USF Libraries
USF Digital Collections

On the color of the Orinoco River plume

MISSING IMAGE

Material Information

Title:
On the color of the Orinoco River plume
Physical Description:
Book
Language:
English
Creator:
Odriozola, Ana L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Cdom
Tsm
Bio-optical properties
Chlorophyll
Ocean color
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: In situ measurements were used to study the bio-optical properties of marine waters within the Gulf of Paria (GOP, Venezuela) and in the Southeastern Caribbean Sea (SEC) as they are affected by the seasonal discharge of the Orinoco River plume. The main purpose of this study was to determine the impact of colored dissolved organic matter (CDOM) (also known as Gelbstoff), phytoplankton, and total suspended matter (TSM) in the color of the Orinoco River plume. This information is essential for regional ocean color algorithms development. Salinity and silica values indicate that the GOP and SEC waters were under the influence of the Orinoco River plume during both seasons. This riverine influence resulted in high values of Gelbstoff absorption, &#945g (&#955), which contributed to up to 90% of the total absorption at 440 nm in both the GOP and SEC regardless of the season.Phytoplankton absorption contributions were normally around 5%, but during the dry season these values reached 20% in the SEC. Ratios of &#945g(440) to &#945ph(440) were extremely large, with most of the values ranging from 10 to 50. Due to the strong absorption by Gelbstoff, light at the blue wavelengths (412 nm, 440 nm and 490 nm) was attenuated to 1% of the subsurface irradiance in the first 5 m of the water column within the GOP, and in the first 10 m of the water column in the SEC. Furthermore, the absorption by Gelbstoff significantly decreased the water leaving radiance (Lw(&#955)) in the blue wavelengths along the Orinoco River plume.As &#945g(&#955) relatively decreased from the GOP to the SEC X≈1.6 m-1 and X≈ 0.9 m-1, respectively), a shift in the maximum peak of Rrs(&#955) spectra (Rrsmax(&#955)), towards shorter wavelengths (from ~ 580 nm to ~500 nm) was observed. Similar to Gelbstoff, concentrations of TSM normally decreased from the stations near the Delta to the stations in the SEC. The impact of TSM on the color of the Orinoco plume was represented by a reduction in the magnitude of Rrsmax(&#955) of ~50% going from the waters near the Orinoco delta to the SEC, indistinctively of the season.
Thesis:
Thesis (M.S.)--University of South Florida, 2004.
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 Ana L. Odriozola.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 198 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:
oclc - 57677678
notis - AJU6882
usfldc doi - E14-SFE0000565
usfldc handle - e14.565
System ID:
SFS0025256:00001


This item is only available as the following downloads:


Full Text

PAGE 1

On the Color of the Orinoco River Plume by Ana L. Odriozola 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. Mller-Karger, Ph.D. Kendall Carder, Ph.D. Chuamin Hu, Ph.D. Date of Approval: November 18, 2004 Keywords: Ocean Color, Bio-optical Properties, Chlorophyll, CDOM, TSM Copyright 2004 Ana L. Odriozola

PAGE 2

Dedication This accomplishment is dedicated to my husba nd, my daughters, and my parents, for their patience and love

PAGE 3

Acknowledgements Many thanks to my major professor Dr Frank Mller-Karger for giving me the opportunity to come as a student to the Univer sity of South Florida (USF) and for all his support and encouragement throughout my time in the Master’s program. I also thank professor Kendall Carder and Dr. Chuanm in Hu for their guidance and support throughout this study. I wish to acknowledge the scientists, tec hnicians, and students at the Estacin de Investigaciones Marinas de Margarita (EDIMA R) Fundacin La Salle, Venezuela, at the Institute for Marine Remote Sensing (IMA RS) USF, and at the College of Marine Science (CMS) USF, who helped in the coll ection and processing of the data used in this study, with special thanks to Professo rs Ramn Varela (EDIMAR) and Yrene Astor (EDIMAR), Glenda Arias (EDIMAR), Natasha Rondn (EDIMAR), and John Akl (IMARS). I thank, Brock Murch (IMARS) for hi s help in revising the manuscript, and my friends Laura Lorenzoni (IMARS) and Damaris Torres-Pulliza (IMARS) for their support. I specially thank my family for their pa tience and encouragement. This work was supported by NASA’s SIMBIOS program, contract # NAS5-31716 and by NASA’s Earth Science Fellowship, grant # NGT5-30354.

PAGE 4

i Table of Contents List of Tables iii List of Figures v Abstract x Introduction 1 Objectives 3 Study Site 4 Orinoco River 4 Orinoco Delta 6 Gulf of Paria 6 The Orinoco River Plume in the Southeastern Caribbean Sea 7 Data Collection and Processing 10 Scientific Expeditions to th e Orinoco River and Plume 10 CTD casts and Water Sample Analyses 10 Absorption coefficients 11 Surface Reflectance and Subsurface Radiometric Measurements 12 Flow-through Measurements 14 Phytoplankton Taxonomy 14 Ocean Color and Bio-optical Algorithms 14 SeaWiFS OC4v4 Algorithm 15 Bio-optical Inversion Model 16 Results and Discussion 25 Temperature and Salinity Distribution 25 Nutrient Concentrations 35 Phytoplankton Pigments 46

PAGE 5

ii Total Suspended Matter (TSM) 54 Phytoplankton Distribution 61 Absorption Coefficients 61 Particle absorption ( ap) 62 CDOM ( Gelbstoff) Absorption ( ag) 82 Continuous, Along-Track (flowthrough) Measurements 93 Light Field 103 Subsurface Measurements 103 Surface Reflectance Measurements 108 GOP-Serpent’s Mouth 108 GOP-Dragon’s Mouth 108 Southeastern Caribbean (SEC) 109 Ocean Color Algorithms 118 Deriving chlorophylla from an empirical algorithm 118 Using a Rrs inversion model 126 Rrs in situ measurements vs. Rrs modeled 130 Summary and Conclusion 132 References 136 Appendices 148

PAGE 6

iii List of Tables Table 1. Orinoco-SIMBIOS Cruises 21 Table 2. Location and date for each station 22 Table 3. Oceanographic and bio-optical measurements collected during each cruise 24 Table 4. Temperature and Salinity surface va lues (~ 1 m) for all the stations 33 Table 5. Mean Values of Temperature and Salinity 34 Table 6. Surface (~ 1m) concentrati on of nutrients for all stations 40 Table 7. Mean Values of Nutrient concentrations ( M) 42 Table 8. Surface (~ 1m) concentration of Chlorophylla and Total Suspended Matter (TSM) for all the stations 51 Table 9. Mean Values of Chlorophylla and Total Suspended solids (TSM) 52 Table 10. T-test results showing the m ean seasonal and spatial variability of Chlorophylla 53 Table 11. Phytoplankton ( aph) and detritus ( ad) absorption coefficients in the blue bands 412 nm, 440 nm, and 492 nm 75 Table 12. Mean values of phytoplankt on and detritus absorption at 440 nm 77 Table 13. Specific absorption coefficient of phytoplankton at 440 nm ( aph*(440)) 79 Table 14. Gelbstoff absorption at 440 nm ( ag(440)), at 300 nm ( ag(300)), and the spectral slope (S) 91 Table 15. Minimum, maximum, and mean values of surface total absorption ( a ), scattering ( b ) and beam attenuation ( c ) coefficients 99

PAGE 7

iv Table 16. Gelbstoff detritus, and phytoplankton contri butions to total absorption at 440 nm 102 Table 17. Diffuse attenuation coefficient (Kd) and depth of maximum penetration ( zmax) per station 106 Table 18. Ratio between the absorption coefficients of Gelbstoff and phytoplankton at 440 nm 116 Table 19. T-test results showing the mean seasonal variability of ag(440), ad(440), and aph(440) in the GOP and SEC 116 Table 20. T-test results showing the mean spatial variability of ag(440), ad(440), and aph(440) 117 Table 21. T-test results showi ng the mean seasonal and spatial variability of Total su spended solids (TSM) 117 Table 22. T-test results showi ng the mean seasonal and spatial variability of the specific atte nuation coefficient at 660 nm ( c (660)) 117 Table 23. Chlorophylla concentrations from in situ measurements [Chla ]mea, and derived from the OC4v4 algorithm [Chla ]oc4v4, and percentage error (%E) 123 Table 24. Results from Rrs inversion model 128 Appendix A. Statistics Summary for surface temperature and salinity values 149 Appendix B. Statistics Summary for surf ace chlorophyll-a (Chl-a) and total suspended matter (TSM) concentrations 155 Appendix C. Statistics Summary fo r surface nutrient concentrations 161 Appendix D. Statistics Summary for surface phytoplankton ( aph( )) and detritus ( ad( )) absorption coefficients in the blue wavelengths 167 Appendix E. Statistics Summary for surface Gelbstoff absorption coefficient at 440 nm ( ag(440)) and spectral slope 173 Appendix F. Statistics Summary for surface contributions (%) of ag(440), ad(440), and aph(440) to the total abso rption coefficient ( a (440)) at 440 nm 179

PAGE 8

v List of Figures Figure 1. Study Region, showi ng the Orinoco River, Orin oco River Delta (ORD), Gulf of Paria (GOP) and South Eastern Caribbean 8 Figure 2. Composite Orinoc o River runoff at Puente Angostura, showing maxima, minima, and mean monthly values 9 Figure 3. Location of the sampli ng stations for each cruise 20 Figure 4. Hydrographic profiles of Temper ature (T) (dotted lin e), Salinity (S) (solid line), Chl-fluorescence (Chl-fl) (dashed line) and total attenuation coefficient at 660 nm (c(660 )) (Dash-dot line) for stations near Serpent’s M outh at about 10.00 N 61.9 W 27 Figure 5. Vertical crosssections showing the me ridional distribution of Temperature (C) for each cruise 31 Figure 6. Vertical crosssections showing the meri dional distribution of Salinity for each cruise 32 Figure 7. Phosphate (PO4), nitrite (NO2), nitrate (NO3), and ammonia (NH4) concentrations for each station. 39 Figure 8. Mean values of nutrient concentrations for the GOP (A) and SEC (B) 42 Figure 9. Dissolved silica con centration at each station 43 Figure 10. Relationship between silica con centrations and salinity during each cruise 44 Figure 11. Relationship between dissolved silica, Si(OH)4, concentration and salinity in the GOP and SEC. 45 Figure 12. Correlation analysis between Chlorophylla and chlorophyllfluorescence in the GOP ( ) and SEC ( ) for each cruise 48

PAGE 9

vi Figure 13. Vertical crosssections showing the meri dional distribution of chlorophyll fluorescence (relati ve values) for each cruise 49 Figure 14. Chlorophylla (mg l-1) concentration at each station 50 Figure 15. Frequency dist ribution of Chlorophylla concentrations 52 Figure 16. Relationship between total suspende d matter (TSM) and chlorophyll concentrations in the GOP ( ) and SEC ( ) for each cruise 57 Figure 17. Vertical cross-se ctions showing the meridiona l distribution of the beam attenuation coefficient at 660 nm (m-1) for each cruise 58 Figure 18. Relationship between Chlorophyll-fluorescence and c (660) in the GOP ( ) and SEC ( ) for each cruise 59 Figure 19. Relationship between TSM and c (660) for each cruise 60 Figure 20. Spatial variability of ap ( ), aph ( ), and ad( ), for some of the tations during SIM1 65 Figure 21. Spatial variability of ap( ), aph( ), and ad ( ), for some of the stations during SIM2 66 Figure 22. Spatial variability of ap ( ), aph ( ), and ad ( ), for some of the stations during SIM3 67 Figure 23. Spatial variability of ap ( ), aph ( ), and ad ( ), for some of the stations during SIM4 68 Figure 24. Spatial variability of ap ( ), aph ( ), and ad ( ), for some of the stations during SIM5 69 Figure 25. Spatial variability of ap ( ), aph ( ), and ad ( ), for some of the stations during SIM6 70 Figure 26. Detritus (blue) and phytoplankt on (green) absorption contributions to particle absorption at 440 nm 71 Figure 27. Particle ap ( ), phytoplankton aph ( ), and detritus ad () absorption coefficients at 412 nm 72

PAGE 10

vii Figure 28. Particle ap ( ), phytoplankton aph ( ), and detritus ad () absorption coefficients at 440 nm 73 Figure 29. Particle ap ( ), phytoplankton aph ( ), and detritus ad () absorption coefficients at 492 nm 74 Figure 30. Phytoplankton specifi c absorption coefficient ( aph *) spectra 78 Figure 31. Relationship between the specific absorption coefficient ( aph *) at 440 nm and chlorophyll-a during the dry (A) and wet (B) season 80 Figure 32. Relationship between phytoplankton absorption ( aph) at 440 nm and chlorophyll-a concentration 81 Figure 33. Spatial variability of ag( ) for some of the st ations during SIM1 84 Figure 34. Spatial variability of ag( ) for some of the st ations during SIM2 85 Figure 35. Spatial variability of ag( ) for some of the st ations during SIM3 86 Figure 36. Spatial variability of ag( ) for some of the st ations during SIM4 87 Figure 37. Spatial variability of ag( ) for some of the st ations during SIM5 88 Figure 38. Spatial variability of ag( ) for some of the st ations during SIM6 89 Figure 39. Relationship between Gelbstoff absorption at 440 nm and Salinity during dry (A) and wet (B) seasons 90 Figure 40. Relationship between Gelbstoff absorption spectral slope and Salinity during dry (A) and wet (B) seasons 92 Figure 41. Along-track measuremen ts of surface Temperature (C), Salinity (psu), CDOM-fluorescence (rela tive values), and Chlorophyllfluorescence (relative values) starting from the GOP near Serpent’s Mouth northward towards the SEC 95 Figure 42. Along-track measurem ents of beam attenuation ( c ), total absorption ( a ), and total scattering ( b ) coefficients, starting from the GOP near Serpent’s Mouth northward towards the SEC 96 Figure 43. Relationship between CDOM and Chlorophyll fluorescence with salinity from alongtrack measurements 97

PAGE 11

viii Figure 44. Relationship between CDOM and Chlorophyll fluorescence with total absorption coefficient at 440 nm from along-track measurements 98 Figure 45. Spectral dependence of total absorption ( a ) and total scattering ( b ) coefficients 100 Figure 46. Gelbstoff (yellow), Detritus (blue) and phytoplankton (green) absorption contributions to total absorption at 440 nm 101 Figure 47. Mean spectral values of the diffuse attenuation coefficient (Kd) in the GOP (A) and SEC (B) 104 Figure 48. Mean values of the euphotic depth ( zmax) (1% of subsurface irradiance) in the GOP (A) and SEC (B) 105 Figure 49. Rrs( ) spectra for stations in the GOP near to Serpent’s Mouth 110 Figure 50. Rrs( ) spectra for stations in the GOP near to Dragon’s Mouth 111 Figure 51. Rrs( ) spectra for stations in the SEC 112 Figure 52. Monthly mean SeaWiFS images processed using the OC4v4 algorithm for the cruises carri ed out during the dry season. 120 Figure 53. Monthly mean SeaWiFS images processed using the OC4v4 algorithm for the cruises carri ed out during the wet season. 121 Figure 54. Chlorophylla concentration derived fr om the OC4v4 algorithm, [Chl-a]oc4v4 versus chlorophyll-a concentrations measured from in situ water samples [Chl-a]mea 122 Figure 55. Frequency distri bution of %E between in situ measurements and derived OC4v4 chlorophyll-a concentrations. 123 Figure 56. Rrs( ) band-ratios used by the OC4v4 algorithm versus in situ chlorophyll-a concentrations in the SEC during the dry season 124 Figure 57. Rrs( ) band-ratios used by the OC4v4 algorithm versus in situ chlorophyll-a concentrations in the SEC during the wet season 125 Figure 58. Local relationship between chlorophylla concentration and aph(440) 128

PAGE 12

ix Figure 59. Relationship between the back scattering coefficient at 660 nm, bb(660), derived from the model and in situ measurements of TSM, and chlorophylla concentrations 129 Figure 60. Comparison between Rrs( ) values uncorrected for sun glint, Rrs_uncorr), the Rrs( ) corrected, and the Rrs( ) derived from the inversion model (Rrs_corr and Rrs_mod, respectively) 131

PAGE 13

x On the Color of the Orinoco River Plume Ana L. Odriozola ABSTRACT In situ measurements were used to study the biooptical properties of marine waters within the Gulf of Paria (GOP, Venezuela) and in the Southeastern Caribbean Sea (SEC) as they are affected by the seasonal discha rge of the Orinoco River plume. The main purpose of this study was to determine the im pact of colored dissolved organic matter (CDOM) (also known as Gelbstoff ), phytoplankton, and total suspended matter (TSM) in the color of the Orinoco river plume. This information is essential for regional ocean color algorithms development. Salinity and silica values indicate that the GOP and SEC waters were under the influence of the Orinoco River plume duri ng both seasons. This riverine influence resulted in high values of Gelbstoff absorption, ag( ), which contributed to up to 90% of the total absorption at 440 nm in both the GOP and SEC regardless of the season. Phytoplankton absorption contributions were normally around 5%, but during the dry season these values reached 20% in the SEC. Ratios of ag(440) to aph(440) were extremely large, with most of the values ranging from 10 to 50. Due to the strong absorption by Gelbstoff light at the blue wavelengths (412 nm, 440 nm and 490 nm) was attenuated to 1% of the su bsurface irradiance in the first 5 m of the water column within the GOP, and in the fi rst 10 m of the water column in the SEC. Furthermore, the absorption by Gelbstoff significantly decreased the water leaving radiance (Lw( )) in the blue wavelengths al ong the Orinoco river plume. As ag( ) relatively decreased from the GOP to the SEC ( X 1.6 m-1and X 0.9 m-1, respectively),

PAGE 14

xi a shift in the maximum peak of Rrs( ) spectra (Rrsmax( )), towards shorter wavelengths (from ~ 580 nm to ~500 nm) was observed. Similar to Gelbstoff concentrations of TSM normally decreased from the stations near the Delta to the stations in the SEC. The impact of TSM on the color of the Orinoco plume was represented by a reduc tion in the magnitude of Rrsmax( ) of ~50% going from the waters near the Orinoco delta to th e SEC, indistinctivel y of the season.

PAGE 15

1 Introduction Ocean color studies have helped develop methods to assess the biomass of marine phytoplankton using remote sensin g techniques. These studies ha ve intensified in the last 10 years, driven by the need to understa nd the role of phytoplankton and dissolved organic matter (DOM) in the global carbon budge t; the spatial and temporal variability of productivity over large and regi onal scales; and, the quality of coastal waters. However, we still know little about the optical characteristics of river plumes. This is an important topic because of the influence of rivers on th e color of adjacent marine waters. In the case of large rivers, such as the Amazon and Or inoco Rivers, this influence extends hundreds to thousands of kilometers from the river' s delta and affects the open ocean as well. Regional and global chlorophyll and primary production estimates are affected by river plumes as is the interpretation of other regi onal processes when using remotely sensed data from ocean-color satellites. Significant accuracy improvements in para meter estimates derived from satellitebased sensors, (such as open ocean chlorophyl l concentration) have taken place since the launch of the Coastal Zone Color Scanner (CZCS) in 1978. A second generation of ocean color satellite sensors were la unched in the late 1990’s; they include the Sea-viewing Wide Field of View Sensor (SeaWiFS) and the Moderate Resolution Imaging Spectroradiometer (MODIS). These second gene ration sensors have higher radiometric sensitivity, higher spectral and spatial resolu tion, and better calibrati on, therefore they are expected to perform better in estimating chlorophylla concentrations (Hooker et al ., 1993; McClain and Fargion, 1999; Doerffer et al. 1999). Pigment concentrations may be derived from radiance measurements collected over the ocean by satellite sensors using bio-opt ical algorithms. Some of the most used algorithms are empirical, and are based on sta tistical regressions of radiance versus

PAGE 16

2 chlorophyll (O’Reilly et al. 1998 and 2000). Satisfactory results can be obtained for oligotrophic Case I waters (Morel et al ., 1977), in which phytoplankton and their derivative products play a dominant role in de termining the bio-optical properties of the ocean. Coastal waters, however, have been recognized as optically complex (Sathyendranath, 2000). Phytoplankton pigments may not be the main cause for changes in the color of these Case II waters (Morel et al. 1977), and other particulate and dissolved substances may be present in concen trations high enough to affect the color of the water, thus masking the signal du e to phytoplankton pigments. Among these substances are colored dissolved organic matter (CDOM, also called Gelbstoff ), detritus, and other suspended sediments. Even the most robust global bio-optical algorithms in use today usually fail to be accurate in turbid coastal waters. This study uses in situ measurements of key bio-optic al properties of the Orinoco River Plume, with the purpose of assessi ng the relative importance of phytoplankton, CDOM, and suspended sediments in defining the color of this plume. The results of this study will help improve algorithms used to study coastal waters using remote sensing techniques. This information would be valu able in the understandi ng of the contribution of dissolved organic matter (DOM) and its colored fraction to th e carbon budget in the SEC.

PAGE 17

3 Objectives The overall focus of this research is to a ssess the bio-optical prope rties of the Orinoco River plume. The general objective is to descri be the seasonal and sp atial variability of bio-optical properties of the Gulf of Paria (GOP) and Southeastern Caribbean Sea (SEC) during low and high Orinoco River discharge. Specifically, this study: 1. Determines the contribution of colo red dissolved organi c matter (CDOM) and of suspended particles (e.g. phytoplankton a nd detritus) to the color of the water observed in the GOP and SEC. 2. Determines the optical patterns characte ristic of the plume relative to the total attenuation of light (absorption dominat ed versus scattering dominated). 3. Determines the impact of the bio-optic al properties that characterize the Orinoco River plume on the performan ce of ocean color and bio-optical algorithms.

PAGE 18

4 Study Site Many studies have examined the impact of the Orinoco River plume on the Caribbean Sea and Atlantic O cean (Mller-Karger et al. 1989; Bidigare et al ., 1993; Bonilla et al. 1993; Blough et al ., 1993; Farmer et al. 1993; Hochman et al ., 1994; Del Castillo et al. 1999; Corredor and Morell, 2001; Morell and Corredor, 2001; Corredor et al. 2004). However, there has been no systematic assessm ent of seasonal changes in the bio-optical properties of the Orinoco’s plume. Orinoco River The Orinoco River originates in the sout hern part of Venezuela (Figure1), and discharges waters from about 31 major a nd 2,000 minor tributarie s into the western tropical Atlantic. The Orinoco is considered to be the third largest river in the world in terms of volumetric discharge (after the Am azon and the Congo) discharging an average of 3.6 x104 m3 s-1 (Muller-Karger et al. 1989) as estimated at Puente Angostura (Venezuela). Therefore, this widely published estimate does not include the contribution from the Caroni River, which enters th e Orinoco downstream of where the hydrograph was monitored historically. Figure 2 shows the Orinoco River hydrogr aph based on data collected from 1923 to 1989. Low discharge occurs during the dry seas on (January – May) and high discharge during the rainy season (July – October) as a result of the meridional migration of the Intertropical Convergence Zone (ITCZ). Maxi mum discharge occurs around August, with a mean of 7x104 m3 s-1. Minimum flow occurs around March with a mean of 1x104 m3 s-1 (Bonilla et al. 1993; Muller-Karger et al. 1989).

PAGE 19

5 Some of the waters of the Orinoco River originate in the ancien t Guyana Shield, and as the river flows toward th e ocean it receives waters fr om the Andes mountains and from the Venezuelan plains or Llanos (Paolin i, 1995). Because of its discharge rate, the Orinoco delivers between 86.3 and 150 x 106 tons yr-1 of suspended sediments to the Atlantic and Caribbean waters (Bonilla et al. 1993 ; Meade et al. 1990; Lewis and Saunders, 1990). Between 85% and 90% of the suspended sediments found in the Orinoco are contributed by waters from the Andes and the Llanos (Meade et al. 1990). Large quantities of suspended solids and dissolv ed substances give the Orinoco a “white” color. In contrast, the waters draining the Gu ayana Shield are typical “black waters”, with very low concentrations of suspended solid s (Monente and Colonnello, 1997) but high concentrations of CDOM (Lewis and Saunders, 1990). The concentration of dissolved and partic ulate matter in the Orinoco River depends on seasonal discharge. According to Pao lini (1995), high con centrations of particulate organic carbon (POC) and part iculate organic nitr ogen (PON) (0.6-1.0 and 0.22-0.38 mg l-1, respectively) are found in the main river stem during rising water (MayJune), while lower values are found during th e low water period (January – April). Mean concentrations of dissolved organic carbon (DOC) and dissolved or ganic nitrogen (DON) in the main stem of the rive r were reported in 4.4 mg l-1 and 160.17 mg l-1 by Lewis and Saunders, 1989 and 1990. Meade et al. (1990) observed that suspended sediment concentrations were characterized by two maxima and two minima, with one of the minima taking place during the peak water discharge (August to September). Lewis and Saunders (1990) reported that n itrate is the dominant form of inorganic nitrogen found in the tribut aries and in the main stem of the Orinoco (around 80 g l-1). Nitrite showed very low values a nd phosphate values were around 10 g l-1. In spite of the relativel y high nutrient concentrations Lewis and Saunders (1990) found low gross photosynthesis values in the Orinoco main stem and in its tributaries (consistently below 50 mg C m-2 day-1). They also reported ch lorophyll concentrations within the main stem ranging from 0.11 g l-1 to 0.18 g l-1 chlorophyll a They finally concluded that the phytoplankton biomass with in the river never reached high levels, and

PAGE 20

6 values were characteristic of oligotrophic lakes. Lewis and Saunders (1990) concluded that light was a limiting factor. The Orinoco River discharges between 3.5 and 6.8 x 106 metric tons of organic carbon into the ocean per year (Lewis and Saunders, 1990; Monente and Colonnello, 1997). Monente and Colonnello (1997) esti mated that while total organic carbon concentrations in the main stem of the Orinoco River are about 3.14 mg l-1 not all of this actually reaches the ocean. They estimated that much of this is retained in the delta, reducing the discharge of organic carbon into the ocean to 3.0 x 106 metric tons per year. Orinoco Delta The Orinoco Delta, also known as th e Lower Orinoco, is geomorphologically complex and large, covering an area of about 22,500 km2. Five or more different but inter-related hydrologic zones can be identified within the delta, based on their hydrochemical characteristics (Monente and Colonnello, 1997). Its la rgest channels are Boca Grande, Mnamo, and Macareo (Figure 1). The Manamo and Macareo tributaries are blocked by dams with regulated flow. Gulf of Paria The Gulf of Paria (GOP) is a semi-enclo sed basin adjacent to the northern region of the Orinoco Delta. The exchange of water and sediment with the Atlantic Ocean and the Caribbean Sea are controlled by two narrow ch annels located to the south (Serpent’s Mouth) and to the north (Dragon’s Mout h) of the Gulf. According to Warne et al. (2002) the GOP receives and retains a significant porti on of the sediments discharged from the Mnamo and the Boca Grande channels. Wate rs and sediments discharged through Boca Grande may mix with waters and sediments from the Amazon River before entering the Gulf. Bonilla et al. (1993) found that, in spite of low nitr ate concentrations in the GOP, primary productivity is high. Primary productivi ty values in Dragon’s Mouth were higher

PAGE 21

7 during the dry season (31.0 g C m-3 h-1 than during the wet season (12.9 g C m-3 h-1) They attributed the high productiv ity to recycling of nutrients. The Orinoco River Plume in the Southeastern Caribbean Sea The Orinoco River plume has significant im pact on the quality and quantity of light available to surface waters of the Caribbea n. Using satellite imagery from the Coastal Zone Color Scanner (CZCS) Muller-Karger et al. (1989) showed the seasonal dispersal of the Orinoco River plume in the eastern Caribbean Sea. This plume covers an area exceeding 3 x 105 km2 every year. Bidigare et al. (1993) found only small s easonal variations in the concentration of chlorophyll a in the subsurface chlorophyll maximu m of the Caribbean Sea. During high river discharge and when the Orinoco plume was present in the southeastern Caribbean Sea, the depth of the subsurface chlorophyll maximum was 39 16 m and diatoms dominated the subsurface phytoplankton commun ities. During low river discharge, the subsurface chlorophyll maximum was deeper (77 12 m), and the phytoplankton communities were distributed in two layers an upper layer (<60m) dominated by lightadapted phytoplankton populations (mainly cy anobacteria), and a lower layer (60 m – 200 m) dominated by shade-ad apted phytoplankton populations (chromophytes and green algae). Farmer et al. (1993) reported that during the fall, all the UV light is absorbed within the upper 5 m of the plume in the Caribbean Se a. The loss of UV light at shallow depths shows the effect of colored dissolved organic matter (CDOM) in the plume. Blough et al. (1993) reported high values of CDOM th roughout the eastern Caribbean during the period of high Orinoco river discharge. They estimated that the annual discharge of CDOM by the Orinoco is around 2.5 x 1012 g C yr-1, or about 1% of the total global transport of dissolved organic carbon to the ocean. Del Castillo et al. (1999) also found that high concentrations of dissolved or ganic carbon (DOC) and a bundance of CDOM in the eastern Caribbean were related to the Or inoco river plume. CDOM was a major factor

PAGE 22

8 controlling light penetration, and therefore, it also cont rolled the position of the chlorophyll maximum under the plume. Due to the interference of CDOM, it has been difficult to obtain good estimates of surface chlorophyll concentrations for that por tion of the Caribbean under the influence of the Orinoco River plume using traditional ocean color algorithms and remotely-sensed data (Muller-Karger et al ., 1989; Hochman et al ., 1994). This study further describes the optical properties of the plume. GOP Orinoco River South Eastern Caribbean Dragon’s Mouth Serpent’s Mouth Boca Grande ORD Macareo channel Mnamo channel Figure 1. Study Region

PAGE 23

9 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 100000 Jan.Feb.Mar.Apr.MayJun.Jul.Aug.Sep.Oct.Nov.Dic. MonthsDischarge (m3 s-1) Figure 2. Composite Orinoco River runoff at Puente Angostura, showing maxima, minima, and mean monthly values, using climatology data from 1923 to 1989 (Source: Muller-Karger et al. 1989; Vrsmarty et al. 1996 and Vrsmarty et al. 1998)

PAGE 24

10 Data Collection and Processing Scientific Expeditions to the Orinoco River and Plume Optical and other oceanographic measuremen ts were collected during six different cruises to the Orinoco Delta, Gu lf of Paria, and southeastern Caribbean Sea (Figure 3) on board of the R. V. "Hermano Gins". Thes e cruises were done as part of NASA's SIMBIOS program (Sensor Intercompari son and Merger for Biological and Interdisciplinary Oceanic Studies). Two annual cruises were carried out in each of three years starting in June of 1998 (Table 1). Th e cruises sought to occupy approximately the same stations during high and low river di scharge (Table 2). Measurements collected during each cruise are summarized in Table 3. Most samples and optical measurements collected in these expeditions were processed at the “Estacin de Investigaciones Marinas de Margarita (EDIMAR)” of La Salle F oundation, with the exception of the nutrient analysis and flow-trough measurements proce ssing which were carried out at the College of Marine Science, Universi ty of South Florida (USF). CTD casts and Water Sample Analyses Vertical profiles of salinity, temp erature, attenuation coefficient ( c660), and chlorophyll fluorescence were carried out at each station. The instruments were placed in a “rosette” which included a Sea-Bird C onductivity, Temperature, and Depth (CTD) sensor, a SeaTech beam transmisometer (660 nm), and a Chelsea chlorophyll fluorometer. Water samples were collected at each station at 1 m de pth using 8L Niskin bottles placed in the “rosette”. Water samples for pigment analyses were filtered through GF/F filters. The pigments were extracted using hot methanol (99.8%) and then measured fluorometrically (HolmHansen et al. 1965) using a Turner Designs Fluorometer model 10-AV.

PAGE 25

11 Nutrient determinations for silica (Si(OH)4), phosphate (PO4), nitrite (NO2) and nitrate (NO3) were carried out following the methods described by Gordon et al. (1993), while the method by Grasshoff (1976) was used for the determination of ammonia (NH4). Total suspended matter (TSM) concentrati ons were determined using the method of Aminot (1983). Absorption coefficients Water samples for particle absorption ( ap) were filtered using Whatman GF/F filters of 25 mm diameter. Enough water was filtered to exceed an optical density of 0.04 at 675 nm (Bissett et al ., 1997). The volumes filtered ranged from 0.05 to 2.0 L. The absorption coefficients were measured following the filter pad method described by Kishino et al. (1985), as modified by Bricaud and Stramski (1990). A PHOTORESEARCH PR-650 (Spectrascan) spec troradiometer with a 4 nm spectral resolution and a band range of 380 nm to 780 nm was used to measure the optical density of the filters. After the first measurement, phytoplankton pigments were extracted by soaking the filters with hot methanol (99.8%), and the optical density of the filters was measured one more time to account for the absorbance due to de-pigmented particles (detritus). Particle and detritus ab sorption coefficients ( ap and ad, respectively) were then obtained by following the Mitchell and Kief er (1988) method, as modified by Bricaud and Stramski (1990). The absorpti on coefficient of phytoplankton ( aph), was then calculated as the difference between ap( ) and ad( ). Gelbstoff or CDOM (colored dissolved organi c matter) are the yellow-brown colored organic compounds present in rive r and seawater. They absorb light in the near ultraviolet and in the blue regions of the spectrum (Pilson, 1998). More specifically, the CDOM consists of humic and fulvic acids, which can originate from local sources (e.g. local phytoplankton degradation) or la nd sources and transported fr om a distant location (e.g. via river runoff). Gelbstoff absorption, ag( ), is usually as a proxy for CDOM concentrations.

PAGE 26

12 Samples for ag estimates were filtered using pre-combusted glass-fiber filters (Whatman GF/F 0.7 m pore size) to remove particulate material. To collect the water samples, the filters were mounted in pre-co mbusted stainless-steel holders and connected directly to the Niskin bottles using silicon tubing. Then the absorbance of the filtrate was measured using a 10 cm long cuvette and an Ocean Optics spectrophotometer with a spectral resolution of 0.23 nm and a spectral range of 185 nm to 475 nm. The absorbance or optical density (D) was estimated as: D = log10( I0 / I) (Kirk, 1994) Where: I0, is the intensity of incident light I, is the intensity of transmitted light The absorption coefficient ( a ) was obtained from the absorbance (D) by a = 2.303 D/r with a pathlength, r = 0.1 m The absorption spectra by CDOM typically decreases with increasing wavelength in an exponential form (Kirk, 1994; Blough and Del Vecchio, 2002); therefore, an exponential function (Bricaud and Prieur. 1981) was used to f it the absorption spectra and calculate the slope (S) using a nonlinear least squares fitti ng routine after a logarithmic transformation of the data. Any spectra for ag that did not follow an exponential curve was excluded from this study. Surface Reflectance and Subsurfa ce Radiometric Measurements Radiometric measurements were collect ed underwater using a PRR-600 radiometer from BIOSPHERICAL INSTRUMENTS, and a bove water using a PR-650 (spectrascan) from PHOTORESEARCH. Software created at the Institute for Marine Remote Sensing (IMARS) using IDL (from Research Systems Inc.) were used to process the data collected by these instruments. The measuremen ts and processing of the radiometric data were carried out following the Ocean Optics Pr otocols for Satellite Ocean Color Sensor Validation (Mueller et al. 2002).

PAGE 27

13 Underwater measurements included pr ofiles of spectral dow nwelling irradiance Ed(z, ), upwelling irradiance Eu(z, ), and upwelling radiance Lu(z, ). To assess Ed(0-, ) and Lu(0-, ), where 0indicates values just below the air-sea interface, the profile measurements of Ed(z, ) and Lu(z, ) were normalized usi ng the sky irradiance Esky( ) measured on deck, and extrapolated to the sea surface using a least squares fit routine from the shallowest depth where a clean pr ofile was collected (i .e. one not obviously contaminated by surface artifacts such as wave focusing of light rays, or bubbles). Diffuse attenuation coefficients for Ed( ), or Kd( ), were derived by integration of Ed(z, ) over depth, using a linear least-squares fit. Kd( ) was used to calculate the wavelengths and the depth of maximum light penetration (zmax), defined here as the depth at which light is reduced to 1% of the subsurface irradiance ( = 4.605) was estimated as (Bukata et al. 1995 and Kirk, 1994): ( ,z) = Kd( )zmax zmax = 4.605 / Kd( ) Remote sensing reflectance, R rs( ), is defined as the ratio of water leaving radiance, Lw( ), to downwelling irradiance, Ed( ) (Carder and Steward, 1985; Mobley, 1994; and Kirk, 1994). R rs( ) values were derived from above-water measurements of Ed( ), total radiance Lt(0+, ), and downwelling sky radiance Lsky( ). Since direct measurements of Lw( ) are not available, this was estimated from Lt( ) and Lsky( ) using a Fresnel reflectance of 0.02 (Mobley, 1994): Lw ( ) = Lt ( ) ( Lsky ( ) 0.02) Measurements of Lt( ) and Lsky( ) were collected with ~ 30 from nadir for Lw( ) and from zenith for Lsky( ), and azimuth ~ 90 from the solar plane.

PAGE 28

14 Flow-through Measurements A WET Labs ac-9 was used to measure tota l attenuation, total ab sorption, and total scattering (by difference) in a flow-through mo de along the cruise tr ack without filtration. Following the WET Labs ac-9 User’s Guide, th e instrument was calibrated with Milli-Q water and the total absorption was corrected for scattering by subtracting the absorption at 715 nm from the absorption at all wavelengths (Zaneveld et al ., 1994 ) Chlorophyll and CDOM fluorescence, temperat ure and salinity were also measured along track using a WETLabs WETStar fluor ometer, a Turner Designs Fluorometer model 10-AV, and a SeaBird Electronics conductivity-temperature-depth (CTD), respectively. The data from the different instruments were merged based on time, along with the geographic position data collected with a GARMIN GPS, using a program created in IDL. Phytoplankton Taxonomy Water samples for phytoplankton taxonomy were collected directly from the Niskin bottles after each CTD cast, and were imme diately preserved using 5% formaldehyde. Back in the lab, phytoplankton sp ecies were identified by li ght microscopy using phase contrast optics (Hasle and Syvertsen et al. 1996). Cell counts were made using sedimentation cameras in an inverted microscope. Ocean Color and Bio-optical Algorithms The term ‘ocean color’ is used to describe the spectral composition of the visible light emitted from the ocean as a result of the irradiance spectra, atmospheric conditions, viewing geometries, and the absorption and scattering properties of the water and its constituents (McClain, 2001). The term ‘bio-optical’ was in itially used to acknowledge the fact that optical properties in the wa ter largely depend on biological activity, mainly on phytoplankton and their deri vatives (Morel, 2001). Thes e two terms are sometimes used interchangeable when referring to al gorithms or models. While ‘ocean color

PAGE 29

15 algorithms’ can be described as those de veloped for the main purpose of deriving geophysical quantities (e. g. chlorophyll conc entrations) from optical measurements, commonly represented by the irradiance refl ectance or remote sensing reflectance (R( ) and Rrs( ), respectively), ‘bio-optical algorith ms (or models)’ are commonly used to predict and/or analyze the inherent optical properties (IOPs), such as absorption and scattering coefficients, of the water and its constituents. Empirical, semi-empirical, or analytical ap proaches are available when developing an algorithm (Morel and Gordon, 1980). The Empiri cal approach is based on statistical regressions and is commonly used on ocean co lor algorithms; the analytical approach is based on the radiative transfer theory and is commonly used on bio-optical models. These approaches can be combined within one al gorithm that can be used both to derive geophysical quantities and optic al properties of water constituents (e.g. Carder et al ., 1999). SeaWiFS OC4v4 Algorithm In this study, the SeaWiFS empirical ocean color (OC) algorithm OC4v4 was used with in situ above water Rrs measurements at 443, 490, 510, and 555 nm to derive chlorophylla concentrations for some of the stations in the SEC. The derived chlorophylla concentrations, [Chl_ader] were compared to i n situ measurements of chlorophylla, [Chl_ain situ] to validate the performance of this OC algorithm, without considering the implications surrounding at mospheric corrections in coastal waters. OC4v4 relates Rrs band ratios to chlorophylla concentrations using a fourth order polynomial equation, as follows: 234 der 4444(0.366 3.067R+ 1.930R + 0.649R1.532R)Chl_a10.0=

PAGE 30

16 where, 443490510 555555555 410R=log(R>R>R), the superscript and subscript indicate the wavelengths used in each band ratio, the maximum band ratio (MBR) is used by the algorithm (O’Reilly et al., 2000). The percentage error %E between [Chl_ader] and [Chl_ain situ] was calculated as: %E = 100*[Chl_ader – Chl_ain situ]/Chl_ain situ Negative values of %E indicate that the deri ved chlorophyll values were underestimated, and positive values indicate that the deri ved chlorophyll values were overestimated. Bio-optical Inversion Model In situ above water, radiometric measurements were used to derive absorption and backscattering coefficients using a hyperspe ctral, remote sensing reflectance inversion model developed by Lee et al. (1999). This model uses an optimization technique to minimize the differences between a measured Rrs() and a modeled Rrs(). The minimization is carried out by adjusting the va lues of a group of variables provided as input to the model. For this purpose, upper an d lower limit values are initially set for each variable. These variables include: total absorp tion and total backscattering coefficients (a() and bb(), respectively), bottom albedo (), and bottom depth (H). When the difference between the measured Rrs() and the modeled Rrs() reaches a minimum, the values of a() and bb(), and H are derived, and can be compared to measured values. This model is based in the assumption th at in a vertical homogenous water column, and ignoring inelastic scattering contributions, Rrs() can be modeled as a function of the variables mentioned above (Lee et al., 1999). Even though, the water column in the southeastern Caribbean is not homogenous due to the presence of a lower salinity layer, the model was applied since measurements of the diffuse attenuation coefficient (Kd) showed most of the light was attenuated in this surface layer (see Light field section).

PAGE 31

17 Therefore, it is assumed that the bottom layer has a negligible influence in the optical signal coming out from the surface. The Rrs inversion method developed by Lee et al. (1999) uses the following approximation: Rrsraw() Rrs() + is a spectrically constant offset Where, 1. Rrsraw() is calculated from above-surface m easurements of upwelling radiance, Lu(), sky radiance, Lsky(), and dowelling irradiance, Ed() (these were the only parameters input into the model): Rrsraw() = Trs() F()Srs(), u rs dL T E s ky rs dL S E Trs, is the total remote sensing reflectance Srs, is the sky input 2. Rrs() is the modeled Rrs() derived from a semi-analytical model developed by Lee et al. (1998) : 0.5 Rrs 11.5rs rsr r 111 1expexp cos()2cos()2CB dp uu rsrs wwDD rrHH (0.0840.170)dp rsru u 051.03(12.4)C uDu

PAGE 32

18051.04(15.4)B uDu /()bubab bab bb = bbw + bbp, a = aw + aph + ag dp rsr is the remote sensing reflectance for optically deep water C uD is the optical path el ongation factor for scatte red photons from the water column B uD is the optical path el ongation factor for scatte red photons from the bottom u and k, are inherent op tical properties bb, is the total backscattering coefficient bbw, is the backscattering coefficient of pure water bbp, is the backscattering coeffi cient of suspended particles a is the total absorption coefficient aw, is the absorption coe fficient of pure water aph, is the absorption coefficien t of phytoplankton pigments ag, is the absorption of gelbstoff A detailed description of the semi-analyti cal model and the remote sensing inversion model can be found in Lee et al. (1998) and Lee et al. (1999). The derived aph and ag values at 440 nm were compared to the Gelbstoff and phytoplankton coefficients at this wavelength determined from water sample analysis. The percentage error between the derived and measured values was computed as: %E=100* /dermeamea iiiQQQ

PAGE 33

19 Where der iQ represents the quantity derived (either aph or ag) and mea iQrepresents the measured quatity (either aph or ag).

PAGE 34

20 Figure 3. Location of the sampling stations for each cruise Cruise: SIMBIOS 1 Jun. 24 –Jun. 28 1998 Cruise: SIMBIOS 2 Oct. 27 –Oct. 30, 1998 Cruise: SIMBIOS 3 Feb. 23, Feb. 26, 1999 Cruise: SIMBIOS 5 Mar. 27, Mar. 31, 2000 Cruise: SIMBIOS 4 Oct. 26, Oct 30, 1999 Cruise: SIMBIOS 6 Oct. 21, Oct 26, 2000

PAGE 35

21Table 1. Orinoco-SIMBIOS Cruises CRUISE IDDATES SEASON SIM1 (9)Jun. 24 – Jun. 28, 1998Dry-wet SIM2 (16)Oct. 27 – Oct. 30, 1998Wet SIM3 (15)Feb. 23 – Feb. 28, 1999Dry SIM4 (14)Oct. 26 – Oct. 30, 1999Wet SIM5 (14)Mar. 27 – Mar. 31, 2000Dry SIM6 (17)Oct. 21 – Oct. 26, 2000Wet ( ) = number of stations sampled

PAGE 36

22Table 2. Location and date for each station Continued on the next page STATION LAT. (N) LON. (W) DATESTATION LAT. (N) LON. (W) DATE SIM1_410.78-62.0025-Jun-98SIM4_310.03-62.0628-Oct-99 SIM1_510.79-61.8525-Jun-98SIM4_410.55-61.8929-Oct-99 SIM1_610.67-61.8326-Jun-98SIM4_510.73-61.8329-Oct-99 SIM1_710.55-61.8926-Jun-98SIM4_610.80-61.8129-Oct-99 SIM1_811.01-61.8726-Jun-98SIM4_710.82-61.9629-Oct-99 SIM1_911.22-62.0026-Jun-98SIM4_810.93-61.7830-Oct-99 SIM1_1111.02-61.8127-Jun-98SIM4_911.11-61.8130-Oct-99 SIM1_1211.10-61.8927-Jun-98SIM4_1011.28-61.8230-Oct-99 SIM1_1311.16-61.9527-Jun-98SIM4_1110.17-62.1028-Oct-99 SIM2_210.11-62.1127-Oct-98SIM4_169.77-61.8127-Oct-99 SIM2_1410.03-62.0127-Oct-98SIM4_179.85-61.8326-Oct-99 SIM2_1510.01-62.0227-Oct-98SIM5_19.98-61.8029-Mar-00 SIM2_1610.00-62.0227-Oct-98SIM5_210.00-61.9329-Mar-00 SIM2_310.43-61.9628-Oct-98SIM5_310.03-62.0629-Mar-00 SIM2_410.55-61.8928-Oct-98SIM5_410.55-61.8930-Mar-00 SIM2_510.68-61.8328-Oct-98SIM5_510.67-61.8330-Mar-00 SIM2_610.81-61.8328-Oct-98SIM5_610.78-61.8330-Mar-00 SIM2_710.78-62.0028-Oct-98SIM5_710.78-62.0030-Mar-00 SIM2_1011.21-61.7929-Oct-98SIM5_810.94-61.7831-Mar-00 SIM2_911.11-61.8029-Oct-98SIM5_911.10-61.7831-Mar-00 SIM2_811.02-61.7829-Oct-98SIM5_1011.27-61.7831-Mar-00 SIM2_5b10.67-61.8330-Oct-98SIM5_1110.17-62.0829-Mar-00 SIM2_6b10.79-61.8330-Oct-98SIM5_1211.43-61.7831-Mar-00 SIM2_1710.95-61.8630-Oct-98SIM5_169.84-61.6127-Mar-00 SIM2_1810.87-61.9430-Oct-98SIM5_179.76-61.7528-Mar-00 SIM3_19.98-61.8024-Feb-99SIM6_19.98-61.8024-Oct-00 SIM3_210.00-61.9224-Feb-99SIM6_210.00-61.9324-Oct-00 SIM3_310.03-62.0624-Feb-99SIM6_310.03-62.0624-Oct-00 SIM3_410.55-61.8925-Feb-99SIM6_410.55-61.8925-Oct-00 SIM3_510.73-61.8325-Feb-99SIM6_510.67-61.8325-Oct-00 SIM3_610.80-61.8125-Feb-99SIM6_610.78-61.8325-Oct-00 SIM3_710.82-61.9625-Feb-99SIM6_710.78-62.0025-Oct-00

PAGE 37

23Table 2. (Continued) STATION LAT. (N) LON. (W)DATESTATION LAT. (N) LON. (W) DATE SIM3_810.93-61.7826-Feb-99SIM6_810.94-61.7826-Oct-00 SIM3_911.11-61.8126-Feb-99SIM6_911.10-61.7926-Oct-00 SIM3_1011.28-61.8226-Feb-99SIM6_1011.27-61.7826-Oct-00 SIM3_1110.17-62.1024-Feb-99SIM6_1110.17-62.0824-Oct-00 SIM3_149.82-61.5923-Feb-99SIM6_1211.43-61.7826-Oct-00 SIM3_159.80-61.8123-Feb-99SIM6_159.90-61.6823-Oct-00 SIM3_169.77-61.8123-Feb-99SIM6_169.84-61.6121-Oct-00 SIM3_179.85-61.8324-Feb-99SIM6_179.58-61.7522-Oct-00 SIM4_1_19.98-61.8028-Oct-99SIM6_199.93-61.6724-Oct-00 SIM4_1_210.00-61.8728-Oct-99SIM6_2010.42-61.9325-Oct-00 SIM4_210.00-61.9228-Oct-99

PAGE 38

24Table 3. Oceanographic and bio-optical measurem ents collected during each cruise Continuous Profiles: 1. CTD Profiles a. Temperature b. Salinity c. In vivo chlorophyll fluorescence d. Attenuation coefficient at 660 nm ( c 660) Discrete Samples: 2. Chlorophyll a and Phaeopigments 3. Total Suspended Matter (TSM) 4. Nutrients (NO3, NO2, NH4, Si(OH)4) 5. Phytoplankton Taxonomy 6. Absorption Coefficients: a. Particle absorption ( ap): i.Phytoplankton absorption ( aph) ii.Detritus absorption ( ad) b. Gelbstoff absorption ( ag) 7. Radiometric Measurements: a. Under water ( Ed, Eu, Esky, Lu) b. Above water ( Ed, Lt, Lsky) Flow-through measurements (only available for SIM6): 8. Total absorption ( a ) and total attenuation ( c ) at nine wavelengths: 412nm, 440nm, 488nm, 510nm, 532nm, 555nm, 650nm, 676nm, 715nm 9. Chlorophyll fluorescence 10. Colored dissolved organic matter (CDOM) fluorescence 11. Temperature 12. Salinity

PAGE 39

25 Results and Discussion Previous studies concerning the seasonal and spatial variability of bio-optical properties in the Orinoco Rive r (OR), Gulf of Paria (GOP) and Southeastern Caribbean (SEC) were usually restricted to one or two parameters (Muller-Karger et al., 1989; Bidigare et al., 1993; Blough et al., 1993; Hofman et al., 1994; Battin, 1998; Del Castillo et al., 1999). This study integrates, for the first time, measurem ents of inherent optical properties (IOPs), apparent opt ical properties (AOPs), and concentrations of pigment and suspended matter. This section begins with an analysis of the distribution a nd variability of temperature, salinity, nutrients, pigments, suspended so lids, and phytoplankton species, and follows with a description of the variab ility of the IOPs and AOPs. Temperature and Salinity Distribution Hydrographic conditions within the GOP and the SEC vary seasonally. The meridional migration of the inter-tropi cal convergence zone (ITCZ) influences precipitation in the region and therefore rive r discharge. The dry season occurs when the ITCZ is at its southernmost extent. It is characterized by maximum wind speed and evaporation and by minimum sea surface temper atures and precipitation. Conversely, the wet season, when the ITCZ is at its northe rn location, is characterized by minimum wind speed and evaporation and by maximum temper ature and precipitati on (Aparicio-Castro, 2003). There are also strong, but as of yet poorly quantified, seasonal changes in circulation of near-surface water masses in the Caribbean Sea (see, for example, MullerKarger et al., 1989).

PAGE 40

26 Hydrographic profiles (Figure 4) show strong spatial and seasonal changes in nearsurface temperature, and particularly, salin ity in the GOP and immediately to the north of Dragon's Mouth. In general, surf ace temperature was lower by 2-3 C during the dry season than during the wet season, and it was lower in the SEC than in the GOP by 0.5 -1 C. Cooler surface temperatures obs erved nearshore in the SEC during the dry season were related to coastal wi nd-driven upwelling (Figure 5). Salinity was highly variable between seasons, as well as in a gradient between river plume waters and oceanic waters. A salinity di fference of 5 to 15 practical salinity units (psu) was found between the dry and wet seasons in the GOP, with hi gher salinity values occurring during the former. During the dry season the difference in salinity between the GOP and SEC ranged from 2 to 6 psu, while during the wet season this difference ranged from 8 to 11 psu (Tables 4 and 5). Stations in the SEC were characterized by strong vertical temperature and salinity gradients during both seasons, with temp erature values decreasing from ~27 C at the surface to ~ 20 C at the bottom, and salinity values increasing from ~ 30 at the surface to ~ 36 at the bottom. This upper portion of the water column, observed in the SEC, was described by Morrison and Nowlin (1982) as the Caribbean Surface Water (CSW), which contains a mixture of Amazon and Orinoco ri ver waters. Morrison and Smith (1990) also noted a marked seasonal difference in temper ature and salinity in the SEC, observing deeper haloclines during the period July–O ctober, which corresponds to the low wind season. The results observed in this study agree with the observations mentioned above: during the wet season, a surface layer (10 20 m deep) of high temp erature (~30 C) and low salinity values (<30) (Figur es 5 and 6) was observed in the GOP and SEC. This is in contrast to western trop ical Atlantic waters outside the ri ver plume, which at that latitude have surface salinities >36.6. The cruise data show changes in the vert ical structure of the Orinoco plume as it disperses into the Caribbean Sea. Figure 6 illustrates a surface layer with salinity values <30 psu extending seawards from the GOP into the SEC. This plume became shoaled

PAGE 41

27 farther north. Satellite imagery confirmed that the plume spread to the west and westnorthwest after entering the southeastern Cari bbean, and the shallower upstream edge of the plume was all that was sampled. Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_2 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_2 Figure 4. Hydrographic profiles of Temperature (T ) (dotted line), Salinity (S) (solid line), Chl-fluorescence (Chl-fl) (dashed line) and total attenuation coefficient at 660 nm (c(660)) (Dash-dot line) for st ations near Serpents Mout h at about 10.00 N 61.9 W

PAGE 42

28 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_7 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_7 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_7 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_4 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_4 Figure 4 (continued). Hydrographic profiles of Temperat ure (T) (dotted line), Salinity (S) (solid line), Chl-fluorescence (Chl-fl) (das hed line) and total attenuation coefficient at 660 nm (c(660)) (Dash-dot line) for stations in the GOP at about 10.55 N 61.89 W

PAGE 43

29 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_6 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_6 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_6 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_5 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_5 Figure 4 (continued). Hydrographic profiles of Temperat ure (T) (dotted line), Salinity (S) (solid line), Chl-fluorescence (Chl-fl) (das hed line) and total attenuation coefficient at 660 nm (c(660)) (Dash-dot line) for stations near Dragon s Mouth at about 10.67 N 61.83 W

PAGE 44

30 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_12 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_12 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM1_12 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM2_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM3_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM4_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM5_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_9 Chl-fl (relative values) and c(660) (m-1) T (C) and S (psu)DepthSIM6_9 Figure 4 (continued). Hydrographic profiles of Temp erature (C), Salinity, Chlfluorescence (relative values) and total attenuation coefficient at 660 nm (m-1) for stations in the SEC at about 11.10 N 61.8 W

PAGE 45

31 SIM1 SIM3 SIM5 SIM2 SIM4 SIM6 Figure 5. Vertical cross-sections showing the meridional distributi on of Temperature (C) for each cruise. Dry season cruises: SIM1, SIM3, SIM5 Wet season cruises: SIM2, SIM4, SIM6 T ( C)

PAGE 46

32 SIM1 SIM2 SIM3 SIM4 SIM5 SIM6 Salinity (PSU) Figure 6. Vertical cross-sections showing the me ridional distribution of Salinity for each cruise. Dry season cruises: SIM1, SIM3, SIM5 Wet season crui ses: SIM2, SIM4, SIM6

PAGE 47

33Table 4. Temperature and Salinity surface values (~ 1 m) for all the stations Station Temperature (C) Salinity (PSU)Station Temperature (C) Salinity (PSU) SIM1_727.9724.67SIM4_1_229.5022.92 SIM1_628.1924.33SIM4_229.6519.54 SIM1_424.8836.42SIM4_330.0615.32 SIM1_527.3529.45SIM4_1129.1620.50 SIM1_5b27.2627.65SIM4_428.6919.23 SIM1_1128.9027.59SIM4_529.8320.04 SIM1_828.3127.36SIM4_629.2622.24 SIM1_8b27.0331.31SIM4_726.9435.61 SIM1_1228.2831.54SIM4_828.4722.75 SIM1_1328.3531.18SIM4_929.5023.78 SIM1_928.1029.94SIM4_1028.8632.25 SIM2_1629.0619.00SIM5_126.6730.34 SIM2_1529.7315.36SIM5_226.8431.21 SIM2_1429.2118.97SIM5_326.8831.24 SIM2_229.0918.18SIM5_1126.9732.55 SIM2_329.0220.00SIM5_426.7631.79 SIM2_429.6020.79SIM5_526.9531.77 SIM2_5b29.0821.32SIM5_724.7036.23 SIM2_529.3721.65SIM5_626.9431.71 SIM2_727.4834.60SIM5_826.4532.61 SIM2_6b27.3531.75SIM5_925.9233.13 SIM2_629.4822.15SIM5_1026.4333.22 SIM2_1828.0934.15SIM5_1226.8835.46 SIM2_1728.1130.38SIM6_128.9215.79 SIM2_829.2725.40SIM6_230.3111.01 SIM2_928.2528.87SIM6_329.4016.28 SIM2_1027.9729.19SIM6_1129.4817.72 SIM3_126.5721.27SIM6_2029.3719.95 SIM3_226.8326.05SIM6_429.5315.61 SIM3_327.2827.19SIM6_530.2816.36 SIM3_1127.6327.06SIM6_628.5626.41 SIM3_427.0624.23SIM6_728.5334.64 SIM3_526.9725.11SIM6_829.9917.49 SIM3_627.2828.33SIM6_929.1424.31 SIM3_725.0233.54SIM6_1028.6029.28 SIM3_826.6827.09SIM6_1228.8332.45 SIM3_926.7929.17 SIM3_1027.0334.94 SIM4_1_128.9620.32

PAGE 48

34Table 5. Mean Values of Temperature and Salinity Temperature (C) Salinity (PSU) CruiseGOP SECGOP SEC SIM128.0827.6124.5030.27 SIM229.2728.2519.4129.56 SIM327.0726.6325.1629.70 SIM429.4128.6119.7027.33 SIM526.8426.2231.4933.73 SIM629.6128.9416.1027.43

PAGE 49

35Nutrient Concentrations Surface nutrient concentrations varied si gnificantly among the six cruises. Figure 7 and Table 6 show the nutrient concentrations at each sta tion, and Figure 8 and Table 7 show the mean concentration of each nutrient in the GOP and SEC per cruise. In general, during both seasons, nitrate (NO3) and ammonia (NH4) were the dominant nutrients within the GOP. In the SEC, NO3 was still the dominant i norganic nitrogen form, and NH4 concentrations were substantially lowe r than those observed in the GOP. Mean values of NO3 within the GOP ranged from 0.20 M (SIM1) to 0.869 M (SIM3) during the dry season and from 0.546 M (SIM6) to 1.21 M (SIM2) during the wet season. In the SEC mean values of NO3 ranged from 0.5 M (SIM3) to 0.9 M (SIM5) during the dry season, and from 0.0 M (SIM6) to 0.5 (SIM4) during the wet season. During SIM1 and SIM5 mean NO3 values were higher in the SEC than in the GOP, whereas, for the rest of the cruises NO3 values were higher in the GOP than in the SEC. An increase in NO3 concentration was observed in the GOP dur ing the wet season. SIM2 presented the highest concentration of NO3 in the GOP while the highest concentration of NO3 in the SEC was observed during SIM5. NH4 concentrations were higher in the GOP than in the SEC in both seasons. Concentrations of NH4 were also typically higher duri ng high river discharge. During the dry season mean cruise values of NH4 ranged from 0.271 M (SIM3) to 0.380 M (SIM5) within the GOP and from 0.083 M (SIM5) to 0.210 M (SIM1) in the SEC. For the wet season mean cruise values of NH4 ranged from 0.281 M (SIM6) to 1.346 M (SIM4) in the GOP and from 0.163 M to (SIM6) to 0.907 M (SIM4) in the SEC. The highest concentrations were obs erved during SIM4, with a maximum of 2.18 M near the Orinoco Delta (SIM4_1_1). High concentrations of ammonia may indicate low inorganic nitrogen demand and/or very lo w nitrification rates (Lewis and Saunders, 1990). The inorganic nitrogen form found in the lowe st concentration, es pecially during the wet season, was NO2. During the dry season mean cruise values of NO2 ranged from 0.029 M (SIM1) to 0.275 M (SIM3) in the GOP and from 0.155 M (SIM5) to 0.2 M

PAGE 50

36 (SIM1) in the SEC. For the wet s eason these values ranged from 0.037 M (SIM6) to 0.105 M (SIM4) in the GOP and from 0.005 M (SIM6) to 0.123 M (SIM4) in the SEC. Since nitrite is the intermediate nitrogen form during the nitrification process, very low values of this nutrient are common and ar e especially expected to occur in well oxygenated waters. Higher concentrations of nitrite are normally associated with upwelling events where a subsurface oxygen minimum is found. Phosphate concentrations were higher during the dry season, with mean cruise values ranging from 0.059 M (SIM1) to 0.245 M (SIM3) in the GOP and from 0.130 M (SIM1) to 0.228 M (SIM5) in the SEC. While during the wet season the mean concentration of this nutrient ranged from 0.003 M (SIM6) to 0.109 M (SIM4) in the GOP and from 0.002 M (SIM6) to 0.096 M (SIM4) in the SEC. Maximum values of PO4 for the GOP were observed during SIM3 and SIM4, and for the SEC during SIM3 (0.36 M), while the minimum values of this nutrient for both the GOP and SEC were observed during SIM6. The high variability in nutrient concentra tions suggested that there is no definite seasonal or spatial pattern in nutrient distri butions. However, two observations are worth mentioning. First, NH4 concentrations were higher duri ng SIM4 than during any other cruise. Second, NO3 and NH4 were generally higher during the wet season than during the dry season, while PO4 and NO2 were higher during the dry season. Mean concentrations of dissolved silica (Si(OH)4) during the dry season ranged from 10.228 M (SIM5) to 26.563 M (SIM3) in the GOP and from 6.150 M (SIM5) to 13.600 M (SIM3) in the SEC. During the wet season, the mean cruise values of Si concentration ranged from 33.093 M (SIM6) to 43.659 M (SIM2) in the GOP and from 14.923 M (SIM2) to 17.095 M (SIM6) in the SEC. Figure 9 shows Si(OH)4 concentrations were higher during the wet seas on, and also higher in the GOP than in the SEC. During the dry season, Si concentrati ons during SIM3 were two to three times higher than during SIM5. Temper ature profiles did not show a significant difference between these two cruises; salinity profiles, on the other hand, showed a surface salinity layer with values below 30 for SIM3 that wa s not observed during SI M5 (Figure 4). Low

PAGE 51

37 salinity values and high concentr ations of Si found during SIM3 indicate that there was a greater riverine influence on areas sampled during this cruise th an during SIM5. Low temperature values, and hi gher concentrations of NO2 and PO4, however, indicate that upwelling events were also taking place during SIM3. Dissolved silica is normally used as an indi cator of freshwater input due to its semiconservative behavior with salinity (Figur e 10). A strong relations hip between silica and salinity was observed in the GOP and SEC during the dry season, with r2 values ranging from 0.91 to 0.95. During the wet seas on, this relationship was weaker (r2 ~ 0.80) for both regions. Biological uptake and dilution by rainfall may have caused silica to behave in a nonconservative manner during the wet s eason. This differs from Bonilla et al. (1993) who reported that Si(OH)4 appears to behave conservative in the GOP during the fall. According to Bonilla et al. (1993), at low discharge the re sidence time of water in the GOP is apparently long, and subs tantial biological removal occurs. However, the silica values they reported for the GOP during the dr y season are substantia lly lower than those obtained in this study. Also, their mixing line indicated an end-member of Si(OH)4 at salinity zero of ~ 120 M in the GOP, which is a value almost twice as high as the one determined by Froelich et al. (1978) (~60.1 M) in the Venezuelan basin, and the one found in this study (71.13 M) (Figure 11). Values of Si(OH)4 within the Orinco River have been reported between 100 M to ~ 130 M (Lewis and Saunders, 1990; Livingstone, 1963). The difference between thes e values and the ri verine end member found using the mixing line in this study c ould be related to the absence of Si(OH)4 concentrations at lower salinity (e.g. <10) in the plot that could indicate the removal of Si(OH)4 by biological uptake. The mixing line obtained in this study falls very close to the one obtained by Froelich et al. (1978) in the eastern Caribbean, who e xplained it as a three-point water mixing process involving contributions from the equa torial surface water, the Amazon River, and rain water without giving too much importa nce to the water form the Orinoco River. They concluded that the dispersal of Amaz on River water was the major factor causing

PAGE 52

38 the seasonal surface salinity changes observe d in the eastern Caribbean, and that the removal of Si(OH)4 by biological uptake in the Am azon water before entering the Caribbean was insignificant. Ten years later, Muller-Karger et al. (1989) demonstrated using satellite images that the seasonal changes observed in the GOP and through the northern Caribbean are caused primarily by the dispersal of Orinoco river water. The si milarity between the Si(OH)4 mixing line obtained in this study for the GOP and the mixing line observed by Froelich et al. (1978) supports the observa tions made by Muller-Karger et al. (1989) that the Orinoco plume, rather than the Amazon, dominates signals in the GOP. The Amazon has substantial influen ce on the oceanography of the Caribbean, however. Hellweger and Gordon (2002) recen tly established that there is a strong correlation between sea surface salinity (SSS) in the Caribbean and the Amazon plume. Hu et al. (2004) also emphasize the impact of the Amazon on the Caribbean Sea.

PAGE 53

39Figure 7. Phosphate ( PO4), nitrite ( NO2), nitrate ( NO3), and ammonia (NH4) concentrations for each station. 0 05 1 15 2 25 3SIM1_7 SIM1_6 SIM1_4 SIM1_5 SIM1_5b SIM1_11 SIM1_8 SIM1_8b SIM1_12 SIM1_13 SIM1_9StationsNutrient Concentration ( M) 0 05 1 15 2 25 3SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5b SIM2_5 SIM2_7 SIM2_6b SIM2_6 SIM2_18 SIM2_17 SIM2_8 SIM2_9 SIM2_10StationsNutrient Concentration ( M) 0 05 1 15 2 25 3SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_7 SIM3_8 SIM3_9 SIM3_10StationsNutrient Concentration ( M) 0 05 1 15 2 25 3SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_7 SIM4_8 SIM4_9 SIM4_10StationsNutrient Concentration ( M) 0 05 1 15 2 25 3SIM5_17 SIM5_16 SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_7 SIM5_6 SIM5_8 SIM5_9StationsNutrient Concentration ( M) 0 05 1 15 2 25 3SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_7 SIM6_8 SIM6_9 SIM6_10 SIM6_12StationsNutrient Concentration ( M)

PAGE 54

40Table 6. Surface (~ 1m) concentration of nutrients for all stations Continued on the next page Station # PO4 ( M)Si(OH)4 ( M)NO2 ( M)NO3 ( M)NH4 ( M) SIM1_70.0721.980.060.280.48 SIM1_60.0521.560.000.120.20 SIM1_40.050.000.020.000.13 SIM1_50.2416.630.531.880.26 SIM1_5b0.1715.360.381.140.26 SIM1_11nannannannannan SIM1_80.1214.870.180.630.15 SIM1_8bnannannannannan SIM1_12nannannannannan SIM1_130.088.120.030.140.19 SIM1_90.119.550.060.250.28 SIM2_160.2158.210.192.681.42 SIM2_150.1551.880.102.031.03 SIM2_140.0954.340.062.190.75 SIM2_20.0756.510.072.191.06 SIM2_30.0038.460.010.460.41 SIM2_40.0031.480.000.000.42 SIM2_5b0.1127.170.070.000.33 SIM2_50.0031.220.060.130.39 SIM2_70.004.160.000.000.00 SIM2_6b0.1319.270.050.000.26 SIM2_60.0026.390.060.480.31 SIM2_180.104.740.000.120.22 SIM2_17nannannannannan SIM2_80.0019.980.010.000.27 SIM2_90.0014.680.020.130.12 SIM2_100.0415.250.060.320.25 SIM3_10.3137.670.522.390.34 SIM3_20.2727.000.291.140.28 SIM3_30.2324.500.240.590.19 SIM3_110.2521.180.120.110.24 SIM3_40.1622.470.200.110.30 SIM3_50.1922.340.250.310.29 SIM3_60.2617.650.330.910.15 SIM3_70.125.820.090.550.00 SIM3_80.2118.020.160.530.23 SIM3_90.3714.980.090.720.00 SIM3 1 00 2 22 8 00 0 20 0 00 0 0

PAGE 55

41Table 6. (Continued) Station # PO4 ( M)Si(OH)4 ( M)NO2 ( M)NO3 ( M)NH4 ( M) SIM4_1_10.3133.070.161.302.18 SIM4_1_20.0731.910.200.551.23 SIM4_20.1239.320.091.161.62 SIM4_30.1034.680.111.071.17 SIM4_110.0634.440.090.411.25 SIM4_40.0428.770.060.000.98 SIM4_50.0629.820.030.001.00 SIM4_60.2219.600.231.950.92 SIM4_70.115.400.020.210.76 SIM4_80.0330.850.210.031.01 SIM4_90.1120.450.110.320.97 SIM4_100.018.560.040.000.88 SIM5_10.2513.590.060.450.00 SIM5_20.0211.250.030.110.11 SIM5_30.2110.590.110.361.92 SIM5_110.208.220.080.310.00 SIM5_40.249.250.110.260.17 SIM5_50.228.470.020.050.08 SIM5_70.150.030.131.200.14 SIM5_60.229.400.120.540.06 SIM5_8nannannannannan SIM5_90.288.330.281.200.08 SIM5_100.266.840.090.660.05 SIM5_12nannannannannan SIM6_10.0033.990.071.590.69 SIM6_20.0049.110.081.770.67 SIM6_30.0231.640.060.460.39 SIM6_110.0036.420.010.000.14 SIM6_200.0029.660.010.000.04 SIM6_40.0025.590.020.000.04 SIM6_50.0025.240.010.000.00 SIM6_60.0018.260.010.000.11 SIM6_70.014.390.000.000.25 SIM6_80.0028.920.010.000.15 SIM6_90.0025.990.010.000.05 SIM6_100.0016.200.000.000.10 SIM6 1 20 0 08 8 10 0 00 0 00 3 2

PAGE 56

42Figure 8. Mean values of nutrient concentr ations for the GOP (A) and SEC (B) Table 7. Mean Values of Nutrient concentrations (M) CruiseGOP SECGOP SECGOP SECGOP SECGOP SEC SIM10.060.1321.7710.75 0.030.200.200.670.340.21 SIM20.080.0443.6614.92 0.070.031.210.150.730.21 SIM30.240.2326.5613.60 0.270.160.870.500.270.11 SIM40.110.1033.1416.97 0.110.120.640.501.350.91 SIM50.190.2310.236.15 0.070.160.260.900.380.08 SIM60.000.0033.0917.10 0.040.010.550.000.280.16 NH4 ( M) PO4 ( M) SIL ( M) NO2 ( M) NO3 ( M) 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 PhosphateNitriteNitrateAmmonium NutrientConcentration ( M) 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 PhosphateNitriteNitrateAmmonium NutrientConcentration ( M) SIM1 SIM2 SIM3 SIM4 SIM5 SIM6A B

PAGE 57

43 0 10 20 30 40 50 60SIM1_7 SIM1_6 SIM1_4 SIM1_5 SIM1_5b SIM1_11 SIM1_8 SIM1_8b SIM1_12 SIM1_13 SIM1_9StationsSi(OH)4 ( M) 0 10 20 30 40 50 60SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5b SIM2_5 SIM2_7 SIM2_6b SIM2_6 SIM2_18 SIM2_17 SIM2_8 SIM2_9 SIM2_10StationsSi(OH)4 ( M) 0 10 20 30 40 50 60SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_7 SIM3_8 SIM3_9 SIM3_10StationsSi(OH)4 ( M) 0 10 20 30 40 50 60SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_7 SIM4_8 SIM4_9 SIM4_10StationsSi(OH)4 ( M) 0 10 20 30 40 50 60SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_7 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12StationsSi(OH)4 ( M) 0 10 20 30 40 50 60SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_7 SIM6_8 SIM6_9 SIM6_10 SIM6_12StationsSi(0H)4 ( M) Figure 9. Dissolved silica, Si(OH)4, concentration at each station

PAGE 58

44 R2 = 080 0 10 20 30 40 50 60 10203040 Salinity (PSU)Si(OH)4 ( M) R2 = 093 0 10 20 30 40 50 60 10203040 Salinity (PSU)Si(OH)4 ( M) R2 = 084 0 10 20 30 40 50 60 10203040 Salinity (PSU)Si(OH)4 ( M) SIM1 SIM2 R2 = 091 0 10 20 30 40 50 60 10203040 Salinity (PSU)Si(OH)4 ( M) SIM3 SIM4 R2 = 095 0 10 20 30 40 50 60 10203040 Salinity (PSU) Si(OH)4 ( M) SIM5 R2 = 084 0 10 20 30 40 50 60 10 20 30 40 Salinity (PSU)Si(OH)4 ( M) SIM6 Figure 10. Relationship between silica concentrati ons and salinity during each cruise

PAGE 59

45 0 20 40 60 80 100 120 0510152025303540 Salinity (PSU)Si(OH)4 ( M) Linear(SIMBIOS) Froelich (1978) Bonilla (1993) SIMBIOS Figure 11. Relationship between di ssolved silica, Si(OH)4, concentration and salinity in the GOP and SEC.

PAGE 60

46Phytoplankton Pigments Continuous (in vivo) chlorophyll fluorescence profiles we re collected at each station during each cruise. Regressions between su rface fluorescence (in relative values) and surface chlorophyll-a concentrations showed good co rrelation for both the GOP (r2 = 0.75 0.99) and SEC (r2 = 0.85-0.96) during both seasons, with exception of SIM1 (Figure 12). The highest r2 (0.99) value was obtained in the GOP during SIM3. Vertical profiles of chlorophyll fluores cence showed high variability among the stations. In general, stations within the GOP were characterized by high surface fluorescence values and no deep fluorescen ce maximum. Surface fluorescence values were normally higher and deeper during the dry season, especially during SIM3, than during the wet season (Figure 4) For stations near Dragon’ s mouth and in the SEC, a maximum in chlorophyll-fluorescence was more marked but still above 20 m. Chlorophyll-fluorescence peaks deeper than 20 m were rare, as in station SIM6_6 (maximum at ~40m), and chlo rophyll-fluorescence d ecreased to a minima around 40 m. Vertical cross-sections of chlorophyll fluorescence were constructed using the vertical profiles obtained during each cruise (Figure 13). This illustrates how the chlorophyll maximum deepens to the north of Dragon’s Mouth. SIM6 (October 2000) showed the lowest chlorophyll-fluorescence values, while SIM3 showed the highest. During the dry season, SIM1 was the crui se with highest concentrations of chlorophyll-a in both the GOP and SEC, while SIM5 showed the lowest concentrations. In the wet season, SIM2 presented the highest values while SIM6 presented the lowest (Figure 14 and Table 8). According to Varela et al. (2003), based on historical data chlorophyll concentrations may reach 8 g l-1 in waters influenced by the Orinoco river plume; in this study, however, only one station was found to have such high concentration (SIM3_11). Average values of surface (~1 m) chlorophyll-a concentrations within the GOP ranged from 0.9 g l-1 (SIM5) to 3.0 g l-1 (SIM3) during the dry season and from 1.1 g l-1 (SIM2) to 1.3 g l-1 (SIM4) in the wet season. In general, chlorophyll-a was higher in the dry season than during the wet season, and higher within the GOP than in the SEC. In

PAGE 61

47 the SEC, these values ranged from 0.7 g l-1 (SIM5) to 1.7 g l-1 (SIM1) and from 0.7 g l-1 (SIM6) to 1.3 g l-1 (SIM4) for the dry and wet seasons respectively (Table 9). Figure 15 shows the frequency distribution of chlorophyll-a concentration for all cruises. While this distribution indicates that 89% of chlorophyll-a samples had concentrations below 2.0 g l-1, the range of values was higher and therefore characteristic of eutrophic waters (Shifrin, 1983). High chlorophyll concentrations in the GOP and SEC should be expected because of nutrients delivery by the Orinoco River and the presence of upwelling foci along the coast of the SEC in the dry season. A t-Test analysis was used to determine wh ether the seasonal a nd spatial variations observed in chlorophyll-a concentration were significant. In the t-Tests, the closer the probability (P) gets to 1, the higher is the probabil ity that the means from two populations are the same, indicating no signi ficant change. Results (Table 10) indicate that there was significant seasonal vari ation in chlorophyll-a concentration throughout the region; however, concentrations in the plume in the SEC were not statistically different from concentrations within the GOP.

PAGE 62

48 Figure 12. Correlation analysis between Chlorophyll-a and chlorophyll fluorescence in the GOP () and SEC () for each cruise R2 = 0.32 0 1 2 0123 Chlorophyll-a ( g l-1)Chl-fluorescence (relative values)SIM1 R2 = 0.96 0 1 2 0246810 Chlorophyll-a (g l-1)Chl-fluorescence (relative values)SIM3 R2 = 0.84 0 1 2 0123 Chlorophyll-a (g l-1)Chl-fluorescence (relative values)SIM5 R2 = 0.85 0 1 2 0123 Chlorophyll-a (g l-1)Chl-fluorescence (relative values)SIM2 R2 = 0.90 0 1 2 01234 Chlorophyll-a (g l-1)Chl-fluoescence (relative values)SIM4 R2 = 0.92 0 1 2 0123 Chlorophyll-a (g l-1)Chl-fluorescence (relative values)SIM6

PAGE 63

49 SIM1 SIM3SIM5 SIM2 SIM4SIM6 Chl-fluorescence (relative values) Figure 13. Vertical cross-sections showing the meridional di stribution of chlorophyll fluorescence (relative values) for each cruise

PAGE 64

50 Figure 14. Chlorophyll-a (g l-1) concentration at each station 0 1 2 3 4SIM1_7 SIM1_6 SIM1_4 SIM1_5 SIM1_5b SIM1_11 SIM1_8 SIM1_8b SIM1_12 SIM1_13 SIM1_9Stations Chlorophylla ( g l-1) 0 1 2 3 4SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5b SIM2_5 SIM2_7 SIM2_6b SIM2_6 SIM2_18 SIM2_17 SIM2_8 SIM2_9 SIM2_10StationsChlorophylla ( g l-1) 0 1 2 3 4SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_7 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12StationsChlorophylla ( g l-1) 0 1 2 3 4SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_7 SIM4_8 SIM4_9 SIM4_10StationsChlorophylla ( g l-1) 0 1 2 3 4SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_7 SIM6_8 SIM6_9 SIM6_10 SIM6_12Stations Chlorophylla ( g l-1) 0 1 2 3 4SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_7 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12StationsChlorophylla ( g l-1)

PAGE 65

51Table 8. Surface (~ 1m) concentration of Chlo rophyll-a and Total Suspended Matter (TSM) for all the stations Station Chlorophylla ( g l-1) TSS (mg l-1)Station Chlorophylla ( g m-1) TSS (mg l-1) SIM1_72.543.45SIM4_1_10.97nan SIM1_62.312.95SIM4_1_20.39nan SIM1_40.153.45SIM4_20.40nan SIM1_51.945.15SIM4_32.38nan SIM1_5b1.993.90SIM4_111.44nan SIM1_111.233.20SIM4_41.30nan SIM1_81.883.10SIM4_51.88nan SIM1_8b1.45nanSIM4_61.06nan SIM1_122.047.50SIM4_73.00nan SIM1_132.417.10SIM4_80.79nan SIM1_92.423.45SIM4_91.12nan SIM2_161.005.94SIM4_100.28nan SIM2_151.024.00SIM5_11.277.80 SIM2_140.773.85SIM5_21.147.33 SIM2_21.303.65SIM5_31.0823.60 SIM2_31.543.00SIM5_110.977.70 SIM2_40.622.25SIM5_40.512.35 SIM2_5b2.124.30SIM5_50.551.50 SIM2_50.782.30SIM5_71.473.35 SIM2_70.633.80SIM5_60.623.90 SIM2_6b1.643.55SIM5_80.563.60 SIM2_61.634.10SIM5_90.663.55 SIM2_180.684.45SIM5_100.543.70 SIM2_171.80nanSIM5_120.184.20 SIM2_81.003.55SIM6_11.0513.67 SIM2_90.794.10SIM6_21.4638.20 SIM2_100.922.60SIM6_31.3910.20 SIM3_11.8797.50SIM6_111.524.18 SIM3_21.7813.33SIM6_200.670.48 SIM3_31.66125.00SIM6_41.093.12 SIM3_118.1145.00SIM6_50.625.96 SIM3_41.6033.00SIM6_60.713.32 SIM3_51.158.33SIM6_70.1811.28 SIM3_60.649.60SIM6_81.533.25 SIM3_73.528.20SIM6_91.132.37 SIM3_81.109.20SIM6_100.172.37 SIM3_91.7010.80SIM6_120.207.60 SIM3_100.173.05

PAGE 66

52Table 9. Mean Values of Chlorophyll-a and Total suspended solids (TSM) Figure 15. Frequency distribution of Chlorophyll-a concentrations TSS (mg l-1) CruiseGOP SECGOP SEC SIM12.431.723.204.61 SIM21.141.143.663.74 SIM33.001.3862.778.20 SIM41.251.25nannan SIM50.920.678.383.72 SIM61.120.6510.835.03 Chlorophylla ( g l-1)

PAGE 67

53Table 10. T-test results showing the mean seasona l and spatial variab ility of Chlorophyll-a GOPSECt-value P Dry 1.421.330.3120.757 WET1.171.010.7940.433 t-value1.221.24 P 0.240.22 Spatial Variability Seasonal Variability Chlorophylla ( g l-1) Chlorophylla ( g l-1)

PAGE 68

54Total Suspended Matter (TSM) Total suspended matter (TSM) includes susp ended particles from terrestrial (e.g. inorganic sediments, minerals ) and biogenic origin (e.g. phytoplankton, detritus). TSM concentrations (Table 8) were typically hi ghest in the GOP near Serpent’s Mouth and they decreased substantially to the north of Dragon’s Mouth in the SEC. Cruise SIM2 was an exception, when TSM concentrations remained similar within the GOP and SEC, with lower values observed near the Delta. During the wet season, mean surface (~ 1m) TSM values ranged from 3.7 mg l-1 (SIM2) to 10.9 mg l-1 (SIM6) within the GOP and from 3.7 mg l-1 (SIM2) to 5.0 mg l-1 (SIM6) in the SEC. During the dry seas on, these values ranged from 3.2 mg l-1 (SIM1) to 62.8 mg l-1 (SIM3) within the GOP and from 3.7 mg l-1 (SIM5) to 18.2 mg l-1 (SIM3) in the SEC (Table 9). The year 1998 (SIM1 and SIM2) represents the year with the lower mean concentrations of TSM. While, the dry seas on of 1999 (SIM3) represents the season with the highest concentration of TSM. Unfort unately, there were no TSM measurements available for SIM4 for a better seasonal comparison. However, in general, TSM concentrations were higher during the wet season than during the dry season. Furthermore, during each cruises, high TSM variability implied significant patchiness. TSM concentrations were compared to chlorophyll-a concentrations to see if phytoplankton was the particle dominating TSM (Figure 16). Regression analyses indicated that there was not a significant relationship (r2 <0.3) between these two variables in the GOP nor in the SEC during any of the seasons. The beam attenuation coefficient at 660 nm, c(660), is commonly used as an indicator of turbidity due to the presence of suspended material, following the principle that larger particles or higher concentration of part icles increase the attenuation of light. Vertical profiles of c(660) were collected during each cr uise (Figure 4). High values of c(660) indicate higher light attenuation. Surface turbidity was highest near Serpent’s Mouth, where profiles show an increase in turbidity between five and ten meters, a sm aller peak near 20 m, and a peak near the

PAGE 69

55 bottom in both seasons. Near Dragon’s Mouth in the GOP, surface (< 10 m) turbidity was higher during the wet season th an during the dry season. At Dragon’s Mouth, a strong peak in c(660) was observed between 10 and 20 m with a stronger signal during the wet season. This peak was not observed in SIM1, when c(660) values remained higher through the water column than during the rest of the cruise s. Near Dragon’s Mouth, attenuation usually increased below 150 m and near the bottom. The surface (~10m) turbidity peak was also present immediately to the north of Dragon’s Mouth during both seasons; however, no significant peaks were observed below the su rface in the water column down to at least 200 m, the depth of our deepest observations, except frequently near the bottom. Further to the north, the near surface turbidity peak was significantly reduced and c(660) remained low through the water column. Vertical cross-sections of c(660) (Figure 17) show strong light attenuation near the surface in the plume (~ 10m) and within th e GOP (latitude < 10.7 N), where chlorophyll fluorescence is also high. Figure 17 also shows that attenuation at the surface decreased north of Dragon’s Mouth. Bottom re-suspens ion, perhaps by the strong tidal currents observed in the GOP, is evident in increased c(660) values observed near the bottom during both seasons close to Serpent’s Mouth (about 10 N) and Dragon’s Mouth (10.7 N), especially during October 1999 (SIM4). Even though, no significant correlation was found between TSM and chlorophyll-a concentrations, surface peaks of c(660) frequently coincided with an increase in chlorophyll fluorescence (Figure 4), especi ally in Dragon’s Mouth and the SEC. However, regressions between chlorophyll fluorescence and c(660) did not show significant correlation between th ese two parameters in either of the two seasons (Figure 18). The correlation between chlorophyll-a fluorescence and partic le concentration may be low when there is a difference between th e vertical distributi on of phytoplankton cells and the deep chlorophyll maximum (DCM). The amount of chlorophyll a per cell can vary depending on depth and photoadaptat ion conditions. Higher fluorescence will

PAGE 70

56 depend on the amount of chlorophyll but not necessarily on particle concentration (Cullen, 1982; Kitche n and Zaneveld, 1990). Surface values of c(660) were also compared by re gression to surface TSM (Figure 19), but all cruises showed larg ely poor correlations with r2 values ranging 0.02 (SIM1) to 0.34 (SIM2) in the SEC and from 0.01 (S IM3) to 0.97 (SIM5) in the GOP. Higher r2 values were observed for the GOP than the SEC during both seasons. Poor correlations between c(660) and TSM are common for coastal and estuarine waters. Short-term variability of particle size, type, and concentration, and th e formation of aggregates may influence the optical properties of the susp ended particles, renderi ng any relation to the beam attenuation coefficient comp lex (Jago and Bull, 2000; Bunt et al., 1999; Wells and Kim, 1991).

PAGE 71

57 Figure 16. Relationship between total suspe nded matter (TSM) and chlorophyll-a concentrations in the GOP () and SEC () for each cruise 0 2 4 6 02468 TSM (mg l-1)chlorophylla ( g l-1) SIM2 0 2 4 6 8 10 050100150 TSM (mg l-1)cholorphylla ( g l-1) SIM3 0 2 4 6 0510152025 TSM(mg l-1)chlorophylla ( g l-1) SIM5 0 2 4 6 0204060 TSM (mg l-1)chlrophylla ( g l-1) SIM6 0 2 4 6 0246810 TSM (mg l-1)chlorophylla ( g l-1) SIM1

PAGE 72

58 Figure 17. Vertical cross-sections showing the meridional di stribution of the total attenuation coefficient at 660 nm (m-1) for each cruise SIM1 SIM3 SIM5 SIM2 SIM4 SIM6 c (660) ( m1 )

PAGE 73

59 Figure 18. Relationship between Chlorophyll-fluorescence and c(660) in the GOP () and SEC () for each cruise 0 1 2 3 4 5 6 7 8 9 00.20.40.60.81 Chl-fl (relative values)c (660) (m-1)SIM1 R2 = 0.82 0 1 2 3 4 5 6 7 8 9 00.20.40.60.81 Chl-fl (relative values)c (660) (m-1)SIM2 R2 = 0.84 0 1 2 3 4 5 6 7 8 9 01234 Chl-fl (relative values)c (660) (m-1)SIM3 0 1 2 3 4 5 6 7 8 9 00.20.40.60.81 Chl-fl (relative values)c (660) (m-1)SIM5 R2 = 0.92 0 1 2 3 4 5 6 7 8 9 00.20.40.60.81 Chl-fl (relative values)c (660) (m-1)SIM4 0 1 2 3 4 5 6 7 8 9 00.20.40.60.81 Chl-fl (relative values)c (660) (m-1)SIM6

PAGE 74

60 Figure 19. Relationship between TSM and c(660) for each cruise 0 2 4 6 02468 TSM (mg l-1)c (660) (m-1)SIM1 0 2 4 6 02468 TSM (mg l-1)c (660) (m-1)SIM2 0 2 4 6 050100150 TSM (mg l-1)c (660) (m-1)SIM3 R2 = 0.97 0 2 4 6 0510152025 TSM(mg l-1)c (660) (m-1)SIM5 0 2 4 6 0204060 TSM (mg l-1)c (660) (m-1)SIM6

PAGE 75

61Phytoplankton Distribution We examined samples to assess phytopl ankton taxonomy for cr uises SIM2, SIM3, and SIM4, but not for SIM1, SIM5 and SIM6. The marine diatoms Skeletonema costatum and Skeletonema tropicum dominated the phytoplankton popul ation for all stations during SIM2 and SIM4, as well as stations close to the delta during SIM3. At station SIM3_11 within the GOP (Table 2), however, a dinoflagelate that could not be identified was the most abundant. Other diatoms such as Guinardia delicatula and Chaetoceros socialis were observed in similar or higher quantities than Skeletonema near Dragon’s Mouth and in the SEC during SIM3 (spring). This crui se featured higher su rface salinity values. Marine diatoms may be expected to be dominant in this region because the availability of river-derived silica. Indeed, Betzer et al. (1977) noticed a shift in the population of primary producers in the eastern Caribbean a nd adjacent western Atlantic, suggesting that diatoms increased in numbers in areas of high river influence. Also, according to Kirk (1976), diatoms are more likely to have a co mpetitive advantage in regions where blue light is absorbed by CDOM because they have the capability to absorb green light using the pigment fucoxanthin. Absorption Coefficients An important step to refine bio-optical models and ocean color algorithms for the Gulf of Paria (GOP) and the Orinoco plume in the southeastern Caribbean Sea (SEC) is to characterize the visible light absorption by particulate and dissolved organic materials in the region. This section addresses this issue by identifying: 1. Seasonal and spatial patterns of partic le and colored disso lved organic matter absorption in the GOP and SEC; 2. Spectral shape parameters that char acterize the absorption by phytoplankton (aph), detritus (ad) and CDOM (ag), and the chlorophyll-sp ecific absorption of phytoplankton (aph*); 3. Impact of the variability in particle and dissolved organic matter absorption on remote sensing reflectance (Rrs()) in the blue wavelengths.

PAGE 76

62Particle absorption ( ap) The absorption coefficient due to particulates, ap(), is defined as the sum of the absorption coefficients due to phytoplankton, aph(), and detritus, ad(). Figures 20 to 25 show the seasonal and spatial variability of ap(), aph(), and ad() for some of the stations. During both th e wet and dry seasons, ad () dominated ap() for the stations around Serpent’s Mouth. Around Dragon’s Mouth, ap() was much smaller than in the southern GOP and phytoplankton ha d a greater contribution to ap(), exceeding that of ad() particularly during the dry season. In the SEC, phytoplankton absorption generally dominated ap(); however, during SIM5 and SIM6 detritus contribution in the SEC stations was almost as high as the cont ribution from phytoplankton (e.g. SIM5_9 and SIM6_9). In the coastal upwelling focus f ound immediately to the west of Dragon’s Mouth, i.e. station 7 in SIM5 and SIM6 (e quivalent to station 4 in SIM1) phytoplankton dominated ap() during both seasons. Figure 26 provides a visual interpretation of the relative contribution of aph and ad to the total ap. To better understand the spatial variability of ap() in the blue wavelengths, ad() and aph() at 412 nm, 440 nm, and 492 nm were plotte d approximately with distance (against station number), starting from the stations cl osest to the river delta to those farthest offshore in the SEC (Figures 27 to 29 ). Table 11 presents the values of ad, and aph at these wavelengths. A strong gradient in ad() was observed between the stations near Serpent’s Mouth (higher ad values) and the stations around Dragon’s Mouth (lower ad values) during both seasons except in 1998 (SIM1 vs. SIM2). Also during both seasons, a small increase in ad() and aph() was observed immediately to the north of Dragon’s Mouth (between 10.75N – 11.0N), after which ap() decreased again w ith latitude. Higher variability in aph() was observed during the dry seas on than during the wet season, during which no major changes in aph() seemed to take place. During the dry season aph(440) ranged from 0.016 m-1 (SIM5_12) to 0.549 (SIM3_11) and during the we t season from 0.013 m-1 (SIM6_10) to 0.113 m-1 (SIM4_3). Minimum values of aph(440) were normally found in the SEC during the dry season,

PAGE 77

63 except for station 7, which showed a regiona l maximum particularly during this season. Minima were found in the GOP during the wet season. Maximum values, however, did not show any pattern with season except fo r station 7, as mentioned above. Overall, cruise average values of surface aph(440) in the GOP and SEC were very similar during both seasons (Table 12). An extreme value of aph(440) (0.549 m-1) was observed during SIM3 at a station in the GOP (SIM3_11), where a maximum in surface chlorophyll-a concentration (8.107 g l-1) was also observed. Spectra of the specifi c absorption coefficient of phytoplankton, aph* (aph() normalized to the concen tration of chlorophyll-a) are shown in figure 30. Changes observed in aph* were largest in the spectral rang e form 400 nm to 550 nm. Changes in aph* usually indicate changes in phytoplankton’s ability to ab sorb light, which may result from changes in light intensity (photoadapta tion), in nutrients availability, in pigment composition, and/or in size and geometry of the cells (change in population, package effect) (Sathyendranath et al., 1987; Carder et al., 1991 and 1999; Kirk, 1994; Bricaud et al., 1995). Any and all of these are possible in this region and there is no clear mechanism to separate these effects. During the dry season aph*(440) ranged from 0.019 m2 mg-1 (SIM3_1) to 0.095 m2 mg-1 (SIM5_11) in the GOP and from 0.020 m2 mg-1 (SIM1_8) to 0.16 m2 mg-1 (SIM5_7) in the SEC (Table 13). During the wet season aph*(440) coefficients ranged from 0.021 m2 mg-1 (SIM4_4) to 0.072 m2 mg-1 (SIM6_20) in the GOP and from 0.017 m2 mg-1 (SIM6_20) to 0.109 m2 mg-1 (SIM6_7) in the SEC. Most of the values were within the aph*(440) range (0.013 m2 mg-1 to 0.077 m2 mg-1) estimated by Prieur and Sathyendranath (1981). Similar to what was observed by Bricaud et al. (1995), there wa s a tendency of aph* in the blue region of the spectra to incr ease with decreasing ch lorophyll concentration (Figure 31). This has been explained in th e past as a “package effect”, whereby the efficiency of light absorpti on per unit chlorophyll decreases with larger cells, where chlorophyll is packaged and self -shaded inside the cells. No aph(440) intercept at null chlorophyll-a concentration was observed (Fig ure 32), suggesting th at the influence by

PAGE 78

64 detritus absorption in aph was not significant (see Bricaud et al., 1995). Variations in aph*(440) were usually accompanied by small variations in aph*(675). Since detritus absorbs less at higher wavelengths, and because of the intercept test shown above, it is assumed that detritus had minimal cont ribution to the changes observed in aph*(675). Changes in aph*(675) observed within the GOP and SEC were very small, with values close to 0.01 m2 mg-1 in most of the cruises. Larger variations in aph*, both at 440 nm and at 675 nm, in the GOP and SEC were observed during SIM3 than during any of the other cruises. These variations could be related to changes in phytoplankton composition due to nutrient availability and light inte nsity, since they were specifically observed at a station (SIM3_11) in which a patch of red tide was observed (as indicated by a extreme concentration of Chlorophyll-a and confirmed from Rrs measurements), and at the station (SIM3_10) farther offshore in the Southeastern Caribbean, where the concentration of TSM was significantly lower in comp arison to the other stations.

PAGE 79

65 Figure 20. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM1

PAGE 80

66 Figure 21. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM2 (note change in scale for S2_16 and S2_3)

PAGE 81

67 Figure 22. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM3 (Note change in scale for S3_1 and S3_11)

PAGE 82

68 Figure 23. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM4 (Note change in scale for S4_1-1b)

PAGE 83

69 Figure 24. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM5 (Note change in scale for S5_1 and S5_2)

PAGE 84

70 Figure 25. Spatial variability of ap(), aph(), and ad(), for some of the stations during SIM6 (Note change ins scale for S6_1 and S6_2)

PAGE 85

71 Figure 26. Detritus (blue) and phytop lankton (green) absorption c ontributions to particle absorption at 440 nm 0% 20% 40% 60% 80% 100%SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5 SIM2_5 SIM2_6 SIM2_6 SIM2_8 SIM2_9 SIM2_10 SIM2_7 SIM2_18StationsPercentage 0% 20% 40% 60% 80% 100%SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_8 SIM3_9 SIM3_10 SIM3_7StationsPecentage 0% 20% 40% 60% 80% 100%SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_8 SIM4_9 SIM4_10 SIM4_7StationsPercentage 0% 20% 40% 60% 80% 100%SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12 SIM5_7StationsPercentage 0% 20% 40% 60% 80% 100%SIM1_7 SIM1_6 SIM1_5 SIM1_8 SIM1_11 SIM1_12 SIM1_13 SIM1_9StationsPercentage 0% 20% 40% 60% 80% 100%SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_8 SIM6_9 SIM6_10 SIM6_12 SIM6_7StationsPercentage

PAGE 86

72Figure 27. Particle ap(), phytoplankton aph(), and detritus ad() absorption coefficients at 412 nm 0000 0400 0800 1200SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_8 SIM3_9 SIM3_10 SIM3_7Stationsap, aph, and ad at 412 nm (m-1) 0000 0400 0800 1200SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_8 SIM4_9 SIM4_10 SIM4_7Stationsap, aph, and ad at 412 nm (m-1) 0000 0400 0800 1200SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12 SIM5_7Stationsap, aph, and ad at 412 nm (m-1) 0000 0400 0800 1200SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_8 SIM6_9 SIM6_10 SIM6_12 SIM6_7Stationsap, aph, and ad at 412 nm (m-1) 0000 0400 0800 1200SIM1_7 SIM1_6 SIM1_5 SIM1_8 SIM1_11 SIM1_12 SIM1_13 SIM1_9Stationsap, aph and ad at 412 nm (m-1) 0000 0400 0800 1200SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5 SIM2_5 SIM2_6 SIM2_6 SIM2_8 SIM2_9 SIM2_10 SIM2_7 SIM2_18Stationsap, aph, and ad at 412 nm (m-1)

PAGE 87

73Figure 28. Particle ap(), phytoplankton aph(), and detritus ad() absorption coefficients at 440 nm 0000 0400 0800 1200SIM1_7 SIM1_6 SIM1_5 SIM1_8 SIM1_11 SIM1_12 SIM1_13 SIM1_9Stationsap, aph and ad at 440 nm (m-1) 0000 0400 0800 1200SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5 SIM2_5 SIM2_6 SIM2_6 SIM2_8 SIM2_9 SIM2_10 SIM2_7 SIM2_18Stationsap, aph, and ad at 440 nm (m-1) 0000 0400 0800 1200SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_8 SIM3_9 SIM3_10 SIM3_7Stationsap, aph, and ad at 440 nm (m-1) 0000 0400 0800 1200SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_8 SIM4_9 SIM4_10 SIM4_7Stationsap, aph, and ad at 440 nm (m-1) 0000 0400 0800 1200SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12 SIM5_7Stationsap, aph,and ad at 440 nm (m-1) 0.000 0.400 0.800 1.200SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_8 SIM6_9 SIM6_10 SIM6_12 SIM6_7Stationsap, aph, and ad at 440 nm (m-1)

PAGE 88

74 Figure 29. Particle ap(), phytoplankton aph(), and detritus ad() absorption coefficients at 492 nm 0000 0400 0800 1200SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5 SIM2_5 SIM2_6 SIM2_6 SIM2_8 SIM2_9 SIM2_10 SIM2_7 SIM2_18Stationsap, aph, and ad at 492 nm (m-1) 0000 0400 0800 1200SIM1_7 SIM1_6 SIM1_5 SIM1_8 SIM1_11 SIM1_12 SIM1_13 SIM1_9Stationsap, aph, and ad at 492 nm (m-1) 0000 0400 0800 1200SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_8 SIM3_9 SIM3_10 SIM3_7Stationsap, aph, and ad at 492 nm (m-1) 0000 0400 0800 1200SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_8 SIM4_9 SIM4_10 SIM4_7Stationsap, aph, and ad at 492 nm (m-1) 0000 0400 0800 1200SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_5 SIM5_6 SIM5_8 SIM5_9 SIM5_10 SIM5_12 SIM5_7Stationsap, aph, and ad at 492 nm (m-1) 0000 0400 0800 1200SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_8 SIM6_9 SIM6_10 SIM6_12 SIM6_7Stationsap, aph, and ad at 492 nm (m-1)

PAGE 89

75Table 11. Phytoplankton (aph) and detritus (ad) absorption coefficients in the blue bands 412 nm, 440 nm, and 492 nm Continued on the next page Station ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) SIM1_70.0490.0500.0360.0550.0180.032 SIM1_60.0470.0500.0340.0540.0170.031 SIM1_50.2420.0610.1780.0540.1000.031 SIM1_80.0320.0340.0240.0370.0120.021 SIM1_110.0350.0320.0250.0380.0120.019 SIM1_120.0390.0440.0290.0530.0140.029 SIM1_130.0370.0520.0250.0650.0120.038 SIM1_90.0400.0440.0290.0570.0130.036 SIM2_160.4090.0370.3070.0330.1890.015 SIM2_150.3100.0300.2270.0310.1380.016 SIM2_140.2330.0360.1690.0460.1030.024 SIM2_20.2160.0370.1600.0390.0960.020 SIM2_30.0870.0510.0640.0570.0360.032 SIM2_40.0260.0290.0180.0340.0100.017 SIM2_50.0240.0260.0180.0290.0090.015 SIM2_5B0.0660.0400.0500.0460.0280.026 SIM2_60.0530.0420.0390.0470.0220.029 SIM2_6B0.0670.0400.0490.0450.0270.025 SIM2_80.0400.0470.0300.0530.0160.031 SIM2_90.0280.0410.0200.0490.0110.028 SIM2_100.0250.0340.0190.0390.0110.021 SIM2_70.0070.0240.0050.0300.0030.017 SIM2_180.0100.0280.0070.0340.0040.020 SIM3_10.7540.0090.5450.0350.3240.035 SIM3_20.2380.0900.1860.0770.1240.030 SIM3_30.2020.0460.1430.0910.0900.058 SIM3_110.2960.4450.2060.5500.1380.294 SIM3_40.0150.0240.0090.0380.0070.024 SIM3_50.0350.0380.0230.0500.0130.031 SIM3_60.0720.0050.0510.0210.0240.019 SIM3_80.0150.0350.0090.0420.0060.025 SIM3_90.0240.0540.0140.0700.0090.041 SIM3_100.0030.0190.0020.0240.0020.015

PAGE 90

76Table 11. (Continued) Station ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) SIM3_70.0780.2230.0530.2730.0300.159 SIM4_1_10.7990.0650.5760.0730.3430.035 SIM4_1_20.1230.0330.0870.0400.0500.023 SIM4_20.1060.0190.0740.0230.0410.013 SIM4_31.0290.0970.7540.1130.4540.071 SIM4_110.0950.0260.0690.0340.0400.017 SIM4_40.0380.0220.0260.0270.0120.015 SIM4_50.0650.0370.0440.0460.0220.027 SIM4_60.1250.0250.0890.0320.0490.018 SIM4_80.0440.0290.0330.0350.0190.014 SIM4_90.0160.0440.0120.0520.0060.027 SIM4_100.0160.0140.0100.0230.0060.011 SIM4_70.0330.0660.0230.0820.0120.046 SIM5_10.1030.0550.0730.0670.0420.036 SIM5_20.0900.0450.0610.0630.0340.035 SIM5_30.3620.0460.2600.0570.1530.033 SIM5_110.0500.0900.0360.0920.0200.051 SIM5_40.0170.0240.0100.0280.0050.017 SIM5_50.0060.0190.0030.0260.0020.014 SIM5_60.0570.0260.0410.0330.0220.020 SIM5_80.0520.0210.0370.0270.0210.016 SIM5_90.0450.0300.0310.0370.0160.019 SIM5_100.0080.0180.0050.0220.0020.009 SIM5_120.0190.0090.0150.0160.0090.007 SIM5_70.0090.0780.0070.0990.0030.057 SIM6_10.5580.0440.4070.0460.2460.031 SIM6_20.5340.0880.4050.0700.2500.038 SIM6_30.3270.0760.2410.0670.1450.042 SIM6_110.0580.0510.0390.0550.0200.032 SIM6_200.0320.0430.0230.0480.0110.029 SIM6_40.0370.0350.0260.0400.0130.023 SIM6_50.0260.0330.0170.0350.0070.021 SIM6_60.0310.0290.0200.0340.0090.020 SIM6_80.0760.0650.0550.0700.0290.040 SIM6_90.0310.0180.0230.0190.0110.011 SIM6_100.0060.0110.0040.0130.0020.007 SIM6_120.0040.0140.0030.0150.0010.009 SIM6_70.0040.0160.0030.0200.0010.011

PAGE 91

77Table 12. Mean values of phytoplankton and detritus absorption at 440 nm Cruise GOP SECGOP SEC SIM10.0350.0510.0550.051 SIM2 0.127 0.0240.0390.043 SIM30.1850.0260.1400.086 SIM40.2330.0330.0510.045 SIM50.0740.0230.0550.039 SIM60.1650.0180.0520.028 ad(440) (m-1) aph(440) (m-1)

PAGE 92

78 Figure 30. Phytoplankton specific absorption coefficient (aph*) spectra 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM2_3 SIM2_4 SIM2_5 SIM2_5B SIM2_6 SIM2_6b SIM2_7 SIM2_18 SIM2_8 SIM2_9 SIM2_10 SIM2_2 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM1_6 SIM1_7 SIM1_8 SIM1_12 SIM1_13 SIM1_9 SIM1_11 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM3_11 SIM3_4 SIM3_5 SIM3_7 SIM3_8 SIM3_9 SIM3_10 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM4_2 SIM4_11 SIM4_4 SIM4_5 SIM4_6 SIM4_7 SIM4_9 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM5_1 SIM5_2 SIM5_3 SIM5_11 SIM5_4 SIM5_6 SIM5_8 SIM5_9 SIM5_10b 0 002 004 006 008 01 012 014 016 400450500550600650700 wavelength (nm)aph* (m2 mg-1) SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_7 SIM6_8 SIM6_9

PAGE 93

79Table 13. Specific absorption coefficien t of phytoplankton at 440 nm (aph*(440)) Station aph* (440) (m2 mg-1) Station aph* (440) (m2 mg-1) SIM1_70.015SIM4_1-20.064 SIM1_60.017SIM4_20.033 SIM1_50.020SIM4_30.029 SIM1_40.073SIM4_40.015 SIM1_80.013SIM4_50.016 SIM1_110.022SIM4_60.021 SIM1_120.019SIM4_70.021 SIM1_130.018SIM4_80.031 SIM1_90.017SIM4_90.033 SIM2_20.021SIM4_100.053 SIM2_30.025SIM4_110.014 SIM2_40.034SIM5_10.036 SIM2_5b0.013SIM5_20.039 SIM2_50.024SIM5_30.037 SIM2_6b0.016SIM5_40.040 SIMIM2_60.019SIM5_50.036 SIM2_70.028SIM5_60.039 SIM2_80.035SIM5_70.052 SIM2_90.043SIM5_80.034 SIM2_100.029SIM5_90.043 SIM2_140.032SIM5_100.031 SIM2_150.020SIM5_110.068 SIM2_160.023SIM5_120.065 SIM2_180.034SIM6_10.030 SIM3_10.011SIM6_200.034 SIM3_20.028SIM6_20.035 SIM3_30.033SIM6_30.035 SIM3_40.016SIM6_40.025 SIM3_50.030SIM6_50.037 SIM3_60.018SIM6_60.032 SIM3_70.050SIM6_70.070 SIM3_80.023SIM6_80.030 SIM3_90.026SIM6_90.012 SIM3_100.086SIM6_100.051 SIM3_110.039SIM6_110.025 SIM4_1-1a0.006SIM6_120.046 SIM41-1b0.052

PAGE 94

80 Figure 31. Relationship between the speci fic absorption coefficient (aph*) at 440 nm and chlorophyll-a during the dry (A) and wet (B) season 0.000 0.020 0.040 0.060 0.080 0.100 0.120 0.1110 Chlorophyll-a (g l-1)aph*(440) (m2 mg-1) SIM1 SIM3 SIM5 0.000 0.020 0.040 0.060 0.080 0.100 0.1200.1110Chlorophyll-a (g l-1)aph*(440) (m2 mg-1) SIM2 SIM4 SIM6A B

PAGE 95

81Figure 32. Relationship between ph ytoplankton absorption (aph) at 440 nm and chlorophyll-a concentration y = 0.0434x0.5961 R2 = 0.5777 0.000 0.100 0.200 0.300 0.400 0.500 0.600 0246810 Chlorophyll-a (g l-1)aph(440) (m-1)

PAGE 96

82CDOM ( Gelbstoff) Absorption ( ag) Figures 33 to 38 show ag() spectra for some of the stations sampled during the cruises. Changes in the magnit ude of the spectra indicate ch anges in the concentration of CDOM in the water. The small shoulder obs erved around 265 nm is attributed to the presence of organic rings of purine a nd pyrimidine (Yentsch and Reichert, 1962), ag() spectra become exponential in the near-UV to visibl e range (350-700) (Bricaud et al., 1981). In general, minimum ag() values were found outside the GOP, while maximum values were found near the Delta. SIM5 and SI M6 were the only two cruises that showed some contrasting seasonal and spa tial distribution. Higher values of ag() were observed during SIM6 during the wet season than duri ng SIM5 (dry season), and in each case a gradient of ag() decreasing with increasing latitude was also observed. CDOM in this area should enter the o cean primarily from two sources, namely terrestrial sources and autochthonous phytoplankton degradation in situ. Because of the strong inverse correlated with salinity fo r the Orinoco plume dispersal region, during both seasons (Figure 39), we may conclude that the autoch thonous production is not very important. A steeper slope in the salinity-ag relationship was observed during the dry season compared to the wet season. Such conservative behavior has been observed previously in the Orinoco River plume as in many other co astal regions (Blough et al., 1993; Vodacek et al., 1997; Del Castillo et al., 1999; Twardowski and Donaghay, 2002; Hu et al., 2003). Deviations from this relationship are usually related to an increase in the concentration of CDOM from biogenic origin, to photo-oxidati on, or to flocculation of organic matter. However, no evidence for e ither process was observed in this study. Values of ag in the GOP and SEC have been previously reported mainly at 300nm (Blough et al., 1993; Del Castillo et al., 1999). Similar values to those obtained in this study during the wet season were reported by Del Castillo et al. 1999 for the same season (Table 14). The spectral slope of ag, S, is commonly used as a proxy or index of changes in the composition of CDOM. Variation in S may result from differ ences in the proportions of humic to fulvic acids (Carder et al., 1989) or by chemical modi fications of CDOM (Del

PAGE 97

83 Castillo et al., 1999). Mean values of S found during th e wet season ranged from 0.014 to 0.015 nm-1 in the GOP and from 0.014 nm-1 to 0.016 nm-1 in the SEC. During the dry season, mean values of S ranged from 0.013 nm-1 to 0.015 nm-1 in the GOP and from 0.014 nm-1 to 0.020 nm-1 in the SEC. SIM5, which was the cruise with the highest salinity values, was the cruise with the broadest rang e of S values observed, namely from 0.012 nm-1 in the GOP to 0.020 nm-1 in the SEC (Table 14). Figure 40 shows how S changes w ith salinity. Larger changes in S seemed to occur at salinities above 30, suggesting an alterati on in the composition of CDOM as it moved from the mouth of the Orinoco towards the Caribbean. Blough et al. (1993) and Del Castillo et al. (1999) made similar observations where they attribute the weak relationship of S and salinities below 30 to the fact that there were no significant changes in the optical properties of CDOM in the river plume closer to shore. Note that all salinity values above 30 shown in Figure 40 duri ng the dry season (SIM5). Values of c(660) indicate that surface turbidity was lower duri ng SIM5 than during any of the other cruise, therefore, it is reasonable to think that photodegradation c ould be one of the processes responsible for the changes obser ved in the spectral slope and therefore in the bio-optical properties of CDOM during this cruise. Because 440 nm is the wavelength at which phytoplankton absorption is near its maximum, it is important to study the absorp tion of light by CDOM at this wavelength since this could seriously bias any bio-opti cal algorithm used to estimate chlorophyll-a. Values of ag(440) are presented in Table 14. During the dry season, ag(440) ranged from 0.415 m-1 to 2.59 m-1 ( = 1.5 m-1) in the GOP, and from 0.23 m-1 to 2.13 m-1 ( = 1.11 m-1 ) in the SEC. During the wet season, ag(440) ranged from 0.40 m-1 to 3.21 m_1 ( = 1.58 m-1) in the GOP, and from 0.38 m-1 to 1.54 m-1 ( = 0.8 m-1) in the SEC. The maximum value of ag(440) was observed during SIM4 at a station near Serpent’s Mouth (SIM4_2) while the minimum value was observe d during SIM5 at a station in the SEC (SIM5_10). In general, SIM5 was th e cruise with lower values of ag(440), which indicates that this was the cruise with the lowest concentration of CDOM. This cruise

PAGE 98

84 was carried out in March, which is the month during which the Orinoco river water discharge is usually at its minimum (Figure 2). Figure 33. Spatial variability of of ag() for some of the stations during SIM1

PAGE 99

85 Figure 34. Spatial variability of of ag() for some of the stations during SIM2

PAGE 100

86 Figure 35. Spatial variability of of ag() for some of the stations during SIM3

PAGE 101

87 Figure 36. Spatial variability of of ag() for some of the stations during SIM4

PAGE 102

88 Figure 37. Spatial variability of of ag() for some of the stations during SIM5

PAGE 103

89 Figure 38. Spatial variability of of ag() for some of the stations during SIM6

PAGE 104

90 Figure 39. Relationship between Gelbstoff absorption at 440 nm a nd Salinity during dry (A) and wet (B) seasons 0 0.5 1 1.5 2 2.5 3 3.5 010203040 Salinity (PSU)ag(440) (m-1) SIM1 SIM3 SIM5A 0 0.5 1 1.5 2 2.5 3 3.5 0102030 Salinity (PSU)ag(440) (m-1) SIM2 SIM4 SIM6B

PAGE 105

91Table 14. Gelbstoff absorption at 440 nm (ag(440)), at 300 nm (ag(300)), and the spectral slope (S) (Slope was measured between 270 nm and 450 nm) Station ag(440) (m-1) ag(300) (m-1) Slope (nm-1) Station ag(440) (m-1) ag(300) (m-1) Slope (nm-1) SIM1_71.148.300.014SIM3_91.178.330.015 SIM1_62.3810.330.010SIM4_1_11.028.160.015 SIM1_50.714.950.014SIM4_1_21.508.560.014 SIM2_161.829.990.013SIM4_23.2118.960.013 SIM2_151.439.570.014SIM4_33.189.470.013 SIM2_141.419.540.014SIM4_111.059.020.015 SIM2_21.5211.590.014SIM4_41.169.060.015 SIM2_31.008.990.015SIM4_51.0710.500.015 SIM2_41.2510.160.015SIM4_81.1119.350.015 SIM2_5B1.137.660.015SIM5_22.593.450.012 SIM2_50.407.450.018SIM5_11D1.102.110.015 SIM2_60.773.800.016SIM5_40.411.930.017 SIM2_80.669.000.016SIM5_50.5614.320.015 SIM2_90.457.520.017SIM5_80.253.150.019 SIM2_100.473.310.016SIM5_100.238.400.020 SIM3_12.2213.060.013SIM6_22.7016.020.013 SIM3_21.6210.310.014SIM6_32.1210.810.013 SIM3_31.247.990.014SIM6_111.4312.630.014 SIM3_111.8311.650.013SIM6_201.6810.620.014 SIM3_42.4814.660.013SIM6_41.4411.350.014 SIM3_52.1413.420.013SIM6_61.5410.320.014 SIM3_61.6710.270.013SIM6_101.007.210.014 SIM3_81.6311.220.014

PAGE 106

92 Figure 40. Relationship between Gelbstoff absorption (ag()) spectral slope and Salinity during dry (A) and wet (B) seasons 0.000 0.005 0.010 0.015 0.020 0.025 010203040 Salinity (PSU)ag Slope (nm-1) SIM1 SIM3 SIM5A 0.000 0.005 0.010 0.015 0.020 0.025 010203040 Salinity (PSU)ag Slope (nm-1) SIM2 SIM4 SIM6B

PAGE 107

93Continuous, Along-Track (flow-through) Measurements Continuous, along-track measurements were collected during SIM6 in October 2000. These measurements show the spatial distribution of surface temperature, salinity, CDOM fluorescence, CHL fluorescence (Figure 41), total absorption, a(),total attenuation, c(), and total scattering, b() (by difference) coefficients at nine different wavelengths (Figure 42). Temperature values ranged from 28.70 C to 29.92 C ( = 29.51C) in the GOP and from 28.49 C to 29.21 C ( = 28.86 C) in the SEC. Salinity values ranged from 9.81 to 32.32 ( = 18.56) in the GOP and from 24.84 to 32.32 ( = 30.70) in the SEC. Relative values of CDOM fluorescence were always higher in the GOP than in the SEC. The lowest CDOM values observed to the north of the Paria Peninsula occurred around the upwelling focus station, SIM6_7. Relative values of chlorophyll fluorescence were also higher in the GOP than in the SE C. Excitation/emission signal of CDOM varies according to its composition (Blough and De l Vecchio, 2002); therefore, the source, photodegradation and mixing with marine wate rs possibly had confounding effects in the CDOM fluorescence signal measured. Patches of high chlorophyll fluorescence were frequently observed in the GOP. A marked decrease in both chlorophyll and CDOM fluorescen ce was observed around station SIM6_20, located approximately in the center of the GOP, and also north of the Paria Peninsula in the SEC. Changes in CDOM fluorescence and chlorophyll fluorescence were both related to changes in salinity, although clearly this relation was stronger for CDOM than for chlorophyll fl uorescence (Figure 43). This is the only evidence observed on this study in which CDOM and chlorophyll-a seem to have a close correlation. Also, both CDOM and chlorophyl l fluorescence showed a strong relation to changes in total absorption at 440 nm (Fi gure 44). This indicates that even though CDOM dominates the total absorption at 440nm, the cont ribution from phytoplankton still has a significant impact on tota l absorption at this wavelength. Minimum, maximum, and mean values of total a(), b(), and c() for the GOP and SEC are presented in Table 15. According to the cruise mean values, at the blue bands

PAGE 108

94 (412nm, 440nm, and 488m) light attenuation was about three times higher in the GOP than in the SEC. At longer wavelengths (510 nm, 532 nm, 555 nm, 650 nm, 676 nm), light attenuation was about two times higher in the GOP than in the SEC. In both the SEC and GOP, b() dominated the light attenuation at all wavelengths. Both a() and b() showed a spectral dependency, with values decreasing with increasing wavelength (Figure 45). Higher values of a(650) and a(676) were due to the increase in water and phytoplankton absorption at these wavelengths. The spectral dependence of a() between 412 nm and 555 nm can be explained by the dominance of ag() over ap() (Figure 46) throughout the st udy region. Contributions of ag(440) to total absorption ranged from 60.95% to 97.89% (Table 16). Spectral dependence of b() has been observed previously in case 1 and case 2 waters by Gould et al. (1999) who reported a linear relationship of b versus wavelength, with the slope progressively decreasing from turbid waters dominated by suspended sediments to clear waters dominated by phytoplankton.

PAGE 109

95 Figure 41. Along-track measurements of surface Temperature (C), Salinity (psu), CDOM-fluorescence (relative values), and Chlorophyll-fluorescen ce (relative values) starting from the GOP near Serpent’ s Mouth northward towards the SEC GOP-Serpent’s Mouth SEC GOP-Serpent’s Mouth SEC

PAGE 110

96 Figure 42. Along-track measurements of total attenuation (c), total absorption (a), and total scattering (b) coefficients, starting from the GOP near Serpent’s Mouth northward towards the SEC GOP-Serpent’s Mouth SEC GOP-Serpent’s Mouth SEC Wavelengths: 412 nm, 440 nm, 488 nm, 510 nm, 532 nm, 555 nm, 650 nm, 676 nm, 715 nm

PAGE 111

97 Figure 43. Relationship between CDOM and Chlor ophyll fluorescence with salinity from along-track measurements -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4s6_19 s6_1 s6_2 s6_3 s6_11 s6_20 s6_4 s6_5 s6_6 s6_7 s6_8 s6_9 s6_10 s6_12Stations Fluorescence (relative values)0 10 20 30 40Salinity (PSU) CDOM fluorescence CHL Fluorescence Salinity R2 = 0.96 R2 = 0.82 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 101520253035 Salinity (PSU)Fluorescence (relative values)

PAGE 112

98 Figure 44. Relationship between CDOM and Ch lorophyll fluorescence with total absorption coefficient at 440 nm from along-track measurements -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4s6_19 s6_1 s6_2 s6_3 s6_11 s6_20 s6_4 s6_5 s6_6 s6_7 s6_8 s6_9 s6_10 s6_12StationsFluorescence (relative values)0 0.5 1 1.5 2 2.5 3Total absorption coefficient at 440 nm (m-1) CDOM Fluorescence CHL fluorescence a(440) R2 = 0.95 R2 = 0.84 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 00.511.522.53 Total absorption coefficient at 440 nm (m-1)fluorescence (relative values)

PAGE 113

99Table 15. Minimum, maximum, and mean values of surface total absorption (a), scattering (b) and beam attenuation (c) coefficients GOPSECGOPSECGOPSEC min0.150.120.350.070.530.32 412 nmmax4.721.5910.091.6613.942.95 mean1.610.551.630.613.241.15 min0.100.040.320.130.460.27 440 nmmax3.261.069.701.7312.372.49 mean1.060.361.570.612.630.97 min0.050.030.320.140.390.23 488 nmmax1.780.528.891.6910.382.05 mean0.530.181.460.592.000.78 min0.050.030.310.130.380.23 510 nmmax1.420.438.571.689.752.01 mean0.420.161.420.591.850.74 min0.040.030.300.130.370.23 532 nmmax1.110.348.291.659.231.90 mean0.340.141.390.581.730.72 min0.040.030.290.140.360.23 555 nmmax0.930.297.981.608.701.80 mean0.270.121.350.571.620.69 min0.310.290.260.110.600.47 650 nmmax0.680.416.931.517.441.89 mean0.380.351.200.531.580.88 min0.430.390.250.120.700.59 676 nmmax1.360.556.661.537.222.04 mean0.490.471.160.521.640.99 a ( ) (m-1) b ( ) (m-1) c ( ) (m-1)

PAGE 114

100 Figure 45. Spectral dependence of total absorption (a) and total scattering (b) coefficients 0 2 4 6 8 10 400450500550600650700750 Wavelength (nm)a (m-1) 0 2 4 6 8 10 400450500550600650700750 Wavelength (nm)b (m-1) SIM6_1 SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_5 SIM6_6 SIM6_7 SIM6_8 SIM6_9 SIM6_10 SIM6_12

PAGE 115

101 Figure 46. Gelbstoff (yellow), Detritus (blue) an d phytoplankton (green) absorption contributions to total absorption at 440 nm 0% 20% 40% 60% 80% 100%Percentage SIM4_1_1 SIM4_1_2 SIM4_2 SIM4_3 SIM4_11 SIM4_4 SIM4_5 SIM4_8Stations 0% 20% 40% 60% 80% 100%Percentage SIM6_2 SIM6_3 SIM6_11 SIM6_20 SIM6_4 SIM6_6 SIM6_10Stations 0% 20% 40% 60% 80% 100%Percentage SIM1_07SIM1_06SIM1_05 Stations 0% 20% 40% 60% 80% 100%Percentage SIM2_16 SIM2_15 SIM2_14 SIM2_2 SIM2_3 SIM2_4 SIM2_5B SIM2_5 SIM2_6 SIM2_8 SIM2_9 SIM2_10Stations 0% 20% 40% 60% 80% 100%Percentage SIM5_2 SIM5_11D SIM5_4 SIM5_5 SIM5_8 SIM5_10Stations 0% 20% 40% 60% 80% 100%Percentage SIM3_1 SIM3_2 SIM3_3 SIM3_11 SIM3_4 SIM3_5 SIM3_6 SIM3_8 SIM3_9Stations

PAGE 116

102Table 16. Gelbstoff, detritus, and phytoplankton contribu tions to total absorption at 440 nm Station ag(440) % ad(440) % aph(440) % Station ag(440) % ad(440) % aph(440) % SIM1_792.172.894.43SIM3_992.801.145.56 SIM1_696.171.382.20SIM4_1_160.9534.304.37 SIM1_574.7918.825.72SIM4_1_291.885.302.43 SIM2_1684.0214.161.53SIM4_296.882.230.70 SIM2_1584.4213.411.80SIM4_378.4918.582.77 SIM2_1486.4210.372.82SIM4_1190.575.922.96 SIM2_288.139.242.26SIM4_495.152.152.18 SIM2_388.695.675.08SIM4_591.683.803.98 SIM2_495.531.412.58SIM4_893.712.812.95 SIM2_5B91.714.073.71SIM5_295.202.262.30 SIM2_588.423.856.33SIM5_11D89.202.887.41 SIM2_689.294.505.48SIM5_490.262.226.13 SIM2_888.033.997.13SIM5_590.262.226.13 SIM2_985.663.859.30SIM5_894.130.524.29 SIM2_1087.873.547.39SIM5_1087.282.068.26 SIM3_179.0919.441.24SIM6_284.8612.742.20 SIM3_285.769.854.05SIM6_387.089.902.76 SIM3_383.739.686.15SIM6_1193.442.583.57 SIM3_1170.597.9521.21SIM6_2095.581.302.75 SIM3_497.890.371.50SIM6_495.241.712.63 SIM3_596.381.062.28SIM6_696.221.282.11 SIM3_695.562.891.19SIM6_1097.760.371.25 SIM3_896.590.532.50

PAGE 117

103Light Field Subsurface Measurements Underwater profiles of downwelling irradiance, Ed(), were used to derive the diffuse attenuation coefficient, Kd(), at the spectral bands used by SeaWiFS. Kd() relates the spectral irradiance at any depth, Ed(,z), to the irradiance just above it. If Kd() is a constant with depth, then downwelling irradiance at any depth can be related to that just beneath the sea surface, Ed(,0-) (Farmer et al., 1993): Ed(,z) = Ed(,0-)exp-[Kd()z] In general, Kd() decreased with increasing wavelength from 412 nm to 555 nm increasing again for the 665nm and 683 nm bands. In the GOP, Kd() values at blue wavelengths were about 6 times higher duri ng SIM3 and SIM4 than during SIM5, and about two times higher than during SIM1 and SIM2. In the SEC Kd() values remained very similar for all the cruises (with exception of SIM3), where Kd() values in the blue wavelengths were about 6 times lower than in the GOP (Figure 47). Depths of maximum light penetration (zmax) were estimated for the GOP and the Orinoco plume in the SEC by pooling observations per cruise. The depth was estimated using the optical depth () at which surface Ed() is reduced to 1% ( = 4.605). Farmer et al. (1993) reported an increase of 15 nm to 90 nm in the wavelengt h of maximum penetrati on (mainly from blue to yellow), for both the SEC and GOP duri ng the wet season. Such seasonal shift in wavelength was not observed in this study. Wavelengths of maximum penetration (532 and 555 nm) and minimum penetration (412 and 443 nm) remained the same for all the stations during both seasons (Figure 48). In the GOP, light at the blue and red wavelengths was attenuated to 1% in the first 5-10 m of the water column. Green light penetrated the deepest, but it also was exti nguished at about 15 m. SIM5 was the cruise with higher depths of lig ht penetration (> 15 m). Plume waters in the SEC were somewhat more transparent than in the GOP. There was an increase in zmax to up to10 m for the blue wave lengths and up to 27 m for the green wavelengths,

PAGE 118

104 while that for the red wavelengths remain ed similar to those observed in the GOP (~10m). Individual values of Kd and zmax are shown in Table 17. Figure 47. Mean spectral values of the diffuse attenuation coefficient (Kd) in the GOP (A) and SEC (B) 0.00 1.00 2.00 3.00 4.00 412443490532555665683 Wavelength (nm)Kd (m-1) 0.00 1.00 2.00 3.00 4.00 412443490532555665683 Wavelength (nm)Kd (m-1) SIM1 SIM2 SIM3 SIM4 SIM5A B

PAGE 119

105 Figure 48. Mean values of the euphotic depth (zmax) (1% of subsurface irradiance) in the GOP (A) and SEC (B) 0.00 10.00 20.00 30.00 412443490532555665683 Wavelength (nm)zmax (m) A 0.00 10.00 20.00 30.00 412443490532555665683 Wavelength (nm)zmax (m) SIM1 SIM2 SIM3 SIM4 SIM5B

PAGE 120

106Table 17. Diffuse attenuation coefficient (Kd) and depth of maximum penetration (zmax) per station Stations K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) SIM1_72.132.171.193.860.726.400.548.540.489.650.785.900.815.68 SIM1_5b0.509.300.3712.280.2617.940.2221.140.2122.220.607.740.627.39 SIM1_40.2320.140.2023.050.1825.530.2121.520.2221.340.617.500.657.08 SIM1_80.676.850.4210.890.2320.060.1726.380.1629.000.499.400.528.82 SIM1_110.499.380.3313.860.2121.770 .1825.330.1826.180.509.280.538.70 SIM1_120.825.630.548.540.3214.370.2419.000.2221.270.568.250.597.79 SIM1_130.766.060.499.380.2816.730.2022.630.1825.130.568.260.597.75 SIM1_90.3214.580.2122.040.1335.610.1237.900.1237.870.519.090.548.45 SIM2_161.373.361.183.910.785.880.607.640.548.580.756.130.785.90 SIM2_150.994.660.994.650.637.280.479.770.4211.060.647.210.676.89 SIM2_141.253.691.064.330.617.590.4 111.270.3413.460.548.480.578.08 SIM2_21.752.631.493.090.895.170.637290.548.500.706.540.736.31 SIM2_31.902.421.223.770.696.670.479.800.4011.430.597.750.627.40 SIM2_41.493.090.964.810.528.890.3513.290.3015.540.509.160.538.65 SIM2_5b0.3114.720.607.710.3812.190.3015.360.2816.640.607.670.657.13 SIM2_51.592.900.984.720.528.930.3513.240.2915.710.538.760.558.36 SIM2_6b0.4210.920.2916.110.1727.500.1432.900.1334.520.489.510.528.89 SIM2_60.2717.150.2320.260.1727.150.1826.100.1924.580.647.240.706.61 SIM2_180.1726.610.1334.720.1047.680.1142.440.1238.390.4510.330.479.78 SIM2_170.726.370.509.180.3214.360.2617.490.2518.180.597.860.627.46 SIM2_81.572.931.034.470.627.480.4610.070.4210.920.785.940.825.64 SIM2_90.845.460.519.080.2518.190.1726.540.1531.310.4310.610.469.97 SIM2_100.736.330.4610.020.2617.890 .1924.440.1826.170.509.210.538.67 SIM3_13.391.362.351.961.443.191.034.460.875.270.954.830.974.76 SIM3_24.031.142.651.732.282.021.872.471.323.501.074.300.974.75 SIM3_113.021.532.112.181.273.620.934.960.885.251.084.251.193.88 SIM3_53.351.382.082.221.124.110.766.040.617.520.726.380.736.28 SIM3_62.262.041.383.340.666.990.3214.500.1825.220.3114.930.3114.72 SIM3_71.074.310.726.430.3214.400.1924.580.1530.310.528.910.627.48 SIM3_91.303.550.746.200.2221.400.1047.600.1237.970.2717.300.2122.20 SIM3_100.3314.030.2915.850.2518.070.2518.780.2419.070.607.650.666.98 SIM4_1_14.381.053.161.462.142.161.582.911.373.371.343.431.343.43 SIM4_1_21.403.280.885.200.499.340.3513.040.3313.940.558.320.617.58 SIM4_21.363.370.875.310.4510.190.3114.970.2716.910.548.460.587.87 SIM4_38.000.585.550.833.651.262.621.762.202.101.802.561.762.61 SIM4_41.732.661.163.960.686.820.469.990.3812.100.627.440.587.96 SIM4_51.862.481.243.710.716.470.509.250.4210.950.666.990.686.80 SIM4_60.865.370.597.770.3712.480.3015.430.2816.460.637.360.667.02 SIM4_70.2518.520.2220.810.1628.480.1531.640.1530.210.568.260.588.00 SIM481.582.910.994.650.587.980.4310.790.4410.530.666.980.568.28 555 nm665 nm683 nm 412 nm443 nm490 nm532 nm Continued on the next page

PAGE 121

107Table 17. (Continued) Stations K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) K d (m-1) zmax (m) SIM4_90.716.520.479.840.3114.870.2716.960.2716.780.558.380.568.28 SIM4_100.4111.220.2419.530.1045.330.0948.480.1431.890.568.180.568.22 SIM5_10.875.300.637.360.4111.270.3214.570.2915.890.578.030.607.63 SIM5_20.676.850.479.810.3015.600.2518.720.2419.180.558.440.568.22 SIM5_30.954.820.676.920.4510.140.3712.380.3513.220.597.840.558.38 SIM5_110.815.670.509.130.2419.520.1433.750.1046.330.4011.440.4011.39 SIM5_40.499.370.3413.560.2022.840.1727.280.1628.190.479.740.489.68 SIM5_50.3911.750.2419.420.1825.470.1726.800.1924.370.568.170.637.31 SIM5_60.3214.310.2618.020.1824.970.1726.640.1826.190.538.640.568.23 SIM5_90.657.110.479.800.3214.290.2915.980.2915.690.667.030.696.71 SIM5100.548.500.3712.280.2518.250 .2418.860.2618.000.726.380.805.79 555 nm665 nm683 nm 412 nm443 nm490 nm532 nm

PAGE 122

108Surface Reflectance Measurements Large differences in Rrs() spectra were observed among th e stations near Serpent’s Mouth, the stations near Dra gon’s Mouth and the stations in the SEC. Specifically, the following patterns were observed: GOP-Serpent’s Mouth: During both seasons, Rrs() spectra in Serpent’s Mout h were characterized by low values in the blue and high values between 500 nm and 680 nm, with three distinctive peaks, namely 576-584 nm, 620-628 nm, and 660-684 nm (Figure 49). Maximum Rrs() values (Rrsmax) usually occurred at about 580 nm ranging from 0.0037 sr-1 (SIM3_11 at 580 nm) to 0.0234 sr-1 (SIM4_1-1b at 584 nm). Rrs(440) ranged from 0.0012 sr-1 (SIM3_11) to 0.0120 sr-1 (SIM5_3). An increase in Rrs() at the blue wavelengths, a shift of Rrsmax wavelength from ~580 nm to 564 nm, and a more pronounced decrease of Rrs values between 580 nm and 600 nm were obser ved during SIM5 near Serpent’s Mouth. Patchiness in this region was visible during two of the crui ses and was detected in the Rrs() measurements. Two Rrs() spectra collected around station 11 during SIM3 showed different spectral feat ures. One of them (SIM3_11R) showed features that are very characteristic of red tides. While during SIM4, a patch of lower Rrs values with Rrsmax below 0.010 sr-1 (SIM4_1-2, SIM4_2a, and SIM4_2b), was observed between stations SIM4_1_1b and SIM4_3 for which Rrsmax were above 0.020 sr-1. GOP-Dragon’s Mouth: With exception of SIM5, Rrs() spectra observed in and near Dragon’s Mouth were similar during the dry and wet season (Fi gure 50). A maximum peak was usually observed between 568 nm and 572 nm, while a second distinctive p eak was observed in the spectral region between 660 nm and 684 nm. The shoulder between 620 nm 640 nm which was very clear in the Rrs() spectra from Serpent’s Mouth was not marked in data from Dragon’s Mouth. This emphasized a pronounced decrease of Rrs()between 580 nm and 600 nm in the Rrs() spectra. In the GOP, Rrsmax values ranged from 0.0009 sr-1 (SIM2_4 at 568 nm) to 0.0034 sr-1 (SIM1_7 at 568 nm). For the secondary peak Rrs values ranged from 0.0004 sr-1 (SIM2_4

PAGE 123

109 at 660 nm) to 0.0022 sr-1 (SIM1_7 at 680 nm). In the blue region, Rrs at 440 nm ranged from 0.0006 sr-1 (SIM2_4) to 0.0024 sr-1 (SIM1_7). The Rrs spectra measured during SIM5 were characterized by high values between 480 nm and 580 nm, with three distinctive p eaks in this spectral region around 504 nm, 544 nm, and 560 nm, and a fourth peak around 680 nm. Rrsmax at 544 nm and the peak around 680 nm were very similar for both stations, namely ~0.0025 sr-1 and ~ 0.001 sr-1 respectively. The main differences between the spectra from SIM5 and those from the other cruises were a shift of the Rrsmax wavelength from ~ 572 nm to 544 nm, and an increase in Rrs values between 440 nm and 520 nm. Southeastern Caribbean (SEC): Rrs() spectra in the SEC were highly variab le in terms of shape and magnitude (Figure 51). Wavelengths of Rrsmax ranged from 488 nm to 568 nm, while Rrsmax values ranged from 0.0016 sr-1 (SIM6_10 at 500 nm) to 0.0092 sr-1 (SIM5_6 at 544 nm). Rrs values at 440 nm ranged from 0.0008 sr-1 (SIM4_8) to 0.0034 sr-1 (SIM4_7), and Rrs values at 680 nm ranged from 0.0004 sr-1 (SIM6_10) to 0.0036 sr-1(SIM3_7). The Rrs spectrum observed in station SIM5_6 (Rrsmax at 544 nm) was quite unique and not reproduced at any other station or crui se in the SEC, however it was very similar to spectra observed in the GOP near Dragon’s Mouth during the same cruise. Unfortunately, there was only one Rrs measurement available fo r the stations outside of the GOP during SIM5.

PAGE 124

110 Figure 49. Rrs() spectra for stations in the GOP near to Serpents Mouth 0000 0005 0010 0015 0020 0025400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM6_1 SIM6_2 SIM6_3 0000 0005 0010 0015 0020 0025400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM4_1-1B SIM4_3 SIM4_1-2 SIM4_2A SIM4_2B 0000 0005 0010 0015 0020 0025400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM3_1 SIM3_2 SIM3_11 SIM3_11R 0 0005 001 0015 002 0025400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM5_1 SIM5_2 SIM5_3 SIM5_11 0000 0005 0010 0015 0020 0025400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM2_16 SIM2_15 SIM2_14

PAGE 125

111 Figure 50. Rrs() spectra for stations in th e GOP near to Dragons Mouth 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM4_4 SIM4_5 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM2_4 SIM2_5b SIM2_5 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM1_7 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM3_4 SIM3_5 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM5_4 SIM5_5 0 0001 0002 0003 0004400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM6_4A SIM6_4B SIM6_5

PAGE 126

112 Figure 51. Rrs() spectra for stations in the SEC 0000 0002 0004 0006 0008 0010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM6_6b SIM6_8 SIM6_9 SIM6_10 SIM6_12 0000 0002 0004 0006 0008 0010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM4_7 SIM4_8 SIM4_9 SIM4_10 0000 0002 0004 0006 0008 0010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM2_18 SIM2_17 SIM2_8 SIM2_9 SIM2_10 SIM2_6 0.000 0.002 0.004 0.006 0.008 0.010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM1_8b SIM1_12 SIM2_13 SIM1_5b 0000 0002 0004 0006 0008 0010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM3_7 SIM3_10 SIM3_6 0000 0002 0004 0006 0008 0010400 440 480 520 560 600 640 680 720 760Wavelength (nm)Rrs (sr-1) SIM5_6

PAGE 127

113 The following spatial patterns in Rrsmax were observed, regardless of season: 1. Stations with wavelengths of Rrsmax between 576 nm and 584 nm were located in the GOP near Serp ent’s Mouth (10.00 N 10.20 N). 2. Stations with wavelengths of Rrsmax between 568 nm and 572 nm were located in the GOP near Dragon’s Mouth (10.50 N 10.70 N). 3. Stations with wavelengths of Rrsmax between 564 nm and 568 nm were located in the SEC (10.8 N 11.20 N). 4. Stations with wavelengths of Rrsmax between 488 nm and 504 nm were located offshore in the SEC (11.20 N 11.40 N). Remote-sensing reflectance (Rrs) is a func tion of the IOPs, namely the absorption and backscattering coefficients, in the form: () Rrs() ()()b bb cons ab ; therefore, spectral variability in Rrs() measurements, such as changes in magnitude and position of Rrsmax, can be explained by changes in these IOPs due to change s in the concentration and composition of particulate and dissolved substances. There were no bb() measurements available for this study. Therefore, the following explanation of the spectral changes observed in Rrs() is somewhat speculative. The data show that CDOM dominated the total absorption in the blue wavelengths in the GOP and SEC indistinctively of the season. In the GOP, during the dry season the mean contributions of ag(440), ad(440) and aph(400) to a(440) were 88.89%, 5.18%, and 5.42%, respectively. While during the wet season, th e mean contribution of ag(440) remained very similar when compared to the dry season (88.46%), the mean contribution of ad(440) increased to 8.13%, a nd the mean contribution of aph(440) decreased to 2.97%. In the SEC, during the dry season, these c ontributions were very similar to those observed in the GOP; however, during the wet season the mean contribution of ad(440) decreased to ~3% and th e mean contribution of aph(440) increased to ~5%. However, the contribution of ad(440) could reach values as high as 34% in the GOP during the wet

PAGE 128

114 season and aph(440) contribution may be as high as 20% in the GOP during the dry season. Mean values of ag(440) to aph(440) ratios ranged from 3.33 to 139.4 in the GOP and from 9.2 to 80.63 in the SEC. These ratios were normally higher in the GOP than in the SEC with the exception of SIM6 (Table 18) In the GOP the highest mean ratio was observed during SIM4 while the lowest one was observed during SI M5. Spatial changes observed in the ratio between ag(440) and aph(440) were mainly due to the spatial changes observed in ag(440). The ratios at 443 nm in our study area are much higher than those reported for the west Florida shelf during summer (Nelson and Guarda, 1995); there they varied from 10 nearshore to about 1, approximately 100 km offshore. Similarly, the ratios in the Mississippi River plume range d from 0.5 to 15 (Nababan, 2004 (personal communication)). According to Kopelevich (2002) SeaW iFS algorithms overestimate chlorophyll concentrations when ag(440)/ aph(440) > 2. Using this ratio as an indicator it could be expected then that SeaWiFS band ratio algor ithms would perform very poorly in the GOP and SEC, regardless of the season. In terms of seasonal and spatial variab ility, there were no significant seasonal changes observed in ag(440) in either the GOP or the SEC. A significant seasonal difference, however, was observed in aph(440) and ad(440) for the GOP and SEC (Table 19). However, the relative changes in ag(440) would have a greater impact in the seasonal spectral variability of a() and Rrs(). On the other hand, ag(440) presented significant spatial variability between the GOP and SEC. Spatial variation in aph(440) was most notable during the wet season (T able 20). Mean values of ag(440) and aph(440) were higher in the GOP than in the SEC, and ag(440) was found to be twice as high in the GOP than in the SEC. Therefore, in terms of spatial distribution, CDOM also have a greater contribution than phytoplankton to the spatia l changes observed in total absorption and therefore in Rrs(). The spatial patterns in c(660) and TSM (Tables 21 and 22) indicate that suspended particles play an important role in the spectral changes observed in Rrs(). Mean values

PAGE 129

115 of c(660) during both seasons were twice as high in the GOP than in the SEC, while the mean value of TSM during the dry season was five times higher in the GOP than in the SEC. Also, measurements with the ac-9 also show that total scattering, b(), was two to four times higher near Serpent’s mouth than in the rest of the GOP and in the SEC. The reduction in the concentration of TS M would result in a reduction of b, and bb() in the green to red channels. Therefore, spatial variability of TSM and CDOM should be the primary factors causing shifts in Rrsmax towards shorter wavelengths, going from Serpent’s Mouth to the stations in the SEC, indistinctively of the season.

PAGE 130

116Table 18. Ratio between the absorption coefficients of Gelbstoff and phytoplankton at 440 nm Bold letters indicate stations in the SEC Table 19. T-test results showing the m ean seasonal variability of ag(440), ad(440), and aph(440) in the GOP and SEC Station ag(440)/ aph(440) Station ag(440)/ aph(440) SIM1_720.81 SIM3_916.69 SIM1_643.76SIM4_1_113.95 SIM1_513.08 SIM4_1_237.80 SIM2_1654.85SIM4_2139.39 SIM2_1546.90SIM4_328.31 SIM2_1430.67SIM4_1130.60 SIM2_238.95SIM4_443.59 SIM2_317.46SIM4_523.04 SIM2_437.01 SIM4_831.81 SIM2_5B24.72SIM5_241.35 SIM2_513.96SIM5_11D12.04 SIM2_616.28 SIM5_414.71 SIM2_812.35 SIM5_514.71 SIM2_99.21SIM5_821.97 SIM2_1011.89SIM5_1010.57 SIM3_163.66SIM6_238.54 SIM3_221.16SIM6_331.59 SIM3_313.61SIM6_1126.16 SIM3_113.33SIM6_2034.73 SIM3_465.45SIM6_436.22 SIM3_542.30 SIM6_645.66 SIM3_680.63SIM6_1077.92 SIM3_838.59 GOPSECGOPSECGOPSEC Dry 1.6130.887 0.0560.055 0.1090.034 WET1.5770.858 0.0470.038 0.1730.025 t-value0.1230.1071.2801.135-1.0520.831 P 0.9040.9170.2130.2700.3010.414 Mean ag(440) (m-1) Mean aph(440) (m-1) Mean ad(440) (m-1)

PAGE 131

117Table 20. T-test results showing the mean spatial variability of ag(440), ad(440), and aph(440) Table 21. T-test results showing the mean seasonal and spatial variability of Total suspended solids (TSM) Table 22. T-test results showing the mean seasonal and spatial variability of the specific attenuation coefficient at 660 nm (c(660)) WETDRYWETDRYWETDRY GOP 1.5771.6130.0470.056 0.1730.109 SEC 0.858 0.887 0.0380.055 0.0250.034 t-value3.2782.1731.3870.1003.4051.742 P 0.0040.0460.1740.9220.0030.105 Mean ag(440) (m-1) Mean aph(440) (m-1) Mean ad(440) (m-1) GOPSECt-value P Dry 2.29 1.23 2.000.06 WET2.38 1.06 2.670.01 t-value-0.130.63 P 0.890.53 c (660) (m-1)Spatial Variability c (660) (m-1) Seasonal Variability GOPSECt-value P Dry 27.06 5.26 2.130.05 WET7.01 4.33 1.080.30 t-value1.911.03 P 0.080.31 Spatial Variability TSS (mg l-1) Seasonal Variability TSS (mg l-1)

PAGE 132

118Ocean Color Algorithms Deriving chlorophylla from an empi rical algorithm Figures 52 and 53, show the dispersal of the Orinoco River plume over the SEC for the months in which the SIMBIOS-Orinoco cruises took place. These images were obtained from the satellite sensor SeaWi FS (Sea-viewing Wide Field-of-view Sensor) and processed using the empirical band ratio ocean color algorithm OC4v4. The dispersal of the Orinoco River plume over the SEC has been documented previously by MllerKarger et al. (1989) and Hochman et al. (1994) using satellite im ages from the Coastal Zone Color Scanner (CZCS). However, the inte rpretation of the optic al signature (color) detected by these satellite sensors over the Orinoco River plume has become a real challenge. Hochman et al. (1994) tried to deconvolve the signatures of co lored dissolved organic carbon (CDOM) and phytoplankton pigmen ts on CZCS images from the Orinoco River plume, concluding the as much as 50% of the chlorophyll derived from the CZCS images within the plume was an artifact due to the presence of CDOM. However, the results found in this study indicate a larger impact of CDOM on ch lorophyll estimations from satellite sensors using band ratio algorithms. When using the OC4v4 algorithm to derive chlorophyll-a concentrations from in situ measurements of Rrs() within the Orinoco River pl ume in the SEC the derived concentrations were largely overestimated (F igure 54) for the wet and dry season. From 23 measurements in the SEC (between both seasons) only 1 value was underestimated, corresponding to the station with the highest Rrs in the blue wavelengths (SIM4_7). The percentage errors (%E) ranged from 20.90% (SIM3_7) to 466.47% (SIM3_10) ( X= 219.67, std = 183.2) during the dry seas on and from 27.33% (SIM4_7) to 1113 % (SIM4_8) ( X=287.4, std = 307.2) during the wet se ason (Table 23). Figure 55 shows the overall (both seasons) frequency distribution of the %E, indicating that most of the chlorophyll-a values derived from OC4v4 algorithm were at least 2 to 3 times higher than the in situ values. The two stations that showed lower %E (SIM3_7, and SIM4_7) were located in an upwelling zone in front of the Paria Peninsul a where the influence of the

PAGE 133

119 Orinoco plume is lower. These stations had Rrs() values in the blue wavelengths above 0.002 sr-1 and Rrs(555) values above 0.004 sr-1. The %E was compared to individual Rrs wavelengths and band ratios, and to the in situ chlorophyll-a concentration. The disp ersion of the few data points available for the comparisons was too scattered, and thus, de fining a trend or systematic error became problematic. Rrs() spectra in the SEC (Figure 51) shows that Rrs() values in the blue wavelengths were mostly around 0.002 sr-1 and Rrsmax values were constantly below 0.010 sr-1. In comparison, in clear blue waters Rrs are typically high in the blue with Rrs(400) values normally above 0.005 sr-1 (e.g. Froidefond et al., 2002). It is evident that the water in the SEC was very dark during both seasons. High values of ag() and ag() to aph() ratios observed in this region (previous section) indicate that Gelbstoff, or dissolved organic matter (CDOM), was the main factor for si gnificant decreases in the Rrs() values. Therefore, it is safe to conclude that ag() at the wavelengths used by the algorithm interfered with the empiri cal relationship be tween chlorophyll-a concentration and the Rrs() band ratios causing the OC4v4 algorithm to fail. An alternative to improve the performan ce of global band ratio algorithms such as OC4v4 is to create regional algorithms by finding site-specific coefficients from statistical regressions between in situ chlorophyll-a concentration and Rrs() band ratios. For this purpose, regressions were performed using in situ chlorophyll-a concentrations and Rrs() values for stations in th e SEC (Figures 56 and 57). Although statistically there seems to be a some relati onship, specifically for the dry season, the dispersion of the poi nts is too close to a flat li ne and it does not approach an asymptote at either end of the plot. This indicates, that changes in chlorophyll-a concentrations are not strongly represented by changes in the band ratios; therefore, very small variations in Rrs() may result in very large errors in the estimation of chlorophyll-a concentrations.

PAGE 134

120 Figure 52. Monthly mean SeaWiFS images, processed using the OC4v4 algorithm, for the cruises carried out during the dry season. SIM1: June, 1998 SIM3: February, 1999 SIM5: March, 2000

PAGE 135

121 Figure 53. Monthly mean SeaWiFS images, processed using the OC4v4 algorithm, for the cruises carried out during the wet season. SIM2: October, 1998 SIM4: October, 1999 SIM6: October, 2000

PAGE 136

122Figure 54. Chlorophyll-a concentration derived from the OC4v4 algorithm, [Chl-a]OC4v4 versus chlorophyll-a concentrations measured from in situ water samples [Chl-a]mea during de dry (A) and wet (B) season 0.1 1 10 0.1110 [Chla ]mea ( g l-1)[Chla ]oc4 ( g l-1) 0.1 1 10 0.1110 [Chla ]mea ( g l-1)[Chla ]oc4 ( g l-1)A B

PAGE 137

123Table 23. Chlorophyll-a concentrations from in situ measurements [Chl-a]mea, and derived from the OC4v4 algorithm [Chl-a]OC4v4, and percentage error (%E) 0 1 2 3 4 5 6 7 8 9 10 0100200300400More %EFrequency Figure 55. Frequency distribut ion of %E between in situ measurements and derived OC4v4 chlorophyll-a concentrations. Dry Season Wet Season Station [Chla ]mea ( g l-1) [Chla ]OC4v4 ( g l-1) %E Station [Chla ]mea( g l-1) [Chla ]OC4v4 ( g l-1) %E SIM1_5b 1.99 5.80 190.54 SIM2_180.68 1.77 160.29 SIM1_8b 1.45 4.23 191.34 SIM2_171.80 2.54 41.11 SIM1_12 2.04 3.22 57.84 SIM2_8 1.00 3.68 268.00 SIM1_13 2.41 3.20 32.78 SIM2_9 0.78 3.03 288.46 SIM3_6 0.64 3.60 460.40 SIM2_6 1.63 3.31 103.07 SIM3_7 3.52 4.26 20.90 SIM2_100.92 2.53 175.00 SIM3_10 0.17 0.98 466.47 SIM4_7 3.00 2.18 -27.33 SIM5_6 0.62 2.71 337.10 SIM4_8 0.79 9.58 1112.66 SIM4_9 1.12 3.31 195.64 SIM4_100.28 1.99 622.06 SIM6_8 1.53 3.95 158.34 SIM6_9 1.13 2.86 153.10 SIM6_100.17 1.29 658.82 SIM6_120.20 0.43 115.00

PAGE 138

124 Figure 56. Rrs() band-ratios used by the OC4v4 algorithm versus in situ chlorophyll-a concentrations in the SEC during the dry season y = 0.11x2 0.51x + 0.96 R2 = 0.53 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs412/Rrs555 y = 0.12x2 0.53x + 1.08 R2 = 0.64 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs443/Rrs555 y = 0.09x2 0.44x + 1.24 R2 = 0.62 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs490/Rrs555 y = 0.06x2 0.29x + 1.16 R2 = 0.60 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs510/Rrs555

PAGE 139

125 Figure 57. Rrs() band-ratios used by the OC4v4 algorithm versus in situ chlorophyll-a concentrations in the SEC during the wet season y = 0.15x2 0.49x + 0.95 R2 = 0.33 0.1 1 10 01110 [chl-a]insitu ( g l-1)Rrs412/Rrs555 y = 0.20x2 0.71x + 1.19 R2 = 0.38 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs443/Rrs555 y = 0.24x2 0.89x + 1.52 R2 = 0.46 0.1 1 10 01110 [chl-a]insitu ( g l-1)Rrs490/Rrs555 y = 0.1588x2 0.5597x + 1.3227 R2 = 0.4895 0.1 1 10 0.1110 [chl-a]insitu ( g l-1)Rrs510/Rrs555

PAGE 140

126Using a Rrs inversion model Another alternative to derive chlorophyll-a concentrations is to use a semi-analytical (SA) algorithm instead of a pure empirical one. Carder et al. (1999) developed a SA algorithm to derive chlorophyll-a concentrations for the M oderate-Resolution Imaging Spectrometer (MODIS) satellite sensor. This al gorithm is based on an inversion model in which absorption coefficients are derived from MODIS Rrs() values. Chlorophyll-a concentrations are then derived from the phytoplankton absorption coefficient (aph) at 440 nm determined by the inversion model. To explore the possibility of using this kind of algorithm, a bio-optical i nversion model developed by Lee et al. (1999) was used to derive ag() and aph() values from in situ measurements of Rrs() in the SEC. Table 24 shows the results from the inversion model. %E for aph(440) ranged from 60.5% (SIM4_8) to 282.96% (SIM2_6). From nine aph(440) data points seven were overestimated by more than 50%, and two poi nts were underestimated by less than 50%. The model had a better performance deriving ag(440) values. %E for ag(440) ranged from –83.23% (SIM6_10) to 36.66% (SIM3_6). From seven ag(440) data points four were underestimated by less than 25%, and one va lue was overestimated by only 3.4%. In general, the model te nded to underestimate ag(440) values when the in situ values of ag(440) were below 1 m-1 and to overestimate when in situ ag(440) values were above 1 m-1. For the inversion model to perform prope rly, it needs to consider the spectral characteristics of the different variables i nvolved in the model, including those of aph, aph*, ag(), and bb() ( Lee et al., 1999; Carder et al., 1999). The spectra l characteristics used by the model in this study were based on a series of in situ measurements collected in the Gulf of Mexico (Ivey, 2004 (persona l communication)) as a representation of coastal waters. Therefore, it is possible that the model failed in the determination of aph() values because the parameters used may not apply to the SEC region. The model could be improved by adjusting (tuning) it to the local conditions found in the SEC. However, as shown previously, due to the riverine influence from the Orinoco, the spectral features that characterize the SE C, were highly variable, especially for aph*().

PAGE 141

127 This variability and the interference of ag() makes it difficult to define the spectral parameters that could be changed (tuned) in the model to adapt it to the local conditions, so that it could effectively work over the whole region. High %E (> 100%) were obtained even by using a lo cal relationship between in situ measurements of aph() and chlorophyll-a concentration (Figure 58). Therefore, a good understandin g of the factors responsible for the variability in aph* () (e.g. package effect) is required to develop a more accurate regional algorithm (see Carder et al., 2004). Even though there were no in situ measurements available to compare to the bb() values derived by the model, these were comp ared to the concentr ation of TSM (Figure 59). The purpose was to determine if a correla tion exists between these two variables in the SEC. In turbid waters, loaded with large quantities of suspended sediments, the scattering coefficient increases at all angl es, which may result in large values of bb() (Mobley, 1994). Therefore, strong correlations between bb() and TSM can be found in these regions. Values of bb(640) ranged from 0.0010 m-1 to 0.0064 nm-1 ( X= 0.0024 m-1, std = 0.0015) (Table 24). No significan t correlation was found between bb(640) and TSM (r2 = 0.38), although some tendency can be observed of bb(640) to increase with TSM the data show significant scatter. It is possible that this poor correlation could result from inaccurate measurements of the in situ radiometric spectra. If the points with suspicious low values of bb(640) are excluded the correlation between bb(640) and TSM increases (r2 = 0.55). A stronger correlation, and less sc attered points, was found between bb(640) and chlorophyll-a concentrations (r2 = 0.76). These results indicate that the concentration of suspended inorganic particles in the SEC was insufficient to have a significant impact on bb(), while phytoplankton, which supposedly scatte rs only weakly at la rge angles (Kirk, 1994), seemed to make a more prominent cont ribution. According to this, it is possible that TSM in the SEC was dominated by phytoplankton particles. However, this conclusion remains somewhat specula tive as the relationship between bb() and the

PAGE 142

128 different particles found in the water continues to be uncertain and controversial, mainly due to the lack of in situ bb() measurements (Stramski et al., 2004). Table 24. Results from Rrs inversion model Figure 58. Local relationship between chlorophyll-a concentration and aph(440) y = 0.1162e43.56x R2 = 0.8 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 0.0000.0200.0400.0600.0800.100aph(440) (m-1)chlorophyll-a (g l-1) aph(440) ag(440) bb(640) STATION in situ (m-1) modeled (m-1) %E in situ (m-1) Modeled (m-1) %E modeled (m-1) SIM2_18 0.034 0.057 67.1 N/A N/A N/A 0.0024 SIM2_8 0.053 0.191 260.2 0.66 0.46 -30.2 0.0023 SIM2_9 0.049 0.154 215.2 0.45 0.42 -5.7 0.0020 SIM2_10 0.039 0.060 54.2 0.47 0.42 -11.4 0.0015 SIM2_6 0.047 0.180 283.0 0.77 0.59 -23.3 0.0021 SIM3_6 N/A N/A N/A 1.67 2.28 36.7 0.0037 SIM3_10 0.024 0.044 85.3 N/A N/A N/A 0.0010 SIM4_7 0.082 0.138 68.8 N/A N/A N/A 0.0048 SIM4_8 0.035 0.014 -60.5 1.11 1.15 3.4 0.0075 SIM6_10 N/A N/A N/A 1.00 0.17 -83.2 0.0011 SIM6_12 0.015 0.011434 -23.8 N/A N/A N/A 0.0014

PAGE 143

129 Figure 59. Relationship between the backscattering coefficient at 660 nm, bb(660), derived from the model and in situ measurements of TSM (A), and chlorophyll-a concentrations (B) R2 = 0.39 0.0010 0.0100 0.1000 024681012 TSM (mg l-1)bb(640) (m-1) R2 = 0.76 0.0010 0.0100 0.1000 00.511.522.533.54 Chlorophylla ( g l-1)bb(640) (m-1)A B

PAGE 144

130Rrs in situ measurements vs. Rrs modeled The Rrs inversion model was supplied with in situ above-water measurements of total upwelling radiance, Lt(); sky radiance, Lsky(); and downwelling irradiance, Ed(). The model calculated Rrs() after correcting for reflecte d sky light and sun glint by subtracting a fraction of the sky reflectance ( based on Fresnel reflectance) and using a spectrally constant offset derived by the model. After Rrs() values were calculated, another Rrs() spectra was modeled using the optim ization technique, from which the values of the absorption and backsca ttering coefficients were derived. The Rrs() spectra showed in Figure 51, we re corrected only by subtracting Lsky using a surface Fresnel re flectance of 0.02 from Lt(). To avoid using the Rrs at 750 nm as an offset, no spectral offset was subtracted, and therefore these Rrs() were not corrected for sun glint and some residual radi ance was present. The improper removal of reflected sky radiance and resi dual signal is a potential source of uncertainties in above water determinations of Rrs() (Toole et al., 2000). These may be the reason why some of them presented unusual high values at wavelengths above 700 nm (e.g. SIM3_6, SIM3_7, SIM6_8). Figure 59 shows the comparison be tween the three suspicious Rrs() spectra (Rrs()uncorrected) and the Rrs() spectra derived from the model (Rrs()corrected, and Rrs()mod). After the Rrs() values were corrected for sun glint, they were significantly lowered; the values went negative at 400 nm and above 700 nm, indi cating that the model may be overcorrecting, especially in the bl ue wavelengths. The final spectra modeled, however, may be a good representation of the “true” surface corrected Rrs(). A problem with the modeled spectra is that it does not account for the chlorophyll-a fluorescence peak around 685 nm. The failure of th e model to solve for a modeled Rrs() that matches the measured one (e.g. SIM6_8) could be an indicative of errors in above-water radiometric measurements; becoming a tool for data quality control.

PAGE 145

131Figure 60. Comparison between Rrs() values uncorrected for sun glint, Rrs_uncorr, the Rrs()corrected, and the Rrs() derived by the inversion m odel (Rrs_corr and Rrs_mod, respectively) 0 0.002 0.004 0.006 0.008 400 450500550600650700750800 Wavelength (nm)Rrs (sr-1) Rrs_mod Rrs_corr Rrs_uncorr 0 0.001 0.002 0.003 0.004 0.005 0.006 400 450500550600650700750800 Wavelength (nm)Rrs (sr-1) Rrs_mod Rrs_corr Rrs_uncorr SIM6_8 0 0.001 0.002 0.003 0.004 0.005 0.006 400 450500550600650700750800 Wavelength (nm)Rrs (sr-1) Rrs_mod Rrs_corr Rrs_uncorr SIM3_7 SIM3_6

PAGE 146

132 Summary and Conclusion A low salinity surface layer observed during each SIMBIOS-Orinoco cruise indicates that the Gulf of Paria (GOP) and southeaste rn Caribbean (SEC) are under the influence of fresh water input during both we t and dry seasons. In the GOP this surface layer is a direct result of the discharge from the Orinoc o River plume; in the SEC this layer may be a mixture of Orinoco and Amazon waters, wh ich are transported in to the SEC by the Guayana current. However, the similarity of the silicate-salinity mixi ng line found in this study for the GOP and SEC to the one found by Froelich et al. (1989) for the Caribbean, suggests that water from th e Orinoco River plume, rather than from the Amazon, dominates this region. This finding supports the observations ma de by Mueller-Karger et al. (1989) using satellite imagery. Higher nitrite ( X= 0.28 mg l-1) and phosphate ( X= 0.25 mg l-1) concentrations observed in the GOP during SIM3, which result ed in high concentrations of chlorophyll-a (~ 8.1 g l-1) in one station (SIM3_11), indicate th at upwelling processes are likely to occur within the GOP duri ng the dry season, causing pa tches of high nutrient and chlorophyll-a values, and even red tides. In general, surface chlorophyll-a concentrations in the GOP and SEC are characteristic of eutrophic waters ( X > 0.9 g l-1). This suggests that there is enough sunlight and nutrients in the surface laye r of the Orinoco plume for phytoplankton to growth during both seasons, with a relative in crease during the dry season. This seasonal variability is accompanied by an increas e in phytoplankton absorption at 440 nm, aph(440), from X 0.04 m-1 in the wet season to X 0.06 m-1 in the dry season. Values of aph() in the blue wavelengths (e.g. 412 nm, 440 nm, and 490 nm) are significantly lower than those observed in Gelbstoff absorption, ag() for the same wavelengths.

PAGE 147

133 Colored dissolved organic matter (CDOM), or Gelbstoff, concentration also remain high in the GOP and SEC, indist inctively of the season, as in dicated by high values of its absorption coefficient, ag(), which at 440 nm ranged from 0.23 m-1 to 3.21 m-1. Spatial changes in CDOM concentrations are represented by a decrease in ag(440) values going from the GOP ( X = 1.6 m-1) to the SEC ( X= 0.9 m-1). Estimations of the contributions by CDOM phytoplankton, and de tritus to the total absorption coefficient, demonstrate that in th e GOP and SEC close to 90% of the light at 440 nm is absorbed by CDOM during both seasons, while the absorption by phytoplankton and detritus are more vari able. During the dry season, phytoplankton absorb approximately half of the remaining 10% and detritus absorb the other half, in the GOP and SEC. During the wet season, the mean contribution by phytoplankton decreases to ~ 3% while the mean contribution by detritu s increases to ~8% in the GOP; and in the SEC, the contribution by phytoplankton rema ins ~ 5% and the mean contribution by detritus decreases to ~3%. However, it is important to emphasize that these values could be as high as 34% for detritus in the GOP during the wet season, and as 20% for phytoplankton in the SEC during the dry season. Ratios between ag(440) and aph(440) are highly variable a nd, in general, extremely large compared to those found in other coas tal areas, with most of the values ranging between 10 and 50. While an extreme minimum value of 3.33 was observed in the station SIM3_11, it should be noted that this stati on was under the influence of an upwelling and red tied event. Concentrations of total suspended matter (T SM) are also highly va riable but normally below 10 mg l-1 in both in the GOP and SEC, with higher values around Serpent’s Mouth in the stations close to the Orinoco Delta. Hi gh values at these stations resulted in an overestimation of the mean concentration of TSM in the GOP. During SIM1 and SIM2 TSM values remained very similar in and out side the GOP with values usually below 5 mg l-1. Concentrations of TSM during SIM3 we re extremely higher when compared to the other cruises, especially within the GOP Unfortunately, there is no clear evidence

PAGE 148

134 about the mechanism that could explain the distribution of TSM in the GOP and SEC, and further study is required. Spectral changes in remote sensing reflectance, Rrs(), are mainly characterized by a shift of the Rrs maximum peak (Rrsmax) towards shorter wavelength. This shift occurred in a spatial transition indistinctivel y of the season. Wavelengths of Rmax decreased from ~580 nm around the Orinoco delta to ~500 nm in the SEC. This shift is the result of the relative reduction in ag() and, to a lesser extent, the decrease observed in TSM going from the GOP to the SEC. The reduction of ~ 50% in the magnitude of Rrsmax, going from waters near the Orinoco Delta to the SEC, is a result of the decrease in the concentration of TSM. The impact of ag() in the color of the Orinoco rive r plume is clearly demonstrated by its strong dominance in the absorption of light at the blue wavelengths over the absorption by suspended particles, significan tly decreasing the water leaving radiance, Lw(), and therefore the Rrs(), at those spectral bands. Th is significant c ontribution of ag() to the total absorption makes using em pirical band ratio algorithms to derive chlorophyll-a concentrations from satellite sensors, in the SEC under the influence of the Orinoco river plume, problematic. The interference of ag() causes the correlations between Rrs() band ratios and chlorophyll-a concentrations, used by the OC4v4 algorithm to be statistically insignificant, and results in algorithm failure with percentage errors >100%. Furthermore, the use of inversion modeling also results in significant overestimations of aph(), and excludes the possibility of estimating chlorophyll-a concentrations from aph(440) values derived from the m odel These findings suggest that a semi-analytical (SA) algorithm based on i nversion modeling from Rrs may not be a feasibly alternative for the determination of chlorophyll-a concentration either. A possible alternative to derive chlorophyll-a concentrations from Rrs() is the use of the fluorescence peak (~ 680 nm) characteristic of this pigment, since Gelbstoff would not have any interference at this wavelength. This alternative has yet to be tested for this region.

PAGE 149

135 Two positive outcomes were obtained by using the SA inversion model. It provided acceptable %errors in the determination of ag(440) values (<20%), suggesting that using an inversion model (e.g. Carder et al., 1999) based on satellite data may be appropriate. Secondly, above water measurements of tota l radiance can be corrected for sky light reflectance and sun glint, thus obt aining more accurate values of Lw() and therefore of Rrs(). In conclusion, the high values of ag(440) suggest that rive rine dissolved organic matter (DOM) has a significant contributi on to the carbon budget in SEC and GOP. More relevance must be given to the study of DOM and its colored fraction to the carbon cycle in these regions and to the estimation of remote sensed ag() values.

PAGE 150

136 References Aminot, A. (1983), Materiel par ticulaire et filtration, in Manuel de Analyses Chimiques en Milieu Marin, edited by A. Aminot and M. Chaussepied, pp. 153-158, Centre National pour L'Exploitation des Oceans, CNEXO, France. Aparicio-Castro, R. (2003), Review of the oceanographic characteristics on the continental shelf of northeastern Venezuela, in The Sardine (Sardinella aurita): Its environment and explotation in Eastern Venezuela, edited by P. Fron, and J. Mendoza, pp. 171-232, Institut de Recher che Pour Le Dveloppement, Paris, France. Battin, T. J. (1998), Dissolved organic matter and its optical properties in a blackwater tributary of the Orinoc o River, Venezuela, Org. Geochem. 28, 561-569. Betzer, P. R., D. W. Eggiman, K. L. Carg er, D. R. Kester, and S. B. Betzer (1977), Seasonal patterns in suspended calcium carbonate concentration during the dry and wet seasons in the eastern Caribbean, in The Fate of Fossil Fuels CO2 in the Oceans, edited by N. R. Andersen, and A. Malahof, pp. 63-79, Plenum Press, New York. Bidigare, R. R., M. E. Ondrusek, and J. M. Brooks (1993), Influence of the Orinoco River outflow on distributi ons of algal pigments in the Caribbean Sea, J. Geophys. Res., 98, 2259-22269.

PAGE 151

137 Bissett, W. P., J. S. Patch, K. L. Carder and Z. P. Lee (1997), Pigment packaging and Chl a-specific absorption in high-light oceanic waters, Limnol. Oceanogr., 42, 961-968. Bonilla, J., W. Senior, J. Bugden, O. Zafiri ou, and R. Jones (1993), Seasonal distribution of nutrients and primary productivity on the eastern continental shelf of Venezuela as influenced by the Orinoco River, J. Geophys. Res., 98, 2245-2257. Blough, N. V., O. C. Zafiriou, and J. Bonilla (1993), Optical absorption spectra of waters from the Orinoco River outflow: Terrestria l input of colored organic matter to the Caribbean, J. Geophys. Res., 98, 2271-2278. Blough, N. V., and R. Del Vecchio ( 2002), Chromophoric DOM in the Coastal Environment, in Biogeochemistry of Marine Dissolved Organic Matter, edited by D. A. Hansell and C. A. Carlson, pp. 509-546, Academic Press, San Diego. Bricaud, A., A. Morel, and L. Prieur ( 1981), Absorption by dissolved organic matter of the sea (yeellow substance) in the UV and visible domains, Limnol. Oceanogr., 26, 43-53. Bricaud, A., and D. Stramski (1990), Sp ectral absorption coefficients of living phytoplankton and nonalgal biogenous matte r: A comparison between the Peru upwelling area and the Sargasso Sea, Limnol. Oceanogr., 35, 562-582. Bricaud, A., M. Babin, A. Morel, and H. Cl austre (1995), Variability in the chlorophyllspecific absorption coefficients of natural phytoplankton: Analysis and parameterization, J. Geophys. Res., 100, 13321-13332.

PAGE 152

138 Bukata, R. P., J. H. Jerome, K. Y. Kondr atyev, and D. V. Po zdnyakov (1995), Optical properties and remote sensing of Inla nd and coastal waters, CRC Press, Boca Raton. Bunt, J. A. C., P. Larcombe, and C. F. Ja go (1999), Quantifying th e response of optical backscatter devices and transmisometers to variations in suspended particulate matter, Continental Shelf Research, 19, 1199-1220. Carder, K. L., and R. G. Steward (1985), A remote sensing reflectance model of a red tide dinoflagellate off West Florida, Limnol. Oceanogr. 30(2), 286-298. Carder, K. L., F. R. Chen, Z. P. Lee, S. K. Hawes, and D. Kamykowski (1999), Semianalytical Moderate-Resolution Imaging Spectrometer algorithms for chlorophyll a and absorption with biooptical domains on nitrate-depletion temperatures, J. Geophys. Res., 104, 5403-5421. Carder, K. L., F. R. Chen, J. P. Canizzaro, J. W. Campbell, and B. G. Mitchell (2004), Performance of the MODIS semi-ana lytical ocean color algorithm for chlorophyll-a, Advances in Space Research, 33, 1152-1159. Carder, K. L., S. K. Hawes, K. A. Baker, R.C. Smith, R. G. Setward, and B. G. Mitchell (1991), Reflectance model for quantifyi ng chlorophyll a in the presence of productivity degradation products, J. Geophys. Res., 96, 20599-20611. Carder, K. L., R. G. Steward, G. R. Harvey and P. B. Ortner (1989), Marine and fulvic acids: Their effects on remote sensing of ocean chlorophyll, Limnol. Oceanogr., 34, 68-81.

PAGE 153

139 Corredor, J. E., and J. M. Morell (200 1), Seasonal variation of physical and biogeochemical features in eastern Caribbean Surface Water, J. Geophys. Res., 106, 4517-4525. Corredor, J. E., J. M. Morell, J. M. Lpez, J. E. Capella, and R. A. Armstrong (2004), Cyclonic eddy entrains Orinoco River Plume, EOS Trans. AGU, 85(20), 201-202. Cullen, J. J. (1982), The deep chlorophyll maximum: Comparing ve rtical profiles of chlorophyll, Can. J. Fish Aquat. Sci., 39, 791-803. Del Castillo, C. E., P. Coble, J. M. More ll, J. M. Lpez, and J. E. Corredor (1999), Analysis of the optical properties of the Orinoco River plume by absorption and fluorescence spectroscopy, Mar. Chem., 66, 35-51. Doerffer, R., K. Sorensen, and J. Aike n (1999), MERIS potential for coastal zone applications, Int. J. Remote Sensing, 20, 1809-1818. 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 the Eastern Caribbean Basin, J. Geophys. Res., 98, 2279-2288. Froelich, P., K. Atwood, and G. Giese ( 1978), Influence of Amazon River discharge on surface salinity and dissolved silicate concentration in the Caribbean Sea, DeepSea Res., 25, 735-744. Froidefond, J., L. Gardel, D. Guiral, M. Parra, and J. Ternon (2002), Spectral remote sensing reflectances of coastal wate rs in French Guiana under the Amazon influence, Remote Sens. Environ., 80, 225-232.

PAGE 154

140 Gordon, L. I., J. C. Jennings Jr., A. A. Ross, and J. M. Krest (1993), A suggested protocol for continuous flow automated anal ysis of seawater nutrients, in WOCE Operation Manual, pp. 1-52, WHP Office Report 90-1, WOCE Report 77. Gould, R. W. Jr., R. A. Arnone, and M. Ma rtinolich (1999), Spectral dependence of the scattering coefficient in case 1 and case 2 waters, Appl. Opt., 38, 2377-2383. Grasshoff, K. (1976), Methods of Seawater Analysis, Verlag Chemie, New York. Hasle, G. R., and E. E. Syvert sen (1996), Marine Diatoms, in Identifying Marine Diatoms and Dinoflagellates, edited by C. R. Tomas, pp. 5-385, Academic Press, Inc., London. Hellweger, F. L., and A. L. Gordon (2002) Tracing Amazon Rive r water into the Caribbean Sea, J. Mar. Res., 60, 537-549. Hochman, H. T., F. E. Mller-Karger, and J. J. Walsh (1994), Interpre tation of the coastal zone color scanner signature of the Orinoco River plume, J. Geophys. Res., 99, 7443-7455. Holm-Hansen, O., C. J. Lorenzen, R. W. Holmes, and J. D. Strickland (1965), Fluorometric determination of chlorophyll, Journal Cons. Perm. Int. Explor. Mer., 30, 3-15. Hooker, S. B., C. R. McClain, and A. Holmes (1993), Ocean color imaging: CZCS to SeaWiFS, MTS journal, 27, 3-15.

PAGE 155

141 Hu, C., E. T. Montgomery, R. W. Schmitt, and F. E. Mller-Karger (2004), The dispersal of the Amazon and Orinoco River water in the tropical Atlantic and Caribbean Sea: Observation from space and S-PALACE floats, Deep-Sea Res. Part II, SeaWiFS-special volume (In press). Hu, C., F. E. Mller-Karger, D. C. Biggs, K. L. Carder, B. Nababan, D. Nadeau, and J. Vanderbloemen (2003), Comparison of ship and satellite bio-optical measurements on the continental margin of the NE Gulf of Mexico, Int. J. Remote Sensing, 24, 2597-2612. Jago, C. F., and C. F. J. Bull (2000), Quantifi cation of errors in transmisometer-derived concentration of suspended pa rticulate matter in the coastal zone: Implications for flux determinations, Marine Geology, 169, 273-286. Kirk, J. T. O. (1976), Yellow Substance (Gel bstoff) and its contribu tion to the attenuation of photosynthetically active radiation in some inland and coastal Soth-Eastern Australian waters, Aust. J. Mar. Freshwater Res., 27, 61-71. Kirk, J. T. O. (1994), Light & Photosynthesis in Aquatic Ecosystems, Cambridge University Press, Cambridge. Kishino, M., M. Takahashi, N. Okami, and S. E. Ichimura (1985), Estimation of the spectral absorption coefficients of phytoplankton in the sea, Bull. of Mar. Sc., 37, 634-642. Kitchen, J.C., and J. R. Zaneveld (1990), On the noncorrelation of th e vertical structure of light scattering and chlorophyll a in case I waters, J. Geophys. Res., 95, 2023720246.

PAGE 156

142 Kopelevich, O. (2000), Bio-opt ical algorithms. Presented at the Remote sensing of ocean color course, IOCCG Training Course, Turkey, September 2000. Lee, Z. P., K. L. Carder, C. D. Mobley, R. G. Steward, and J. S. Patch (1998), Hyperspectral remote sensing for shallo w waters: 1. A semianalytical model, Appl. Opt., 37, 6329-6338. Lee, Z. P., K. L. Carder, C. D. Mobley, R. G. Steward, and J. S. Patch (1999), Hyperspectral remote sensing for shallow waters: 2. Deriving bottom depths and water properties, Appl. Opt., 38, 3831-3843. Lewis, W. M. jr., and J. F. Saunders, III (1989) Concentration and transport of dissolved suspended substances in the Orinoco River, Biogeochemistry, 7, 203-240. Lewis, W. M. jr., and J. F. Saunders, III ( 1990), Chemistry and element transport by the Orinoco main stem and lower tributaries, in The Orinoco River as an ecosystem, edited by F. H. Weibezahn, H. Alvarez, and W. M. Lewis, Jr., pp. 211-236, Impresos Rubel C.A., Caracas. McClain, C. (2001), Ocean Color from Satellites, in Encyclopedia of Ocean Sciences, edited by J. H. Steele, K. K. Ture kian, and S. A. Thorpe, pp. 1946-1959, Academic Press, New York. McClain, C. R., and G. S. Fargion ( 1999), SIMBIOS Project 1998 annual report, NASA Tech. Memo. 1999-208645, NASA Goddard Space Flight Center, Greenbelt, MD.

PAGE 157

143 Meade, R. H., F. H. Weibezahn, W. M. Lewis jr., and D. Perez Hernandez (1990), Suspended-sediment budget for the Orinoco River, in The Orinoco River as an ecosystem, edited by F. H. Weibezahn, H. Alva rez, and W. M. Lewis, Jr., pp. 5579, Impresos Rubel C.A., Caracas. Mitchell, B. G., and D. A. Kiefer (1988), Chlorophyll a specific absorption and fluorescence excitation spectra for light-limited phytoplankton, Deep-Sea Res., 35, 639-663. Mobley, C. D. (1994), Light and Water: Radiative transfer in natural waters, Academic Press, New York. Monente, J. A., and G. Colonnello (1997) Hidroquimica del delta del Orinoco, Memoria, 57(148), 109-131. Morel, A. (2001), Bio-optical Models, in Encyclopedia of Ocean Sciences, edited by J. H. Steele K. K. Turekian, and S. A. T horpe, pp. 317-326, Academic Press, New York. Morel, A., L. Prieur, E. Boss, S. W. Pegau, R. V. Zaneveld, and A. H. Barnard (1977), Analysis of variations in ocean color, Limnol. Oceanogr., 22, 709-722. Morell, J. M., and J. E. Corredor (2001), P hotomineralization of fluorescent dissolved organic matter in the Orinoco River pl ume: Estimation of ammonium release, J. Geophys. Res., 106, 16807-16813. Morrison, J. M, and O. P. Smith (1990), Ge ostrophic transport vari ability along the Aves Ridge in the eastern Caribbean Sea during 1985-1986, J. Geophys. Res., 95, 699710.

PAGE 158

144 Morrison, J., and W. Nowlin (1982), Genera l distribution of water masses within the eastern Caribbean Sea during the winter of 1972 and fall of 1973, J. Geophys. Res., 87, 4207-4229. Muller, J. M. (2002), In-water radiometric profile measurements and data analysis protocol, in NASA Tech. Memo. 2002-210004, Rev. 3, Vol. 1, edited by J. L. Mueller and G. S. Fargion, NASA Goddard Space Flight Center, Greenbelt, MD. Mller-Karger, F. E., C. R. McClain, T. R. Fisher, W. E. Esaias, and R. Varela (1989), Pigment distribution in the Caribbean Sea: Observations from space, Prog. Oceanog., 23, 83-64. Nelson, J. R., and S. Guarda (1995), Partic ulate and dissolved spectral absorption on the continental shelf of the southeastern United States, J. Geophys. Res., 100, 87158732. O'Reilly, J. E., S. Maritonera, B. G. Mitchell, D. A. Siegel, K. L. Carder, S. A. Garver, M. Kahru, and C. McClain (1998), O cean color chlorophyll algorithms for SeaWiFS, J. Geophys. Res., 103, 24937-24953. O’Reilly, J.E., S. Maritorena D.A. Siegel, M.C. O’Brien, D. Toole, F. P. Chavez, P. Strutton, G.F. Cota, S.B. Hooker, C.R. Mc Clain, K.L. Carder, F.E. Mller-Karger, L. Harding, A. Magnuson, D. Phinney, G. F. Moore, J. Aiken, K.R. Arrigo, R. Letelier, and M. Culver (2000), Ocea n Color Chlorophyll a Algorithms for SeaWiFS, OC2 and OC4: Version 4, in NASA Tech. Memo. 2000-206892, Vol. 11, Part 3, edited by S.B. Hooker, and E.R. Firestone, NASA Goddard Space Flight Center, Greenbelt, MD.

PAGE 159

145 Paolini, J. (1995), Particul ate organic carbon and nitrog en in the Orinoco River (Venezuela), Biogeochemistry, 29, 59-70. Pilson, M. E. Q. (1998), An introduction to the chemistry of the sea, Prentice Hall, New Jersey. Prieur, L., and S. Sathyendranath (1981), Optical Classification, Limnol. Oceanogr. 26, 671-689. Sathyendranath, S. (Ed.) (2000), Remote Sensing of Ocean Col our in Coastal, and Other Optically-Complex, Waters, IOCCG Report 3, IOCCG, Dartmouth. Sathyendranath, S., L. Lazzara, and L. Prieur (1987), Variations in the spectral values of specific absorption of phytoplankton, Limnol. Oceanogr., 32, 403-415. Shifrin, K. S. (1983), Physical optics of ocean water, American Institute of Physics, New York. Stramski, D., E. Boss, D. Bogucki, and K. J. Voss (2004), The role of seawater constituents in light backscattering in the ocean, Prog. Oceanog., 61, 27-56. Toole, D. A., D. A. Siegel, D. W. Menzies, M. J. Neumann, and R. C. Smith (2000), Remote-sensing reflectance determinati ons in the coastal ocean environment: impact of instrumental characteristics and environmental variability, Appl. Opt., 39, 456-468. Twardowski, M. S., and P. L. Donaghay (2002), Photobleaching of aquatic dissolved materials: Absorption removal, spectral alteration, and their interrelationship, J. Geophys. Res., 107, 6-1/6-11.

PAGE 160

146 Varela, R., F. Carvajal, and F. E. Mller-Kar ger (2003), El fitoplanct on en la plataforma nororiental de Venezuela, in The Sardine (Sardinella aurita): Its Environment and Exploitation in Eastern Venezuela, edited by P. Fron, and J. Mendoza, pp. 263294, Institut de Recherche P our le Dveloppement, Paris. Vodacek, A., N. V. Blough, M. D. DeGrandpre, E. T. Peltzer, and R. K. Nelson (1997), Seasonal variation of CDOM and DOC in the Middle Atlantic Bight: Terrestrial inputs and photooxidation, Limnol. Oceanogr., 42, 674-686. Vrsmarty, C. J., B. Fekete, and B. A. Tu cker (1996), River discharge Database, version 1.0 (RivDIS v1.0), Technical Documents in Hydrology Series, UNESCO, Paris. Vrsmarty, C. J., B. Fekete, and B. A. Tucker (1998), River discharge Database, Version 1.1 (RivDIS v1.0 supplement), Inst itute for the Study of Earth, Oceans, and Space, University of Ne w Hampshire, Durham, NH. Warne, A. G., R. H. Meade, W. A. White, E. H. Guevara, J. Gibeaut, R. C. Smyth, A. Aslan, and T. Tremblay (2002), Regi onal controls on geomorphology, hydrology, and ecosystem integrity in the Orinoco Delta, Venezuela, Geomorphology, 44, 273-307. Wells, J. T., and S. Y. Kim (1991), The re lationship between beam transmission and concentration of suspended pa rticulate material in the Neuse Rver Estuary, North Carolina, Estuaries, 14, 395-403. WET Labs Inc. (1995), ac-9 User's Guide, ve rsion 1.0, Wet Labs, Inc. Philomath, OR.

PAGE 161

147 Yentsch, C. S., and C. A. Reichert (1962) The interrelationshi p between water-soluble yellow substances and chloroplastic pigments in marine algae, Bot. Mar., 3, 6574. Zaneveld, J. R. V., J. C. Kitchen, and C. Moore (1994), Scatteri ng error correction of reflecting-tube absorption meter, Ocean Optics XII, 2258, 44-55.

PAGE 162

148 Appendices

PAGE 163

149Appendix A. Statistics Summary for surface temperature and salinity values A-1 SIM1-GOP TEMPERATURE (C) SALINITY (psu) Mean 28.08 24.50 Standard Error 0.11 0.17 Median 28.08 24.50 Standard Deviation 0.15 0.24 Sample Variance 0.02 0.06 Range 0.21 0.35 Minimum 27.97 24.33 Maximum 28.19 24.67 Count 2 2 A-2 SIM1-SEC TEMPERATURE (C) SALINITY (psu) Mean 27.61 30.27 Standard Error 0.40 0.95 Median 28.10 29.94 Standard Deviation 1.19 2.84 Sample Variance 1.42 8.08 Range 4.02 9.06 Minimum 24.88 27.36 Maximum 28.90 36.42 Count 9 9

PAGE 164

150Appendix A. (Continued) A-3 SIM2-GOP TEMPERATURE (C) SALINITY (psu) Mean 29.27 19.41 Standard Error 0.10 0.72 Median 29.15 19.50 Standard Deviation 0.27 2.04 Sample Variance 0.07 4.16 Range 0.70 6.28 Minimum 29.02 15.36 Maximum 29.73 21.65 Count 8 8 A-4 SIM2-SEC TEMPERATURE (C) SALINITY (psu) Mean 28.25 29.56 Standard Error 0.27 1.49 Median 28.10 29.78 Standard Deviation 0.76 4.22 Sample Variance 0.58 17.83 Range 2.13 12.45 Minimum 27.35 22.15 Maximum 29.48 34.60 Count 8 8

PAGE 165

151Appendix A. (Continued) A-5 SIM3-GOP TEMPERATURE (C) SALINITY (psu) Mean 27.06 25.15 Standard Error 0.15 0.91 Median 27.02 25.58 Standard Deviation 0.37 2.22 Sample Variance 0.13 4.92 Range 1.06 5.93 Minimum 26.57 21.27 Maximum 27.63 27.19 Count 6 6 A-6 SIM3-SEC TEMPERATURE (C) SALINITY (psu) Mean 26.56 30.62 Standard Error 0.40 1.53 Median 26.79 29.17 Standard Deviation 0.89 3.43 Sample Variance 0.79 11.74 Range 2.26 7.85 Minimum 25.02 27.09 Maximum 27.28 34.94 Count 5 5

PAGE 166

152Appendix A. (Continued) A-7 SIM4-GOP TEMPERATURE (C) SALINITY (psu) Mean 29.41 19.70 Standard Error 0.19 0.86 Median 29.50 20.04 Standard Deviation 0.49 2.27 Sample Variance 0.24 5.16 Range 1.38 7.60 Minimum 28.69 15.32 Maximum 30.06 22.92 Count 7 7 A-8 SIM4-SEC TEMPERATURE (C) SALINITY (psu) Mean 28.61 27.33 Standard Error 0.45 2.76 Median 28.86 23.78 Standard Deviation 1.01 6.17 Sample Variance 1.02 38.07 Range 2.55 13.37 Minimum 26.94 22.24 Maximum 29.50 35.61 Count 5 5

PAGE 167

153Appendix A. (Continued) A-9 SIM5-GOP TEMPERATURE (C) SALINITY (psu) Mean 26.84 31.49 Standard Error 0.05 0.30 Median 26.86 31.51 Standard Deviation 0.11 0.74 Sample Variance 0.01 0.55 Range 0.30 2.20 Minimum 26.67 30.34 Maximum 26.97 32.55 Count 6 6 A-10 SIM5-SEC TEMPERATURE (C) SALINITY (psu) Mean 26.22 33.73 Standard Error 0.34 0.71 Median 26.44 33.18 Standard Deviation 0.83 1.74 Sample Variance 0.69 3.03 Range 2.25 4.52 Minimum 24.70 31.71 Maximum 26.94 36.23 Count 6 6

PAGE 168

154Appendix A. (Continued) A-11 SIM6-GOP TEMPERATURE (C) SALINITY (psu) Mean 29.61 16.10 Standard Error 0.19 1.02 Median 29.48 16.28 Standard Deviation 0.51 2.70 Sample Variance 0.26 7.30 Range 1.39 8.94 Minimum 28.92 11.01 Maximum 30.31 19.95 Count 7 7 A-12 SIM6-SEC TEMPERATURE (C) SALINITY (psu) Mean 28.94 27.43 Standard Error 0.23 2.52 Median 28.72 27.84 Standard Deviation 0.56 6.17 Sample Variance 0.32 38.07 Range 1.46 17.16 Minimum 28.53 17.49 Maximum 29.99 34.64 Count 6 6

PAGE 169

155Appendix B. Statistics Summary for surface chlo rophyll-a (Chl-a) and total suspended matter (TSM) concentrations B-1 SIM1-GOP Chl-a (g l-1) TSM (mg l-1) Mean 2.43 3.20 Standard Error 0.12 0.25 Median 2.43 3.20 Standard Deviation 0.17 0.35 Sample Variance 0.03 0.13 Range 0.23 0.50 Minimum 2.31 2.95 Maximum 2.54 3.45 Count 2 2 B-2 SIM1-SEC Chl-a (g l-1) TSM (mg l-1) Mean 1.72 4.61 Standard Error 0.24 0.63 Median 1.94 3.68 Standard Deviation 0.71 1.79 Sample Variance 0.50 3.19 Range 2.27 4.40 Minimum 0.15 3.10 Maximum 2.42 7.50 Count 9 8

PAGE 170

156Appendix B. (Continued) B-3 SIM2-GOP Chl-a (g l-1) TSM (mg l-1) Mean 1.14 3.66 Standard Error 0.18 0.42 Median 1.01 3.75 Standard Deviation 0.50 1.20 Sample Variance 0.25 1.43 Range 1.50 3.69 Minimum 0.62 2.25 Maximum 2.12 5.94 Count 8 8 B-4 SIM2-SEC Chl-a (g l-1) TSM (mg l-1) Mean 1.14 3.74 Standard Error 0.17 0.23 Median 0.96 3.80 Standard Deviation 0.48 0.60 Sample Variance 0.23 0.36 Range 1.18 1.85 Minimum 0.63 2.60 Maximum 1.80 4.45 Count 8 7

PAGE 171

157Appendix B. (Continued) B-5 SIM3-GOP Chl-a (g l-1) TSM (mg l-1) Mean 1.61 53.69 Standard Error 0.12 19.32 Median 1.66 39.00 Standard Deviation 0.28 47.32 Sample Variance 0.08 2238.78 Range 0.72 116.67 Minimum 1.15 8.33 Maximum 1.87 125.00 Count 5 6 Extreme value: 8.11 mg l-1 (SIM3_11) Not included in statistics B-6 SIM3-SEC Chl-a (g l-1) TSM (g l-1) Mean 1.43 8.17 Standard Error 0.58 1.35 Median 1.10 9.20 Standard Deviation 1.30 3.01 Sample Variance 1.69 9.06 Range 3.35 7.75 Minimum 0.17 3.05 Maximum 3.52 10.80 Count 5 5

PAGE 172

158Appendix B. (Continued) B-7 SIM4-GOP Chl-a (g l-1) TSM (mg l-1) Mean 1.25 N/A Standard Error 0.28 N/A Median 1.30 N/A Standard Deviation 0.73 N/A Sample Variance 0.54 N/A Range 1.99 N/A Minimum 0.39 N/A Maximum 2.38 N/A Count 7 0 B-8 SIM4-SEC Chl-a (g l-1) TSM (mg l-1) Mean 1.25 N/A Standard Error 0.46 N/A Median 1.06 N/A Standard Deviation 1.03 N/A Sample Variance 1.07 N/A Range 2.73 N/A Minimum 0.28 N/A Maximum 3.00 N/A Count 5 0

PAGE 173

159Appendix B. (Continued) B-9 SIM5-GOP Chl-a (g l-1) TSM (mg l-1) Mean 0.92 8.38 Standard Error 0.13 3.25 Median 1.03 7.52 Standard Deviation 0.32 7.97 Sample Variance 0.10 63.45 Range 0.76 22.10 Minimum 0.51 1.50 Maximum 1.27 23.60 Count 6 6 B-10 SIM5-SEC Chl-a (g l-1) TSM (mg l-1) Mean 0.67 3.72 Standard Error 0.17 0.12 Median 0.59 3.65 Standard Deviation 0.43 0.30 Sample Variance 0.18 0.09 Range 1.29 0.85 Minimum 0.18 3.35 Maximum 1.47 4.20 Count 6 6

PAGE 174

160Appendix B. (Continued) B-11 SIM6-GOP Chl-a (g l-1) TSM (mg l-1) Mean 1.12 10.83 Standard Error 0.14 4.86 Median 1.09 5.96 Standard Deviation 0.36 12.86 Sample Variance 0.13 165.35 Range 0.90 37.72 Minimum 0.62 0.48 Maximum 1.52 38.20 Count 7 7 B-12 SIM6-SEC Chl-a (g l-1) TSM (g l-1) Mean 0.65 5.03 Standard Error 0.23 1.48 Median 0.45 3.29 Standard Deviation 0.57 3.63 Sample Variance 0.33 13.19 Range 1.36 8.91 Minimum 0.17 2.37 Maximum 1.53 11.28 Count 6 6

PAGE 175

161Appendix C. Statistics Summary for su rface nutrient concentrations C-1 SIM1-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.06 21.77 0.03 0.20 0.34 Standard Error 0.01 0.21 0.03 0.08 0.14 Median 0.06 21.77 0.03 0.20 0.34 Standard Deviation 0.02 0.30 0.04 0.11 0.20 Sample Variance 0.00 0.09 0.00 0.01 0.04 Minimum 0.05 21.56 0.00 0.12 0.20 Maximum 0.07 21.98 0.06 0.28 0.48 Count 2 2 2 2 2 C-2 SIM1-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.13 10.75 0.20 0.67 0.21 Standard Error 0.03 2.56 0.09 0.29 0.03 Median 0.11 12.21 0.12 0.44 0.22 Standard Deviation 0.07 6.27 0.21 0.72 0.06 Sample Variance 0.00 39.35 0.04 0.52 0.00 Range 0.19 16.63 0.52 1.88 0.15 Minimum 0.05 0.00 0.02 0.00 0.13 Maximum 0.24 16.63 0.53 1.88 0.28 Count 6 6 6 6 6

PAGE 176

162Appendix C. (continued) C-3 SIM2-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.08 43.66 0.07 1.21 0.73 Standard Error 0.03 4.55 0.02 0.41 0.14 Median 0.08 45.17 0.06 1.25 0.58 Standard Deviation 0.08 12.87 0.06 1.16 0.41 Sample Variance 0.01 165.72 0.00 1.34 0.17 Range 0.21 31.03 0.19 2.68 1.10 Minimum 0.00 27.17 0.00 0.00 0.33 Maximum 0.21 58.21 0.19 2.68 1.42 Count 8 8 8 8 8 C-4 SIM2-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.04 14.92 0.03 0.15 0.21 Standard Error 0.02 3.07 0.01 0.07 0.04 Median 0.00 15.25 0.02 0.12 0.25 Standard Deviation 0.06 8.12 0.03 0.18 0.11 Sample Variance 0.00 65.95 0.00 0.03 0.01 Range 0.13 22.24 0.06 0.48 0.31 Minimum 0.00 4.16 0.00 0.00 0.00 Maximum 0.13 26.39 0.06 0.48 0.31 Count 7 7 7 7 7

PAGE 177

163Appendix C. (continued) C-5 SIM3-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.24 25.86 0.27 0.78 0.27 Standard Error 0.02 2.51 0.06 0.36 0.02 Median 0.24 23.49 0.25 0.45 0.28 Standard Deviation 0.06 6.14 0.14 0.88 0.05 Sample Variance 0.00 37.70 0.02 0.77 0.00 Range 0.15 16.50 0.40 2.28 0.14 Minimum 0.16 21.18 0.12 0.11 0.19 Maximum 0.31 37.67 0.52 2.39 0.34 Count 6 6 6 6 6 C-6 SIM3-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.23 11.85 0.14 0.54 0.08 Standard Error 0.04 3.16 0.05 0.15 0.05 Median 0.22 14.98 0.09 0.55 0.00 Standard Deviation 0.09 7.07 0.12 0.34 0.11 Sample Variance 0.01 49.96 0.01 0.12 0.01 Range 0.24 15.22 0.31 0.91 0.23 Minimum 0.12 2.80 0.02 0.00 0.00 Maximum 0.37 18.02 0.33 0.91 0.23 Count 5 5 5 5 5

PAGE 178

164Appendix C. (continued) C-7 SIM4-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.11 33.14 0.11 0.64 0.64 Standard Error 0.04 1.33 0.02 0.21 0.21 Median 0.07 33.07 0.09 0.55 0.55 Standard Deviation 0.09 3.51 0.06 0.54 0.54 Sample Variance 0.01 12.31 0.00 0.30 0.30 Range 0.27 10.55 0.17 1.30 1.30 Minimum 0.04 28.77 0.03 0.00 0.00 Maximum 0.31 39.32 0.20 1.30 1.30 Count 7 7 7 7 7 C-8 SIM4-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.10 16.97 0.12 0.50 0.91 Standard Error 0.04 4.56 0.04 0.37 0.04 Median 0.11 19.60 0.11 0.21 0.92 Standard Deviation 0.08 10.20 0.09 0.82 0.10 Sample Variance 0.01 104.12 0.01 0.67 0.01 Range 0.21 25.46 0.21 1.95 0.25 Minimum 0.01 5.40 0.02 0.00 0.76 Maximum 0.22 30.85 0.23 1.95 1.01 Count 5 5 5 5 5

PAGE 179

165Appendix C. (continued) C-9 SIM5-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.19 10.23 0.07 0.26 0.38 Standard Error 0.03 0.83 0.02 0.06 0.31 Median 0.22 9.92 0.07 0.29 0.10 Standard Deviation 0.08 2.03 0.04 0.15 0.76 Sample Variance 0.01 4.11 0.00 0.02 0.57 Range 0.23 5.37 0.09 0.40 1.92 Minimum 0.02 8.22 0.02 0.05 0.00 Maximum 0.25 13.59 0.11 0.45 1.92 Count 6 6 6 6 6 C-10 SIM5-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.23 6.15 0.16 0.90 0.08 Standard Error 0.03 2.11 0.04 0.17 0.02 Median 0.24 7.59 0.13 0.93 0.07 Standard Deviation 0.06 4.21 0.09 0.35 0.04 Sample Variance 0.00 17.75 0.01 0.12 0.00 Range 0.13 9.37 0.19 0.66 0.09 Minimum 0.15 0.03 0.09 0.54 0.05 Maximum 0.28 9.40 0.28 1.20 0.14 Count 4 4 4 4 4

PAGE 180

166Appendix C. (continued) C-11 SIM6-GOP P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.00 33.09 0.04 0.55 0.28 Standard Error 0.00 3.09 0.01 0.30 0.11 Median 0.00 31.64 0.02 0.00 0.14 Standard Deviation 0.01 8.17 0.03 0.79 0.30 Sample Variance 0.00 66.71 0.00 0.63 0.09 Range 0.02 23.87 0.07 1.77 0.69 Minimum 0.00 25.24 0.01 0.00 0.00 Maximum 0.02 49.11 0.08 1.77 0.69 Count 7 7 7 7 7 C-12 SIM6-SEC P04 ( g l -1 ) Si(OH)4 ( g l -1 ) NO2 ( g l -1 ) NO3 ( g l -1 ) NH4 ( g l -1 ) Mean 0.00 17.10 0.01 0.00 0.16 Standard Error 0.00 3.88 0.00 0.00 0.04 Median 0.00 17.23 0.01 0.00 0.13 Standard Deviation 0.00 9.50 0.01 0.00 0.10 Sample Variance 0.00 90.23 0.00 0.00 0.01 Range 0.01 24.53 0.01 0.00 0.27 Minimum 0.00 4.39 0.00 0.00 0.05 Maximum 0.01 28.92 0.01 0.00 0.32 Count 6 6 6 6 6

PAGE 181

167Appendix D. Statistics Summary fo r surface phytoplankton (aph()) and detritus (ad()) absorption coefficients in the blue wavelengths D-1 SIM1-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.048 0.050 0.035 0.055 0.018 0.032 Standard Error 0.001 0.000 0.001 0.000 0.000 0.001 Median 0.048 0.050 0.035 0.055 0.018 0.032 Standard Deviation 0.002 0.000 0.001 0.000 0.001 0.001 Sample Variance 0.000 0.000 0.000 0.000 0.000 0.000 Range 0.002 0.000 0.002 0.001 0.001 0.001 Minimum 0.047 0.050 0.034 0.054 0.017 0.031 Maximum 0.049 0.050 0.036 0.055 0.018 0.032 Count 2 2 2 2 2 2 D-2 SIM1-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.071 0.045 0.051 0.051 0.027 0.029 Standard Error 0.034 0.005 0.025 0.005 0.015 0.003 Median 0.038 0.044 0.027 0.054 0.013 0.030 Standard Deviation 0.084 0.011 0.062 0.011 0.036 0.008 Sample Variance 0.007 0.000 0.004 0.000 0.001 0.000 Range 0.210 0.030 0.154 0.028 0.088 0.019 Minimum 0.032 0.032 0.024 0.037 0.012 0.019 Maximum 0.242 0.061 0.178 0.065 0.100 0.038 Count 6 6 6 6 6 6

PAGE 182

168Appendix D. (continued) D-3 SIM2-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.171 0.036 0.127 0.039 0.076 0.021 Standard Error 0.050 0.003 0.038 0.003 0.023 0.002 Median 0.152 0.036 0.112 0.036 0.066 0.019 Standard Deviation 0.143 0.008 0.106 0.010 0.066 0.006 Sample Variance 0.020 0.000 0.011 0.000 0.004 0.000 Range 0.385 0.025 0.290 0.028 0.179 0.018 Minimum 0.024 0.026 0.018 0.029 0.009 0.015 Maximum 0.409 0.051 0.307 0.057 0.189 0.032 Count 8 8 8 8 8 8 D-4 SIM2-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.033 0.037 0.024 0.043 0.013 0.024 Standard Error 0.008 0.003 0.006 0.003 0.003 0.002 Median 0.028 0.040 0.020 0.045 0.011 0.025 Standard Deviation 0.022 0.008 0.016 0.009 0.009 0.005 Sample Variance 0.000 0.000 0.000 0.000 0.000 0.000 Range 0.060 0.023 0.044 0.024 0.025 0.014 Minimum 0.007 0.024 0.005 0.030 0.003 0.017 Maximum 0.067 0.047 0.049 0.053 0.027 0.031 Count 7 7 7 7 7 7

PAGE 183

169Appendix D. (continued) D-5 SIM3-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.257 0.109 0.185 0.140 0.116 0.079 Standard Error 0.110 0.068 0.079 0.082 0.047 0.043 Median 0.220 0.042 0.165 0.064 0.107 0.033 Standard Deviation 0.268 0.167 0.194 0.202 0.116 0.106 Sample Variance 0.072 0.028 0.038 0.041 0.013 0.011 Range 0.739 0.436 0.535 0.515 0.317 0.270 Minimum 0.015 0.009 0.009 0.035 0.007 0.024 Maximum 0.754 0.445 0.545 0.550 0.324 0.294 Count 6 6 6 6 6 6 D-6 SIM3-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.038 0.067 0.026 0.086 0.014 0.052 Standard Error 0.015 0.040 0.011 0.047 0.006 0.027 Median 0.024 0.035 0.014 0.042 0.009 0.025 Standard Deviation 0.034 0.089 0.024 0.106 0.012 0.061 Sample Variance 0.001 0.008 0.001 0.011 0.000 0.004 Range 0.075 0.219 0.051 0.252 0.029 0.144 Minimum 0.003 0.005 0.002 0.021 0.002 0.015 Maximum 0.078 0.223 0.053 0.273 0.030 0.159 Count 5 5 5 5 5 5

PAGE 184

170Appendix D. (continued) D-7 SIM4-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.322 0.043 0.233 0.051 0.137 0.029 Standard Error 0.155 0.011 0.114 0.012 0.069 0.008 Median 0.106 0.033 0.074 0.040 0.041 0.023 Standard Deviation 0.411 0.028 0.300 0.032 0.182 0.020 Sample Variance 0.169 0.001 0.090 0.001 0.033 0.000 Range 0.991 0.078 0.728 0.090 0.442 0.058 Minimum 0.038 0.019 0.026 0.023 0.012 0.013 Maximum 1.029 0.097 0.754 0.113 0.454 0.071 Count 7 7 7 7 7 7 D-8 SIM5-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.047 0.036 0.033 0.045 0.018 0.023 Standard Error 0.020 0.009 0.014 0.010 0.008 0.006 Median 0.033 0.029 0.023 0.035 0.012 0.018 Standard Deviation 0.045 0.020 0.032 0.023 0.018 0.014 Sample Variance 0.002 0.000 0.001 0.001 0.000 0.000 Range 0.109 0.052 0.079 0.059 0.043 0.035 Minimum 0.016 0.014 0.010 0.023 0.006 0.011 Maximum 0.125 0.066 0.089 0.082 0.049 0.046 Count 5 5 5 5 5 5

PAGE 185

171Appendix D. (continued) D-9 SIM5-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.105 0.046 0.074 0.055 0.043 0.031 Standard Error 0.054 0.010 0.039 0.010 0.023 0.006 Median 0.070 0.045 0.049 0.060 0.027 0.034 Standard Deviation 0.132 0.025 0.095 0.025 0.056 0.014 Sample Variance 0.017 0.001 0.009 0.001 0.003 0.000 Range 0.355 0.071 0.257 0.066 0.151 0.038 Minimum 0.006 0.019 0.003 0.026 0.002 0.014 Maximum 0.362 0.090 0.260 0.092 0.153 0.051 Count 6 6 6 6 6 6 D-10 SIM5-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.032 0.030 0.023 0.039 0.012 0.021 Standard Error 0.009 0.010 0.006 0.012 0.004 0.007 Median 0.032 0.024 0.023 0.030 0.013 0.018 Standard Deviation 0.022 0.025 0.016 0.030 0.009 0.018 Sample Variance 0.000 0.001 0.000 0.001 0.000 0.000 Range 0.048 0.069 0.036 0.083 0.019 0.050 Minimum 0.008 0.009 0.005 0.016 0.002 0.007 Maximum 0.057 0.078 0.041 0.099 0.022 0.057 Count 6 6 6 6 6 6

PAGE 186

172Appendix D. (continued) D-11 SIM6-GOP ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.225 0.053 0.165 0.052 0.099 0.031 Standard Error 0.092 0.008 0.069 0.005 0.043 0.003 Median 0.058 0.044 0.039 0.048 0.020 0.031 Standard Deviation 0.244 0.021 0.182 0.013 0.113 0.007 Sample Variance 0.060 0.000 0.033 0.000 0.013 0.000 Range 0.533 0.055 0.390 0.035 0.243 0.021 Minimum 0.026 0.033 0.017 0.035 0.007 0.021 Maximum 0.558 0.088 0.407 0.070 0.250 0.042 Count 7 7 7 7 7 7 D-12 SIM6-SEC ad(412) (m-1) aph(412) (m-1) ad(440) (m-1) aph(440) (m-1) ad(492) (m-1) aph(492) (m-1) Mean 0.025 0.026 0.018 0.028 0.009 0.016 Standard Error 0.011 0.008 0.008 0.009 0.004 0.005 Median 0.018 0.017 0.012 0.020 0.005 0.011 Standard Deviation 0.028 0.020 0.020 0.022 0.011 0.012 Sample Variance 0.001 0.000 0.000 0.000 0.000 0.000 Range 0.072 0.054 0.052 0.057 0.028 0.033 Minimum 0.004 0.011 0.003 0.013 0.001 0.007 Maximum 0.076 0.065 0.055 0.070 0.029 0.040 Count 6 6 6 6 6 6

PAGE 187

173Appendix E. Statistics Summary for surface Gelbstoff absorption coefficient at 440 nm (ag(440)) and spectral slope E-1 SIM1-GOP ag(440) (m-1) Slope (nm-1) Mean 1.76 0.012 Standard Error 0.62 0.002 Median 1.76 0.012 Standard Deviation 0.88 0.003 Sample Variance 0.77 0.000 Range 1.24 0.004 Minimum 1.14 0.010 Maximum 2.38 0.014 Count 2 2 E-2 SIM1-SEC ag(440) (m-1) Slope (nm-1) Mean 0.71 0.014 Standard Error N/A N/A Median N/A N/A Standard Deviation N/A N/A Sample Variance N/A N/A Range N/A N/A Minimum N/A N/A Maximum N/A N/A Count 1 1

PAGE 188

174Appendix E. (continued) E-3 SIM2-GOP ag(440) (m-1) Slope (nm-1) Mean 1.25 0.015 Standard Error 0.15 0.001 Median 1.33 0.015 Standard Deviation 0.42 0.001 Sample Variance 0.18 0.000 Range 1.42 0.005 Minimum 0.40 0.013 Maximum 1.82 0.018 Count 8 8 E-4 SIM2-SEC ag(440) (m-1) Slope (nm-1) Mean 0.59 0.016 Standard Error 0.08 0.000 Median 0.56 0.016 Standard Deviation 0.16 0.000 Sample Variance 0.02 0.000 Range 0.32 0.001 Minimum 0.45 0.016 Maximum 0.77 0.017 Count 4 4

PAGE 189

175Appendix E. (continued) E-5 SIM3-GOP ag(440) (m-1) Slope (nm-1) Mean 1.92 0.013 Standard Error 0.18 0.000 Median 1.98 0.013 Standard Deviation 0.45 0.001 Sample Variance 0.20 0.000 Range 1.25 0.001 Minimum 1.24 0.013 Maximum 2.48 0.014 Count 6 6 E-6 SIM3-SEC ag(440) (m-1) Slope (nm-1) Mean 1.49 0.014 Standard Error 0.16 0.001 Median 1.63 0.014 Standard Deviation 0.28 0.001 Sample Variance 0.08 0.000 Range 0.50 0.002 Minimum 1.17 0.013 Maximum 1.67 0.015 Count 3 3

PAGE 190

176Appendix E. (continued) E-7 SIM4-GOP ag(440) (m-1) Slope (nm-1) Mean 1.74 0.014 Standard Error 0.38 0.000 Median 1.16 0.015 Standard Deviation 1.01 0.001 Sample Variance 1.01 0.000 Range 2.18 0.002 Minimum 1.02 0.013 Maximum 3.21 0.015 Count 7 7 E-8 SIM4-SEC ag(440) (m-1) Slope (nm-1) Mean 1.110 0.015 Standard Error N/A N/A Median N/A N/A Standard Deviation N/A N/A Sample Variance N/A N/A Range N/A N/A Minimum N/A N/A Maximum N/A N/A Count 1 1

PAGE 191

177Appendix E. (continued) E-9 SIM5-GOP ag(440) (m-1) Slope (nm-1) Mean 1.17 0.015 Standard Error 0.50 0.001 Median 0.83 0.015 Standard Deviation 0.99 0.002 Sample Variance 0.99 0.000 Range 2.18 0.005 Minimum 0.41 0.012 Maximum 2.59 0.017 Count 4 4 E-10 SIM5-SEC ag(440) (m-1) Slope (nm-1) Mean 0.24 0.020 Standard Error 0.01 0.000 Median 0.24 0.020 Standard Deviation 0.01 0.001 Sample Variance 0.00 0.000 Range 0.01 0.001 Minimum 0.23 0.019 Maximum 0.25 0.020 Count 2 2

PAGE 192

178Appendix E. (continued) D-11 SIM6-GOP ag(440) (m-1) Slope (nm-1) Mean 1.87 0.014 Standard Error 0.24 0.000 Median 1.68 0.014 Standard Deviation 0.54 0.001 Sample Variance 0.29 0.000 Range 1.27 0.001 Minimum 1.43 0.013 Maximum 2.70 0.014 Count 5 5 D-12 SIM6-SEC ag(440) (m-1) Slope (nm-1) Mean 1.27 0.014 Standard Error 0.27 0.000 Median 1.27 0.014 Standard Deviation 0.38 0.000 Sample Variance 0.15 0.000 Range 0.54 0.000 Minimum 1.00 0.014 Maximum 1.54 0.014 Count 2 2

PAGE 193

179Appendix F. Statistics Summary for su rface contributions (%) of Gelbstoff absorption (ag(440)), detritus absorption (ad(440)), and phytoplankton absorption (aph(440)) to the total absorption coefficient (a(440)) at 440 nm F-1 SIM1-GOP ag(440) % ad(440) % aph(440) % Mean 94.17 2.13 3.31 Standard Error 2.00 0.75 1.12 Median 94.17 2.13 3.31 Standard Deviation 2.83 1.07 1.58 Sample Variance 7.99 1.14 2.49 Range 4.00 1.51 2.23 Minimum 92.17 1.38 2.20 Maximum 96.17 2.89 4.43 Count 2 2 2 F-2 SIM1-SEC ag(440) % ad(440) % aph(440) % Mean 74.79 18.82 5.72 Standard Error N/A N/A N/A Median N/A N/A N/A Standard Deviation N/A N/A N/A Sample Variance N/A N/A N/A Range N/A N/A N/A Minimum N/A N/A N/A Maximum N/A N/A N/A Count 1 1 1

PAGE 194

180Appendix F. (continued) F-3 SIM2-GOP ag(440) % ad(440) % aph(440) % Mean 88.42 7.77 3.26 Standard Error 1.34 1.67 0.59 Median 88.27 7.46 2.70 Standard Deviation 3.80 4.71 1.68 Sample Variance 14.41 22.21 2.82 Range 11.51 12.75 4.80 Minimum 84.02 1.41 1.53 Maximum 95.53 14.16 6.33 Count 8 8 8 F-4 SIM2-SEC ag(440) % ad(440) % aph(440) % Mean 87.71 3.97 7.33 Standard Error 0.75 0.20 0.78 Median 87.95 3.92 7.26 Standard Deviation 1.51 0.40 1.56 Sample Variance 2.28 0.16 2.44 Range 3.63 0.96 3.81 Minimum 85.66 3.54 5.48 Maximum 89.29 4.50 9.30 Count 4 4 4

PAGE 195

181Appendix F. (continued) F-5 SIM3-GOP ag(440) % ad(440) % aph(440) % Mean 85.57 8.06 6.07 Standard Error 4.24 2.85 3.12 Median 84.75 8.82 3.17 Standard Deviation 10.38 6.98 7.64 Sample Variance 107.67 48.71 58.40 Range 27.30 19.07 19.97 Minimum 70.59 0.37 1.24 Maximum 97.89 19.44 21.21 Count 6 6 6 F-6 SIM3-SEC ag(440) % ad(440) % aph(440) % Mean 94.98 1.52 3.08 Standard Error 1.13 0.71 1.30 Median 95.56 1.14 2.50 Standard Deviation 1.96 1.23 2.24 Sample Variance 3.85 1.51 5.03 Range 3.79 2.37 4.37 Minimum 92.80 0.53 1.19 Maximum 96.59 2.89 5.56 Count 3 3 3

PAGE 196

182Appendix F. (continued) F-7 SIM4-GOP ag(440) % ad(440) % aph(440) % Mean 86.51 10.33 2.77 Standard Error 4.81 4.53 0.46 Median 91.68 5.30 2.77 Standard Deviation 12.73 11.99 1.21 Sample Variance 161.99 143.78 1.47 Range 35.93 32.16 3.68 Minimum 60.95 2.15 0.70 Maximum 96.88 34.30 4.37 Count 7 7 7 F-8 SIM4-SEC ag(440) % ad(440) % aph(440) % Mean 93.71 2.81 2.95 Standard Error N/A N/A N/A Median N/A N/A N/A Standard Deviation N/A N/A N/A Sample Variance N/A N/A N/A Range N/A N/A N/A Minimum N/A N/A N/A Maximum N/A N/A N/A Count 1 1 1

PAGE 197

183Appendix F. (continued) F-9 SIM5-GOP ag(440) % ad(440) % aph(440) % Mean 91.23 2.39 5.50 Standard Error 1.35 0.16 1.11 Median 90.26 2.24 6.13 Standard Deviation 2.70 0.32 2.21 Sample Variance 7.27 0.10 4.89 Range 6.01 0.66 5.11 Minimum 89.20 2.22 2.30 Maximum 95.20 2.88 7.41 Count 4 4 4 F-10 SIM5-SEC ag(440) % ad(440) % aph(440) % Mean 90.70 1.29 6.27 Standard Error 3.43 0.77 1.99 Median 90.70 1.29 6.27 Standard Deviation 4.85 1.09 2.81 Sample Variance 23.49 1.19 7.89 Range 6.85 1.54 3.97 Minimum 87.28 0.52 4.29 Maximum 94.13 2.06 8.26 Count 2 2 2

PAGE 198

184Appendix F. (continued) F-11 SIM6-GOP ag(440) % ad(440) % aph(440) % Mean 91.24 5.65 2.78 Standard Error 2.21 2.37 0.22 Median 93.44 2.58 2.75 Standard Deviation 4.94 5.29 0.50 Sample Variance 24.42 28.04 0.25 Range 10.72 11.44 1.37 Minimum 84.86 1.30 2.20 Maximum 95.58 12.74 3.57 Count 5 5 5 F-12 SIM6-SEC ag(440) % ad(440) % aph(440) % Mean 96.99 0.82 1.68 Standard Error 0.77 0.45 0.43 Median 96.99 0.82 1.68 Standard Deviation 1.09 0.64 0.60 Sample Variance 1.18 0.41 0.36 Range 1.54 0.91 0.85 Minimum 96.22 0.37 1.25 Maximum 97.76 1.28 2.11 Count 2 2 2


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 2200421Ka 4500
controlfield tag 006 m d s
007 cr mnu
008 041209s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000565
035
(OCoLC)57677678
9
AJU6882
b SE
040
FHM
c FHM
090
GC11.2 (ONLINE)
1 100
Odriozola, Ana L.
0 245
On the color of the Orinoco River plume
h [electronic resource] /
by Ana L. Odriozola.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (M.S.)--University of South Florida, 2004.
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 198 pages.
520
ABSTRACT: In situ measurements were used to study the bio-optical properties of marine waters within the Gulf of Paria (GOP, Venezuela) and in the Southeastern Caribbean Sea (SEC) as they are affected by the seasonal discharge of the Orinoco River plume. The main purpose of this study was to determine the impact of colored dissolved organic matter (CDOM) (also known as Gelbstoff), phytoplankton, and total suspended matter (TSM) in the color of the Orinoco River plume. This information is essential for regional ocean color algorithms development. Salinity and silica values indicate that the GOP and SEC waters were under the influence of the Orinoco River plume during both seasons. This riverine influence resulted in high values of Gelbstoff absorption, αg (λ), which contributed to up to 90% of the total absorption at 440 nm in both the GOP and SEC regardless of the season.Phytoplankton absorption contributions were normally around 5%, but during the dry season these values reached 20% in the SEC. Ratios of αg(440) to αph(440) were extremely large, with most of the values ranging from 10 to 50. Due to the strong absorption by Gelbstoff, light at the blue wavelengths (412 nm, 440 nm and 490 nm) was attenuated to 1% of the subsurface irradiance in the first 5 m of the water column within the GOP, and in the first 10 m of the water column in the SEC. Furthermore, the absorption by Gelbstoff significantly decreased the water leaving radiance (Lw(λ)) in the blue wavelengths along the Orinoco River plume.As αg(λ) relatively decreased from the GOP to the SEC X≈1.6 m-1 and X≈ 0.9 m-1, respectively), a shift in the maximum peak of Rrs(λ) spectra (Rrsmax(λ)), towards shorter wavelengths (from ~ 580 nm to ~500 nm) was observed. Similar to Gelbstoff, concentrations of TSM normally decreased from the stations near the Delta to the stations in the SEC. The impact of TSM on the color of the Orinoco plume was represented by a reduction in the magnitude of Rrsmax(λ) of ~50% going from the waters near the Orinoco delta to the SEC, indistinctively of the season.
590
Adviser: Mller-Karger, Frank E.
653
Cdom.
Tsm.
Bio-optical properties.
Chlorophyll.
Ocean color.
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
949
FTS
SFERS
ETD
GC11.2 (ONLINE)
nkt 1/26/05
4 856
u http://digital.lib.usf.edu/?e14.565