Chapter 2 Literature Review

The concentration of Chl-a can provide insight about the environmental conditions of an area. Variation in the concentration of these pigments in the water column is mainly caused by the abundance of phytoplankton in the water column (Yoder et al. 2001).

2.1 Estimation of Phytoplankton Biomass From Space

Phytoplankton can be detected by satellite sensors because their photosynthetic pigment (Chl-a) absorbs blue, violet and red light and reflects green (Joint and Groom, 2000). Light from the sun enters the water and interacts (part of it being either scattered, absorbed or reflected) with the water itself or other water constituents such as suspended or dissolved matter [Robinson (2004). The fraction of light which is reflected back to the sensor is the one being used for remote sensing purposes. The measured radiance may be quantitatively related to various constituents in the water column that interact with visible light (Fett 2000).

Colour change can be observed using a radiometer measuring the reflected light in the visible part of the electromagnetic spectrum between 400-700 nm wavelength; red (650 nm), green (520 nm) and blue (450 nm) (Robinson 2004). Light in the range 400-700 nm is known as photosynthetically active radiation (PAR).

Satellite remote sensing of Chl-a concentrations rely on the absorption and scattering characteristics of phytoplankton and the way these optical properties affect the underwater light field and the reflectance or radiance values measured by the satellite sensor (Brando et al. 2006). Information on Chl-a concentrations is obtained using appropriate algorithms that relate measurements of remote sensing reflectance spectra either directly to Chl-a concentration (empirical algorithms), or to optical properties of phytoplankton and other optically active materials in the water using semi analytical algorithms or based on radiative transfer and theoretical relationships (Mansor et al. 2001).

Satellite data can be used to characterize the habitat and ecosystem properties that influence marine resources at large, temporal and spatial scales, and high temporal and spatial resolution. There are two principal ways of using satellite data for fisheries management. The first is based on environmental monitoring; with a view to better understand ecosystem processes or stock biology (Solanki et al. 2003). The other is based on locating fish populations with a view to increase fishing efficiency or enhances conservation by mitigating human interactions (Solanki et al. 2003). Additionally, satellite data are used to monitor a number of issues that impact fisheries, such as harmful algal blooms and coastal pollution (Solanki et al. 2003).

Satellite chlorophyll-a measurements are the primary or principal components of the algorithms used to calculate the primary productivity (PP) of the ocean (Pauly and Christensen 1995). Global PP measurements, in conjunction with fish catch statistics and food web models, can be used to estimate the carrying capacity of the world’s fisheries (Pauly and Christensen 1995). In the open ocean, 2% of the PP is needed to support the fishery catch. However in coastal regions, the requirement ranges from 24-35%, suggesting that these systems are at or beyond their carrying capacity (Pauly and Christensen 1995). The exhaustive capacity of coastal waters to support fishery is among the major issues of concern for fishery management as the bulk of the world’s fish catch comes from coastal areas.

In other studies, discrepancies between the values of satellite derived PP and reported fish catches have been used to demonstrate spurious trends in global fish catches (Watson and Pauly 2001). Thus, satellite Chl-a data provides important objective baseline information which could be used to manage coastal and marine socio-economic activities.

2.2 The History of Satellite Ocean Colour

The earliest sensor that has been used for mapping and measuring Chl-a concentration was Coastal Zone Colour Scanner (CZCS) (Babin, Morel, and Gentili 1996b). Realizing the importance of synoptic global observation of marine parameters, several ocean colour sensors were then launched; Sea-viewing Wide Field-of-view Sensor (SeaWIFS) in July 1997; Moderate Resolution Imaging Spectroradiometer (MODIS Terra) in December 1999 and MODIS Aqua in May 2002 (Babin, Morel, and Gentili 1996b). The Medium Resolution Imaging Spectro-meter (MERIS) was launched by Europe and it started to operate in March 2002. The MODIS sensor is described in detail below since the data from this sensor was used in this thesis.

2.3 Moderate-resolution Imaging Spectrometer (MODIS) Aqua/Terra

The MODIS sensors are placed on NASA’s EOS Terra and Aqua satellites. The sensor yields simultaneous observations of high atmospheric, oceanic (Sea Surface Temperature and chlorophyll) and land surface features (Barnes, Pagano, and Salomonson 1998). MODIS is placed in a 705 km sun-synchronous orbit. The temporal resolution is 1-2 days and has a field of view of \(\pm\) 55º off-nadir, which provides a swath width of 2330 km. MODIS is a whiskbroom scanning imaging radiometer consisting of a cross-tracking scan mirror, collecting optics, and a set of linear array of detector with spectral interference filters located in four planes. The sensor collects data in 36 spectral bands and provides a spatial resolution from 250 × 250 m (bands 1 and 2) to 500 × 500 m (bands 3 through 7) and 1 × 1 km (bands 8 through 36) (Barnes, Pagano, and Salomonson 1998).

2.4 Spatial and Temporal Variation of Phytoplankton Biomass

Phytoplankton biomass and composition varies with space and time (Raymont 1980). In temperate zones, phytoplankton biomass is characterized by high biomass and predominance of few species (Lugomela 1996). In contrast, the phytoplankton biomass in tropical areas is characterized by low biomass and a relatively higher richness of species. Phytoplankton biomass and productivity are also highly variable on short time scales as a result of short term fluctuations in the water column stability (Raymont 1980). These are generated by physical processes which affect the progression of seasonal conditions.

Phytoplankton biomass in the coastal waters of Tanzania is generally high during the southeast (SE) monsoon season, particularly when it is raining and low during the northeast (NE) monsoon season, particularly when it is dry (McClanahan 1988). The higher concentration during the SE monsoon season is mainly attributed to relatively stronger mixing of the water column, greater runoff and nutrient input from rivers and greater availability of biologically assimilable nitrogen. Consequently, the total fish catch and reproduction rate are highest during this period (McClanahan 1988).

2.5 Factors Affecting Phytoplankton Variations

There are several environmental (physico-chemical) variables which influence the distribution pattern of phytoplankton (Raymont 1980). The chemical variables include nutrients (nitrates, phosphate, silicates), salinity, pH, and dissolved oxygen (DO). The physical variables include temperature, water circulation, turbidity and light. In addition to these, biological factors such as grazing, competition and diseases also influence the distribution of phytoplankton (Lugomela 1996). Moreover there are three major environmental variables that control photosynthesis in water bodies, which include light, temperature and nutrient availability (Peterson et al. 1987), and there is a general correlation between the nutrient status of a body of water and its primary productivity (Riley and Prepas 1985).

2.5.1 Nutrients

Availability of nutrient in the water column is one of the major environmental factors limiting phytoplankton abundance in the ocean. Although a wide range of chemical elements are required by the phytoplankton, nitrogen and phosphorus are the most critical elements as they are needed in quite large amounts, but are generally present in low concentrations in sea water (Kitheka et al. 1996). Several studies show a limiting effect of phosphorus for phytoplankton growth in temperate lakes and reservoirs but in most tropical regions, nitrogen is reported to be among the most critical nutrient element needed for phytoplankton growth (Hupfer and Lewandowski 2008). Generally, inorganic nitrogen is a limiting nutrient in the oceans while inorganic phosphorus limits phytoplankton growth in freshwater environments (Kitheka et al. 1996).

Nutrient levels are also higher than normal along the coast for tropical seawaters indicating anthropogenic inputs. Concentration values ranging from 3.75 to 15.17 µg/l PO_4^3-, 26 µg/l NH⁴⁺ and 1.3 to 7 µg/l NO^3- have been reported (Mohammed, Ngusaru, and Mwaipopo 1993). Conversely, Kyewalyanga (2002) found a relatively higher phytoplankton biomass of about 12.14 mgm^3 at Chwaka bay Zanzibar.

In aquatic environment along the tropics, phytoplankton biomass is generally higher during the rainy season of SE monsoon. Higher productivity in this season is mainly attributed by greater runoff and nutrient input from rivers and availability of biologically assimilable nitrogen (Wang et al., 2006).

2.5.2 Sea Surface Temperature (SST)

Sea surface temperature (SST) is a very important environmental parameter for phytoplankton growth. The phytoplankton thrives only when the sea water is within the optimum temperature ranges (McClanahan 1988). The mixing processes in the ocean are generally influenced by the temperature conditions of the sea surface water. When surface waters are cold, it is easier for deeper water to rise to the surface, bringing nutrients to sunlit areas where phytoplankton can use them (McClanahan 1988). When surface water is warm, cooler, nutrient-rich water is trapped below. Because the vertical layers of the ocean are not mixing, nutrients that have built up in deep waters cannot reach the surface.

In places where ocean currents cause upwelling, SST is often cooler than nearby waters, and Chl-a concentration is higher. In coastal upwelling areas, the rising slope of the sea floor pushes cold water from the lowest layers of the ocean to the surface. The rising, or upwelling water carries iron and other nutrients from the ocean floor (McClanahan 1988). Cold coastal upwelling and subsequent phytoplankton growth are most evident along the west coasts of North and South America and Southern Africa (Raymont 1980).

Temperature of tropical sea depends on the weather conditions such as rainfall, relative humidity, air temperature, wind velocity and light radiation intensity. Sea surface temperature follows an annual cycle and is greatly influenced by the monsoon winds. Several studies (Lugomela 1996, Bryceson (1977); Julius 2005; Newell 1959) showed higher SST in the coastal waters of East Africa within the latitudes 0 - 10° S during the NE monsoon season and lowest during the SE monsoon season and exhibits seasonality that is influenced by changes in the water masses and by climate related factors.

The differences in temperature during the different monsoon seasons are caused by the South Equatorial Current which brings water of relatively low temperature from the Pacific Ocean during the southeast monsoon and draws in waters of high temperature during the northeast monsoon (Francis, Shigala, and Aledaide 2001). Wang et al. (2006) indicated that, sea surface water temperatures between 20°C and 30°C are favourable for the growth of the phytoplankton. The SST within Zanzibar and Pemba channels reach a minimum of 25°C in September and rises to a maximum of 29°C in March (McClanahan 1988). This indicates that phytoplankton production and growth is favoured throughout the year within this region given that other environmental variables are constant.

2.5.3 Salinity

Salinity is the mass fraction of salts in seawater (Fofonoff 1985). Salinity used to be expressed in parts per thousand. In practical terms, salinity is expressed as PSU (practical salinity units) which are based on water temperature and conductivity measurements (Fofonoff 1985). For oceanic seawater, ppt and PSU are very close. However PSU is unit-less, thus salinity now days has no unit.

In general, salinity values around Tanzanian waters are relatively low during April and May, following the peak in freshwater outflow, and highest in November. The salinity values start to decrease in February before the beginning of the rains (Nyandwi and Dubi 2001). Lowest salinities occur at the onset of the southeast monsoon when the river discharges, cloud cover and rainfall are high. Immediately following the rainy season, salinity can drop drastically in near shore areas and some studies had indicated measurements as low as 26 (McClanahan 1988). The highest salinities occur during northeast monsoon when air temperatures and solar insolation are high and rainfall and discharge low (McClanahan 1988).

In estuarine system salinity usually increases away from a freshwater source such as a river, although evaporation sometimes may cause the salinity at the head of the estuary to exceed that of seawater (Paula et al. 1998). The observed abnormality suggest that, rainfall is not the only driving force for salinity changes. The multiple factors that influence the sea surface salinity in near shore areas include surface runoff, dilution from rainfall and evaporation (Paula et al. 1998).

2.5.4 Dissolved Oxygen

Dissolved oxygen (DO) refers concentration refers to the amount of oxygen contained in water, and defines the living conditions for oxygen-requiring (aerobic) aquatic organisms. Oxygen has limited solubility in water, usually ranging from 6 to 14 mgl-1^ (Connell, Miller, and Miller 1984). DO concentrations reflect the equilibrium between oxygen-producing processes such as photosynthesis and oxygen-consuming processes (aerobic respiration, nitrification, chemical oxidation), and the rates at which DO is added to and removed from the system by atmospheric exchange (aeration and degassing) and hydrodynamic processes (accrual/addition from rivers and tides against export to ocean) (Connell, Miller, and Miller 1984). Oxygen solubility varies inversely with salinity, water temperature and atmospheric and hydrostatic pressure.

Dissolved oxygen consumption and production are influenced by plant and algal biomass, light intensity and water temperature (because they influence photosynthesis), and are subject to diurnal and seasonal variation (Connell, Miller, and Miller 1984). DO concentrations naturally vary over a twenty-four hour period due to tidal exchange, and net production of oxygen by plants and algae during the daytime when photosynthesis occurs. Coastal discharges of wastes rich in organic carbon (from sewage treatment plants, paper manufacturing, food processing and other industries) are produced in large quantities in urban population centers, and can substantially reduce dissolved oxygen concentrations (Connell, Miller, and Miller 1984).

Most aquatic organisms require oxygen in specified threshold values for respiration and efficient metabolism, and if DO is not within the threshold conditions adverse physiological effects may occur (Best, Wither, and Coates 2007). Even short-lived anoxic and hypoxic events can cause major mortality of aquatic organisms. Exposure to low oxygen concentrations can have an immune suppression effect on fish which can elevate their susceptibility to diseases for several years (Mellergaard and Nielsen 1988).

Moreover, the toxicity of many toxicants (lead, zinc, copper, cyanide, ammonia, hydrogen sulfide and pentachlorophenol) can double when DO is reduced from 10 to 5 mgl^-1 (Best, Wither, and Coates 2007). The death of immobile organisms and avoidance of low-oxygen conditions by mobile organisms can also cause changes in the structure and diversity of aquatic communities.

Furthermore, if dissolved oxygen becomes depleted in bottom waters (or sediment), nitrification and therefore denitrification may be terminated, and bioavailable orthophosphate and ammonium may be released from the sediment to the water column. These recycled nutrients can give rise to or reinforce algal blooms. Ammonia and hydrogen sulfide gas (also the result of anaerobic respiration), can be toxic to benthic organisms and fish assemblages in high concentrations (Connell, Miller, and Miller 1984).

In Tanzanian water, dissolved oxygen concentration approaches saturation throughout the year in the surface water but with some reduction in oxygen tension before the thermocline is reached (Wagner, 2002). The dissolved oxygen concentration in the coastal waters of Tanzania has been reported to range between 2-7 mgl^-1 (Bryceson 1982).

2.5.5 pH

The pH is a measure of acidity or alkalinity of water on a logarithmic scale, varying from 0 (extremely acidic) through 7 (neutral) to 14 (extremely alkaline). It is the negative base-10 logarithm of the hydrogen ion (H+) activity in moles per liter. Hydrogen ions predominate in waters with pH less than 7; hydroxyl ions (OH^-) predominate in waters with pH greater than 7.

The pH of marine waters ranges from 7.5 to 8.5 whereas most natural freshwaters have pH values ranging from 6.5 to 8.0 (Sotthewes 2008). Most waters have some capacity to resist pH change through the effects of the carbonate-buffer system (Chen and Durbin 1994). In this system, hydroxyl ions produced during the hydrolysis of bicarbonate neutralize H^+ ions, and maintain pH at a near constant level (Sotthewes 2008). Bicarbonate ions (HCO^3-) are acquired from the weathering of silicate or carbonate minerals as rainwater passes through the soil zone.

Some studies have demonstrated that pH also changes significantly in marine systems despite the strong buffering capacity of the carbonate system in seawater (Frithsen et al., 1985, Pegler and Kempe, 1988). This may be an important factor regulating algal abundance and distribution in many marine systems.

Generally, changes in pH levels in marine systems appear to correlate with changes in temperature, dissolved oxygen, and phytoplankton production. Conditions of high pH, high phytoplankton production, and high oxygen conditions are characteristic of nutrient enriched systems and often are found in coastal waters or enclosed bodies of water (lagoons, salt ponds and embayments) which receive anthropogenic inputs such as sewage effluent or agricultural runoff (Park, Hood, and Odum 1958; King 1970; Oviatt et al. 1986). In these systems, pH ranges approaching 9 have been measured. The pH levels exceeding 9.0 have been recorded in a number of coastal marine systems (Hires, Stroup, and Seitz 1963; Emery 1972; Frithsen, Keller, and Pilson 1985)

References

Yoder, JA, JE O’Reilly, AH Barnard, TS Moore, and CM Ruhsam. 2001. “Variability in Coastal Zone Color Scanner (Czcs) Chlorophyll Imagery of Ocean Margin Waters Off the Us East Coast.” Continental Shelf Research 21 (11-12). Elsevier: 1191–1218.

Robinson, Ian S. 2004. Measuring the Oceans from Space: The Principles and Methods of Satellite Oceanography. Springer Science & Business Media.

Fett, Robert W. 2000. “The Remote Sensing Data Book.” Bulletin of the American Meteorological Society 81 (1). American Meteorological Society: 107.

Brando, Vittorio E, AG Dekker, A Marks, Y Qin, and K Oubelkheir. 2006. “Chlorophyll and Suspended Sediment Assessment in a Macro-Tidal Tropical Estuary Adjacent to the Great Barrier Reef: Spatial and Temporal Assessment Using Remote Sensing.” Indooroopilly, QLD, CRC for Coastal Zone, Estuary; Waterway Management.

Mansor, Shattri, CK Tan, HM Ibrahim, and Abdul Rashid Mohd Shariff. 2001. “Satellite Fish Forecasting in South China Sea.” In Paper Presented at the 22nd Asian Conference on Remote Sensing, 5:9.

Solanki, HU, RM Dwivedi, SR Nayak, VS Somvanshi, DK Gulati, and SK Pattnayak. 2003. “Fishery Forecast Using Ocm Chlorophyll Concentration and Avhrr Sst: Validation Results Off Gujarat Coast, India.” International Journal of Remote Sensing 24 (18). Taylor & Francis: 3691–9.

Pauly, Daniel, and Villy Christensen. 1995. “Primary Production Required to Sustain Global Fisheries.” Nature 374 (6519). Springer: 255–57.

Watson, Reg, and Daniel Pauly. 2001. “Systematic Distortions in World Fisheries Catch Trends.” Nature 414 (6863). Nature Publishing Group: 534.

Babin, M, A Morel, and B Gentili. 1996b. “Remote Sensing of Sea Surface Sun-Induced Chlorophyll Fluorescence: Consequences of Natural Variations in the Optical Characteristics of Phytoplankton and the Quantum Yield of Chlorophyll a Fluorescence.” International Journal of Remote Sensing 17 (12). Taylor & Francis: 2417–48.

Barnes, William L, Thomas S Pagano, and Vincent V Salomonson. 1998. “Prelaunch Characteristics of the Moderate Resolution Imaging Spectroradiometer (Modis) on Eos-Am1.” IEEE Transactions on Geoscience and Remote Sensing 36 (4). IEEE: 1088–1100.

Raymont, John EG. 1980. Plankton and Productivity in the Oceans: Vol. 1: Phytoplankton. Vol. 489. Pergamon press New York.

Lugomela, C. 1996. “Studies of Phytoplankton in the Near Shore Waters of Zanzibar.” PhD thesis, PhD thesis, University of Dar es Salaam, Tanzania.

McClanahan, Timothy R. 1988. “Seasonality in East Africa’s Coastal Waters.”

Peterson, David L, Michael A Spanner, Steven W Running, and Kurt B Teuber. 1987. “Relationship of Thematic Mapper Simulator Data to Leaf Area Index of Temperate Coniferous Forests.” Remote Sensing of Environment 22 (3). Elsevier: 323–41.

Riley, ET, and EE Prepas. 1985. “Comparison of the Phosphorus–chlorophyll Relationships in Mixed and Stratified Lakes.” Canadian Journal of Fisheries and Aquatic Sciences 42 (4). NRC Research Press: 831–35.

Kitheka, Johnson U, BO Ohowa, BM Mwashote, WS Shimbira, JM Mwaluma, and JM Kazungu. 1996. “Water Circulation Dynamics, Water Column Nutrients and Plankton Productivity in a Well-Flushed Tropical Bay in Kenya.” Journal of Sea Research 35 (4). Elsevier: 257–68.

Hupfer, Michael, and Jörg Lewandowski. 2008. “Oxygen Controls the Phosphorus Release from Lake Sediments–a Long-Lasting Paradigm in Limnology.” International Review of Hydrobiology 93 (4-5). Wiley Online Library: 415–32.

Mohammed, SM, AS Ngusaru, and OU Mwaipopo. 1993. “Determination of the Effects of Pollutants on Coral Reef Areas Around Zanzibar Town.” Report Submitted to National Environment Management Council, Dar Es Salaam.

Kyewalyanga, MNS. 2002. “Spatial-Temporal Changes in Phytoplankton Biomass and Primary Production in Chwaka Bay, Zanzibar.” Tanzania Journal of Science 28 (2). College of Natural; Applied Sciences (CoNAS) of the University of Dar es Salaam (UDSM): 11–26.

Bryceson, I. 1977. “An Ecological Study of the Phytoplankton of the Coastal Waters of Dar Es Salaam.” University of Dar Es Salaam, Dar Es Salaam, Zanzibar.

Julius, Anita. 2005. “Monitoring Programme for Resource Condition, Environmental and Biological Parameters for Mnazi-Bay Ruvuma Estuary Marine Park (Mbremp) Tanzania.” Reykjavik, Iceland.

Newell, BSA. 1959. “The Hydrography of British East African Coastal Waters (Part 2).”

Francis, Julius, Mahongo Shigala, and Semesi Aledaide. 2001. “The Coastal Environment: Eastern Africa Atlas of Coastal Resources of Tanzania.”

Wang, Zhaohui, Yuzao Qi, Jufeng Chen, Ning Xu, and Yufeng Yang. 2006. “Phytoplankton Abundance, Community Structure and Nutrients in Cultural Areas of Daya Bay, South China Sea.” Journal of Marine Systems 62 (1-2). Elsevier: 85–94.

Fofonoff, NP. 1985. “Physical Properties of Seawater: A New Salinity Scale and Equation of State for Seawater.” Journal of Geophysical Research: Oceans 90 (C2). Wiley Online Library: 3332–42.

Nyandwi, Ntahondi, and Alfonse M Dubi. 2001. “Episodic Atmospheric Changes and Their Impact on the Hydrography of Coastal Waters in Tanzania.” Climate Research 18 (1-2): 157–62.

Paula, Joseé, Isabel Pinto, Isabel Guambe, Soónia Monteiro, Domingos Gove, and Joseé Guerreiro. 1998. “Seasonal Cycle of Planktonic Communities at Inhaca Island, Southern Mozambique.” Journal of Plankton Research 20 (11). Oxford University Press: 2165–78.

Connell, Des W, Gregory J Miller, and Gregory K Miller. 1984. Chemistry and Ecotoxicology of Pollution. Vol. 65. John Wiley & Sons.

Best, MA, AW Wither, and S Coates. 2007. “Dissolved Oxygen as a Physico-Chemical Supporting Element in the Water Framework Directive.” Marine Pollution Bulletin 55 (1-6). Elsevier: 53–64.

Mellergaard, Stig, and Else Nielsen. 1988. The Influence of Oxygen Deficiency on the Dab Populations in the Eastern North Sea and the Southern Kattegat. Danmarks Fiskeri-og Havundersøgelser.

Bryceson, I. 1982. “Seasonality of Oceanographic Conditions and Phytoplankton in Dar Es Salaam Waters.” Univ. Sci. J.(Dar Univ.) 8: 66–76.

Sotthewes, W. 2008. “Forcing on the Salinity Distribution in the Pangani Estuary.” PhD thesis, MSc. Thesis, Delft University of Technology, The Netherlands. 82p.

Chen, Celia Y, and Edward G Durbin. 1994. “Effects of pH on the Growth and Carbon Uptake of Marine Phytoplankton.” Marine Ecology-Progress Series 109. INTER RESEARCH: 83–83.

Park, Kilho, Donald W Hood, and Howard T Odum. 1958. “Diurnal pH Variation in Texas Bays, and Its Application to Primary Production Estimation.”

King, Darrell L. 1970. “The Role of Carbon in Eutrophication.” Journal (Water Pollution Control Federation). JSTOR, 2035–51.

Oviatt, Candace A, Aimee A Keller, Peter A Sampou, and Lynn L Beatty. 1986. “Patterns of Productivity During Eutrophication: A Mesocosm Experiment.” Marine Ecology Progress Series 28 (1-2). Inter-Research: 69–80.

Hires, RI, Edward D Stroup, and Raymond Carlton Seitz. 1963. “ATLAS of the Distribution of Dissolved Oxygen and Ph in Chesapeake Bay 1949-1961.” JOHNS HOPKINS UNIV BALTIMORE MD CHESAPEAKE BAY INST.

Emery, By KO. 1972. “A Coastal Pond.” JAWRA Journal of the American Water Resources Association 8 (3). Wiley Online Library: 632–32.

Frithsen, Jeffrey B, Aimee A Keller, and Michael Edward Quinton Pilson. 1985. Effects of Inorganic Nutrient Additions in Coastal Areas: A Mesocosm Experiment: Data Report. Vol. 3. Marine Ecosystems Research Laboratory, Graduate School of Oceanography, University of Rhode Island.