USF Libraries
USF Digital Collections

The influence of local rivers on the eastern cariaco basin, Venezuela

MISSING IMAGE

Material Information

Title:
The influence of local rivers on the eastern cariaco basin, Venezuela
Physical Description:
Book
Language:
English
Creator:
Lorenzoni, Laura
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Bottom nepheloid layer
Colored dissolved organic matter
Subtropical underwater
Remote sensing reflectance
Anoxic basin
Dissertations, Academic -- Marine Science -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Two oceanographic cruises were conducted during September 2003 and March 2004 in the eastern half of the Cariaco Basin. Specific objectives were to examine the hydrography of the seasonal upwelling plume characteristic of this region, the spatial distribution of particles in the area, and to help determine the source and relative importance of in situ particle production vs. terrigenous particles delivered laterally from the coast.During September 2003, average surface salinities within the basin were higher (36.6) relative to Caribbean Sea waters outside the basin (35.6). Salinity patterns indicated that the Orinoco and Amazon River plumes did not enter or influenced the basin directly.The upwelling plume in March 2004 stimulated primary productivity. Beam attenuation and CDOM fluorescence profiles showed marked vertical structure in biomass of microbial populations, particularly near the oxic-anoxic interface typically located between about 250 and 300 m.There is an increasing difference in temperature and salinity between the Cariaco Basin and the adjacent Caribbean Sea below 200 m. Inside the Basin temperatures and salinities were higher by 4ʻC and 0.5.The influence of local rivers on the Cariaco Basin was evident during September 2003. Low salinity plumes with high beam attenuation (1m1) lined the southern margin of the Basin. The primary rivers that affected the basin were the Unare and Never.̕ Their sediment input affected the shelf near the river mouths, and a surrounding radius of up to 40 Km. Their low salinity plumes were carried northwestward toward the CARIACO time series station. In March 2004, there was minimal or no terrigenous input from local rivers. Near the Manzanares River, off the city of Cuman,̀ and near Cubagua Island, located south of Margarita Island, attenuation due to suspended particles (0.09 m-1) was observed at depth (70-150 m) during both cruises (0.09-0.15 m-1).
Thesis:
Thesis (M.S.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Laura Lorenzoni.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 89 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001680969
oclc - 62469296
usfldc doi - E14-SFE0001152
usfldc handle - e14.1152
System ID:
SFS0025473:00001


This item is only available as the following downloads:


Full Text

PAGE 1

The Influence Of Local Rivers On Th e Eastern Cariaco Basin, Venezuela by Laura Lorenzoni A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: Frank E. Muller-Karger, Ph.D. Kendall L. Carder, Ph.D. Gabriel A. Vargo, Ph.D. Date of Approval: April 1, 2005 Keywords: bottom nepheloid layer, colore d dissolved organic matter, subtropical underwater, remote sensing reflectance, anoxic basin Copyright 2005 Laura Lorenzoni

PAGE 2

Acknowledgements If I wanted to thank all the people that have made this possible by enriching my knowledge, lending a hand when in need or just sharing special moments during my stay at the College of Marine Science, I’d have to write another manuscript, with twice the number of pages (including figur es). Unfortunately, that is no t an option, and I’ll have to settle for this short page. I want to thank first an d foremost my husband; who has stood by me, in the good and the bad times. To my family and Gea, I would not be here if it was not for their love, constant support and faith in me. Dr. Frank Mller-Karger, my advisor, gave me the opportunity to come to the College of Marine Science, and for that I’ll be forever grat eful. He believed in me and gave me the means to achieve what I aspired. One individual that de serves a very special recognition is Jim Ivey, who always had tim e for my questions, putting aside his own work to help me get through mine. I also wish to acknowledge Jennifer Cannizzaro, Robyn Comry, Dr. Chuanmin Hu, Dr. Robert Weisberg (and the Ocean Circulation Group), Dr. Peter Howd, Dr. Carlos del Castil lo, Dr. Emmanuel Boss, Dr. Collin Roesler and Dr. Dennis Hansell for their help both in the data processing and analysis. I want to thank my committee members, Dr. Ken Carder and Dr. Gabriel Vargo, for their wisdom and support. To our friends and colleagues at ED IMAR, especially Yrene Astor, Ramn Varela, Glenda Arias, Aitzol Arrellano a nd Chuchu, without them and the crew of the Hermano Gins this work would have never be en possible. Thank you for bearing with us during those long sampling days and making this project work. To Luchi, Damaris. Marina, Lucia, Michelle, Ester and the entirety of the IMaRS lab, thank you for those helping hands. To my friends, near and far, you mean the world to me. For those that I failed to mention, I know who you are, and I thank you all so much. Financial support for this work was pr ovided by The Nationa l Science Foundation OCE0118566 and OCE 0326268.

PAGE 3

i Table of contents List of figures................................................................................................................ ......ii List of tables................................................................................................................. ......iii Abstract....................................................................................................................... .......vi Foreword....................................................................................................................... ....vii List of symbols................................................................................................................ .viii CHAPTER I 1. Introduction................................................................................................................ .....1 1.1. Objectives................................................................................................................ 3 2. Methods..................................................................................................................... ......4 2.1 Data processing.........................................................................................................7 3. Results..................................................................................................................... ........9 3.1. Hydrography............................................................................................................9 3.1.1 Surface Hydrography.........................................................................................9 3.1.2. Water mass distribution..................................................................................14 3.2. Distribution of DOC and CDOM around the Cariaco Basin.................................19 3.3. Vertical distribution of par ticulate and dissolved matter.......................................25 4. Discussion.................................................................................................................. ...32 5. Conclusion.................................................................................................................. ..35 CHAPTER II 1. Introduction................................................................................................................ ...37 1.1 Objectives...............................................................................................................38 2. Methods..................................................................................................................... ....39 2.1. Data processing......................................................................................................40 3. Results..................................................................................................................... ......45 3.1. Surface distribution of CD OM near the coast.......................................................45 3.2. River influence on the optical propert ies of Cariaco's coastal waters...................49 3.3. Particles and sediment transport from the coast....................................................53 3.4. Remote sensing reflectance of Cariaco waters......................................................59 4. Discussion.................................................................................................................. ...64 5. Conclusions................................................................................................................. ..70 General conclusions..........................................................................................................72 References..................................................................................................................... ....73

PAGE 4

ii List of tables Chapter I Table 1.1: Station location and cas t depth for COHRO and COHRO2..............................5 Table 1.2: ag, DOC and S values for selected stations........................................................8 Table 1.3: Sources of histori cal temperature and salinity measurements for waters below 1200 m in the Eastern Cariaco Basin. The CA RIACO time series measurements are yearly averages..........................................................................................................16 Table 1.4: Temperature, salinity, CDOM fluor escence and nutrient data for selected stations sampled during March 2004........................................................................21 Chapter II Table 2.1: Station location and cas t depth for COHRO and COHRO2............................42 Table 2.2: CDOM absorption slopes, S............................................................................49 Table 2.3: Error obtained using the Lee et al (1999) and Carder et al. (1999) algorithms to estimate aph(440) at the Cariaco time-series station..............................................62 Table 2.4: Error obtained using the Lee et al (1999) and Carder et al. (2004) algorithms to estimate chlorophyll concentration and ag(440) in waters of the Cariaco Basin during September 2003 and March 2004..................................................................63

PAGE 5

iii List of figures Chapter I Figure 1.1: Schematic of the Cariaco Basi n, showing location of the CARIACO time series station (from As tor et al., 2003)........................................................................2 Figure 1.2: Station map location for Sept ember 2003 (A) and March 2004 (B). Major local rivers are shown......................................................................................................... 6 Figure 1.3: Surface temperature (A, C) and sali nity (B, D) of the Cariaco Basin during September 2003 and March 2004, respectively. No te the different scales used for each cruise.................................................................................................................11 Figure 1.4: Vertical distributi on of the east/west component of current velocity (cm/s), measured at the CARIACO time-series st ation (see figure 1.1 for location) for September 2003. Positive (yellow/red) is ea stward flow. Black lines indicate the cruise date.................................................................................................................12 Figure 1.5: SeaWiFS images of CDOM absorption at 440 nm (ag(440) and chlorophyll concentration. (A) 30 September 2003; (B) 10 March 2004....................................13 Figure 1.6: T/S diagrams for station 9, September 2003 (x) and station 26, March 2004 ( ). Maximum sampling depth: 1300 m and 1382 m, respectively Error! Bookmark not defined. 4 Figure 1.7: Salinity transect s for September 2003 (A) and March 2004 (B). Red lines correspond to casts performed at stations.................................................................15 Figure 1.8: T/S diagrams for September 2003 (A) and March 2004 (B). Observed ( ) and predicted (-) temperature (C) and salinity (D) increase for waters below 1200 m inside the Cariaco Basin (see table 1.3)....................................................................17 Figure 1.9: Oxygen profiles for selected stations during September 2003.......................18 Figure 1.10: Oxygen profiles for select ed stations during March 2004...........................18 Figure 1.11: Property-propert y plot of (A) DOC ( MC) and Temperature (oC) (B) ag(400) and DOC ( M C) (Stations used are shown in Table 1.2). Both figures refer to March 2004 samples.................................................................................................19 Figure 1.12: Correlation plot of CD OM absorption coefficient, ag (m-1) at 370 nm, and corresponding fluorescence emission (expresse d in arbitrary units, AU) for stations 6, 29 and 35, sampled during March 2004. Circles ( ) correspond to sampled stations at depths of 35m, 55m and 100m. Other symbols correspond to surface measurements: Station 35, 1m. Station 6, 1m. Station 29, 1m......................20 Figure 1.13: Correlation plot of FCDOM (Arbitrary Units, AU) and FChl (Arbitrary Units, AU) for surface (1-25 m) waters sampled during March 2004. Stations: 29 ( ) and 35 ( )..................................................................................................................22 Figure 1.14: CDOM fluorescence (FCDOM) and te mperature for stati ons inside (14, 19, 20, 29, 38) and outside (26, 35) the Cariaco basin in March 2004...........................23

PAGE 6

iv Figure 1.15: Vertical profiles of oxygen and FCDOM at stations 19 (A) and 14 (B) in March 2004...............................................................................................................24 Figure 1.16: (A) Profile of tota l absorption at 412 nm (a412) and attenuation at 650 nm (c650) (B) Profiles of total ab sorption at 412 nm (blue), sa linity (black) and density (red). Both profiles were measured at station 9 in September 2003.........................26 Figure 1.17: Profiles of total absorption at 412 nm (a412) for different stations inside the Cariaco Basin, sampled during September 2003......................................................28 Figure 1.18: Total beam attenuation coe fficient at 650 nm (c650, black) and beam attenuation spectral slope ( blue) measured at sta tion 9 in September 2003........29 Figure 1.19: Profiles of total absorption at 412 nm (a412) measured with the AC-9 (solid line) in September 2003, and CDOM abso rption derived from CDOM fluorescence at 412 nm (ag(412), dotted line) in March 2004 for stations sampled at the same location......................................................................................................................3 0 Chapter II Figure 2.1: Schematic of the Cariaco Basi n, showing location of the CARIACO time series station (from As tor et al., 2003)......................................................................38 Figure 2.2: Sample location for the Sept ember 2003 (A) and March 2004 (B) cruises. Major local rivers are shown.....................................................................................43 Figure 2.3: Visible changes in water color in September 2003. Station 48 is located near the Manzanares River, station 14 close to the Never River, and stations 49 and 50 near the Unare River.................................................................................................46 Figure 2.4: Relationship between CDOM absorption coefficient (ag(400)) and salinity for stations 50, 13, 14, 22, 21 and 20 sampled in September 2003................................47 Figure 2.5: (A) Vertical profile (from 1-400m) of salinity at the CARIACO time-series station from January 1996 to July 2004. Contour lines are interpolated between measurements. (B) Vertical distribution of the east/west component of current velocity (cm/s), at the CARIACO time-seri es station (see Figure 2.1 for location) for September 2003. ......................................................................................................48 Figure 2.6: Relationship between the spect ral slope, S, and CDOM absorption, ag at 400 nm. Samples used include those collected during September 2003 ( ) and March 2004 ( )....................................................................................................................49 Figure 2.7 Total absorption measured during September 2003 at (A) 1m (B) 5m (C) 10m (D) 25m.....................................................................................................................51 Figure 2.8: Optical depth at selected stations sampled during September 2003. Station 50 ( ) Station 51 ( ) Station 24 ( ) Station 14 ( )......................................................52 Figure 2.9: SeaWiFS images of CDOM absorption at 443 nm (ag(443)), chlorophyll concentration (chl a ) and aph (443), processed with Lee et al. (1999), Carder et al. (1999) and Carder et al. ( 2004) algorithms. Images are from September 30, 2003 and March 10, 2004.........................................................................................................54 Figure 2.10: Beam attenuation (660nm) transect from station 13 to station 21 sampled during September 2003. Contour lines are inte rpolated between stations. Black lines indicate profile locat ions and depths.........................................................................55

PAGE 7

v Figure 2.11: Horizontal distribut ion of the beam attenuati on (660 nm) near the eastern coast of the Cariaco Basin during Septem ber 2003 at the surface (A) and at the maximum depth sampled at each station (B)............................................................56 Figure 2.12: Beam attenuation (660nm) profiles near the Never River during September 2003. For station location re fer to Figure 2.2A........................................................57 Figure 2.13: (A) Profile of tota l absorption at 412 nm (a412 – dotted line) and attenuation at 650 nm (c650 solid line) at station 2 (10o29’ N 64o21’ W) during September 2003 (B) Beam attenuation profiles at 660 nm (c660) at 10o29’ N 64o21’ W for September 2003 (dotted line) and March 2004 (s olid line). Red circles indicate INL (Internal Nepheloid Layer) location (C) To tal beam attenuation coefficient at 650 nm (c650, black) and beam attenuation spectral slope ( blue) at station 2 (10o29’ N 64o21’ W ) during September 2003. Arrow indi cates the location of the maximum phytoplankton biomass; red circle indicates location of the INL.............................58 Figure 2.14: Remote sensing reflectance spectr a measured in situ in September 2003...60 Figure 2.15: SeaWiFS images of chlorophyll a for the eastern coast of Venezuela (A) September 25, 2000 (B) September 26, 2000 (C) October 25, 2001 (D) October 24, 2003 The approximate location of the Tuy River is indicated by the red arrow....66 Figure 2.16: Bathymetry of the Cariaco Basin. The red segmented line marks approximate sea level during the Last Glacial Maximum (Fairbanks, 1989)...........67 Figure 2.17: Relationship between chlorophyll measured in situ during March 2004 at 26 stations, and MODIS chlorophyll fluorescence for corresponding locations (image from March 16, 2004)................................................................................................................ 69

PAGE 8

vi The Influence of Local Rivers on th e Eastern Cariaco Basin, Venezuela Laura Lorenzoni ABSTRACT Two oceanographic cruises were conduc ted during September 2003 and March 2004 in the eastern half of the Cariaco Basi n. Specific objectives were to examine the hydrography of the seasonal upwelling plume ch aracteristic of this region, the spatial distribution of particles in the area, and to help determine the source and relative importance of in situ particle production vs. terrigenous particles delivered laterally from the coast. During September 2003, average surface sali nities within the basin were higher (~36.6) relative to Caribbean Sea waters outside the basin (~35.6). Salinity patterns indicated that the Orinoco a nd Amazon River plumes did not enter or influenced the basin directly. The upwelling plume in March 2004 stimulated primary productivity. Beam attenuation and CDOM fluorescence profiles showed marked vertical structure in biomass of microbial populati ons, particularly near the oxic-anoxic interface typically located between about 250 and 300 m. There is an increasing difference in temperature and salinity between the Cariaco Basin and the adjacent Caribbean Sea below 200 m. Inside the Basin temperatures and salinities were higher by 4oC and 0.5. The influence of local rivers on the Ca riaco Basin was evident during September 2003. Low salinity plumes with hi gh beam attenuation (~ 1m–1) lined the southern margin of the Basin. The primary rive rs that affected the basin we re the Unare and Never. Their sediment input affected the shelf near the ri ver mouths, and a surrounding radius of up to 40 Km. Their low salinity plumes were carried northwestward toward the CARIACO time series station. In March 2004, there was minimal or no terrigenous input from local rivers. Near the Manzanares River, off th e city of Cuman, and near Cubagua Island, located south of Margarita Island, attenuation due to suspended particles (~0.09 m-1) was observed at depth (70-150 m) during both cruises (~0.09-0.15 m-1). Therefore, sediment transport from the shelf into the basin s eems to occur year-round. More observations are necessary to determine the natu re and origin of these part icles. In March 2004, there was minimal or no terrigenous input

PAGE 9

vii Foreword The study is based on data collected during two cruises, carried out in September 2003 and March 2004, in the fr amework of the CARIACO (CA rbon R etention I n A C olored O cean) time series project. Specifica lly, the objective of the study was to identify sources of particulate material th at may contribute to sediments accumulating in the Cariaco Basin (10o30’ N 64o40’ W). This knowledge helps explain the paleoclimatology record stor ed in these sediments. The thesis is divided into two chap ters. The first focuses on oceanographic interactions between the Cariaco Basin and the open Caribbean Sea, and the second on the impact of local rivers and their sediment load on the basin. Each chapter contains an introduction that explains the rationale of the work conducted, as well as methods, results and discussion, and conclusions. An overall conc lusion chapter is incl uded at the end of the thesis.

PAGE 10

viii List of symbols Symbol/Acronym Definition Units a( ) Absorption coefficient at wavelength m-1 aCDM Absorption coefficient of colored dissolved and detrital material m-1 ad Absorption coefficient of detritus m-1 ag Absorption coefficient of CDOM m-1 aT Total absorption m-1 aw Absorption coefficient of water m-1 BNL Bottom nepheloid layer c( ) Attenuation coefficient at wavelength l m-1 CDM Colored dissolved and detrital material CDOM Colored Dissolved Organic Matter COHRO Campaa "ptica e Hidrogrfica Regin Oriental (September 2003) COHRO2 Campaa "ptica e Hidrogrfica Regin Oriental 2 (March 2004) CSW Caribbean Surface Waters DOC Dissolved organic carbon Ed Downwelling irradiance W m-2 nm-1 FCDOM CDOM fluorescence AU FChl Chlorophyll fluorescence AU INL Intermediate nepheloid layer Kd Diffuse attenuation coefficient m-1 OD(zeu) Optical Depth at which 1% downward irradiance is reached m PSD Particle size distribution Rrs Remote sensing reflectance S Spectral slope of the exponential fit to ag nm-1 SeaWiFS Sea-viewing Wide Field-of-View Sensor SUW Subtropical Underwater TACW Tropical Atlantic Central Waters Hyperbolic slope of c Dimensionless Wavelength of light nm o Reference wavelength nm Hyperbolic slope of the PSD Dimensionless Optical depth Dimensionless

PAGE 11

CHAPTER I HYDROGRAPHY OF THE EASTER N CARIACO BASIN, VENEZUELA

PAGE 12

1 1. Introduction The Caribbean Sea does not contain a si ngle, uniform water mass. Rather, it contains several water masses that show spa tial and temporal changes. Understanding the distribution and variability of these water masses is relevant to the interpretation of the climate record contained in sediments of the Cariaco Basin in the s outhern margin of the Caribbean Sea. Here we examine this variability by studying hydrographic properties associated with different water masses and patterns in the color of marine waters including factors that change them, such as variation in colored di ssolved organic matter concentration. The upper portion of the water column in the Caribbean Sea, Caribbean Surface Water (CSW; Wust, 1964; Morrison and Nowli n, 1982), is characterized by low salinity (<35.5) and higher concentratio ns of silicates, as compar ed to underlying waters, attributed to the influence of the Amazon and Orinoco Rivers. Fr oelich et al. (1978) concluded that 60% of the freshwater in th e Caribbean is from the Amazon River. The Orinoco and Amazon rivers affect the sea surfac e salinity (SSS) in the tropical Atlantic at great distances from their poin t of discharge (Muller-Karger et al., 1989; Hu et al., in press). Both rivers reach th e Caribbean Sea (Muller-Karger & Varela, 1989; Hellinger & Gordon, 2002). The Orinoco plume grows in August through November (Muller-Karger et al., 1989; Hellweger and Gordon, 2002). T ogether, these rivers influence surface salinities (Morrison and Smith, 1990) and pi gment distribution of the Caribbean Sea (Muller-Karger et al., 1989; Bidigare et al. 1993). SeaWiFS and other surface spectral reflectance (SSR) satellite imagery are able to trace the circulation of Amazon and Orinoco rivers in the near-s urface Caribbean waters (Mul ler-Karger and Varela, 1989, Hu et al, in press). Their joint pl umes have been traced with sa tellite ocean color images at least as far north as Puerto Rico. Though the dispersa l of the Orinoco and Amazon is fairly well understood in the open Caribbean, there is a lack of information on what is the effect of these large river plumes on the southern margin of the Caribbean Sea. Below the CSW, between 75 and 150 m, is the Subtropical Underwater (SUW). The SUW is characterized by a salinity maximum (36.8; Morrison and Smith, 1990). Also of importance are the 18C Sargasso Sea Water (SSW; located in the central Caribbean between 200 and 400 m, characteri zed by an oxygen maximum; Kinard et al, 1974), the Tropical Atlantic Central Water (TACW; between 400-700 m and recognizable by an oxygen minimum), and Anta rtic Intermediate Water (AAIW, below 700 m with low salinity and high silicate cont ent; Morrison and Nowlin, 1982). There is very little known about the spa tial and temporal vari ability, or even the origin, of some of these water masses. Much less is known about their influence on pr operties around the margin of the Caribbean basin. The Cariaco Basin (Figure 1.1) is located off the coast of Eastern Venezuela. The Basin is formed by the strike slip fault zone s of El Morn and El Pilar (Schubert, 1982). It contains two depressions, th e Eastern deep and the Wester n deep, separated by a sill of

PAGE 13

2 less than 900 m. Each depression reaches de pths of ~1400 m. The basin is bound to the south by the Unare platform, to the north by the Tortuga Bank, and to the East and West by continental Venezuela. The Cariaco Basin is connected to the Ca ribbean Sea through a sill of about 100 m deep, which is cut by the Centinela Channel to the west and the Tortuga Channel to the north, both of a depth of about 140 m. Processes in the Caribbean Sea affect th e exchange of waters with the Cariaco Basin, its productivity and ver tical carbon flux (Muller-Karge r et al., 2001). The Cariaco Basin is anoxic below 250 m because of deco mposition of the flux of particulate organic matter settling from surface waters and the reduced circulation inside the basin. The origins of the basin date back to the Quat ernary, and since then it has seen several alternating episodes of anoxia during interglacials and oxicity mostly during glacial times (Peterson et al., 1991; Hughen et al, 1996; Lin et al., 1997; Haug et al., 1998; Peterson et al., 2000). The sediment record extracted to date under the Ocean Drilling Program extends back an estimated 600,000 years (Hughen et al, 1996; Yari ncik et al., 2000). Figure 1.1: Schematic of the Cariaco Basin, showing location of the CARIACO time series station (from Astor et al., 2003). Due to its present anoxic state and se dimentation rate (1 cm/1000 years), the Cariaco Basin conserves in its sediments a re cord of processes that cover the present interglacial period (Thunell, 1997; Black et al., 1999; Muller-Kar ger et al., 2000; Peterson et al., 2000). In order to correctly inte rpret the record of this and gain insight about past interglacials, it is necessary to ha ve a clear unders tanding of the processes that occur in the water column of the Cariaco Basin.

PAGE 14

3 The CARIACO project has over 9 years of data which have provided invaluable information on primary productivity and geoche mical cycles in the tropics, as well as a better understanding of processe s that occur in anoxic envi ronments (Muller-Karger et al., 2001; Astor et al., 2003; Kessler et al., 2004; Muller-Karger et al. 2004). However, the one-location sampling is not enough to ga in a full understanding of the processes that affect the basin, and the important role th at the Caribbean and large South American Rives may play on the productivity and possibl e sediment and nutrient transport of the region. 1.1. Objectives The objective of this chapte r is to understand the influence of the Caribbean Sea and large South American Rivers, like th e Orinoco, on the Cariaco Basin. More specifically, we want to an swer the following questions: • Which water masses enter the Cariaco Basi n from the adjacent Caribbean Sea? • Do the Orinoco and Amazon rive rs contribute dissolved and/ or detrital material to the Cariaco Basin? What is the role of small, local rivers? • What is the vertical distribution of co lored dissolved organic matter (CDOM) in the Cariaco basin? Is the Caribbean S ea a source of CDOM to the Basin?

PAGE 15

4 2. Methods Two cruises were conducted to th e Cariaco Basin on board of the R/V Hermano Gines (Fundacin La Salle de Ciencias Naturales de Venezuela). The first cruise was conducted in September 2003 (COHRO – C ampaa O ptica e H idrografica R egin O riental). The objective of this cruise was to study the impact of rivers, both local and distant, specifically the Orinoco and Am azon, on the Cariaco Basin. Stations offshore were separated by a distance of approximately 20 km, but were more closely spaced near the coast (Figure 1.2A). The second cruise was carried out in March 2004 (COHRO2). The objective of the second cruise was to study the distribution of upwelling waters within the Cariaco Basin, and to determine th e characteristics of the water entering the basin from the Caribbean Sea during this period. Stations were separated by approximately 20 km in a grid pattern. Eigh t stations were located outside the basin, north of the Tortugas-Margarita sill (Figure 1.2B). Table 1 provides a list of location and maximum depth sampled in both cruises. Salinity and temperature profiles were collected using a Seabird SBE25 CTD, deployed at each station in a rosette en semble. Dissolved oxygen and chlorophyll fluorescence data were obtaine d with a Seabird SBE 43 oxygen sensor and a Chelsea fluorometer (Chelsea, Inc.) attached to th e CTD. The ensemble also had a C-Star transmissometer (WetLabs) that measured b eam attenuation at 660 nm (c660). The data were processed using SeaBird’s SBE Data Processing software. CDOM (colored dissolved organic matte r) fluorescence (hereafter referred to as FCDOM) was measured during March 2004 only, using a WetStar fluorometer (WetLabs). The excitation and emission wa velengths were set to 370 nm and 460 nm, respectively. A linear relationship between CDOM absorption and fluorescence permitted derivation of an empirical estimate of CDOM absorption throughout th e region, similar to the approach used by Vodacek et al. (1997) and Blough and Del Vecchio (2002). Samples were filtered on board with a 0.2 m pore-size anotop filter, using a glass syringe. Filtrate samples were fro zen in acid cleaned am ber-colored bottles. Absorption was measured within one month as described below under Absorption Spectroscopy In September 2003, an absorption/atte nuation meter (AC-9, WetLabs) was deployed as part of the verti cal hydro-casts. A pump was used to draw water through the absorption and attenuation sensors. At 14 sta tions, double casts were performed, one with an unfiltered AC-9, and the second using a 0.2 m filter (Propor PES filter capsules) attached to the inlet of the absorption tube The instrument was ca librated before, during and after the cruise using distilled water as a reference. DOC (dissolved organic carbon) samples were collected onl y during March 2004, by gravity filtering though GF/F precombusted filters, directly from the Niskin bottles and using silicone tubing. Samples and tubi ng were handled using polyethylene gloves.

PAGE 16

5Table 1.1: Station location and cast depth for COHRO and COHRO2. COHRO, September 16-20, 2003 COHRO2, March 15-20, 2004 Latitude Longitude Station Maximum profiling depth (m) Latitude Longitude Station Maximum profiling depth (m) 10.83 -64.37 0 225 10.83 -64.36 1 217 10.66 -64.36 1 157 10.66 -64.36 2 160 10.50 -64.36 2 400 10.50 -64.36 3A 400 10.33 -64.55 3 120 10.50 -64.36 3B 1220 10.49 -64.55 4 400 10.45 -64.26 4 120 10.66 -64.55 5 400 10.41 -64.38 5 190 10.83 -64.55 6 220 10.33 -64.55 6 127 10.83 -64.70 7 225 10.40 -64.61 7 160 10.66 -64.71 8 400 10.45 -64.55 8A 521 10.49 -64.66 9 400 10.45 -64.55 8B 798 10.33 -64.71 10 40 10.56 -64.55 9A 400 10.33 -64.88 11 56 10.56 -64.55 9B 1220 10.23 -64.88 12 37 10.76 -64.55 10 400 10.14 -65.03 13 15 10.76 -64.71 11 383 10.19 -64.79 14 26 10.76 -64.88 12 400 10.28 -64.8 15 45 10.60 -64.88 13 403 10.49 -64.88 16 400 10.48 -64.88 14 400 10.66 -64.88 17 400 10.33 -64.71 15 50 10.83 -64.88 18 224 10.28 -64.80 16 50 10.66 -65.05 19 400 10.38 -64.93 17 76 10.49 -65.05 20 400 10.50 -65.05 18 400 10.41 -65.05 21 63 10.63 -65.10 19 400 10.33 -65.05 22 59 10.76 -65.05 20 400 10.23 -65.05 23 25 10.93 -65.05 21 67 10.23 -65.21 24 24 11.05 -65.05 22 75 10.33 -65.21 25 56 11.16 -65.05 23 276 10.41 -65.21 26 75 11.33 -65.05 24 400 10.33 -65.38 27 55 11.33 -64.88 26A 400 10.41 -65.38 28 77 11.33 -64.88 26B 1382 10.49 -65.21 29 113 11.16 -64.88 28 276 10.83 -65.04 30 220 11.05 -64.88 29 123 11.00 -65.04 31 80 10.93 -64.88 30 181 11.00 -64.88 32 130 10.93 -64.71 31 310 11.00 -64.71 33 174 11.05 -64.71 32 90 10.99 -64.54 34 273 11.16 -64.71 33 90 10.44 -64.26 48 70 11.33 -64.71 34 210 10.14 -65.11 49 15 11.33 -64.55 35 380 10.16 -65.21 50 10 11.16 -64.55 36 50 10.41 -64.40 51 180 11.05 -64.55 37 84 10.93 -64.55 38 350

PAGE 17

-65.4-65.3-65. 2 -65. 1 -65-64. 9 -64. 8 -64. 7 -64. 6 -64. 5 -64.4-64.3-64. 2 -64.1 1010.110.210.310.410.510.610.710.810.91111.111.211.3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 28 29 30 31 32 33 34 35 36 37 38 -65.4-65.3-65 2 -65. 1 -65-64. 9 -64. 8 -64. 7 -64. 6 -64. 5 -64.4-64.3-64. 2 -64.1 1010.110.210.310.410.510.610.710.810.91111.111.211.3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 48 49 50 51atitude Unare Never Manzanares L Gulf of Santa Fe Longitude Figure 1.2: Sample map location for September 2003 (A) and March 2004 (B). Major local rivers are shown. 6

PAGE 18

7 The samples were stored in Nalgen e wide-mouthed translucent 60 ml acidcleaned HDPE polyethylene bottles and frozen to -20 C until analysis. Between casts, both the filter holders and tubing were rinsed with 10% HCl and di stilled water. They were then sent to the Rosens tiel School of Marine and At mospheric Science (RSMAS) at the University of Miami (Dr. De nnis A. Hansell) for analysis. 2.1 Data processing AC-9 processing and validation AC-9 data were corrected for temperat ure and salinity following Pegau et al. (1997). A scattering correction was also app lied to the absorption data. The scattering correction was done by subtracting a refere nce wavelength (715 nm) from all data (hypothesis 3 in the AC-9 WetLabs User’s Gu ide, 2003). As a data quality check, the AC-9 beam attenuation data at 650 nm were compared to c660 taken with the transmissometer at all stations. Absorption Spectroscopy Water samples from the March 2004 cruise were transported back to the University of South Florida (USF), and analy zed within one month of the cruise. Samples were re-filtered once thawed to remove any remaining particles. The data were scanned between 200 and 800 nm at 1 nm intervals using a Perkin-Elmer Lambda 18 spectrophotometer equipped with 10 cm quar tz cells. The detection limit of the instrument was 0.002, or equiva lent absorption of 0.046 m-1. Milli-Q water was used as blank. For each sample, two scans were pe rformed. Running means were used to smooth the data prior to comparison, and, where n ecessary, the blank absorption values were subtracted from the sample. The criterion used for the subtraction of the blank was quantitative: at 650 nm the absorption had to be zero. If the curve was above the zero line, the blank was subtracted from the sample. Each sample was looked at individually, and the best of the two scans selected. The selecti on of the best scan was based on the smoothness and shape of the curve, whether it ha d suspicious features and whether it had been necessary to subtract the blank from the sample. Those scans where no subtractions had been done were preferred. Absorption coefficients were calculated from absorbance, according to the relationship: a( ) = 2.303 A( )/r where A is the absorbance or optical densit y and r is the pathlength. Light absorption by CDOM decreases exponentially from the UV to the visible part of the spectrum, and can be approximated by: ag( ) = ag( o)exp[–S( o)]

PAGE 19

8 where ag( ) and ag( ) are the absorption coefficients at wavelength and at a reference wavelength ; was set to 400 nm. S is the spectral slope, which describes the decrease in absorption of ag with increasing wavelength (Bl ough et al, 1993; Green and Blough, 1994; Ferrari, 2000, Blough and Del Vecchio, 2002) Each spectrum was plotted in two ways: through a linear least s quares regression of the log-tr ansformed data (ln(a) vs. J. Cannizzaro, pers. comm and in an exponential form, using a non-linear least squares fitting (Dr. C. Hu, pers. comm.). The method that provided the single best fit for each individual spectrum was selected. The sl ope S was calculated be tween 330 and 400 nm. Beyond 400 nm, the signal to noise ratio was t oo large. Absorption coefficients at 370 nm, 400 nm, and 412 nm (ag(370), ag(400) and ag(412); Table 1.2) were used for the different characterizations. 3 stations samp led in March 2004 are shown in Table 1.2. These stations were selected because they are located outside the Basin (station 35), at the northern sill (station 29) and inside the Basin, at the upwelling focus (station 6). Table 1.2: ag, DOC and S values for selected stations sampled during March 2004 Station Depth (m) ag(370) (m-1) ag(400) (m-1) ag(412) (m-1) S (330-400) (nm-1) DOC ( M of C) Temp. (C) Salinity 35 1 0.1349 0.0619 0.0492 0.0234 66.42 24.73 36.76 35 35 0.1096 0.0447 0.0342 0.0254 56.75 22.40 36.75 6 1 0.0823 0.0407 0.0334 0.0191 59.29 22.90 36.94 6 35 0.0626 0.0442 0.0393 0.0127 53.58 21.94 36.92 29 1 0.1352 0.0673 0.0532 0.0215 65.75 25.21 36.84 29 35 0.1116 0.0624 0.0517 0.0184 59.83 22.65 36.88 29 55 0.0758 0.0354 0.0278 0.0236 21.60 36.91 Satellite imagery Two SeaWiFS (Sea-viewing Wide Field-of -View Sensor) images were processed for chlorophyll using the OC-4 algorithm (O’Reilly et al., 1998). The dates of the images were chosen as close as possible to the dates of the cruises. The only image with low enough cloud cover for September 2003 was of September 30. For March 2004, the image of March 10 was selected. Sea Surf ace Temperature (SST) images were also processed. Sea Surface Temperature (SST) wa s derived from infrar ed (IR) observations collected by the Advanced Very High Reso lution Radiometer (AVHRR) sensors flown on the National Oceanic and Atmospheric Administration's (NOAA) Polar Orbiting Environmental Satellite (POES) series. SST was computed using the multi-channel seasurface temperature (MCSST) algorithm deve loped by McClain et al. (1985). Due to cloud cover, three-day composites were used instead of daily images. Running means were used to generate the composites of September 2003 (using September 29, 30 and October 1st), and March 2004 (using March 9, 10 and 11th).

PAGE 20

9 3. Results 3.1. Hydrography 3.1.1 Surface Hydrography A marked seasonal difference was found in the Cariaco Basin in both surface temperature and salinity distribution between the sampling carried out in the dry (March 2004) and the rainy seasons (September 2003) The patterns and seasonal differences were similar to those observed in previous studies (Herre ra and Febres-Ortega, 1975; Herrera et al., 1980; Astor et al. 1998; Scra nton et al, 2001, Mull er-Karger et al, 2001; Astor et al., 2003). Figure 1.3 shows the surface distribution of temperature and salinity during September 2003 and March 2004. During the rai ny season, surface temperatures ranged between ~26 to ~28 C (Figur es 1.3A and B). Salinities o ffshore varied from ~36.4 to ~36.8. Influence from local rivers was visible as lower salinity near the coast. In Figure 1.3A, the water coming from the Unare River was warmer than the Cariaco Basin water by ~0.5 C; this river’s signal was also visible as a low (~35.5) salinity area near the coast (Figure 1.3B). In contrast the waters in the region of the plume of the Never River were approximately 0.2-0.3 C colder than the surrounding Cariaco Basin water (Figure 1.3A). During September 2003 winds were we aker and the water was stratified. During the rainy season (September 2003), warm water was observed immediately to the southwest of Margarita Island. Figure 1.4A is a current velocity profile, measured at the CARIACO time series station (see Figure 1.1 for reference), for the period of September 2003. During this month, modera te to strong (8-18 cm/s) eastward currents were observed between 30-40 m. The depth of the thermocline at this location was around 100m, and because the wind was weak, it was possible to use these observations to infer the approximate flow cl oser to the surface (Dr. R. Weisberg, pers. comm.). The eastward current apparently help ed contain warm surface water over the continental shelf in the northeastern part of the basin. In the dry season (March 2004, Figure 1.3C and D), temperatures in the basin ranged from ~22 to ~26 C, and salinity from 36.7 to ~37. Colder water was observed near the coast, and the lowest temperatures in March 2004 were recorded near the Gulf of Santa Fe (22.86 C, Figure 1.3C. These results are in agreement with model predictions of the location of the upwelling focus (Wal sh et al., 1999). In March 2004, a slight decrease in surface salinity to ~36.83 (relativ e to >36.86 in most of the Basin) was seen near the northern sill. This was part of th e CSW (<36.75) observed in the Caribbean Sea north of the sill. Figure 1.5 shows SeaWiFS images from the sampling periods, processed for chlorophyll a and CDOM absorption (ag). In Figure 1.5A the combined plume of the Orinoco and Amazon is visible on the extreme eastern margin of the image (right side of image) as a red patch, but it was distant fr om Margarita Island and the Cariaco Basin and

PAGE 21

10 extended to the northwest. Its inclination is very pronounced, almost completely in a northward direction. In the SST image, the sl iver of lower temperature (~ 26 C) visible corresponds to the patch of low color in the chlorophyll image. The Orinoco/Amazon river plume is of higher temperature (~27-28 C) located in the upper right-hand side of the image. In March 2004, the patch ex tending from the east to the west just north of Margarita Island shows the high (> 4 mg/m3) chlorophyll concentratio ns that occur in the region due to upwelling. The Orinoco River plume is in a similar location as seen in September 2003, but it is much smaller and in clined towards the northwest. SST shows a pattern of colder water similar to that of chlorophyll, indicating that this cold (~24 C) water was upwelled water and not riverine wa ter, which is warmer. The Orinoco/Amazon plume is harder to see in this SST image, as compared to September 2003, probably due to the occurrence of some de gree of mixing between warm er river waters and cold, upwelled ones.

PAGE 22

11 Figure 1.3: Surface temperature (A, C) and salinity (B, D) of the Cari aco Basin during September 2003 and March 2004, respectively. Note the different scales used for each cruise. -65.4-65. 2 -6 5 -64. 8 -64.6-64.4-64.2 10 10.2 10.4 10.6 10.8 11 11.2 11.4 22.8 23 23.2 23.4 23.6 23.8 24 24.2 24.4 24.6 24.8 25 25.2 25.4 -65.4-65.2-65-64.8-64.6-64.4-64.2 10 10.2 10.4 10.6 10.8 11 11.2 11.4 36.71 36.73 36.75 36.77 36.79 36.81 36.83 36.85 36.87 36.89 36.91 36.93 36.95 36.97 35.45 35.55 35.65 35.75 35.85 35.95 36.05 36.15 36.25 36.35 36.45 36.55 36.65 36.75 36.85 -65.4-65. 2 -6 5 -64.8-64. 6 -64.4-64. 2 10 10.2 10.4 10.6 10.8 11 11.2 11.4 27.3 27.4 27.5 27.6 27.7 27.8 27.9 28 28.1 28.2 28.3 28.4 28.5 -65.4-65. 2 -6 5 -64.8-64. 6 -64.4-64.2 10 10.2 10.4 10.6 10.8 11 11.2 11.4 B D A C Longitude Latitude

PAGE 23

-150-140-130-120-110-100-90-80-70-60-50-40-30 09/16/0309/26/03 09/06/03 -18-16-14-12-10-8-6-4-2024681012141618202224u (cm/s)Depth Figure 1.4: Vertical distribution of the east/west component of current velocity (cm/s), measured at the CARIACO time-series station (see Figure 1.1 for location) for September 2003. Positive (yellow/red) is eastward flow. Black lines indicate cruise dates. 12

PAGE 24

Chlorophyll September 2003 Sea Surface Tem p erature Se p tember 2003 Chlorophyll March 2004 Sea Surface Tem p erature March 2004 Figure 1.5: SeaWiFS images chlorophyll concentration for September, 30 2003 and March 10, 2004. Sea Surface Temperature composites are also shown. 13

PAGE 25

14 3.1.2. Water mass distribution Figure 1.6 is a temperature/salinity (T/S) diagrams for stations 9 (September 2003) and 26 (March 2004, see Figure 1.2 for reference). Following He rrera et al. (1980) and Morrison and Nowlin (1982), the water masses visible in these diagrams are: (1) CSW, surface waters with high temperatures (>25 oC) and lower salinity (~ 36.6) (2) SUW, characterized by a salinity maximum (~ 37) (3) TACW, monot onically decreasing temperature and salinity below the SUW (4) An tarctic Intermediate Water (AAIW), seen as a salinity minimum (~ 34.7) (5) Upper North Atlantic Deep Water (NADW), visible in the lower portion of the diagra m as a salinity maximum (~ 35). Figure 1.6: T/S diagrams for station 9, September 2003 (x) and station 26, March 2004 ( ). Maximum sampling depth: 1300 m and 1382 m, respectively Figure 1.7 is a meridional tr ansect along 64.88W showing salinities in September 2003 (7A) and March 2004 (7B). The depth of the SUW varied with season. During September 2003, it lay below the CSW, while in March 2004 it occupied the upper 100 m. Herrera et al. (1980) and Astor et al. ( 2003) also noted that the SUW reached its shallowest depths during the first months of the year. Surface salinities were therefore higher during this pe riod (Figure 1.3D). Figure 1.8 shows T/S diagrams from differe nt stations in September 2003 (Figure 1.8A) and March 2004 (Figure 1.8B) In any one cruise, varia tions between stations at temperatures above 21oC (depths above 200 m) were visibl e mainly in salinity. The most

PAGE 26

15 pronounced variations (Figure 1.8B) were cause d by differences between stations inside (19, 6, 14, red and yellow colors) vs. stations outside the basin(35, 26, blue colors; station 29 was located at the sill). Okuda (1976) obs erved similar differences between surface waters inside and outside the basin, and attr ibuted such variations in salinity to the influences of large South American Rivers on Caribbean waters. The smaller differences visible in Figure 1.8A were mostly caused by th e influence of local rivers, such as in station 11 (gray squares; see Figure 1.2A for reference). Figure 1.7: Salinity transects for September 2 003 (A) and March 2004 (B). Red lines correspond to casts performed at stations. Temperature and salinity between 250 m and 400 m in the Cariaco Basin were higher than in the Caribbean st ations north of the sill by approximately 4C and ~0.5. No spatial temperature or salinity changes we re noted below the 17-degree, 36.3 salinity values inside Cariaco (Figure 1.8A). Figure 1.8 C and D, respectively, show th e predicted and measured temperature and salinity increase according to Scranton et al. (1987) for waters below 1200m. Table 1.3 lists the sources of in situ historical data used. The temperature measured during March 2004 followed Scranton et al’s pred iction. However, Marc h 2004 values exceed the expected salinity. In Figure 1.8D there is a noticeable cha nge in the salin ity slope in Stations B A

PAGE 27

16 the past 6 years, which cannot be attributed to errors in the measurements. Salinity for the CARIACO project is measured by duplicates automatically using a CTD and with discrete samples; both sets of measurements are consistent. During September 2003, dissolved oxygen intr usions were observed at stations 19, 17, 16 and 8 (Figure 1.9). At station 19, at l east two intrusions are apparent between 200 and 300 m. In stations 16 and 17 they we re smaller single peaks at ~ 250 and 280 m, respectively. At the CARIACO time-series st ation, oxygen intrusions have been reported (Scranton et al., 2001; Astor et al., 2003). These intrusions, or ventilations, manifest themselves as peaks of dissolved ox ygen concentration between 5 and 50 M. Until now, the only information regarding these ventila tions was what could be observed in the CARIACO time-series records. During March 2004, subsurface oxygen peaks were also observed throughout the basin (Figure 1.10). Two well-defined subs urface oxygen maxima were recorded at station 19 (at ~100 and 160 m). Other subsur face oxygen maxima were seen at other stations throughout the basi n (e.g. stations 13, 14, 11, 20, Figure 1.10). The oxygen profile for station 26 is also shown. Station 26 was located outside the Basin. The Cariaco Basin is anoxic below~250m. The small offs et observed in both Figures 1.9 and 1.10 (ca. 4 M of oxygen) is caused by the CTD’s oxygen sensor, which, despite the fact that there is no oxygen in part of the basin, still pr ovides a minimal read ing. Calibration with discrete measurements has been difficult, and discrete measurements are not enough in numbers to pick up small vertical features like the oxygen intrusions. Table 1.3: Sources of historical temperature and sa linity measurements for waters below 1200 m in the Eastern Cariaco Basin. The CARIACO time se ries measurements are yearly averages Year Potential temperature ( ) Salinity Reference 1955 16.667 Richards and Vaccaro (1956) 1965 16.742 Richards (1970) 1968 16.724 Fanning and Pilson (1972) 1971 16.752 Spencer and Brewer (1972) 1973 16.765 36.196 Bacon et al. (1980) 1982 16.825 36.204 Scranton et al. (1987) 1986 16.864 Casso et al. (1986) 1990 17.041 36.212 Zhang and Millero (1993) 1995 16.881 36.223 CARIACO time series 1996 16.915 36.224 CARIACO time series 1997 16.924 36.225 CARIACO time series 1998 16.934 36.227 CARIACO time series 1999 16.944 36.231 CARIACO time series 2000 16.952 36.232 CARIACO time series 2001 16.961 36.236 CARIACO time series 2002 16.968 36.239 CARIACO time series 2003 16.978 36.240 CARIACO time series 2004 16.989 36.242 This study/CARIACO time series

PAGE 28

18 Figure 1.9: Oxygen profiles for selected stations during September 2003 Figure 1.10: Oxygen profiles for selected stations during March 2004

PAGE 29

19 3.2. Distribution of DOC and CDOM around the Cariaco Basin Concentrations of dissolved organic car bon (DOC) at the surface ranged from ~70 M C (Table 1.2) in stations outside the basin to ~60 M C near the Gulf of Santa Fe (see Figure 1.2 for reference). Surface DOC sa mpled at locations outside the basin came from the Subtropical Underwater mass (Fi gure 1.7). Concentrations of DOC measured outside the basin agreed with SUW DOC concentration meas ured by Del Castillo (1998) near Trinidad. There was a correlation (r2 = 0.87) between temperature and DOC concentration (Figure 1.11A), where highest va lues of DOC were associated with warmer water, and lower values with colder, recently upwelled, waters. Figure 1.11B shows the relationship betw een DOC and CDOM in the Cariaco Basin for March 2004 (r2 = 0.8). Similar high correlations have been noted elsewhere, especially in systems under riverine in fluence, like the West Florida Shelf (r2 = 0.8, Del Castillo et al., 2000), in the Orinoco River plume (r2 = 0.7, Del Castillo et al., 1999) and the Baltic Sea (r2 ~ 0.9, Ferrari et al., 1996). However, no evidence of riverine influence was observed in the basin during March 2004. No DOC samples were collected during the rainy season, and therefore th e relationship between DOC and ag could not be examined. Figure 1.12 shows CDOM fluorescence (FCDOM) and ag sampled during March 2004. Samples collected below 35 m showed a linear correlation between these parameters (r2 = 0.9). However, surface samples showed no cl ear relationship between the variables. Surface CDOM (1-25 m) had a different fluor escence yield than that immediately below the surface at 35 m depth, due to possibl e photochemical altera tions and different sources. Figure 1.11: Property-property plot of (A) DOC ( MC) and Temperature (oC) (B) ag(400) and DOC ( M C) (Stations used are shown in Table 1.2). Both figures refer to March 2004 samples Surface (1-25 m) CDOM fluorescence showed a linear correlation (r2 > 0.8) with chlorophyll fluorescence (FChl) measurements (F igure 1.13), indicating that most of the CDOM observed at this depth range was au tochthonous, i.e. released by phytoplankton and bacteria in the water column (Mague et al, 1980; Bricaud et al, 1981; Carder et al., 1986; Coble et al., 1998). B A

PAGE 30

20 Figure 1.14 is a temperature/FCDOM diagra m for stations sampled during March 2004. The low FCDOM values observed in the upper 25 m of the water column, associated with the highest temperatures (21-24 C), correspond to that portion of the water column where there was high chlorophy ll fluorescence. This was observed at all stations except station 26. From SeaWiFS images of chlorophyll and CDOM absorption, it was possible to see how, during March 2004 (Figure 1.5B), there was a front located between stations east and west of the cha nnel between Tortuga and Margarita Islands. Waters located to the west, such as at st ation 26 showed significa ntly lower nutrient concentration (Table 1.4). This suggests that they also had lower primary production and hence lower CDOM production and ag values. Below 19oC, there was a considerable differenc e between stations inside the basin (stations 14, 19, 20 and 38) and those outside (stations 35 and 26). Inside the basin, FCDOM increased rapidly immediately belo w the oxic-anoxic interface (~200 m), and temperature remained more or less consta nt. Outside the basin, the increase in CDOM was observed deeper, around ~250-270 m. There was a correlation between nutrient concentration and CDOM fluorescence both in side and outside the basin (data not shown), supporting the idea that nutrients in the Cariaco Basin were remineralized simultaneously with CDOM (Chen and Bada, 1992). Figure 1.15A and 1.15B show profiles of FCDOM and oxygen with depth for stations 19 and 14 in Marc h 2004. The relationship between oxygen and CDOM is useful for understanding origins of water masses. Refrac tory CDOM is altered with exposure to Figure 1.12: Correlation plot of CDOM absorption coefficient, ag (m-1) at 370 nm, and corresponding fluorescence emission (expressed in arbitrary units, AU) for stations 6, 29 and 35, sampled during March 2004. Circles ( ) correspond to sampled stations at depths of 35m, 55m and 100m. Other symbols correspond to surface measurements: Station 35, 1m. Station 6, 1m. Station 29, 1m.

PAGE 31

21Table 1.4 : Temperature, salinity, CDOM fluorescence and nu trient data for selected stations sampled during March 2004 Station Depth (m) NO2 ( M) NO3 ( M) PO4 ( M) Si(OH)4 ( M) Salinity Temp. (C) CDOM Fl. (AU) 6 1 0.13 5.91 0.35 3.27 36.94 25.44 1.0313 6 35 0.12 6.50 0.38 3.07 36.92 25.69 1.0385 6 55 0.07 7.98 0.48 3.72 36.85 25.83 1.0406 6 75 0.03 8.33 0.53 4.55 36.80 25.94 1.0387 6 100 0.03 9.66 0.69 6.43 36.71 26.09 1.0425 8 1 0.15 3.72 0.17 1.57 36.94 25.45 1.0208 8 100 0.05 7.47 0.49 4.04 36.77 26.03 1.0287 8 200 0.02 10.09 1.69 24.49 36.45 26.37 1.0403 8 225 0.05 5.10 1.92 33.14 36.42 26.39 1.0481 8 300 0.00 0.15 2.70 45.83 36.37 26.41 1.0857 8 350 0.11 0.19 2.70 46.05 36.35 26.42 1.1004 8 400 0.00 0.15 2.81 54.03 36.34 26.43 1.1086 26 1 0.03 0.10 0.03 0.36 36.80 24.62 26 35 0.24 5.78 0.34 2.39 36.93 25.49 1.0265 26 55 0.22 6.84 0.37 2.48 36.85 25.90 1.0183 26 100 0.02 8.35 0.46 2.83 36.64 26.21 1.0214 26 125 0.02 9.66 0.55 3.27 36.51 26.35 1.0236 26 150 0.06 10.86 0.62 3.77 36.42 26.44 1.0224 26 175 0.01 13.10 0.77 4.88 36.22 26.56 1.0248 26 200 0.00 14.02 0.83 5.18 36.12 26.64 1.0275 26 400 0.00 26.74 1.73 13.96 35.12 27.06 1.0637 35 1 0.03 0.07 0.02 0.46 36.76 24.75 35 35 0.38 7.40 0.51 6.55 36.75 25.43 1.0353 35 55 0.11 7.74 0.44 3.45 36.80 26.03 1.0287 35 100 0.07 8.79 0.49 3.30 36.64 26.24 1.0293 35 230 0.04 16.03 0.98 6.71 35.87 26.74 1.0325 35 350 0.02 25.98 1.68 13.64 35.18 27.02 1.0798

PAGE 32

22 Figure 1.13: Correlation plot of FCDOM (Arbitrary Units, AU) and FChl (Arbitrary Units, AU) for surface (1-25 m) waters sa mpled during March 2004. Stations 29 ( ) and 35 ( ) are shown

PAGE 33

23 Figure 1.14: CDOM fluorescence (FCDOM) and temperature for stations inside (14, 19, 20, 29, 38) and outside (26, 35) the Cariaco Basin in March 2004.

PAGE 34

24Figure 1.15: Vertical profiles of oxygen and FCDOM at stations 19 (A) and 14 (B) in March 2004 B A

PAGE 35

25 light only or slow decompos ition, although labile CDOM is also affected by microbial decomposition or algae uptake (Bricaud et al ., 1981; Carder et al., 1989; Blough and Del Vecchio, 2002). Vertical dissolved oxygen concen trations change in relation to different water masses (Morrison and Nowlin, 1982) and in situ biological processes. Inside the Cariaco Basin during March 2004, average fl uorescence values for the upper 5 m were ~1.03 arbitrary units (AU). Between 5 150 m they increased to ~ 1.04 AU. Values in stations outside the sill varied from ~ 1.01 AU at station 26 to 1.03 AU at station 35 for the upper 5 m. Between 5 200 m, these valu es ranged from 1.02 (station 26) to 1.03 AU (station 35). The low FCDOM peaks associat ed with the oxygen intrusion (Figure 1.15A and B) were of approximately 1.03 AU. 3.3. Vertical distribution of pa rticulate and dissolved matter Figure 1.16A shows absorption and beam a ttenuation coefficient (a, c) profiles collected during the September 2003 cruise at the CARIACO time-seri es station (station 9). The wavelength for absorption observations was 412 nm., which is strongly affected by both CDOM and phytoplankton. At 650 – 6 60 nm, there is minimal absorption by CDOM and particles in the water, and atte nuation at this wavele ngth is commonly used as a descriptor of bulk particle concentrati on (Boss et al., 2001; Bricaud et al., 2002). The peaks in absorption (blue dots) in the upper 50 m of the wa ter column were caused by phytoplankton. The a and c maxima located at 130-150 m in Figure 1.16A coincided with the bottom of the pycnocline that exists in the basin at this depth (Figure 1.16B) (Scranton et al., 1987; Zhang a nd Millero, 1993). Stra tification inhibits vertical mixing above this depth, and at the bottom of th e pycnocline particles may be temporarily retained before sinking. The series of peaks observed betwee n 200 m and 300 m in Figure 1.16A in the beam attenuation profile corr esponded to bacterial layers (Taylor et al., 2001). Similar observations were made by Naqvi et al. (2001 ) in the Arabian Sea and Garfield et al. (1983) and Spinrad et al. (1989) off the coast of Peru. Absorption at 412 nm also showed a maximum between 200-300 m that marked th e location of the oxic-anoxic interface. This absorption peak was observed th roughout the Eastern basin (Figure 1.17). Figure 1.18 shows the relationshi p between beam attenuation and the beam attenuation slope. The spectral shape of the b eam attenuation coefficient can be linked to the particle size distribution (PSD) approximately through = – 3 (Diehl and Haardt, 1980) where is the exponent of the PSD, which in the ocean can be well approximated by a hyperbolic distribution (Junge-like) (Diehl and Haardt, 1980; Boss et al., 2001). As particles become larger, decreases according to the concentration and size of the particles. In the Cariaco Basin, at the surface was larger th an at 150 m, suggesting that particles at the surface were smaller than thos e at the bottom of the pycnocline. It is likely that particles retained at the pycnocline were combined to form larger aggregates (Curran et al., 2003 ).

PAGE 36

26 Below ~200 m, the absorption of CDOM at 412 nm increased (Figure 1.19, dotted line). The ag (412) calculated based on FCDOM measurements in March 2004 was matched to the total absorption measuremen ts (a(412)) taken w ith the AC-9 during September 2003, in stations sampled at the same locations. Water below the oxic-anoxic interface is very stable (Deuser, 1973); due to slow turnover, absorption observed below the interface should be sim ilar year-round. Below 300 m, ag(412) values from March 2004 matched the total absorption measuremen ts at 412 nm taken during September 2003 (Figure 1.19), indicating that below the oxic-an oxic interface most of the absorption was caused by CDOM. The ag (412) profile, as estimated from fluorescence measurements, did not exhibit a maximum at the oxic-anoxic interface. However, a(412) did present a maximum of absorption at th e interface (Figure 1.17). Figure 1.16: (A) Profile of total absorption at 412 nm (a412) and attenuation at 650 nm (c650) (B) Profiles of total absorption at 412 nm (blue), salinity (black) and density (red). Both profiles were measured at station 9 in September 2003. (continued next page) A

PAGE 37

27 Figure 1.16 (Continued) B

PAGE 38

28Figure 1.17: Profiles of total absorption at 412 nm (a412) for different stations inside the basin, sampled during September 2003

PAGE 39

29 Figure 1.18: Total beam attenuation coeffi cient at 650 nm (c650, black) and beam attenuation spectral slope (g blue) measured at station 9 in September 2003.

PAGE 40

30 Figure 1.19: Profiles of total absorption at 412 nm (a412) measured with the AC-9 (solid line), in September 2003 and CDOM absorption derived from CDOM fluorescence at 412 nm (ag(412), dotted line), in March 2004 for stations samples at the same location. (Continued next page)

PAGE 41

31 Figure 1.19: (Continued)

PAGE 42

32 4. Discussion 4.1 Water mass exchange between the Cariaco Basin and the Caribbean Sea Water exchange between the Cariaco Basin and the Caribbean Sea is restricted by a series of sills. The deepest pass on the sill between Tortuga and Margarita Islands is about 140 m. Because of this restriction, onl y CSW, SUW and TACW can move into the basin (Okuda, 1969; Richards and Vaccaro, 1956). During September 2003, all three water masses were observed inside the Cari aco Basin. However, during March 2004, no trace of CSW was found. In March 2004, the SU W entering the basin was being pushed to the surface, occupying the upper 100 m of the water column. Herrera et al. (1980) and Astor et al. (2003) also noted that the SUW reached its shallo west depths during the first months of the year. Higher salinities in March 2004 were associated with the upwelled, colder water, which was the surface expressi on of the SUW. Subsurface water intrusions were observed throughout the basin, bot h during September 2003 and March 2004. These intrusions manifested themselves as peak s of higher dissolved oxygen and lower CDOM concentration than surrounding wa ter. Intrusions were observed both above (March 2004) and below (September 2003) the oxic-anoxic interface. Intrusions of oxygenated water below the oxic-anoxic interface have been obs erved before (Scranton et al., 1987; Astor et al., 2001) and the working hypothesis is that this water originates outside the sill, forced inside the Basin by processes occurr ing in the Caribbean Sea. The observations made by the CARIACO time series program and during these crui ses suggest that subsurface intrusion of oxygenated water into the basin may be more frequent than previously believed. The fact that these events can be traced with both CDOM and oxygen may allow better estimates of volumes of water mixing between the Caribbean Sea and the Cariaco Basin. The above paragraphs indicate that, desp ite the fact that the major influence exerted on the Cariaco Basin by the Caribbean Sea seem to be limite d to the upper 200 m, the deep waters of the Basin are far from a st eady-state. Slight increases with time in the deep temperature and salinity of the basin ha ve been noted by several authors (Scranton et al., 1987; Holmen and Rooth, 1990; Zha ng and Millero, 1993, Scranton et al., 2001). Scranton et al. (1987) calcul ated that the temperature below 1200 m increased by 0.006 C per year. This would predict an in crease of 0.132 C from 1982 to 2004. Similarly, Scranton et al. (1987) noted an increase in salin ity of approximately 8.89 x10-4 per year. March 2004 values follow Scranton’s et al ( 1987) temperature prediction, but exceed the expected salinity. Scranton et al. (1987) attrib uted the increase in temperature and salinity to the diffusive mixing of warm, saline water from the upper layers downwards; they also suggested that geothermal heating plays a sma ll role in the warming process. Holmen and Rooth (1990) modeled the effect of episodic water injection into the deep basin. They determined that, in order to explain the temp erature and salinity changes at depth, water penetration into the deep was necessary. The difference in salinity slope observed during the past 6 years suggests that th ere may have been an increase in the input of saltier water from the surface to the deep basin in more recent years (Scranton, pers. comm.).

PAGE 43

33 4.2. Influence of the Orinoco and Amazon Rivers on the Cariaco Basin and role of small, local rivers During September 2003 and March 2004, no major intrusions of low salinity water were observed from the north, indica ting that, while Amazon and Orinoco river water was present immediately to the north of the sill, it did not have major direct influence on the Cariaco Basin. The presence of Orinoco and Amazon river water outside the Cariaco Basin was evident as part of a low salinity plume (<36.75) observed in the Caribbean Sea north of the sill during March 2004. Howeve r, the Orinoco plume during this time was small and spread west along the northern coast of Vene zuela, flowing north of Margarita Island (Muller-Karger and Va rela, 1990). Caribbean surface waters are regularly influenced by fresh water inputs from the large So uth American rivers. During the period of high discharge of the Amazon, in May and June (Froelich and Atwood, 1978; Hellweger and Gordon, 2002), the surface NE CC is not active and waters along the Brazilian coast flow straight into the Caribbean (Halliwell et al. 2003). The combined plume of the Orinoco and Amazon rivers was visible through SeaWiF S images, but even during the rainy season (September 2003), it wa s distant from Margarita Island and the Cariaco Basin. Waters outside the sill were more exposed to such influences, whereas those inside the Cariaco Basin seemed isol ated. Within the Cariaco Basin, CDOM and chlorophyll concentrations m easured in March 2004 were mostly the result of local upwelling. Influence from local rivers was visible as lower salinity near the coast. Though the low salinity plumes of local rivers were limited mostly to the coast, less saline water was transported by local currents eastward towards the center of the basin, influencing directly the CARIACO time-se ries station. Regularly in the CARIACO time-series station, the lowest salinity is recorded in the month of Se ptember (Astor et al., 1998). Currents in the Cariaco Basin are still poo rly characterized, and though the observations of September 2003 support the above statement, further research is necessary to fully understand water movement in and around the basin. 4.3. Vertical distribution of CDOM in the Cariaco Basin Inside the Cariaco Basin, the upper 200 m of the water column are affected by both open exchange with the Caribbean Sea and local processes, such as local river discharge and primary production. During Se ptember 2003, most of the CDOM observed in the SeaWiFS image was located near the co ast, and areas near the northern sill had very low concentrations of both CDOM and chlorophyll a indicating that there is no significant amount of CDOM en tering the basin from the Caribbean Sea during this period. In March 2004, most of the CDOM ob served in the upper 25 m of the water column was autochthonous, i.e. released by phytoplankton and ba cteria in the water column (Mague et al, 1980; Br icaud et al, 1981; Carder et al ., 1986; Coble et al., 1998). Surface DOC concentrations declined from outside the basin to the coast by approximately 10 M. The relationship between te mperature and DOC suggests that higher DOC concentrations were associated with waters that had already sustained

PAGE 44

34 prolonged primary productivity. DOC is rele ased by phytoplanktonic organisms upon breakdown and decomposition, accumulating du ring the decline of phytoplankton blooms (Chen et al. 1996; Hansell and Waterhous e, 1997). Alvarez Salgado et al. (1999) observed a similar DOC surface distribution in waters off the coast of the Iberian Peninsula, and associated this pattern to wind relaxation and upw elling "spin down". In Figure 1.5, high (> 4 mg/m3) chlorophyll concentrations are visible near the eastern coast of the Cariaco Basin. This chlorophyll was associated with colder temperatures and higher salinities, indicating this water was r ecently upwelled SUW water. To the North of Margarita Island high chlorophyll was also visible, with some filaments extending towards the West. This chlorophyll was in wa rmer waters, most likely an older plume that was already being degraded Below 200 m, there was a difference in CDOM distribution between stations located inside and outside th e Cariaco Basin. Inside the basin, CDOM concentration, measured as CDOM fluorescence, increased rapidly immediately be low the oxic-anoxic interface (~200 m), whereas outside the basin, the increase in CDOM concentration was observed at depths of around ~250-270 m. CDOM fluorescence has been shown to increase with depth, in part due to the rege neration of fluorescence in the absence of high radiation (Kouassi and Zika, 1990, Kouassi et al, 1990), to remineralization (Chen and Bada, 1992) and to microbial activity (Nelson et al., 1998; Mague et al.1980, Coble et al., 1998). The correlation between nutrient conc entration and CDOM fluorescence both inside and outside the basin supported the idea that nutrients in the Cariaco Basin were remineralized simultaneously with CDOM (Table 1.4) (Chen and Bada, 1992). At the oxic-anoxic interface, between 200 and 300 m, the peaks observed in the beam attenuation coefficient corresponded to bacterial layers (Taylor et al., 2001). Similar observations were made by Naqvi et al (2001) in the Arabian Sea and Garfield et al. (1983) and Spinrad et al. (1989) off the coast of Peru. Absorption at 412 nm also showed a maximum between 200-300 m that marked the location of the oxic-anoxic interface, suggesting the presence of particles at the interface with pi gments that absorb blue light, most likely bacteria and viruses. Repeta and Simpson (1991) measured a suite of pigments in bacteria inhabiting the re doxcline of the Black Sea, associated with bacterial anoxygenic photosynthesis. The Cari aco Basin’s bacteria may have a similar suite of pigments, but its purpose would not be photosynthetic, since the Cariaco redoxcline is located much deeper than in the Black Sea (Taylor et al., submitted). Below 300 m, most of the absorption in the Basin was caused by CDOM. This supports the idea that CDOM is generated throughout the water column and accumulates below the interface due to the lack of mixing. The age of this material and its cycling period remains to be determined. The lack of a maximum in the ag (412) profile at the oxic anoxic interface suggests that either CDOM is not produced at a faster rate in this location in the water column, or the CDOM pr oduced by the bacterial population is labile and used up immediately.

PAGE 45

35 5. Conclusion Water exchange between the Cariaco basin and the Caribbean Sea is restricted by a series of sills. Because of this restrict ion, only CSW, SUW and TACW can move into the basin (Okuda, 1969; Richards and Vaccaro, 1956). Waters deeper than ~160 m in the Caribbean Sea do not access the basin. In March 2004, water from outside the Cariaco Basin was observed entering th e basin at depth and upwelli ng near the coast. Surface waters were pushed offshore, eliminating CS W from the study site. The water entering the Basin brought in DOC and dissolved oxygen. Oxygen intrusions similar to those documented by Scranton et al. ( 2001) and Astor et al. (2003) were observed at different locations around the basin in September 2003, between 200 and 300 m. During March 2004, subsurface oxygen peaks were also observed throughout the basin. Oxygen intrusions into the anoxic portion of the basin may be more frequent in time than previously believed. A seasonal difference in both surface temperature and salinity distribution was found between September 2003 and March 2004. During September 2003, surface temperatures in the basin ranged from ~26 to ~28 oC, and salinity from ~36.4 to ~36.8. The influence of local rivers on the basin, as c ontributors of less saline water, was seen in September 2003 near the coast. During March 2004, temperatures were ~3 oC lower than in September 2003; salinities were higher on average by 0.2 throughout the entire study region. No major intrusion of low salinity wa ters was observed from the north during either the rainy or dry season, indicating that the Amazon and Orinoco rivers do not have major direct influence on the Cariaco Basin. There is a difference in temperature a nd salinity between th e Cariaco Basin and the open Caribbean Sea below 200 m. Inside the basin, temperatures and salinities averaged between 200 and 400 m were higher by ~4oC and by ~0.5. Scranton et al. (1987) calculated that deep (~1200 m) temp erature and salinity in side the basin were rising by 0.006 oC and 8.89 x 10-4 per year. Temperature me asurements in March 2004 agree with these observations. However, sali nity at 1200 m was higher than predicted by Scranton et al. (1987), indicating an input of saltier water to the deep basin (Scranton, pers. comm.). Concentrations of dissolved organic carbon (DOC) at surface ranged from ~70 M C in stations outside the basin to ~60 M C near the Gulf of Santa Fe. The lower DOC concentrations were associated with co ld, recently upwelled water, while the higher ones were observed in warmer waters where phytoplankton blooms were likely already in decline. CDOM concentrations increased rapidl y below the oxic-anoxic interface (~200 m) inside the Cariaco Basin. Outside the basin, the increase in CDOM was observed around ~250-270 m. Below 300 m, ag(412) values from Marc h 2004 match the total absorption measurements at 412 nm taken during September 2003, indicating that below the oxic-anoxic interface most of the absorption is caused by CDOM. This supports the idea that CDOM is generated throughout the water column, and accumulates below the

PAGE 46

36 interface due to the lack of mixing. Total absorption profiles exhibited an absorption maximum at the oxic-anoxic interface, suggests the presence of particles with pigments that absorb blue light, most likely bacteria and viruses. The ag (412) profile did not exhibit a maximum at the oxic-anoxic inte rface, making it unclear whether CDOM is produced faster at the redoxcline than elsewhere in the water column or the majority of the CDOM produced is labile.

PAGE 47

CHAPTER II THE INFLUENCE OF LOCAL RI VERS ON THE BIO-OPTICAL PROPERTIES OF THE EASTERN CARIACO BASIN

PAGE 48

37 1. Introduction Continental margins play an important role in the global cycles of carbon and other biogeochemical elements. Most land mate rials enter the sea th rough river runoff in particulate or dissolved form (Liu et al., 2000), and some fraction is deposited on the adjacent continental shelf. The delivery of terrigenous materials via rivers varies according to changes in weather patterns, such as seasonal rainfall and ocean circulation. Ultimately, climate changes have impacts on all of these processes. The Cariaco Basin, because of its anoxic condition and high sedimentation rate, conserves one of the best sediment records available in the marine environment (Hughen et al., 1996; Lin et al., 1997; Bl ack et al., 1999; Peterson el al 2000; Goi et al., 2003). In order to carry out an accurate interpretation of past climate variation based on this record, it is necessary to understand the sources of particulate matter to the basin. Here we examine the contributions of dissolved and detrital material of local rivers to the Cariaco Basin, in an effort to help provide such information. The Cariaco Basin is situated on the cont inental shelf off Venezuela (Figure 2.1), and is composed of two depressions of ~ 1400 m each. It is connect ed to the Caribbean Sea by a shallow sill (~140 m maximum depth; Richards, 1975). The CARIACO oceanographic time series project has collected over 9 years of data, including sediment traps, optics and hydrography in the Cariaco Basin (Muller-Karger et al., 2005). The lowest surface salinities and highest terrig enous inputs at the CARIACO time series station occur in September (Astor et al., 1998) There has been a li ngering question as to whether these changes are associated with lo cal rivers or to influence of the larger Orinoco plume that enters the Caribbean Sea (see Muller-Karger et al., 1989). Three main rivers discharge onto the s outhern Cariaco shelf known as the Unare Platform: the Manzanares, Never and Unare Rivers (Figure 2.1), all originating in the nearby Coastal Mountain Range (Cordillera de la Costa). These rivers show high discharge during the rainy season (July to November) with maximum discharge in August and September. The Unare is the largest of these rivers, with a drainage basin area of 22.3 Km2 and an approximate averag e discharge rate of 56 m3/sec (Zinck, 1977). The Never and Manzanares are sma ller (drainage basi n of 3.9 and 1.0 km2, respectively), discharging approximately 35 and 22 m3/sec, respectively, into the Cariaco Basin (Zinck, 1977). Discharge rates were measured dur ing 1958-1967. Unfortunately, data are not available to assess more recent discharge or ev en variation in flow. Several of the rivers have since been dammed, while others now re ceive contributions from sewer systems, industry and urban areas. In this study we examine the distribution of riverine material near the coast and around the Cariaco basin to help address the question of the relativ e importance of local vs. remote river inputs.

PAGE 49

38 1.1 Objectives The main objective of this chapter is to understand the impact of local rivers on the Cariaco Basin, and whether this is more important than the Orinoco River with respect to sediment delivered to the basin. Sp ecifically, we asked the following questions: What is the distribution of colored dissolv ed organic matter (CDOM) near the coast? What is the distribution of partic ulate material near the coast? What is the accuracy of regional CDOM and chlorophyll satellite observations? Figure 2.1: Schematic of the Cariaco Basin, showing location of the CARIACO time series station (from Astor et al., 2003)

PAGE 50

39 2. Methods Two cruises (Table 2.1) were conducted to the Cariaco Basin in different seasons using the R/V Hermano Gines of the Fundacion La Salle de Ciencias Naturales de Venezuela. The first cruise was conducted in September 2003 (COHRO – C ampaa O ptica e H idrografica R egin O riental; Figure 2.2A). The object ive of this cruise was to study the impact of rivers, bot h local and distant, specifi cally the Orinoco and Amazon, on the Cariaco Basin. 19 stations were pl aced between the coast and the 200 m isobath (Figure 2.2A). The rest of the stations were placed in a grid-like pattern, separated by a distance of approximately 20 km. The second cruise was carried out in March 2004 (COHRO2; Figure 2.2B). The objective was to study th e distribution of upwe lling waters within the Cariaco Basin, and to help understand the characteristics of wa ters entering the basin from the Caribbean Sea. Eight stations were located outside the basin, north of the To rtuga-Margarita sill. Salinity and temperature profiles were measured using a Seabird SBE25 CTD, deployed at each station in a rosette ensemb le. Oxygen data and chlorophyll fluorescence were obtained with a Seabird SBE 43 oxygen se nsor and a Chelsea fluorometer (Chelsea, Inc.), respectively, attached to the CTD. The ensemble also had a C-Star transmissometer (WetLabs) that measured beam attenuation at 660 nm (c660). The data were processed using SeaBird’s SBE Data Processing software. During the September 2003 cruise, one ab sorption/attenuation meter (AC-9 WetLabs) was deployed as part of the hydr ocasts. A pump was used to draw water through the absorption and attenuation t ubes. At 14 stations, double casts were performed, one with an unfiltered AC-9, and the second using a 0.2 m filter (Propor PES filter capsules) attached to the inlet of the absorption tube. The instrument was calibrated before, during, and after the cruise using distilled water as a reference. Optical measurements were taken during both cruises in a nd above the water, using a PRR-600 submersible radiomet er and a surface Spectrascan PR-650 spectroradiometer, respectively. Samples for optical measurement of absorp tion of light due to colored dissolved organic matter (CDOM) were collected at 15 stations during Septem ber 2003 (depth of 1, 10, 15, 25, 50, 100; depth of sampling varied in so me stations due to depth constrains or features of interest), and at 4 stations during March 2004 (depth s of 1, 35 and 100 m). During September 2003, samples were filt ered onboard the ship with a 0.2 m pore-size anotop filter, using a glass syringe and meas ured immediately (see section 2.1). During March 2004 samples were also filtered on board in the same manner as described above, but were later stored in acid cleaned amber-c olored bottles and frozen for transportation. Absorption measurements were conducted with in one month of sample collection at the University of South Florida using an Ocean Optics spectrophotometer (see section 2.1)

PAGE 51

40 2.1. Data processing AC-9 processing and validation AC-9 data were corrected for temperat ure and salinity, following Pegau et al. (1997). A scattering correction was also app lied to the absorption data. The scattering correction was done by subtracting a refere nce wavelength (715 nm) from all data (hypothesis 3 in the AC-9 WetLabs User’s Gu ide, 2003). As a data quality check, the AC-9 beam attenuation data at 650 nm were compared to c660 taken with the transmissometer at all stations. In the filtered samples, a lag in the data was observed, caused by the reduction in flow due to the filter. This lag was adjusted by comparing patterns and features between the ac-9 and th e transmissometer data. Only those spectra where this procedure was possible were used. Because attenuation at 412 nm was lower than absorption at 412 nm, caused most likely by a lack of internal calibration of th e instrument or filter degradation, a new c value was calculated for 412 nm, according to the hyperbolic model c( ) = c( o)*( o)where c( ) is the beam attenuation, c( o) is the attenuation at the reference wavelength and is the spectral slope of c (Boss and Zane veld, 2003), computed he re using all bands except 412 nm; o was 440 nm. Absorption Spectroscopy In September 2003, absorption spectra we re measured between 200 and 800 nm, with a 0.3 nm interval, using a dual fibe r optic spectrometer (O cean Optics) equipped with 10-cm quartz cuvettes. The detection li mit of the spectrophotometer was 0.002, or equivalent absorption of 0.046 m-1. Distilled water was used as a blank. The distilled water was collected the day prior to the start of the cruise and kept in Nalgene polycarbonate carboys inside the airconditioned wet lab of the ship. A running mean was used to smooth the absorption spectra, which was later binned in 1-nm intervals. Each spectrum was examined individually, and the blank absorption was s ubtracted from the sample absorption when absorption at 650 nm was greater than zero. March 2004 CDOM samples were transporte d to the University of South Florida (USF), and analyzed within one month of the cruise. Samples were refiltered once thawed to remove any particles. Samples were scanned between 200 and 800 nm at 1 nm intervals, using a Perkin-E lmer Lambda 18 spectrophotome ter equipped with 10 cm quartz cells. The detection limit of the instru ment was 0.002, or e quivalent absorption of 0.046 m-1. Milli-Q water was used as blank, and for each sample two scans were performed. Running means were used to sm ooth the data, and, where necessary, the blank absorption values were subtracted fr om the sample. The best of two scans was selected. The selection of th e best scan was based on the smoothness and shape of the curve, whether it had features out of the ordina ry, and if it had been necessary to subtract

PAGE 52

41 the blank from the sample. Those scans where no subtractions had been done were preferred. Absorption coefficients were calculated from absorbance, according to the following relationship: a( ) = 2.303 A( )/r (2) where A is the absorbance or optical density and “r” is the pathlength. Light absorption by CDOM was approximated by ag( ) = ag( o)exp[–S( o)] (3) where ag( ) and ag( ) are the absorption coefficients at wavelength and at a reference wavelength S is the spectral slope, which descri bes the decrease in absorption of ag with increasing wavelength (Blough et al, 1993; Green and Blough, 1994; Ferrari, 2000, Blough and Del Vecchio, 2002). Each spectrum wa s plotted in two ways: through a linear least squares regression of the log-transformed data (ln(a) vs. J. Cannizzaro, pers. comm and in an exponential form, using a non-lin ear least squares fit (Dr. C. Hu, pers. omm..). The method that provid ed the single best fit for each individual spectrum was selected. The slope S was calcu lated between 350 and 450 nm. Absorption by phytoplankton (aph) was obtained from in situ fluorometric measurements. The CARIACO project has ov er nine years of data, which include chlorophyll fluorescence and absorption from filt er pads. These measurements were used to calibrate the chlorophyll fluorometer and obtain, for the September 2003 cruise, quantitative values of aph( ) from fluorescence. aph for the CARIACO dataset was measured in situ through filtration following Kishino et al. (1985). The from Mitchell and Kiefer (1988) was used for the correction of the optical path elongation due to filter pads. Reflectance observations PRR and Spectrascan data were colle cted following the NASA Ocean Optics Protocols for SeaWiFS Validation (Mueller an d Austin, 1995). The data were processed with software developed in-house (Dr. C huanmin Hu, USF, personal communication). The diffuse attenuation coefficient, Kd, was computed from downwelling irradiance profiles (Ed) at 412, 443, 490, 532, 555, 665 and 683 nm, following the relationship: Ed ( z ) = Ed(0-) e -Kd* z (Baker and Smith, 1979) (4) Where z is depth and Ed(0-) refers to downwelling irradiance just below the ocean’s surface. Remote sensing reflectance, Rrs, defined as Rrs = Lw/Ed (Mobley, 1992) (5)

PAGE 53

42Table 2.1: Station location and cast depth for COHRO and COHRO2 COHRO, September 16-20, 2003 COHRO2, March 15-20, 2004 Latitude Longitude Station Maximum profiling depth (m) Latitude Longitude Station Maximum profiling depth (m) 10.83 -64.37 0 225 10.83 -64.36 1 217 10.66 -64.36 1 157 10.66 -64.36 2 160 10.50 -64.36 2 400 10.50 -64.36 3A 400 10.33 -64.55 3 120 10.50 -64.36 3B 1220 10.49 -64.55 4 400 10.45 -64.26 4 120 10.66 -64.55 5 400 10.41 -64.38 5 190 10.83 -64.55 6 220 10.33 -64.55 6 127 10.83 -64.70 7 225 10.40 -64.61 7 160 10.66 -64.71 8 400 10.45 -64.55 8A 521 10.49 -64.66 9 400 10.45 -64.55 8B 798 10.33 -64.71 10 40 10.56 -64.55 9A 400 10.33 -64.88 11 56 10.56 -64.55 9B 1220 10.23 -64.88 12 37 10.76 -64.55 10 400 10.14 -65.03 13 15 10.76 -64.71 11 383 10.19 -64.79 14 26 10.76 -64.88 12 400 10.28 -64.8 15 45 10.60 -64.88 13 403 10.49 -64.88 16 400 10.48 -64.88 14 400 10.66 -64.88 17 400 10.33 -64.71 15 50 10.83 -64.88 18 224 10.28 -64.80 16 50 10.66 -65.05 19 400 10.38 -64.93 17 76 10.49 -65.05 20 400 10.50 -65.05 18 400 10.41 -65.05 21 63 10.63 -65.10 19 400 10.33 -65.05 22 59 10.76 -65.05 20 400 10.23 -65.05 23 25 10.93 -65.05 21 67 10.23 -65.21 24 24 11.05 -65.05 22 75 10.33 -65.21 25 56 11.16 -65.05 23 276 10.41 -65.21 26 75 11.33 -65.05 24 400 10.33 -65.38 27 55 11.33 -64.88 26A 400 10.41 -65.38 28 77 11.33 -64.88 26B 1382 10.49 -65.21 29 113 11.16 -64.88 28 276 10.83 -65.04 30 220 11.05 -64.88 29 123 11.00 -65.04 31 80 10.93 -64.88 30 181 11.00 -64.88 32 130 10.93 -64.71 31 310 11.00 -64.71 33 174 11.05 -64.71 32 90 10.99 -64.54 34 273 11.16 -64.71 33 90 10.44 -64.26 48 70 11.33 -64.71 34 210 10.14 -65.11 49 15 11.33 -64.55 35 380 10.16 -65.21 50 10 11.16 -64.55 36 50 10.41 -64.40 51 180 11.05 -64.55 37 84 10.93 -64.55 38 350

PAGE 54

43 Figure 2.2 Sample location for the September 2003 (A) and March 2004 (B) cruises. Major local rivers are shown. -65.4-65.3-65. 2 -65. 1 -65-64. 9 -64. 8 -64. 7 -64. 6 -64. 5 -64.4-64.3-64. 2 -64.1 10 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11 11.1 11.2 11.3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 28 29 30 31 32 33 34 35 36 37 38Gulf of Santa Fe -65.4-65.3-65. 2 -65. 1 -65-64. 9 -64. 8 -64. 7 -64. 6 -64. 5 -64.4-64.3-64. 2 -64.1 10 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11 11.1 11.2 11.3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 48 49 50 51Longitude Latitude Manzanares Never Unare

PAGE 55

44 was derived from above-water measurements for wavelengths between 380-780 nm, at 1 nm intervals. Beam attenuation profiles taken with th e WetLabs CStar transmissometer were used to study the thre e-dimensional distribution of part iculate matter across the basin. A statistical analysis was conduc ted to determine possible rela tionships between the beam attenuation and chlorophyll fluorescence profil es, to identify variab ility corresponding to inorganic or detrital partic les as opposed to phytoplankton. Satellite imagery Two SeaWiFS (Sea-viewing Wide Field-of -View Sensor) images were processed for chlorophyll and CDOM absorption (ag) at 440 nm, using the optimization algorithm of Lee et al. (1999) and the Carder et al. (1999) MODIS algorithm (Dr. C. Hu, Pers. Comm). The dates of the images were chosen as close as possible to the dates of the cruises. For September 2003, the only imag e with low enough cloud cover was for September 30, almost 2 weeks after the cr uise. For March 2004, the image of March 10 was selected.

PAGE 56

45 3. Results 3.1. Surface distribution of CDOM near the coast Observations collected during Sept ember 2003 show that the two main contributors of CDOM to the Cariaco Basin were the Neve r and Unare rivers. There were low (~0.04 m-1 at 400 nm) concentrations of CDOM in surface waters near the mouth of the Manzanares river (Figure 2.3). Tr aces of leaves, plastic and other types of debris were seen at station 48 off the Manzan ares (see Figure 2.2 for reference), but the water was blue, similar to what is expected in open ocean areas away from the coast with no visible suspended matter. Figure 2.4 shows the relationship between CDOM absorption (ag) and salinity for stations sampled during September 2003. A lin ear, conservative mixing is expected between salinity and CDOM concentration from rivers if there are no additional sources or sinks of CDOM (Carder et al, 1989; Del Castillo et al., 1999; Boss et al., 2001; Boss et al., 2001; Cauwet, 2002, Hu et al., 2003). The relationship between CDOM and salinity observed near the coast in Cariaco was not robus t, in part due to the quick mixing of the river water with the coastal seawater. The lowest salinity recorded near the coast was close to 36, indicating that at that locati on there was mostly seawater. A relationship between CDOM and salinity probably exists cl oser to the river mouths. Despite the fact that even near the coast sa linity was not lower than ~ 36, at the CARIACO time-series station lower than average (~ 36.5) salinity is always obser ved during September (Figure 2.5 A) (Astor et al., 1998). This was also th e case for September 2003. Figure 2.5B is a current profile measured near the CARIAC O time-series station for September 2003; a relatively strong eastward curr ent (~22 cm/s) was observed between 30-50m. The depth of the thermocline at this location was around 100m, and because the wind was weak (~2-4 m/s as measured from the R/V Hno. Gi nes in stations near the coast), it was possible to use these observations to infer the approximate flow at the surface (Dr. R. Weisberg, pers. comm.). Less saline water from local rivers is the cause of the low salinity observed yearly at th e CARIACO time-series stati on, which is transported by local currents as far north as 1030’.

PAGE 57

46 Figure 2.3: Visible changes in water color in September 2003. Station 48 is located near the Manzanares River, station 14 close to the Never River and stations 49 and 50 near the Unare River

PAGE 58

47 Figure 2.4: Relationship between CDOM absorption coefficient (ag(400)) and salinity for stations 4, 11, 12, 13, 14, 16, 20, 21, 22, 50, and 51 sampled in September 2003. Figure 2.6 shows the relationship of S, the slope of ag (equation 3), and ag(400) in the Cariaco Basin. Variations in S contain inform ation about the nature of CDOM (Nelson et al., 1998; Blough and Del Vecchio, 2002). Farther offshore, in the central and northern Cariaco Basin, S was higher (>0.02 nm-1) than in coastal regions under the influence of rivers (0.010-0.018 nm-1). Offshore waters not affected by rivers, sampled during March 2004, showed low ag(400) (~0.03 m-1) and high S (0.024 nm-1). Areas near the coast, sampled during September 2003, showed high ag(400) (~0.25 m-1) and low S (~0.011 nm1). Table 2.2 compares the range of S obser ved in the Cariaco Basin during September 2003 and March 2004 with data from other locations. The range of S measured was similar to that reported by other authors, both for coastal and open waters (Bricaud et al., 1981; Nelson and Guarda, 1995; Carder et al., 1999; Cannizzaro, 2004). Some extreme slope values (e.g. 0.0105 nm-1) were associated with high (~ 0.25 nm-1) concentrations of CDOM, suggesting that the estimate of S was sensitive to particle scattering and baseline offsets (Green and Blough, 1994). When ag values were too low (~0.02 m-1) there was a signal to noise problem, in which the instruments used to measure ag was not sensitive enough for such low concentrations. This yielded a higher than normal S (e.g. 0.028 nm1) (Dr. K. Carder, pers. comm.). Such extremes should be treated carefully.

PAGE 59

48 Figure 2.5: (A) Vertical profile (from 1-400m) of salinity at the CARIACO time-series station from January 1996 to July 2004. Contour lines are in terpolated between measurements. (B) Vertical distribution of the east/west component of current velocity (cm/s), at the CARIACO time-series station (see Figure 2.1 for location) for September 2003. B -150 -140 -130 -120 -110 -100 -90 -80 -70 -60 -50 -40 -30 09/16/0309/26/03 09/06/03 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24u (cm/s)Depth -400 -300 -200 -100 199619972001199820001999200220032004DepthYears 36.1 5 36.25 36.35 36.45 36.55 36.65 36.75 36.85 36.95 37.05A

PAGE 60

49 Figure 2.6: Relationship between the spectral slope (S) and CDOM absorption (ag) at 400 nm. Samples used include those collected during September 2003 ( ) and March 2004 ( ) Table 2.2: CDOM absorption slopes, S Location S (nm -1) Reference Range (nm) Mediterranean and Baltic Sea 0. 012-0.018 Bricaud et al. 1981 375-500 GOM 0.011-0.017 Carder et al. 1989 370-440 South Atlantic Bight 0.017-0.023 Nelson and Guarda, 1995 300 ~500 West Florida Shelf 0.0140.025 Cannizzaro, 2004 350-450 Cariaco Basin 0.011-0. 026 This study 330-400 3.2. River influence on the optical prop erties of Cariaco's coastal waters Optical properties of the water can be us ed to determine the spatial variability of riverine influence in the Cariaco Basin. Tota l absorption at depths of 1, 5, 10, and 25 m, measured during September 2003, are shown in Figure 2.7. Changes in absorption of light in natural waters are caused mainly by phytoplankton (aph), CDOM (ag) and detritus (ad) (Roesler and Perry, 1995; Green et al., 2003). Absorption by water (aw) is well know and constant (Pope and Fry, 1997). The absorpti on of light in the ocean is additive: aT = aph + aCDOM + ad [m-1] where aT stands for total absorption (Roesler et al., 1989; Green et al., 2003). At stations near the coast (22, 21 and 11, Figure 2.7) small absorption peaks were noted in all depths

PAGE 61

50 at 676 nm. These peaks indicated the presen ce of phytoplankton. The typical absorption shape of phytoplankton in the blue part of the spectrum was masked by CDOM. Chlorophyll concentrations in the co astal area varied between 0.04-1.58 mg/m3; concentrations were highest between 10-20 m, where the deep chlorophyll maximum was located. Station 21 showed some of the strongest absorption at 10 and 25 m, which corresponded to higher chlorophyll concentrations. The combined percentage contribution of detritus and CDOM (aCDM) to total absorption was assessed and compared to the contribution by phytoplankton. The relationship between aph at 676 nm and chlorophyll fluor escence was calculated using data from the CARIACO time-series project. This regression resulted in the following equation: aph(676) = 0.05778 FChl +0.00003 (r2 = 0.92) Absorption at 676 nm was linearly related (r2 = 0.98) to aph(412) by aph(412) = 1.5719 – aph(676) + 0.0071 Other authors have used power functions in these relationships (Claustre et al., 2000), to account for packaging effects and accessory pigments. Because the concentration of phytoplankton was very low, in this case a linear relationship is probably more accurate. aph(412) was subtracted from total absorption to obtain the absorption at 412 nm effected by detr ital and dissolved matter. The percentage contribution of CDM at 412 nm to total absorp tion at 412 was then calculated (Claustre et al., 2000): %CDM(412) = aCDM/aT(412) 100 At some stations, absorption estimates by CD OM at 412 nm were available from the AC9 filtered data, and the actual percentage contribution only by CDOM was calculated. In coastal areas, over 90% of the absorption was caused by detrital and dissolved matter. In stations close to the coast (22, 21 and 11) over 80% of this absorption was caused by CDOM alone.

PAGE 62

51 Figure 2.7: Total absorption measured during the September 2003 at (A) 1m (B) 5m (C) 10m (D) 25m A C B D

PAGE 63

52 The contribution of CDOM decreased with distance from the coast. In stations away from the influence of rivers, aT(412) decreased by half (Figure 2.7, station 48 and 2), and absorption was likely dominat ed by CDOM and non-algal particles. Figure 2.8 shows changes of light with de pth, expressed as optical depth (OD) and derived from diffuse attenua tion coefficient measurements (Kd) as follows: = Kdz (Kirk, 1994) Figure 2.8: Optical depths at selected stations sampled during September 2003. Station 50 ( ) Station 51 ( ) Station 24 ( ) Station 14 ( ). Kd can be used as an indicator of bio-opt ical state of natural waters (Baker and Smith, 1982; Farmer et al., 1993). Similar to absorption, the combined effects of dissolved and particulate matter affect Kd and optical depth. Of sp ecial interest is the OD where the downward irradiance of PAR (photos ynthetically available ra diation) is 1% of that just below the surface (Kirk, 1994). Th is depth corresponds to the bottom of the euphotic zone and is of importance for photos ynthetic organisms. At this point in the water column equals to 4.6 (Kirk, 1994); this depth will hereafter be referred to as OD(zeu). A difference in light penetration thr ough the water column was observed in September 2003 between stations close to th e coast and those farther away (Figure 2.8). Stations 50 and 14, roughly 10 km from the m ouths of the Unare and Never rivers, respectively, reached the 1% light level at 412 nm around 30 m. Stations 51 and 24, further away from the coast, reached this point at 50 and 59 m, respectively. The

PAGE 64

53 transparency window for all the stations wa s near the green, at 488 nm. The depths at which the 1% light level in the blue was r eached far exceeded th e real depth of each station, indicating that light penetrates the entire water column, despite the high concentrations of CDOM and detrital suspended matter. Figure 2.9 shows SeaWiFS images processed for ag(440) and chlorophyll, from September 30, 2003 and March 2004. The images were processed using two different algorithms, discussed in Section 3.4. In the September 2003 images, a narrow rim of color is visible near the coast. This was most likely suspended and dissolved material ejected by the rivers into the basin. Figure 2.10 is a cross-shelf profile of beam attenuation coefficient collected during September 2003 from station 13 (located near the mouth of the Unare River) to station 21 (see Figure 2.2A for refere nce). Higher attenuation (0.6 m-1) was observed at ~15 m depth at station 13, as compared to station 21 (0.2 m-1). At station 23, located approximately 10 Km from station 13, the attenu ation was almost half of that measured at station 13 at 15 m depth (0.4 m-1). Figure 2.11A shows the distribution of th e beam attenuation coefficient at the surface for September 2003. The plume from the Unare River caused high light attenuation (~ 1 m-1) near its mouth (station 49), due to the suspended material carried along by the river. As distance increased from the river mouths, attenuation was reduced, suggesting that there were fewer suspended par ticles. Most particles associated with the rivers seem to have precipitated from surf ace waters within 10 km from their mouths. 3.3. Particle and sediment tr ansport from the coast In Figure 2.10, higher attenuation was observe d near the bottom of stations 21 and 22 (~0.2-0.3 m-1) compared to waters just above that (~0.15 m-1). Figure 2.11B shows the distribution of beam attenuation coefficient ne ar the bottom of each station in September 2003. Higher attenuation was obs erved throughout the Unar e Platform, with higher values (~ 0.6 1 m-1) near the bottom of stations 49 and 13. Figure 2.12 is a beam attenuation plot of the st ations surrounding the Never River (see Figure 2.2 for reference). Higher beam attenuation (0.5-0.6 m-1) was also visible at two stations near the bottom: station 12 and 14. The interpolation of beam attenuation va lues of Figure 2.11B suggests that the higher coeffi cients observed at station 12 were most likely caused by sediments coming from the Unare River and not the Never. The thickness of the bottom attenuation maximum varied. Near the Never River, the bottom nepheloid layer ranged from 6 to 15 m, while near the Unare it was <5m. Intermediate nepheloid layers were seen at stations 15 and 11 at depths of 35 and 45 m, respectively (Figure 2.12).

PAGE 65

54 Figure 2.9: SeaWiFS images of CDOM absorption at 443 nm (ag(443)), chlorophyll concentration (chl a ) and aph (443), processed with Lee et al. (1999), Carder et al. (1999) and Carder et al. (2004) algorithms. Images are from September 30, 2003 and March 10, 2004 Carder et al. (1999), September 2003 Carder et al. (2004), March 2004 Lee et al. (1999), September 2003 Lee et al. (1999), March 2004

PAGE 66

55 Figure 2.10 : Beam attenuation (660nm) transect from station 13 to station 21 sampled during September 2003. Contour lines are interpolated between stations. Black lines indicate profile locations and depths. Other stations sampled during Septembe r 2003 also exhibited attenuation maxima both near the bottom and at intermediate wate r column depths (stations 0, 1, 3 and 51). The plume of the Manzanares River was less apparent than that of the Unare and Never Rivers. However, intermed iate nepheloid layers (~0.09 m-1) (INL) were observed in September 2003 near the Manzanares Ri ver at stations 48 and 2, located over a submarine canyon (Figure 2.13). These nepheloid layers were less pronounced than those observed near the southern coast at stations 15 and 11 (~0.2 m-1). The peaks in beam attenuation corresponding to the INL did not show a corresponding variation in absorption. The INL observed at 10o29’ N 64o21’ W was measured both in September 2003 (station 2, Figure 2.13A) and March 2004 (station 3, Figure 2.13B). During September 2003, this peak (located at ~220 m) was thic ker (~ 50 m) than the one observed at a similar depth in March 2004 (~ 30 m). These laye rs were similar to those reported by Pak et al. (1980) for the West coast of South America. During March 2004, a shallower, thinner and more a ttenuating (~ 0.15 m-1) layer was also seen around 100 m. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 10.210.310.4 -60 -40 -20 LatitudeDepthm-113232221 Station

PAGE 67

56Figure 2.11: Horizontal distribution of the beam attenuation (660 nm) near the eastern coast of the Cariaco Basin during Septem ber 2003. At the surface (A) and at the maximum de pth of each st ation (B). m-1 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 48 49 50 51 -65.4-65.3-65. 2 -65. 1 -6 5 -64.9-64. 8 -64. 7 -64. 6 -64.5-64.4-64. 3 -64. 2 10 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11 11.1 11.2 11.3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 48 49 50 51 -65.4-65.3-65.2-65.1-65-64.9-64.8-64.7-64.6-64.5-64.4-64.3-64.2 10 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11 11.1 11.2 11.3Longitude Latitude A B

PAGE 68

57 Figure 2.12: Beam attenuation (660nm) profiles near the Never River during September 2003. For station location refer to Figure 2.2A The waters inside the basin during Ma rch 2004 were well mixed, due to the intense winds that blow in this region dur ing the upwelling season, and this shallower nepheloid layer was not associat ed with any density structure. It was not linked to the deep chlorophyll maximum either, leav ing its formation an open question. For the nepheloid layer observed near 200 m at 10o29’ N 64o21’ W in September 2003, the spectral shape of the b eam attenuation coefficient ( ) was calculated. can be linked to the particle size distri bution (PSD) approximately through = – 3 (Diehl and Haardt, 1980) where is the PSD exponent (Boss et al ., 2001). As particles become larger, decreases according mainly to the si ze of the particles. at the nepheloid layer was smaller, indicating it was probably co mposed of larger partic les, perhaps aggregates. The range of attenuation coefficients observed in the water column of the Cariaco Basin in September 2003 and March 2004 varied between 0.1 and 0.22 m-1. This included locations near the rivers and over the Ea stern Deep, but exclude d the intermediate nepheloid layers. Puig and Palanques (1998) observed attenu ation between 0.5 and 0.9 m1 in the Foix Canyon on the Barc elona continental margin. McPh ee-Shaw et al. (in press) and Snyder and Carson (1986) obs erved a similar range over the California continental margin and the Quinault Canyon, respectively.

PAGE 69

58Figure 2.13: (A) Profile of total absorption at 412 nm (a412 – dotted line) and attenuation at 650 nm (c650-solid line) at station 2 (10o29’ N 64o21’ W ) during September 2003 (B) Beam attenuation profiles at 660 nm (c660) at 10o29’ N 64o21’ W for September 2003 (dotted line) and March 2004 (solid line). Red circles indicate INL (Intermediate Nepheloid Layer) location (C) Total beam attenuation coefficient at 650 nm (c650, black) and beam attenuation spectral slope ( blue) at station 2 (10o29’ N 64o21’ W ) during September 2003. Arrow indicates the location of the maximum phytoplankton biomass; red circle indicates location of the INL (Continued next page) C B A

PAGE 70

59 Figure 2.13 (Continued) The low-end, or background values of b eam attenuation measurements of the authors mentioned above, which would corres pond to clear, particle-free waters, were close to 0.1 m-1 or above. The low-end in the Ca riaco Basin was approximately 0.05 m-1. Taking into account this offset between meas urements, beam attenuation values observed in the Cariaco Basin were similar to t hose reported for other near shore regions. 3.4. Remote sensing reflectance of Cariaco waters Figure 2.14 shows remote sensing reflectan ce (Rrs), at different stations during September 2003. Rrs(750) was used as an of fset. The contribution of dissolved and suspended particulate material from local ri vers to the Cariaco Basin lead to marked changes in the color of surface waters (Figure 2.3). Some coastal stations were dominated by reflectance in the green (e.g. stations 50 a nd 49). Rrs at 532 nm. at station 49 (0.0109 sr-1) was somewhat higher than at station 50 (0.0101 sr-1). At station 49, located in front of the Unare River mouth, surface (~1m) beam attenuation (~0.9 m-1) was higher than at station 50 (~0.5 m-1, see Figure 2.11), suggesting the c oncentration of suspended matter was higher. Rrs for station 14 was ~0.008 sr-1, considerably lower than at stations 49 and 50. Figure 2.14 also shows low reflectance in the blue at stations near the coast. This was most likely due to the strong absorption by CD OM at these locations. Waters only a few (~30) kilometers farther offshore showed high Rrs in the blue. C

PAGE 71

60 Figure 2.14: Remote sensing reflectance spectra measured in situ in September 2003 Lee's et al. (1999) bio-optical optimization m odel and the Carder et al. (1999) global biooptical algorithm developed for MODIS were applied to Cariaco waters to test the retrieval of inherent optical properties (IOP's) from reflectance data. Their performances are later compared. To initially test the algorithms, Rrs and aph(440) measurements from the CARIACO time-series project (10o30’N, 64o40’W) were used. Due to the large dataset the CARIACO Projec t has gathered, only three years (2001, 2002 and 2003) of the CARIACO Project Rrs measurements were used for testing the model. At the CARIACO time-series station, Rrs measurements are collected monthly; a total of 34 Rrs spectra were analyzed. Each selected spectr um was carefully examined and those that were suspicious, based on the shape of the Rr s and sky reflectace curves, were discarded. The Rrs data were compared with SeaWiFS images of the corresponding cruise day or closest available. The satellite imag es rarely matched the day of the in situ sampling, but the Rrs spectra measured with SeaWiFS for the available dates were always within 10% of the Rrs gathered in situ aph was derived from in situ measurements, as described in the Data Processing section. To obtain more accurate results using th e Lee et al. (1999) model, empirical coefficients for ao( ) and a1( ) (Lee et al, 1998) were derived from the in situ data. Due to the high noise in some of the ap and ad measurements, averaged ao and a1 were used. Error to assess the models’ effectiveness wa s calculated between the measured and the modeled data using:

PAGE 72

61 %Error = (aph_mo-aph_mea)/aph_mea *100 where aph_mea was the measured value of aph and aph_mod was the value obtained through the model. Table 2.3 presents some of the results obtained using the CARIACO Project Rrs measurements. The Lee et al. (1999) mode l error estimated, based on the CARIACO Project data, ranged between 2 and 135%. The e rror was lowest when data from either the dry or the rainy season were used (T able 2.3). These months included January, February and March during the dry season, a nd July, August and September during the rainy season. The error obtained using th e model was generally highest during “transition” months between the rai ny and dry season, when there was low aph(440) (e.g. 0.01 mg/m3). aph was generally underestimated by th e model during these “transition” months. The Carder et al. (1999) algorithm ha d less error variation between the selected dates (Table 2.3). Its errors ranged between 40 and 80%, with no clear reason for such variations. Chlorophyll was generally underes timated as well with Carder et al.'s algorithm. This systematic underestimation of a ph (440) is consistent with Carder et al. (1999) The Lee et al. (1999) and Carder et al. (1999) algorithms were then used for estimating ag(440) and chlorophyll a concentration directly from SeaWiFS images from September 30, 2003 and March 10, 2004 (Figur e 2.9). Rrs measurements from the satellite were in good agreement (r2 >0.9) with Rrs measured in situ during the September 2003 cruise. During March 2004 Rrs from SeaWi FS was one order of magnitude lower in the blue than Rrs in situ (~0.004 vs. 0.0004 sr-1 for stations in the upwelling plume). This problem arise most likely from errors in th e atmospheric correction of SeaWiFS. When aerosols in the atmosphere are non-absorbi ng, atmospheric correct ions using the NIR (near infrared) part of the spectrum work well, but when the aerosol absorption is not negligible in the ne ar infrared, this technique fa ils (Chomko and Gordon, 2001). African dust is a strong absorber and is transported across the A tlantic year round. Though during the northern hemisphere summer months the amount of dust transported peaks, reaching altitudes of 5-7 km in the atmosphere and extending over large areas (from the Caribbean Sea and the south-east United States) (Car lson and Prospero, 1972) during the northern hemisphere winter there are al so large quantities of dust carri ed out of North Africa and across the Atlantic, but its transport is pr imarily to South America (Prospero et. Al., 1981). In September 2003, both algorithms perfor med very similarly (less than 10% difference from one another) for the estimation of both ag and chlorophyll a At locations away from the coast where both ag(440) and chlorophyll were low (< 0.1 m-1 and 0.01 mg/m3), ag(440) and chlorophyll con centration were underestim ated by more than 80% by the algorithms relative to the in situ validation observations (T able 2.4, stations 18 and 32). In stations located betw een the CDM-dominated waters and clear waters (stations 4, 20 and 21), both algorithms underestimated in situ CDOM observations by >50%. Chlorophyll was both under and overestimate d, depending on location, by more than 80%. It is not clear what made the models under or overestimate chlorophyll concentrations in this region. In areas highly influenced by rivers (stations 11 and 13),

PAGE 73

62 ag(440) was underestimated by up to 70%. The Lee et al. (1999) algorithm performed better in the estimat ion of chlorophyll than the Ca rder et al. ( 1999) algorithm, underestimating concentrations by less than 50% (Table 2.4). In March 2004, in areas of high ch lorophyll concentration (> 4 mg/m3), neither the Lee et al. (1999) nor the Carder et al. (1999) algorithms were able to derive any values, mainly because the water leaving radiance in the blue was very low (~0.02 Wm2sr-1). The Carder et al. (2004) expanded algorithm was used instead, which was able to provide values of chlorophyll and ag for the upwelling plume. Table 2.3: Error obtained using the Lee et al. (1999 ) and Carder et al. (1999 ) algorithms to estimate aph(440) at the Cariaco time-series station CARIACO project Month % Lee et al. (1999) error % Carder et al. (1999) error Aug-01 102.492 44.891 Aug-02 109.502 60.717 Feb-03 62.179 48.185 Jan-03 23.838 65.377 Jul-00 14.608 70.744 Jul-01 135.000 57.807 Jul-02 21.607 71.058 Jul-03 -2.256 70.989 Jun-02 99.615 63.008 Jun-03 118.605 60.039 Mar-02 36.989 54.160 May-02 52.593 59.933 Oct-02 63.087 67.195

PAGE 74

63Table 2.4: Error obtained using the Lee et al. (1999 ) and Carder et al. (2004 ) algorithms to estimate chlorophyll con centration and ag(440) in waters of the Cariaco Basin during September 2003 and March 2004. Undet. refers to undetermined value Cruise Location Station Latitude Longitude % Lee et al. (1999) error for ag % Lee et al. (1999) error for chl % Carder et al. (1999) error for ag % Carder et al. (1999) error for chl COHRO 32 11.00 -64.88 90.95 >100 91.93 87.53 COHRO 11 10.33 -64.88 61.13 48.47 67.52 >100 COHRO 13 10.14 -65.03 17.05 19.51 20.24 >100 COHRO 20 10.49 -65.04 78.89 >100 82.26 82.20 COHRO 18 10.83 -64.88 80.79 40.83 83.53 66.85 COHRO 4 10.49 -64.55 59.61 28.38 66.00 68.71 COHRO 21 10.41 -65.04 81.07 >100 83.74 82.60 COHRO 2 14 10.48 -64. 88 16.62 99.33 77.88 62.45 COHRO 2 21 10.93 -65.05 Undet. Undet. 10.14 88.28 COHRO 2 22 11.05 -65.05 72.74 65.93 >100 25.23 COHRO 2 8 10.33 -64.55 >100 -98.99 29.53 29.53 COHRO 2 18 10.5 -65.05 >100 90.44 >100 24.06 COHRO 2 1 10.63 -64.36 Undet. Undet. >100 57.87 However, when chlorophyll concentr ations are higher than ~1.52.0 mg/m3, when Rrs (412) is too low for the semianalytical retrievals, the algori thm defaults to an empirical expression, which uses a band ratio between Rrs(488) and Rrs(551), along with empirically derived constants, to provi de chlorophyll concentrations. For high chlorophyll concentration areas, the Carder et al. (2004) algo rithm underestimated chlorophyll concentrations by ~50-70%. This underestimation was caused by the problems with the atmospheric correction, wh ich returned a lower than real Rrs. ag was overestimated by more than 90%, in part cau sed by the same low water leaving radiance, and in part to the fact that absorption by detritus and CDOM are combined when retrieving them using inversion models. In areas where the in situ chlorophyll concentration was between 2 – 3 mg/m3, the Lee et al. (1999) algorithm calculated only ag(440), which was highly overestimated (> 100%). In those areas where chlorophyll concentrations and ag(440) were estimated by the Carder et al. (1999) algorithm (where chlorophyll concentrations were below 3 mg/m3 – refer to Figure 2.9), chlorophyll concentrations were underestimated by 30-90%. ag(440) was generally overestimated by >50% (Table 2.4)

PAGE 75

64 4. Discussion 4.1. Colored dissolved organic matter (CDOM) distribution in the eastern Cariaco basin The two main contributors of CDOM to the Cariaco Basin during September 2003 were the Never and Unare ri vers. Riverine CDOM affect ed primarily a narrow (30-40 Km) rim of the basin along the coast. Away from the coast, CDOM concentrations in Cariaco were generally low (absorp tion coefficients at 412 nm of 0.03 m-1). The range of S measured was similar to that reported by other authors, both for coastal and open waters (Bricaud et al., 1981; Nelson and Gu arda, 1995; Carder et al., 1999; Cannizzaro, 2004). The changes in CDOM slope from the co astal area to the open basin were likely due to different sources of CDOM, and were indicative of the mixing between riverine and autochthonous material. CDOM distribution has generally been stud ied in relation to lig ht distribution and its effect on remote sensing, primary produc tion and the carbon cycle. The presence of CDOM in surface waters of the Cariaco Basi n changed significantly its color. Coastal stations affected directly by local rivers were dominated by a strong reflectance in the green, whereas waters farther offshore showed high Rrs in the blue. The contribution of CDOM to the Rrs signal will have an impact on the estimates of IOP’s from satellite remote sensing data (see S ection 4.4), especially in areas close to the coast. The plume of the Manzanares River was less apparent than that of the Unare and Never Rivers. There appeared to be little CDOM inject ed to the Basin through the Manzanares River. To the north of the Basin, CDOM was low (~0.03 m-1), indicating that the CDOM ejected by the Amazon and Orinoc o does not influence the basin directly. 4.2 The distribution of particulate mat erial in the eastern Cariaco basin The Unare and Never rivers were the largest source of particulate matter to the Eastern Deep of the Cariaco Basin. Beam attenuation measurements suggest that the Unare River discharged more suspended particulate matter th an the Never River. The distribution of detrital matter in the Cariaco Basin for September 2003 was similar to that of CDOM. Most particulate ma terial was located close to the mouths of the rivers, decreasing with distance from the coast. Su spended particles associ ated with the river largely precipitated from surface waters with in 10 km from the river mouths. Maximum beam attenuation coefficient was observed near the bottom and was associated with bottom nepheloid layers (Pak et al., 1984; Boss et al., 2001; Xu et al., 2002; McPheeShaw et al., in press). The Unare’s bottom nepheloid layer was obser vable to at least 40 km from the mouth of the river, while the Ne ver’s was close to 23 km. This was a larger radius of influence than seen in the susp ended and surface distri bution of particulate matter.

PAGE 76

65 Thunell et al. (2000), Goni et al. (2003) and Goi et al. (2004) have reported terrigenous material in sediment traps lo cated near the CARIACO times-series station. Local river water affected th e salinity measured at the CARIACO time-series station. However, sediments discharged by the local rivers in September 2003 did not affect the optical properties of the waters at the timeseries station. The mate rial observed in the traps is probably derived from local rivers, but is likely transported at intermediate levels in layers moving off the bottom of the continental margin. More intensive sampling is necessary to reach a better understanding of sediment transport and distribution within the basin. Although the Unare and Never are li kely the largest sources of terrigenous material to the Basin, contri butions from the Manzanares and Tuy Rivers may also be important. The Tuy River affects mostly the western Cariaco Basin. Its drainage basin occupies an area of 6.6 km2 and has an approximate discharge of 82 m3/sec. (Zink, 1977). The Tuy River is larger than the Unare, a nd its plume has been observed flowing toward the center of the Western basin (Figure 2.15). Sediment cores analyzed by Hughen et al. (1996) were recovered from the 900m deep saddle, located betw een the Eastern and Western Deep. Given the distan ce from the southern coast to the saddle, and current observations of the Unare and Never Rive r plumes, it seems unlikely that terrigeous material from the southern coast would be ca rried this far into the basin. The sediment observed by Hughen et al. (1996) could have b een transported by the Tuy’s river plume. Nepheloid layers present at intermediate depths (150-300 m) could constitute an important pathway of coastal sediment transp ort to the central Ca riaco Basin. Several stations sampled around the Eastern Deep exhib ited turbid layers at intermediate depths. Of particular importance was a layer observed at 10o29’ N 64o21’ W over a submarine canyon. Hickey et al. (1986) determined that the dominant mode of sediment transport off the shelf in Quinault Canyon was by epis odic formation of turbidity layers at intermediate depths caused by horizontal adv ection. Puig et al. (2003) also observed such a persistent layer, which contributed to th e off-shelf sediment transport in Eel Canyon. Puig and Palanques (1998) suggested that in canyons located on continental margins, such as the Foix Canyon, sediment trans port could be dominated by intermediate nepheloid layer detachments and internal waves. The intermediate nepheloid layer observed at 10o29’ N 64o21’ W in Cariaco did not seem to be associated with any temperatur e or salinity feature. It is unlikely that it formed from only vertical particulate flux, a nd indeed waters above and below the layer were relatively devoid of susp ended particles. The ~220 m turbidity layer could have been an offshore extension of a shelf-break intermediate nephelo id layer, and could constitute an important means of terrigenous se diment transport into the deeper portions of the Eastern Cariaco Basin. During glacial times, enhanced terrigenous input to the Cariaco Basin may have occurred (Peters on et al., 2000). The s ources were probably local rivers and aeolian dust (Yarincik et al., 2000). Part of this enhanced sediment input could also be responsible for the large exte nsion of the Unare platform. During glacial times the sea level would have been a pproximately 121 m lower (Fairbanks, 1989), exposing the Unare platform (Figure 2.16). The Unare and Never Rivers would have had a more direct drainage into the basin, a nd sediment transport through the submarine canyon near the Manzanares River would ha ve been enhanced, driven by gravity mechanisms such as turbidity currents and debris flows.

PAGE 77

66 Figure 2.15: SeaWiFS images of chlorophyll a for the eastern coast of Venezuela (A) September 25, 2000 (B) September 26, 2000 (C) October 25, 2001 (D) October 24, 2003. The approximate location of the Tuy River is indicated by the red arrow D A C B

PAGE 78

67 Figure 2.16: Bathymetry of the Cariaco Basin. The red segmented line marks appr oximate sea level during the Last Glacial Maximum (Fairbanks, 1989) -65.4-65.3-65.2-65.1-65-64.9-64.8-64.7-64.6-64.5-64.4-64.3-64.2-64.1 10 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11 11.1 11.2 11.3 mLon g itudeLatitude -1350 -1300 -1250 -1200 -1150 -1100 -1050 -1000 -950 -900 -850 -800 -750 -700 -650 -600 -550 -500 -450 -400 -350 -300 -250 -200 -150 -100 -50 0 248

PAGE 79

68 4.3. Remote sensing of CDOM and chlorophyll in the Cariaco Basin It is a challenge to retrieve inherent optical properties using remote sensing data in coastal areas, and particularly it is di fficult to properly differentiate between chlorophyll and CDOM absorpti on (Carder et al. 1989). In South America, an extra setback comes from Saharan Dust, which is tr ansported year round from the Africa coast and across the Atlantic, and which affects aer osol composition in the atmosphere. Two algorithms were used in the Cariaco Ba sin to test retrie val of chlorophyll a and ag from satellite remote sens ing measurements duri ng this study. Specifi cally, the Lee at al. (1999) and Carder et al. (1999) algorithms were used. Errors in the algorithms varied with season. In September 2003, both algor ithms performed very similarly (<10% difference between them). Both algorithms were off by at least 50% in estimates of chlorophyll and CDOM absorption. In March 2004, the water leavi ng radiance measured by SeaWiFS, especially in the blue part of the visible spec trum, was very low. This was caused by the combined effects of high phyt oplankton absorption and errors in the atmospheric correction, most likely due to a hi gh concentration of Saharan Dust in the atmosphere. This lower-than-real water l eaving radiance was the main culprit of erroneous estimations in both chlorophyll and ag during March 2004. In areas of high chlorophyll concentration (>4 mg/m3), neither algorithm was ab le to derive values. In areas of intermediate chlorophyll concentration (between 2 – 3 mg/m3), the algorithm of Lee et al. (1999) attri buted most of the low remote se nsing reflectance to CDOM, while they were largely due to high concentrati ons of chlorophyll (C. Hu, pers. comm.). The Carder et al. (1999) algorithm provide d an estimate of chlorophyll and ag(440) only in areas where the chlorophyll a concentration was below 3 mg/m3.Chlorophyll and ag(440) were underestimated by 30-90% and overest imated by more than 50%, respectively. In order to improve the algorithm performances for this region, it will be necessary to improve atmospheric correctio ns, especially taking into account the influence of absorbing aerosols, such as Sahara n Dust. It will be necessary also to adjust certain parameters, such as S, the ag(440) slope, to match local values. Changes will most likely also have to be done taking into consid eration the season, and it is possible one single algorithm may not suffice (Gilbes et al., in preparatio n). Unfortunately, there is still no single algorithm that can su ccessfully estimate chlorophyll and CDOM concentrations in coastal areas. One possibility, for chlorophyll retrieval in future applications, can be to use the chlorophyll fluorescence product availabl e from sensors like MODIS (Moderate Resolution Imaging Spectroradiometer) or MERIS. MODIS bands 13, 14 and 15 (662 – 672 nm., 673 – 683 nm. and 743 – 753 nm.) pr ovide a measure of solar-stimulated fluorescence. MODIS has been actively taki ng measurements sin ce 1999 and 2002 on board of the Terra and Aqua sa tellites, respectively, and da ta processing and validation has improved significantly ove r the past 3 year s. Figure 2.17 shows the relationship between fluorescence determined with MODIS (image from March 16, 2004) and chlorophyll, measured during March 2004, fo r 26 stations. The linear relationship (r2 = 0.6266) between variables is encouraging, but fu rther validation has to be done in order to determine the accuracy a nd viability of chlorophyll deri vation from remote sensing images using this method.

PAGE 80

69 Figure 2.17: Relationship between chlorophyll measured in situ during March 2004 at 26 stations, and MODIS chlorophyll fluorescence for corresponding locations (image from March 16, 2004)

PAGE 81

70 5. Conclusions The two main contributors of CDOM a nd terrigenous particles to the Cariaco Basin were the Never and Un are rivers. Their influence was most pronounced along a narrow (30-40 km) band along the southern coas t of the Cariaco Basin. There was only a weak relationship between CDOM and salinity when all stations were considered, but correlation improved near the mout hs of rivers.The slope of ag varied from high values (0.025 nm-1) in areas depleted of CDOM away fr om the coast, to low values (0.011 nm-1) in areas influenced by rivers. The changes obs erved in S were likely due to different sources of CDOM. The range of S observed in Cariaco in September 2003 and March 2004 was similar to that reported by other authors, both for coastal and open waters (Bricaud et al., 1981; Nelson and Guarda, 1995; Carder et al., 1999; Cannizzaro, 2004). Near the coast, over 90% of the light absorption was caused by detrital and dissolved matter, particular ly during September 2003. In so me stations, over 80% of absorption was caused by CDOM. As distance from the coast increased, CDM absorption decreased rapidly. The contribution of suspe nded and particulate material from local rivers to the Cariaco Basin also lead to ch anges in the color of surface waters. Some coastal stations were dominated by a refl ectance in the green; waters 30 Km farther offshore showed high Rrs in the blue. The Unare River discharged into the Ca riaco Basin more suspended particulate matter than the Never River, as seen th rough optical measurements. Light attenuation was higher adjacent to the rivers, as compared to offshore waters in September 2003. At 15 m, beam attenuation 20 km away from the Unare’s mouth was almost half (0.4 m-1), as compared to beam attenuation 10 km from the river’s mouth (0.6 m-1), indicating that a considerable portion of the suspended material that caused light atte nuation precipitated within that distance. Bottom nepheloid la yers, visible throughout the Unare Platform, were strongest near the point of discharge. The radius of influence of the Unare’s bottom nepheloid layer was approximately 40 km from the mouth of the river, while the Never’s was close to 23 km. The Manzanares River di d not show any visible suspended matter or high concentrations of CDOM in surface wate rs. However, internal nepheloid layers (INL) were observed near the Manzanares River, over a submarine canyon, both in September 2003 and March 2004. This may be an important mechanism of sediment transport off the shelves and into the Eastern Deep. Local rivers are the probable source of the terrigenous material to the Eastern Deep. Though the Unare and Never were the most obvious sources of terrigenous material in September 2003, contributions fr om the Tuy River should not be ignored. Lee et al. (1999) and Carder et al. ( 1999) algorithms were used for estimating ag and chlorophyll a concentration from SeaWiFS imag es from September 2003 and March 2004. Rrs measurements from the satellite were in good agreement (r2 >0.9) with Rrs measured in situ during the September 2003 cruise. During March 2004, water leaving radiances measured from the sa tellite were lower than real in situ measurements, especially in the blue part of the visible spectrum. This was most likely due to the

PAGE 82

71 combined effect of high absorption by chlorophy ll in the blue part of the spectrum and a problem in the atmospheric correction, caused by absorbing aerosols in the atmosphere, such as Saharan Dust. In September 2003, both algorithms performed similarly. ag(440) and chlorophyll concentration were generally underestimated by at least 50%. In March 2004, neither algorithm was able to derive ag(440) or chlorophyll conc entration values in highly (> 4 mg/m3) productive areas. In areas where chlorophyll concentrations and ag(440) were calculated, using the Carder et al. (1999) algorithm (chlorophyll a concentrations of less than 3 mg/m3), they were usually underestimated by 30-90% and overestimated by more than 50%, respecti vely. Most of the problems retrieving chlorophyll and ag values during March can be attributed to the low water leaving radiance, and hence Rrs.

PAGE 83

72 GENERAL CONCLUSIONS The temporal variability of physi cal and biogeochemical oceanographic parameters of the Cariaco Basin, located along the southern margin of the Caribbean Sea, was studied using data collected during two cr uises to the eastern half of the Basin in September 2003 and March 2004. Water excha nge between the Cariaco basin and the Caribbean Sea was restricted to the upper ~150 m. In March 2004, water from outside the Cariaco basin was observed entering the basin and upwelling near the coast. This water brought nutrients and DOC into the basin. A ma rked seasonal difference in both surface temperature and salinity distribution wa s found between Sept ember 2003 and March 2004. During September 2003, surface temperatures in the basin ranged from ~26 to ~28 oC, and salinity from 36.4 to ~36.8. There wa s a difference in temperature and salinity between the Cariaco Basin and the open Caribb ean in waters below 200 m. Inside the basin temperatures and salinities, averag ed between 200 and 400 m, were higher by ~4oC and by ~0.5, respectively. Deep (1200m) water temperature agreed with the predictions of Scranton et al. (1987). Salinity below 1200 m was higher than predicted by Scranton et al. (1987), indicating an input of saltier water to the deep basin. The influence of local rivers on the basin, as contributors of less saline water and dissolved and detrital matter, was seen in Septembe r 2003 near the coast. No major intrusion of low salinity waters was observed fr om the north during either the rainy or dry season, indicating that the Amaz on and Orinoco rivers do not have major direct influence on the Cariaco Basin. The two main contributor s of CDOM and freshwater to the Cariaco Basin were the Never and Un are rivers, located on the sout hern coast of the Cariaco Basin. The suspended and dissolved colored matter brought by the rivers affected the optical properties of the water, but were conf ined to the southern margin of the basin. Near the coast, over 90% of the absorption was caused by detrital a nd dissolved matter, as measured during September 2003. The low sa linity plumes were dispersed towards the NE, reaching the CARIACO time-series station (10o30’ N, 64o40’W). Local rivers are responsible for the seasonal salinity vari ation observed in surf ace waters at the CARIACO time-series station. Local rivers are the probable source of the terrigenous material to the Eastern Deep. Though the Unare and Never were the most obvious sources of terrigenous material, contributions from the Tuy and the Manzanares River should not be ignored. There is a deep CDOM pool below the oxi c-anoxic interface (~200 m) inside the Cariaco Basin. Below 300 m, mo st, if not all, of the abso rption measured in September 2003 and March 2004 was caused by CDOM. This supports the idea that CDOM accumulates below the interface due to the lack of mixing. The presence of lightabsorbing particles was also noted at the oxi c-anoxic interface, most likely bacteria and viruses.

PAGE 84

73 REFERENCES Alvarez-Salgado, X.A., M.D. Doval and F.F. Prez. 1999. Dissolved organic matter in shelf waters off the Ra de Vigo (NW Iberian upwe lling system). J. of Mar. Chem., 18: 383-394. Astor, Y., F. Muller-Karger and M. Scranton. 2001. Seasonal and Interannual Variation in the Hydrography of the Cariaco Basin: Implications for Basin Ventilation. Cont. Shelf Res. 23:1, 125-144 Astor, Y., J. Meri, F. Muller Karger. 1998. Va riabilidad estacional hidrogrfica en la Fosa de Cariaco. Memoria de la Sociedad de Cien cias Naturales La Salle, 63 (149): 61-72. Baker, K. and R. C. Smith. 1982. Bio-optical classification and model of natural waters. Limnol. Oceanogr., 27: 500509. Bidigare, R. R., M. E. Ondrusek, and J. M. Brooks. 1993. Influence of Orinoco River outflow on distributions of algal pigments in the Caribbean Sea, J. Geophys. Res., 98, 2259–2269. Black, D. E.; L. C. Peterson; J. T. Overpeck; A. Kaplan; M. N. Evans and M. Kashgarian. 1999. Eight centuries of North Atlantic ocean-atm osphere variability. Science, 286, 1709-1713. Blough, N.V. and R. Del Vecchio.2002. Chromopho ric DOM in the Coastal Environment. In: Biogeochemistry of Marine Dissolved Organic Matter. D.A. Hansell and C.A. Carlson eds. Academic Press. 774 p. Blough, N.V., O.C. Zafiriou, and J. Bonilla 1993. Optical Absorption Spectra of Waters From the Orinoco River Outflow: Terrestrial Input of Colored Organic Matter to the Caribbean. J. Geophys. Res. 98: 2271-2278 Boss E. and J. R. V. Zaneveld, 2003. The eff ect of bottom substrate on inherent optical properties; evidence of biogeochemical processes. Limnol. Oceanogr., 48, 346-354. Boss E., M. S. Twardowski and S. Herring, 2001. The shape of the particulate beam attenuation spectrum and its relation to the size distributi on of oceanic particles. Applied Optics,40, 4885-4893. Boss, E., S.W. Pegau, R. Zaneveld, and A. H. Ba rnard. 2001. Spatial and temporal variability of absorption by dissolved material at a continental shelf. J. Geophys. Res. 106: 9499-9507 Boss, E., W. S. Pegau,W. D. Gardner, J. R. V. Zaneveld, A. H. Barnard., M. S. Twardowski, G. C. Chang and T. D. Dickey, 2001. The spectral particulate attenuation and particle size distribution in the bottom boundary layer of a continental shelf. J. Geophys. Res., 9509-9516. Bricaud, A., A. Morel, and L. Prieur 1981. Ab sorption by Dissolved Organic Matter of the Sea (Yellow Substance) in the UV and Visible Domains. Limnol. Oceanogr. 26(1): 43-53. Bricaud, A., C. Roesler, J. Ishizaka, and J. Pa rslow. 2002. Bio-optical studies during the JGOFSEquatorial Pacific program: A contribution to the knowledge of the Equatorial system. Deep Sea Res. 49: 2583-2599. Cannizzaro, J. P. 2004. Det ection and Quantification of Karenia brevis Blooms on the West Florida Shelf from Remotely Sensed Ocean Colo r Imagery. Master Thesis, University of South Florida, Florida, USA. Carder, K. L.; R.F. Chen; J. P. Cannizzaro; J. W. Campbell and B. G. Mitchell. 2004. Performance of the MODIS semi-analytical ocean color algorithm for chlorophylla Adv. in Space Res. 33: 1152-1159. Carder, K.; F.R. Chen; Z. P. Lee; S. K. Hawes and D. Kamykowski. 1999. Semianalytic Moderate-Resolution Imaging Spectro meter algorithms for chlorophyll a and absorption with

PAGE 85

74bio-optical domains based on nitrate-depletion temperatures. J. of Geophys. Res., 104 (C3): 5403-5421. Carder, K.L., R.G. Steward, G.R. Harve and R. B. Ortner. 1989. Marine humic and fulvic acids: their effects on remote sensing of ocean chlorophyll. Limnol. Oceanogr. 34, 68-81. Carlson, T. N. and J. M. Prospero. 1972. The Large-Scale Movement of Saharan Air Outbreaks over the Northern Equatorial Atlantic. J. appl. Met. 11: 283–297. Cauwet, G. 2002. DOM in the Coastal Zone. In: Biogeochemistry of Marine Dissolved Organic Matter. D.A. Hansell and C.A. Carlson eds. Academic Press. 774 p. Chang, G.; T. D. Dickey; C. D. Mobley, E. Boss and W. S. Pegau. 2003. Toward closure of upwelling radiance in coastal waters. Applied Optics, 42(9): 1574-1582. Chen, R. F., B. Fry, C. S. Hopkinson, D. J. Repeta, E. T. Peltzer. 1996. Dissolved organic carbon on Georges Bank. Cont. Shelf Res., 16: 409-420. Chen, R.F., and J.L. Bada. 1992. The fluorescence of dissolved organic matter in seawater. Mar. Chem. 37, 191–221. Chomko, R.M. and H. R. Gordon. 2001. Atmospheri c correction of ocean color imagery: test of the spectral optimization algorithm with the S ea-viewing Wide Field-of-View Sensor. Appl. Optics, 40 (18): 2973-2984. Claustre, H.; F. Fell; K. Oubelkheir; L. Prie ur; A. Sciandra; B. Gentili and M. Babin. 2000. Continuous monitoring of surface optical properties across a geostrophic front: Biogeochemical inferences. Lim nol. Oceanogr., 45(2): 309–321. Coble, P. G., C. E. Del Castillo, and B. Avril. 1998. Distribution of DOM in the Arabian Sea During the SW Monsoon. Deep Sea Research II 45, 2195-2223 Coble, P.G. 1996. Characterization of marine a nd terrestrial DOM in seawater using excitationemission matrix spectroscopy. Mar. Chem. 51(4): 325-346. Curran, K. J., P. S. Hill, and T. G. Milligan. 2003. Time variation of floc properties in a settling column. J. of Sea Res., 49: 1-9. Del Castillo, C. 1998. Optical characteristics of the colored dissolved organi c matter in the Easter Caribbean, West Florida Shelf and the Arab ian Sea: relationship between chemical characteristics and optical response. Ph.D. thesis, University of South Florida. 131p. Del Castillo, C.E., Coble P.G., Morel J.M., Lopez J. M., and Corredor J.E. 1999. Analysis of the optical properties of the Orinoco River Plume by absorption and fluorescence spectroscopy. Mar. Chem. 66: 35-51. Del Vecchio, R. and N.V. Blough. 2002. Chrom ophoric DOM in the Coastal Environment. In: Biogeochemistry of Marine Dissolved Organic Matter. D.A. Hansell and C.A. Carlson eds. Academic Press. 774 p. Deuser, W. G. 1973. Cariaco Trench: Oxidation of organic matter and residence time of anoxic water. Nature, 242, 601-603. Diehl, P. and H. Haardt. 1980. Measurement of the spectral attenuation to support biological research in a ‘plankton tube’ expe riment. Oceanol. Acta 3, 89–96. Fairbanks, R.G. 1989. A 17,000 year glacio-eustatic sea level record: Influence of glacial melting rates on the Younger Dryas event and deep-o cean circulation. Nature 342, 637-642. Farmer, C.T.; C.A. Moore; R.G. Zika and R.J. Sikorski. 1993. Effects of low and high Orinoco River Flow on the Underwater Light Field of th e Eastern Caribbean Basin. J. Geophys. Res., 98 (C2): 2279-2288. Ferrari, G.M. 2000. The relationship between chromophoric dissolved organic matter and dissolved organic carbon in the European Atlantic coastal area and in the West Mediterranean Sea (Gulf of Lions). Mar. Chem. 70: 339-357. Ferrari, G.M., M.D. Dowell, S. Grossi, and C. Targa. 1996. Relationship between the optical properties of chromophoric dissolved organic ma tter and total concentration of dissolved organic carbon in the southern Baltic Sea Region. Mar. Chem. 55: 299-316

PAGE 86

75Froelich Jr., P. N. and D. K. Atwood, 1978. Dissolved silicate and salinity structure of the upper waters of the Venezuela Basin, Caribbean Sea, In: Progress in Marine Research in the Caribbean and Adjacent Regions (ed. H. B. St ewart, Jr.), 1976. FAO Fisheries Report No. 200 Supplement, p. 299-314. Froelich Jr., P. N., D. K. Atwood and G. S. Gie se. 1978. Influence of Amazon River discharge on surface salinity and dissolved s ilicate concentration in the Ca ribbean Sea. Deep Sea Res., 25:735-744. Goi, M.A.; E. Tappa; H. Aceves; B. BenitezNelson; F. Muller-Karger; L. Lorenzoni; M. McIntyre; Y. Astor, R. Varela and R. Thunell. 2004. Seasonal to Climate-Scale Variability in the Vertical Flux of Terrigenous Organic Matter in the Cariaco Basin. Presented at the 2004 AGU meeting, San Francisco. Goi, M.A.; H. L. Aceves; R. C. Thunell; E. Tappa ; Y. Astor; R. Varela and F. Muller-Karger. 2003. Biogenic fluxes in the Cariaco Basin: A combined study of sinking particulates and underlying sediments. Deep-Sea Res. I 50: 781-807. Gordon, A.L. 1967. Circulation of the Caribbean Sea. J. of Geophys. Res., 72, 6207-6223. Green, R. E.; H. M. Sosik and R. J. Olson, 2003. Contributions of phytoplankton and other particles to inherent optical properties in New England continental shelf waters. Limnol. Oceanogr., 48 (6), 2377-2391. Green, S.A. and Blough, N.V., 1994. Optical absorption and fluorescence properties of chromophoric dissolved organic matter in na tural waters. Limnol. Oceanogr. 38, 1903–1916. Halliwell, Jr., G.R., R. H. Weisberg, and D.A. Ma yer, 2003. A synthetic float analysis of upperlimb meridional overturning circul ation interior ocean pathways in the tropical/subtropical Atlantic. Interhemispheric Water Exchange in th e Atlantic Ocean, G. Goni and P. MalanotteRizzoli, ed., Elsevier Publishing Company. Hansell, D.A. and T. Y. Waterhouse. 1997. C ontrol of the distribution of organic carbon and nitrogen in the Eastern Pacific Ocean. Deep-Sea Res. I, 44:843-857. Haug, G., Pedersen, T., Sigman, D., Calver t, S, Nielsen, B., Peterson, L., 1998. Glacial/interglacial: variations in production a nd nitrogen fixation in the Cariaco Basin during the last 580 kyr. Paleoceanography 13, 427–432. Hellweger, F. L. and A. L. Gordon, 2002. Traci ng Amazon River Water into the Caribbean Sea. J. Mar. Res., 60(4), 537-549. Herrera, L.E. and G. Febres-Ortega. 1975. Procesos de surgencia y de renovacin de aguas en la Fosa de Cariaco, Mar Caribe. Bol. Inst. Oceanogrfico Univ. de Oriente 14, 31–44. Herrera, L.E., G. Febres-Ortega and J. M. Andrs. 1980. Distribucin de las masas de agua y sus vinculaciones dinmicas en el sector Centro-Occidental Venezolano, Mar Caribe. Bol. Inst. Oceanogrfico Univ. de Oriente, 19, 93–118. Hickey, B.; E. Baker and N. Kachel. 1986. Suspen ded particle movement in and around Quinault Submarine Canyon. Mar. Geol. 71, 35–83. Hill, P. S., T. G. Milligan, and W. R. Geyer. 200 0. Controls on effective settling velocity in the Eel River flood plume. Cont. Shelf Res., 20: 2095-2111. Hill, P. S.; G. Voulgaris and J. H. Trowbridge. 2001. Controls on floc size in a continental shelf bottom boundary layer. J. of Geophys. Res., 106: 9543-9549. Hoge, F. E., and P. E. Lyon. 1996. Satellite retr ieval of inherent optical properties by linear matrix inversion of oceanic radiance mode ls: An analysis of model and radiance measurement errors, J. Geophys. Res., 101: 16,631-16,648. Hu, C., E. T. Montgomery, R. W. Schmitt and F. E. Muller-K arger. The Amazon and Orinoco River plumes in the tropical Atlantic a nd Caribbean Sea: Observation from space and SPALACE floats. Submitted to Deep-Sea Research-II SeaWiFS special volume (in press).

PAGE 87

76Hu, C.; F. E. Mller-Karger; D. Biggs; K. L. Carder; B. Nababan; D. Nadeau and J. Vanderbloemen. 2003. Comparison of ship a nd satellite bio-optical measurements on the continental margin of the NE Gulf of Mexico. Int. J. of Rem. Sens., 24 (13): 2597-2612. Hughen, K. A.; J. T. Overpeck; L. C. Peterson a nd R. F. Anderson. 1996. The nature of varved sedimentation in the Cariaco Basin, Venezuel a, and its palaeoclimatic significance. In: Palaeoclimatology and Palaeoceanography from La minated Sediments. A. E. S. Kemp, ed. Geological Society Special Publication No. 116, pp. 171-183. Hughen, K.A., Overpeck, J.T., Peterson, L.C., Tr umbore, S., 1996b. Rapid climate changes in the tropical Atlantic region during the last deglaciation. Nature 380, 51}54. Karl, D.M. and G. A. Knauer. 1991.Microbial production and particle flux in the upper 350 m of the Black Sea. Deep Sea Res. Supl. 2, 38: S921-S942. Kinard, W.F., D.K. Artwood and G.C. Giese. 1974. Dissolved oxygen as evidence for 18oC Sargasso Sea Water in the Eastern Caribb ean Sea. Deep Sea Res., 21: 79-82. Kirk, J.T.O. 1994. Light and photos ynthesis in aquatic ecosystems. Cambridge University Press, 401p. Kishino, M.; M. Takahashi; N. Okami and S. Ichimura. 1985. Estimation of the spectral absorption coefficients of phytoplankton in the sea. Bull. of Mar. Sci., 37: 634-642. Lee, Z., Carder, K.L., Mobley, C.D., Steward, R.G., Patch, J.S., 1999. Hyperspectral remote sensing for shallow waters: 2. deriving bottom de pths and water properties by optimization. Applied Optics 38, 3831-3843. Lin, H. L., L. C. Peterson, J. T. Overpeck, S. E.Trumbore and D. W. Murray. 1997. Late Quaternary climatic change from d-18 O records of multiple species of planktonic foraminifera: High-resolution records from the anoxic Cariaco basin, Venezuela. Palaeoceanography, 12(3): 415-427. Liu, K.K; K. Iseki and S. Y. Chao. 2000. Co ntinental margin carbon fluxes. In: The Changing Ocean Carbon Cycle: A Midterm Synthesis of the Joint Global Ocean Flux Study. R. B. Hanson, H. W. Ducklow and J. G. Field, eds. Cambridge University Press. 528 p. McClain, E.; W. Pichel and C. Walt on. 1985. Comparative Pe rformance of AVHRRBased Multichannel Sea Surface Temperatur es. Journal of Geophysical Research (90): 11,587-11,601. McPhee-Shaw, E. E.; R.W. Sternberg; B. Mullenb ach and A. S. Ogston. 2004. Intermediate nepheloid layers and internal tide reflection on the Northern California Margin. Cont. Shelf Res. (in press) Morrison, J., and W. Nowlin. 1982. General di stribution of water masses within the eastern Caribbean Sea during the winter of 1972 and fall of 1973. J. of Geophys. Res., 87: 42074229. Morrison, J.M., and O.P. Smith. 1990. Geostrophi c transport variability along the Aves Ridge in the eastern Caribbean Sea during 1985-1986. J. of Geophys. Res., 95: 699-710. Mueller, J. L. and R.W. Austin (1995). Ocean Optics Protocols for SeaWiFS Validation, Revision 1, SeaWiFS Technical Report Series, Vol. 25, S. B. Hooker, E. R. Firestone and J. G. Acker (eds.),NASA Technical Memorandum 104566, Greenbelt, Maryland. Muller-Karger, F. E. R. Varela, R. Thunell, M. Scranton, R. Bohrer, G. Taylor, J. Capelo, Y. Astor, E. Tappa, T. Y. Ho, and J. J. Wals h. 2001. Annual Cycle of Primary Production in the Cariaco Basin: Response to upwelling and implica tions for vertical export. J. Geophys. Res. 106 (C3): 4527-4542 Mller-Karger, F. E., and R. Aparicio. 1994. Mesoscale Processes Affecting Phytoplankton Abundance in the Southern Caribbean Sea. Cont. Shelf Res. (14:2/3). 199-221. Mller-Karger, F. E., C. R. McClain, T. R. Fish er, W. E. Esaias, and R. Varela. 1989. Pigment distribution in the Caribbean Sea: Observation from space. Prog. Oceanogr. 23: 23-64.

PAGE 88

77Muller-Karger, F., R. Varela, R. Thunell, M. Scrant on, R. Bohrer, G. Taylor, J. Capelo, Y. Astor, E. Tappa, T.Y. Ho, M. Iabichella, J. J. Wals h, and J. R. Diaz. 2000. Sediment record linked to surface processes in the Cariaco Basin. EO S. AGU Transactions. American Geophysical Union, 81 (45): 529, 534-535 Muller-Karger, F.E., and R.J. Varela. 1989/1990. Influjo del Ro Orinoco en el Mar Caribe: observaciones con el CZCS desde el espacio. Me m. Soc. Cienc. Nat. La Salle 49/50: 361390. Nelson, J.R. and S. Guarda. 1995. Particulate a nd dissolved spectral abso rption on the continental shelf of the southeastern United States. J. of Geophys. Res. 100, 8715-8732. Nelson, N.B., D.A. Siegel, and A.F. Michaels. 1998. Seasonal dynamics of colored dissolved material in the Sargasso Sea. Deep-Seas Res. I. 45:931-957 O’Reilly, J.E.; S. maritorena; B.G. Mitchell; D.A. Siegel; K. L. Carder; S.A. Garver; M. Kahru and C. McClain. 1998. Ocean color algorithms for SeaWiFS. J. of Geophys. Res. (103): 24937-24953. Okuda, T. 1978. Condiciones hidroqumicas de las a guas superiores de la Fosa de Cariaco y reas adyacentes. In: Progress in Marine Research in the Caribbean and Adjacent Regions (ed. H. B. Stewart, Jr.), 1976. FAO Fisheries Report No. 200 Supplement, p. 349-362. Pak, H.; J.R.V. Zaneveld and R.W. Spinrad. 1984. Vertical distribution of suspended particulate matter in the Zaire River, Estuary and Plume. Neth. J. of Sea Res., 17 (2-4):412-425. Pak, H.; L.A. Codispoti and J. R. V. Zaneveld 1980. On the intermediate particle maxima associated with oxygen-poor water pff west S outh America. Deep Sea Res., 27A: 783-797. Pegau, W. S.; D. Gray and J. R. V. Zaneveld 1997. Absorption and attenuation of visible and near-infrared light in water: dependance on temperature and salinity. Applied Optics 36: 6035-6046. Peterson, L. C., G. H. Haug, R.W. Murray, K. M. Yanicik, J. W. King, T. J. Bralower, K. Kameo, S. D. Rutherford and R. B. Pearce. 2000. Late Quaternary stratigraphy and sedimentation at site 1002, Cariaco Basin (Venezuela). In: Leck ie, R. M., Sigurdson, H., Acton, G. D. & Draper, G. (eds.), Proceedings of the Ocean Drilling Program, Scientific Results, 165: 85-99. Peterson, L.C., Overpeck, J.T., Kipp, N.G., Im brie, J., 1991. A high-resolution late quaternary upwelling record from the anoxic Cariaco Basi n, Venezuela. Paleoceanography 6, 99–119. Prospero, J., R.; Glaccum and R. Nees. 1981. Atmo spheric transport of soil dust from Africa to South America. Nature (289): 570 – 572. Puig, P. and Palanques, A. 1998. Nepheloid stru cture and hydrographic control on the Barcelona continental margin, northwestern Me diterranean. Mar. Geol. 149, 39–54. Puig, P., Ogston, A.S., Mullenbach, B.L., Nittrouer, C.A., Sternberg, R.W. 2003. Shelf-to-canyon sediment-transport processes on the Eel continenta l margin (northern California), Mar. Geol. 193, 129-149. Repeta, D.J. and D.J. Simpson. 1991. The distribution and recycling of chlorophyll, bacteriochlorophyll and carotenoids in the Black Sea. Deep Seas Res. Suppl. 2, 38: S969S984 Richards, F. A. 1975. The Cariaco Basin (Trench) Oceanogr. Mar. Biol. Ann. Rev., 13, 11 –67. Roesler, C. S. and M. J. Perry. 1995. In situ phytoplankton absorption, fluorescence emission, and particulate backscattering spectra determined from reflectance. J. of Geophys. Res., 100:13279-13294. Roesler, C. S., M. J. Perry, and K. L. Carder. 1989. Modeling in situ phytoplankton absorption from total absorption spectra. Limnol. Oceanogr. 34: 1512-1525. Scranton, M. I., Y. Astor, R. Bohrer, T.-Y. Ho, and F. E. Muller-Karger. 2001. Controls on temporal variability of the geochemistry of th e deep Cariaco Basin. Deep Sea Res. I, 48: 1605-1625.

PAGE 89

78Snyder, G. W. and B. Carson. 1986. Bottom a nd suspended particle sizes: implications for modern sediemtn transport in Quinault Submarine Canyon. Mar. Geol., 71:85-105. Taylor, G., M. Scranton, M. Iabichella, T.Y. Ho and R. Varela. 2001. Chemoautotrophy in the redox transition zone of the Cariaco Basin, a significant source of midwater organic carbon production. Limnol. Oceanogr., 46:148-163 Thunell, R., R. Varela, M. Llano, J. Collister, F. Muller-Karger, and R. Bohrer. 2000. Organic carbon flux in an anoxic water column: sediment trap results from the Cariaco Basin. Limnol. Oceanogr., 45: 300-308 Thunnell, R. 1997. Continental margin particle flux: Seasonal cycles and archives of global change. Oceanus. 40: 20-24 Vodacek, A., N.V. Blough, M.D. DeGramdpre, E. T. Peltzer, and R.K. Nelson. 1997. Seasonal variation of CDOM and DOC in the Middle Atlantic Bight: Terrestrial inputs and photooxidation. Limnol. Oceanog. 42: 674-686 Walsh, J.J., D.A. Dieterle, F.E. Muller-Karger, R. Bohrer, W.P. Bissett, R. Aparicio, R.J. Varela H.T. Hochman, C. Schiller R. Diaz, R. Thune ll, G.T. Taylor, M.I. Scranton, K.A. Fanning, and E.T. Pelzer. 1999. Simulation of carbon/ni trogen cycling during spring upwelling in the Cariaco Basin. J. Geophys. Res., 104 (C4):7807-7825. WetLabs AC-9 User’s Guide. Revision L, January 10, 2003. (D. Stahlke and H. Van Zee). 37p. Wust, G. 1964. Stratification and circulation in the Atillean-Caribbean Basins. Columbia University Press, Palisades, NY, 201 pp. Xu, J.P., M. Noble, S. L. Eittreim, L. K. Rosenfeld, F. B. Schwing and C. H. Pilskaln. 2002. Distribution and transport of suspended partic ulate matter in Monterey Canyon, California. Mar. Geol. 181: 215-234. Yarincik K. M., Murray R. W., and Peterson L. C. 2000. Climatically controlled eolian and hemipelagic deposition in the Cariaco Basin, Ve nezuela, over the past 578,000 years: Results from Al/Ti and K/Al. Paleoceanography, 15: 210-228.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200409Ka 4500
controlfield tag 001 001680969
005 20060215071039.0
006 m d s
007 cr mnu uuuuu
008 051206s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001152
035
(OCoLC)62469296
040
FHM
c FHM
049
FHMM
090
GC11.2 (Online)
1 100
Lorenzoni, Laura.
4 245
The influence of local rivers on the eastern cariaco basin, Venezuela
h [electronic resource] /
by Laura Lorenzoni.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 89 pages.
520
ABSTRACT: Two oceanographic cruises were conducted during September 2003 and March 2004 in the eastern half of the Cariaco Basin. Specific objectives were to examine the hydrography of the seasonal upwelling plume characteristic of this region, the spatial distribution of particles in the area, and to help determine the source and relative importance of in situ particle production vs. terrigenous particles delivered laterally from the coast.During September 2003, average surface salinities within the basin were higher (36.6) relative to Caribbean Sea waters outside the basin (35.6). Salinity patterns indicated that the Orinoco and Amazon River plumes did not enter or influenced the basin directly.The upwelling plume in March 2004 stimulated primary productivity. Beam attenuation and CDOM fluorescence profiles showed marked vertical structure in biomass of microbial populations, particularly near the oxic-anoxic interface typically located between about 250 and 300 m.There is an increasing difference in temperature and salinity between the Cariaco Basin and the adjacent Caribbean Sea below 200 m. Inside the Basin temperatures and salinities were higher by 4C and 0.5.The influence of local rivers on the Cariaco Basin was evident during September 2003. Low salinity plumes with high beam attenuation (1m1) lined the southern margin of the Basin. The primary rivers that affected the basin were the Unare and Never. Their sediment input affected the shelf near the river mouths, and a surrounding radius of up to 40 Km. Their low salinity plumes were carried northwestward toward the CARIACO time series station. In March 2004, there was minimal or no terrigenous input from local rivers. Near the Manzanares River, off the city of Cuman, and near Cubagua Island, located south of Margarita Island, attenuation due to suspended particles (0.09 m-1) was observed at depth (70-150 m) during both cruises (0.09-0.15 m-1).
590
Adviser: Dr. Frank Muller-Karger.
653
Bottom nepheloid layer.
Colored dissolved organic matter.
Subtropical underwater.
Remote sensing reflectance.
Anoxic basin.
0 690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
949
FTS
SFERS
ETD
GC11.2 (Online)
856
u http://digital.lib.usf.edu/?e14.1152