregon coast data, where Z is Secchi depth and U is upwelling -7 3 index. A similar equation, Z =14.01 - .03 U + 1.80 x 10 U , n ' s with a correlation coefficient of .72 was derived using com- bined data from both the Oregon coast area and Monterey Bay. An inverse trend between phytoplankton wet volume and upwelling index occurred for data obtained for Monterey Bay. However, considerable scatter was observed, possibly result- ing from the technique employed in the wet volume measurement. 46 VI. PROPOSED FUTURE RESEARCH Work should continue to provide better world-wide cover- age of Secchi disc measurements with the other oceanographic parameters normally sampled. To ensure this, a program should be established to provide better dissemination of oceanographic data - especially older data - from the various existing oceanographic institutions to NODC . Studies relating Secchi disc measurements to other ocean parameters such as the diffuse attenuation coefficient should be continued to check further the empirical relationships that have been formulated and their spacial variability. Both disc diameter and reflectance should be standardized or specified. Investigations should also be conducted to determine quantitatively the effects of varying sun altitude and wire angle on data taken with the Secchi disc. As more Secchi data become available, time series analyses should be performed in shallow seas and coastal regions to determine any long term effects in water trans- parency due to the various forms of pollution. Further research should be conducted in the Atlantic open ocean areas to study consistencies between the linear correlations between Secchi depth and other simultaneously measured parameters for the different regions. Additional study should be devoted to the various upwelling regions of the world oceans to determine relationships 47 between Secchi measurements and upwelling indices to the end that standing crops may be predicted directly from the latter. 48 o •H GO KA O CM o O P4 CM O P I Oj E fc«0 •H CO 00 Pi B CO EH E ra o A p p P Pi O CO Pn P o r-H O O ■h ro o p O Pi 0) 0) 00 P) a1 CO I rO Pi 00 00 « p, cd vo KA VO VO in CM LTv CM 00 CM CM r- VD VO CT> co in o o in ocn-d-r-r-ino t- v- i- tn CM CM CM VO O «*• to» vo o VO o O tn CM vo in in ^ — tn VO tn CM tn VO VO CM vo CM CTi CM CM tn CM t — tn v- t- in t- tn r- tn t- cinmcrvcMCOCOC— O in k> tn r~- en •*- tn tn -3- cm U~\ tn a\ 00 CO r- 00 t> en tn o CM i— i \ — CM vo CM r~ Cn * — VO tn co tn o T— * — CM CM in CO CM CO -si- CM O tn tn CO o CM CO en 00 tn CM O vO in 1 — CM Cn CO x — tn CM VO o CM vO ^ — CM o CM CM CM Cn CM CM o o CM CO en CM CM O 00 CO VO O cn c- CTi o o CM o r— CM Cn <^- " CM CM CM CM CM CM CM O i — in tn tn CO CT» O T~ t — CM CM tn tn 49 o •H CO c*- m C\J CM m CM CTi f- in r~- oo CM 00 CM IO o in 00 in CM T— to LPi ■sf- T— to CM r* r- to 00 to o in CM en CM o 525 O CM O C\J 00 in to O ON ■* to o CM o c- CM VO 00 CM 00 in t — f- en i — to in O CM -st- O O r- CM m o s — CM <[ — VO CM 00 CM VD CO en in in vo to O tO VO CM CM en co +> I Ctj B •H CO • I— I ni co ft B CD EH [>- "<* to -* VO LO T- en t — o f- «* ITN VO CM o VD c- 00 in i* CM VO vo CT> en en ■*— r- O CM [O CM CM to CM in CM LT\ t— T- LP\ T- o to t> "Sf »*■ «*■ r- to v- o T- i— CO **■ in en VO o VO r- 00 LO "* CM tr- VO CT> en cn T— r- O CM to CM CM to CM io CM in r~ **"" in to O P- 00 •3- o LT\ VD to 'v— ^_ CM p- VD T_ VO r- o VD co CM VO «* CM o VO CH Cn cr> vo en ■«*■ CM to T~ to CM to CM VD C\J in T~ 1 — in to to o S co OA -P +> +» ft O CD to t- •t— r- O r- CO *=d- in o 00 en CM ^ — c- in ^ — VD CO in to CM ^ — •* en 00 00 VD vo ■*— r- ^ T- T- T— r- "* u o rH O in O to en ^ — O T- VO P- ■ O ft CO 0) co p c— CO in o VD VX» to CM ^ — to CO VD •* — cr> T— o V£i CO CM V£> "* CM O VD CT\ CTi 0> VO en in C\J to V- to CM to CM *£> CM in T_ -I_ in to to O co I CO r- o to ^h m VD c- o x^- in VD 00 cn in en CM to to to to to to vd- ■* ■> o p ,0 ■H >H -P (0 ■H O -p «5 P I ctf E fcuO •H CO ctf CO e p p p p< O 0) O O o •H W ,d ^! O P '-0Q CT1 co J CO CT1 CO * ctf s cm CM CM to, VO LO CO VO CM CM CM en VO LTN to CO lo CM VO CO lO CM * — to VO CM CM CM CO VO tO CO en VO to CM VO CM vo o CM to VO to VO to to to CM CM LO tT> CO to to CM CM LTv VO O ** — o o CM CO o to i — VO CM CM CM O to en T— CO VO O VO CO to to VO ^ — 00 CM VO to CM to CM CM VO 00 o LT\ c^ c- o CO CO en to o T — CM to LT\ T — to CM O CM to VO VO O CO to LO LO CM en to to t — VO CM CO CM VO CO T — VO CO CO CM o CO to LO Cn to to r-- o o to CM CM VO vo LTv Cn to LO to LO VO to LO CM CO CM CM Cn CO to VO Cn co CM CM CM to O CO LO -3- 1 — o CM to VO CM C-- 00 to o CM to VO to CM T~ o Cn CM CM ^ VO CO co CM VO o o CT< Cn o CO CM T— CM ■5— to T- CM LO -=d- C^ to c- 00 LO LTv T— CM CM T — to c- cn *=f D- VO *vf VO 00 Cn CM r— to C~- 00 ^1- c- VO cn C- CM T— T— CM CM C~- t- LO -3- LO LO to Cn to Cn CM [O CO CM c— n — p- o to •H to CM CM T- CM r- ^ — CM to r_ CO T~ to O J25 C\J O CO to C- r~ to P- e'- ^X) LO -* r— fe T_ \ — CM T_ " o «* to ^t" co r~- MD t — en VD to CO LO to Ph •"4- CM T— CM v~ ^ — CM to T — CM CM "3- v£> ^ — cn Cn -3- CM CO LO CO c- -3- O (M C- CM T— CM to " CM CM to P ; 1 CM to O cn *d" p- to T- cn CO O to CM r~ to e **3- v£> f~ CM CTi LO to v£> CM o CM o CO 00 to W) ^ — cn CM T— sfr CM CM to ^ — to CM T — "=J- ^~ •H co • 3 CM ■<*• O CM sj" P- to ,_ Cn CO O to CM c- St" Cfl -3- VD C— •* — Cn LO to ^D CM o CM o CO co CM CO i — CTi CM "«*■ CM CM to ^ — to CM ,— ^4- • PM cn C— ^ — o P- t— CM CO o Sf" LO LO \ — xd- CJ s ^ CO cn ■«* to P- LO Cn sj- • v£> to sf LO to CM P- o VO en o ^d- to •p ft *3- VO to 1— P- ^ — O VC T — cn cn to r- co LO O CD ^ — VO CM T- to CM CM V- CM ■<— m r Pi o o CO VD s}" O \o CM "=}" to O IT- KD to cn CM rH to 00 to T — CO T- CO o T- en CO c~- MD r- LO O ^ — Cn CM \ — CM CM t— to v— CM T— to O •H CO o -P Cn CO T_ o p- T_ CM CD o «d- LO LO T- •^i- to O PL, "vf CO cn xtf" to c— LO Cn >=d- >=t ■<* T- o o to 0) CD cn CM ■5— LO CM CM to T— to CM T- LO v— s— CO PI • ex1 CO 1 CO cm LO MD cn LO VO cn to •vi- LO CO cn T — CO ,o -=J- ^ LO LO LO \D VO V£> r~ p- p- c— r- CO 00 2 CO • o< CO • Pi CM CM CM CM CM CM CM CM CM CM CM . CM CM CM CM a3 to to to to to to to to to to to to to to to 52 *=»- O VO CO to •H ^ — CO NA i O S CM o !2S -tf o vO CO to P4 T- Q> u a p* cm VO co to XT' o T- CO 1 i rQ ?i -p CO 1 cd r- 1 — CO £ E Lf\ CO c- 0> t*D 1— to •V •H CO co U i cd • r- 00 a LTv to t- >> CO v- tO £> d « o p< CM o to •H S VO V£> CO •p CD ^ — tO 2 EH rO •rH M S CO P O^ CO -p p V0 O to •H -p ft cn P O CD PPP 03 ■P R5 rH P O t> CO CTi O rQ en CTi en o 2 CO • M • c^ CD CO rH • £> rH C\J CM CM cd cd tO to to EH S v— % — T- 53 V) CD U, rt if) 3 CD O to CO CD ' ^ ^ 4-> 0 e. CO CD !h £ oj CD (X 10 OS or, C O •H P t/l rt a? > CD — ■ CM- CM- — -vt" CSS CM w CM- CO CO U3 VD tOO tO CM M0 CTi tOCT\ M0 CO tO- M0 CO v- CM M0 00 ^=J- t- C0 " CO C- r- O t— CO CO LO ■vfCO tOMO CM en -d- CM Lf\ CTi LO- — LfY — LO- — ^W MO t- LO- — LO- — M0 MO — ' MO — ' O to 00 CM CO M0 Si" LO t- tO to •<- LOLO CO CM O %i CTiO C— LT\ U0 CM t- CO 00 ^ CJl CNl CM O O 00 ■^tO CM- tOCTi LO tO CTvCn 1- tO r- lo \s\ to cm r— CO LO tovo o \o «^-to CO O O tot— O CO md r~ LO'd- T- O to to T- to CM--' CO "* CH'vt- CM ■* M3 «=3- t-^ LOLO LOLO r— kd tO M0 MO CTi O O e o o e CTi CTN ■<- en t— i- cno C- LTv CM tO C"-^- ■p- CT\ LO tO O T- t- LO tO-r- OMO ^}-M0 toto torj- CM v- ^J- T- tO v- tOt- •^-T- tO T- to^ -d- t- ■=i--r- -=d" T- CM cn O to 5 4 V CD co p; C co g3 }-i aj ^ p; 0 -H •P a pi rt O rt +-> P-, rt •H > ^° 5 rt £ rt Op ow 0) -P *h- rt o » rt EH -p to rt O I S2J tjC ft iH +3 cm rt o 55 I r-i P> •^ rt o L CM\ O H rt e M •H CO £ o •H O r-H\ rto co- — p^-^ E o ao EH—' E^ o -P^-> -P Ph E P CD — ' oo pq u o H o o I • rO CT1 pS CO CO fOr- >s)->«* «sf-<* KMO c— cn Cn r— O CO cno VO f- CM CO vr> to ^ -tt oo t- 00 MO C— MD O T- t— CM cn-=t 00 LO CM fO CO LO t- CM "*t LT\ r— cn CM CM CM MD T- O -=t- CM o o coo O T- r-r- o o O CO CO LO CM CM CM CO O LOMD t- CM 00 o *- cn T- Mr- CO' CM- CM CM' CO CM- CO- CM- CO- CM CO' CM- CO' CM • tOCM CM- CO Cr- tO vO LOCO t- co CM Lf\ V£> CM CM CO -3-io C"-CT\ ^j-to cm O vd-CM CM CO LOLO CM *vt" O 00 CO CO- tO CO CO CO tO' co- tO' CO- tO' to- ^-oo o lo "=d-r- oo to to CM-—' ^ CM CM — ' to to CM'—' tONf CM--' CM CO • e to [O CM- — v- cn • • to to CM-—' to to « • cntn CM CM CO-3- CM-— ' CM i- • t CM "3" CM^-- in en envo CM r— ^ en oo in cm co me— ovd LT\ t- CM t- o o ■tf- 00 to t- O to t- KD OvD Oh tohn r— --' LO— ' *3-w LO- — ' -=1-—' lO^-' ■^•s^ LTW-' LO- into CM ^ cT>r— cn-^t- ■^}- CM xH-CM i- r- LO^l- tO-r- ■>j-to o r- LOCM r— lo co cn cn t^. o o enco VD 00 o c^ 4j- LO CM t- tO CM- — CM CM CO t- CM tO- cm hn CM- — ■ LO tO O v- CM KC tor- LOCO CM i- ^ T- tO t- to LO r- O to «*f to CM CM CM to to to to to 55 CD Rj o co i CO 0 "ti [fl , i-t in pj CD 0 ^^ ,Q -P P! (/) (D S H O P* ■H c pj M p > ^ a ^T3 o +_, H H rH •si- -P O nj ■H I Pi -p to cd o I -p o I -p •st- cd O I CM\ O rH B -P I rt E •H CO £ O •H O r-H\ rto CO PV^ e o a>o LH — ' E ^ o -p-— > •P Ph E •P 0-— OR PP M o rH O o I • 2 CO CO a? co oo c\i en to co o \o to v- 00 ^ 00 o -si- C — OvJ -si" *4- r- oo r-T- oo lo T- LO vO -si- en-sj- en LO OMA tO C\J CTiC-~ t- O enc— o o CT\KD o o envo o in c\j t- lo en C\J r- VJ3 v- CM C\J -=t O VO fO CM CM T- t- tOOO CM CM 00 CM CM CM o ^ mo c- o toto CM t- \S\KD en ^t- KD CM C*- en enc- cn LO t- CM •st"LO lo to VO 00 LTV LO LO- LO- LO LO- LO- LTV LO ■ CO LO O to •si- 03 KD tO cn-st- C- CO CM v- tor- cno co cm enr-- co co LOtO CO o i- CM [O CM enio CM to- CM- to- CM- CM CM CM > — O "sh tO-r- CM -^ CM — to CM- 'si-- CM to cm- co ^ r— cm cno "=f«vt- i- CM lAr- t- cm r~ to O en c-o CM CM t-. r-o t- CM O 00 "=i--r- cno t- t- • e e * ■vt-CO CM O tor-- r- to ^— " -s— to ■sj- cm cnto ^ to cn-r- \o cm en co LT\ v- v- -=i- C— MD fO*3- tr-co CMv— - c*-co CM^—' t- co cn-r- to-— - r- cm cm en to-— • O en • * to^— ■ CM CM • e tO T- cm r~- IAt- 00 •* LOVO kd c~- en lo t- t- tO CM KD CM • • lO' — en en to CM CM' — envo -st-vo CM t- '=i-'=i- O VD co en LOCO CO T~ T- tO LP>r>- CM VO toco T- O CM V£> CM t- cm r-- cn cm r— io 00 CO to «=i- — to tO-—' to t- CM tOv-^ CM' — tOT- tO— ' to to o *4- to-^ CM CM tO*-' O O tO T- enr— VD \D LOCM CM tO o o to en CM t- cm en CM — CM -st CM'— • en-si- i — ' — * 00 LO ^ — v— ' CO VD CO LO T — ** CM "=tf- CM'-' O to CM'-^ 00 VD ^ — ^s CO -^ t — N— ■* ■st- CM • • CM KD CO V£> t- CM enr- o « f- -si- MD to to o ■st" ■sl- to ■vt- U5 "St" CO ■si- en LO LO en LO to to to to to to to to to to 56 V) 13 Cfl t U v\ ctf CD CO 10 CD p: ^ cc5 hs (D Pu +J in pj o ■H CD e cd PL| > CD ^G DW M cd Eh O cd •H I co W pi to cd O I fH -P OJ cd O I H P -3- cd O I 3 CM" O I cd E CO Pi •H cdo co — E o cdo EH'-' 6^ o P'-' P ft E P CD--' on pq fH o o I • 7i CO CO fn cr1 ci co >x> ■>=*■ tOuO CO o to cm ttf- CM to cm \o to O 00 CO t- ^-O CO tr\ T- C\J — ' U0 C- ■*t CM U3 O VD C\J t- tOCM to co CM v- CM OJ VD -* ^ O to MO LT\ CO CM to mo 00 to Cn r- toto CM ■**• tn t- LT\ t- tO LT\ tO CM CM (T\ 00 CM O O O to LTV- LPv- LT\- in' ITV LT\' LT\- UTV v- r- t- O LTWO CT.00 LT\ tO inr- ^t-cn O v- M0 O in CM tOin O r- 1AC- C-O cn <=* cm o mCTi CM'-' tO CM' -3" i- CM— ' tO ,— CM — ' KMA CM"-'' CM — ' CM • CM CM" tO' CM' IAIA cn^i- 00-^- UD O LfAO N"\*3- CM UD CM >=^- md •<- cn in m -^- cm t- cm cn r-oo •st CM CM C— cm \s\ CM t- CM VD • • IAC0 tO CM o • C~C7> »X) CM -vi-O tor- • e tO'-' O mo ^c-^v — ro^-- OMO CM v- C0 r- tO r— tO—' r>- cm ^^^T- cr\c- CM'— - totr- tO' — to in to O to to 00 r- to oo c» to cn cn to cn O CM to cn CM to t- o CM O ■*nt cn cm T~ O T- O P- to cn <=^ 1- -vi- m CM tO in co tO' CM - into r- t- >x> vo >vt-MD enen toco t- cn c-o -=d--^ oo cn co r- r-v£) COM3 CM VD C- CM oo ^t- t— to cm r- O CO co cn tO— ' to> — to- — tOt- CM tO to— ' tO T- ^j-— ' ■stf-w tO—' to to to tO'-' tO—' to to^ to to to tOi- v- C— enc— in oo O to ^t-co IOt- -vj-in c- cn CM -^- • « * e • « » « • • e • © • • • • • • • c- in cnin \£> in CTN'vt- C- VD c~- in inMD •^ to T~ ■<■ T- "vf KO ^t MD CO VD O v- — too tO v- 00 CM CM to 57 m C 5 co I CO Pi CD t3 en . f-< to cti CD S w CD [/: CD Pi i- a oj CD (X, r<~- Pi CD 4-> Cfl Pi O •H 4-> cti cti (X • > CD CD'CI 3 ?- ■HI) -P Pi Pi rt o *-* o to CD rH cti EH *=h +» O cti •H I CO fat 3- H ■P o I S3 WJ ^-v -p C\J cti o i (3 M 3- — ' -p -=d- Cti O I Ph M H O H a •p i a! e no CO Pi o •H O H\ CtiO CO- — E o CDO E-l— ' 5 ^ O -P"-> ■PftE -P CD — OR pq o rH O O ,0 D< 3 CO CO cti to o en [— tO CM to ^»- LO "Si- T- cm r- 1- CM tO^ 00 CM T" CM tO -^ t— t— >x> •*tt~- to Of- • e CO C— « < -=f^0 LOCT> CT»-=J- ^00 • o CTvtO » e -=!- CO i=i-i=i- i=i- "=}- Mio i— cti cm to i- co cj\ "=j- to CO "=i- 03 CM tO O T" C— to MD V£> CO—' t— CM i- •=}- f- cm t- ro O -=t CM t— •*to \D tO MD i- i- CT> co en tO—' O to "=1- 1- cnto CM CVi •sfr 00 to— ^ V£J T- CM VD CM -=i-T- cn to- O CT\ LO i- O VD t— i- tOO tO CM LOCO CM -if i- to LOO i- to CM ^ CO cti 1=1- 1— too toco T- i=l- CM 00 CM t— tO CM i=f- t- CM -^ CM i=l- to U3 O • • tOi- IO— ' CM MD OMO e e • • -vt-CO I— >=t- too » ft LOMD i- to LO CT» CM 00 to cn to i=l- -=t LO CO <7\ 1=1- LO LO IO cn LO CM to CM to CM to CM to CM to CM to CM to CM to CM to CM to 58 ^— V -P ^-^ . — v *■ — \ * -s /"^ ^-N ^-N , v sf- a3 mst" CO CM lp\ m r- m O sf cnro r-r— V£» vO % — T" ^i ct ^ — t — t- CM *~ *~ *~ T~ ^ — 1 — *~ T~ *" ,_ T— ^-^ ^-^ ■p m rt O 1 H Is; W " 5. •p _ ^^ cm a) CO ^D OMD0 r*ACTi <£> si- KNtn UD CM vo in O 1 H O O o o •<- o O O t— \— o o O O K W e • • • • • a * — ' v-~-' s— + s--^ ■ — ' s-s* 1 — -* ^_^ ■p •~-N /—N S~ >V •— ^ , — n ^^v *^^ •— 'X st- aj r-^- KO ^f CO in GOVT) LPvO ltm~- C^-O tno V) O l rH t^OJ ■^r^ C\J r- meg rncM Sl-CM st- st- CO t- CD PM tiC • • i-i o 2 v*"*' N v~' * "" w _ ^^^ _ _ ^^ _ _ ^_^ i rH m-^- O>00 •«*•*- CM LO X) co m r~ cm cti t>0 mcv t<"\CM -«drv- CM CM h<> — CM^— ' CM^— ' CM^— ' CM--' CM CM w r>-r£ CO rC 4-> ^ 0 CD £ O KMPv CO CTv t- VD lf\KD cr\ in CO "* ^ ir\ VX> KN ^ — O c-A •H O o si- LT\ sf- co r- LT\LTv VD f~ ^<-^c^l ^CO r~MD (T\ CM ^c^o Cti p. , ctio ^ — t — tO»v^ ^_ , •v— * — ' t — s — ** rrvv-^ fO>*-- O ■s — m • 05 ^— ' t>A tA-^ to KA ^c^ lA m K^ m ^t- C0 <-V£> ITNVD sJ-CO ffMT- ■I-) cti cti Pu ■H E,G LO.VO i- cr> ^^ IAh ino m-vj- o o cnr- LTN T— CO m > o +»*-> e c t • • e * • 0 • • • ft ft • « a ■ ft ft CD ■PP.E St-|-- ir\ cj\ cm in o t- O v- KNCM cno mm CO MD r- co s — -. C -p Q> in-~ ' CM ■ — O en [— — ' r- 1- "=J-T- CTiO LT\r- sf ^— ' CT\ tA T* on 1~ — ' - — v^ \ — CM K> CM sf (Did m — ' - — ' -— x 3 f-l Pi cti •HT3 fn -P £ o ' K\ CM>JD CTiCTi VD t- sf-vf KO KD O IT\ co r- "t — CM CO CO G cti rH m . o 4-> O • ^ — VD CM N>^-^ ■^-t- ITv-r- t~- r- St- — (C\« in T— roi--' o co O st- — ' v ** N*~-' ' v — ' » H 1 • ,C1 O^ LT\ VD CT» ra sf LO CO K\ tn EH S 59 O n! •H I CO bL rH 5-4 cr en B CO a> ■ts sn 5-! to OS 0) S -^ Pi en tp p: f- CD PU Pi 5-. -H o Pi e o as -H a$ ai ■ > 03 T3 Pi U Pi as ■H-t3 +> Pi Pi 03 O 4-> o co t-i M V O rH s p I E no ■H CO Pi o •H O rH\ ceo PV— £ o cuo EH—' E ^ o +»'-» •PftE -p O)^ — on pq O rH O O I • fit cr1 p CO Pi cr1 aj co t--r- r- m r— in O CM -si" c\j CT>t- OJ CM LA "sh CM t- r- ^j- C— CM CO r- LTV men mo • © KN-^r- IOC— • e [Oy—' CO CM to tOLTi O C- M3' CO C- mt— ^r- ^v- CM — ' C\] — - ^-^ ^ (T>m 0\r- LT\ -3" -=*c- • * » • tO— ' to* — tO to CM O • « [OC-- o O CM CT\t- C\J C— t— LT\ -^ [O i— to U3 KD » © fO— ' CO CM tO CM -r- LTV iO^ m^o • • CM — ' -=1- ITS • • to T- CO tAin torn UA tO O MD CM— ' to- rn m CM to 60 -3- o •H CO tA O CM o £5 O CM O -P I aJ E •H CO Pi •H H c c— \o «* 1— o O T— CO v© LTv LA T — ■>* ^ — cn tA CM O o CM >sf c- o *~ O CM CM to v£> «3- to to LA 1 1 l i 1 1 1 1 1 1 • i 1 1 tA cn 00 to o LA cn CM CM Cn c- «sf CO o «tf- tA O 00 c- CM IA "3" CM t — CM T — LO ro CM r- CO CO OJ CM o CM CM O CM CM tO T~ •=*■ O LA •vf- "5* LA i 1 1 J 1 1 1 1 1 1 1 CO ,_ 'd- tO <*■ CM tA O tA CM Cn CM VO CO VD tA tA KO •t — CM tA UO r— r- ^D CM t — O r- tA VD 'li- CM T — tA CM o tA CM CM 1 to to 1 t CM i t — ! CM ch- CT\ 00 '*■ CM •«*■ o to O CM VD MA CM ■* — ,_ CM en ^J- MD o <* *4- 'd- to LA CO ">*■ LA '*■ T— CM CM T~ CM CM ^ — r" CM T — to CM "Sh CM LA P 4-> a o P PQ to *vj- o cn 1— CM CO r- to ■**■ tA CM cn r- to ^ — LA KO VD V£> ■<* CM tA VD LA CM t"- to LA O c- VD LA •=t M3 to "*■ o v- O CM tA o ^ — O u o r-i O O r- r— cn LA cn T- CM 00 -=f t— KD CO cn CM ■sf «* CM y£> 'sj- "St- vo CO f- tA "«*■ U) \r~ cn t— f- 00 V£) LA LA LA Nt" LA tA 'vi- tA CM tA T— VD CM "* [A CM C3^ CO I pi CO o O \ — CM O ■sf LA LA LA VO VD O 00 CM CO CM cn o o LA tA CM LA CM CM CO U S o o o o O o o o o o tA tA tA tA to fA tA tA tA tA tA tA tA tA tA tA 61 o tj- O -3" CM NX NX CO ^D r~- tf\ r- cr* LT\ o l^- CM CM o 1 *" •tf- i O 1 1 1 i I 1 1 1 1 NX V£> o V£> o T- • 1 o • O UX • Tj- CO CTi -^ in C^ NX LPi o ^ CO ^ — T — ox O CM NX a o CM O O CM vl- -* o o '- NX m CO o NX sf O 0 Ph 1 T3 1 1 1 I 1 l/l P o CM c- o <* CM "* CO o f- CM ^D NX 00 O -3" *=r" CM CTN UX vO CO t~- kO o ■t— NX CM CT\ •< — CTi vO CM 00 o ^ ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 V) P -p 1 UX UX NX o LT\ r- M3 -^1- T_ V£) cn CO cn r- VD PI K) UX M3 t— CO C- c- ^ — ox NX CO CO CO T— •i- o -=J- 0 s O O NX o *t LPl NX NX "=1" o CM o NX ■vf NX ux •H w U •H 1 1 1 1 1 1 1 1 •H CO |

CO UX UX CO CO ox o r- r— o NX T— CM VO O ■H LT\ r- t— v£> CM •* ux CM 00 NX CO CM OX O o t~- CJ rH (XI nx LTi NX t — CM CM ** ^d- VP •*- CM •* UX NX NX Pi O cd CO 1 1 1 1 1 •H P ft C— v£> UX CO ux UD CO T NX ■^ — -=d- OX "«*■ VD UX 00 CC S 0) C\J NX VD CM T— ^ — CO CT» I> ^ NX CM CM ■^ U5 NX rH 0) CM CM CM UX >X> CM t — o KO ^f o NX o CM UX P Pi EH i 1 1 1 o u p CD o -P >st "sj- CM "* CO NX c— CM ITS UX ^D o LT\ rO NX NX NX NX NX NX <5j" •vl- "5* ■«^- -* "* UX IT\ M3 MD 0 • 'J) M M V. M • Q) CO rH t 03 p a) ^ — NX NX NX nx NX NX NX NX NX NX NX N~x NX NX NX NX EH S 62 o •H CO 00 o o LP o CM CTi to, CM c*- LP tp C\J ^D O O CM o o to to ^ — o ^ — o O to, O CM O !23 O CM O B w •H CO ■H CO ft S H • i CO CM o 1 O o o o o o to, r— I o o o cn CM 1 CO CO 1 tP CM i — 1 i — 1 o C\J CM CM in O CP Lf\ o tp CM o CO CM LTv cn o T — r- CM M3 ^ — T — a* CM o CM cn cn O LP ^ — 1 1 l 1 1 1 1 ' i 1 1 1 to, LP LP CO CTi CM CM o LP CO CM o Cn to, cn LP LP tP CM M5 Cn CM o tp cn CM o VD tP CO LP, rp M3 cn LP, o LP, ^ — CO co 1 1 1 1 1 1 1 1 1 i i 1 1 i 1 1 c— LP cn o rp LP LP CM o lp CM O LP, LTv Cn to, O ^3- v£5 lp LP O CM CO LP, o CO o C- O (P M3 CM i 1 1 1 l 1 ' 1 i 1 1 t 1 1 CM CM tp CM LP tO O T — ■<3- CM r— CM CM O CTi rp t — CM LP CM ^ — o to, r- LP, T — CM CM CM o 1 1 I 1 l 1 1 1 1 1 to, MD -3" tP CM lp LP co CM to lp CM LTv tp CM CO M5 MD t — LP CO CM t — tO, to, o o -p -P P- +> (D O R pq o H O O CO CM to, CO M3 O LP o cn CO CM 1 CM CO >x> C\J to cn O 1 o LP to, o cn O CO LP CM VX) to, CM to, o o c- to, CM CM O to, LP LP cn LP M3 M5 00 r— >X5 LP o PJ CO cn CO LP cn CO en LP LP 1 1 i 1 i 1 1 l 1 1 1 1 I 1 l CO I CO o c- 00 00 cn en cn CO cn CO cn "* LP CO cn LP r- r~- r- CO cn o *, — CM CM to, to, •vf '* -* *d- LP cr1 CO • U cd S to, to, to. to. to, CM CM CM CM CM CM CM CM CM CM CM to, to, to, tP to, to to to, to, to to, 63 CD tn 03 3 cr co i C/^ PI CD (/) p. w PI CD •H O ■rl <+J m CD O u p o •H 4-> Oj rH (D Pi Pi O cu P •H -! 0) P5 Pi •H •P Pi O o 0) H cO o CT\ o o "* LCA cn o C— CM CO LP. f- o •H T- C\J en c-~ CO en CM r~ O o CM LP O CO t o CM CM o o in r- MA O «a- o T — v— O 00 M3 CM Ph C\J o 1 CM 1 CM • I • • » • 1 • 1 • 1 • 1 s»- cn LP, CM CM r- 00 IP. VO LP. LP, VD CO OJ MA C\J T— ■st p- CM LP. LP. CM CO OA \ — p- o VD CM VD LT\ ^ — O VO CM o CO LP. P- cn 1 1 1 1 I 1 1 1 1 1 1 l ■p 1 in t— c- vO -* -3" "vl- VO LP. cn VO cn CM MA VD Oj en CA en VO LP, P- VO o VD MA t- cn vj- o ** e ma \ — vo ■vt- CM LP, MA MA CM m CM o Nl" MA MA tiO •H i 1 i 1 1 I 1 | 1 i | I 1 1 | CO > Pi cu CT> CM MA CO ^ CM VO CO cn CM o -* IP. OA •H o CM VO p- CM MA -st- LP. p- MA LP. CO LP, VD c- rH t — O CM O CM MA O o r — MA T— T— T— v- CM Oj CO I 1 1 1 1 1 1 1 1 1 1 1 # ft LT\ CM MA CM CO T_ ** VO MA cn CM CO MA VD P- B -=t x — CM Cn -=d- CM "* t— CM LP. LP, xtf- VD CO 00 CD VD ma p- LP, CM VO -=* MA CM CM MA CM -3" CM CM EH EX O -P O m p- cn CM r— MA ^ — **■ -sj- P- o CM CO cn +> ft "<*■ CM on CO LP. ■vt" O P- P- MA MA p- MA CO CM -p CD o VO t— CM i — MA «5* T- o T— x — o <■ CM LP, OP e • m 1 h LP, MA CO c ,_ CM cn CM cn LP. X — CO V T— oo o •==)- T— o *tf- CM P- VO CM CO t— CM MA VD o cn H VD c- VO X — LP, LPv LP, "=f -3- UA MA •vf CM LP, CM O o 1 1 1 1 1 1 I 1 i 1 1 1 1 « l • CO 1 VD cn LP, VO cn MA ■■*■ LP. CO cn i— CO cn CO cn rO LP, LP, VO VO VO r- P- P- c— c- CO CO CO cn cn 3 CO • DH CO • Pi CM CM CM CM CM CM CM CM CM oj CM CM CM CM C\J q5 to, MA MA MA MA MA MA MA MA MA MA MA MA MA MA 64 Table IV. Data Density Code Used in Figures 5-19 The following table shows the symbols used in plotting frequencies of data in BIOMED 02D graphical output. For example, a symbol K represents twenty data points at a particular x-y coordinate. DATA POINTS SYMBOL DATA POINTS SYMBOL 1 1 21 L 2 2 22 M 3 3 23 N 4 4 24 0 5 5 25 P 6 6 26 Q 7 7 27 R 8 8 28 S 9 9 29 T 10 A 30 U 11 B 31 V 12 C 32 w 13 D 33 X H E 34 Y 15 F 35 Z 16 G 36-41 - 17 H 42-47 + 18 I 48-54 * 19 J 55-62 $ 20 K 63+ / 65 Table V. Open Ocean Area Delineations AREA LATITUDE LONGITUDE 1 40°-60° N 20°-40° W 2 20°-40° N 50- 70° W 3 10°-40° N 30°-50° W 4 40°-50° N 140°-150° W 5 30°-40° N 160°-180° E 6 20°-30° N 140°-160° Ji 7 10°-20° N 130°- 150° E 8 0°-20° N 150°-160° E 9 0°-10° N 160°-180° E 10 0°-20° S 10°-30° W 10°-20° S 0°-10° W 11 40°-60° S 10°-50° W 40°-50° S 0°-10° W 66 «=* o lp >£> 00 LP CO ^ LP, ■H CM OJ t- LP CM CO ">*• co to o (25 OJ OJ LT\ CO LP O CM CM CM S ^t c- o CJN CM Cn ^=1- o CM LP OJ sf CO ■^t Ph OJ o S W) •H CO cci co ft E CD EH E to -p -p -p ft O CD o rH O o •H CO O -P O ft cop> v U < en CO LP LTv CO 00 r- co LP 00 CM to to ^D 00 CO o en o o en en LP Cn Cn o t- T- T- CM C- CM r- o to en o o en o LP en en o r- 1— 1— t — CM r- 'd- LP rp LP co r>- cp c rp CM CO 00 MD c- t— O MD CO c- o v* -*t LP o LP LP LP CO r- rp CM ** *=*■ 00 -3- <5f LP T— o OJ CM VD c- o VD r- -=J- o LP o en en en LP ^ — en **■ vt * — V v- \ — CM CO *=t- ** 1- LP 00 en vo -vt- -si" C- en -vt" o O *=J- en V£) LP -* en o VD CM v~ T~ OJ \ — CM \— "tf- o V£> r- CO c- t~- LP f- en CO cp CM CM CM •«=h CM r*- ^t" O " CO *=* o \T\ o cn en en LP T — en "St ■<*■ T— T— T~ ^ — CM CO -st -a- T— co LP VD CO en 67 CO rt CD CD CO f-l CD -i U rt O Ph C CD P C CO Pi c •H -P CO CC Pi' 05 > CD CD S Q CD i-. +-> 05 CD T3 E Pi re! tf rt CO M M CD rH cti -3- -p o cti •H I CO td H •P tO Cti O I (25 WJ CM cti O I ^ Cti O I Ph w* H ^ CM" O i t Cti E W> •H CO Pi O ■H O ctio •JO—' P-~- E O CDO EH—' E ^ O -P-— P P E P CD--' o O o H O o cti CD u < lpc- lp cv [At- r- CM to cm F- cm O 00 CM — ' VO CO VO "* O cn C- 00 VO CO VO •**- o o vo -^ oo vo c- CTi-ct- lp-n1- o o -st-to o o o o VO r- O O LP CTi r- O 00 t- O T- t- CO T" O t-oo O T~ ^-CTi LP tO O O CO vo o o lp lp f- >«d- "«d- CM O O o LP tO 00 cm en o C- to VO 00 r- O LP en CO LP C- 00 f— in c-r- Is- CM ^oo VO VO LPOA CTvCM -^ tO r- to to^t- VO -sT O to LO T- LP- VO' cn t- o t- VO VO -vfO f- to CM CO — IP CM VO LPt^- O en t— en C- 00 tOO O IP VO CO 00 to VO' CM CM ■ LP- CM LP- CM CM CM- CM CM- CM CM CO CM •>=)- t- CM LP CM Is- LP LP VO "3" r- cm LPtO CM O CO to r- en VO CM r- oo LP tO CM -=d" CM VO CM ^ -^ ^f CM tO CM- — LPOA CT\tO tO CM O to CM — ' VO CM CM-—' 00 CM- 00- CM • o r- r-co LOVO VO CM "* -r- <* CO CTvLP tO t- to to O CM ■<3- ^~ ^S- CM -vj" CM cn to- vo "d- CM v- (TiC^ r~ lp r- co CM vo c— vo VO CM to LP VO r- oo VO CM -^t CM VO to CM r- VO o> sJ-O vo to c- to CM •sh to LP VO— ' VO--' CM--^ ^ r^-v_^ 'it— ' ■— ' •^— ' to to to to [O to to to tn CM 0^ CM VO VO oo r— VO CO- CM coco o ■* oo^- r~-o> LPC^ LPVO LP LP CM «sl- O O v-- — CM--' OA 68 xtf" ^— -x <^*^> o -P CT> 00 VD O •H a3 • • • • CO 1 rH "*■ v- mO W v— ^ t- CM 3 "* -P *~n r<-\ 0j C\J T" O 1 rH cor- » -P : • • CM cd ^~n " — *» O 1 rH CO in r- en S3 w 1- o v- o P • • • * ■p •=t rt ' — » '-■^ o I H CO o t^T- s--^ Ph ti T~ ^~ O m in P • • • • Ct CD v— * T* v — ■" CD en f-i CD • — v /"■^n * S <,£ rH ^-cn IT\C^~ -p c\j - — m -*t moo F. c o rH » • • • K (D E ^-v_^ VD^ CU fn u rt O Ph s s -p .^ ^-, CD -H 1 CO VD CM t— P, cd LP\C~- \0 K~\ C f> E • • • • £ W) tOw ^^ ^ c •H CM CM rC -H CO -P OT rt fi -H • • .— -v - V Oj > £ o l> -=t VO MD CD CD 2 Pi •H o C-rn KN1XN r-C • a » t K)C LfV ^- _-< CD H -J> tn K\ -P rt CD t3 ^_^ 6 FJ rt rt P^ t— tO CM v- U P rt co P* "— S o CDO • • M3 CM • • CO ir\ EH CM--^ s~~* ^^ ^.^ i — \ e& tnen CM CTi CD 0-P'~- • e • • P 5 -p P-. E CM "=*- LTv-sJ- -p o- — O LTv tOCTv •H Ofl Lf\0 incTi •P FP "4-v- ■sfw S — ' o u O rH KMT- • V * O C<"\ ^~ M o ^-^ M > CD rH cd ^> ! 0) o ^ — rt f-< v- T— EH C- C- m CO 1— CM VO CM «* o in CM in st o CM sj- to to sj- VO in *~ in CM T— O CM VO o to en to CO CO en vo T— ■(— ■( — VO r— o CO ■st- VO VO o r- CO t- O v~ to o to to o CM ^ — o 'r~ O 70 Table IX. Parameter Means by Marsden Sub -Square Using Values Averaged to the Secchi Depth fStandard Deviations in Parentheses) Mar. Sq. Sub-Sq. KV Salin. r/oo) Sigma-t °2 (ml/1) 130 51 19.0 (4.5) 34.24 ( .90) 24.37 (1.29) 5.55 (.61) 130 60 16.0 (4.5) 33.30 (2.05) 24.35 (1.97) 6.25 (1.12) 130 61 16.7 (4.6) 34.05 (.65) 24.75 (1.19) 5.80 (.74) 130 92 13.0 (6.1) 33.53 (.80) 25.10 (1.32) 6.56 (.97) 131 30 19.1 (5.0) 33.82 (1.02) 24.00 (1.80) 5.12 (.47) 131 34 22.1 (4.0) 33.25 (1.04) 22.90 (1.40) 5.18 (.53) 131 36 20.9 (4.0) 34.52 (.43) 24.11 (1.29) 5.09 (.45) 131 40 19.7 (4.6) 33.83 (.84) 23.86 (1.75) 5.38 (.50) 131 45 17.9 (5.8) 31.08 (1.94) 22.17 (2.22) 5.46 (.83) 131 49 20.1 (3.8) 34.43 (.36) 24.25 (1.14) 5.42 (.50) 131 55 18.0 (5.6) 32.58 (1.51) 23.26 (1.90) 5.50 (.72) 131 59 18.0 (4.5) 33.34 (1.74) 23.93 (1.80) 5.55 (.61) 131 65 18.8 (5.4) 33.73 (.63) 24.00 (1.71) 5.33 (.59) 131 70 16.6 (5.5) 33.88 (.63) 24.64 (1.65) 5.64 (.48) 131 77 19.4 (4.8) 33.27 (1.66) 23.50 (1.86) 5.40 (.42) 71 Table IX. (Continued) Linear Correlation Coefficients by Marsden Sub-Square Using Values Averaged to the Secchi Depth (Standard Deviations in Parentheses) Mar.Sq. Sub-Sq, Tgmp. re) Salin. (b/oo) Sigma-t °2 (ml/1) 131 132 132 v 132 132 132 78 16.7 (6.0) 33.29 (1.43) 24.07 (1.89) 5.14 (.26) 29 20.7 (4.4) 19.1 (5.0) 33.77 (.83) 33.82 (.98) 23.58 (1.65) 24.02 (1.82) 5.24 (.47) 5.22 (.61) 49 19.1 (4.6) 33.85 (.98) 24.04 (1.84) 5.27 (.45) 59 17.6 (4.4) 33.72 (.92) 24.35 (1.67) 5.53 (.42) 79 14.6 (5.6) 33.81 (.56) 25.02 (1.49) 5.91 (.62) 72 Table X. Linear Correlation C Sub -Square Using Val Depth oef f icients lies Avpragec by Marsden I to the Secchi Mar.Sq. Sub-Sq. Temp. Salin. Sigma- t °2 130 51 .260 .310 -.117 -.356 130 60 .104 .349 .222 -.725 130 61 .149 .226 -.051 .028 130 92 .266 .260 -.165 -.198 131 30 .207 .284 -.054 -.350 131 34 .179 .709 .524 -.209 131 36 .521 -.235 -.505 -.507 131 40 .297 -.292 -.523 -.317 131 45 -.029 .550 .380 -.025 131 49 -.008 .225 .061 -.185 131 55 .304 .474 .067 -.320 131 59 .041 .573 .402 -.058 131 65 .529 -.356 -.532 -.517 131 70 .471 -.327 -.473 -.629 131 77 .456 .175 -.161 -.516 131 78 .226 .430 .093 -.746 132 29 .183 .170 -.079 -.256 132 39 .167 .110 -.079 .057 132 49 .340 -.253 -.318 -.159 132 59 .356 -.037 -.258 -.125 132 79 .276 -.578 -.367 -.054 73 Table XI. Parameter Means by Open Ocean Area Using Values Averaged to the Secchi Depth (Standard Deviations in Parentheses) Area Tern ( c V Salin. Sigraa-t °2 (ml/1) 1 2 3 4 5 6 7 8 9 10 11 12.4 (4.5) 24.4 (4.D 22.6 (3.D 9.4 (3.8) 20.3 (3.1) 26.2 (2.9) 28.2 (1.5) 28.6 (.8) 28.7 (.5) 25.9 (2.4) 8.7 (4.9) 35.29 (.52)- 36.28 (.63)"" 36.65 (.45) 32.70 (.21) 34.50 (.36) 34.82 (.29) 34.60 (.33) 34.58 (.37) 34.48 (.34) 35.78 (.35) 34.39 (.56) 26.64 (.68)- 5.82 (.57) 24.43 U98)^ 4.78 (.56) 25.27 (.83) 4.77 (.24) 25.21 (.63) 6.68 (.57) 24.07 (.95) 4.94 (.29) 22.80 (.88) 4.75 (.31) 22.05 (.50) 4.62 (.36) 21.87 (.37) 4.51 (.16) 21.78 (.32) 4.44 (.05) 23.64 (.74) 4.59 (.26) 26.58 (.37) 6.45 (.80) 74 Table XII. Linear Correlation Coefficients by Open Ocean Area Using Values Averaged to the Secchi Dept^ Area Temp. Salin. Sigma-t °2 1 .528 .530 -.485 -.510 2 .664 .496 -.453 -.719 3 .543 .428 -.382 -.023 4 -.142 .265 .176 .083 5 .556 .147 -.501 -.429 6 .247 .139 -.236 -.150 7 .090 .121 .003 -.461 8 -.006 -.187 -.216 -.228 9 .235 -.468 -.490 .343 10 .629 .341 -.496 -.512 11 -.071 -.057 .173 -.104 7 5 p CQ o o fl o 0 rH o CO -p rH Pi to « CO •H CO !>> rH a) o •H CQ to CD U «50 O Pi HH tx! 0) rH crj EH o •H P rH a> rH O P O C o a) -h rH O ■H r- I m vO I O v- X t"- o 00 H C- K\ in m • m P • in + P3 1 + ^_ I c- l o \ — 1 O ,~ 1 !=> P5 v" ><; P *=r C— X CM VD m • rn in VD • o T— • • \ — CM ft ^ V* T— + m + 1 + 13 + P3 P3 P3 £> {=> in t=> P3 m in in tn ^ CTi tr\ "^- O O O O o O O o 1 l vo I rn 1 VO 1 o 1 1 00 tn c^ cr> CM VD CT* CM r- co » • • • • • • • m m "vl" tn m ^t" tn tn ii .u II .i' ii r li ii t/i 10 CO CO 10 to co i tsi CSI tsi tSl tsi tsi ts: tsi w CM m CM tn CM VD CM VD CM CTi V* id xi CD Pi Pi c\j •H £> ^ — a VD o s o o o CD CO m E X ^—^ CD T3 ^! Pi P h: Pn 0) to fi Pi •H •H rH ,£ rH O 0) o ? CD Pa CO p2 en o is] II II co v-, 76 m a> u CD -P fl o g co -P H ra 0) co •H CO >> H 03 fl I o CM I o to i=> p- i o to P- I o "l_ £> CM X T~ T~ X 00 i=> o MO X X ** cn in • O o •<* • I T— O CO • T— -si- o ^ — 1 • • ^- | CM r— 1 T — T~ T~ + tD + X £> 1 £> + + i=> t=>" •& i=> £> o CM p LO t=3 i=> to CM M3 M3 T~ CM CO v- to T~ CM CM to o O * O o • • en o o o o • 1 • 1 CO to • • 1 to 1 + + 1 1 1 1 LO LO + T— en MD LO P- LO o p- T- to CT> T — mo O MO V MO to M0 co o •«* **■ • o CM ■^ CM CM to to to ■ < P- CO MO P- -P E ci> P-i o en c en en c fl o 1/3 £D ' — ' 1- -H l £> t- t- o o o 1 I WS. II II O T— T — CM to o e t- O CD d P- P- p- P- P- p- o vof- fn rd -P "& cn cn cn cn cn en o en cno £ o3 to in — *^ j ti ! -p D i u^,in i in , in ' w i 1 ^-^4J)_ r . /t* >d - t» •H _ V . . -P V ctj -\ ,n> _ _J_ " CD sh \co i / \ "Vc •H rH 0) D o v : V: o O X3 CD ■^ \ j -CN fH S y ^ i" ctf 3 ? y CO 1 ^x pi co (!) CD h < o M • in z sin 0) CO *— — *• co CD -P s c cd O £ u • CD u 0 o S lO •H T- 1 1 1 1 1 1 1 Ph *->S~ - "o «* n m 81 t o in o in o © 1 ' O 1 -in 1 1 i i i 1 i i ! i D in 1 ! 1 : i 1 > in 1 j ! | i ,t 1 1 1 CM O CM CO 1 CM j i in I ! t* s ! s 1 !' i o o iri"-\ o c> CO 1 i i i i ! ' ' D 1? "9 \ 3i &T inC Ii ! i ; i . l/l * 1 1 i i *\Y !S o ±i_ m cs 1 1 \ 1 0 1 1 i m -ih f VvV1^ in in cs i «- Sn i Y™ i j ( T i 1 ' ^ ' J3 °o 0 O 6 o * 1* <^S^- 0 O (>) z i 1 1 i i i 1 CO co co « cd P< cO u o «H -P cd CJ C o •H -P cd co CO CO u & CO t CO CO CO fH M • CO CO CO CO -P o ■* CO O in O co p4 82 o 10 * aivoiiis MSDiXO Q--VW3IS NI1VS dwai Hi to. W -p r~- C CV) CD •H *» O LT\ •H CV cO o fn W *, -P o C ^f CD •H •» O O •H t^ «H «H CO CD CD O u O CO 2 C a1 O ra •H l "P r^ CD 2 rH CO CD U «. fH T- O KN O V" e W ^ CD ^ 2 bl) •H pq JLN3IDI:Ua03 NOiiV13li 87 aivorcis • N-aoixo ^•-vwois NITVS dWaj; Hiaaa-a.- HOIOD' a> u TO P* cr1 w a CD to o •H •H 1 O K\ *r-{ KN C K~n O CO •H to tH X; .n ^ ft •« rt LT\ f-l _r, W) »* -p ir\ C ^J- CD •H •\ O "vT •H ^}' - ,3 ft m> cd ^t- Pi c- to 9k •P M3 C VC Q) •H •■. O LTi •H UD O Pi o cC P* Pi cr< o CO •H 1 -P ^ CC 2 rH CO O ■t"" • Hi «* tt) Pi P bO •H F^ m i tn IN3IDIJJ30D NQIIVII^OD aivonis N30AXO i-vwois NI1VS dwai raaaa-a HOTODr CD u cd 3 c< CO C a, 'Ci CO h rt &: • CJ •H *— _ A Pi *. Cti OD fn r- to r. -p c^ C VD CD •H •> O CJ' •H .Tv bO •* ■p c^ C 00 a) •H - O CO •H CO j ti! cc >r ITl -J in ro cn U> > o -P CD -H CO CO «H CD O T3 O •H -P CO n cC CD TD r-\ CD P P> Cti P &H O ca, • a; cu •C •CS o •H O o o cu C/3 P P O iH P- CO EH O -P O CU H «H O CO u; 0 U Pi •H P<4 HOIOD 98 1 1 l 1 { 1 1 l t-t 1 2 (M CM CM CM — i -* —t — i •-t •-* CM <\J •4- CM m CM in t— < cm —-I - r-l CM CM cm CM CD u ? c co C CD M cri CD tJ (H 0) ,0 -P ct -P E-\ o H O ft-P U h o 0) rH 0 •-"in r»- ifM —4 ^-l(\l f-r-4 CD m .— t r-t (NJ — i •— i CO f--* ,-j i\j r- ^J-r-l— 1 r-H,— 1 CM m ro cm r\jm.-< CM i-H r-l —j r-"^-l (M —< oE cn"~" to i C\J CD U a cr1 ra c cu (0 -P Ph •H o • CD CU CO T3 O 'CM' CM- -•■ chcm i .— i-hcmcm Ul £2 u o I 'J W1 o o o in O a o •H -P O ? ctf to T3 O P -P > O H rH P CD H P CD PEh 0) a p •H p, CL> ■P fi •H O P o P c -p f-l -P Cm O Q) (X) 0) •H (m)Hiciia woiioa 100 in m i m 2 tNJ CNjrg — i o ~j — • su — i r\i u\»j'-^— -i-t — j •— i —• r»-i— icmcm ~< «— f> ,-< r«i!\| ro in sy —j -j — i — t r-l fO— (—<■-<— 'CM-— lUUM f\Jf\J * — ,<-t\j!,i'-*fMn —i f-im c\j m — >o~. ^rmrvvytM cmx\— ■ «=* -'r-i^i - r\j rn •-- 1~- in r"u vm u.' 2: in '-'fT-'fl — iClsn o i u i u i j_ c°«-» im'm ivjrvj^j— ^<=. cc i •— — ■> —4 ^ — > - — • . — i .— «Lno.*-~ — ' — <1NJCN) LT\ I CD Pi CF pi cr1 co P O M rH P-i CD ) ,C + O •H o CO CD C CO CD •xi CC M CO CD CD H rQ 'O CD 0 EH -P P O O P rH ft fH 0 CD «i- Sh 0 2 « P CD fH • 0 ■<*■ ft'tf" £ | CD OJ P CO, \ — CD O 0 CD U «H CD P. 0 2 C71 CO CC o 0 U to •H (3.) d W 3 1 102 i 1 1 1 r-l r-l ~ •-I mm i -« c\j ro ▼- — i .— i to -J r-l ^H r-< _| ^-1—1 — r-4 H^H (\|-T r-l f\J "— r-l r-4 r-1 r-1 r-l . — If — 1 (NJ— t rJr-1 r-t r-t — 1 -J (NJ — -lr-4(NJ f— tr-l •— 1 r-t 1 r— 1 r-4 r-4(NJ fNjrNJ r-l r-t — r-t r-l r-i _|— 4 r-t r-t r-t— « —t rj r-l — 1 -J (NJ (NJ r-t r-l r-f-4 — 'r-l r-l r-l r-l _t r-l r-t — tt\)r-l— 1 —1 ,-tr-tf\l. _ r-lr-l —1 C\J — KNI— 1 — | — 1 CNJ r-i <"\J r-l — » — 1 (\| — t OK —1 r-lr_|f\|r-lr-lt\J —> r-< CM r-l (NJ rj — ' r-l r-t r-l „, -_l — 1 ,-t _| r-l —t r-t r-l —1 r-| tN) r-l r-l _| r-t r-t r-t (NJ — t 1 \J r-l O •— 1 _| r-t — • r-l — 1 r-l mm r-t r-t — • — 'r-l IMrJ r-l r-l —• (NJr~ 1 — ! —4 r-l r-l r-l _| r-l mm — < mm i— i r-l r— 1 . c CD TC CD Jh cO r£ o n a, UJ oU UJ 00 o 6 Ci q oi CN o r£ • -P CD Pt *C CD o T3 o ■H r^ rC "P o •H CJ CO 0 C 73 CD -C C fi •H o O •H Pt -P O U C o 2 «H cd M CO 0 0i r- rQ 13 fij CD EH -P P O O P rH ft *- CD CD c»- r-l CD 3 tf P a u • cd r— ftKN B 1 CD t — -P r<"N T— CD CJ CD cC u «H cC fH 2 pi o< CO CO CD Sh •r-t P4 (o.) dW3i 103 IT <<\J (\J ^J_4vJ- OJ- 2 rg »r\j <(NJ f^ CM .-CMf.-)— < —i— "CI— • CM C\J IN CM _rr--'\ -ifc o 6 o 0) o 11 co 03 g; — . »—* A • -P cd P, X -P O -H O CO 3C CC H H- CO cu 0. Cf r-l !•! 'ti co Q CD EH -P MB— -P O X O -P H u ft p, u Q) CD l CC P. • CD C-- PhCM E 1 CD t- -P N"\ \ — CD O CD cc p. >- — • j. I (j^\rm-i — « 1-1 —< 'UJ I 5;U-01UA —"CM —J 13t I | CDCTfM J"(\.1— t— < — «— < —> r03T I OIQ^- '-J r~> — r — • (\i •-< •-* '-* .-h ^a | -_*t.o coxier- i-itMr-i— 'in — > T3 P •H •r-l CO & C o cd O no CO o c •l-i P CJ C > 0 M

M CD rH .O CT ct Eh t3 c 0) -p -P -P p, o > -P •H • C UA •H T— rH | cC •5— CD Ks T— CD O CD CC U tp ro p, 2 2 CT1 CO CO ^3- i" m^-" cm ti> i— i o r— < r~ u > < r,o *u — < T>< X'~'OC3COu-'invi T— — < r-i U — J—JCOlVC^'Mrn"- — ■• m— ' •— ' »-l«-l CNILJ IL-'CocUvJ-iMmom.— •— I — '--(_,;j)":-i oir or-^i- ten — * — * "M{JLlV^C,'J5<^LU-r'X|f--'MrM — 4 •—••—* — i COM 2-5.. C_)(7>C)OCja,,r\0'Nl(M— 'CM <\J>-«CM ,-t.-4_| r\jtj>x. > •— to ~3 o cj *~ * CO r^ cvj rQ u'LHnijtt'>y>i;ririM-'t,o-i-j _» (\j r\( »~ .;vj —j 1 _ w-A vT t\J — ■ -T l_' O 50JCL' — )U,U Ul Kj m Ul Tj CD CD H -P ^ -P tf O EH H Ph O -P -P 1 f-i C? CD E r-l CI CM —1 r-4 — 4 —1 —4 Ga» . — l «— * — — r— < — 1— « —4 a hkmm 1 1 l |_ MMWB o o u a pi a1 CO CO CfJ o in O l-VWDJS P P,— Q) . T) CD T3 •H O ^ o CJ o >> CD -P CO -H CO _^ u re h u T3 0) UJ CD rH t/> -P r^ ■P 03 © O EH CN rH Ph o -P -P 1 ^ tf CD E vn M 0) •H K CO 0) • O LO 03 v- rc> CO *- o C4 VD CD fH Pi HO •H 108 rg m m i to -J m .-4 "O <}•,}■.-• in corn #— i— < ,_! — i— <-- 1 — ici ^-irvjO'— i ■-I — I — I v!" — I r-J r-< IT> — I vj" CO — • r-« — -i — i r\) — i CM •— ' rO ^J" — « •-< m f"i >-< •-h-j — ir-i — '(MLn-Mirvrn^f — imcirv] ci rur-v*iii\00^- ii— I -J- i—i »-< rnr'J'(rOm-T' ri eg cl> '-" ' *J ~£) — < •— ■ -j r-r--u'*>,),rtir\<"'~>— <(>o — < — km ;->j > ^vr i> iu> v." t\; m-' umiju" vu t/KJ uiM'aj(M «ro»M— -r\j i\i m(MiMM_Mi> J * >uvy o i — r\j CO c -a to X -P'—. ft • o , O -P a> -h X CO w h- u CC M LU CD to ygen plotted efer to Tabl o X Pi o \«- Surface 131-55. c— Q) u (|/|lu) N 39AXO o 109 {\/\vj) N39AXO 110 m m i m to w-ir- (\JC\J— I r\j c\ir\. —>— >nfi~W -4 oit-i — •— i ■— •— irMNDinCloOOOr- r-l —•— * •— MM ^Ti'XltL.- 5-^^U »-^ ■— J~> VT "J-1 '— ' ^- l—' t*~ *" ' ,Vr_i ,_! — i(m ir\r>— 'vj-nuM ■— i voi\i> T3 -P o •H ft f Secchi int dens E c o P( X C rH r- •H O P, P* tH LI! £ > a 2 M na a> o (N :s u cc a:' u o: Eh Ul t o U) ilicate plotte 1-55. Refer t o m k\ T" Surface square 1 o o -o o 6 OD CT. 0) .H r-rf M ■H O o (l/?D-6ri) 3 J.VDI1 IS 111 aivoms N30AX0 ^-vwois NITVS awaj; Hidsa-a- N O"- HOTOD1- K> CO CO CO 0) u cc cu o o CJ ■H -P CO H -P O o if) o 1N3IDHJ30D m o -I- o C\J CD Pi W •H ft, NOI1V13880D 112 aivoms N30AXO- 3--VWOIS NITVS iwai Hiaaa-a CO U tn q i PQ c CM G) & •H p-4 1N3IDIJJ30D NOIlV13S^OD 113 o> 00 - aivoiTis waoixo ^-vwois NI1VS awax Hiaaa-a H0100 o^ CO c~- CO CO cC CD CJ o CJ •H U •H in in !N3IDIii30D NOIIV13MHOD 114 1 :y— p 15? 149 14G 143 140 !37 13 4 !3t: ~ T " i 128 125 !2 2 119 II6; 113 110 107 1)04 10 j 63" ~60 I MS7 I.'. . ■'._ 4 ■ J .1 1 | j i i i ■ V ■ i \ ! i i T . Y ... Up j: — Sf|i V ■■ i 54 51 — u j ! |- >-" — r ■ ■ .V. Si: -A . < • ■ 1 j j ( -■-■*- i y — i tw * ,...,. — — - 48 i- ' i •■ > s. 1 : x-l'vv: : 1 j: 1 • 1 ft > ' ■ 3 ■ J 1 1 ■■;•■ I -4tr 42 • ■ i . 1 t < ....Vj- _ i s ■■■ ■'■) 1 1 j n 39 1 ' 1 U i 5- ■ X- ■-*■■■«■■ si 1 I i 36 1 V T-aI — L j s* 33 ! • • 1 ■ '■i #££-. a 1 :| :\ P i i : ' :i ~3t> , "v _ i 'J - -II ■ ■ | ::-: 1:: ! .... 27 I ■ . i 1 . i \-V- "V : -Jfl- ■~~"~ -¥ 24 •91 : i i r~\v >; . . ■ is £ 1 .-^ fil ^ -i ' It! ! : : 1 i i ; ; ; ; ' .... .... .... .... ::::! Figure 21 Points for which upwelling indices were computed hy Bakun (1973). (Large circles indicate locations off Oregon and Cali- fornia for which comparisons were made with Secchi data. ) 115 UPV/ELLING INDEX o o o o o o I o o T3 ra £> O 25 X 2 a, a 2 ■ — , _1 5 3 :s Ui &; -P n- o •H r-H o o CD en o CM oo n cd o C M-H CD rH fH O t3 •H CD rH -C o P CO CO o •H rH O cO bfl •H l-H rH H O X a CD r^ C) CO ai Eh CD CO «| « — >> CtiKN o - CO co b. o cti +3 Cli a> •H -PI 0 W) £ ^d •H T3 o o CD tv-. •H rH H CD 5 CO • £ + O Ph!3 CO CD ^J fH O rH • O I v to CO CD O bQ CD o o CO cti CD O PnrQ 2 o I . • ■Hv- ,C t- rH II CD ■pvo co,c; Ph cr. t3 rs3 P CD v- CD TJ & Pi £ I CO o o •H Cti -H O CO Cti CD O Cti CD 2 CO cd o ^ cr1 cc; DO O EH CD r=> CM CM CD U •rH (uOKld3<3 IKDD3S 116 UPWELLING INDEX 8 o o o o o o t> A to -P fi P^-» © CD T3 n- T o r— n •H r-i LU & ni (-1 o o o- — CJ CD U-r .CO ^ CD CO C CD CD -H OC^ bjO-H P ^ Q)H O P, O -H O-d-P^ •H £ CU rH bO X O fi'-* -F CO -H r-H T3 H p n) P M o ox oaa > O ,000; P W rH Ct P ^ CU > T3 nj AON >, o oe ^ -P \ ^ P o--' t>> ■p a' h m p cio P O CO o o das p co a; . o,^;r- -P CO CD P 1 cc > nd PhO nd onv 0) i- X -H T3 P CD ,Cl X o Td O O bJ P O P to CD inr _ -H CD v- P CO CO • O bo T2 to R'OCl+'H •H CD O ■"• O Nllf rH > PhP rH P CO -. • CO CD CD CD -3" P £ CO P O c6 p1rQ p .CO AVW - £ O O 1 >> O to T3\| CO O a cd • > cd p to-p HJV • •rlT- ,p C\J H II CD -p \£) K^ Pi (T. fO tS3 P CD \- CD HVIa X) & P P 1 CO O O •H Cti -H ^•Pti-PT) O CO CO CD aa^ O CO' CD p CO CD OrC C1 03 W (J B (D P NVl • to C\J CD U P bj •H (3q °2 2 £ W Hid3Q IHDD3S 1 1 7 o o o in -P (0 03 o o - o o 0 o Q) o -P in i ^ X O UJ -O t3 ; i i | •H o o 0) CO : o in • 7 i sj- CM 0) tjD •H 1 2 (in ? ("O Hid3Q I HDD3S 118 CM V£> -P W a o o o O 0 ■p o X 0) T3 C to CD 3 CO -p o o o CD in cm CD ^i pi bo •H Cui3 HidBQ IHDD3S 119 X UJ O 2 z 3 CM en I CO cti o o c o CD o CD -P O > Hld3Q IHDD3S 120 I22°00' I2I°55 ■ ■■' ->w**"*vrw I22°00' I2I°55 I2I°50' Figure 27. Monterey Bay, showing locations of CALCOFI stations occupied by Hopkins Marine Station of Stanford University. 121 UPWELLING INDEX QJ Pi CO +> Pi O J5 Pi o Pi cd CD O .Ej +> Pi o E to > CD Pi •H M PI •H (H Pi 13 O pi r- 03 G> I XI -p Pi CD t^v T3 Pi o •H 41 O o CD CO -P a -p CO CO C\J CD Pi g, •H CuOHldHQ IHDD3S 122 UPWELLING INDEX >> a 33 >> u CO -P O o h 0) >> O P Pi o E co > CO PI PI •H CD ■5 ft • PJC- co" cr> I p ft co t^ t3 P! •H O & -H O P o rt 03 -P CO CO CM 0) P) •H P4 CuOHld3<] IHDD3S 123 UPWELLING INDEX m >> CD u ■P O o 03 CD >> > cd pq >> > >> 0 -P O O > CO FQ >> 0> u CD -P o o +» p< (1) •H A V O CO IO CD w •H P4 Cui)Hld3 0 IHDD3S 130 o •H ■P Cti -P CO a) pq CD 0 -P o o +> CD T3 o o 0 CO 0 u •H Pn CuOHldia IHDD3S o •* — in t> c> > cd CN U cu -p o o o s»: d ^H X o UJ o o D T- • CO > A -P ft o CD m T3 •H .C O CJ •H o F^ G»0 HIJ3Q IHDD3S 132 r- i o o o -f + C\J o o m N + + + + ■+■ CM T CO o in rn in cm (uO HldBQ IHDD3S to ra o a o m o •H -P -p CO o ^ in ctf ON cu a) -p O o O s in z •H *— _J ttO _l £ UI •H £ r-{ a. CD O o D pth vs. upw 0-1973. o 0) r- in Secchi d and 4 19 O • (D o in •H 177 CM \D cr> T- o 1 - o *~ m C7> T— -P to cd o o o _o CN e Oregon 0-1973. o a r- -O -P CT> X ^ *" UJ o ■<*■ a «H z -O x a cd a5 o T3 C to -o z ■H »- to —J UJ ling tion CL rH CO 3 E ps rH O > -P 0) O VO -p en cti vo rH LTv O T- "P >>!>> .cj 03 Q) •H p4 C|"0 3WniOA NOONVldOlAHd 135 APPENDIX A AVERAGING PROGRAM C C c c c c c c c c c c c c c c c c c c c c c c THIS PROGRAM READS SEQUENTIAL A-SHEET DATA, SCREENS THE DATA AND THEM STORES THE SCREENED DATA IN PREPAR- ATION FOR BIMED02D ANALYSIS. EACH STATION LACKING A SECCHI DEPTH MEASUREMENT IS OMITTED FROM ANALYSIS. LATITUDE AND LONGITUDE CORRESPONDING TO EACF STATION ARE SCREENED FOR THE AREA TO BE ANALIZED. TEMPERA- TURE, SALINITY, SIGMA-T, AND OXYGEN ARE THEN SCREENED AND AVERAGED OVER THE SECCHI DEPTH. INTEGER*4 XD,XM, XTM, YD , YM , YTM DIMENSION TP(3500),TEMP(3500) , SAL (3500) , S IGMAT ( 3500) , 5)OXY(3500) DATA T P /3 5 00* 0. 0 /, TEM P/ 3500* 0. 0 /, SAL/3 5 0 0*0. 0/,S IGMAT 3/ 3500* 0. 0/, OX Y/ 3 50 0*0.0/ CALL REREAD THIS SECTION CF THE PROGRAM SETS THE SUBSCUARE BOUN- DARIES AND READS SEQUENTIAL DATA. N = l DSS=35.43 DNN=36. 0 DEE=135.0 DWW=136.0 1 READ( 5, 10,END=80) 3ASAL,ASIGMA, AOXY, 10 FORMAT ( IX, 212, II , a5X,F4.0,24X, I 1) IFtKD.GT.l ) GO TO 15 CONTINUE XD, XM, XTM, YD, YM, YTM, DEPTH, ATEMP, KD IX, 13, 12, II, 5X,F4.0,F5. 2, F4. 2, F5.0, THIS SECTION CONVERTS MINUTES AND SECONDS TO TENTHS DEGREES. LATITUDE AND LONGITUDE IS THEN CHECKED AGAINST THE SUBSQUARE BOUNDARIES. A = 0.0 B=0.0 C=0.0 D = 0.0 IF( (XM IF( (YM OF ) XTM=XTM+1 YTM=YTM+1 EQ.O) . AND. (XTM.EQ.O) EQ.O) .AND. (YTM. EQ.O) ) ALAT=FLOAT(XD)+( FLOAT (XM)+( FLO AT ( XTM ) *. 1 ) J/60.0 ALON=FLGAT(YD)+( F L OAT ( YM )+( FLOAT { YTM )*. 1 ) J/60.0 IF(ALAT .LT.DSS) GO TO 1 IF(ALON.LT.DEE) GO TO 1 IF(ALAT.GT.DNN) GO TO 1 IF(ALON.GT.DWW) GO TO 1 ThIS SECTION SCREENS THE DATA, AVERAGES THE DATA OVER THE SECCHI DEPTH, AND THEN PLACES THE DATA IN STORAGE FOR BIMED02D ANALYSIS. BTEMP=0.0 DTEMP=0.0 ITEMP=0 ETEMP=0.0 FTEMP=0.0 GTEMP=0.0 BSAL=0.0 ISAL=0 DSAL=0.0 SECCHI=0.0 IF( (ATEMP. EQ. 0.0 ) .AND. (ASAL.EQ. 0.0) ) GO TO 1 I F(ATEMP.EQ.O.O) GO TO 20 BTEMP=ATEMP*100.0 DTEMP=ATEMP*10.0 136 APPENDIX A (CON'TJ ITEMP=OT£MP ITEMP=ITEMP*10 ETEMP=FLOAT(ITEMP) FTEMP=BTEMP-ETEMP GTEMP=FTEMP*10.0 GO TO 2 5 20 GTEMP=0.0 25 IF(ASAL.EQ.O.O) GO TO 30 BSAL=ASAL/10.0 ISAL=BSAL DSAL=FLOAT(ISAL) GO TO 35 30 DSAL=0.0 35 SECCHI=GTEMP+DSAL IF(SECCHI.LE.O.l) GO TO 1 CTEMP=0.0 CSAL=0.0 CSIGMA=0.0 COXY=0. 0 40 READ< 5, 10, END=80) XD, XM, XTM f YD f YM.YTM, DEPTH ,ATEMP , aASAL,ASIGMA,AOXY,KD IF(KD.EQ.l) GO TO 65 IFUTEMP.EQ.O .0 ) GO TO 45 A=A+1.0 CTEMP=CTEMP+ATEMP 45 IF(ASAL.EQ.O.O) GO TO 50 B=B+1.0 CSAL=CSAL+ASAL 50 IF(ASIGMA.EQ.O.O) GO TO 55 C=C+1.0 CSIGMA=CSIGMA+ASIGMA 55 IF(AOXY.EQ.O.O) GO TO 60 D=D+1.0 COXY=COXY+AOXY 60 CONTINUE GO TO 40 65 CONTINUE TP(N)=SECCHI IF(TP(N).LE.0.1) GO TO 70 IF{TP(N).GE.99.0J GO TO 70 IF(A.EQ.O.O) TEMP(N)=0.0 IF(A.GT.O.O) TEMP(N)=CTEMP/A IF(B.EQ.O.O) SAL(N)=0.0 IF(B.GT.O.O) SAL(N)=CSAL/B IF(C.EQ.O.O) SIGMAT(N)=0.0 IF(C.GT.O.O) SIGMAT(N)=CSIGMA/C IF(D.EO.O.O) OXY(N)=0.0 IF(D.GT.O.O) OXY(N)=COXY/D GC TO 75 70 CONTINUE TP(N)=0.0 GO TO 15 75 CONTINUE N = N+1 GO TO 15 80 CONTINUE L = N-1 NUM = L WRITE (8f90)(TP(N) ,TEMP(N) ,SAL(N) ,SIGMAT(N) , aOXY(N) .N=1,L) 90 F0RMAT(F8.0,F6.2,F5.2,F6.0,F5.0) WRITE(6,100) NUM 100 FCRMATf «0' , 'TOTAL COUNT THIS SUBSQUARE : • , I 7 ) STOP END 137 APPENDIX B SAMPLE BIMED02D OUTPUT BMD02D CORRELATION WITH TRANSGENE RAT I ON REVISED JANUARY 29, 1970 HEALTH SCIENCES COMPUTING F AC I L I TY, UCL A PROBLEM CODE TEMP NUMBER OF VARIABLES 2 NUMBER OF CASES 3399 CASE SELECTION CARDS A CASE IS ACCEPTED IF (VAR( 2) NE 0.0000) ** VARIABLE FORMAT CARD (S) (F8.0,F6.2) REMAINING SAMPLE SIZE= 3399 SUMS 36683.0000 61336.2617 MEANS 10.7923 18.0454 CROSS PRODUCT DEVIATIONS COL. COL. 1 2 1178283.2500 41609.7930 2141609.7930 105265.5625 STANDARD DEVIATIONS 7.2434 5.5658 VARIANCE-COVARIANCE MATRIX COL. COL. 1 2 1 52.4671 12.2454 2 12.2454 30.9787 CORRELATION MATRIX COL. COL. 1 2 1 1.0000 0.3037 2 0.3037 1.0000 138 APPENDIX C TIME SERIES ANALYSIS C THIS PROGRAM READS SEQUENTIAL A-SHEET DATA AND SORTS C SECCHI DEPTH OBSERVATIONS ACCORDING TO YEAR AND MONTH C OF OBSERVATION. SECCHI DEPTH MEASUREMENTS ARE THEN C AVERAGED FOR EACH MONTH IN THE AREA OF ANALYSIS. C HSiSiSoS T8fsE6T?^12),DIVIDE(72a2),AVSEC(72,12) DATA NLAT/43/,NL0NG/124/, NLA/46/, NLO/128/, N/O/, N I/O/, SHYEAR/O/ CALL REREAD DO 2 1=1,72 DO 1 J=l,72 TOTSEC( I, J)=0.0 D1VIDEC I, JJ=0.0 AVSEC( I ,J)=0.0 1 CONTINUE 2 CONTINUE C THIS SECTION READS IN SEQUENTIAL STATIONS AND CHECKS C LATITUDE AND LONGITUDE TO INSURE THEY ARE WITHIN THE C AREA OF ANALYSIS. STATIONS ARE THEN SCREENED FOR C ERRONEOUS MONTHS AND YEARS AND CHECKED FOR ZERO SECCHI C DEPTHS. SECCHI DEPTHS ARE THEN AVERAGED FOR EACH C MONTH. 5 READC5.10,END=15) XDj XM, YD, YM, SECCHI , I YEAR , MONTH , KD 10 FORMAThx,2I2,2X,I3,i2,14X,F2.0,13X,2I2,24X,IlJ IF(XD.LT.NLAT) GO TO 5 1FCXD.GE.NLA) GO TO 5 IF(YD.LT.NLONG) GO TO 5 IF(YD.GE.NLC) GO TO 5 IFCKD.GT.l) GO TO 5 N = N+1 IFCSECCHI.LT. 0.1) GO TO 5 IF(MONTH.EQ.O) GO TO 5 IFC IYEAR.EQ.O) GO TO 5 IFCMCNTH.GT.12) GO TO 5 IFCIYEAR.GT.72) GO TO 5 TOTSECCIYEAR, MONTH )=TOTSEC( I YEAR, MONTH) +SECCHI DI VIDE CI YEAR I MONTH )=DI VIDE (I YEAR, MONTH) +1.0 GO TO 5 15 CONTINUE DO 30 IYEAR=1,72 DO 25 M0NTH=1,12 _ „c IFCDIV1DEC IYEAR, MONTH) .EQ. 0.0) GO JO 25 AVS ECC I YEAR, MONTH )=TOTS EC (I YEAR, MONTH) /DIVIDE CI YEAR, SKCNTH) 25 CONTINUE 30 CONTINUE 50 FORMAT?' 1« ,10Xj ' TOTAL NUMBER OF STATIONS ',15) 100 FORMATC'O' ,10X,« TOTAL NUMBER OF S-DEPTHS ',15) 150 F0RMATtJ0^,20X,«TIME SERIES OF SECCHI MEASUREMENTS') DO 600 IYEAR=32,72 MYEAR=0 MYEAR=1900+IYEAR WRITE<6,200) MYEAR 20 0 FORMAT C ' 0' ,15X, 'YEAR=« ,14,//) 300 FGRMATCJ0°?5X, 'MONTHS 5X,'NR. S-DEP ' , 5X, ' A VER. S-DEP') SRITE?6?400y_M6NTH,DIVIDE( IYEAR, MONTH) , A VSEC C I YEAR , SMONTH) 139 APPENDIX C (CON'T) 400 500 600 FORMAT ( • CONTINUE CONTINUE STOP END •,6X,I2,10X,F4.0,11X,F4.1) 140 BIBLIOGRAPHY 1. Anderson, G. C, "The Seasonal and Geographic Distri- bution of Primary Productivity off the Washington and Oregon Coasts," Limnology and Oceanography, 9(3), 284-302, 1964. 2. Arsen'yev, V. S., and Voytov, V. I., "Relative Trans- parency and Color of Bering Sea Water," .Dccanology , 8_(1) 41-43, 1968. 3. Atkins, W. R. G., Jenkins, P. G., and Warren, F. J., "The Suspended Matter in Sea Water and its Seasonal Changes as Affecting the Visual Range of the Secchi Disk," Journal of the Marine Biological Association of the OnTted Kingdom, 33, 497-508, 19S4. 4. Bakun, A., National Oceanic and Atmospheric Administra- tion Report 671, Coastal Upwelling Indices, West Coast of North America, 1946-71, 103 p., June 1973. 5. Bakun, A., Personal Communication to S. P. Tucker, February, 1974. 6. Brown, P. J., Correlation Coefficients Calculated on a World Wide Basis Between Observed Secchi Depths and Other Simultaneously Measured Standard Oceanographic Parameters, M.S. Thesis, U.S. Naval Postgraduate School, Monterey, 1973, 123 p. 7. Cialdi, A. and Secchi, P. A., "On the Transparency of the Sea," Tr . by A. Collier, Limnology and Oceanography, 1_3(3) , 391-394, 1968. 8. Clark, G. L., Ewing, G. C, and Lorenzen, C. J., "Spectra of Backscattered Light from the Sea Obtained from Air- craft as a Measure of Chlorophyll Concentration," Science, 1_6J7_(3921) , 1119-1121, 1970. 9. Dixon, W. J., Biomedical Computer Programs, University of California Press, Berkeley, 1973, 773 p. 10. Dixon, W. J. and Massey, Jr., F. J., Introduction to Statistical Analysis, McGraw-Hill Book Co. , Inc. , New York, 19 57, 488 p. 11. Duntley, S. Q., The Visibility of Submerged Objects, Visibility Laboratory, Mass. Inst, ot Tech., 74 p., Final Report Under Contract N5ori 07864, August 1952. 141 12. Duntley, S. 0., Oceanography from Manned Satellites by Means of Visible Light, Woods Hole Oceanographic Insti- tution Report 10,- 1965. 13. Duntley, S. Q. , Underwater Lighting by Submerged Lasers and Incondescent Sources, Scripps Institution of Ocean- ography , Reference Number 71-1, 1971. 14. Fonselius, S. H., "On Eutrophication and Pollution in the Baltic Sea," Marine Pollution and Sea Life, ed . by M. Ruivo, 23-28, Fishing News (Books) Ltd. , England, 1970. 15. Frederick, M. A., An Atlas of Secchi Disc Transparency Measurements and Forel-Ule Color Codes for the Oceans of the World, M.S. Thesis, U.S. Naval Postgraduate School, Monterey, 1970, 188 p. 16. Graham, J. J., "Secchi Disc Observations and Extinction Coefficients in the Central and Eastern North Pacific Ocean," Limnology and Oceanography, 1_1_(2) , 184-190, 1966. 17. Holmes, R. W. , "The Secchi Disc in Turbid Coastal Waters," Limnology and Oceanography, 1_5_(5) , 668-694, 1970. 18. Jerlov, N. G., Optical Oceanography, Elsevier, Amsterdam, London, New York, 194 p. , 1968. 19. Luksch, Josef, "Expeditionen S.M. Schiff POLA im Mittel- landischen, Agaischen und Rothen Meere in den Jahren 1890-1898. Wissenschaftliche Ergebnisse XIX. Unter- suchungen iiber die Transparenz und Farbe des Seewassers." Denkschrif ten der Kaiserlichen Akademie der Wissen- schatten. Mathematisch-Maturwissenschaf tliche Classe (Wien) 69, 400-485, 1901. 20. Murphy, G. I., "Effect of Water Clarity on Albacore Catches," Limnology and Oceanography, 4 (1) , 86-93, 1959. 21. Pal:, H. and Zaneveld, J. R. V., "The Cromwell Current on the East Side of the Galapagos Islands," Journal of Geophysical Research, 78_(33), 7845-7859, 1973. 22. Petri, K. J. and Starry, R. F., Remote Measurements of Sea Surface Wind Velocity, American Soc. o± Photogrammetry , Oceanography Symposium, Orlando, Florida, October 1973. 23. Poole, H. H. , and Atkins, W. R. G., "Photoelectric Mea- surement of Submarine Illumination Throughout the Year," Journal of the Marine Biological Association of the United Kingdom, 16, 297-324", 1929. 24. Postma, H., "Suspended Matter and Secchi Disc Visibility in Coastal Waters," Netherlands Journal of Sea Research, 1(3), 359-390, 1961. 142 25. Ryther, J. H. and Yentsch C. S., "The Estimation of Phytoplankton Production in the Ocean from Chlorophyll and Light Data," Limnology and Oceanography, 2_(3) , 281-286, 1957. 26. Tyler, J. E. and Preisendorf er , R. W. , "Transmission of Energy Within the Sea," The Sea, Vol I, ed . by M. N. Hill, 397-445, Interscience Publishers, New York, 1963. 27. Tyler, J. E., "The Secchi Disc," Limnology and Ocean- ography, 1_3(1), 1-6, 1968. 28. Visser, M. P., "Secchi Disc and Sea Colour Observations in the North Atlantic Ocean During the Navado III Cruise, 1964-1965, Aboard H. Neth. M.S. 'Snellius' (Royal Nether- lands Navy)," Netherlands Journal of Sea Research, 3_(4) , 553-563, 1967. 29. Voytov, V. I., and Dement 'yeva, M. G., "The Relative Transparency of the Indian Ocean Water," Oceanology , 10(1), 35-37, 1970. 143 INITIAL DISTRIBUTION LIST No. Copies 1. Defense Documentation Center 12 Cameron Station Alexandria, Virginia 22314 2. Library, Code 0212 2 Naval Postgraduate School Monterey, California 93940 3. Department of Oceanography 3 Naval Postgraduate School Monterey, California 93940 4. Dr. Ned A. Ostenso , (bde 480D 3 Office of Naval Research Arlington, Virginia 22217 5. Oceanographer of the Navy 1 Hoffman II 200 Stovall Street Alexandria, Virginia 22332 6. Naval Oceanographic Office 1 Attn: Library, Code 3330 Washington, D.C. 20373 7. Dr. Robert S. Andrews 1 Department of Oceanography Naval Postgraduate School Monterey, California 93940 8. Mr. Roswell W. Austin 1 Visibility Laboratory Scripps Institution of Oceanography La Jolla, California 92037 9. Dr. Hasong Pak 1 Department of Oceanography Oregon State University Corvallis, Oregon 97331 10. Dr. Wayne V. Burt 1 Department of Oceanography Oregon State University Corvallis, Oregon 97331 144 11. Dr. Peyton Cunningham Department of Physics Naval Postgraduate School Monterey, California 93940 12. Dr. Seibert Q. Duntley Visibility Laboratory Scripps Institution of Oceanography La Jolla, California 92037 13. Mr. George Eck Naval Air Development Center Johnsville, Warminster, Pennsylvania 18974 14. Mr. Gus Griffin Code 14 Naval Undersea Center San Diego, California 15. Dr. R. C. Honey Stanford Research Institute Menlo Park, California 16. Mr. Kenneth V. Mackenzie Naval Oceanographic Office NRL, Code 8050 Maury Center, Bldg. 58, Rm. 117 4555 Overlook Avenue, S.W. Washington, D.C. 20390 17. Dr. Robert E. Morrison Code 102 - OSC Office of Naval Research Arlington, Virginia 22217 18. Mr. Jerry Norton Oceanography Department Naval Postgraduate School Monterey, California 93940 19. Mr. Larry Ott Naval Air Development Center Johnsville, Warminster, Pennsylvania 18974 20. Mr. Thomas J. Shoppie Naval Air Development Center Johnsville, Warminster, Pennsylvania 18974 21. Dr. S. P. Tucker Department of Oceanography Naval Postgraduate School Monterey, California 93940 145 22. Mr. John E. Tyler Visibility Laboratory Scripps Institution of Oceanography La Jolla, California 92037 23. Mr. Alan Baldridge, Librarian Hopkins Marine Station Pacific Grove, California 93950 24 . Mr. Ted Petzold Visibility Laboratory Scripps Institution of Oceanography La Jolla, California 92037 25. Mr. W. J. Stachnik Optical Systems U.S. Navy Underwater Sound Laboratory Fort Trumbull New London, Connecticut 06320 26. Dr. Robert W. Holmes Marine Science Institute University of California Santa Barbara, California 93106 27. LCDR Margaret Anne Frederick, USN Command § Staff Course Naval War College Newport, Rhode Island 02840 28. Mr. Henry Odum Oceanographers Services Branch National Oceanographic Data Center Washington, D.C. 20390 29. Dr. Ronald Zaneveld Department of Oceanography Oregon State University Corvallis, Oregon 97331 30. Mr. David Norman W. R. Church Computer Center Naval Postgraduate School Monterey, California 93940 31. Dr. Kendall Carder Marine Sciences Center University of South Florida St. Petersberg, Florida 33701 32. LCDR Calvin Dunlap Department of Oceanography Naval Postgraduate School Monterey, California 93940 146 33. LT Gerald York 3 1265 Chelsea Way Concord, California 94521 34. Mr. Andrew Bakun 1 National Marine Fisheries Service c/o Fleet Numerical Weather Central Monterey, California 93940 35* Office of Naval Research 1 Code U60 Arlington, Virginia 22217 36. Office of Naval Research 1 Code 102 -OS Arlington, Virginia 22217 37. Office of Naval Research 6 Code 105 Arlington, Virginia 22217 38. Dr. Robert E. Stevenson 1 Scientific Liason Office Scripps Institution of Oceanography- La Jolla, California 92037 39* Commander, Oceanographic Systems, Pacific 1 PoO, Box 1390 FPO, San Francisco, California 966IO k0o Director, Naval Research Laboratory 6 Attention: Library, Code 2620 Washington, D* C. 20375 ill* Commander, Naval Oceanographic Office 1 Attention: Code I6I4.O Washington, D. C. 20390 i|2. Commander, Naval Oceanographic Office 1 Attention: Code 70 Washington, D. C. 20390 U3. NODG/NOAA 1 Rockville, Maryland 20882 bk» Director of Defense Research and Engineering 1 Office of the Secretary of Defense Attention: Assistant Director of Research Washington, D. C. 20301 U5. SIO Library . 1 University of California, San Diego P. 0. Box 2367 La Jolla, California 92037 147 ii60 Department of Oceanography Library University of Washington Seattle, Washington 98105 il7» Department of Oceanography Library Oregon State University Corvallis, Oregon 97331 Ut8 UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PAGE (When Datm Entrred) REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM 1 1. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER 4. TITLE (end Submit) STATISTICAL STUDIES OF WORLD-WIDE SECCHI DATA 5. TYPE OF REPORT & PERIOD COVERED Master's Thesis; March 1974 6. PERFORMING ORG. REPORT NUMBER 7. AUTHORf»; Gerald L. York; Lieutenant, USN 8. CONTRACT OR GRANT NUMBERf»; ONR Project Order # P04-0121 9. PERFORMING ORGANIZATION NAME AND ADDRESS Naval Postgraduate School Monterey, California 93940 10. PROGRAM ELEMENT. PROJECT. TASK AREA & WORK UNIT NUMBERS II. CONTROLLING OFFICE NAME AND ADDRESS Naval Postgraduate School 1 Monterey, California 93940 12. REPORT DATE March 1974 13. NUMBER OF PAGES 149 14. MONITORING AGENCY NAME & ADDRESSf// dl Iterant from Controlling Ollice) i 1 Naval Postgraduate School Monterey, California 93940 IS. SECURITY CLASS, (ol thla report) Unclassified ISa. DECLASSIFICATION/ DOWN GRADING SCHEDULE 16. DISTRIBUTION STATEMENT (otthla Report) Approved for public release; distribution unlimited. 17. DISTRIBUTION STATEMENT (of the abstract entered In Block 20, II dlllerent from Report) IB. SUPPLEMENTARY NOTES 19. KEY WORDS (Continue on reverae aide It neceaafuy and Identity by block number) Secchi Depths Upwelling Index Sea Water Transparency Phytoplankton Wet Volume Correlation Coefficient Analysis Optical Properties of Sea Multiple Regression analysis „ , t, -, h ' Forel Color 20. ABSTRACT (Continue on reverae aide It neceaoery end Identity by block nunxber) An investigation was made to determine possible correlations between Secchi depths and other simultaneously measured ocean- ographic parameters which were on file at the National Ocean- ographic Data Center as of March 1972. Sixty- three one-degree subsquares occurring in Japanese and Korean waters and eleven Atlantic and Pacific open ocean areas were chosen for linear correlation analysis using both sea surface data and mean DD 1 jan'73 1473 EDITION OF I NOV 65 IS OBSOLETE (Page 1) S/N 0102-014- 6601 I UNCLASSIFIFD 149 SECURITY CLASSIFICATION OF THIS PAGE (Khen Date Snterad) INCLASSIFIFD CkCIJHITY CLASSIFICATION OF THIS PAGECWhon Dull Entered) values of some fourteen different oceanographic parameters averaged over the Secchi depth. In particular, oxygen measurements exhibited trends toward an inverse proportionality with Secchi depth while temperature data indicated aspossible direct proportionality. Time series analyses of Secchi depths were performed and compared with upwelling indices computed for the Oregon coast and near Monterey Bay, California. An inverse proportionality and possible phase lag of mean Secchi depth compared to monthly upwelling index was observed. Multiple regression equations relating Secchi depth and upwelling index were calculated for both locations. DDlJann73 14?3 ^^ UNCLASSIFIFD S/N 0102-014-6601 SECURITY CLASSIFICATION OK THIS PAGEfWhen Deli Entered) 150 14 FE.&79 5 0Ci79 S OCT79 -&&& 2 5 U 5 9 152' '■■ Thesis \5l ^Statistical studies of >r Id-wide Seech, data. wot J4&W — e-5** -**} 25U5 9 Thes i s Y53 York c.l Statistical studies of world-wide Secchi data. lhesY53 Statistical .-.Jijdies of world-wide Secchi iii^'i'iiiiniiiiiiiiiiwiiiiiipiiiiiiiir' hi in mil i mi 3 2768 001 90527 6 DUDLEY KNOX LIBRARY