Concentration of particulate matter in the Eastern Gulf of Mexico; an indicator of surface circulation patterns.

Citation
Concentration of particulate matter in the Eastern Gulf of Mexico; an indicator of surface circulation patterns.

Material Information

Title:
Concentration of particulate matter in the Eastern Gulf of Mexico; an indicator of surface circulation patterns.
Creator:
Schlemmer, Frederick C.
Place of Publication:
Tampa, Florida
Publisher:
University of South Florida
Publication Date:
Language:
English
Physical Description:
vi, 82 leaves. 29 cm.

Subjects

Subjects / Keywords:
Mexico, Gulf of ( lcsh )
Oceanography -- Mexico, Gulf of ( lcsh )
Ocean currents -- Florida ( lcsh )
Dissertations, Academic -- Marine science -- Masters -- USF ( FTS )

Notes

General Note:
Thesis (M.S.)--University of South Florida, 1971. Bibliography: leaves 58-61.

Record Information

Source Institution:
University of South Florida
Holding Location:
Universtity of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
028406001 ( ALEPH )
06964832 ( OCLC )
F51-00010 ( USFLDC DOI )
f51.10 ( USFLDC Handle )

Postcard Information

Format:
Book

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

CONCENTRATIONS OF PARTICULATE MATri'EPL IN THE EASTERN GULF OF MEXICO: AN INDICATOR OF SURFACE CIRCULATION PATTERNS by Frederick C. Schlemmer II A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts in the Department of Oceanography in The University of South Florida December, 1971 Thesis supervisor: Dr Kendall L Carder

PAGE 2

Graduate Council University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Vl.AS TER. Is THESIS This is to certify that the Master1s Thesis of FREDERICK C. SCHLEMMER II with a major in Oceanography has been approved by the Examining Committee as satisfactory for the thesis requirement for the Master of Arts degree at the convocation of June 11, 1972 Thesis committee: Thesis supervisor Member Member

PAGE 3

ACKNOWLEDGMENTS The author wishes to express his appreciation to Dr. Kendall L. Carder for his advice and support in the research and development of this thesis. He also wishes to thank Dr. Thomas E. Pyle and Dr. Thomas L Hopkins for their advice concerning the geological and biological aspects of this investigation. The author also wishes to thank Mr. M. 0. Rinkel for coordination of the EGMEX Program which made this study possible. This research was supported by the Oceanography Section, National Science Foundation, NSF Grants GA 25991 and GA 29590. ii

PAGE 4

TABLE OF CONTENTS LIST OF 'fABLES iv LIST OF FIGURES v I. INTRODUCTION 1 II. PREVIOUS STUDIES 4 Physical Characteristics of the Eastern Gulf of Mexico 4 Indicators of Circulation Patterns 5 III. THEORY 7 Particle Analysis 7 Light-Scattering Theory 9 IV. EXPERIMENTAL PROGRAM 12 Introduction 12 Sampling Program 12 Laboratory Analysis 15 v. RESULTS AND DISCUSSION 17 VI. CONCLUSIONS 54 BIBLIOGRAPHY 58 APPENDIX A 62 APPENDIX B 71 iii

PAGE 5

Table 5.1 5.2 5.3 5.4 A.l B.l B.2 LIST OF TABLES Station position and hydrographic data associated with particle samples. Particle and optical data. Specific scattering-to-area ratios of some marine particles. Correlation coefficients and equations of lines of linear regression among particle parameters. Volumes and equivalent spherical diameters corresponding to current, amplification, and threshold Total particle volumes of preserved and unpreserved samples. R, m, and log a of Junge distribution for particles in Bayboro Harbor and MIZAR samples. iv 18 20 36 44 63 73 76

PAGE 6

Figure 4.1. 5.1. 5.2. 5.4. 5.5. 5.6. 5. 7. 5.8. 5.9. 5.10. 5.11. 5 12 5.13. 5.14. A.l. LIST OF FIGURES Station positions of particle samples. Sea-surface temperatures synoptic with particle samples. ( C). Topography of 22 isothermal surface synoptic with particle samples. Distribution of surface plankton concentration during EGMEX (ml/M3) (after Rinke 1, 1971). Contours of total volumes at sea surface. (xlO fJ?/ml). Contours of particle numbers at sea surface. (#lml). Contours of specific scattering-toarea ratio at sea surface (xlo-lm-6ster-l). Contours of the slope of the particle size distribution. Scatter diagram of N(50) vs. TV. Scatter diagram of N(lOO) vs. TV. Scatter diagram of /3(45) vs. TV. Scatter diagram of /5(45) vs. N(50) Scatter diag ram of /5(45) vs. N(lOO). Scatter diagram of N(lOO) vs. N(50) Scatter diagram of TV(lOO) vs. TV(50). Typical log-log graph of size parameter, x versus cumulative p article number, N. v Page 14 25 26 29 30 31 35 41 45 46 48 49 50 52 53 64

PAGE 7

Figure B.l. B.2. B.3. B.4. B.S. Graph of total particle volumes versus time. Log-log plot of particle size distribution of Bayboro Harbor samples. Log-log plot of particle size distribution envelope for Bayboro Harbor sample. Log-log plot of particle size distribution of MIZAR sample. Log-log plot of particle size distribution for MIZAR sample. vi 74 77 78 79 80

PAGE 8

I. INTRODUCTION The distribution of suspended particles is a recognized method of tracing circulation patterns and distinguishing water masses. This technique is most frequently used over relatively small areas such as near river deltas and estuaries, where a large active particle source flows into waters of lower particle concentration (Jerlov, 1953; Pak, 1970). It has also been used over larger areas (Carder et al., 1971), where there was only one predominant source of particles, providing basically only one particle type (e.g., organic or inorganic). This thesis will study the particle concentration and size distribution over a large oceanic area with multiple particle sources providing various particle types. Knowledge of the distribution of concentrations, sizes, and types of particles in the ocean has wide application in studies of light scattering, sedimentation, biological activity, and water mass identification. To date, there has been no reported large-scale investigation of the horizontal distribution of particle sizes or volume in the Gulf of Mexico. With its multiple particle sources (Loop Current, Mississippi River, and regions of upwelling over the Campeche Banks and Florida Escarpment (Bogdanov et al., 1

PAGE 9

1967), the Gulf should provide a somewhat more complex problem in investigating the recognizable differences in the distribution of sizes, concentrations, and types of particles over a large area of the open sea. The differentiation between areas of varying turbidity is relatively easy and is possible by various methods based on light scattering, particle-volume concentration, or filtration techniques. However, distinguishing among the particle types is less easily accomplished. This thesis will suggest a method for particle type differentiation based on a combination of the variations of size distribution and light scattering properties of the suspended particles. Several hypotheses are made concerning the distribution of particle types and sizes in the Gulf of Mexico. Variations in the particle types present are expected to exist as a result of the influence of the different particle sources. The Caribbean-derived waters of the Loop Current should be mostly organic with some inorganic particles resulting from scouring of sediments as it begins to pass through the Yucatan Channel. The organic particles indigenous to the Loop Current would be mostly detrital as compared with the relatively more living organic material introduced into the Loop Current as it passe s the regions o f upwelling along the Campeche Banks or the Florida shelf. The particles of the Mississippi River plume should have a large inorganic component. Th e s e varia t ions in p article types are expected to be indicated by variations in values of specific light-2

PAGE 10

scattering due to differences in the size distribution or indices of refraction of the particles (Pak et al., 1970). --The primary purpose of this thesis is to determine if the distribution of particles can be used to imply circulation patterns in the Gulf of Mexico. This will be deter-mined by comparing particle size distributions and optical characteristics with other proven circulation indicators such as sea-surface temperature and the depth of the 22 isotherm (Leipper, 1970). In order to determine this, the following questions need to be answered: 1. Can circulation patterns be discerned from the spatial variation of the following parameters? a. Total particle volume b. Particle number c. Ratio of specific light scattering to mean particle size 2. Do various size and light-scattering properties correlate well enough that the more easily ob-tained parameters can be chosen for use at sea? Using these questions as an outline of the problems to be investigated in this thesis, each will be developed and answered in the sections to follow. 3

PAGE 11

II. PREVIOUS STUDIES Physical Characteristics of the Eastern Gulf of Mexico The Gulf of Mexico has been characterized as a small ocean basin with generally restricted flow. For hydrographic studies it can be divided into eastern and western areas by a line drawn from the Mississippi River Delta to the Yucatan Peninsula (Ichiye, 1962; Grose, l9b6). Use of this division is advantageous in describing water-mass and circulation characteristics of the two areas. Several hydrographic surveys and many physical descriptions have been made of the eastern Gulf of Mexico (Parr, 1935; Leipper, 1954; Austin, 1955; Franceschini, 1955; Collier et al., 1958; Armstrong and Grady, 1967; Nowlin and McLellan, 1967; Leipper, 1970; Nowlin, 1971). Each one found that the primary circulation feature of the upper waters was the Loop Current, a tongue of Caribbean-derived water which enters through the Yucatan Channel and loops clockwise varying distances into the Gulf of Mexico, dependent primarily upon the season. It exits through the Straits of Florida as the Florida Current. Other notable hydrographic features in the eastern Gulf of Mexico include the regions of upwelling. Extensive upwelling has been reported over the Campeche Banks (Cochrane, 4

PAGE 12

5 1966; Bogdanov 1967; Drennan, 1967; Armstrong, 1969) along with reports of abundant phytoplankton (Zernova, 1969). Upwelling along the west Florida shelf is the subject of some disagreement. Chew (1955; 1961) suggested that upwelling in that area was improbable. However, Bogdanov al. (1967) reported upwelling along the shelf west of Tampa Bay to be active all year with that north of Tampa Bay occurring mostly in the summer. The major source of runoff in the eastern Gulf of Mexico is the Mississippi River. Its discharge in the upper layers is found to move to the northeast and east along the shelf regions (Scrutton and Moore, 1953; Chew, 1955; Ichiye, 1960; ., 1962; Ichiye, 1962; Tolbert and Salsman, 1964; Gaul, 1967). Parr (1935) found a tongue of lower salinity water, probably indicative of the Mississippi River water, following the edge of the Florida shelf. Indicators of Circulation Patterns There has been a continual search for indicators which can be used to imply or specify current patterns or to characterize water masses. Current-velocity determinations can be accomplished with current meters or geostrophic calculations; however, each method has significant limitations. Current meters are difficult to use in the deep sea and the number required to make a synoptic study of the eastern Gulf of Mexico would be prohibitively expensive. Geostrophic calculations require

PAGE 13

numerous lengthy oceanographic stations and considerable data reduction. The most expedient method for delineating circulation patterns in the eastern Gulf of Mexico is through the use of temperature. Lee (1967) and Stakhiv (1968) summarized the use of temperature distribution for tracing major oceanic currents. Lee (1967) found that the circulation in the eastern Gulf of Mexico could be outlined by the contours of 6 sea-surface temperatures during the fall, winter, and spring, but not in the summer as the surface waters over the Gulf of Mexico are heated to almost the same temperature. Leipper (1970) concluded that the topography of the 22C isothermal surface was the best indicator of the Loop Current in all seasons. In areas where particle sources exist such as where a river discharges turbid water into a less turbid region (Jerlov, 1953; Pak, 1970), suspended particle concentration has been used for tracing the river outflow. Of course, salinity is a good tracer of river discharge; but particle concentrations have also been used in regions where two water masses of different particulate loads meet. Carder et al. (1971) noted excellent agreement between the distri---bution of particle concentration a nd the circulation patterns where the turbid Peru Current and the relatively clear water of an anti-cyclonic gyre fed by the Equatorial Counter Current meet i n the eastern equatorial P acific Ocean.

PAGE 14

III. THEORY Particle Analysis It has been found that the particle size distributions in the ocean waters often can be described by the hyperbolic or Junge distribution (Bader, 1970; Brun-Cottan and Ivanoff, 1970; Brun-Cottan, 1971). This curve can be expressed by the equation 3.1 -m N=ax where N is the cumulative number of particles larger than the size parameter x (i.e. volume, diameter, radius, surface area, etc.) with a and m being constants. This distribution will describe a straight line when N and x are plotted on log-log paper (see Appendix A for further discussion and application). Definitions of the particle statistics used in this study are provided below. The three statistics are cumulative frequency distribution, total volume, and the mean particle cross section. The cumulative frequency tion is the distribution of the number of particles equal to or smaller than a certain volume. Since the Coulter Counter provides the number of particles larger than some 7

PAGE 15

volume, N0-N represents the cumulative frequency distribution, where N0 is the total number of particles in the sample. For the sake of simplicity, N will be called the inverse cumulative frequency in this thesis. Two points in the inverse cumulative frequency distribution will be referred to in this study: N(50) and N(lOO). N(50) is the total number of particles counted using the 50 micron aperture of.the Coulter Counter and corresponds to the number of particles larger than 1.34 cubic microns. N(lOO) is the number of particles counted using only the 100 micron aperture of the Coulter Counter and represents those particles larger than 6.14 cubic microns. By measuring the number of particles between successive size increments, the total particle volume for each increment can be calculated. The total volume, TV, is the summation of the particle volumes of each size increment and represents the total volume of the particles counted by the Coulter Counter. Computation of this quantity is covered in the data reduction section of Appendix A. The mean particle cross section, S, is the mean crosssectional area of the particles in the sample. It is calculated from the total particle cross section divided by the total number of particles. Further discussion of the computation of this statistic is provided in the data reduction section of Appendix A. 8

PAGE 16

Light-Scattering Theory A thorough and readable summation of the practical and theoretical techniques used in light-scattering studies in the ocean is given by Jerlov (1968). A detailed coverage of these techniques is unnecessary in the scope of this study. However, certain definitions and relationships need to be briefly described. When working with the amount of light scattered at an angle e from the direction of incident-beam propagation by the particles suspended in some scattering volume V, it is convenient and useful to relate the to a volume scattering function tJ for that angle. It is defined as 3.2 where I (e) is the radiant intensity scattered in the direction e when the scattering volume is illuminated by a light beam of irradiance E. It has the units (meter-steradian)-1 Jerlov (1953) showed empirically that the volume scattering function for e = 45 {3 ( 45)' was directly proportional to the total scattering coefficient, b, for seawater which is proportional to the total particle cross section (Jerlov (1968, p. 30). Deirmendjian (1963) verified this relationship theoretically for scattering ate= 40 by several different polydisperse distributions of particles. This relation has some usefulness in particle studies. 9

PAGE 17

10 Beardsley et al. (1970) studied samples taken in the eastern equatorial Pacific Ocean and attempted to determine how well various particle parameters correlated withAJ(45). In relating/J(45) to particle number, total particle cross section, and total particle volume, correlation coefficients of 0.90, 0.82, and 0.56 respectively were obtained. They explained the lower correlation with total volume by the fact that particles too small to be detected using the Coulter Counter were still important in the scattering process. As a method of distinguishing among particles with different indices of refraction, Pak et al. (1970) suggested the use of a light-scattering vector. The volume scattering function was divided by the number of particles in suspension to obtain specific light scattering,/5(45), an estimate of the light scattered by a single particle. The mean particle cross section, S, was determined from the particle size distribution. These two quantities form a unique two-dimensional vector representation of the water sample. Since the light-scattering vectors depend upon the indices of refraction and the diameters of the particles in sus-pension, particles of dissimilar sizes and indices of re-fraction can be distinguished. This method may provide in-sight into the differences in origin of particle samples if they are of significantly different sizes or indices of refraction. Variations in the light-scattering vector were used by Carder et al. (1971) in a study of particles in the

PAGE 18

eastern equatorial Pacific Ocean. They hypothesized that the increase in the specific light-scattering for a given particle cross section was the result of the decay of the soft organic parts of the particles, leaving the harder inorganic materials. 11 Although the light-scattering vector is not applied in this study, a numerical representation for it is determined by calculating the ratio between specific light-scattering and mean particle cross section. This ratio is termed the specific scattering-to-area ratio, /3(45)/S (see Appendix A for development and further discussion).

PAGE 19

IV. EXPERIMENTAL PROGRAM Introduction Since very little is known about the suspended matter in the eastern Gulf of Mexico on a volume or number basis, it should be very informative to study their variations. With particle distribution having been proven to be a suitable water-mass marker in areas such as near river mouths (Jerlov, 1953; Pak, 1970) or over a large oceanic area gl., 1971), it is hypothesized that variations in particle distribution could be used to imply circulation patterns in the eastern Gulf of Mexico. Because the circulation pattern of the eastern Gulf of Mexico is dominated 12 by the Loop Current, which is known to vary in position and current intensity in time and space, a survey using particles as circulation indicators requires synoptic particle samples to be taken over a large area both in and out of the Loop Current. The Eastern Gulf of Mexico (EGMEX) Program provided an excellent method for accomplishing this. A complete description of the philosophy and organization of EGMEX is provided by Rinkel (1971). In brief, it was a cooperative, near-synoptic, multi-ship, interdisciplinary study of the Loop Current in the eastern Gulf of Mexico. The survey was coordinated so that the maximum

PAGE 20

13 amount of data could be acquired simultaneously over the entire area. Positioning of the stations along the transects was based on historical information about the Loop Current for the month of May and on an aerial survey of sea-surface temperatures taken shortly before commencement of the program. Sampling Program The sampling program for EGMEX I covered almost the entire eastern Gulf of Mexico (see Figure 4.1). This survey included Coast Guard airborne radiation thermometer overflights to help define the boundaries of the Loop Current. and to determine the final positioning of the EGMEX stations. Standard hydrographic measurements of temperature and salinity in addition to the particle samples were taken at each of fifty-one stations. Concurrent chemical and biological sampling programs are described in Rinkel (1971). With synopticity of data of primary concern, the optimum method for obtaining samples over such a large area is to use several research vessels, each with a technician and particle-sizing instruments aboard to analyze the samples immediately. The difficulty in supplying the particlesizing instruments and personnel for every ship is apparent. The particle-sizing instrument used in this study was the Model "B" Coulter Counter (Coulter Electronics, 1968). This particular instrument is not suitable for use at sea except on large stable ships in the calmest of weather (Beardsley et al., 1970). For this reason, preservation of the particle samples was required.

PAGE 21

.. . 64P 1051 ........ .. ..... ... 86 . . ..... 59P 6 I P P 62P ... f5P 66P 1081 63P 69P 58 P 1151 1ll1 661 641 941 102l . 117f . 821 . I 19T 20T : 21 T : 22T. : . .. ..lo Oo. ., .. L '"T.. 26T .. ".._ 27T ............ 3T 2T IT -..54 A : 23T 25T 24T -- .T -:.. . . .. asA_/ A ,:' 57A.: . . .45A .. 0 4 4A -43A 0 o42A Figure 4 1 Station positions of particle samples. 14

PAGE 22

Lugol's solution, a preservative and staining agent used in work with phytoplankton, was chosen as the preservative because it was believed that the majority of the particles which would need preservation would be phytoplankton detritus. Since preservation of the samples was required, the effect of Lugol's solution upon the particle size distribution was determined (see Appendix B for discussion and development). 15 Sixteen ounce polyvinylchloride sample bottles had been prepared by washing with deionized water and filled with 10 milliliters of Lugol's solution (I-KI), also made using deionized water. The particles in the deionized water were counted later and found to be negligible. The bottles were sealed with plastic electrical tape to prevent evaporation, and were provided to each research vessel. Samples of surface water for particle analysis were drawn directly from the Niskin sampling bottles into the bottles with preservative to eliminate potential contamination from the use of an intermediate transfer v essel. Laboratory Analysis Particle size distributions of the samples w ere measured with the Coulter Counter (se e Appendix A for data reduction). A 50 micron aperture was used to measure particles with volumes between 1.34 and 6.14 cubic microns A 100 micron a perture measure d particles large r tha n 6 .14 cubic microns. The thresholds selecte d for use were based on a

PAGE 23

trial-and-error method using several samples to determine those providing best resolution in measuring the particle size distribution. Light-scattering measurements were made with the Brice Phoenix Light Scattering Photometer (Phoenix Precision Instrument Company, 1963). 16

PAGE 24

17 V. RESULTS AND DISCUSSION The particle size distributions of preserved surface water samples from the fifty-one stations were measured, and their light-scattering characteristics were determined. Table 5.1 lists position, date, time, sea surface temperature, surface salinity, and depth of the 22C isotherm for each station. The station numbers and hydrographic data are those supplied by the sampling ship. The letters, A, I, P, and T following the station number refer to the research vessel, ALAMINOS, ISLAND WATERS, PILLSBURY, and TURSIOPS, respectively. Table 5.2 lists the particle and light-scattering data measured or calculated for each sample. These data include the number of particles larger than 1.34 cubic microns, N(50); number of particles larger than 6.14 cubic microns, N(lOO); total volume of all particles counted, TV; volume of particles larger than 6.14 cubic microns, TV(lOO); an estimate of the volume of particles below the limits of resolution of the Coulter Counter, TV (<50); the ratio of the "uncounted" particles to the total volume measured, TV (<50) /TV; mean particle cross section, S; the volume scattering function,/](45); the specific volume scattering function,t$(45); and the specific scattering-to-area ratio,

PAGE 25

18 TABLE 5.1 Station position and hydrographic data associated with particle samples. Sta. Lat. Date Time SST s 22 (N) (W (GMT) (oC) ( 0/oo) Depth (M) 40A 23 06.0 83 48.5 5-5-70 04.0 27. 20 35.95 203 41A 23 21.0 83 42.0 5-5-70 07 .o 27.11 35.94 200 42A 23 34.0 83 35.5 5-5-70 12.0 27.12 35.92 188 43A 23 50.5 83 28.0 5-5-70 16.0 27.33 35.87 160 44A 24 07.5 83 20.7 5-5-70 22.0 27.15 36.01 116 45A 24 22.0 83 13.0 5-6-70 06.0 26.38 72 54 A 22 18.5 86 49.5 5-7-70 16.0 24.60 36.16 15 55A 21 34.7 86 26.0 5-8-70 06.0 27.12 36.02 45 56A 21 56.0 86 09.0 5-8-70 11.0 27.12 36.00 150 57 A 21 52.1 85 43.8 5-8-70 18.0 27.28 35.84 194 58A 21 59.5 85 23.5 5-8-70 23.0 27.26 35.96 213 59 A 22 06.5 85 11.5 5-9-70 04.0 26.37 35.98 195 57P 28 07.5 85 50.5 5-5-70 02.2 24.45 36.21 29 58P 27 50.0 86 20.0 5-5-70 08.6 24.23 36.26 35 59P 28 17.0 86 53.0 5-5-70 18.4 25.40 36.31 37 61P 28 05.0 87 20.0 5-7-70 10.5 25.20 36.38 48 62P 27 50.0 87 49.0 5-7-70 18.0 25.27 36.37 38 63P 27 33.4 87 17.0 5-8-70 00.2 27.13 36.46 37 64P 27 17.0 88 45.0 5-870 07.0 24.29 36.19 29 65P 26 55.5 88 28.0 5-8-70 13.9 24.80 36.38 36 66P 26 42.0 88 14.0 5-8-70 19.4 25.87 36.25 44 67P 26 58.5 87 47.5 5-9-70 01.3 26.04 36.15 43 68P 27 14.0 87 20.0 5-9-70 07.9 24.43 36.29 38 69P 27 33.0 86 51.0 5-9-70 14.2 25.65 36.22 53 64I 25 15.0 87 25.0 5-6-70 20.6 27.07 35.87 174 66I 25 30.0 86 45.0 5-7-70 00.7 27.10 35.90 187 69I 25 42.0 86 03.0 5-7-70 06.5 27.12 35.88 167 72I 25 54.0 85 19.0 5-7-70 12.5 26.79 36.02 110 75I 26 07.0 84 56.0 5-7-70 16.2 26.10 36.14 108 79I 26 17.0 84 28.0 5-7-70 18.7 2 6 .32 36.19 62 82I 26 40.0 84 oo.o 5-7-70 23.0 26.16 36.38 42 94I 27 30.0 84 15.0 5-9-70 20.7 24.32 36.25 36 102I 27 19.0 84 54.0 5-11-70 02.7 24.50 36.35 49 105I 25 19.0 88 51.0 5-11-70 18.7 24.12 36.51 49 lOBI 25 51.0 88 01.0 5-12-70 03.8 25.72 36.35 46 llli 26 21.0 87 11.0 5-12-70 12.5 27.02 36.03 120 115I 26 44.0 86 29.0 5-12-70 22.7 26.32 36.25 69 117I 26 58.0 85 31.0 5-13-70 06.2 25.11 36.57 41 lT 22 30.0 87 30.0 5-1-70 06.0 23.80 36.21 16 2T 22 51.0 86 55.0 5-1-70 15.7 26.80 47 3T 23 05.0 86 28.0 5-2-70 02.0 27.57 35.97 188

PAGE 26

19 TABLE 5 1 {con1d ) Sta. Lat. Date Time SST s 22 { N ) {W {GMT) { oc) ( 0/oo) Depth (M) 4T 23 29 0 85. 35 0 5-270 17 o 27. 30 36 00 208 19T 25 42 0 83 27 0 5-9-70 16 5 24 50 36.24 44 20T 25 29 0 83 47 0 5-9-70 20 9 25 50 36 19 41 21T 25 20 0 84 08 0 5-10-70 01 7 25 .50 36 .22 55 22T 25 03. 0 84 32. 0 5 -10-70 07 2 25. 50 36. 20 51 23T 24 41.0 84 59 0 5 -10-70 13 7 27. 00 36 00 121 24T 24 30 0 85 25 0 5 -10-70 19 2 27. 00 36. 01 189 25T 24 20 0 85 54 0 .. 5 -11-70 01 0 27 00 35 .92 209 26T 23 54 0 86 55. 0 5-11-70 11 2 27 40 35.91 175 27T 23 40 0 87 37.0 5-11-70 23 0 26.30 36.06 42

PAGE 27

20 TABLE 5.2a Particle and optical data.* .Station N(50) N(lOO) TV TV(lOO) TV(<50) (ml-1) (ml1 ) (xloV> (xlotL3> (xloV> 40A 11530 2152 9.26 6.88 8.60 41A 9764 1886 6.50 4.46 14.22 42A 9874 2094 7.90 5.92 9.98 43A 39890 5030 26.00 16.50 233.76 44A 15868 2910 11.44 8.04 26.62 45A 399386 145804 1233.82 1162.02 116.78 54 A 11734 3364 15.54 13.24 6.32 55A 12212 2826 15.58 13.08 11.40 56A 2610 788 5.60 5.02 1.18 57 A 5952 1210 5.12 3.84 7.00 58A 7158 1570 12.74 11.20 6.22 59 A 9686 2826 24.60 22.72 5.14 57P 1730 524 4.04 3.70 0.84 58P 4022 880 3.70 2.86 4.28 59P 3482 708 2.92 2.18 4.44 61P 1786 436 1.80 1.44 1.52 62P 5806 1262 7.04 5.82 5.42 63P 4122 864 3.54 2.70 5.14 64P 1832 400 1.50 1.12 2.06 65P 6264 1162 4.50 3.18 10.30 66P 5162 1304 6.30 5.26 3.88 67P 4544 1080 5.40 4.48 3.96 68P 3516 782 3.16 2.42 3.50 69P 2286 538 2.62 2.14 1.92 64I 8846 1634 5.22 3.38 16.02 66I 8584 1512 5.38 3.58 16.84 69I 5552 1116 3.40 2.28 6.88 72I 6010 690 3.54 2.02 8.40 75I 6832 1170 4.32 2.80 11.20 79I 5080 1176 3.88 2.86 4.72 82I 4122 1026 4.20 3.38 3.16 94I 9402 1560 5.78 3.72 17.56 102I 5610 1456 5.86 4.72 3.68 105I 7202 1372 6.34 4.98 28.90 108I 7176 1546 6.38 4.84 6.50 llli 5972 1170 4.42 3.10 7.02 115I 10512 1322 6.40 3.98 32.04 ll7I 12928 2420 9.30 6.52 19.44 lT 17296 5798 38.54 35.10 5.38 2T 6620 2098 8.34 7.10 2.78 3T 5420 1280 .4.68 3.48 4.62 4T 5416 1744 4.80 3.84 2.82 *(see text for explanation of symbols)

PAGE 28

21 TABLE 5.2a (con'd.) Station N(50) N(100) TV TV(100) TV(< 50) (m1-1) (m11 ) 19T 4814 1430 6.34 5.50 4.10 20T 5156 1502 5.28 4.32 3.12 21T 7912 1618 5.50 4.06 20.16 22T 9692 2006 7.00 5.00 12.40 23T 8474 2162 6.92 5.38 9.24 24T 6020 1462 4.80 3.62 5.72 25T 6020 1462 5.22 4.04 5.72 26T 8502 2000 6.46 4.86 9.02 27T 6424 2208 10.12 8.96 2.20

PAGE 29

22 TABLE 5.2b Particle and optical data., Station s {3<45) {45)/S TV (1_,2) ( w-2) ( 10 l l xlo-1 m:ster : 4-ster -6sterl 40A 0. 9285 3.6533 0.208 0.180 0.50 41A 2.1877 3.4584 0.254 0.260 o. 75 42A 1.2646 3.5508 0.227 0.230 0.64 43A 8.9879 3.1883 0.503 0.126 0.40 44A 2.3283 3.4893 0.290 0.183 0.52 45A 0.0946 7.3295 21.200 0.531 0.72 54 A 0.4070 4.8070 0.436 0.372 0.78 55A 0.7324 4.3215 0.383 0.314 0.72 56A 0.2114 5.6672 0.189 0.726 1.78 57 A 1.3672 3. 8128 0.171 0.288 0.76 58A 0.4877 4.9964 0.378 0.528 1.06 59A 0.2091 6.1737 1.280 1.322 2.14 57P 0.2066 5.7191 0.143 0.826 1.45 58P 1.1615 3.9415 0.164 0.408 1.03 59P 1.5200 3.6686 0.146 0.419 1.15 61P 0.8426 4.1359 0.159 0.891 2.15 62P 0.7696 4.2399 0.368 0.634 1.50 63P 1.4470 3.7707 0.211 0.512 1.36 64P 1.3629 3.7389 0.163 0.889 2.38 65P 0.4375 3.4676 0.226 0.361 66P 0.6152 4.4305 0.271 0.525 1.38 67P 0.7331 4.2481 0.127 0.280 0.66 68P 1.1118 3.9144 0.286 0.828 2.12 69P 0.7322 4.1441 0.135 0.591 1.83 64I 3.0658 3.2986 0.236 0.267 0.81 66I 3.1239 3.2744 0.288 0.356 1.02 69I 2.0177 3.3248 0.259 0.466 1.40 72I 2.3697 3.1805 0.192 0.320 1.00 75I 2.5997 3.3819 0.551 0.806 2.38 79I 0.1926 3.7433 0.292 0.575 1.54 82I 0.7520 4.2929 0.254 0.616 1.44 94I 3.0424 3.3272 0.764 0.812 2.44 102I 0.6266 4.4227 0.532 0.948 2.14 l05I 4.5640 3.8182 0.265 0.368 0.96 l08I 1.0177 4.0049 0.484 0.674 1.68 llli 1.5915 3.6552 0.257 0.340 1.18 115I 5.0061 3.2192 0.287 0.273 0.85 ll7I 2.0912 3.4829 0.739 0.572 1.64 lT 0.1396 6.1483 0.408 0.236 0.38 2T 0.3331 4.8017 0.169 0.256 0.53 *(see text for explanation of symbols)

PAGE 30

23 TABLE 5.2b (con 1 d. ) Station s /J(45)/f TV (1.12) ( 10) ( 0-1 J x10) m4-ster ( m -6ster 3T 0.9865 3.9201 0.132 0.244 0.62 4T 0.5900 4.1703 0.126 0.232 0.56 19T 0.6462 4.9592 0.203 0.422 0.85 20T 0.5913 4.4256 0.201 0.390 0.88 21T 3.6626 3.5002 0.675 0.853 2.44 22T 1.7722 3.5800 0.443 0.457 1.28 23T 1.3354 3.7099 0.225 0.266 0.72 24T 1.1956 3.8242 0.206 0.342 0.90 25T 1.0605 3.9545 0.219 0.364 0.92 .26T 1.3949 3.6952 0.322 0.378 1.02 27T 0.2183 5.5012 0.248 0.386 o. 70

PAGE 31

)3(45)/S. Development and calculation of these parameters are discussed in Appendix A. 24 Comparisons of the particle and optical property distributions with the circulation patterns, as indicated by the surface and 22 isotherms, follow. A discussion of the correlations among the various parameters is included for selection of the best parameters for future use. The storage, handling, and resuspension of the particles caused a loss in total particle volume and an increase in the slope of the particle size distribution plot due to particle disintegration (see Appendix B for development and discussion). This loss of volume and increase in slope are indio-ative of some particle disintegration to sizes below the resolution of the Coulter Counter. The particles counted in the preserved samples do not represent the actual particle concentrations in the Gulf of Mexico. However, the distri-bution of their volumes and numbers is indicative of the relative distribution of particle concentration because the total range of particle concentrations found in the Gulf of Mexico is much greater than even the maximum (35% ) loss of particles encountered. Therefore, the total volume and particle number of the preserved samples used in this study and compared with hydrographic indicators of circulation patterns should be relative representations of particle concentrations in the Gulf of Mexico. Figures 5.1 and 5 2 respectively are contours of the sea-surface-temperature distribution and depth of 22C

PAGE 32

.. ... .. I ., .. 3o .; oo ;. 27 11181 ': e,.$ .. ,2e. I .. 26 . \ ...._ . Figure 5.1. Sea-surface temperatures synoptic with particle samples. ( C). 25 28 28 24 22

PAGE 33

. BB .. ,. 86 "'! 26 .. . 28 26 4 ... . . ... 0 0 0. 0 .. 0 . .. : .. 22 Figure 5.2. Topography of 22 isothermal surface synoptic with particle samples.

PAGE 34

isothermal surface based on the hydrographic data synoptic with the particle samples. The Loop Current is indicated from the sea-surface temperature (Lee, 1967) as the band of 27 water entering through the Yucatan Channel, looping into the Gulf of Mexico, and exiting through the Straits 27 of Florida. It can also be identified by sharp topographic gradients of the 22C isotherm (Leipper, 1970). Upwelling over the Campeche Banks and along the Florida shelf is indicated by the lowered sea-surface temperature and the shoaling of the 22C isotherm. The "eye" of the Loop Current is the region bounded by the inner 27C surface isotherm extending northward from Cuba. In Figure 5.2 the relatively low gradient area at the center of the topography of the 22 isotherm also defines the "eye" of the Loop Current. The regions of maximum current speed are indicated by the areas having larg e 22 C topographic gradients (contour bunching). A comparison of Figures 5.1 and 5.2 indicates that the deep circulation of the Loop Current is outlined by the 100 meter contour of the 22C isotherm. It shows a deep meander to the east until it nears the edge of the Florida shelf where it begins to follow the 200 meter bottom contour south and throug h the Straits of Florida. The surface temperatures indicate that surface waters of the Loop Current boundaries are intruding over the shelf regions along the eastern edge of Campech e Banks and along the Florida shelf near Fort Myers. The 25C contour gives evidence that the

PAGE 35

boundary waters are overrunning the shelf regions 1n the north also. However, the contours cannot be connected due to lack of data. Two tongues of cold water appear to isolate a pocket of 27C water and isolate the meander. 28 Figure 5.3 is a plot of the surface distribution of biomass measured in total displacement in milliliters of zooplankton materials sampled per cubic meter of water (after Rinkel, 1971). High plankton concentrations are noted over the Campeche Banks, along the edge of the Florida shelf, and in a tongue centered about 28N-87' W. A low plankton concentration is found in the region of the Loop Current and another low concentration is found in a tongue centered about 27N-87' W. Figures 5.4 and 5.5, respectively, are contoured areal plots of total particle volume and particle number, N (100). Comparisons of these contour patterns to those of sea surface temperature show good agreement. Upwelling over the Campeche Banks and at the southwest tip of the Florida shelf is well defined by the increased particle volume and number. A tongue of lower particulate load is seen entering the Gulf of Mexico through the center of the Yucatan Channel, corresponding well to the entry of the Loop Current. An increase in the particle volume and number is noted at the station at the western tip of Cuba with a decrease in a westerly direction toward the center of the Yucatan Channel. An increase was also found at the southernmost station in the transect across the Straits of Florida. The increase

PAGE 36

Figure 5.3. Distribution of surface plankton concentration during EGMEX (ml/M3) (after Rinkel, 1971). 29

PAGE 37

---. /'.... I 0-:-:---.._ ... .. ..... .... -. 2 Figure 5.4. Contours of total particle volumes at sea surface. (x 30 28 24 22

PAGE 38

. .. ..... . .... .... 1 2000 : : . ; ' . ... Figure 5 5 Contours of particle numbers at sea surface. (#/ml). 31 28 26 4 2 2

PAGE 39

32 is generally consistent with the increase in surface plankton concentration found along the northern coast of Cuba (Figure 5.3), although the total particle volume concentration at the tip of Cuba is much higher than would be expected on the basis of plankton concentration. The increase at the western tip of Cuba could be attributed to the scouring of sediments by the Loop Current as it passes close to land or to terrigenous input. An area of high particle is found in the northern portion of the entrance to the Straits of Florida. This could be the result of a combination of a high surface plankton concentration shown in this region in Figure 5.3 and upwelling as the L oop Current leaves the shelf to the north. The tongue of water with lower particle content extending from the Yucatan Channel into the Gulf of Mexico is in good agreement with the isotherm delineatingthe Loop Current. The contour of total volume almost parallels the boundary of the Loop Current to about 27N indicating entrainment of the highly productive waters over the Campeche Banks. A slight bulge in both the total volume and number contours near the northeast corner of the Campeche Banks could be attributed to the scouring and upwelling action of the water crossing the shelf regions in addition to the generation of particles in upwelling regions. The 5.0 (xlo+]!3/ml) contour of total volume follows the axis of both the surface and deep circulation of Loop Current as it extends north. A tongue of cold, less

PAGE 40

turbid water is found pushing in from the west at 271 N and joining the less turbid waters of the Loop Current. A wedge of more turbid water extends from the north at 281 N, 88 1 r.r and i t f th n n rudes onto the boundary o e Loop Current. A northern meander is seen in the contours of sea-surface temperatures as extending northward toward the edge of the west Florida shelf. This is reflected as a decrease in both the volume and number of particles. The tongue of clear water, the wedge of turbid water, and the pocket of lower particle content are in agreement with pat-terns of surface plankton concentration in this region (Figure 5.3). The boundary of the Loop Current is found to leave the Florida shelf near 28N and 85' W, loop into the Gulf, and then meander back over the Florida shelf at about 27N It begins to leave the shelf again at about 26N with its axis of flow following the edge of the Florida shelf as it proceeds toward the Straits of Florida. These meanderings are reflected in the contours of both particle volume and number. The area between the departure from the Florida shelf of the Loop Current boundary at 28N and its return at 27N is marked by a tongue of water with higher particle content indicative of more turbid shelf water. As the Loop Current returns to the shelf the contours of lower particle volume and number follow the same path as the surface iso-therm. The particle volume and number contours in the area where the Current leaves the Florida shelf extend farther 33

PAGE 41

34 into the Gulf than the isotherms. This could be the result of scouring upwelling as the boundary of the Loop Current parallels the edge of the Florida shelf where it nears the Straits of Florida. It is seen from these comparisons that the contours of particle volume and number are in good agreement with the shallow circulation patterns delineated by the sea-surface-temperature contours. Particle volume is in better agreement than is particle number with the deep circulation patterns outlined by the 100 meter contour of the 22 isothermal surface topography. Figure 5.6 is a contoured areal plot of the distribution of the specific scattering-to-area ratio. The specific scattering-to-area ratio is the ratio of the components of the light-scattering vector described by Pak &1 (1970). It is useful in implying relative fractions of organic and refractory organic/inorganic material in the particle samples because of the difference in their indices of refraction. The value of this ratio should increase with an increase in the inorganic fraction. Table 5.3 lists the specific scattering-to-area ratios generated from the light-scattering vectors of a number of marine particle types reported by Pak et &1 (1970). A range of values can b e seen in this table with the specific scattering-to-area ratio of the calcareous ooze being approximately sixteen times that of the Isochrysis galbana, a small (4-lqL) flagellated, n a k e d phytoplankter. This represents a range that seems to vary directly with the

PAGE 42

88 ', ... . 1.2 1.5 ---- 0. 0 .. '(. 1.0 0.5 .. I. 2 5 o. 75 1.0 84 .... 0 0 ..... 0 0 .. 5 . .. Figure 5.6. Contours of specific ratio at sea surface (x lo-1 m-6ster-l). 35 28 26 24 2

PAGE 43

TABLE 5.3 Specific scattering-to-area ratios of some marine particles. Particle type !1(45)/S (m -ster-1 ) Calcareous ooze Kaolinite Pacific clay Quartz Coccolithophore, Coccolithus sp. Coccolithophore, Syracosphaera sp. Thalassiosira nordenskioldii Carteria sp. Isochrysis ga1bana 1.00 0.90 0.40 0.30 0.25 0.10 0.38 0.20 0.27 0.23 0.14 0.12 0.06 36

PAGE 44

index of refraction or refractory properties of the particles. The specific scattering-to-area ratios of the particle samples used in this study cannot be directly related to.Table 5.3. Those in Table 5.3 were generated from samples of a single particle type while the samples used in this study represent a mixture of various particle types and sizes. Additionally the particle samples used by Pak (1970) contained a concentration of particles which were of sizes large enough to be counted by the Coulter Counter. The particle samples used in this study contain many particles which are below the resolution limits of the Coulter Counter but are still optically active. Although the specific scattering-to-area ratios in this study be directly related to those in Table 5.3, they do exhibit 37 a four-fold range of values. The range of values expected in a non-homogeneous, "natural", particle sample would be smaller than that of homogeneous particle samples. It can be seen in Table 5.3 that the specific scattering-to-area ratios of the phytoplankton tend to be lower than those of the inorganic particles. Although the types of particles in the oceanic water samples could not be identified from the specific scattering-to-area ratios, these values are useful in showing trends in variations in the mixtures of various particle types as the fractions of organic and refractory organic/inorganic particles change. There is insufficient information about the light scattering properties of many types of suspended marine particles to make a definite

PAGE 45

sta.tement about the compos.i ti_on of the particle samples, consequently only inferences can be made at this time using this method. 38 The areas of upwelling over the Campeche Banks and at the southwest tip of the Florida shelf are distinguished as having a low value for the ratio. The contours of low ratios along the eastern edge of the Campeche Banks extend northward as if entrained in the current along the Campeche Banks. This is consistent in position with the band of water with high surface plankton concentration found extending north (Rinkel, 1971) and the tongue of 24.5C water looping north in Figure 5.1. The "eye" of the Loop Current is also marked by particles with low ratio. This could be attributed to the carrying of organic material reported by Rinkel (1971) along the north coast of Cuba .into the 11eyeff as the Loop Current moves north. The low ratio of the tongue of water extending into the Gulf through the Straits of Florida is consistent with the increase in surface plankton concentration (Rinkel, 1971). A tongue of water ( -1 -6 -1) an average value ratio 1.25xl0 m -ster is found extend.ing into the Gulf through the Yucatan Channel indi cat.ing that the Caribbean waters probably are not predominately either organic or refractory organic/inorganic (probably biogenous detritus). The increase in surface plankton concentration along the north coast of Cuba (Rinkel, 1971) is found to have a low value for the specific scattering-to-area ratio. The

PAGE 46

ratio at the station at the western tip of Cuba is high, consistent with the hypothesis that the increase in total volume and number here could result from scouring or land runoff at the west tip of Cuba. 39 A tongue of water with a higher ratio centered about 26N is wedging in from the west, indicating that water with a different particulate composition could be influencing the total volume in that region. This is not seen in the total volume contour of Figure 5.6. The cold tongue centered about 271 N has a high ratio, indicating that although it is relatively clean, the dominant fraction could be inorganic or refractory detrital. A tongue or water with a high ratio is found originating near the area of high particle load at about 28N and extending south along the edge of the Florida shelf to about 25N. Although the water in this bulge might be expected to have a high ratio value because of a possible terrigenous contribution of the Mississippi River to this area, its effects are not expected to extend quite so far south as indeed the high salinity in this region indicates. The continuous nature of the tongue may be due to vertical transport of refractory organic/ inorganic particles from the bottom as a result of upwelling along the Florida shelf. The water in the center of the Gulf of Mexico has an average value (1.0-1.5 meters-6-ster-1 ) for the ratio indicating that one particle type probably does not predominate. An increase in refractory detritus would show a specific scattering-to-area ratio value between

PAGE 47

that of soft organic material and that of refractory organic or inorganic material. This would also lead to more average values in the less productive interior of the Gulf of Mexico. 40 Good agreement is found among the contours of the specific scattering-to-area ratio and the regions of varying particle types. The best agreement is in the south where the regions of upwelling and high surface plankton concentration are well defined in relation to the Loop Current. Agreement is also seen with the meanderings of the boundary of the Loop Current as it weaves across the edge of the Florida shelf. Figure 5.7 is a contour plot of the slope of the loglog size distribution graph (see Appendix A) of the population of the smaller particles (1.34-3.88fLin diameter) at each station. Variations in the slope of this graph are indicative of tendencies toward relatively larger or smaller particle sizes in the distribution. A low value of the slope would indicate the presence of larger particles, probably coincident with regions of high productivity. The high values would be indicative of a paucity of largersized particles in the distributions. The Loop Current is characterized by particle size distributions of intermediate slope compared to those of surrounding waters. A tongue of water with a slope of less than 3.0 is found extending along the edge of the Campeche Bank almost paralleling the contours ofj5(45)/S. A tongue

PAGE 48

. ... .. , ... ' .. -' --3.0 -3.0 . . ..... . ...... -3.0 -.0 ... ... : -4.0 . ... Figure 5 7 Contours of the slope of the particle size distribution. 41 28 26 24 22

PAGE 49

of water with a large distributional slope is found to intrude from the west at about 25 301 N. Water in the "eye" of the Loop Current is also shown to have smaller slopes than those of the surrounding Loop Current water. These contours agree well with the contours ofj$(45}/S for this area found in Figure 5 .6. The contours of particle Size distributional slope would imply a difference in the type of particle population found at stations 43A and 44A. Although particles at station 43A probably are predomininantly organic as suggested by the low value oftj(45}/S, they appear to be predominantly small particles. Station 45A is also an area of high productivity but apparently includes a population of larger-sized particles.in the. distribution. This could be due to the bloom of a largersize plankter. 42 Variations in particle distributional slopes appear to be a good indicator of circulation patterns. They also seem to be another indicator of differences among the sources of the particles present and could give additional insight in their origins. Having studied the worth of the particle and optical parameters as indicators of .circulation patterns, an interest arises in determining how well various particle param-eters correlate. High correlation would allow for the selection of the more easily acquired parameters for future use and the elimination of the ones which are more difficult to obtain.

PAGE 50

43 In order to interrelate the various particle parameters, scatter diagrams were and correlation coefficients were calculated to compare each parameter with the others. As a result of the large range of values for the various parameters, the scatter diagrams contain the maximum number of points that can be plotted allowing for an adequate separation of points. As an example, to include the points for particle number, total particle volume, or volume scattering function for station 45A would require that the majority of the other points be plotted as an indistinguishable lump near the origin. Table 5.4 is a summary of the scatter diagrams, their associated correlation coefficients, and the equation for the line of linear regression through the data points. Two correlation coefficients and two linear regression lines are listed, one without station 45A and one with station 45A. This is done because values from 45A are two orders of magnitude greater than all other data. Consequently these data will not be included in the statistics discussed here. Figures 5.8 and 5.9 compare the particle number found using the 5 0 micron and 100 micron apertures respectively to the total particle volume. The plot relating TV to N(50) shows much more scatter than that of TV and N(lOO). The correlation coefficient for TV versus N(50) is 0.7082 where that of TV versus N (lOO) is 0.9211. The difference s between these correlations is the result of variations in

PAGE 51

44 TABLE 5.4 Correlation coefficients and equations of lines of linear regression among particle parameters. Figure Parameters Correlation Linear Regression Number Compared Coefficient Equation (y-x) With Station 45-A 5.8 N(50)-TV 0.9964 y=2772.6587+319.6x 5.9 N(lOO)-TV 0.9997 Y=379.2119+118x 5.10 {3(45)-TV 0.9936 y=0.2237+.03398x 5.11 /J(45)-N(50) 0.9890 y=0.05917+1.054x 5.12 /](45)-N(lOO) 0.9930 y=0.1147+2.888x 5.13 N(l00)-N(50) o. 9971 Y=-622.8825+3656x 5.14 TV(l00)-TV(50) 0.99997 y=6114.1235+9427x Figure Parameters Correlation Linear Regression Number Compared (y-x) Coefficient Equation Without Station 45-A 5.8 N(50)-TV 0.7082 y=l500.0674+619.3830x 5.9 N(lOO)-TV o. 9211 y=288.9414+142.948lx 5.10 /](45) -TV 0.4766 y=0.2007+0.03017x 5.11 /]< 45) -N (50) 0.3575 y:0.2149+ 0.2588xlo-4 x 5.12 {3< 45) -N ( 100) 0.3992 y=O.l795+ O.l629xlo-3x 5.13 N(l00)-N(50) 0.8343 y=2 59.2403+0.1480x 5.14 TV(l00)-TV(50) 0.9847 y=0.2872+0.8646x

PAGE 52

N(50) (#/m1) 16000 14000 12000 10000 8000 6000 4000 2000 2 4 .. ... . 6 8 10 12 14 16 18 TV (j.L3x1o4 /m1) Figure 5.8. Scatter diagram of N(50) vs. TV. 45

PAGE 53

46 4000 3000 N(100) (#/m1) .. 2000 . .. ' 1000 .. 2 4 6 8 10 12 14 16 18 TV CfJ}x104 /m1) Figure 5.9. Scatter diagram of N(100) vs. TV.

PAGE 54

the size distribution among the samples. A sample with a relatively large number of small particles (large slope} will have a high N(50), a lower N(lOO), and a lower volume than does a sample with a distribution with relatively more large-size particles. For this reason N(lOO) is a better indicator of TV than N(50). 47 Figure 5.10 is a plot of/5(45) versus TV. Considerable scatter can be seen in this diagram. The correlation coefficient for these data is 0.4766. The scatter in this plot is the result of variation in the indices of refraction of the particles as well as the relative fractions of particles too small to be detect.ed by the Coulter Counter but still optically active. Beardsley et al. (1970) reported a correlation of 0.56 between/5(45) and the particle volume at stations in the eastern equatorial Pacific Ocean. They attributed the low correlation to uncounted particles that were causing light scattering. The addition of Lugol's solution to the sample probably causes a distortion due to selective absorption by the various particles. This should cause an even greater variation in the apparent refractive indices of the particles. Figures 5.11 and 5.12 are comparisons oftJ(45) with N(50) and N(lOO). Considerable scatter is evident in both plots with correlation coefficients being 0.3575 and 0.3992 forJJ(45) versus N(50) and N(lOO) respectively. Beardsley et (1970) found a correlation coefficient of 0.90 between light scattering at 45 and the particle number in

PAGE 55

48 1.4 1.2 1.0 0.8 0.6 0.4 .. .. : 0.2 . . .. 2 4 6 8 10 12 14 16 18 TV (fi3x104/m1) Figure 5.10. Scatter diagram of {J( 45) vs. TV.

PAGE 56

49 1.6 1.4 1.2 1.0 0.8 0.6 0.4 . 0.2 2000 4000 6000 8000 10000 N(50) (#/m1) Figure 5.11. Scatter diagram of /](45) vs. N(50).

PAGE 57

{j(45) (x1o2)(m-ster)-1 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 .. . . . .... . . . .. . 1000 2000 3000 N(100) (#/m1) Figure 5.12. Scatter diagram of/5(45) vs. N(100). 50

PAGE 58

the eastern equatorial Pacific Ocean. It is believed that the poor correlation between these parameters found in this study is caused by a combination of differences in indices of refraction, accentuated by the selective staining of the Lugol's solution, and the variations in the shape of size distribution curves among the particle samples. Figure 5.13 shows a graph of N(lOO) versus N(50) The plot indicates a fair grouping around a linear function. A correlation coefficient of 0.8343 is found between these parameters. Figure 5.14 is a plot of TV(lOO) versus TV(50). An excellent correlation exists between these two parameters with the coefficient of correlation being 0.9847. This is interpreted as an indication of the minor influence upon total particle volume of the additional particles counted with the 50 micron aperture. From the preceding discussions, certain conclusions have been made concerning the utilization of particle concentrations as relative circulation indicators. These conclusions are summarized in the following section. 51

PAGE 59

N(100) (#/m1) 3000 2000 1000 .. 2000 . .. .. 4000 6000 8000 N(50) (#/m1) Figure 5.13. Scatter diagram of N(100) vs. N(50). 52 10000

PAGE 60

TV(100)
PAGE 61

54 VI. CONCLUSIONS The stations sampled in this study covered a wide range of oceanic conditions: the upwelling over the Campeche Banks and along the Florida shelf, the intrusion of the Caribbean-derived Loop Current, and the eastern Gulf shelf water. These diverse water masses would suggest significant variations in particle volume, number, and type. With the use of contoured maps of particle volume and number, specific scattering-to-area ratio, and the slope of the particle size distributions, the regions of upwelling, shelf water, and the Loop Current were readily identifiable. The upwelling regions were found to have high particle concentrations with low specific scattering-to-area ratios and low size distributional slopes relative to the particles in the Loop Current. The regions of scouring or upwelling of sediments had high particle concentrations associated with high specific scattering-to-area ratio and low size distributional slopes. The Loop Current was characterized by low particle content, average specific scattering-to-area ratio, and intermediate slopes of particle size distributions. From these observations, it would seem that particle volume and number are each reasonably good indicators of circulation patterns in regions of strong circulation with varying

PAGE 62

particle sources such as the eastern Gulf of Mexico. In addition, specific light scattering-to-area ratios and slopes of the particle size distributions appear to provide some insight into the type of particles predominating in each area. The correlation among the various particle parameters suggests several methods of approach in future studies. 55 Most obvious was the low correlation of light scattering at 45 degrees with total volume or number in the preserved particle samples. This eliminates the use of this optical method as a substitute for these parameters. On the other hand a high correlation would have limited the potential value of the specific scatter-to-area ratio. The excellent correlation of the total volume using both the 50 and 100 micron apertures with that using only the 100 micron aperture suggests that the inclusion of the 50 micron aperture is unnecessary and use of the 100 micron aperture will be adequate in future surveys of this type. The high correlations of particle number determined using only the 100 micron aperture, N{lOO), with total volume, TV, as well as with the number of particles counted with the 50 micron aperture, N{50), support the use of only the 100 micron aperture in the future. The relative ease with which the cumulative number of particles can be counted recommend it as the most useful parameter in implying circulation patterns with particles. A complete particle size distribution

PAGE 63

could be run and a total volume calculated if time and operational considerations permit. With the use of more advanced and automated equipment in particle sizing, more work can be done on shipboard and the need for preservation of particle samples will be reduced. However, synopticity will still be a problem unless equipment and personnel are placed aboard each vessel for immediate sample analysis or particle preservation techniques are perfected. It would be of great interest to continue particle studies in the Gulf of Mexico equipped with automated particle sizing equipment suitable for shipboard use. In this way unpreserved particle samples could be taken from the water column to determine the suitability of inferring vertical variations in the circulation patterns using particles a s Austin (1971) did using biological indicators. 56 It is believed that the use of staining techniques in conjunction with light scattering has some potential in differentiating among particles of different types if light of varying wavelengths is used. By selectively staining a particular type of material, the imaginary (absorptive) part o f the index of refraction relative to a particular wavelength of light can b e changed and the particles of that material might be easily identified. Further studies of the specific scattering-to-area ratios of various plankton cultures a nd suspe n de d minerals

PAGE 64

57 would provide a better basis for the use of this ratio in discussing the composition of particle samples. This also would be useful in improving the light-scattering techniques for the estimation of particulate load in water containing heterogeneous particle types.

PAGE 65

BIBLIOGRAPHY Armstrong, Reed. 1969. Late-winter waters of Yucatan Straits-a 1968 GERONIMO survey in the Gulf of Mexico. Commercial Fisheries Review. 31(2): 33-36. 58 Armstrong, Reeds. and Grady, J. R. 1967. GERONIMO cruises entire Gulf of Mexico in late winter. Commercial Fisheries Review. 29(10): 35-40. Austin, G. B. 1955. Some recent oceanographic surveys of the Gulf of Mexico. Transactions, American Geophysical Union. 36(5): 885-892. Austin, H. M. 1971. The characteristics and relationships between the calculated geostrophic current component and selected indicator organisms in the Gulf of Mexico Loop Current system. Ph.D. thesis. Florida State University. 369 p. Bader, H. 1970. The hyperbolic distribution of particle sizes. Journal of Geophysical Research. 75(20): 2822-2830. Beardsley, G. F., Jr., Pak, H., Carder, K. and Lundgren, B. 1970. Light scattering and suspended particles in the eastern equatorial Pacific Ocean. Journal of Geophysical Research. 75(15): 2837-2845. ------Bogdanov, V., Sokolov, A., and Khromov, N. s. 1967. Regions of high biological and commercial productivity in the Gulf of Mexico and Caribbean Sea. Oceanology. 8(3): 371-380. Brun-Cottan, J. c. 1971. Etude de la granulometrie des particules marines mesures effectuees avec un Compteur Coulter. Cahiers Oceanographigues. 23(2): 193-205. Brun-Cottan, J. D. and Ivanoff, A. 1970. Particles' size distribution in sea water. (abstract) presented at 'The Ocean World': Joint Oceanographic Assembly. IAPSO, IABO, CMG, SCOR. Tokyo, Japan. September, 1970.

PAGE 66

Burt, W. V. 1956. A light scattering diagram. Journal !Marine Research. 15: 76-80. Carder, K. L. 1970. Particles in the eastern equatorial Pacific Ocean: Their distribution and effect upon optical parameters. Ph. D. thesis. Oregon State University. Corvallis, Oregon. 140 p. Carder, K. L., Beardsley, G. F., Jr., and Pak, H. 1971. Particle size distributions in the eastern equatorial Pacific Ocean. Journal of Research. 76(21): 5070-5077. -Chew, Frank. 1955. On the offshore circulation and convergence mechanism in the Red Tide region off the west coast of Florida. Transactions, American Geophysical Union. 36(6): 963-974. Chew, Frank. 1961. Some implications of the highly saline water off the southwest coast of Florida. Journal Q! Geopnysical Research. 66(8): 2445-2454. Chew, F., Drennan, K. L., and Demoran, W. J. 1962. Some results of drift bottle studies off the Mississippi Delta. Limnology and Oceanography. 7(2): 252-257. Cochrane, John D. 1966. The Yucatan Current, upwelling of northeastern Yucatan, and currents and waters of western equatorial Atlantic. Texas A&M University Reference 66-23 T. 59 Collier, A., Drummon, K., and Austin, G. 1958. Gulf of Mexico, Physical and chemical data from ALASKA Cruises, with a note on some aspects of the physical oceanography of the Gulf of Mexico. Special Scientific Report, Service. Bulletin No. 249. 417 p. Coulter Electronics. 1968. Instruction and Service Manual for the Model B Coulter Counter. Hialeah, Florida. -Deirmendjian, D. 1963. Scattering and polarization properties of poly-disperse suspensions with partial absorption. In: I.C.E.S. Electromagnetic Scattering. London: Pergamon. 5: 171-189. Drennan, K. L. 1967. Investigations of sea surface temperature patterns in the Gulf of Mexico as determined by an airborne infrared sensor. Gulf South Research Institute, Department of Oceanography. Environmental Sciences and Engineering Laboratory, New Iberia, Louisiana. Reference 67-0-1.

PAGE 67

60 Franceschini, G. A. 1955. Reliability of commercial vessel reports of sea surface temperatures in the Gulf of Mexico. Bulletin of Marine Science 2f the Caribbean. 5(1): 42-51. Gaul, R. D. 1967. Circulation over the continental margin of the northeast Gulf of Mexico. Ph. D. dissertation. Texas A&M University. Grose, P. L. 1966. The stratification and circulation of the subsurface waters of the Gulf of Mexico. M. S. thesis. Florida State University. Department of Oceanography. 84 p. (not seen) cited in Lee, T. 1967. Sea surface temperatures as related to circulation in the Gulf of Mexico. M. s. thesis. Department of Oceanography. Florida State University. 42 p. Ichiye, T. 1960. On the hydrography near the Mississippi Delta. Oceanographical Magazine. 11(2): 65-78. Ichiye, T. 1962. Circulation and water mass distribution in the Gulf of Mexico. Geofisica International. 2(3): 47-76. Jerlov, N. G. 1953. Particle distribution in the ocean. Report Q! Deep-Sea Expedition. 3: 73-97. Jerlov, N. G. 1968. Optical Oceanography. New York: Elsevier. 194 p. Lee, T. 1967. Sea surface temperatures as related to circulation in the Gulf of Mexico. M. S. thesis. Department of Oceanography. Florida State University. 42 p. Leipper, D. 1954. Physical oceanography of the Gulf of Mexico. In: QB1f Q! Mexico, its origin, waters, and marine life. U. s. Fisheries Bulletin Number 89. VOlume 54-5s:--p. 119-137. Leipper, D. 1970. A sequence of current patterns in the Gulf of Mexico. Journal Q! Geophysical Research. 75(3): 637-657. Nowlin, W. D. 1971. Water masses and general circulation of the Gulf of Mexico. Oceanology International. February, 1971: 28-33. Nowlin, W. D. and McLellan, H. J. 1967. A characterization of the Gulf of Mexico waters in winter. Journal !Marine Research. 25(1): 29-59. Pak, H. 1970. Th e Columbia Rive r a s a source of marine light scattering particles. Ph. D. thesis. Oregon State University, Corvallis. 110 p.

PAGE 68

Pak, H., Beardsley, G. F., Jr., Heath, G. R., and Curl, H. 1970. Light-scattering vectors of some marine particles. 15(5): 683-687. Parr, A. 1935. Report on hydrographic observations in the Gulf of Mexico made during Yale Oceanographic Expeditions on the MABLE TAYLOR in 1932. Bulletin of the Bingham Oceanographic Collection. Yale University. 5(11): 1-93. Phoenix Precision Instrument Company. 1963. Brice Phoenix Light Scattering Photometer Operation Manual-OM2000. Gardiner, New York. 60 p. 61 Rinkel, M. o. 1971. Results of cooperative investigationsA pilot study of the eastern Gulf of Mexico. Caribbean Fisheries Institute, 23rd Annual Proceedings. November, 1970. Scrutton, P. C., and Moore, D. G. 1953. Distribution of surface turbidity off Mississippi Delta. Bulletin of the American Association of Petroleum Geologists. 37(5): 1067-1074. -Sheldon, R. 1rf., Evelyn, T. P. T., and Parsons, T. R. 1967. On the occurrence and formation of small particles in seawater. Oceanography. 12(3): 367-375. Stakhiv, E. 1968. The dependence of the circulation in the Gulf of Mexico upon the horizontal distribution of surface temperatures. M. S. thesis. Department of Oceanography. Florida State University. Tolbert, W. H. and Salsman, G. G. 1964. Surface circulation of the eastern Gulf of Mexico as determined by drift bottle studies. Journal of Geophysical Research. 69(2): 223-230. Zernova, V. V. 1969. The horizontal distribution of phytoplankton in the Gulf of Mexico. Oceanology. 9(4): 565-575.

PAGE 69

APPENDIX A DATA REDUCTION Reduction of Particle Data The description, calibration, and operation of the Model B Coulter Counter is thoroughly developed by Coulter Electronics (1968) and will not be discussed here. In the calibration of the Coulter Counter used in this thesis, latex spheresof mean diameters 3.49 microns and 9.5 0.9 microns were used. Calibration was performed on the 50 micron and 100 micron apertures used obtaining calibration constants of 2.86 cubic microns per threshold unit and 19.64 cubic microns per threshold unit respectively. Table A.l lists the volumes and equivalent spherical diameters corresponding to the aperture current, amplification and threshold settings used. 62 At least two replicate measurements were made at each threshold to ensure data repeatability. Only particles larger than that of the threshold volume were counted. Thus by varying the threshold value a cumulative particle distribution was obtained. To analyze the particle size distribution of the samples, the cumulative number of particles was plotted against the equivalent spherical diameters. Brun-Cottan and Ivanoff

PAGE 70

Table A.l Volumes and equivalent spherical diameters corresponding to current, amplification, and threshold settings. 63 Aperture 1 1 Threshold Volume Equivalent CfJ)) Size Ampli-Aperture Diameter (fJ..) fication Current (fL) 50 1/8 1/8 30 1.34 1.37 40 1.79 1.51 60 2.68 1.72 80 3.58 1.90 100 4.47 2.04 100 1/8 1/8 20 6.14 2.27 30 9.21 2.60 50 15.34 3.08 70 21.48 3.45 100 30.69 3.88 100 1/4 1/4 40 49.10 4.54 60 73.65 5.20 100 122.75 6.17 100 1/2 1/2 40 196.40 7.21 60 294.60 8 .26 80 392.80 9.09 100 491.00 9.79 100 1/2 1 60 589.20 10.40 80 785.60 11.45 100 982.00 12.33 100 1 1 60 1178.40 13.10 80 1571.20 14.42 100 1964.00 15.54

PAGE 71

(1970) and Brun-Cottan (1971) found that the particles between 1 and 20 microns in diameter in the western Mediterranean were fit by the Junge cumulative distribution: A.5 N = ax-m 64 where N is the number of particles larger than a certain diameter, x, and a and m are constants. Bader (1970) showed this relationship to hold for many natural collections of small particles including.mineral and organic particles suspended in sea water, but he allowed x to represent any size parameter (volume, diameter, surface area, etc.). This distribution of particle size when plotted on a log-log scale results in a straight line with a slope of m and a y-intercept of the log10 a (see Figure A.3). log N log x Figure A.l. Typical log-log graph of size parameter, x, versus cumulative particle number, N. The y-intercept represents the logarithm of the cumulative number of particles larger than a size parameter of 1. This

PAGE 72

65 relation was found to hold for the distributions of particle sizes in preserved samples taken over the varied ocean areas in this study. By making the assumption that this distribution would hold for all the particles throughout the study, particle counts of dubious reliability could be detected and eliminated. These counts were near the limits of Coulter Counter resolution with both the 50 and 100 micron apertures and electrical interference produced erroneous counts as a result of the high sensitivity required to count the smaller particles. It was found that the counts at the less sensitive settings did not deviate significantly from this relation. Unreliable points of the plot were discarded and an estimate of the correct count was made. The estimate of the correct count was calculated by determining the slope and y-intercept of the best fit straight line through the points not considered to be in error. The equation of this line would be: A.6 log N = log a-m log x where N is the cumulative number of particles larger than the diameter x and a and m are the number of particles large r than a diameter of 1 and slope of the line respectively, as in the distribution of particle sizes of BrunCottan (1971). The logarithm of particle diameter for the counts in error was used in equation A.6 and the logarithm of the estimate of the correct number was calculated. An

PAGE 73

estimate of the correct number was obtained from its logarithm and substituted for the discarded count at that volume. A correlation coefficient of the points making up the best fit line was calculated as an indication of the closeness of fit of the points used. 66 The total volume of the particulate load was calculated by the summation of the products of the number of particles in a particular size increment and average volume in that increment. The number of particles in each increment was determined by the difference between successive threshold counts. The average volume of the size range was the un-weighted mean between the two volumes defining the size range. The calculation can be expressed by the equation n A.7 TV = L (Ni-l -Ni) {vi + vi-1> i=l 2 where TV is the total volume, Ni is the number of particles larger than volume, Vi, and Ni-l is the number of particles larger than the volume, Vi_ 1 The unweighted mean was used because the Coulter Counter measures the volume of particles without regard to the shape. The average cross-sectional area of particles in the sample was calculated for comparison with the specific volume scattering function (discussed below). This was done by calculating the mean particle cross section in an increment by the formula

PAGE 74

67 A 8 where Si is the average cross-sectional area of a particle in the increment and Di and Dil are the diameters corresponding to the volume defining the size increment, assuming the particles to be spherical. The mean cross section was then calculated by A 9 s t (Ni-l -Ni) (si) = i=l n (Nil -Ni) i=l Since the Coulter Counter has a limit of particle size below which it cannot count, a number of particles smaller than the limit are not included in the particle distribution counting al., 1970; al. 1971). An estimate of the number of particles and total volume of the suspended particles not counted, TV (<50), would be use-ful in light-scattering comparisons. These could be calcu-lated in the same manner as the extrapolated points for the unreliable data in the particle size distribution of the particle sample if the assumption were made that the hyperbolic distribution of particle sizes can be extended to sizes less than 1 micron. Since the logarithm of the particle diameter approaches negative infinity as the particle

PAGE 75

68 diameter goes to zero it would be impossible to estimate the number of all particles less than the threshold of the Coulter Counter with this method. A lower limit of 0.3 microns was used in these calculations. This limit was chosen because Pak et (1971) found theoretically that the middle 90% of the volume scattering function, for e= 45 was generated by suspended clay particles between 0.3 and 8.5 microns in diameter with a relative index of refraction of 1.15. Reduction of Light Scattering Data The description, calibration, and operation of the Brice Phoenix Light Scattering Photometer is thoroughly developed by Phoenix Precision Instrument Company (1963), Pak (1970), and Carder (1970) and will not be discussed here. The specific volume scattering function for an angle of 45 degrees, J5 (45), is the mean scatter per particle. It is found by dividing tJ(45) by the total number of particles, N(50), in the scattering sample. The scattering volume is not the same as the volume of the sample counted, which in this study is 0.5 milliliters. The number of particles counted differs from the number of scattering particles. However, they would be proportional if the scattering volumes and particle sample volumes are constant. Consequently, if varying volumes of particle samples are measured, their specific volume scattering function can be compared by

PAGE 76

69 scaling the number of particles to reflect a constant volume of sample measured. This would be an estimate of the light scattered by a single particle in the sample. The quantity is plotted against the area of the average particle cross section, s. The point defined by these values is the optical vector as developed by Pak et al. (1970). A numerical representation of this optical vector can be found by dividing the specific volume scattering function, /5 (45), by the average particle cross section, s. This value is termed the specific scattering-average cross section ratio, j] (45)/S, or henceforth, the specific scattering-to-area ratio. The value of this ratio could be used as an indicator of relative amounts of organic and inorganic materials as a result of the difference in their indices of refraction (Burt, 1956; Pak et al., 1970; 1971). This ratio should vary directly with the inorganic or refractory fraction of the particles. As the volume fraction of the inorganic materials increases the amount of light scattered per particle would increase in relation to other samples with the same average particle cross-sectional area (Pak al., 1970; Carder et al., 1971). Addition of Lugol1s solution should enhance the dif-ference in scattering because the organic material would more readily absorb the Lugol's solution and become stained thereby causing an increase in the absorption of light by

PAGE 77

the organic particles, decreasing the amount of light available to be scattered. 70 Less change would be expected in scattering by the inorganic fraction as it would be less likely to absorb the Lugol's solution. This effect could be useful in the characterization of the particles in suspension as to the relative fractions of organic material, living and detrital, and the inorganic material.

PAGE 78

APPENDIX B EFFECTS OF LUGOL'S SOLUTION Introduction The necessity for preservation and storage of the particle samples made a study of the effects of Lugol's solution on the particles imperative. Lugol's solution is a preservative and staining agent used in work with phytoplankton. It is made by dissolving one gram of iodine and two grams of potassium iodide in fifteen milliliters of low-particle deionized water, then diluting this to a volume of 100 milliliters. The Lugol's solution is then filtered through a 0.45 micron Millipore filter. 71 The method used to determine the effects of the Lugol's solution was to take samples from two widely separated areas and measure their particle size distribution and total volume over a period of time. One sample was taken in the Bayboro Harbor area of Tampa Bay. This sample would be representative of the more productive areas sampled. The other was taken in the Gulf of Mexico in the Loop Current aboard the USNS MIZAR and represents the less productive regions encountered. Each sample was divided into subsamples. The particle size

PAGE 79

distribution was determined immediately using the Coulter Counter as previously described. The other subsamples were put in particle-free sample bottles which contained 10 milliliters of particle-free Lugol's solution for storage. Test samples were taken using Niskin water bottles. Division into the subsamples was performed by immediately drawing samples from the Niskin bottle directly into either an acid-cleaned beaker for the raw sample count, or the particle-free bottles for the samples to be stored. The concentration of the particles found in the sample taken in Bayboro Harbor was found to be too high to be counted on the Coulter Counter because of the problem of coincidence. This occurs when two or more particles pass through the aperture simultaneously resulting in their being counted as one large particle. For this reason, all subsamples of Bayboro Harbor water were diluted on a 1:10 basis with filtered sea water before their particle size distributions were measured. The preserved subsamples were counted at intervals of one, two, and three weeks after they were taken. The sample taken in the Loop Current was subdivided as previously described and the unpreserved sample was counted using a Model T .Coulter Counter. The preserved subsamples were counted at one, two, three, and four weeks using the Model B Coulter Counter. The particle size distribution and total volume of the particles were calculated in the same manner as for the 72

PAGE 80

73 EGMEX samples. These were compared with each other graphically and statistically. The graphical comparison was a log-log plot of the size distribution to note changes in the size distribution. The counts of the Lugol1s samples at each threshold were averaged and an envelope of one standard deviation about these lines was developed for comparison with the untreated subsample. The total volumes of the sub-samples were graphed versus time. Results Table B.l lists the total volumes calculated from the subsamples. TABLE B.l Total particle volumes of preserved and unpreserved samples. Time Total Volume Total Volume Interval Bayboro % Change MIZAR % Change (Weeks) Harbor (xlo}J?) 0 26.2 3.55 1 20.4 22.1 2.67 24.7 2 17.8 32.5 3.27 8.1 3 19.7 25.0 2.30 35.2 4 2.31 34.8 Figure B.l is a graph of these values with respect to time. It can be that there is a significant decrease in the total volume between weeks 0 to 1. At week 2 there is a

PAGE 81

3 2 1 TV (x1ofl3) 3 2 1 0 0 Bayboro Harbor sample 1 2 3 MIZAR sample 1 2 3 Time interval (weeks) Figure B l Graph of total particle volumes versus time. 74

PAGE 82

slight decrease in the Bayboro Harbor sample and an unaccountable increase in the MIZAR sample. By the third week the Bayboro Harbor sample has stabilized and the MIZAR sample has decreased. At the fourth week, the MIZAR sample has stabilized. Table B.2 lists the correlation coefficient, R, for closeness of fit to a Junge distribution, (Bader, 1970; Brun-Cottan, 1971), the slope, m, of the size distribution plot, and the logarithm of the number of particles larger than 1.0 micron in diameter, log a, of the Bayboro Harbor and MIZAR samples. Figure B.2 and B.3 are the log-log plots of cumulative particle number versus the particle diameters for the Bayboro Harbor sample respectively. Figure B.2 is the plot of the actual counts measured. Figure B.3 is a plot of the envelope of one standard deviation around the mean of the preserved particle numbers. The particle distributions of the preserved samples appear to fall into a small range as shown by the envelope calculated. 75 Figures B.4 and B.5 are the plots of size distribution and envelope respectively found in the MIZAR samples. The size distribution of the preserved samples are generally close, although the distribution of third week sample drops abnormally in 1.9 and 2.2 microns range and the second week sample distribution shows a slight hump between 3.9 and 8.3 microns in diameter. The envelope calculated from the preserved samples is found to be wider than that of the Bayboro

PAGE 83

76 TABLE B.2 R, m, and log a of Junge distribution for particles in Bayboro Harbor and MIZAR samples. Sample Time R m log a (Week) Bayboro 0 -.9802 -2.59 4.5865 Harbor 1 -.9715 -2.80 4.5722 2 -.9569 -2.89 4.6204 3 -.9813 -2.87 4.5511 Average for preserved samples -2.8571.0509 4.5812.0355 MIZAR 0 -.9999 -2.83 3.908 1 -.9916 -3.23 3.8526 2 -.9904 -2.75 3.6929 3 -.9784 -3.11 3.6651 4 -.9983 -3.01 3.7462 Averag e for preserved samples -3.0244.2082 3.7392!0.0828

PAGE 84

77 g 0 0 week (unpreserved) 10000 0 +1 week (preserved) g 0 +2 week (preserved) 08 8 +3 week (preserved) gB0 8ffi0 '() [:] 1000 8 G 0 N g (iml-1 ) El 0 8 G B 100 0 &0 Be0 90 0 8o8 10 5 Q 1 5 2 2 5 3 4 5 6 7 8 10 12 14 Particle diameter (f.L) Figure B 2 Log-log plot of particle size distribution of Bayboro Harbor samples.

PAGE 85

10000 1000 100 10 5 0 Standard (unpreserved) N o-I [J GJ G G I 1.5 2 2.5 3 4 5 6 7 8 (fl.) 10 12 14 Particle diameter Figure B.3. Log-log plot particle size distribution envelope Bayboro Harbor sample. 78

PAGE 86

1000 100 10 5 & 0 0 08 0 0 G 0 tJ 0 8 0 0 week 8 +1 week 0 +2 week 0 +3 week 'i& +4 week 8G f!Jfl YJ 8 G 0 G 8 (unpreserved) (preserved) (preserved) (preserved) {preserved) % (J 79 1.5 2 2.5 3 4 5 6 7 8 1 0 12 14 Particle diameter (fL) Figure B.4. Log-log plot of particle size distribution of MIZAR sample.

PAGE 87

1000 100 10 5 G Standard (unpreserved) -N '" 80 1.5 2 2.5 3 4 5 6 7 8 10 12 14 Particle diameter (f.L) Figure B.S. Log-log plot of particle size distribution envelope for MIZAR sample.

PAGE 88

Harbor sample as a result of these two deviations from the Junge Distribution observed. Conclusion The particle volume shows an initial decrease but appears to stabilize by the fourth week. There is a slight change in the particle size distribution which also stabilizes by the fourth week. Therefore, the preserved distributions would differ from the original distributions. 81 Sheldon al. (1967) reported the formation of small particles in seawater and hypothesized that they were caused by the presence of bacteria. Although their presence would be difficult to discount, they were not evi-dent from the samples run in the comparison of the preserved samples nor were they reflected in the plots of the EGMEX particle samples measured. It is uncertain whether or not they occurred. If they did, they were of such small numbers that they were masked by the other particles counted or, if abundant, they were for the most part attached to the surfaces of other particles. The decrease in particle volume measured and the in-crease in the slope of the particle size distribution is indicative of the disintegration of the particles in the samples. This disintegration resulted in parts of the particles which comprised a portion of the original total volume to be of sizes below the resolution limits of the Coulter Counter.

PAGE 89

82 Although the measure of the original concentration of the particles is lost in the storage of particle samples, the percentage loss in total volume and the shift in the slope of the size distribution and cumulative particle number is less than the range of the particle volumes and numbers expected in the Gulf of Mexico. Therefore, contouring the volumes and numbers would be adequate for showing trends in actual particle concentrations occurring in the Gulf of Mexico.


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.