Louisiana State University LSU Digital Commons
LSU Master's Theses Graduate School
February 2020
Role of Coastal Environmental Conditions During Austral Winter On Plankton Community Dynamics And The Occurrence Of Pseudo-nitzschia Spp. And Domoic Acid In Inhambane Province, Mozambique
Holly Kelchner
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_theses
Part of the Aquaculture and Fisheries Commons, Oceanography Commons, and the Terrestrial and Aquatic Ecology Commons
Recommended Citation Kelchner, Holly, "Role of Coastal Environmental Conditions During Austral Winter On Plankton Community Dynamics And The Occurrence Of Pseudo-nitzschia Spp. And Domoic Acid In Inhambane Province, Mozambique" (2020). LSU Master's Theses. 5056. https://digitalcommons.lsu.edu/gradschool_theses/5056
This Thesis is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Master's Theses by an authorized graduate school editor of LSU Digital Commons. For more information, please contact [email protected]. ROLE OF COASTAL ENVIRONMENTAL CONDITIONS DURING AUSTRAL WINTER ON PLANKTON COMMUNITY DYNAMICS AND THE OCCURRENCE OF PSEUDO-NITZSCHIA SPP. AND DOMOIC ACID IN INHAMBANE PROVINCE, MOZAMBIQUE
A Thesis
Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Master of Science
in
The School of Renewable Natural Resources
by Holly Ann Kelchner B.S., University of Alaska Southeast, 2017 May 2020 Acknowledgements
First and foremost of my lengthy list of those to be acknowledged for the work that went into this thesis, I cannot begin without the acknowledgement of my graduate advisor, Dr. Reagan Errera. Without her support, inspiration and heartfelt mentorship, my degree would not be possible. Reagan not only went beyond expectation of a graduate advisor when spending countless hours planning and packing for three months in Africa or reading and extensively editing drafts till all hours of the night, but welcomed me into her home and cared for me as family. I am so honored to be her first graduate student and be part of her academic family tree. In addition, my minor advisor, Dr. Sibel Bargu Ates, deserves nothing but the opposite of a “minor” acknowledgement for her experience, loving support and kindness. I’d also like to acknowledge my other committee members, Dr. Megan La Peyre, for her knowledge and feedback that never failed to interest me further, and Dr. Nan Walker, for endless inspiration that reminds me to look for the bigger picture (or image in the case of satellite data) and be patient for great things to come. Beyond my committee, I’d like to acknowledge LSU faculty Dr. Bret Collier, Dr. Mike Kaller, and Dr. Chris Green, who were supportive of my research, degree, and beyond with fervor, authenticity and generosity. A special thank you to Bret for getting me out of Africa alive. I’d like to acknowledge Monique Boudreaux, for being a great pseudo lab-mate and always laughing at my bad plankton jokes. I’d also like to thank other LSU friends and colleagues for their support, especially Hannah Masood, Margret Foy, and Lauren McFarland for being willing to look at plankton for hours and still having a great attitude with me. I’d like to acknowledge my other research collaborators, Dr. Kathryn Shreiner from the University of Minnesota-Duluth and Dr. Amy Wagner from California State University- Sacramento. I thank both for accompanying me on my journey to Africa and adding greater depth to my knowledge and this project. A large acknowledgement needs to go to All Out Africa, especially Kim Roques for sharing his passion for conservation and Katie Reeve-Arnold for her enthusiasm and guidance throughout the field season and while halfway around the world. In addition, thank you to all the All Out volunteers and staff who helped with fieldwork, especially France Armando Gomba, Elisabeth Boon, George Bevan, Franzi Spahn, Naomi Martineau and Willie Viljoen. A huge acknowledgement goes to Peri Peri Divers for their collaboration, literally giving me the ability to get out on the water and do science. I’d like to thank Steve and Nick for the thumbs up in the morning to get on the boat, and Martina for her bigheartedness beginning the minute I arrived in Tofo. I’d like to thank all the skippers, especially Não for his skilled and gentle maneuvering while sampling, all the staff for always offering to carry my “heavy” sampling equipment, and Helen, Sinda, Moises, and Jerry who’s inspiring friendship drives me to inspire others. I would not have made it through grad school without my family and friends for their endless emotional support. To list all their names would distract from the depth of gratitude I feel for each phone call or extended chats that kept me going the last three years. In particular, I’d like to
ii thank my undergraduate biochemistry lab partner and valued friend, Josh Russell, who never fails to challenge me intellectually and answer my seemingly simple statistics questions and help with R code. Also deserving a special mention is my once roommate and forever best friend, Sam Rice for always being there through every mood, emotion, triumph and failure faced in grad school. I’d also like to thank Andrew Poley from Michigan Tech Research Institute and Ashley Burtner from the Cooperative Institute for Great Lakes Research for their gracious map-making GIS skills. Last but not least, I’d like to acknowledge the Louisiana State University AgCenter and Department of Renewable Natural Resources for funding this project.
iii Table of Contents
Acknowledgements...... ii
List of Tables...... v
List of Figures...... vi
Abstract...... viii
Chapter 1. Literature Review...... 1 Introduction...... 1 Harmful Algal Blooms...... 1 Study Site: Mozambique...... 9 Proposed Research...... 18
Chapter 2. Environmental influences on plankton community composition of coastal Inhambane Province, Mozambique...... 22 Introduction...... 23 Methods...... 27 Results...... 35 Discussion...... 49
Chapter 3. Environmental influences on domoic acid production by Pseudo-nitzschia spp. along coastal Inhambane Province, Mozambique...... 59 Introduction...... 60 Methods...... 68 Results...... 74 Discussion...... 81
Chapter 4. Concluding remarks on plankton community dynamics effect on Pseudo-nitzschia spp. and domoic acid: implications for ecosystem and human health...... 89 Summary...... 89 Suggestions for Future Work...... 94
Appendix. Supplementary Data...... 100
References...... 111
Vita...... 135
iv
List of Tables
Table 2.1. Identified plankton by microscopy and frequency of occurrence across all sites and sampling events…...... 43
Table 2.2. Shannon Weaver Diversity Index and Equitability...... 44
Table 2.3. Average particulate C:N ratios of samples collected within the four study regions before and after the major upwelling event...... 49
Table 3.1. Pearson correlation for physiochemical and biological variables over study period (May – August 2018) ...... 74
Table A.1. Sampling data…...... 100
Table A.2. Subsample filter amounts…...... 102
Table A.3. Pigment: chlorophyll a starting and output ratios in the CHEMTAX analysis of HPLC pigments…...... 104
Table A.4. Phytoplankton verification counts (shown as a percent of total community) and listed with pigment proportions derived from CHEMTAX...... 105
Table A.5. Identified phytoplankton genera and corresponding group...... 105
v List of Figures
Figure 1.1. Political map of Mozambique...... 9
Figure 1.2. Surface circulations around the Mozambique Channel and the greater Agulhas Current system (from Lutjeharms 2006)...... 11
Figure 1.3. Particulate domoic acid detected within waters of coastal Inhambane Province known for high sightings of Rhincodon typus (whale shark) between January 2017 – May 2018 (Errera, unpublished data)...... 17
Figure 2.1. Inhambane Province, Mozambique...... 28
Figure 2.2. Sea surface temperature (SST) and chlorophyll a (chla) images from 28 July 2018 at 13:15 Local Standard Time (LST) reveal the possible occurrence of a cyclonic eddy along the coast of Inhambane Province...... 35
Figure 2.3. SST and chla images from 13 August 2018 at 13:15 LST show coastal production along the shelf of Inhambane Province...... 36
Figure 2.4. A) Subsurface and bottom temperature and B) dissolved oxygen concentration during sampling period...... 37
Figure 2.5. A) Wind vectors and frequency of B) wind direction and C) wind speed along Praia do Tofo, Mozambique throughout the sampling period...... 38
Figure 2.6. Pigment community concentrations measured by HPLC analysis and derived using CHEMTAX...... 40
Figure 2.7. Proportion of each phytoplankton group contributing to total chla at A) northern, B) central, C) whale shark hotspot, and D) southern regions...... 41
Figure 2.8 Copepod size and abundance counted from zooplankton fraction (>80µm ) based on organism size...... 42
Figure 2.9. Ratio of particulate organic carbon (C) to particulate organic nitrogen (N) concentration quantified within phytoplankton and zooplankton fractions A) prior to MUE and B) after MUE...... 45
Figure 2.10. Particulate organic C and N concentrations measured within phytoplankton and zooplankton showed increased C:N ratios among regions before and after the upwelling event at A) Northern reefs, B) Central reefs, C) WSH, and D) Southern reefs...... 46
Figure 2.11. A) Inorganic phosphorus concentrations and B) inorganic nitrogen concentrations measured during field season...... 47
vi Figure 2.12. Relationship between inorganic nitrogen and inorganic phosphorus A) prior to MUE and B) post MUE...... 48
Figure 2.13. Silica concentrations (average 1.4 ±0.39mg L-1) during sampling period...... 50
Figure 3.1. Particulate domoic acid detected within waters of coastal Inhambane Province known for high sightings of Rhincodon typus (whale shark) between January 2017 – May 2018 (Errera, unpublished data)...... 67
Figure 3.2. Inhambane Province, Mozambique...... 69
Figure 3.3. Sea surface temperature (SST) and chlorophyll a (chla) images from 28 July 2018 at 13:15 Local Standard Time (LST) reveal the possible occurrence of a cyclonic eddy along the coast of Inhambane Province...... 75
Figure 3.4. Pseudo-nitzschia spp. cell abundance and pDA from phytoplankton fraction...... 76
Figure 3.5. Light microscopy revealed at least three species of Pseudo-nitzschia spp. present in samples from coastal Inhambane Province based on width and shape...... 77
Figure 3.6. Particulate DA concentrations measured along the coast of Praia do Tofo during the austral winter months in 2018...... 78
Figure 3.7. Particulate DA concentrations separated by regions, A) north reefs, B) central reefs, C) whale shark hotspot and D) south reefs...... 79
Figure 3.8. Principle component analysis (PCA) variable factors from May to August 2018 defined by the two first axes showing the link between physiochemical parameters, chla, pDA and Pseudo-nitzschia spp. abundance...... 80
Figure 3.9. PCA individual factors by region as a function of physiochemical and biological parameters...... 82
Figure 4.1. pDA measured along Inhambane Province between January 2017 - August 2018....90
Figure A.1 SST and Chla images from 25 May 2018 at 13:15 LST…...... 107
Figure A.2 SST and Chla images from 1 June 2018 at 13:15 LST…...... 107
Figure A.3 SST and Chla images from 17 June 2018 at 13:20 LST…...... 108
Figure A.4 SST and Chla images from 19 June 2018 at 13:05 LST…...... 108
Figure A.5 SST and Chla images from 20 June 2018 at 13:15 LST…...... 109
Figure A.6 SST and Chla images from 24 June 2018 at 13:25 LST…...... 109
Figure A.7 SST and Chla images from 1 July 2018 at 13:30 LST…...... 110
vii Abstract
Harmful algal blooms (HABs) are increasing globally in frequency, persistence, and geographic extent. HABs pose a threat to economic stability, and ecosystem and human health.
To date no incidences of marine toxins produced by phytoplankton have been recorded in
Mozambique, which may be due to the absence of a monitoring program and general awareness of the potential threat. This study is the first documentation of the occurrence of a neurotoxin, domoic acid (DA), produced by the diatom genus Pseudo-nitzschia spp. along the east coast of
Africa. The coast of Inhambane Province is a biodiversity hotspot where year-round Rhincodon typus (whale shark) sightings are among the highest in the world, supporting an emerging ecotourism industry. Links between primary productivity and biodiversity in this area have not previously been considered or reported. My research focused on identifying environmental factors, specifically nutrients, influencing coastal productivity and DA concentrations and highlights variations within the system across four regions during May-August 2018. During this time, the coastal phytoplankton community was diatom-dominated, with high abundances of
Pseudo-nitzschia spp. which were influenced by nutrient pulses resulting from wind-driven upwelling. In late July 2018, primary production was enhanced and corresponded with a peak in
DA located within a biodiversity hotspot for Rhincodon typus (whale shark). Increases in DA concentration were correlated to phosphorus limitation. Domoic acid was also found to be present in mesozooplankton samples, providing evidence for trophic transfer of the toxin and the potential for bioaccumulation within a system, serving as a vector for higher trophic level organisms. Continued and comprehensive monitoring along southern Mozambique would provide critical information to develop predictive models to assess ecosystem and human health
viii threats and impacts to the local economy from marine toxins under challenges posed by global change.
ix Chapter 1. Literature Review Introduction
Phytoplankton play an essential role in aquatic ecosystems serving as the primary base in food webs, and enhancing biogeochemical processes through recycling. Globally, marine phytoplankton contribute less than 1% of photosynthetic biomass, yet they account for approximately 50% of primary production (Field et al. 1998, Simon et al. 2009, Boyce et al.
2010). Phytoplankton production is influenced by multiple factors such as temperature, light irradiance, nutrient availability, competition, and salinity (Reynolds 2006, Simon et al. 2009,
Sigman & Hain 2012), which vary globally across latitude gradients and aquatic systems. Areas of high abundance often correlate to highly productive fisheries (Pauly & Christensen 1995,
Chassot et al. 2010). However, some phytoplankton can reach levels in which they can cause harm to the ecosystem or human health, known as harmful algal blooms (HABs) or species
(HA). Among HA species, some are known to produce toxins, which can accumulate in the aquatic food web (Anderson 2002, Davidson et al. 2014, Berdalet et al. 2015, Willis et al. 2018).
Harmful Algal Blooms
Harmful algal blooms (HABs) are a natural phenomenon, but are increasing globally in frequency, persistence, and geographic extent, likely due to anthropogenic and climatic influences (Smayda 1990, Van Dolah 2000, Graneli & Turner 2006, Heisler et al. 2008, Paerl &
Paul 2012, Davidson et al. 2014, McKibben et al. 2015, Wells et al. 2015, Hallegraeff 2019).
Increases in HAB events coincide with global climate change and coastal eutrophication from human development along coastlines and waterways (Van Dolah 2000, Anderson et al. 2008).
Marine HABs are most noted to cause direct harm through the production of potent and persistent toxins that can lead to human illnesses or mortality. They also may indirectly harm
1 coastal societies through economic impairment resulting from increased costs associated with bloom management and decreased revenue during fisheries closures (Anderson et al. 2000,
Berdalet et al. 2015, Grattan et al. 2016, Willis et al. 2018). For example, studies in Florida revealed economic losses between $868 to $3734 in daily sales (13.7% - 15.3% of average daily sales) among coastal communities during a bloom of toxic Karenia brevis (Morgan et al. 2009).
In developing countries, shellfish closures due to accumulation of toxins from a HAB event could result in an entire community without a source of revenue or food (Sidharta 2005). These consequences to ecosystem services can have major effects on human well-being (Berdalet et al.
2015). Understanding coastal dynamics influencing HABs is important for economic security and protection of human health.
Humans are exposed to HAB-related toxins through ingestion of contaminated seafood products, skin contact through contaminated water, or through the inhalation of aerosolized toxins (Anderson 2002, Graneli & Turner 2006, Landsberg et al. 2009, Berdalet et al. 2015).
Historically referred to as “red tides,” HABs have been documented as far back as early 1800’s, where many native American tribes currently avoid contamination by not harvesting shellfish during times of the year when blooms are likely to occur (Anderson 2002, McKibben et al.
2015). As blooms have increased in frequency and severity, global monitoring efforts have increased to test for toxins among harvested shellfish and fish (Trainer et al. 2003, Bauer 2006,
McKibben et al. 2015, Grattan et al. 2016).
Toxic blooms can have deleterious economic effects through contamination and mortalities of wild or farmed seafood and decreases in tourism revenue (Bauer 2006, Berdalet et al. 2015).
The US averages $1912 per kilometer of coastline per year in commercial fisheries and aquaculture revenue losses due to HABs (Hoagland & Scatasta 2006). In 2015, a HAB event
2 spanning the entire west coast of the United States drastically impacted the Dungeness crab and razor clam fisheries and resulted in over $120 million in revenue losses (Ritzman et al. 2018).
The EU averages $9743/km per year in recreation and tourism losses due to HAB events, while the US reports a loss of $225/km per year (Hoagland & Scatasta 2006). The difference is likely an indicator of a higher level of tourist-related income. However with HABs increasing in frequency and persistence (Wells et al. 2015) and systems experiencing emerging HABs, these economic impacts will continue to increase (Anderson et al. 2000). One of the well-documented and cosmopolitan harmful algal species is Pseudo-nitzschia, which produces extensive blooms around the globe, most notably along the west coast of North America.
Role of Nutrients on Phytoplankton Dynamics
Phytoplankton blooms have been linked to increases in nutrient availability and coastal eutrophication (Anderson et al. 2002, Anderson et al. 2008, Heisler et al. 2008, Davidson et al.
2014). Eutrophication refers to the input of nutrients, either from anthropogenic runoff into coastal systems or physical processes such as coastal upwelling. Changes in nutrient stoichiometry, particularly nitrogen (N) and phosphorous (P), impact phytoplankton community composition and potentially promote the growth of harmful species (Anderson 2002, Glibert et al. 2005, Heisler et al. 2008). For example, increases in agricultural runoff from the Maumee
River watershed have been linked to increases in harmful cyanobacteria blooms in the western basin of Lake Erie (Michalak et al. 2013). Increased nutrient use and runoff in the Mississippi-
Missouri River watershed, which stretches from the Montana-Idaho border to northern Gulf of
Mexico off the coast of Louisiana, have been linked to hypoxia in the northern Gulf of Mexico and an increase in PN blooms (Turner & Rabalais 1991, Parsons et al. 2002). Areas of upwelling can also bring in nutrients and promote the growth of harmful species. Pitcher and Nelson (2006)
3 indicated that nutrients from upwelling in the Benguela system in the south Atlantic Ocean promoted the dominance of toxic dinoflagellate Gymnodinium spp. Similarly, the Pacific
Northwest and California experience seasonal toxic blooms of PN when nutrient-rich waters are upwelled along the coast (Trainer et al. 2000, McKibben et al. 2015, McCabe et al. 2016, Smith et al. 2018). Globally, degraded water quality has left phytoplankton communities more vulnerable, providing an opportunity for HA species dominance (Heisler et al. 2008).
Nutrients play a large role in plankton dynamics as phytoplankton species have distinct nutrient requirements that lead to growth, limitation, and intra- and inter-specific competition.
Limiting macronutrients for phytoplankton growth include nitrogen, phosphorus, silica, and iron
(Tilman et al. 1982, Harris 1986, Reynolds 2006, Litchman & Klausmeier 2008). Within freshwater systems, P is noted as biologically unavailable causing limitation, however, in coastal marine systems, N is typically suggested as the limiting nutrient (Howarth 1988, Kilham &
Hecky 1988, Anderson 2002). Historically, a single nutrient was suggested as regulating the phytoplankton assemblage within systems. Based on the Liebig’s Law of the Minimum theory maximum productivity is proportional to the concentration of the limiting nutrient (Hutchinson
1961), providing regulatory authorities a single environmental parameter to govern in an effort to decrease algal growth. However, the ratio of key macronutrients has been shown to be important regulators of growth, particularly when we consider individual species requirements (Hodgkiss
& Ho 1997, Danger et al. 2008). The stoichiometric relationship of nutrients within a system is often referred to as the Redfield ratio (106C:16N:1P) (Redfield 1958) and has been used to determine nutrient limitation. Individual phytoplankton species require specific and individual amounts of macronutrients for growth, revealing preferred stoichiometric ratios. For example, diatoms favor environmental conditions where the N:P ratio is high and silica is abundant
4 compared to dinoflagellates who dominate in low N:P systems (Hodgkiss & Ho 1997, Hu et al.
2008). Competition for available nutrients will lead to shifts in species dominance and abundance, typically favoring the species that can utilize the available nutrients most efficiently for their growth (Reynolds 2006, Humphries 2009).
Increased anthropogenic nutrients and shifts in nutrient stoichiometry from human population growth and agriculture development stimulate phytoplankton growth leading to higher productivity, these events are termed a “bloom” (Anderson et al. 2008, van der Molen et al.
2016). As the human population continues to grow, estimated to reach 10 billion people by 2050, there will be an increased demand for agriculture and aquaculture production (FAO 2018a, b), and increasing the non-point source pollution which often enhance phytoplankton blooms
(Carpenter et al. 1998). For example, increased nitrate loading from agricultural derived non- point source runoff from the Mississippi watershed has led to an increase in relative abundance of PN spp. in the sediment record along the northern Gulf of Mexico (Turner & Rabalais 1991).
Trainer et al. (2003) found that since the beginning of monitoring efforts in the Pacific
Northwest, increasing populations in counties surrounding Puget Sound are highly correlated (r2
= 0.987) with an increase in HAB-related events, suggesting that activities associated with human population growth, such as land clearing, logging, direct sewage outfalls, and non-point source agriculture and aquaculture runoff (Howarth 2001) influence HAB development.
Domoic Acid
Domoic acid (DA) is a potent neurotoxin produced by several species of Pseudo-nitzschia
(PN), leading to both acute and chronic toxicity in mammals, fish and birds (Work et al. 1993,
Lefebvre et al. 2002, Bargu et al. 2010, Fire et al. 2010, Zabaglo 2016, D'Agostino et al. 2017).
DA is a water soluble, polar, non-protein excitatory amino acid with a similar structure to
5 glutamic acid and kainic acid, which binds to glutamate receptors causing overstimulation in the central nervous system in vertebrates (Berman & Murray 1997, Hampson & Manalo 1998,
Lefebvre & Robertson 2010). Glutamate is the principle excitatory neurotransmitter within the brain and is a critical component for all synaptic transmission, however excessive glutamate is attributed to neurodegeneration and seizures (Lefebvre & Robertson 2010). Overactivation of glutamate receptors can lead to neuronal injury or cell death. Having a similar but more rigid structure, DA has a similar effect on these receptors but has a higher binding affinity and efficacy than glutamate, therefore permanently binding to the receptor (Hampson & Manalo
1998, Lefebvre & Robertson 2010, Hiolski et al. 2014). Human exposure through consumption of the toxin can induce gastrointestinal distress, confusion, disorientation, seizures, short-term memory loss, and, in high concentrations, death (Bates et al. 1989, Jeffery et al. 2004, Lefebvre
& Robertson 2010, Saeed et al. 2017). This condition is known as Amnesic Shellfish Poisoning
(ASP). In 1987, DA was found to be the cause of 4 deaths and over 100 acute illnesses following the consumption of contaminated mussels (Mytilus edulis) from Prince Edward Island in eastern Canada (Bates et al. 1989, Wright et al. 1989). The event led to the establishment of a regulatory monitoring program for algal toxins. No human fatalities have been reported since the implementation of monitoring programs, however, DA still causes harm to local economies and the environment through the accumulation of toxin within the food web during blooms
(Goldstein et al. 2008, Trainer et al. 2012, D'Agostino et al. 2017, Saeed et al. 2017, Bates et al.
2018).
The presence of DA in the ecosystem can impact the whole food web through bioaccumulation during PN blooms. Well-established monitoring programs along the Pacific coast of the United States have furthered our understanding of the impact of PN blooms and DA
6 on the ecosystem (Lefebvre et al. 2017). One bloom of PN in 2015 led to the mortality of over
400 California sea lions (Zalophus californianus) all exhibiting signs of neurological dysfunction
(Trainer et al. 2000, Bauer 2006, McKibben et al. 2015). During the bloom, toxin was also detected in whales, dolphins, porpoises, seals and sea lions ranging from southern California to northern Washington, and noted globally as the largest geographic extent of DA detection in marine mammals (McCabe et al. 2016). In addition to marine mammals, DA can also affect birds and other marine organisms. In 1991, a large mortality of brown pelicans (Pelecanus occidentalis) and Brant’s cormorants (Phalacrocorax penicillatus) was attributed to the consumption of anchovies which had bioaccumulated DA (Lefebvre & Robertson 2010,
McKibben et al. 2015). This phenomenon led to the finding that the event along Monterey Bay in
1961 causing “crazed seabirds” which influenced Alfred Hitchcock’s thriller The Birds, actually correlated to a toxic bloom of Pseudo-nitzschia (Bargu et al. 2011a). The relationship between
DA and the harm it will cause on the environment reflects the complexity of food webs and differences in degradation products and rates of accumulated toxin in planktivorous fish or shellfish.
To predict the potential harm caused by a bloom, it is necessary to assess the accumulation of toxin within vector organisms. It is important to note that a toxic bloom of PN might occur but not be consumed by predators. For DA accumulation to occur, pelagic and/or benthic organisms must feed on toxic PN cells (Bargu et al. 2002, Lefebvre et al. 2002, Bargu et al. 2003, Kvitek et al. 2008, Bargu et al. 2010, Blanco et al. 2010, Del Rio et al. 2010, Harethardottir et al. 2015,
Baustian et al. 2018). For example, when given the choice, oysters will selectively feed on non- toxic PN cells (Thessen et al. 2010), avoiding accumulation. The retention of toxin is an important factor in determining the effect on the ecosystem, as depuration rates vary depending
7 on the organism. Razor clams can retain DA for over a year after a bloom (Schultz et al. 2008), while anchovies (Engraulis mordax) are able to depurate DA within a week (Lefebvre et al.
2002). Many studies have quantified the trophic transfer of DA to zooplankton (Lincoln et al.
2001, Bargu et al. 2002, Del Rio et al. 2010, Leandro et al. 2010, Tammilehto et al. 2012,
Veschasit et al. 2017, Bates et al. 2018) which have been found to retain up to 50% of consumed
DA within tissues (Bates et al. 2018). In addition to zooplankton, toxins are also commonly measured in shellfish, echinoderms, large crustaceans, and fishes (Trainer et al. 2012). The relationship between trophic transfer and toxins remains complex, and is often studied alongside biological and environmental factors affecting toxin production.
Factors triggering toxin production are complicated and have been widely studied, yet not fully elucidated (Pan et al. 1998, Trainer et al. 2002, Wells et al. 2005, Trainer et al. 2012,
Zabaglo 2016, McKibben et al. 2017, Bates et al. 2018, Smith et al. 2018). Toxin production by
PN is not always correlated to cell concentration and can vary among strains, environmental conditions, and biological conditions (Graneli & Turner 2006, Thessen et al. 2009, Trainer et al.
2012). Many theories regarding DA production emphasize physiological stress (Pan et al. 1998).
Domoic acid often exhibits seasonal dynamics (Trainer et al. 2002, Trainer et al. 2012, Smith et al. 2018), corresponding to physiochemical changes such as temperature and irradiance
(reviewed by Lelong et al. (2012)) and shifts in nutrient availability. The ratio of nitrogen to phosphorus has been found to play an important role in the production of DA (Pan et al. 1998,
Hu et al. 2008). Nitrogen is important for synthesis of nitrogenous based toxin and phosphorous is known to limit the growth of PN cells (Pan et al. 1998, Hu et al. 2008). Recent studies have also added copepod grazing pressure as a factor in DA production (Olson et al. 2008, Smayda
2008, Bates et al. 2018). Factors that promote DA production are extensively studied with little
8 consensus, suggesting a complex relationship between environmental and biological conditions controlling PN toxicity.
Study Site: Mozambique
Figure 1.1. Political map of Mozambique
Mozambique is located in southeast Africa sharing boarders with South Africa and Eswatini in the south, Zimbabwe and Zambia, in the west, and Tanzania and Malawi in the north (Figure
1.1). Within sub-Saharan Africa, Mozambique has one of the most extensive coastlines stretching 2,700 km along the Indian Ocean. The country has experienced significant political
9 and economic changes since the ending of their civil war in 1992. The population has more than doubled from 14.07 million people in 1992 to 29.67 million in 2017 (World Bank 2018).
Agriculture has also increased by over 50% in the last two decades (Menezes et al. 2011a, Porter et al. 2017). It is considered a least developed country (LDC) by the United Nations,
Mozambique remains one of the world’s poorest countries with over 70% of the coastal population living below the poverty line (Menezes et al. 2011a) and 26.6% of the population considered undernourished (FAO 2017, World Bank 2018b). Most of the population (65%) live along the coast for food security where many coastal communities derive their food and income from farming and fishing practices. Following the end of the civil war, gross domestic product
(GDP) in agriculture rose at a rate of 8% annually until 2010, when the rate slowed to an annual increase of 3.9% (FAO 2018a). Aquaculture has developed at a slower pace, with only an annual increase of 1% due to financing constraints and challenges, such as infrastructure, availability of seed source, and economic incentives (FAO 2005), supporting increased fishing pressure on wild caught populations. This is reflected in employment opportunities, as 71% (86% in rural communities) of jobs in Mozambique are in agriculture, while only 10% in fisheries (World
Bank 2018a). Recently outside countries (Iceland and Norway) have invested in increasing both wild caught and aquaculture fisheries production, in an effort to support food and economic security (FAO 2005).
Physical Ocean Dynamics
The physical mesoscale ocean dynamics off Mozambique provide insights to its unique coastal processes. In the Indian Ocean, the South Equatorial Current flows west and diverges
10 around Madagascar with one current flowing around the north end of the island and another flowing south (Figure 1.2), forming the East Madagascar Current (Quartly & Srokosz 2004, Raj et al. 2010). Water flowing south within the Mozambique Channel is dominated by southward propagating anti-cyclonic eddies along the narrow shelf with a steep slope (Swart et al. 2010).
Westward-moving cyclonic eddies are a common feature along southern Madagascar during austral winter and can vary in duration from a week to months (Lamont et al. 2014, Barlow et al.
2017), forming between 4-7 eddies per year (Schouten et al. 2003). The southern end of the channel and East Madagascar Current merge between Maputo and Durban, forming the Agulhas
Current, the strongest western boundary systems (WBS) in the world (Lutjeharms 2006), moving at a rate of -69.7 ±21.5 Sv (Beal & Bryden 1997, Bryden et al. 2005), compared to the Gulf
Stream at 31 Sv (Lund et al. 2006). The interactions of mesoscale physical dynamics along the
Figure 1.2. Surface circulation around the Mozambique Channel and the greater Agulhas Current system (from Lutjeharms 2006).
11 Mozambican coast cause upwelling of cooler, nutrient-rich water, resulting in elevated phytoplankton biomass (Lamont et al. 2014).
On a smaller, localized scale, coastal upwelling along Southern Mozambique will occur when northerly winds are prevalent (Jury 1981, Schumann et al. 1982). Note, along the eastern coastal boundaries within the Southern Hemisphere northerly winds result in longshore wind stress, transporting surface waters offshore, i.e. coastal surface water is moved offshore to the east and replaced by upwelled deeper water. Indicated by the difference in temperature of these water masses, coastal upwelling can be studied using thermal infrared satellite data (Schumann et al. 1982, Walker 1986, Lutjeharms & Stockton 1991, Varela et al. 2018). Studied extensively in
South Africa, this coastal band can be tens of kilometers wide, demarcated by the thermocline with colder water on the landward side as the thermocline is pulled to the surface (Schumann et al. 1982). Southern Mozambique subsurface topography plays an important role in the prevalence of upwelling (Lutjeharms & Stockton 1991) as a steep continental slope serves as a pathway for deeper cold water to move up onto the narrow shelf.
Upwelling caused by wind stress and cyclonic eddies replenish nutrients in oligotrophic waters and promote primary production along the southeastern coast of Africa (Lamont et al.
2010, Barlow et al. 2014). While general phytoplankton biomass throughout the channel is low, anticyclonic eddies (warm core) are common within the western side of the Mozambique
Channel and found to decrease phytoplankton biomass, whereas mesoscale processes including cyclonic eddies found throughout the channel and coastal wind-driven upwelling are sources of seasonal increases in phytoplankton biomass (Raj et al. 2010, Barlow et al. 2014, Lamont et al.
2014). The nutrient pulses provided by these frontal systems are essential to support the diverse trophic webs within the region (Tew Kai & Marsac 2010, Ternon et al. 2014).
12 Plankton Dynamics
Coastal phytoplankton dynamics along the majority of Mozambique’s coastline are under studied, but the plankton community contributing to the highly productive Delagoa Bight in
Maputo Bay and Sofala Bank have been broadly considered (Sá et al. 2013). The majority of coastal waters off Mozambique are oligotrophic (Lutjeharms 2006), having low nitrogen concentrations (Sá et al. 2013). However, Delagoa Bight is characterized by a shallow shelf and cyclonic eddies which replenish nutrients throughout most of the year (Barlow et al. 2007). In contrast, the Sofala Bank receives nutrient input from the Zambezi River. The area is an important fishery ground and primarily dominated by highly edible diatoms (70-90%), including
Chaetoceros spp., Pseudo-nitzschia spp., Proboscia alata, Cerataulina pelagica and
Thalassionema nitzschioides and a small abundance of dinoflagellates (Sá et al. 2013). Potential energy transfer from the basal trophic level is defined by phytoplankton type and size. Larger phytoplankton, such as diatoms, have higher concentrations of particulate organic carbon (POC) and provide more energy to primary consumers (Fry & Wainright 1991). Northern Mozambique, between 10 – 17°S, is characterized by warm, oligotrophic waters, low phytoplankton biomass, and dominated by nanoplankton (2-20 µm). Based on nutrient kinetics, smaller phytoplankton are better competitors in nutrient-depleted waters due to their high surface area to volume ratio
(Bonachela et al. 2011). Waters further south are colder and nutrient-rich, allowing for microphytoplankon (20-200 µm) to dominate (Tew-Kai & Marsac 2009, Tew Kai & Marsac
2010, Ternon et al. 2014).
Praia do Tofo Biodiversity Hotspot
Among Mozambique’s three coastal regions, Inhambane Province and the parabolic dune coast is known as a biodiversity hotspot. Aerial surveys along the coast have identified roughly
13 200km along Inhambane Province as having a relatively high density of whale sharks, with a pronounced hotspot adjacent to Praia do Tofo (Rohner et al. 2018). Rhincodon typus (whale shark) are important to the economy of southern Mozambique as an internationally recognized hot-spot for encounters (Pierce et al. 2010) and are the focus of Mozambican government’s effort in the developing tourism sector (World Bank 2006, Tibiriçá et al. 2011). In addition to R. typus,
Praia do Tofo is also frequently visited by planktivorous reef manta rays (Manta alfredi) and giant manta rays (M. birostris) (Rohner et al. 2013b). Venables et al. (2016) estimated an annual contribution of $34 million USD to the province’s economy from manta tourism. Coastal
Inhambane Province is also a hotspot for migrating humpback whales, bottlenose dolphins, humpback dolphins, dugongs, sea turtles, small-eye stingray, guitar sharks, corals, and many reef fishes (Pierce et al. 2008, Pereira et al. 2014, Williams et al. 2017, Fordyce 2018), however the primary driver of tourism in the area is the large planktivores, particularly R. typus (Tibiriçá et al.
2011).
Whale sharks are known to exhibit site fidelity (Cagua et al. 2015) and have been repeatedly sighted along Inhambane Province through photo-ID techniques (Rohner et al. 2018). Sightings of R. typus, which consist primarily of juvenile males (Rohner et al. 2018), have decreased by
79% between 2005 and 2011 in Praia do Tofo (Rohner et al. 2013b). Increased mortality of marine megafauna from fisheries activities has increased globally (Capietto et al. 2014), and bycatch in purse seine and gillnet fisheries has been documented to cause mortality to R. typus in the Indian Ocean (Speed et al. 2008). Bycatch could be a significant source of mortality associated with the decline, however, bottom-up interactions with food availability could also be an unstudied factor in the decline of R. typus.
14 These planktivores are considered opportunistic filter feeders and habitat choice is highly dependent on localized productivity (Colman 1997, Stevens 2007, Couturier et al. 2012).
Aggregations of large planktivores occur in sub-tropical and tropical regions worldwide and are driven by foraging opportunities connected to periodic or seasonal productivity (Hoffmayer et al.
2007, Stevens 2007, Rowat et al. 2009a, Rohner et al. 2013b, Rohner et al. 2015). Whale sharks and manta rays are often seen feeding at the surface on small, planktonic organisms such as phytoplankton, copepods, krill, small fishes, and fish and coral larvae (Stevens 2007, Couturier et al. 2012). Colman (1997) suggests the majority of energy consumption consists of nutrient- rich zooplankton whereas algal matter is swallowed accidentally. Using fatty acid analysis, whale sharks and manta rays off the coast of Mozambique have been found to primarily feed on mysids, shrimp (sergestids), copepods, and myctophid fishes (lantern fish) (Couturier et al. 2013,
Rohner et al. 2013a), suggesting that not only do these creatures feed at the surface, but also in bathypelagic waters. The high content of mysids suggests that these planktivores extensively feed at night, when these zooplankton vertically migrate and tend to be in surface waters.
Changes in sightings of these planktivores are suggested to be linked to changes in primary production. The high density of planktivorous megafauna that aggregate and feed along the coast of Praia do Tofo are an indicator of the coastal productivity. To date, no studies have assessed the plankton community in this area, which can be an indication of overall ecosystem health and provide insight to the future stability of sightings in which the local community relies on for a source of tourism.
HAB-related events within Africa
The effect of HABs within the Southwestern Indian Ocean (SWIO) are severely understudied due to a lack of monitoring infrastructure. To date, there are no reported HAB-related events in
15 Mozambique (Tamele et al. 2019), despite the presence of HAB species and documented accounts of marine toxins within the surrounding region. Okadaic acid and ciguatoxin are common toxins produced by dinoflagellates and have been recorded along the east coast of
Madagascar and nearby islands (Boisier et al. 1995, Bouaïcha et al. 2001, Tamele et al. 2019).
The occurrence of toxic freshwater cyanobacteria blooms have raised concerns in Kenya and
Tanzania (Lugomela et al. 2006, Hamisi & Mamboya 2014). South Africa has the most developed HAB monitoring program within the SWIO. Multiple derivatives of okadaic acid and yessotoxins have been reported within mussels along the western coast of the country, within the
Benguela Current. However no marine toxins have been cited as the cause of shellfish closures along the eastern coast (DAFF 2016). One potential emerging HAB species in South Africa is
PN, which has been reported with DA in areas where shellfish are harvested, however, measured concentrations have not resulted in shellfish closures (Pitcher et al. 2014).
Pseudo-nitzschia
Among the diatom assemblage of coastal Mozambique is Pseudo-nitzschia (PN), a known producer of domoic acid (DA) (Sá et al. 2013), however species have not been identified.
Pseudo-nitzschia spp. exhibit species-specific dynamics with regards to toxin production and responses to available nutrients (Trainer et al. 2012, Thorel et al. 2017). Three generalized groups have been suggested based on morphological characteristics: P. pungens/P. multiseries group, P. australis/P. seriata group, and P. delicastissima group (Trainer et al. 2008, Lelong et al. 2012). A study by Lugomela (2013) identified three species of Pseudo-nitzschia along the coast of Tanzania, including P. pungens, P. cuspidate, and P. seriata. Pseudo-nitzschia pungens is considered to have a cosmopolitan distribution with over 137 described strains (Casteleyn et al. 2008) and was responsible for the first reported DA poisoning in eastern Prince Edward
16 Island (Bates et al. 1989). P. pungens has since been identified in all major oceans except for
Antarctica (Bates et al. 2018). Most of these strains produce toxin, leading the species to be the most commonly reported and potentially toxic representatives of the genus (Hasle 2002). Both P. cuspidate and P. seriata have also been documented to produce toxin in the Pacific Northwest,
Scotland and the Arctic (Fehling et al. 2004, Trainer et al. 2009, Tammilehto et al. 2015). In addition, P. multiseries has been identified to be responsible for DA accumulation in mussels within Algoa Bay, South Africa (Pitcher et al. 2014). While a specific species has not been identified within Mozambique, due to the proximity and physical dynamics along the east coast of Africa, it is likely these species are present along the coast of Mozambique.
Figure 1.3 Particulate domoic acid detected within waters of coastal Inhambane Province known for high sightings of Rhincodon typus (whale shark) between January 2017 - May 2018. (Errera, unpublished data)
17 Recently, particulate DA has been documented along the coast of Inhambane Province
(Figure 1.3) and patterns indicate possible seasonal influences in production at the beginning of austral winter (Errera, personal communication). This is the first documentation of DA production within the Southwestern Indian Ocean and while concentrations are low, could have broader impacts to ecosystem health through bioaccumulation and magnification within the food web.
Proposed Research
My overarching goal is to enhance understanding of coastal plankton dynamics along the
Inhambane Peninsula and explore the influence of inorganic nutrients on plankton community shifts. Influence of environmental factors (e.g. inorganic nutrients, salinity, and temperature) along the coast of Inhambane Peninsula will likely have a direct impact on the phytoplankton composition within the coastal region, as seen in the developing country of Tanzania, where development of the coastal community has led to increased anthropogenic nutrients promoting growth of PN spp. (Lugomela 2013, Hamisi & Mamboya 2014). Systems around the world are currently changing due to multiple and compounding anthropogenic pressures; there is a pressing need to study systems that have experienced minor anthropogenic influences to establish a baseline understanding of the ecosystem and provide essential information on how global
changes may be impacting coastal systems.
Globally, biotoxin production by PN is increasing in geographic extent and intensity with changes to coastal environments (Sekula-Wood et al. 2011, Trainer et al. 2012, Trainer et al.
2019), and western boundary upwelling systems are no exception (Trainer et al. 2008). The recent emergence of DA contamination of shellfish fisheries has been noted in the Southern
18 Hemisphere along Argentina (Almandoz et al. 2007, D'Agostino et al. 2017), Uruguay (Méndez et al. 2012), Brazil (Fernandes & Brandini 2010), New Zealand (Rhodes et al. 2013), as well as
South Africa (Pitcher et al. 2014). Currently, there are limited studies related to phytoplankton dynamics in coastal areas of Mozambique with the majority of these studies focused primarily on the oceanic waters within the Mozambique Channel (Raj et al. 2010, Sá et al. 2013, Barlow et al.
2014, Barlow et al. 2017, Olofsson et al. 2017). In addition, very few of these studies have quantified inorganic nutrients shifts within the coastal system of Mozambique (Olofsson et al.
2017). Given the importance of the planktivores in the area and the community dependence connected to the productivity of the system, this research could provide essential data to support economic growth within the region. Furthermore, the focus of this research will serve as a baseline to identify key shifts in coastal Inhambane Providence but also provide insight on a potential toxic phytoplankton blooms within the region which could lead to ecosystem and human health effects. Specifically, my research focused on two main objectives:
Objective 1: To identify the plankton community composition and influencing environmental factors (temperature, salinity, and nutrients) of coastal Inhambane Province.
Plankton samples were collected at four regions along the coast of Inhambane Peninsula and analyzed for pigments, particulate organic matter, and phyto/zooplankton identification for community composition. Environmental parameters including temperature, salinity, dissolved oxygen, and inorganic nutrients, were collected to establish environmental conditions that regulate productivity.
The following hypotheses were tested:
H1: Sea surface temperature along Inhambane Province will decrease throughout study
time period (austral winter).
19 H0-1: Sea surface temperature along Inhambane Province will remain steady during
austral winter.
H2: Phytoplankton biomass will increase with coastal upwelling.
H0-2: Phytoplankton biomass will not increase with coastal upwelling.
H3: The coastal phytoplankton assemblage will be primarily diatom-dominated.
H0-3: The coastal phytoplankton assemblage will be dominated by a phytoplankton group
other than diatoms.
H4: Regions along the coast of Inhambane Peninsula will exhibit independent
characteristics in phytoplankton composition, zooplankton diversity, plankton
biomass and nutrients.
H4a: North reefs will exhibit less plankton diversity and biomass than other
regions.
H4b: Whale shark hotspot (sandy bottom area) will exhibit more plankton
diversity and biomass than reef regions.
H0-4: Regions along the coast of Inhambane Peninsula will be part of a well-mixed system
and exhibit similar characteristics in phytoplankton composition, zooplankton
diversity, plankton biomass and nutrients.
Objective 2: Identify the environmental drivers and primary influences of domoic acid production by Pseudo-nitzschia spp. during austral winter.
Plankton samples, were size fractionated into phytoplankton and zooplankton fractions, and collected at four regions along the coast of Inhambane Penninsula and analyzed for particulate domoic acid (pDA) concentration and PN spp. abundance. Environmental parameters such as temperature, salinity, dissolved oxygen, and inorganic nutrients were also collected to establish
20 the environmental conditions which influenced toxin production. The following hypotheses were tested:
H1: pDA concentration will increase with coastal upwelling.
H0-1: pDA concentration will not increase with coastal upwelling.
H2: All regions along the coast of Inhambane Peninsula will have the same
concentrations of pDA during the study period.
H0-2 All regions along the coast of Inhambane Peninsula will not have the same
concentrations of pDA during the study period.
H3: The whale shark hotspot region will have the highest concentration of pDA.
H0-3: The whale shark hotspot region will not have the highest concentration of pDA.
H4: pDA concentration will positively correlate with PN abundance.
H0-4: pDA concentration will not positively correlate with PN abundance.
H5: pDA concentration is influenced by nutrients.
H5a: pDA concentration increases during nitrogen limitation.
H5b: pDA concentration increases during phosphorous limitation.
H0-5: pDA concentration is not influenced by nutrients.
21
Chapter 2. Environmental influences on plankton community composition of coastal Inhambane Province, Mozambique
Limited studies along southern Mozambique highlight the seasonal influence of mesoscale cold-core eddies and wind-driven upwelling along the continental slope and shelf, increasing localized coastal primary productivity. In these bottom-up driven trophic systems, primary production drives biodiversity by supplying energy at the base of the food web. Inhambane
Province is noted as a biodiversity hotspot for year-round aggregations of large planktivores, which have experienced a decrease in sightings of up to 79% between 2005 and 2011, raising concerns for the future economic stability of ecotourism in the region. Here I present the first assessment of the plankton dynamics and nutrient profiles along the narrow shelf region of
Inhambane Province during austral winter. During the austral winter of 2018, plankton were sampled at four regions along a north-south gradient to characterize phytoplankton assemblages, including biomass and community proportion. Satellite imagery of SST and chlorophyll a combined with coastal measurements of wind and sea surface temperature were used to investigate the impact of coastal upwelling. Decreases in subsurface and bottom temperature between July 14-19, 2018 followed by upwelling-favorable winds corresponded with increased primary productivity (chlorophyll a and particulate organic matter). Satellite imagery suggested the coastal zone and adjacent shelf was also impacted by a cyclonic eddy during this time.
Following this upwelling event, Pseudo-nitzschia spp., a potentially toxin-producing diatom, increased in biomass and community proportion at all regions but made up the highest portion of the phytoplankton community in the area most frequented by planktivores. This study serves as a baseline dataset and suggests the need for continued monitoring to create modeled scenarios affecting the biodiversity of the region.
22 Introduction
Phytoplankton are the base of aquatic food webs and play a critical role in biogeochemical processes. Marine phytoplankton contribute less than 1% of photosynthetic biomass, yet they account for approximately 50% of global primary production (Field et al. 1998, Simon et al.
2009, Boyce et al. 2010). Higher trophic level productivity is linked to cycles in lower food web production, at both the phytoplankton-zooplankton scale as well as zooplankton-fish interactions
(Runge 1988, Chassot et al. 2010). This dependence varies spatially due to environmental conditions, patterns of plankton biomass, and/or species competition (Mackas et al. 1985). Areas of high primary production, such as coastal upwelling areas and coral reefs, are associated with highly productive fisheries (Pauly & Christensen 1995, Chassot et al. 2010).
Plankton production is influenced by competition, predation, migration, and environmental factors such as temperature, salinity, light irradiance, and resource availability (Reynolds 2006,
Simon et al. 2009, Sigman & Hain 2012). Large phytoplankton are better competitors for light in polar regions where nutrients tend to be high (Cermeño et al. 2005). Within oligotrophic waters, such as tropical latitudes, the phytoplankton community tends to be dominated by picoplankton (>3µm) (Buitenhuis et al. 2012), where low surface-to-volume ratio makes them the best competitor for limited nutrients (Raven 1998). These relationships can provide an understanding of the plankton community for areas across latitudinal gradients.
Nutrients play a large role in plankton dynamics as phytoplankton species have distinct nutrient requirements that lead to growth and intra- and inter-specific competition. In systems not impacted by cultural eutrophication or large freshwater inputs (either from rivers or groundwater), phytoplankton growth is supported by the recycling of nutrients within the system or nutrients from deeper waters, i.e. upwelling (Bristow et al. 2017). Key nutrients for
23 phytoplankton growth include nitrogen, phosphorus, silica, and iron (Tilman et al. 1982, Harris
1986, Reynolds 2006, Litchman & Klausmeier 2008, Bristow et al. 2017) and can vary between systems. In coastal marine systems, nitrogen (N) has been suggested to be the primary limiting nutrient (Howarth 1988, Kilham & Hecky 1988, Anderson 2002). However, this relationship is complex, often revealing that primary production does not rely on a single nutrient, but the ratio between key macronutrients (Hodgkiss & Ho 1997, Danger et al. 2008). Individual phytoplankton species require specific and individual amounts of macronutrients for growth, revealing preferred stoichiometric ratios. For example, diatoms favor environmental conditions where the ratio of N:P is high and silica is abundant compared to dinoflagellates which dominate in low N:P systems (Hodgkiss & Ho 1997, Hu et al. 2008). Competition for available nutrients leads to shifts in species dominance and abundance, typically favoring the species that can utilize the available nutrients most efficiently for growth (Reynolds 2006, Humphries 2009).
This stoichiometric view of phytoplankton species composition leads to implications for grazer-mediated nutrient dynamics (Hessen 1992, Sterner & Schultz 1998). Nutrient enrichment greatly increases phytoplankton growth and alters phytoplankton community structure (Kenitz et al. 2013, Bužančić et al. 2016), however, this does not always lead to high zooplankton diversity
(Vanni 1987). While phytoplankton are limited by inorganic macronutrients, zooplankton growth is a function of nutritional quality. Hessen (1992) observed that copepod grazers can be limited in growth from low C:N ratio of ingested phytoplankton, a possible side effect of low phytoplankton diversity. Larger phytoplankton, such as diatoms have a high C:N ratio (Garcia et al. 2018) and are considered more nutritious when compared to smaller phytoplankton. Coastal systems dominated by picoplankton are often recognized as having decreased production, for
24 example, in the southern Benguela upwelling system, where the picoplankton-dominated community is presumed to be the cause of low pelagic fish yield (Probyn 1992).
Zooplankton link primary productivity to fisheries abundance (Runge 1988). Runge (1988) suggests that small copepods and other microzooplankton drive larval fish viability. This idea is supported by the match/mismatch hypothesis (Cushing 1975, 1990) where seasonal dynamics in phytoplankton lead to ebbs and flows of zooplankton production. However, oligotrophic waters are dominated by small plankton with low biomass and driven by nutrient recycling, as seen in coral reef habitats (Muscatine & Porter 1977, Wyatt et al. 2013). Mesoscale oceanic processes such as localized wind-driven shelf currents and Ekman-driven upwelling are key sources of nutrient input in these areas (Hanson et al. 2005, Wyatt et al. 2013, Bristow et al. 2017). With these processes in mind, zooplankton biomass models have been developed to identify biodiversity hotspots (De Monte et al. 2013, Putra et al. 2016).
Seasonal variability in plankton productivity has led to dynamic predator-prey models that highlight predator migration in response to this variability (Huang & Diekmann 2001, Chen &
Zhang 2013). Satellite tagging of planktivorous Rhincodon typus (whale shark) and Manta birostris (manta ray) have identified movement patterns relative to boundary currents, thermal fronts, and seasonal upwelling events, where abundant prey is available (Hsu et al. 2007, Graham et al. 2012). These planktivores are considered opportunistic filter feeders and habitat choice is highly dependent on localized productivity (Colman 1997, Stevens 2007, Couturier et al. 2012).
Aggregations of large planktivores occur in sub-tropical and tropical regions worldwide and are driven by foraging opportunities connected to periodic or seasonal productivity (Hoffmayer et al.
2007, Stevens 2007, Rowat et al. 2009a, Rohner et al. 2013b, Rohner et al. 2015). For example, resident R. typus populations along the Mesoamerican Barrier Reef and within the Gulf of
25 Mexico exhibit movement patterns and sightings relating to seasonal spawning of corals and fishes (Graham & Roberts 2007, Hoffmayer et al. 2007). Oligotrophic systems often support high diversity of planktivores, such as the Ningaloo Reef along Western Australia where R. typus and M. birostris are commonly sighted (Preen et al. 1997, Taylor & Pearce 1999, Wilson et al.
2001). Shelf dynamics of the Ningaloo Reef area and sporadic wind-generated countercurrents lead to localized coastal upwelling (Hanson et al. 2005), prompting increased primary productivity supporting the large diversity of marine megafauna (Sleeman et al. 2007).
The greatest known concentration of R. typus occurs off the southern coast of Mozambique and northern coast (KwaZulu-Natal) of South Africa (Gifford et al. 2007a, Gifford et al. 2007b), where sightings occur year-round. Aerial surveys have identified roughly 200km along
Inhambane Province as having a relatively high density of whale sharks, with a pronounced hotspot adjacent to Praia do Tofo (Rohner et al. 2018). In addition to R. typus, Praia do Tofo is frequently visited by planktivorous reef manta rays (Manta alfredi) and giant manta rays (M. birostris) (Rohner et al. 2013b). Coastal Inhambane Province is also a biodiversity hotspot supporting migrating humpback whales, bottlenose dolphins, humpback dolphins, dugongs, sea turtles, small-eye stingray, guitar sharks, corals, and many reef fishes (Pierce et al. 2008, Pereira et al. 2014, Williams et al. 2017, Fordyce 2018). Sightings of R. typus have decreased by 79% between the years of 2005 and 2011 in Praia do Tofo (Rohner et al. 2013b). Mortality of marine megafauna from fishing activities has increased globally (Fowler et al. 2002, Capietto et al.
2014). Bycatch in purse seine and gillnet fisheries has been documented to cause mortality to R. typus in the Indian Ocean (Speed et al. 2008), which could be a significant source of mortality associated with the decline. However, bottom-up interactions with food availability could also be an unstudied factor in the decline of R. typus.
26 The purpose of this work is to provide the first comprehensive study on the phytoplankton community composition and environmental factors driving production within the coastal system.
Currently, there are limited studies considering phytoplankton dynamics in coastal areas of
Mozambique with the majority of studies focused primarily on the oceanic waters within the
Mozambique Channel (Raj et al. 2010, Sá et al. 2013, Barlow et al. 2014, Barlow et al. 2017,
Olofsson et al. 2017). In addition, very few of these studies have quantified inorganic nutrients shifts within the coastal system of Mozambique (Olofsson et al. 2017). Given the importance of the planktivores to the area, an understanding of the plankton dynamics is a critical component to interpreting ecosystem dynamics. Specifically, this research will investigate phytoplankton and zooplankton productivity and relationships with inorganic nutrients (N, P and Si) at four regions along the coast of Inhambane Province from field measurements obtained in winter 2018. The findings will provide the first suite of critical baseline information to identify plankton assemblage dynamics and serve as an indication of overall ecosystem health.
Methods
Study Site
The study was conducted along the coast of Inhambane Province, Mozambique, stretching approximately 40 km from Barra Beach (2347’35.00S, 3531’05.00 E) to Praia de Jangamo
(2403’52 S, 3529’33 E) (Figure 2.1). The region is defined by a narrow continental shelf with a steep slope (Swart et al. 2010), warm waters (21-29C), and intermittent coastal upwelling
27 (Lamont et al. 2010), providing for a diverse coral reef complex. The area is a popular destination for recreational SCUBA, snorkeling, surfing, and fishing.
Figure 2.1. Inhambane Province, Mozambique. Four reef regions were chosen for detailed study. Coral reef regions are outlined in green and the soft bottom area, termed the whale shark hotspot (WSH), is outlined in red.
28 Four regions for detailed study were selected based on the geophysical and benthic characteristics: northern reefs, central reefs, central soft bottom, and southern reefs (Figure 2.1).
The northern reefs are dominated by hard plated corals such as Acropora spp. (plate coral),
Porites spp. (porous coral) and Favia spp. (false honeycomb coral). The site is marked by strong currents typically flowing from the northwest to the southeast and is also the closest location to
Baia de Inhambane and estuary system. Reefs within the north site range between depths of 23 and 31 m. While the site provides habitat for large aggregation of schooling fish, due to the variable currents the site is not frequented by recreational divers and fisherman.
The central reef sites are sheltered within Praia do Tofo coastal bay. Moderate currents flow from north to south along the coast allowing for the development of branching hard corals, such as Tubastrea micranthus (green tree coral), as well as soft corals such as Dendronephthya spp.
(branching soft coral). Depth at reefs ranged between 18 and 30 m. Based on its close proximity to Praia do Tofo and depth, this location is heavily trafficked by recreational divers and fisherman
The central soft bottom site ranging from 14m to 18m depth, is located on the leeward side of
Tofino Point, approximately 5km south of Praia do Tofo. The site is an established feeding location of R. typus, M. birostris, and M. alfredi, and frequently visited by recreational snorkelers. Hereafter, this region is referred to as a whale shark hotspot (WSH).
The southern reefs are dominated by rocky substrate and soft corals, such as Lobophytum spp. (leather coral), Sarcophyton spp. (fleshy soft coral), and Dendronephthya spp. (branching soft coral). The reefs are located approximately 20 km south of Praia do Tofo and at a depth of
18-27m.
29 Satellite Imagery and Analysis
Satellite imagery from Landsat 8, MODIS-Aqua, MODIS-Terra, VIIRS-JPSS1, and VIIRS-
Suomi-NPP were examined over the study period. However, only MODIS-Aqua satellite Level 2 images were used in this study. They were obtained from the NASA Goddard Space Flight
Center (GSFC) Ocean Color data (https://oceancolor.gsfc.nasa.gov). All images within the sampling period were assessed for cloud contamination. Data were processed using SeaDAS v7.4 and re-projected with a 1km spatial resolution.
Sampling
The study used ships of opportunity, through Peri-Peri Divers, to conduct environmental and plankton sampling between May 20, 2018 and August 10, 2018. Each region was sampled once per week within an 11-week sampling period, however weather and space limited sampling events (See Appendix Table A – 1 for site-specific sampling date). Nine different sites were visited within the four categorized regions (Figure 2.1). Samples were collected as the boat drifted above the reef site. Physiochemical parameters were recorded at GPS waypoint markings at the beginning of each sample collection. Measurements were recorded using YSI ProDSS multiparameter water quality meter (model #626870-1, #627150-4); parameters included temperature, conductivity, and dissolved oxygen (DO). Calibrations for conductivity and DO were completed prior to each use. Bottom temperature data were recorded using a Suunto D4i dive computer (accuracy ±2⁰C). The thermistor YSI and dive computer were compared on the same day and time at well-mixed, shallow sites and found to be comparable (n = 2). Weather activity, wind direction, percent cloud cover and surface activity (Beaufort scale) were also recorded (listed in Table A.1 in Appendix). Wind data were gathered from World Meteorological
Organization station #67323 located at the Inhambane Airport (FQIN). Sequential 5L discrete
30 samples were collected at a depth of 5m (General Oceanics Model 1010). Approximately 30 ±
5.1 L of water was collected during each sampling event (Table A.1). Samples were gently poured through consecutive 80µm mesh (i.e. zooplankton fraction) and 36µm mesh (i.e. phytoplankton fraction). Particulate matter from 80µm and 36µm samples were then collected and transported back to All Out Africa Marine Research Station (AOA) for processing and preservation. In addition, 1 L of whole water was collected in an acid-washed container for inorganic nutrients, dissolved organic carbon (DOC) and total organic carbon (TOC).
Plankton Analysis
Upon return to AOA, the particulate matter fractions were divided into subsamples (Table
A.2) for pigment analysis and particulate organic matter (POM). Phytoplankton community composition was determined through analysis of signature pigments and verification via microscopy from the 36µm fraction (hereafter referenced as “phytoplankton”). Duplicate samples were filtered through glass fiber filters (GF/F) under gentle vacuum. Filters were frozen and stored for photopigment analysis by high performance liquid chromatography (HPLC) following the method from Pinckney et al. (1996) at the HPLC Photopigment Analysis Facility at University of South Carolina. Briefly, filters were sonicated in 3mL of 100% acetone for 30 seconds and extracted in the dark for 20 to 24 hours at -20°C. Extracts were filtered through
0.2µm and 300µL injected into an HPLC system equipped with reverse-phase C18 columns in series (Rainin Microsorb-MV, 0.46 ×10 cm, 3mm, Vydac 201TP, 0.46 × 25 cm, 5 mm). A nonlinear binary gradient adapted from Van Heukelem et al. (1994) was used for pigment separations. Solvent A consisted of 80% methanol and 20% ammonium acetate (0.5M adjusted to pH 7.2), and solvent B consisted of 80% methanol and 20% acetone. Absorption spectra and
31 chromatograms were acquired using a Shimadzu SPD-M10av photodiode array detector, where pigment peaks were quantified at 440nm.
Photopigment concentrations were examined with CHEMTAX (v.1.95) (Mackey et al. 1996) to determine the absolute abundance of major phytoplankton groups (µg chl a-1) using the convergence procedure outlined by Latasa (2007). The initial pigment matrix was derived from
Barlow et al. (2017) and modified using Higgins et al. (2011) (Table A.3). Measured pigments include chlorophyll a (Chl a), chlorophyll b (Chl b), chlorophyll c1 (Chlc1), chlorophyll c2
(Chlc2), chlorophyll c3 (Chlc3), peridinin (Per), 19’-butanoyloxyfucoxanthin (But), fucoxanthin
(Fuc), neoxanthin (Neo), violaxanthin (Viol), prasinoxanthin (Pras), 19’-hexanoylfucoxanthin
(Hex), alloxanthin (Allo), zeaxanthin (Zea), antheraxanthin (Anth), and lutein (Lut). Previous research suggest that the majority of the phytoplankton community is represented within 20-
200µm size range (van der Molen et al. 2016), phytoplankton groups whose size were <36µm were excluded from the matrix. Pigment ratio values for cyanobacteria were derived from
“cyano-1” pigment values from Higgins et al. (2011) and reported as Trichodesmium because other representation within this group are solely freshwater species. Diatoms are distinguished by the photopigment fucoxanthin, and PN can be further distinguished from this group by containing chlorophyll c3 (Higgins et al. 2011). A subsample of the phytoplankton fraction was preserved with a 5% Lugols solution, additional Lugols was added once the samples were returned to the US.
A portion of the phytoplankton samples (13%) was analyzed to verify CHEMTAX results
(Table A.4). A settling procedure was used to concentrate the samples to 5mL and cell counts were preformed using Sedgwick-Rafter counting cell on an inverted microscope at 10x objective
(ZIESS Axis Observer-A1 axiovert 135), and identified to genus level. A subsample of the
32 zooplankton fractionated (>80µm) plankton was collected and preserved with 5% acidified
Lugols and organisms were identified to genus level using a dissecting scope
(ZIESS SteREO Discovery.V8). Chaetocerous spp. were noted at several sites in the zooplankton fraction, however, due to preservation effects we were unable to accurately quantify the biomass, therefore presence/absence was noted.
Particulate Organic Matter
Duplicate subsamples from both phytoplankton and zooplankton fractions (Table A.2) were used for POM analysis. Duplicate samples were filtered through combusted GF/F under gentle vacuum, frozen and transported to the US. 훿13C and 훿15N analyses were administered by the
University of Minnesota-Duluth Department of Chemistry and Biochemistry at the Large Lakes
Observatory following standard procedures by Bianchi and Bauer (2011).
Nutrient Analysis
Total organic carbon (TOC) samples were collected in 30mL combusted amber borosilicate glass scintillation vials. For dissolved organic carbon (DOC) analysis, particulates were filtered through GF/F using a bell jar vacuum and filtrate was collected. All samples were collected in duplicate. DOC and TOC samples were acidified using 85% phosphoric acid to a pH of >2.0 and processed at University of Minnesota-Duluth Department of Chemistry and Biochemistry at the Large Lakes Observatory following procedures established in Bianchi et al. (2004).
Samples for inorganic nitrogen (NO2, NO3, and NH4), phosphorous (PO4), iron (Fe) and silica (Si) analysis were filtered through GF/F using gentle vacuum and collected in water quality sterile sample bags (VWR # 89085-546) and frozen. Nutrients were analyzed at the Wetland
Biogeochemistry Analytical Services at LSU. Ammonium, nitrate + nitrite and orthophosphate
33 concentrations were determined using standard auto analyzer techniques (EPA 353.4, EPA 350.1 and EPA 365.5). Dissolved inorganic nitrogen (DIN) was calculated as the sum of nitrate, nitrite and ammonium concentrations, and dissolved inorganic phosphorus (DIP) was considered equal to the orthophosphate concentration. Fe and Si were analyzed by ICP-OES using standard methods (EPA #200.7). Quantification limits were 0.1µM for nitrate, 0.2µM for nitrite, 0.1µM for ammonia, 0.04µM for phosphate, 0.012mg L-1 for silica, and 0.007mg L-1 for iron.
Statistical Analysis
Multivariate analyses were performed using SigmaPlot (version 14). Pearson correlations for physiochemical and biological variables in surface water throughout the study were tested.
Significance was defined at p < 0.05. Numbers are reported as the mean ± standard deviation.
Distinct one-way analysis of variance (ANOVA) for temperature, dissolved oxygen and salinity parameters were used to identify differences between regions. A Mann-Whitney t-test was used to identify differences between subsurface and bottom temperatures. Based on the apparent coastal upwelling event, statistical analysis was also performed on datasets collected before and after July 14th, 2018. A Kruskal-Wallis One-Way Analysis of Variance (ANOVA) was used to determine differences between subsurface and bottom temperatures both before and after the July
14th date using Dunn’s Method. For particulate organic matter, linear regressions were performed for phytoplankton and zooplankton C vs. N concentrations before and after July 14th, 2018.
A Shannon Weaver Diversity Index was performed to compare diversity within the zooplankton samples. The following equation was performed on each region:
푠 Shannon Index (퐻) = − ∑푖=푙 푝푖 ln 푝푖 S is equal to the number of species present within the population. Pi is equal to the proportion of the population made up of species i (Pi = n/N). This fraction is multiplied by the natural log of itself. The Shannon Index value (H) (Table 2.2) is equal to the inverse of total sum of these
34 products. Shannon Equitability (EH) is also reported by dividing H by ln(S) with S being the number of species identified within each region. Equitability assumes a value between 0 and 1 with 1 being complete evenness.
Results
Satellite Imagery
A total of 9 cloud-free MODIS-Aqua images between May 20, 2018 and August 15, 2018 were obtained and used to identify the presence of coastal upwelling, cyclonic and anticyclonic eddies, and coastal fronts. An image from July 28, 2018 revealed the occurrence of localized high chla concentrations (thus, elevated productivity) along the coast of Inhambane Province
(Figure 2.2). The circular nature (approximately 100km in diameter) and clock-wise rotation of
Figure 2.2. Sea surface temperature (SST) and chlorophyll a (Chla) images from 28 July 2018 at 13:15 Local Standard Time (LST) reveal the occurrence of a cyclonic eddy along the coast of Inhambane Province. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
35 the the chla feature suggest the existence of a cyclonic eddy. The feature is not apparent in the corresponding SST image, however that could be due to the daytime warming of the surface layer (<1m) measured by satellite sensors. The measurement of SST is confined to 1m whereas chla a can extend deeper. The cyclonic eddy could have begun to impact the adjacent coastal system several days prior, however cloud contamination prevented observation (Blondeau-
Patissier et al. 2014).
Following the cyclonic eddy, chla imaging from August 13 (Figure 2.3) reveal coastal production along Inhambane Peninsula after the cyclonic eddy had dissipated. The high concentration of chla is confined to the narrow shelf along a thermal front seen in the corresponding SST image. This image also indicates that high production continued after sampling period conducted in this study. Supplementary images processed within the sampling period can be found within the Appendix.
Figure 2.3. SST and Chla images from 13 August 2018 at 13:15 LST show coastal production along the shelf of Inhambane Province. Estimate of chlorophyll a inferred from MODIS Aqua image produced using SeaDAS with NASA standard algorithms.
36 Environmental Parameters
Subsurface temperature (T), ranging from 22.7-25.6°C; bottom T, ranging from 21-25°C; and dissolved oxygen, ranging from 6.36-7.04mg L-1, decreased throughout the study period from
Figure 2.4. A) Subsurface and bottom temperature and B) dissolved oxygen concentration during sampling period. Dashed line represents when major upwelling event occurred (July 14-19, 2018).
37 May through August (Figure 2.4), while salinity remained relatively consistent, at an average of
36.58 ± 0.4 at all sites. No significant difference in temperature (p = 0.81), DO (p = 0.76), or
A
B
C
Figure 2.5. A) Wind vectors and frequency of B) wind direction and C) wind speed along Praia do Tofo, Mozambique throughout the sampling period. Data gathered from timeanddate.com and visualized using Python. Vectors point in the direction of wind movement. Wind intensity is conveyed through the length of the vector.
38 salinity (p = 0.75) was observed between regions. A Mann-Whitney t-test revealed a significant difference (p = 0.003) in temperatures recorded at the surface compared to the bottom of the water column. However, these data failed the Shapiro-Wilk normality test. A decrease in T was observed following July 14, 2018, which corresponded to a prolonged period of strong southerly- driven winds (ranging from 2.68 to 6.69 m sec-1) shifting to strong northerly winds (2.25 ±1.1 m s-1) on July 19, 2018 (Figure 2.5), resulting in upwelling-favorable winds along the coast of
Inhambane Province. This event seemed to have an apparent effect on nutrients and biological parameters, therefore is hereafter referred to as a major upwelling event (MUE).
Before the MUE, subsurface T averaged 24.8 ± 0.43̊°C and average bottom T was 23.9 ±
0.67̊°C (Figure 2.4 A). Following the MUE, the averages of subsurface ant bottom T dropped to
23.1 ± 0.31°C and 22.4 ± 0.93°C, respectively. By separating data recorded before and after the
MUE, the data passed the normality test and a one-way ANOVA revealed a difference (p <
0.001) between subsurface and bottom temperature. Multiple comparison was significant for all pairs except for bottom T before the MUE vs. subsurface T after the UME and subsurface T vs. bottom T after the MUE.
Pigment Analysis
Measured chlorophyll a (chla) concentration was used as a proxy for productivity and was significantly affected by SST (p < 0.05). Low chla concentrations were observed early in the sampling period and increased following the MUE event (Figure 2.6). Measured concentration ranged from 4 – 234 ng L-1 prior to the MUE and 19 – 794 ng L-1 after the MUE. A large spike in productivity on August 6, 2018 was observed in the central region on a day with calm conditions and followed by low concentration during an increase in turbulence (Table A.1).
39 Figure 2.6. Pigment community concentrations measured by HPLC analysis and derived using CHEMTAX. Stacked bars represent total chlorophyll a. Prior to MUE, low concentrations are observed and increase by a factor of 2.6 following the MUE.
The phytoplankton community was dominated by diatoms (Figure 2.7), with a significant portion composed of PN spp. Through microscopic analysis, the diatom assemblage was composed of chain-forming genera including Bacteriastrum spp., Thalassiosira spp., Guinardia spp., Chaetoceros spp., and the non-chain forming Rhizosolenia spp. Prevalent dinoflagellates included Protoperidinium spp., Ceratium spp. and Pyrophacus spp., however Alexandrium spp.,
Dinophysis spp., Prorocentrum spp., and Akashiwo spp. were also observed in low abundance, which have the ability to produce phycotoxins. The primary species identified within the “other” category was Phaeocystis spp., a prymnesiophyte, and Dictyocha spp., a unicellular flagellate.
(Table A.5).
40 Figure 2.7. Proportion of each phytoplankton group contributing to total chl-a at A) northern, B) central, C) whale shark hotspot, and D) southern regions. Break in time represents wind-driven upwelling event between July 14-19, 2018.
The northern region tended to have a higher proportion of dinoflagellates (Figure 2.7A), in particular the week prior to the MUE, the community became dominated by dinoflagellates
(92%) as opposed to the diatom-dominated assemblage throughout the remaining sampling period. Trichodesmium spp. tended to have a higher proportion (prior to UE = 4.62 ± 60%, post
MUE = 0.66 ± 1.5%) prior to the MUE, whereas PN increased in community proportion across all regions following the MUE. Across all regions, PN contributed to 13.96 ± 15.6% of the community before the MUE, and 40.11 ± 17.5 % after the MUE. The WSH region had the highest change in PN proportion, with an average 25.2 ± 1.15% before the MUE and became the dominant species with an average of 54.5 ± 4.9% following the MUE.
41 Zooplankton Identification
Copepods represented the most abundant microzooplankton (62.9%) (Table 2.1). In total,
85% of copepods were less then 1000µm, indicating a smaller size fraction, 42.7% (n = 524) of which were nauplius individuals within the zooplankton fraction. Of adult copepods (n = 703), the most abundant genera were Clausocalanus spp. (25.0%, n = 176), Oithona spp. (23.6%, n =
166), Paracalanus spp. (20.6%, n = 145), Macrostella spp. (6.1%, n = 43), Acartia spp. (5.9%, n
= 40) and Oncaea spp. (7.5%, n = 53). No significant difference (p = 0.85) in abundance was found before and after MUE (Figure 2.8. Copepod abundance counted from zooplankton fraction
(>80µm) based on organism size.).
Figure 2.8. Copepod abundance counted from zooplankton fraction (>80µm) based on organism size.
42 Table 2.1. Identified plankton by microscopy and frequency of occurrence across all sites and sampling events.
Type Genus Region Occurrence Frequency Chaetognath Flaccisagitta WSH 1 Chaetognath Sagitta North, South 13 Copepod Acartia All 40 Copepod Clausocalanus All 186 Copepod Calanus All 12 Copepod Halicyclops All 28 Copepod Macrostella All 49 Copepod Oithona All 168 Copepod Oncaea North, WSH, South 52 Copepod Paracalanus All 94 Copepod Subeucalanus North, Central, South 10 Copepod Tortanus North, WSH 3 Diatom Chaetocerous All 11 Diatom Cosinodiscus All 51 Dinoflagellate Ceratium All 29 Dinoflagellate Noctiluca All 78 Dinoflagellate Prorocentrum WSH, South 16 Dinoflagellate Protoperidinium Central, WSH 24 Foraminafera Globigerina All 115 Gastropod Limacina All 47 Lancelet Brachiostoma Central 1 Larvacean Fritillaria North, Central, WSH 28 Larvacean Oikopleura All 62 Larvae Pluteus WSH, South 2 Larvae Balanus All 14 Polychaete Myrianida Central, South 12
43 Large phytoplankton species were also identified in the >80m sample fraction, including both diatom and dinoflagellate genera (Table 2.1). Identification did show that some PN cells were captured within the large fractionation, however cell counts were minimal (16 total cells within 10 samples). Large dinoflagellates such as Ceratium spp. (n = 29), Noctiluca spp. (n =
80), Prorocentrum spp. (n = 16), and Protoperidinium spp. (n = 24) were present in low abundance. Large diatoms such as Coscinodiscus spp. (n = 103) and Chaetocerous spp. were mostly present in low abundance with the exception of 6 samples dominated by Chaetocerous spp.
A Shannon Weaver Diversity Index identified the relative diversity across the four regions
(Table 2.2). The northern region had the lowest diversity index out of all the sites. The sites with the most similar index were the central and south reef regions, indicating similar diversity. The whale shark hotspot region had an index closest to the central reef region.
Table 2.2. Shannon Weaver Diversity Index and Equitability.
Region H EH North 2.71 0.82 Central 2.97 0.83 WSH 2.88 0.80 South 3.04 0.85
Particulate Organic Matter
Prior to the MUE, phytoplankton had lower concentrations of both carbon (9.24±5.0 mg mL-
1) and nitrogen (2.41±1.2 mg mL-1) than zooplankton (13.8±5.0 mgC mL-1, 3.15±1.1 mgN mL-1)
(Figure 2.9). Regressions between cellular C:N revealed coefficients of 0.144 (R2 = 0.395) and
0.196 (R2 = 0.836) for phytoplankton and zooplankton, respectively. Zooplankton showed a tight linear relationship in C:N ratio. Average C:N ratio was 4.07 ± 1.59 for phytoplankton and 4.42 ±
44 Figure 2.9. Ratio of particulate organic carbon (C) to particulate organic nitrogen (N) concentration quantified within phytoplankton and zooplankton fractions A) prior to MUE and B) after MUE.
0.61 for zooplankton. Following the MUE, phytoplankton showed a decreased coefficient of
0.102 (R2 = 0.881) whereas zooplankton maintained a similar relationship of 0.197 (R2 = 0.90).
Cellular nutrient concentrations of phytoplankton (20.9 ± 12 mgC mL-1, 3.78 ± 1.3 mgN mL-1) and zooplankton (22.4 ± 8.9 mgC mL-1, 4.56 ± 1.8 mgN mL-1) increased and average C:N ratios
45 Figure 2.10. Particulate organic C and N concentrations measured within phytoplankton and zooplankton showed increased C:N ratios among regions before and after the upwelling event at A) Northern reefs, B) Central reefs, C) WSH, and D) Southern reefs. exhibited an increase of 0.24 for phytoplankton and 0.12 fold-change for zooplankton. POM concentrations increased across all regions following the MUE (Figure 2.10).
There were two samples with high POM concentrations within the data, but not considered outliers. The high value measured within the central region phytoplankton fraction corresponds to the high productivity seen on 6 August 2018 (Figure 2.6), which contributes to the high average C:N ratio (Table 2.3). One zooplankton sample was recorded to contain a large crustacean larvae (~10mm), attributing for the large concentration of C and N.
Nutrient Analysis
Inorganic nutrient concentrations had no significant difference in concentrations (N: p =
0.573, P: p = 0.300, Si: p = 0.479), which suggest that the coastal system of Inhambane
Province is well-mixed (Figure 2.11, Figure 2.13). Iron concentrations were below detection
46 Figure 2.11. A) Inorganic phosphorous concentrations and B) inorganic nitrogen concentrations measured during field season. Symbols represent the sampled region. Dashed line represents when the MUE took place. limits (0.007 mg L-1) and not reported. Both inorganic nitrogen (NO3 + NH4) and silica were temporally and spatially variable.
NH4 form of nitrogen dominated concentrations with no NO2 detected. Total nitrogen concentrations (N) fluctuated within each region and ranged between 1.60 µM L-1 to 4.15 µM L-
1 (Figure 2.11 B). The Whale Shark Hotspot region averaged lower concentrations (average 2.15
± 0.62 µM L-1) than the reef sites (average 2.66 ± 0.78 µM L-1). The WSH was not affected by
47 the MUE. The North and South regions experienced a 19.7% and 6.9% decrease respectively in
N concentration following the UE (North prior MUE [N] = 2.66 ± 1.15 µM L-1, North following
MUE [N] = 2.14 ± 0.58 µM L-1; South prior MUE [N] = 2.76 ± 0.73 µM L-1, South following
UE [N] = 2.14 ± 0.58 µM L-1). The central region experienced an 8.1% increase in average N following the MUE (prior MUE [N] = 2.73 ± 0.71 µM L-1, following MUE [N] = 2.95 ± 1.1 µM
L-1).
Inorganic phosphorous concentration ranged from 0.161 µM L-1 to 0.36 µM L-1 throughout the study period with an average concentration of 0.21 ± 0.04 µM L-1 (Figure 2.11 A). The whale shark hotspot region had the highest concentrations with a mean of 0.25 ± 0.07 µM L-1, however the reef regions had a collectively similar mean of 0.196 ± 0.02 µM L-1. Following the MUE, the
WSH increased in P concentration by 12.5% (prior MUE [P] = 0.24 ± 0.06 µM L-1, following
MUE [P] = 0.27 ± 0.09 µM L-1).
Prior to the upwelling event, the average N:P ratio was 13.4 with a wide variation (std =
3.92) and following the MUE, this ratio decreased to 10.48 (std = 2.44) with the exclusion of two samples from the central region high in nitrogen (Table 2.3). In addition, the WSH region, which
Figure 2.12. Relationship between inorganic nitrogen and inorganic phosphorous A) prior to UE and B) post MUE. Symbols represent the sampled region.
48 tended to have higher P concentrations, did not follow the trend and had the lowest ratios both before and after the MUE (combined average ratio of 9.07±3.63).
Table 2.3. Average particulate C:N ratios of samples collected within the four study regions before and after the major upwelling event.
Before Following Region Type MUE MUE North Phytoplankton 4.01±0.41 3.77±0.65 Zooplankton 3.85±2.56 2.36±1.86 Central Phytoplankton 4.87±0.69 4.24±1.43 Zooplankton 3.58±1.41 3.80±1.96 WSH Phytoplankton 4.47±0.10 5.34±1.41 Zooplankton 4.62±1.46 4.27±0.47 South Phytoplankton 4.42±0.61 4.98±3.20 Zooplankton 4.28±1.40 3.86±1.02 Inorganic silica concentrations ranged from 0.99 mg L-1 to 2.27 mg L-1 with an average value of 1.41 ± 0.39 mg L-1 (Figure 2.13). The upwelling event had no apparent effect on silica concentration (p = 0.82). Silica was highest in the southern region in late July. Samples collected on the same day within the south region revealed high variability within regions, “Manta” reef had a silica concentration of 2.27 mg L-1 while “Robs” reef had a concentration of 1.13 mg L-1, one of the lowest concentrations. This difference could possibly be an indicator of different processes affecting the reefs in similar proximity or simply indicate that there is high variability within the system.
Discussion
Several coastal processes including wind-driven coastal upwelling are working together to enhance productivity and create community shifts. Many temperate systems experience seasonal increases in productivity during spring and summer months associated with increases in SST and expanded photoperiod (Sigman & Hain 2012), however Inhambane Province, subtropically (or semi-temperate) located near the Tropic of Capricorn (23.5°S), exhibited an increase in
49 Figure 2.13. Silica concentration (average 1.4 ±0.39 mg L-1) during the austral winter of 2018. Symbols represent the sampled region. The upwelling event had no effect on concentration (p = 0.82). productivity during austral winter months (June-August). A decrease in SST throughout the field season was indicative of a seasonal shift into austral winter, and increased productivity (chl-a), suggesting this semi-temperate system is controlled by different factors.
Temperature was similar across all regions and indicated the system is well mixed.
However, it is possible that depth of the dive site (where bottom T was recorded) could play a role in differences between surface and bottom T. Across the sampling period, the difference in T between subsurface and bottom recordings indicates that there is a T gradient within the water column, but these values are not normally distributed, which is likely due to the drop in T following July 14th, 2018. When separated before and after this date, the data are found to be normally distributed and significantly different. The comparison between bottom T before the MUE and subsurface T after the MUE found these values to be statistically similar, which
50 could indicate that the colder bottom temperatures below the surface moved upwards, reflecting the noted rise of thermocline felt by SCUBA divers in late July and early August. Bottom and subsurface T following the MUE was not significantly different, which corresponds to the noted shallow thermocline in the top 5 meters of the water column noted by divers. A rise in the thermocline is an indication of upwelling processes affecting a coastal system and often used as an identifying indicator in satellite imaging (Schumann et al. 1982).
Pigment proportions and cell counts revealed slight differences in patterns, which could be attributed to cell sizes and pigment composition. Pigment ratios were verified with microscope identification (Table A.4). Percentage of cell counts were not always aligned with pigment percentages, in particular, Trichodesmium spp. percentage by cell count is often higher than pigment concentrations, this is due to the small cell size of Trichodesmium spp. create a disproportionate dominance over biomass of other taxa. While pigment ratios were assessed to accurately represent the proportion of Trichodesmium spp. biomass within the sample, it should be noted that due to low concentrations of overall pigments, CHEMTAX analysis may underestimate rare groups. Dinoflagellates, aside from “type 1” containing peridinin
(photosynthetic dinoflagellates), are difficult to distinguish among pigments because they have derived their pigments from other taxa, causing the group to possibly be invisible to pigment analysis (Higgins et al. 2011). Pigment starting ratios for dinoflagellates were adjusted to best represent the community, however, identification counts revealed a general under representation of this group. In general, diatoms were well represented through pigment analysis. PN proportion corresponded with cell counts with the exception of August 6, 2018, which had a higher abundance of non-PN diatoms than suggested via CHEMTAX. Rhizoselenia spp. and
Pseudosolenia spp. were contributors to this sample, which are also considered “type 2” diatoms
51 containing fucoxanthin (Higgins et al. 2011) and accounted for the misrepresentation. These species were not large contributors among other samples, therefore general trends discussed regarding PN still provide an accurate representation of the system.
As found within many coastal systems, our data indicated that Inhambane Province coastal waters are diatom-dominated (Figure 2.6) and reflected similar identified genera (Table A.5) to other studies within the Southwestern Indian Ocean (Barlow et al. 2007, Sá et al. 2013, van der
Molen et al. 2016, Olofsson et al. 2017), including a large proportion of biomass comprised of the pennate diatom, Pseudo-nitzschia spp., a known toxin producing species. Other dominating diatoms were Bacteriastrum spp. and Chaetoceros spp., which were present in all phytoplankton samples. Diatoms tend to have a high C:N ratio (Garcia et al. 2018) which make them a favorable prey item for zooplankton because of their high energy content (Hessen 1992).
Zooplankton identification revealed qualitative insight to the mesozooplankton assemblage but is not a quantitative analysis of this group due to low abundance collected. The primary goal of this identification was to assess the diversity of this size class along the coastal system; therefore, a Shannon Weaver Diversity Index was used to assess differences of present species between the four regions in the study. Diversity and evenness between the regions were similar, which was anticipated based on the small spatial scale of the study, however the northern reefs exhibited a trend for lower diversity. While the WSH had the lowest evenness at 0.80, the region still exhibited a uniform distribution. Copepods made up a large portion of this diversity, which likely contributed to equitability scores ranging from 0.80-0.85, indicating that diversity was not completely even, however relatively high. Given that diversity is important in predicting ecosystem stability (Ptacnik et al. 2008), the data suggests that the northern region is less stable
52 than the central and southern region, while the WSH remains in the middle. Differences in this index are an indicator that slightly variable coastal processes are affecting these areas.
Based on global averages in ambient nutrient ratios, the Redfield Ratio of
106C:16N:1P:1Si:0.01Fe was developed as a baseline for determining nutrient limitations within aquatic systems (Redfield 1958). Nutrient analysis along coastal Inhambane Province revealed an average ratio of 13:1 (N:P), suggesting this system is primarily controlled by nitrogen availability. Silica is an important nutrient for diatoms (Reynolds 2006, Bristow et al. 2017), with overall ratios of 2:1 (N:Si) and 7:1 (Si:P). The low ratio between N:Si further suggests that nitrogen is the primary limiting nutrient for diatom growth, and the high Si:P ratio suggests that this macronutrient is likely not a limiting nutrient. Iron is also a limiting nutrient in systems, however these systems are primarily characterized with high nutrients (Coale et al. 1996); among the oligotrophic coastal system of this study, Fe was undetected and due to its close proximity to shore was not considered a limiting nutrient within the system.
Following the MUE, all sites, with the exception of two samples from the central region, exhibited a reduction in available nitrogen (Figure 2.13) suggesting an influx of new nitrogen during the MUE promoted an increase in productivity and resulted in a decrease in ambient nutrient ratios likely due to nutrient uptake, which is supported by the increase in particulate organic nitrogen (Figure 2.10). The high N:P ratio (>20N:P) within two samples from the central region indicated possible phosphorus limitation within the system and corresponded to a high proportion of PN spp. dominance, supporting previous research which suggested PN spp. was able to maintain high growth under phosphate-limited conditions (Hu et al. 2008).
Following the MUE event, both inorganic N:P ratios (Figure 2.12) and phytoplankton C:N ratios decreased (Table 2.3) suggesting possible nutrient limitation causing a shift to smaller
53 phytoplankton species within the community (Litchman & Klausmeier 2008). Smaller phytoplankton are adapted for rapid growth rates at lower nutrient concentrations (Litchman &
Klausmeier 2008), leading to an increase in trophic structure and an overall decrease in energy movement through the trophic system (Probyn 1992). Energetically, high C:N ratios are more efficient at converting energy between trophic levels, therefore considered more nutritious among primary consumers (Jones & Flynn 2005), however mono-species diet can have decreased nutritional effects on zooplankton. Increased C:N ratios following the MUE could be a crude indication of higher quality prey source, however upon further inspection, data showed PN became dominant in response to the MUE, which could have deleterious nutritional consequences on consumers from decreased diversity. Fatty acid analysis is suggested as a better indicator of nutritional quality (Sterner & Schultz 1998), which captures the essential nutrients often not provided within a monodiet (Jones & Flynn 2005). While fatty acids were not available for this study, they should be included in future research.
Copepods were the most abundant type of zooplankton within samples, which are important for supporting higher trophic webs (Turner 2004). Copepods make up a portion of the diets of planktivorous megafauna feeding along the coast of Inhambane Province, revealed by fatty acid signature analysis (Couturier et al. 2013, Rohner et al. 2013a), however a larger portion is shown to be composed of mysids and sergestids. These demersal zooplankton species were not captured and identified with my sampling methods, which suggests a need for sampling methods that would capture diel migrating zooplankton at night when these species are more active in order to fully capture the trophic interactions between plankton and planktivores.
Variability in environmental parameters such as nutrients was low between regions, however there were slight trends in varying coastal processes. The Whale Shark Hotspot (WSH) region
54 tended to be low in nitrogen and higher in phosphorous, suggesting greater nitrogen limitation for this site. There was a correlation between the low nutrient ratios and PN dominance. This dominance within the phytoplankton community could be the basis for a lower diversity index among the zooplankton compared to the central and southern reef regions. Productivity and zooplankton diversity were lower within the northern region, with productivity identified as the lowest within the area and a low C:N ratio was apparent within this region. The northern reefs are the only region where large megafauna (with the exception of migrating humpback whales) are not sighted. While the differences seen in the northern region compared to the other regions are minute, they indicate key factors that drive the biodiversity within this system.
One explanation for the variability between the regions relates to the small-scale physical dynamics of these locations that promote primary productivity. Localized upwelling from wind- driven dynamics influence coastal systems, which are highly influenced by depth of shelf region and geography of the coast (Hanson et al. 2005). The near-shore dynamics of the WSH region support productivity within turbid waters, which are highly favorable to diatoms (Margalef 1978,
Estrada & Berdalet 1997, Compton 2011) and is a consistent source of mixing within the shallow sandy region, resulting in a higher source of phosphorus to the area. The central and southern regions are also highly influenced by the geography of the peninsula, which acts as a headland, shown to play a pivotal role in the nature of upwelling (Lutjeharms & Stockton 1991). This theory is consistent with the depth profiles of these regions as well, where a slight depth gradient from north (deepest) to south (shallower) could potentially drive variation water mixing properties. These variations are exacerbated as energy is transferred up the food chain and could explain the year-round nature of planktivore sightings within the shallow, diatom-dominant, and productive near-shore region of the WSH region. Central and southern regions were also highly
55 productive, responding to the nutrient pulses, which could be reflected in the increased diversity along the coral reef habitats. To test this idea, sampling during contrasting seasons would help identify if these processes are consistent year-round and could identify further seasonal influences to the plankton community and influence dynamics under warmer surface water conditions found in summer months.
Additional satellite images (see Appendix) also provided insight on coastal productivity within coastal areas outside of the immediate study area, such as within the Baia de Inhambane estuary or further south along the coast by Praia de Zavora. Both these areas are characterized with high concentrations of chlorophyll a. Satellite imagery is a helpful tool for studying large, ocean processes, however coastal systems are much more variable in terms of light interactions with the bottom, with dissolved and suspended pigments and suspended material and thus require in situ verification for accurate analysis of chlorophyll a in particular. During this study, I attempted to ground truth these images along the coast of Inhambane Province, however, was unsuccessful due to logistics with satellite passover times. To understand how these areas are affected by similar processes, there is a need for more collaborative research to assess the use of satellite imaging used to expand the spatial range of the coastal study.
Anthropogenic nutrient loading from terrestrial sources can increase growth of harmful algae, such as PN spp. (Trainer et al. 2010). Data from this research suggest that the coastal system along Inhambane Province is not highly impacted by freshwater input. Salinity throughout the season remained relatively constant, which took place during the dry season. Additional research during the wet season (austral summer) would verify if this trend is present year-round.
Mozambique is forecasted to continue to grow in agriculture (Porter et al. 2017) and aquaculture
(FAO 2005, World Bank 2018a). As the coastal communities develop and increase agriculture
56 and aquaculture production, determining the impact of freshwater runoff can help predict changes in nutrient input.
In addition, the Praia do Tofo region would benefit from the understanding of seasonal dynamics on plankton assemblage and community dynamics. Warmer surface temperatures and increased irradiance (not measured within this study) would likely influence the community and present different trends in environmental parameters, especially nutrients. The influence of
Trichodesmium spp. was not significant within the study period, however is noted to be seasonally affected by the monsoon season in nearby Tanzania (Lugomela et al. 2002), promoting the recycling of limited N through nitrogen fixation properties. This species is also recognized as a harmful species in these areas (Tamele et al. 2019) and is predicted to increase in prevalence under elevated surface temperature conditions.
Changes to mesoscale processes affecting environmental conditions and nutrient transport are predicted to occur within the Southwestern Indian Ocean (Backeberg & Reason 2010, Behagle et al. 2014), possibly impacting the factors influencing primary productivity. The Indian Ocean is warming faster than any of the other global oceans and will likely have profound effects on surrounding coastal communities (Hermes et al. 2019). Annual variability in monsoon events is predicted to intensify and become more frequent. Monsoon events affect Mozambique in late austral Summer (January – March), including Cyclone Favio in February 2007 and Cyclone
Dineo in 2017 affecting Inhambane Province. These events have shown to increase productivity in the Northern Indian Ocean (Madhu et al. 2002) and south of Madagascar (Uz 2007). Increased nitrate sources during monsoon season have shown to promote PN booms in Tanzania (Hamisi &
Mamboya 2014). Warming is predicted to have slowed effects on the Indian Ocean Dipole, which contributes to interannual variability in rainfall and is likely to result in decreased rainfall
57 (Saji et al. 1999). Overall, rainfall in southern Africa is expected to become less frequent, however intensity of rainfall events is expected to increase (Usman & Reason 2004). These processes are likely to affect coastal systems, resulting in decreased overall terrestrial runoff, while increase the intensity of nutrients being inputted into the coastal system, which have increased the frequency of harmful cyanobacteria blooms within Tanzania (Lugomela et al.
2002).
58 Chapter 3. Environmental influences on domoic acid production by Pseudo- nitzschia spp. along coastal Inhambane Province, Mozambique
Harmful algal blooms (HABs) are increasing globally in frequency, persistence, and geographic extent and threaten the economy of coastal communities and ecosystem and human health. However, in many parts of the globe, especially in developing countries, there is insufficient monitoring of HAB events, as this is evident in eastern Africa. With an extensive coastline and an abundance of marine natural resources, Mozambicans rely on the coastal ecosystem for food and economic security. This is especially true in Southern Mozambique, which is a known biodiversity hotspot supporting large megafauna such as Rhincodon typus
(whale shark), who’s year round presence supports an expanding ecotourism industry. Previous studies have identified Pseudo-nitzschia spp. (PN), a known producer of the neurotoxin, domoic acid (DA), within phytoplankton communities along the Mozambican coast. This is the first study to document the occurrence of DA along the coast of Inhambane Province in Southern
Mozambique, specifically within an area with the highest sightings of R. typus. While concentrations of DA were low, ranging between 0.102 and 11.7 pg particulate DA L-1, impacts to ecosystem health through bioaccumulation and magnification could have broader impacts to coastal communities. Wind-driven upwelling along the coast promoted primary production and increased the abundance of PN. Peak DA concentrations corresponded to a decrease in subsurface temperature and upwelling-favorable wind conditions. DA concentrations tended to be higher under conditions of phosphorus limitation. Domoic acid was also quantified in mesozooplankton samples, providing evidence for bioaccumulation within a system, as mesozooplankton function as a vector for higher trophic level biomagnification. These findings suggest a need for continued and comprehensive monitoring of DA along southern Mozambique.
59 Introduction
Harmful algal blooms (HABs) are a natural phenomenon, which are increasing globally in frequency, persistence, and geographic extent due to anthropogenic influences (Smayda 1990,
Van Dolah 2000, Graneli & Turner 2006, Heisler et al. 2008, Paerl & Paul 2012, Davidson et al.
2014, McKibben et al. 2015, Wells et al. 2015, Hallegraeff 2019). Increases in HAB events coincide with global climate change and coastal eutrophication from human development along coastlines and waterways (Van Dolah 2000, Anderson et al. 2008). These blooms can cause substantial harm to coastal societies through economic impairment by increased costs associated with bloom management and decreased revenue during fisheries closures (Anderson et al. 2000,
Grattan et al. 2016, Willis et al. 2018). For example, in the United States, the economic impact of
HABs is reported to be over $49 million per year from public health (45% of total economic impact) and commercial fisheries (37%) (Anderson et al. 2000), however this estimate has likely grown in the last 20 years with increased HAB events. Understanding coastal dynamics contributing to the presence and duration of HABs is important for economic security and human health.
Domoic acid (DA) has received increased attention among global monitoring efforts since the first outbreak resulting in the death of at least three people following the consumption of contaminated shellfish from Prince Edward Island in eastern Canada (Bates et al. 1989). Domoic acid is a potent neurotoxin produced by several species of the genus Pseudo-nitzschia, leading to both acute and chronic toxicity in mammals, fish and birds (Work et al. 1993, Lefebvre et al.
2002, Bargu et al. 2010, Fire et al. 2010, Zabaglo 2016, D'Agostino et al. 2017). The neurotoxic effects of DA on mammals is due to its chemical structure similar to glutamic acid, an excitatory neurotransmitter. DA permanently binds to the glutamate receptors, eventually causing cell death
60 (Goldstein et al. 2008). Human consumption of highly contaminated shellfish can lead to
Amnesic Shellfish Poisoning (ASP); symptoms include gastrointestinal distress, confusion, disorientation, seizures, short-term memory loss, and in rare cases can cause death (Bates et al.
1989, Jeffery et al. 2004, Lefebvre & Robertson 2010, Saeed et al. 2017). No human fatalities have been reported since the implementation of monitoring programs, however DA still causes harm to the ecosystem through the accumulation of toxin within the food web (Goldstein et al.
2008, Trainer et al. 2012, D'Agostino et al. 2017, Saeed et al. 2017, Bates et al. 2018).
Domoic acid poisoning can affect marine mammals, birds, and other organisms through the bioaccumulation of DA within vector organisms (Bargu et al. 2010, McKibben et al. 2015,
McCabe et al. 2016). Humans are most directly exposed to DA through shellfish, however, vector species within the ecosystem can include zooplankton, benthic organisms and small fishes
(Bargu et al. 2002, Del Rio et al. 2010, Leandro et al. 2010, Bargu et al. 2011b, Tammilehto et al. 2012, Baustian et al. 2018). Vector organisms often exhibit no physical effect from DA accumulation (Bejarano et al. 2008), which allow the toxin to become biomagnified within the food web, impacting higher trophic organisms such as birds and marine mammals. Along the
California coast in 2015, the consumption of toxic anchovies led to the mortality of over 400
California sea lions (Zalophus californianus) all exhibiting signs of neurological dysfunction
(Trainer et al. 2000, Bauer 2006, McKibben et al. 2015, Bates et al. 2018). Similarly, DA detected in fecal samples was implicated in southern right whales calf mortality off the coast of
Argentina (D'Agostino et al. 2017). The interactions of DA within the food web are complex, but occur under environmental conditions that promote PN growth.
Individual species of PN have specific growth and toxin production dynamics in response to temperature, salinity, irradiance, photoperiod, and variation in nutrients (Hasle 1995, Pan et al.
61 1998, Wells et al. 2005, Fehling et al. 2006, Thessen 2007, Cochlan et al. 2008, Hu et al. 2008,
Seeyave et al. 2009, Thessen et al. 2009, Anderson et al. 2010, Auro & Cochlan 2013, Downes-
Tettmar et al. 2013, Terseleer et al. 2013, Martin-Jezequel et al. 2015, McCabe et al. 2016,
Shanks et al. 2016, Lema et al. 2017, McKibben et al. 2017, Thorel et al. 2017, Delegrange et al.
2018, Radan & Cochlan 2018, Van Meerssche et al. 2018). For example, P. delicatissima is documented to dominate waters with lower salinity and irradiance and cold temperatures in systems off Scotland (Fehling et al. 2006), whereas P. seriata bloom during warmer temperatures with higher ammonium levels. Through lab experiments, Fehling et al. (2004) documented increased DA production under phosphorus (P) and silica (Si) limitation by P. seriata, and which has also been seen in the natural system within the Bay of Seine (France)
(Thorel et al. 2017). Experiments with nutrient ratios have demonstrated that P. pungens is able to maintain high growth under low P conditions (Hu et al. 2008) and globally have expanded to coastal systems that have experienced nitrogen loading (Casteleyn et al. 2008, Lundholm et al.
2010). Experiments in the Pacific Northwest have shown P. multiseries, noted as one of the most efficient DA producers (Hasle 2002), to exhibit higher DA production under ammonia and urea sourced N as opposed to nitrate (Radan & Cochlan 2018), whereas cultures from European waters revealed DA production at its highest under urea enrichment (Martin-Jezequel et al.
2015). These complex species interactions highlight the importance of understanding species- specific factors contribute to DA toxicity.
Pseudo-nitzschia spp. is a globally cosmopolitan diatom that is particularly prevalent in
Eastern Boundary Systems (EBS), including the California, Humbolt, Canary, and Benguela current (Pitcher & Pillar 2010, Trainer et al. 2010). In these areas, blooms typically occur between late spring and summer when upwelling introduces key nutrients, particularly nitrogen
62 (Trainer et al. 2000, Trainer et al. 2002, Fawcett et al. 2007). Shellfish closures from PN blooms have become an increasing trend within European shellfish fisheries, particularly in the UK,
Ireland and France (EFSA 2009, Trainer et al. 2012). The seasonal variation of these bloom events have been extensively studied along the Canary current and shown to result in species succession, primarily controlled by increasing photoperiod, beginning with the onset of spring nutrient upwelling.
Incidences of DA within western boundary systems (WBS) have historically been less prevalent. Blooms of Pseudo-nitzschia spp. along WBS typically occur in shallow bays or estuaries where nutrients are terrestrially sourced, such as along the coast of China and eastern
United States (Thessen 2007, Anderson et al. 2010, Van Meerssche & Pinckney 2017, Jiang &
Wang 2018). Blooms of PN along the continental shelf of Maine are driven by the Maine Coastal
Current causing nutrient upwelling and limitation of growth by silica (Fernandes et al. 2014). In the southern hemisphere, DA emerged as a potential threat to fisheries among WBS in New
Zealand beginning in the 1990’s (Rhodes et al. 1998, Rhodes et al. 2013). In eastern South
America, a bloom of P. australis produced detectible levels of DA within the Argentine Sea in
2000 (Negri et al. 2004, Almandoz et al. 2017). In 2001, DA was reported in mussels in Uruguay
(Méndez et al. 2012), followed by a reported contamination of farmed mussels in Brazil, leading to a shellfish fishery closure for 25 days (Fernandes & Brandini 2010). While worldwide, EBS have received attention for increases in PN abundance and DA events, coastal zones worldwide are being impacted by DA due to increased frequency of warm ocean anomalies (McKibben et al. 2017) and WBS could be a potential area of concern for future toxic blooms.
To protect human health from the threats posed by DA contamination, many global monitoring efforts have set a regulatory limit of 20 µg DA/kg shellfish, which is based on the
63 lowest level of DA reported to cause toxic effects in humans (Lefebvre et al. 2017). The purpose of these efforts is to mitigate the acute exposure of DA, however this creates a gap in the understanding of health impacts through chronic exposure at low doses (Hiolski et al. 2014). A study on mice by Lefebvre et al. (2017) found that repeated DA exposure in low doses impacted cognitive function and memory, however, the mice were able to recover following the cessation of exposure. Other studies have found that populations where subsistence shellfish harvest is a common practice could be at risk for chronic exposure and long term memory loss (Grattan et al.
2016, Ferriss et al. 2017). With this in mind, it is important to study toxin production even in areas where toxin production could be low but sustained for long periods of time.
Harmful algal blooms within the Southwestern Indian Ocean (SWIO) are understudied due to a lack of monitoring infrastructure, despite the presence of HA species and documented accounts of marine toxins within the region (Tamele et al. 2019). Okadaic acid and ciguatoxin, produced by dinoflagellates, have been recorded along the east coast of Madagascar and nearby islands including Mauritius, the French Islands, Comoros, and Seychelles (Boisier et al. 1995, Bouaïcha et al. 2001, Tamele et al. 2019). An outbreak in 1993 along the east coast of Madagascar following the consumption of a single shark contaminated with ciguatoxin led to the hospitalization of 188 people and resulted in the death of 50 people, including a dog that was reported to have licked the blood from the harvested shark (Boisier et al. 1995). South Africa has the most developed HAB monitoring program within the SWIO. Multiple derivatives of okadaic acid and yessotoxins have been reported within mussels along the western side of the country associated with the Benguela current, however, no marine toxins have reported to cause shellfish closures along the eastern coast (DAFF 2016). With the global trend of expanding distributions
64 of toxic algal species, monitoring efforts in South Africa are actively exploring phytoplankton distribution and ecology.
The coastal waters of Southern Mozambique feed the Agulhas Current, one of the strongest
WBS in the world (Bryden et al. 2005, Lutjeharms 2006, Lamont et al. 2010). This current system is greatly influenced by the East Madagascar Current and eddies within the Mozambique
Channel (Lutjeharms 2006). In the Indian Ocean, the South Equatorial Current flows west and diverges around Madagascar with one current flowing around the north end of the island and another flowing south, forming the East Madagascar Current (Quartly & Srokosz 2004, Raj et al.
2010). Water flowing south within the Mozambique Channel along the steep slope, narrow shelf
(Swart et al. 2010) forms both cyclonic and anticyclonic eddies (Lutjeharms 2006). Westward cyclonic (cold core) eddies are a common along the south of Madagascar during austral winter with a lifetime from a weeks to months (Lamont et al. 2014, Barlow et al. 2017). The southern end of the channel and East Madagascar Current merge between Maputo and Durban, forming the Agulhas Current. These waters are primarily oligotrophic, however mesoscale features such as coastal upwelling caused by wind stress and cyclonic, cold-core eddies promote increased primary production along the southeastern coast of Africa (Lamont et al. 2010, Barlow et al.
2014).
Coastal phytoplankton dynamics along the majority of Mozambique’s coastline are under studied, but the plankton community contributing to the highly productive Delagoa Bight in
Maputo Bay and Sofala Bank have been broadly studied (Sá et al. 2013). The Sofala Bank area is an important fishery ground associated with nutrient input from the Zambezi River primarily dominated by diatoms (70-90%), including Chaetoceros spp., Pseudo-nitzschia spp., Proboscia alata, Cerataulina pelagica and Thalassionema nitzschioides and small abundance of
65 dinoflagellates (Sá et al. 2013). Contrastingly, the Delagoa Bight productivity is driven by mesoscale features which replenish nutrients throughout most of the year (Barlow et al. 2007), which allow microphytoplankon (20-200 µm) to be dominant. These features influence the biological productivity supporting higher trophic level diversity along southern Mozambique
(Tew-Kai & Marsac 2009, Tew Kai & Marsac 2010, Ternon et al. 2014).
To date, there have been no reported HAB-related events in Mozambique (Tamele et al.
2019). PN is among the diatom assemblage of coastal Mozambique (Sá et al. 2013), however species level identification has not been confirmed. Of the 52 identified species of PN, 26 (50%) are toxigenic (Bates et al. 2018). PN spp. exhibit species-specific dynamics with regards to toxin production and responses to available nutrients (Trainer et al. 2012, Thorel et al. 2017). Three generalized groups have been suggested based on morphological characteristics: P. pungens/P. multiseries group, P. australis/P. seriata group, and P. delicastissima group (Trainer et al. 2008,
Lelong et al. 2012). A study by Lugomela (2013) identified three species of PN along the coast of Tanzania, including P. pungens, P. cuspidate, and P. seriata, all of which are documented to produce toxins in other parts of the world (Hasle 1995, Hasle 2002, Fehling et al. 2004,
Casteleyn et al. 2008, Trainer et al. 2009, Trainer et al. 2012, Tammilehto et al. 2015, Bates et al.
2018). In addition, P. multiseries has been identified to be responsible for DA accumulation in mussels within Algoa Bay, South Africa (Pitcher et al. 2014).
Particulate DA has been documented along the coast of Inhambane Province (Figure 3.1) and suggests a seasonal spike in production at the beginning of austral winter (Errera, personal communication). This is the first documentation of DA production within the Southwestern
Indian Ocean and while concentrations are low the presence of DA could lead to broader impacts on ecosystem health. Coastal Inhambane Province is a biodiversity hotspot for whale sharks,
66 manta rays, migrating humpback whales, bottlenose dolphins, humpback dolphins, dugongs, sea turtles, small-eye stingray, guitar sharks, corals, and many reef fishes (Pierce et al. 2008, Rohner et al. 2013b, Pereira et al. 2014, Williams et al. 2017, Fordyce 2018, Rohner et al. 2018). This biodiversity supports a prominent tourism industry for SCUBA diving, snorkeling, surfing, and fishing. In addition, roughly two-thirds of Mozambique’s population lives along the coast and relies on marine resources as a source of food and income (FAO 2017, World Bank 2018a), with nearly 1.5 million people living in Inhambane Province (Instituto Nacional de Estatística, 2017).
The reliance on the coastal system within the community highlights the importance to assess the drivers and influences of coastal productivity.
The purpose of this research is to provide the first comprehensive study on PN spp. and the occurrence of DA in Inhambane Province during a potential peak season. Globally, DA
Figure 3.1. Particulate domoic acid detected within coastal waters of Inhambane Province, Mozambique in area with high sightings of Rhincodon typus (whale shark) between January 2017 - May 2018 (Errera, unpublished data) from WSH.
67 production by PN is increasing in geographic extent and intensity with changes to coastal environments, including western boundary upwelling systems (Trainer et al. 2008, Sekula-Wood et al. 2011, Trainer et al. 2012, Trainer et al. 2019). Currently there are no studies related to the coastal ecology of PN spp. in Mozambique, however limited studies related to phytoplankton dynamics have identified a cosmopolitan occurrence within the Mozambique Channel and coastal community (Raj et al. 2010, Sá et al. 2013, Barlow et al. 2014, Barlow et al. 2017,
Olofsson et al. 2017). In addition, very few of these studies have quantified inorganic nutrients shifts within the coastal system of Mozambique (Olofsson et al. 2017). Specifically this research will compare the occurrence of DA and environmental conditions across four regions, which differ in sightings of whale sharks. I aim to identify differences in the dynamics where planktivores are commonly sighted versus the areas where they are less common. Given the importance of planktivores in the area and societies dependence on the productivity of the system, this research could provide essential data to support economic growth within the region.
Furthermore, this research will serve as a baseline for future changes to the coastal area and identify environmental factors that promote DA production.
Methods
Study Site
The study was conducted along the coast of Inhambane Province, Mozambique, stretching approximately 40 km from Barra Beach (2347’35.00S, 3531’05.00 E) to Praia de Jangamo
(2403’52 S, 3529’33 E) (Figure 3.2). The region is defined by a narrow continental shelf with a steep slope (Swart et al. 2010), warm waters (21-29C), and coastal upwelling, providing for a
68 diverse coral reef complex. The area is a popular destination for recreational SCUBA, snorkeling, surfing, and fishing.
Figure 3.2. Inhambane Province, Mozambique. Four reef regions were chosen for detailed study. Coral reef regions are outlined in green and the soft bottom area, termed the whale shark hotspot (WSH), is outlined in red.
Based on the geophysical and benthic characteristics, four regions were chosen for detailed study: northern reefs, central reefs, central soft bottom, and southern reefs (Figure 3.2). The northern reefs are dominated by hard plated corals such as Acropora spp. (plate coral), Porites spp. (porous coral) and Favia spp. (false honeycomb coral). The site is marked by strong currents
69 typically flowing from the northwest to the southeast and is also the closest location to Baia de
Inhambane and estuary system. Reefs within the north site range between depths of 23 and 31 m.
While the area provides habitat for large aggregations of schooling fish, due to the variable currents, this region is less frequented by recreational divers and fisherman.
The central reef sites are sheltered within Praia do Tofo coastal bay. Moderate currents flow from north to south along the coast allowing for the development of branching hard corals, such as Tubastrea micranthus (green tree coral), as well as soft corals such as Dendronephthya spp.
(branching soft coral). Depth at reefs ranged between 18 and 30 m. Based on its close proximity to Praia do Tofo and depth, this location is heavily trafficked by recreational divers and fisherman.
The central soft bottom site ranging from 14m to 18m depth is located on the leeward side of
Tofino Point, approximately 5km south of Praia do Tofo. The site is an established feeding location of R. typus, M. birostris, and M. alfredi, and frequently visited by recreational snorkelers. Hereafter, this region is referred to as a whale shark hotspot (WSH).
The southern reefs are dominated by rocky substrate and soft corals, such as Lobophytum spp. (leather coral), Sarcophyton spp. (fleshy soft coral), and Dendronephthya spp. (branching soft coral). The reefs are located approximately 20 km south of Praia do Tofo and at a depth of
18-27m.
Satellite Imagery and Analysis
MODIS-Aqua satellite Level 2 images were obtained from the NASA Goddard Space Flight
Center (GSFC) Ocean Color data (https://oceancolor.gsfc.nasa.gov). All images within the sampling period were assessed for cloud contamination. Data were processed using SeaDAS v7.4 and re-projected with a 1km spatial resolution.
70 Nutrient Analysis
Samples for inorganic nitrogen (NO2, NO3, and NH4), phosphorous (PO4), iron (Fe) and silica (Si) analysis were filtered through GF/F using gentle vacuum and collected in water quality sterile sample bags (VWR # 89085-546) and frozen. Nutrients where analyzed at the Wetland
Biogeochemistry Analytical Services at LSU. Ammonium, nitrate + nitrite and orthophosphate concentrations were determined using standard auto analyzer techniques (EPA 353.4, EPA 350.1 and EPA 365.5). Dissolved inorganic nitrogen (N) was calculated as the sum of nitrate, nitrite and ammonium concentrations, and dissolved inorganic phosphorus (P) was considered equal to the orthophosphate concentration. Fe and Si was analyzed by ICP-OES using standard methods
(EPA #200.7). Quantification limits were 0.1µM for nitrate, 0.2µM for nitrite, 0.1µM for ammonia, 0.04µM for phosphate, 0.012mg L-1 for silica, and 0.007mg L-1 for iron.
Sampling
Plankton surveys along the coast of Praia do Tofo were designed to investigate the temporal and spatial patterns of DA-producing PN abundance. Sampling occurred opportunistically in conjunction with Peri-Peri Divers. Each region was sampled approximately once per week within an 11 week period, however weather and access limited sampling events (site-specific sampling dates listed with Table A.1). Nine different sites were visited within the four regions
(Figure 3.2). Samples were collected as the boat drifted above the reef site. Physiochemical parameters were recorded at GPS waypoint markings at the beginning of each sample collection.
Measurements were recorded using YSI ProDSS multiparameter water quality meter (model
#626870-1, #627150-4); parameters included temperature, conductivity, and dissolved oxygen
(DO). Calibrations for conductivity and DO were completed prior to each use. Weather activity, wind direction, percent cloud cover and surface activity (Beaufort scale) were also recorded
71 (Table A.1). Wind data were gathered from World Meteorological Organization station #67323 located at the Inhambane Airport (FQIN).
Sequential 5L discrete samples were collected at a depth of 5m (General Oceanics Model
1010). Approximately 30 ± 5.1 L of water was collected during each sampling event (Table A.1).
Samples were gently poured through consecutive 80 µm mesh (i.e. zooplankton fraction) and 36
µm mesh (i.e. phytoplankton fraction). I used a 36 µm mesh to target PN cells for particulate domoic acid (pDA) analysis. An 80 µm size mesh was used to separate zooplankton from PN cells based on the assumption that PN cells would pass through, however microscopic verification confirmed most cells were captured within the small fractionation.
Once all collected water had passed through the column, the particulate matter from
80 µm and 36 µm (hereafter referred to as “zooplankton” and “phytoplankton” respectively) was collected and transported with ice packs to All Out Africa Marine Research Station for processing and preservation. Subsamples of phytoplankton (duplicate) and zooplankton (single) for particulate DA (pDA) were filtered onto 25 mm glass fiber filters (GFF) and frozen for future processing. A subsample of the phytoplankton fraction was preserved with acidified Lugols and used to determine PN abundance using the Utermöhl method (Utermöhl 1958).
Pigment Analysis
Phytoplankton productivity was determined by the concentration of chlorophyll a (chla) concentration. Duplicate samples were filtered through glass fiber filters (GF/F) under gentle vacuum. Filters were frozen and stored for photopigment analysis by high performance liquid chromatography (HPLC) following the method from Pinckney et al. (1996) at the HPLC
Photopigment Analysis Facility at University of South Carolina. Filters were sonicated in 3mL of
100% acetone for 30 seconds and extracted in the dark for 20 to 24 hours at -20°C. Extracts were
72 filtered through 0.2µm and 300µL injected into an HPLC system equipped with reverse-phase
C18 columns in series (Rainin Microsorb-MV, 0.46 ×10cm, 3mm, Vydac 201TP, 0.46 × 25cm,
5mm). A nonlinear binary gradient adapted from Van Heukelem et al. (1994) was used for pigment separations. Solvent A consisted of 80% methanol and 20% ammonium acetate (0.5M adjusted to pH 7.2), and solvent B consisted of 80% methanol and 20% acetone. Absorption spectra and chromatograms were acquired using a Shimadzu SPD-M10av photodiode array detector, where pigment peaks were quantified at 440nm.
Particulate Domoic Acid Analysis
Particulate DA (pDA) was assessed by Biosense ASP ELISA (Amnesic Shellfish Poison, enzyme-linked immunosorbent assay) (Biosense #A31300401) using the protocol described in
Del Rio et al. (2010), Baustian et al. (2018), expressed as pg pDA L-1. Samples were extracted using 20% methanol and sonicated, and cell debris was removed. Plates were read using
Thermo Varioskan Flash spectral scanning multimode reader and SkanIt Software version 2.4.1.
Toxin concentrations (pDA) were calculated using a non-linear 4-parameter logistic curve fit provided by Abraxis (Abraxis, Inc, Warminster, PA). Detection limit was 9.33 ±1.1 pg DA mL-1 across the analyzed plates; concentrations below detection limit were reported as zero.
Statistical Analysis
Multivariate analyses were performed using SigmaPlot (version 14). Distinct one-way analysis of variance (ANOVA) for temperature, dissolved oxygen and salinity parameters were used to identify differences between regions. I also used a Mann-Whitney t-test to analyze these data sets by values recorded before and after July 14th, 2018 to assess the effect of upwelling.
Pearson correlations for physiochemical and biological variables in surface water throughout the
73 study were tested (Table 3.1). Significance was defined as a p value < 0.05. Principle component analyses (PCA) were performed to characterize the relationships between physiochemical parameters (temperature, salinity, and nutrients) and biological parameters (chlorophyll a, PN cell counts, and domoic acid).
Table 3.1. Pearson correlation for physiochemical and biological variables over study period (May - August 2018). Values in bold with an asterisk are significant at α = 0.05.
Temp Salinity NO3 NH4 PO4 Si N:P N:Si Si:P Chla pDA PN cells Temp 1 -0.608* 0.0105 0.113 -0.174 -0.0116 0.174 -0.00731 0.0954 -0.562* -0.558* -0.348 Salinity 1 0.0935 -0.341* 0.396 -0.0609 -0.489* -0.129 -0.283 0.271 0.574* 0.164 NO3 1 -0.0475 0.513* 0.0225 -0.0590 0.121 -0.164 0.209 -0.0545 0.0788 NH4 1 -0.215 -0.0749 0.918* 0.794* 0.0499 0.190 -0.373 0.335 PO4 1 -0.0561 -0.511* -0.0249 -0.489* 0.154 -0.0233 -0.0320 Si 1 -0.0160 -0.610* 0.878* -0.270 0.0415 -0.140 N:P 1 0.681* 0.256 0.156 -0.319 0.322 N:Si 1 -0.499* 0.431 -0.278 0.459* Si:P 1 -0.301 0.0207 -0.120 Chla 1 0.111 0.863* pDA 1 -0.0111 PN Cells 1
Results
Environmental Parameters
No significant difference in temperature (p = 0.81), DO (p = 0.76), or salinity (p = 0.75) was observed between regions, therefore environmental conditions were assumed to be similar throughout the study area. Between July 14-19, 2018, a significant decrease (p < 0.001) in subsurface temperature (T) was observed. Also during this time, there were strong southerly winds, which shifted to strong northerly winds, a change in direction that would induce wind- driven coastal upwelling. This event seemed to have an apparent effect on nutrients and biological parameters, therefore is hereafter referred to as a major upwelling event (MUE). Slight trends in nutrients reflected this theory, primarily observed with a decreased N concentration following the MUE (p = 0.124). This decrease likely reflects the uptake in nutrients by increased
74 primary productivity, which also saw a significant increase (p < 0.001) in chla following this period.
Satellite Imagery
A total of 9 cloud-free MODIS images between May 20, 2018 and August 15, 2018 were obtained and used to identify the presence of coastal upwelling, cyclonic and anticyclonic eddies, and coastal fronts. An image from July 28, 2018 revealed the occurrence of localized productivity along the coast of Inhambane Province (Figure 3.3). The circular nature
(approximately 100km in diameter) and clock-wise rotation of the the chla feature suggest the existence of a cyclonic eddy. The feature is not apparent in the corresponding SST image, however that could be due to the daytime warming of the surface layer (<1m) measured by satellite sensors. The measurement of SST is confined to 1m whereas chla a can extend deeper.
Figure 3.3. Sea surface temperature (SST) and chlorophyll a (chla) images from 28 July 2018 at 13:15 LST reveal the occurrence of a cyclonic eddy along the coast of Inhambane Province. Estimate of chlorophyll a inferred from MODIS Aqua image produced using SeaDAS with NASA standard algorithms.
75 The cyclonic eddy could have begun to impact the adjacent coastal system several days prior, however cloud contamination prevented observation (Blondeau-Patissier et al. 2014).
Pseudo-nitzschia Abundance
Cell concentrations ranged from 1 – 528 cells L-1 and did not correspond to pDA concentration (Figure 3.4). The highest value was recorded on August 6, 2018 in the central region corresponding with the highest recorded productivity. The concentrations across the study region was 33 ± 33 cells L-1 (n=33) throughout the season. Prior to the MUE, cell concentrations averaged 28 ± 34 cells L-1 (n=17), and exhibited a 0.37 fold increase to 38 ± 32 cells L-1 (n=16) following the MUE.
Figure 3.4. Pseudo nitzschia spp. cell abundance and pDA from phytoplankton fraction.
Among individual regions, the northern region had the lowest concentration of cells (average
16 ± 15 cells L-1, n=7) and exhibited a 3.89 fold increase in concentration following the MUE
(prior average 6 ± 5 cells L-1, n=4; post average 30 ± 12 cells L-1, n=3). The central region experienced the highest incerase (9.93 fold increase) in cell abundance following the MUE (prior
76 average 5 ± 5 cells L-1, n=4; post average 54 ±49 cells L-1, n=6). Both the southern region and the WSH exhibited a decrease (-0.44 and -0.21 fold change, respectively) in PN spp. abundance following the MUE. Averages for these regions prior to upwelling were 49 ± 42 cells L-1 (n=7) and 44 ± 27 cells L-1 (n=2), respectively and following the MUE decreased to 27 ± 27 cells L-1
(n=4) and 35 ± 18 cells L-1 (n=3), respectively.
Light microscopy revealed the presence of at least three species within the study area (Figure
3.5). Distinguishing characteristics of these types included variable width, shape, and length.
One type (Figure 3.5 A/B) had a width of 3 µm and a short lenth (>100 µm). Another type
(Figure 3.5 C/D/E) had a similar length (>100 µm), but a width of 5µm. The third type (Figure
3.5 F/G) was distinguished by the very long cell shape (200+ µm) with finely pointed ends and a width of 5µm.
Figure 3.5. Light microscopy revealed at least three species of Pseudo-nitzschia spp. present in samples from coastal Inhambane Province based on width and shape.
77 Particulate DA
During the sampling period, pDA was detected in 79% of phytoplankton samples and 74% of zooplankton samples with a range of 0.166 to 8.33 pg pDA L-1 and 0.102 to 5.05 pg pDA L-1 respectively (Figure 3.6). The highest total concentration of 11.7 pg pDA L-1 was observed on
July 27, 2018 at the whale shark hotspot region.
Figure 3.6. Particulate DA concentrations measured along the coast of Praia do Tofo during the austral winter months in 2018.
There was a 2.11 fold increase of pDA following the MUE (Figure 3.6). Prior to the MUE, combined pDA concentrations averaged 0.903 pg L-1 (70% detected, n=20). All samples following the MUE (n=15) had detectable pDA with a total concentration of 4.02 pg L-1. All regions experienced a similar trend (Figure 3.7), with the reef regions having a greater increase in pDA compared to the whale shark hotspot (WSH) region. The northern reefs experienced a
9.13 fold increase from 0.484 ±0.92pg L-1 (40% detected, n=5) to 4.91 ±1.3pg L-1 (n=4). The
78 central sites experienced a 3.56 fold increase from 0.507 ±0.48pg L-1 (80% detected, n=5) to 2.31
±0.44pg L-1 (n=6). The southern reefs experienced a 9.2 fold increase from 0.377 ±0.97pg L-1
(71% detected, n=7) to 3.84 ±1.0pg L-1 (n=3). All samples (n=5) at the WSH region measured detectable levels of pDA during the sampling period and experienced a 1.21 fold increase from
2.70 ±1.5pg L-1 (n=2) to 5.96 ±4.9pg L-1 (n=3) following the MUE.
The percentage of pDA concentration measured within the two size fractions varies among regions (Figure 3.7). The zooplankton fraction (>80 µm) comprised an average of of 69%
(±25%) of measured pDA (n=6) among the northern reefs. The proportion of pDA measured within the zooplankton fraction at the central (n=9) and WSH (n=5) regions were similar with an
Figure 3.7. Particulate DA concentrations separated by regions, A) north reefs B) central reefs C) whale shark hotspot and D) south reefs. Dashed line identifies when upwelling event occurred.
79 average of 47.0% (±25%) and 49.6% (±20%) respectively. The southern reefs had the lowest proportion of zooplankton pDA being 39.6% (±21%, n=9).
Principle Components
Principle components analysis (PCA) identifying the first two axes accounted for 56% of the total variance between physiochemical and biological parameters (Figure 3.8). Chlorophyll a
(chla) and Pseudo-nitzschia spp. (PN) are both positively influenced within the system.
Figure 3.8. Principle component analysis (PCA) variable factors from May to August 2018 defined by the two first axes showing the link between physiochemical parameters, chla, pDA and Pseudo-nitzschia spp. abundance.
80 Abundance of PN was also positively correlated with N:Si ratio, which is expected because N tended to be limiting in the environment and Si is a required nutrient by diatoms. In contrast, Si is negatively influenced within the system and negatively correlated to high phytoplankton biomass (chla and PN abundance) due to high diatom growth. pDA was most closely related to phosphorus and salinity, suggesting that pDA concentration is positively influenced by these elements. Salinity is known to have a positive effect on DA (Bargu et al. 2016, Van Meerssche &
Pinckney 2017), however these studies took place in areas with high freshwater input, unlike coastal Inhambane Province, which stayed relatively constant throughout the study period.
Therefore, I concluded that salinity does not have an influence on pDA in this region. PCA analysis suggests that pDA is negatively influenced by the N:P ratio, or rather when N is high, causing the N:P ratio to be low, pDA is higher. From PCA analysis, it is difficult to distinguish whether the positive influence of P or negative influence of N:P ratio has a greater effect on pDA, however given that N is typically limiting within the system, the ratio of these nutrients may play a large role in DA concentration. Temperature and ammonia (NH4) did not seem to have a clear relationship with DA production or chla concentration.
Individual factors by region revealed slight variations (Figure 3.9). Central reefs were positively controlled by principle component (PC) 1 and the WSH region was more positively controlled by PC 2. Both northern and southern regions were negatively influenced by PC 2, however the northern region was more influenced by PC 1 than the southern sites.
Discussion
Attention to PN has increased over the last few decades due to their production of the neurotoxin, domoic acid, which has the potential to cause harm to human health and the environment. On a global scale, DA has been particularly prevalent among eastern boundary
81 Figure 3.9. PCA individual factors by region as a function of physiochemical and biological parameters. systems and in temperate climate regions where upwelling of nutrients promote toxicity (Trainer et al. 2008, McCabe et al. 2016). Long-term monitoring has revealed increases in community dominance by PN and increased toxicity under warming global climate conditions and increased nutrients from terrestrial sources (Anderson et al. 2010, Lundholm et al. 2010, Trainer et al.
2012, Rhodes et al. 2013, McCabe et al. 2016, Smith et al. 2018). There has been a recent emergence of DA contamination of shellfish fisheries within WBS such as Argentina (Almandoz et al. 2007, D'Agostino et al. 2017), Uruguay (Méndez et al. 2012), Brazil (Fernandes &
Brandini 2010), New Zealand (Rhodes et al. 2013), as well as South Africa (Pitcher et al. 2014), supporting the trend for increased spatial extent of this HAB species. This study documents the
82 first occurrence of a biotoxin along the coast of Mozambique and the first documentation of DA along the eastern Africa within a subtropical WBS.
The variability in DA concentration could be linked to the intensity of upwelling events.
Northerly winds along the coast of Inhambane Province appear to create upwelling-favorable conditions, which likely provide year-round sources of small-scale nutrient pulses. These localized upwelling processes could explain the year-round detection of toxin within the preliminary study. The peak concentration in 2018 occurred on 27 July 2018, following a wind- driven upwelling event and aligned with a cyclonic eddy identified using ocean color satellite imagery (Figure 3.3). Preliminary data during 2017 also revealed a peak in mid-May aligning with the beginning of austral winter. Upwelling is known to play an important role in the production of DA, most notably along the western coast of the United States (Trainer et al. 2000,
Trainer et al. 2002, McCabe et al. 2016, Smith et al. 2018), but in these areas, PN exhibits a seasonal bloom during the spring transition into summer. In Mozambique, cyclonic eddies are known to occur during austral winter (Lamont et al. 2014), as opposed to spring along the
California coast (Anderson et al. 2006). During this sampling, it appeared that that coastal upwelling increased pDA concentration along Inhambane Province, however other systems with established monitoring programs exhibit spatial and temporal variation in DA concentration
(Trainer et al. 2002, Husson et al. 2016, Shanks et al. 2016). Therefore to further understand these processes and variation, long-term continued monitoring is needed to fully assess the seasonality of DA concentration along the coast of Inhambane Province.
While the overall concentrations within this study period are low (0.62 pg pDA L-1 – 11.71 pg pDA L-1), toxin has been detected year-round (Errera, personal communication). The water- soluble nature of the toxin prevents DA from accumulating within the ecosystem without a
83 source of production (Wohlgeschaffen et al. 1992, Quilliam et al. 2006). In areas where DA is known to occur with a seasonal bloom, monitoring is focused during bloom periods because the ecosystem is exposed within a shorter window. However along Inhambane Province, there could be concern for chronic exposure to the ecosystem and possibly human health, which has been shown to cause declines in cognitive function to the central nervous system (Hiolski et al. 2014,
Ferriss et al. 2017, Lefebvre et al. 2017). This study only focused on quantifying particulate DA within plankton samples and further investigation into vector organisms is needed to assess the threat of chronic DA exposure to the ecosystem.
DA concentration was found to be positively related to P concentration and negatively associated with ambient N:P ratios, suggesting higher DA concentration under greater phosphorus-limited conditions consistent with prior studies (Pan et al. 1996). Another previously explored aspect to this relationship suggests a restriction of DA production under nitrogen limited conditions resulting from an insufficient pool of cellular free N for synthesis of the nitrogenous toxin (Pan et al. 1998). However, studies with multiple species of PN reveal various growth responses to nutrient ratios (Hu et al. 2008), suggesting a more complex relationship with nutrients is species-specific.
Concentration of DA did not correspond to abundance of PN, similar to previous studies
(Sekula-Wood et al. 2011, Husson et al. 2016, Bates et al. 2018), which suggest both cell abundance and toxin concentration must be included in monitoring efforts. High variance in cell abundance suggests high day-to-day fluctuations within the study site. High variability could also be the result of low sampling volume or low abundance overall within the system.
Increasing the volume of water collected could help with this variance in abundance. During preliminary DA sampling, only 175 mL of water was collected for analysis. In efforts to best
84 quantify toxin concentration, the volume processed was increased to an average 4.5 ±1 L during this study, however recorded concentrations were similar in magnitude, suggesting the increased processing volume did not increase DA detection sensitivity. Cell counts were not conducted during preliminary sampling, however it is likely that larger processing volume is helpful in quantifying PN cells.
Based on light microscopy identification of PN spp. cells, at least three species of PN were suggested simultaneously present along the coast of Inhambane Province. Based on prior reports of P. multiseries, P. pungens, P. cuspidate, and P. seriata along the eastern coast of Africa
(Lugomela 2013, Pitcher et al. 2014), these species are likely to be similar, however difinitive identification of PN species is based upon variations in frustule morphology and require the use of electron microscopy (Trainer et al. 2008). Thorel et al. (2017) has demonstrated the importance of identifying species-specific influences to DA production, noting that simultaneously occurring species are affected by different conditions. Based on light microcopy measurements of cell length and width (Trainer et al. 2008, Pitcher et al. 2014, Almandoz et al.
2017), these species are suggested to be similar to Pseudo-nitzschia multiseries (Figure 3.5 A/B),
P. australis (Figure 3.5 C/D/E), and P. turgidula (Figure 3.5 F/G). All three species have been recorded as toxin producers, however P. multiseries is known to be a highly toxic species
(Trainer et al. 2008).
Along the coast of Inhambane Province, the highest toxin concentration was observed in the whale shark hotspot, a common area for aggregations of Rhincodon typus, Manta alfredi, and M. birostris (Rohner et al. 2013b). On average, half the pDA concentration was measured within the zooplankton fraction, providing evidence for toxin bioaccumulation within a region of high importance for reoccurring megafauna. With sightings of these megafauna being the main driver
85 of ecotourism in the area (World Bank 2006, Tibiriçá et al. 2011, Haskell et al. 2015, Venables et al. 2016), further assessment of toxin trophic transfer is needed to understand and manage for ecosystem health.
Overall there were slight variations between the four regions in DA concentration and PN cell abundance with the largest differences being between the North and South regions. As seen in other systems (Trainer et al. 2002, Negri et al. 2004, Fehling et al. 2006, Sekula-Wood et al.
2011, Downes-Tettmar et al. 2013, Husson et al. 2016), cell concentrations did not correspond to toxin concentrations. For example, prior to the MUE, the southern reef area had a high abundance of PN cells, but decreased in average cell concentration following the MUE, at the same time as the cell decrease pDA increased by 920% (Figure 3.7). In contrast, following the
MUE, both cell concretion and pDA in the northern reefs increased significantly. The northern reefs are absent of large planktivores, however could be a potential first colonizing reef site for juvenile fish since it is closest to the estuary and is an area commonly fished by local fishermen.
Differences in pDA between phytoplankton and zooplankton proportions among regions indicate potential differences in accumulation processes within the food web. Specifically, a greater proportion of pDA sampled within the northern region was measured in the zooplankton fraction compared to the southern region. One explanation for lower cell abundance in the north could be explained by the presence of grazers feeding upon PN spp. causing cell counts to be low. Small zooplankton feeding upon toxic diatoms are able to accumulate toxins (Bargu et al.
2003, Harethardottir et al. 2015, Bates et al. 2018), posing a threat to higher trophic levels, which suggests that while concentrations in this region are low, there could be a higher potential for biomagnification. Prior studies have also documented the ability for copepod grazers to induce toxin production (Bates et al. 2018, Lundholm et al. 2018), possibly accounting for the higher
86 concentration of DA, however in order to assess this interaction, further studies would require the quantification of zooplankton grazers and feeding experiments using cells from the area.
When assessing the impact of DA within the food web, it is important to identify potential vector species and the bioaccumulation dynamics leading to biomagnification within the food web. The coastal waters of Inhambane Province are an important biodiversity hotspot for megafauna and fishes, which the community depends on for a source of tourism, recreation and livelihood (World Bank 2006, Rohner et al. 2013b, Fordyce 2018, World Bank 2018a). Using fatty acid analysis, whale sharks and manta rays off the coast of Mozambique have been found to primarily feed on mysids, shrimp (sergestids), copepods, and myctophid fishes (lantern fish)
(Couturier et al. 2013, Rohner et al. 2013a), suggesting that not only do these creatures feed at the surface, but also in bathypelagic waters. The high content of mysids suggests that these planktivores extensively feed at night, when these zooplankton are known to vertically migrate to surface waters. Therefore, further studies are needed to access diel migrating zooplankton and their role as potential vectors. In addition to demersal zooplankton, benthic bottom feeders are especially important for their role in enhancing food capture by resuspending fallen matter back into the food web (Gili & Coma 1998). Prior studies have found DA to affect benthic food webs
(Parsons et al. 2002, Kvitek et al. 2008, Baustian et al. 2018). My study along Inhambane
Province identified DA within the water column directly above highly diverse coral reef habitats.
Changes to mesoscale processes affecting environmental conditions and nutrient transport are predicted to occur within the Southwestern Indian Ocean (Backeberg & Reason 2010, Behagle et al. 2014), possibly impacting the factors influencing PN assemblage dominance and pDA concentration. The Indian Ocean is warming faster than any of the other global oceans and will likely have profound effects on surrounding coastal communities (Hermes et al. 2019). Annual
87 variability in monsoon events are predicted to intensify and become more frequent. Monsoon events affect Mozambique in late austral Summer (January – March), including Cyclone Favio in
February 2007 and Cyclone Dineo in 2017 affecting Inhambane Province. These events have shown to increase productivity in the Northern Indian Ocean (Madhu et al. 2002) and south of
Madagascar (Uz 2007). Increased nitrate during monsoon season have shown to promote PN booms in Tanzania (Hamisi & Mamboya 2014), however cyclones are a relatively recent threat to southern coastal systems, have not been studied along the coast of Mozambique. Warming is predicted to have slowed effects on the Indian Ocean Dipole, which accounts for interannual variability in rain fall and is likely to contribute to decreased rainfall (Saji et al. 1999). Overall, rainfall in southern Africa is expected to become less frequent, but increase in intensity (Usman
& Reason 2004). These processes are likely to affect coastal systems with decreased terrestrial runoff. Although it is important to note that increased rain intensity has promoted harmful cyanobacteria along coastlines in Tanzania (Lugomela et al. 2002).
88 Chapter 4. Concluding remarks on plankton community dynamics effect on Pseudo-nitzschia spp. and domoic acid: implications for ecosystem and human health Summary
The data collected during the austral winter in 2018 reveal that the Inhambane coastal region is a well-mixed system with dynamics that support a high biodiversity along Inhambane
Province. The system is primarily driven by localized wind-derived upwelling as a source of nutrient input into the oligotrophic coastal system. Here I present findings that nitrogen (N) is the primary driver of phytoplankton dynamics based on low inorganic nutrient ratios. These types of nutrient dynamics support a diatom-dominant community, known to energetically enhance diverse trophic webs. A key shift in the phytoplankton community occurred in late July when upwelling-favorable wind conditions corresponded to elevated chlorophyll a (chla) concentrations, increased assemblage dominance of Pseudo-nitzschia spp. (PN) and increased concentration of neurotoxin domoic acid (DA) within the system. Principle component analysis highlighted the positive relationship between pDA and phosphorus (P) and negative relationship between pDA and N:P ratio, suggesting that overall DA concentration increases with either the limitation of P within the environment, the addition of N, or a combination of these two processes regulates toxin concentration. Slight variations in nutrient profiles, pDA, and PN abundance are noted among the four sampled regions suggesting factors influencing biotoxin concentration are complex and vary by location-specific environmental dynamics and species.
As the first documentation of DA along the coast of east Africa, this study highlights the need for a continued and comprehensive toxin monitoring program to further assess ecosystem and human health impacts.
89 The variation among the four regions seen in the nutrient dynamics and biological diversity
could be small-scale evidence for support of the intermediate disturbance hypothesis (IDH) (Dial
& Roughgarden 1998), which states that highly diverse systems are most stable with moderate
disturbances. In the case of my study, upwelling acts as a physical disturbance driving nutrient
input, promoting diversity within the coastal system. The estuary within Baía de Inhambane is
low-flowing and tidally influenced (Reeve-Arnold, personal communication), therefore is not a
significant contributor of nutrients to the coastal reef systems. This idea could explain the lower
diversity among the north reefs, which is suggested to have less physical disturbance based on
the deeper nature of the water column compared to the regions further south and the further
proximity to shore (Hanson et al. 2005). In contrast, the WSH is located within close proximity
to shore with little to hard benthic structure, resulting in more turbid water conditions favored by
diatoms (Margalef 1978, Estrada & Berdalet 1997). In this region, the apparent dominating
nature of PN is supported by this theory that high disturbance facilitates competitively dominant
species (Smith et al. 2018). The potential seasonal peaks in DA within the system (Figure 4.1),
Figure 4.1. pDA measured along Inhambane Province between January 2017 - August 2018.
90 suggest that bioaccumulation within the food web could have a seasonal signature impacting local coastal predators when conditions favoring DA production are reached.
The IDH theory suggests that moderately disturbed systems are better fit to recover from a disturbance. The presence of a HAB species can disturb food web structure, through shifts in food preferences (Bargu et al. 2003, Harethardottir et al. 2015, D'Agostino et al. 2017).
Copepods, found to be a significant component to mesozooplankton assemblage, are known to selectively feed on non-toxic species of PN when community diversity lends the opportunity
(Olson et al. 2006), i.e. when the assemblage is diverse, copepods will feed on non-toxic species.
The north region was lowest in diversity (Table 2.2) and had the highest toxin concentration within the zooplankton fraction (Figure 3.7), suggesting the low diversity was effected by the disturbance of toxic PN in the area resulting in higher bioaccumulation of DA. However, pigment analysis showed that PN spp. was not the dominant diatom within the system, therefore, the zooplankton could be experiencing an ecological trap (Dwernychuk & Boag 1971, Ratti &
Reese 1988), feeding on the emerging toxic diatoms while it negatively effects their fecundity
(Lincoln et al. 2001). This warrants further study to investigate the grazing behavior of zooplankton to accurately make accumulation predictions.
The high biodiversity along Inhambane Province is an indicator that this area receives a relatively steady source of nutrients to support higher trophic levels. Mozambique is unique from other western boundary systems (WBS) which typically have a wide, shallow continental shelf and are less productive than eastern boundary systems. Mozambique has a narrow, steep shelf particularly along Inhambane Province, which allows access for deep, nutrient-rich water to become upwelled into coastal waters (Lutjeharms & Stockton 1991). The biological impact of upwelling has been found to be a function of geographical location, primarily with respect to the
91 width of the continental shelf and the strength and duration of upwelling-favorable winds (i.e. intensity of upwelling) (Hanson et al. 2005). Cool waters with elevated chlorophyll a were observed along Inhambane Province during the study period in response to strong, upwelling- favorable wind conditions, which supports the idea that the geographical location (narrow, steep shelf) and upwelling-favorable winds are the foundation for the biodiversity of this area. The combination of these types of biophysical factors present along Inhambane Province support increased biodiversity, which has been shown in other oceanic systems to support a higher abundance of megafauna (Sleeman et al. 2007).
Southern Mozambique is home to the largest aggregations of juvenile male whale sharks
(Rohner et al. 2013b). Rhincodon typus (whale shark) is known to globally occur in frontal and boundary currents (Colman 1997, Eckert & Stewart 2001, Wilson et al. 2001, Duffy 2002,
Rohner et al. 2013b), where primary productivity is enhanced by these physical dynamics.
Mozambique is unique in regards to the year-round nature of R. typus sightings (Rohner et al.
2018). Other R. typus hotspot areas typically have a seasonal component that aligns with seasonal productivity. Other planktivores such as Manta alfredi are common residents year- round and are more frequently seen when there is a prevailing poleward current (Rohner et al.
2013b), which is a feature of upwelling-favorable conditions. Being a resident species, M. alfredi is more likely to exhibit selective behavior in response to local conditions and take advantage of enhanced productivity, similar to the Ningaloo Reef (Sleeman et al. 2007). These local trends are more difficult to correlate to the behavior of highly migratory species such as R. typus, however satellite tagging has revealed a preference for cooler shelf waters with higher chlorophyll a
(Rohner et al. 2018). Photo identification and genetic data has suggested that individuals frequenting aggregation sites along Mozambique are part of a mixed population within the
92 Western Indian Ocean and Arabian Gulf (Prebble et al. 2018), suggesting that trends along
Inhambane Province have larger implications for the species.
Rhincodon typus is listed as a globally threatened species by the IUCN with a decreasing population (Pierce & Norman 2016). Global analysis of coastal upwelling has predicted the coastal waters of Southern Mozambique to warm with global climate changes (Varela et al.
2018), which could have an impact on the habitat preference of the species. Decreases in sightings along Southern Mozambique have been suggested to be related to foraging activities
(Rohner et al. 2018) and could be impacted by the warming prediction. Other areas of the Indian
Ocean have reported declines in R. typus sightings (Theberge & Dearden 2006, Bradshaw et al.
2008, Rowat et al. 2009b, Sequeira et al. 2013), which have been linked to climatic warming predictions.
Suggesstions for Future Work
The results of the present study are informative for understanding the environmental factors and nutrient dynamics that promote toxic Pseudo-nitzschia spp. blooms along the coast of
Inhambane Province. The next step could consist in providing scenarios for ecosystem-based management of nutrients as the coastal system further develops and developing early warning tools for potential increases in DA. Predictive models have been previously developed for
Pseudo-nitzschia spp. blooms in other ecosystems (Blum et al. 2006, Anderson et al. 2009, Lane et al. 2009, Cusack et al. 2015) and similar models for the coastal ecosystem of Inhambane
Province could predict the probability for a toxic bloom based on environmental predictor variables. The development of these models would require continued and comprehensive data collection and monitoring including PN abundance, plankton assemblage diversity, pDA concentration, and physiochemical data. The development of such a model would be a valuable
93 model within a WBS influenced by localized wind-driven upwelling and mesoscale eddies and aid in understanding the environmental factors that impact similar systems and how these systems will change with climate.
In this study, I have documented the presence of a harmful biotoxin primarily influenced by coastal physical dynamics, however the absence of a monitoring program in Mozambique has created a gap in education among health professionals regarding the symptoms of toxin- contaminated seafood illnesses, which could then lead to misdiagnosis. For example, malaria is a common illness that affects coastal populations of Mozambique and exhibits similar symptoms including headaches, nausea and vomiting, which are also common first symptoms of seafood- related illnesses, including Amnesic Shellfish Poisoning. In addition to monitoring efforts, health professionals within coastal communities need to me trained to recognize symptoms of common HAB-related seafood illnesses. Paralytic (PSP) and Diarrhetic Shellfish Poisoning
(DSP) are common in South Africa (Pitcher & Calder 2000) and Ciguatera Fish Poisonings
(CFP) are commonly reported along Madagascar and surrounding islands within the
Southwestern Indian Ocean (Bouaïcha et al. 2001, Puech et al. 2014, Diogene et al. 2017,
Tamele et al. 2019). In addition to PN found along Inhambane Province, other toxin-producing species were present within the plankton assemblage (Table A.5), such as Dinophysis spp. and
Prorocentrum spp., which produce Okadaic acid (DSP); Prorocentrum lima, which produces
Ciguatoxins (CFP); Protoperidinum spp., which produces Azaspiracids; and Alexandrium spp., which makes saxitoxin (PSP). The Indian Ocean is known for endemic production of ciguatoxins and other derivatives (Tamele et al. 2019), and yet the plankton dynamics associated with these toxins are understudied. Mozambique’s extensive coastline has the potential for these toxins to
94 emerge, however further ecosystem monitoring is needed to understand the nature of plankton assemblage dynamics that promote the toxic nature of these species.
I was able to identify slight variations among the regions within my study area, but unable to determine if this variation is actually between these regions or simply day-to-day variations within the system. My study was limited by the access to sampling sites based on vessels of opportunity. This allowed me to identify slight variations among the regions, with large variation within and among sites. To further identify spatial variation with regards to plankton and nutrient dynamics, future studies should expand to include sampling in all regions and multiple sites within each region on the day of sampling. Sampling multiple regions on the same day, will help reveal if this variation is primarily temporal (day to day) or spatial (site-specific or region- specific). Increased sampling could also reduce variation at sites within regions and eliminate the day-to-day variation within the plankton community that limited the analysis during my study.
Additionally, this study would benefit from additional sampling within summer months to identify seasonal shifts in plankton dynamics and provide further understanding on drivers of productivity during the wet season when terrestrial runoff would increase. This information would enhance the ability to create nutrient-based ecosystem management suggestions for coastal systems, helping to sustain healthy ecosystems as development of Mozambique continues. Remote sensing can provide time-series observations along this extensive under- sampled coastline to aid in the understanding of linkages between the wind-driven coastal upwelling physical, chemical and biological processes.
Ideally, further field studies along this coastline would include preferred sampling days to coincide with the passover times of ocean-monitoring satellites, such as NASA’s MODIS-Aqua satellite and Suomi VIIRS (as examples), to create datasets able to be used to create and verify
95 algorithms used in ocean color satellite-derived data. Remote sensing is a powerful tool used to study remote areas and expand temporal inferences through available access to long term datasets and has been used to monitor and model HAB events within aquatic systems worldwide
(Anderson 2009, Hu 2009, Hu et al. 2010a, Hu et al. 2010b, Klemas 2011, Zhang et al. 2014).
However, these images need to be combined with in-situ data to correct for variation in reflectance along the shallow shelf regions and other optical interferences, such as color dissolved organic matter (Tilstone et al. 2013, Zhang et al. 2014). In pelagic systems, drifting
ARGOS floats provide this data used in calibration, however these drifting floats are limited in data along coastal systems. Reflectance corrections require additional sampling collection of hyperspectral properties including hyperspectral reflectance (Rrs) and fluorometric pigment signatures throughout the water column. Specific algorithms created for coastal Inhambane
Province would be used to correct and validate satellite images allowing for remote monitoring of bio-optical properties and contributing pigments in real-time and expand temporal datasets of the coastal system. Algorithms for satellite imaging would also allow for spatial expansion of coastal ecosystem monitoring. NASA-corrected Level 2 MODIS-Aqua images used in this study suggested high productivity along Inhambane Province outside of the study region, including off the coast of Praia do Zavora and within Baia de Inhambane. However, in-situ data would distinguish if these high values were due to optical interferences such as sediments, CDOM or shallow bottom reflectance, which can cause inaccuracies during high and low chlorophyll a conditions.
While satellite images can be used to help identify physical and biological processes affecting these areas, they have limitations in monitoring capabilities. Optical satellite sensors are only capable of observing the thin surface layer, therefore phytoplankton dynamics below
96 this surface layer are not captured. For example, remote sensing has expanded spatial and temporal bloom dynamics of harmful cyanobacteria across Lake Erie, and in situ monitoring has aided in the understanding bloom dynamics of vertically stratified phytoplankton in the water column (Bosse et al. 2019). Along Inhambane Province, low productivity was observed in the top 5m of the water column, however the addition of fluorometric vertical stratification of phytoplankton over complex coral reef habitats would further the understanding of dynamics and diversity within these habitats. In addition to understanding the plankton dynamics, toxin concentration is not correlated with bio-optical properties, therefore in situ monitoring is the only source of data regarding toxin dynamics within the plankton community.
The study would benefit from expansion of phycotoxin monitoring efforts in these areas. The
Baia de Inhambane is centrally located between the two largest populated cities in Inhambane
Province. Along the coast of California where HABs exhibit a strong seasonal marker, environmental sampling processors (ESP) have aided in the remote detection of domoic acid
(Doucette et al. 2009), this technology could also be useful within Inhambane Province and aide in expanding the temporal and spatial dataset, however this technology is costly. Implantation of a monitoring network using solid phase adsorption toxin tracking (SPATT) bags would be a cost- effective, simple in-situ method for monitoring DA and other phycotoxins within the system
(MacKenzie et al. 2004). SPATT provide a passive sampling method and an integrated snapshot of DA in the water column which would be easily deployed and recovered.
Increases in population coupled with increasing agriculture and aquaculture practices have left coastal communities around the world more vulnerable to nutrient pollution, increasing coastal eutrophication and harmful algal blooms (Paerl 1997, Turner et al. 2002, Heisler et al.
2008, Lundholm et al. 2010). Coastal communities in Mozambique experiencing changes at an
97 alarming rate (World Bank 2018b), and particularly has interest in developing aquaculture for shrimp farming (FAO 2018a), which is a practice associated with nutrient increases to coastal systems (Hamisi & Mamboya 2014). In addition to DA, increased temperatures due to climate change and nutrient runoff from agriculture could promote growth of harmful cyanobacteria along coastal areas, however in the absence of monitoring and education about the effect of algal toxins, signs and symptoms of toxin contamination could be already occurring. For instance, cyanobacteria blooms are increasing in occurrence within freshwater lake systems in South
Africa (Matthews & Bernard 2015), Tanzania (Lugomela et al. 2002, Lugomela et al. 2006), and
Kenya (Kotut et al. 2006). Within Inhambane Province, agriculture and freshwater tilapia aquaculture have expanded in recent years at the consequence of wetland removal surrounding the Baia de Inhambane, an important nursery habitat for larval fishes. This habitat is critical for diversity and lack of understanding of nutrient dynamics and terrestrial runoff currently exists for this area. Changes to nutrient loads into this system would likely result in faster change than the nearby coastal system along Praia do Tofo, where mesoscale physical disturbances vary annually and seasonally, compared to estuarine systems which experiences daily variations in tidal flux and greater terrestrial water influx. The shallow and stagnant nature of this ecosystem emphasize the potential for eutrophic conditions favored by harmful cyanobacteria blooms, which could cause not only an effect within the estuary but also affect the recruitment of larval fishes dependent on this habitat, leading to bottom-up decreases in fisheries abundance and biodiversity.
Mozambique is a developing country with one of the fastest growing economies in sub-
Saharan Africa (World Bank 2018b) following suppression of population growth due to a prolonged civil war. As the country focuses on the enhancement of its economic independence,
98 sustainable development is critical for alleviating poverty among economically unstable communities. Roughly 94% of the poor population primarily engages in agriculture practices
(World Bank 2018b) and increases in agriculture by 50% since the early 1990’s have helped alleviate poverty (Porter et al. 2017). Poverty is widespread among coastal communities, which depend on coastal resources for food security and/or livelihood (Menezes et al. 2011b). The government has identified fisheries and aquaculture as a target for economic growth, currently these industries are constrained by lack of infrastructure, including access to clean water and electricity for fish preservation and roads accessing rural fishing communities to markets. Non- renewable resources markets such as offshore mineral (Lehto & Gonçalves 2008) and hydrocarbon exploration (Nairn et al. 1991, Zhou et al. 2013), are a threat to conservation along southern Mozambique and have high potential for rapid economic development (Yager 2015).
Increased development in these areas could lead to negative consequences to coastal ecosystems through increased pollution including but not limited to noise pollution from seismic exploration, marine debris from expanding coastal populations, and nutrient loading from agriculture and aquaculture production. Mozambique has the opportunity to support current monitoring efforts for HABs and plankton assemblage dynamics that can be used to create necessary correction algorithms for remote monitoring by satellite and suggest best management practices to prevent impairment to coastal communities.
99 Appendix. Supplementary Data
Table A.1. Sampling data.
Time Water DO DO % Cloud Wind Total Location Date Salinity Condition Beaufort Start Temp (%) (mg/L) Cover Direction Filtered (L) Sherwood 2018-05-20 10:50 25.4 100.7 6.72 36.43 0 Sunny SE 4 15 Robs 2018-05-21 8:53 25.3 100.6 6.71 36.54 25 Sunny SE 1 27 Robs 2018-05-28 8:28 25.6 101.5 6.73 37.14 25 Sunny NE 1 38.4 Giants 2018-05-30 8:00 25.6 101.1 6.68 37.55 0 Sunny none 1 39 Reggies 2018-06-02 11:12 25.3 102.8 6.9 35.75 0 Sunny N 1 32.2 Manta 2018-06-03 11:38 25 102.5 6.91 35.85 25 Sunny N 2 34.2 Salon 2018-06-06 8:40 24.5 102.4 6.89 36 90 Raining S 5 34.7 Reggies 2018-06-11 11:23 24.9 100.3 6.76 35.79 25 Sunny SE 2 32.2 Manta 2018-06-12 8:48 24.8 100 6.74 35.96 75 Cloudy E 4 26.5 Robs 2018-06-15 8:32 24.7 99.8 6.75 35.91 0 Sunny NE 2 32.2 Salon 2018-06-17 14:29 24.6 101.1 6.85 35.86 10 Sunny SE 2 32.1 Office 2018-06-21 11:35 24.9 102.3 6.89 36.29 25 Sunny S 1 28.3 Marble 2018-06-22 11:55 24.6 101.1 6.84 36.41 10 Sunny N 1 32.3 Arches Robs 2018-06-25 9:52 24.7 103.3 6.95 36.44 0 Sunny NE 2 21.2 Office 2018-06-29 11:20 25.3 102.6 6.86 36.89 25 Sunny SE 3 26.5 Giants 2018-07-02 14:25 24.7 101.2 6.81 36.4 0 Sunny NE 4 26.2 Manta 2018-07-09 10:30 24.2 102.2 6.94 36.49 50 Sunny SE 5 27.1 Manta 2018-07-10 9:39 24.1 97.5 6.69 36.4 25 Sunny SE 4 32.5 Office 2018-07-11 11:14 24.6 102.3 6.88 36.51 25 Sunny SE 3 32.5 Marble 2018-07-14 11:35 24.5 103.4 7.04 36.4 5 Sunny NE 1 32.4 Arches (table cont’d)
100 Time Water DO DO % Cloud Wind Total Location Date Salinity Condition Beaufort Start Temp (%) (mg/L) Cover Direction Filtered (L) Giants 2018-07-19 15:15 23.6 102.2 6.96 36.5 75 Cloudy SE 5 27.3 Reggies 2018-07-20 14:34 23.5 102.1 6.99 36.56 50 Cloudy SE 4 21 Giants 2018-07-23 8:43 23 100.7 6.95 36.73 25 Sunny S 5 26.4 Manta 2018-07-24 8:57 22.9 99 6.92 36.69 50 Cloudy E 3 26.5 Robs 2018-07-24 11:10 23 99.3 6.87 36.71 75 Cloudy E 3 26.8 Manta 2018-07-26 8:50 22.8 100.6 6.98 36.73 75 Cloudy S 1 32.3 Chimney 2018-07-27 13:40 23.5 100.7 6.93 36.64 10 Sunny NE 1 32.7 Giants 2018-07-30 13:55 23.6 96.5 6.65 36.57 0 Sunny N 2 32.9 Office 2018-07-31 8:56 22.7 96.5 6.72 36.53 0 Sunny S 1 32.8 Manta 2018-08-01 9:01 22.8 95 6.6 36.44 0 Sunny N 2 33.4 Marble 2018-08-03 14:12 23.3 95.1 6.56 36.55 25 Sunny E 1 39.6 Arches Sherwood 2018-08-04 11:30 23.1 96.8 6.7 36.8 10 Sunny NE 3 27.3 Giants 2018-08-06 12:14 22.7 92.7 6.49 36.56 25 Sunny none 1 22 Office 2018-08-07 9:19 23.2 92.8 6.36 36.77 0 Sunny NE 1 33.3 Marble 2018-08-09 14:20 23.2 94.1 6.5 36.8 100 Cloudy SE 4 33.8 Arches Giants 2018-08-10 12:44 23 92.6 6.45 36.71 50 Sunny NE 4 32.3 (table cont’d.)
101
Table A.2. Subsample filter amounts.
Total (L) Phytoplankton Zooplankton Sample: C:N Pigment DA Count Total C:N DA Count Total Date Location Replicate: A B A B A B (mL) A B (mL) 20-May-18 Sherwood 15 9.6 10 10 9.8 9.8 9.6 145 25 24.5 50 288.5 21-May-18 Robs 27 16.5 14 15 15 15 16 20 142.5 51 50.5 89 290.5 27-May-18 Office 28 15 15 11 10 15.5 16 48 151.5 50 51 28 60 289 28-May-18 Robs 38.4 15 15 11 11 15 15 49 151.5 50 50 30 60 290 30-May-18 Giants 39 15 15 11 10.5 15 15 50.5 152 50 51 22 60 284.5 2-Jun-18 Reggies 32.2 50 50 30 60 290 3-Jun-18 Manta 34.2 15 15 10.5 10 15 15 50 150.5 49 51 32 60 291 6-Jun-18 Salon 34.7 16 15 9 10 16 16 49 151 49 49 31 60 288 11-Jun-18 Reggies 32.2 15 15 10.5 10 25 24 34 152.5 51 49 30 60 290 12-Jun-18 Manta 26.5 15 14.5 10 10.5 24.5 25.5 33 153 49 49 36 60.5 295.5 15-Jun-18 Robs 32.2 15 14.5 10 10.5 25 25 33 153 50 49 30 60 291 17-Jun-18 Salon 32.1 15 16 10 9.5 26 25 33 153.5 50 50 31 60 291 21-Jun-18 Office 28.3 15 15 10 10.5 25 25 32 154.5 50 49 32 60 292.5 22-Jun-18 Marcble Arches 32.3 16 14 10 9 25 26 32 152 50 50 33 60 295 25-Jun-18 Robs 21.2 14.5 16 11 11 25.5 23 32 152.5 50 51 30 60 292 29-Jun-18 Office 26.5 15 15 10 10.5 24.5 26 33 154 50 50 33.5 60 294 2-Jul-18 Giants 26.2 10 10 10.5 9.5 30 30 33 153.5 50 50 35 60 293 9-Jul-18 Manta 27.1 15 15 10 9.5 25.5 24.5 31 151 51 49 32 60 293 10-Jul-18 Manta 32.5 16 16 10.5 10 22 23 32.5 150.5 50.5 50 32 62 294 11-Jul-18 Office 32.5 15 16 10 10 24 26 31 152 50 49 31 60 291 (table cont’d)
102 Total (L) Phytoplankton Zooplankton Sample: C:N Pigment DA Count Total C:N DA Count Total Date Location Replicate: A B A B A B (mL) A B (mL) 14-Jul-18 Marcble Arches 32.4 15 14 9 11 24.5 25 33 151.5 50 50 33 61 295 19-Jul-18 Giants 27.3 15 15 10 10.5 25 25 33 154.5 51 50 33 60 292 20-Jul-18 Reggies 21 14.5 16 10.5 9.5 25 25 33 153.5 50 50 33 61 292 23-Jul-18 Giants 26.4 14 14 9.5 10.5 25 25 33 153 50 50 33 60 292.5 24-Jul-18 Manta 26.5 16 16 10 10 25 21.5 32 152 49.5 52 29 61 292 24-Jul-18 Robs 26.8 15 16 10 10 25 24 33 152 49 49 33 61 292 26-Jul-18 Manta 32.3 15 15 10 10.5 26 26 32 153 50 51 30 61 290.5 27-Jul-18 Chimney 32.7 14 15 10 10 25 26 33 154 50 51 31 61 294 30-Jul-18 Giants 32.9 15 16 10.5 9 25 24 32.5 152 50 50 34 60 293 31-Jul-18 Office 32.8 15 16 9 9 25 27 32 153 50 50 33 61 294 1-Aug-18 Manta 33.4 16 14 10 10 25 26 33 154 50 51 30 61 291 3-Aug-18 Marcble Arches 39.6 15 15 10 10 25 29 32 155 51 50 32 60 293 4-Aug-18 Sherwood 27.3 15 15 10 10 26.5 25.5 32 155 50 50 33 60 293 6-Aug-18 Giants 22 16 16 9.5 9 25 27 33 155 49 50 35 60 295 7-Aug-18 Office 33.3 15 16 10 10 26 27 33 155 49 50 37 60 295 9-Aug-18 Marcble Arches 33.8 15 15 10 9 20 18 33 154.5 49 50 32.5 60 293 10-Aug-18 Giants 32.3 15.5 15 9 10 26.5 30 32 158 49 49 33 59 293 (table cont’d.)
103
Table A.3. Pigment: chlorophyll a starting and output ratios in the CHEMTAX analysis of HPLC pigments. Starting ratios derived from Barlow et al. (2017) and Higgens et al. (2011). Chl a-chlorophyll a; Chl b-chlorophyll b; Chl c1-chlorophyll c1; Chl c2- chlorophyll c2; Chl c3-chlorophyll c3; Per-perodinin; But-butanoyloxyfucoxanthin; Fuc-fucoxanthin; Neo-neoxanthin; Viol- violaxanthin; Pras-prasinoxanthin; Hex-19’-hexanoyloxyfucoxanthin; Allo-alloxanthin; Zea-zeaxanthin; Anth-antheraxanthin; Lut- lutein
Group Chl a Chl b Chl c Chl c 3 Per But Fuc Neo Viol Pras Hex Allo Zea Anth Lut 1+2 Starting Ratios Diatoms-1 1 0 0.022 0 0 0 0.201 0 0 0 0 0 0 0 0 (ex: chaetoceros) Diatoms-2 1 0 0.125 0.062 0 0 0.371 0 0 0 0 0 0 0 0 (Pseudo-nitzschia) Dinoflagellates-1 1 0 0.124 0 0.315 0 0 0 0 0 0 0 0 0 0 Dinoflagellates-2 1 0 0.050 0.196 0 0.033 0.103 0 0 0 0.076 0 0 0 0 Cryptophytes 1 0 0.127 0 0 0 0 0 0 0 0 0.241 0 0 0 Pelagophytes 1 0 0.125 0.024 0 0.292 0.245 0 0 0 0 0 0 0 0 Prasinophytes 1 0.324 0 0 0 0 0 0.037 0.071 0 0 0 0.013 0.012 0.029 Chlorophytes 1 0.194 0 0 0 0 0 0.040 0.030 0 0 0 0.019 0.008 0.103 Cyanobacteria 1 0 0 0 0 0 0 0 0 0 0 0 0.030 0 0 (Trichodesmium) Final Ratio Diatoms-1 1 0 0.017 0 0 0 0.210 0 0 0 0 0 0 0 0 (ex: chaetoceros) Diatoms-2 1 0 0.116 0.049 0 0 0.352 0 0 0 0 0 0 0 0 (Pseudo-nitzschia) Dinoflagellates-1 1 0 0.086 0 0.219 0 0 0 0 0 0 0 0 0 0 Dinoflagellates-2 1 0 0.035 0.146 0 0.023 0.071 0 0 0 0.035 0 0 0 0 Cryptophytes 1 0 0.093 0 0 0 0 0 0 0 0 0.176 0 0 0 Pelagophytes 1 0 0.073 0.014 0 0.187 0.143 0 0 0 0 0 0 0 0 Prasinophytes 1 0.218 0 0 0 0 0 0.025 0.048 0 0 0 0.009 0.008 0.020 Chlorophytes 1 0.139 0 0 0 0 0 0.029 0.021 0 0 0 0.014 0.006 0.074 Cyanobacteria 1 0 0 0 0 0 0 0 0 0 0 0 0.027 0 0 (Trichodesmium)
104
Table A.4. Phytoplankton verification counts (shown as a percent of total community) and listed with pigment proportions derived from CHEMTAX.
Date Type Diatoms PN Dino. Trich. Other 21-May-18 Counts 83.10 14.36 0.05 2.03 0.00 Pigments 52.43 39.41 2.15 5.79 0.23 11-Jun-18 Counts 49.22 3.91 20.20 26.67 0.00 Pigments 89.02 0.00 0.05 10.86 0.07 11-Jul-18 Counts 37.92 7.03 50.12 10.95 0.97 Pigments 7.06 0.00 92.94 0.00 0.00 20-Jul-18 Counts 50.75 19.56 27.38 2.29 0.00 Pigments 49.87 7.71 42.42 0.00 0.00 6-Aug-18 Counts 28.73 62.31 4.75 3.74 0.04 Pigments 0.00 77.70 22.30 0.00 0.00 9-Aug-18 Counts 61.11 25.88 3.21 9.80 0.00 Pigments 31.94 58.57 8.37 0.00 1.12
Table A.5. Identified phytoplankton genera and corresponding group. **indicates dominant species *indicates identification within multiple samples
Identified Genera and Phytoplankton Group *Trichodesmium Cyanobacteria Stephanopyxis Diatom Actinoptychus Diatom *Thalassionema Diatom *Bacillaria Diatom *Thalassiosira Diatom **Bacteriastrum Diatom Triceratium Diatom *Cerataulina Diatom *Akashiwo Dinoflagellate **Chaetoceros Diatom *Alexandrium Dinoflagellate *Coscinodiscus Diatom *Ceratium Dinoflagellate *Cylindrotheca Diatom Dissodinium Dinoflagellate Detonula Diatom Dinophysis Dinoflagellate Eucampia Diatom *Gymnodinium Dinoflagellate *Guinardia Diatom Heterosigma Dinoflagellate *Leptocylidrus Diatom *Noctiluca Dinoflagellate Nanoneis Diatom *Ornithocercus Dinoflagellate Navicula Diatom *Prorocentrum Dinoflagellate *Odontella Diatom *Protoperidinium Dinoflagellate Paralia Diatom Pyrocystis Dinoflagellate Plagiogrammopsis Diatom Pyrophacas Dinoflagellate Pleurosigma Diatom Scrippsiella Dinoflagellate **Pseudo nitzschia Diatom *Chlamydomonas Green algae Pseudosolenia Diatom *Dictyocha Heterokont algae **Rhizosolenia Diatom *Phaeocystis Prymnesiophyte
105
Additional Satellite Imagery
At the beginning of the sampling period, satellite imagery shows warm (>25ºC) surface water temperatures surrounding Inhambane Peninsula (Figure A.1). Temperature gradients reveal a coastal front along the coast north of Inhambane Province, narrowing along Inhambane
Peninsula, then reforming south of the study area. The chlorophyll a (chla) image identifies this frontal zone as an area of increased productivity.
Surface temperature reveals small mesoscale features with decreased surface temperature along the coast (Figure A.2). These small features correlate to pockets of elevated chla within the centers of cool water masses, which is likely where upwelling is occurring. The clockwise rotation of the masses suggest these could be cold-core (cyclonic) eddies along the coast due to wind-driven Ekman transport.
By mid-June, SST has decreased along the coast (<25 °C), yet coastal front seen as a thermal gradient remains strong barrier for productivity (Figure A.3). Wind-driven longshore currents along the coast of Inhambane Province caused decreased sheer along the coastal front and induced upwelling along the shallow shelf which appeared to induce increased productivity on the leeward side of the headland peninsula (Figure A.4). Coastal upwelling by wind stress can be distinguished from terrestrial nutrient input in the thermal image, which would likely be warmer than the surrounding water. However, SST indicates cold water is being upwelled as an eddy forms south of Inhambane Peninsula, as well as a possible counter-current causing northward progression of increased productivity along the coast (Figure A.5). The thermal front along the narrow shelf creates a boundary layer for productivity (Figure A.6), which becomes suppressed when the front hugs the shoreline.
106
Figure A.1. SST and Chla images from 25 May 2018 at 13:15 LST reveal a coastal front within a thermal gradient with coastal production along the shelf of Inhambane Province. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
Figure A.2. SST and Chla images from 1 June 2018 at 13:15 LST reveal the occurrence of mesoscale features along the shelf of Inhambane Province, corresponding to increased productivity. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
107 Figure A.3. SST and Chla images from 17 June 2018 at 13:20 LST show coastal production along the shelf of Inhambane Province. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
Figure A.4. SST and Chla images from 19 June 2018 at 13:05 LST show coastal production along the shelf of Inhambane Province and wind- derived coastal upwelling south of the peninsula. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
108
Figure A.5. SST and Chla images from 20 June 2018 at 13:15 LST show coastal production along the shelf of Inhambane Province and wind- derived coastal upwelling south of the peninsula. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
Figure A.6. SST and Chla images from 24 June 2018 at 13:25 LST show coastal production along the shelf of Inhambane Province. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
109
Figure A.7. SST and Chla images from 1 July 2018 at 13:30 LST reveal the suppression of productivity along Inhambane Province when the thermal front hugs the shoreline. Estimate of chlorophyll a inferred from MODIS-Aqua image produced using SeaDAS with NASA standard algorithms.
110 References
Almandoz GO, Fabro E, Ferrario M, Tillmann U, Cembella A, Krock B (2017) Species occurrence of the potentially toxigenic diatom genus Pseudo-nitzschia and the associated neurotoxin domoic acid in the Argentine Sea. Harmful Algae 63:45-55
Almandoz GO, Ferrario ME, Ferreyra GA, Schloss IR, Esteves JL, Paparazzo FE (2007) The genus Pseudo-nitzschia (Bacillariophyceae) in continental shelf waters of Argentina (Southwestern Atlantic Ocean, 38–55°S). Harmful Algae 6:93-103
Anderson CR, Brzezinski MA, Washburn L, Kudela R (2006) Circulation and environmental conditions during a toxigenic Pseudo-nitzschia australis bloom in the Santa Barbara Channel, California. Marine Ecology Progress Series 327:119-133
Anderson CR, Sapiano MRP, Prasad MBK, Long W, Tango PJ, Brown CW, Murtugudde R (2010) Predicting potentially toxigenic Pseudo-nitzschia blooms in the Chesapeake Bay. Journal of Marine Systems 83:127-140
Anderson CR, Siegel DA, Kudela RM, Brzezinski MA (2009) Empirical models of toxigenic Pseudo-nitzschia blooms: Potential use as a remote detection tool in the Santa Barbara Channel. Harmful Algae 8:478-492
Anderson DM (2002) Harmful algal blooms and eutrophication: nutrient sources, composition and consequences.
Anderson DM (2009) Approaches to monitoring, control and management of harmful algal blooms (HABs). Ocean Coast Manag 52:342
Anderson DM, Burkholder JM, Cochlan WP, Glibert PM, Gobler CJ, Heil CA, Kudela R, Parsons ML, Rensel JE, Townsend DW, Trainer VL, Vargo GA (2008) Harmful algal blooms and eutrophication: Examining linkages from selected coastal regions of the United States. Harmful Algae 8:39-53
Anderson DM, Glibert PM, Burkholder JM (2002) Harmful algal blooms and eutrophication: nutrient sources, composition and consequences. Estuaries 25:704-726
Anderson DM, Hoagland P, Kaoru Y, White AW (2000) Estimated Annual Economic Impacts from Harmful Algal Blooms (HABs) in the United States.
Auro ME, Cochlan WP (2013) Nitrogen Utilization and Toxin Production by Two Diatoms of the Pseudo-nitzschia pseudodelicatissima Complex: P. cuspidata and P. fryxelliana. J Phycol 49:156-169
Backeberg BC, Reason CJC (2010) A connection between the South Equatorial Current north of Madagascar and Mozambique Channel Eddies. Geophys Res Lett 37
111 Bargu S, Baustian MM, Rabalais NN, Del Rio R, Von Korff B, Turner RE (2016) Influence of the Mississippi River on Pseudo-nitzschia spp. Abundance and Toxicity in Louisiana Coastal Waters. Estuaries and Coasts 39:1345-1356
Bargu S, Marinovic B, Mansergh S, Silver MW (2003) Feeding responses of krill to the toxin- producing diatom Pseudo-nitzschia. J Exp Mar Biol Ecol 284:87-104
Bargu S, Powell CL, Coale SL, Busman M, Doucette GJ, Silver MW (2002) Krill: a potential vector for domoic acid in marine food webs. Marine Ecology Progress Series 237:209- 216
Bargu S, Silver M, Goldstein T, Roberts K, Gulland F (2010) Complexity of domoic acid-related sea lion strandings in Monterey Bay, California: foraging patterns, climate events, and toxic blooms. Marine Ecology Progress Series 418:213-222
Bargu S, Silver MW, Ohman MD, Benitez-Nelson CR, Garrison DL (2011a) Mystery behind Hitchcock's birds. Nature Geoscience 5:2-3
Bargu S, Smith E, Ozhan K (2011b) Toxic Diatom Pseudo-nitzschia and Its Primary Consumers (Vectors). The Diatom World
Barlow R, Lamont T, Gibberd MJ, Airs R, Jacobs L, Britz K (2017) Phytoplankton communities and acclimation in a cyclonic eddy in the southwest Indian Ocean. Deep-Sea Res Pt I 124:18-30
Barlow R, Lamont T, Morris T, Sessions H, van den Berg M (2014) Adaptation of phytoplankton communities to mesoscale eddies in the Mozambique Channel. Deep-Sea Res Pt Ii 100:106-118
Barlow R, Stuart V, Lutz V, Sessions H, Sathyendranath S, Platt T, Kyewalyanga M, Clementson L, Fukasawa M, Watanabe S, Devre E (2007) Seasonal pigment patterns of surface phytoplankton in the subtropical southern hemisphere. Deep-Sea Res Pt I 54:1687-1703
Bates SS, Bird CJ, Defreitas ASW, Foxall R, Gilgan M, Hanic LA, Johnson GR, McCulloch AW, Odense P, Pocklington R, Quilliam MA, Sim PG, Smith JC, Rao DVS, Todd ECD, Walter JA, Wright JLC (1989) Pennate Diatom Nitzschia-Pungens As The Primary Source Of Domoic Acid, A Toxin In Shellfish From Eastern Prince Edward Island, Canada. Canadian Journal of Fisheries and Aquatic Sciences 46:1203-1215
Bates SS, Hubbard KA, Lundholm N, Montresor M, Leaw CP (2018) Pseudo-nitzschia, Nitzschia, and domoic acid: New research since 2011. Harmful Algae 79:3-43
Bauer M (2006) Harmful Algal Research and Response: A Human Dimensions Strategy. National Office for Marine Biotoxins and Harmful Algal Blooms, Woods Hole, MA
112 Baustian MM, Bargu S, Morrison W, Sexton C, Rabalais NN (2018) The polychaete, Paraprionospio pinnata, is a likely vector of domoic acid to the benthic food web in the northern Gulf of Mexico. Harmful Algae 79:44-49
Beal LM, Bryden HL (1997) Observations of an Agulhas Undercurrent. Deep Sea Research Part I: Oceanographic Research Papers 44:1715-1724
Behagle N, du Buisson L, Josse E, Lebourges-Dhaussy A, Roudaut G, Menard F (2014) Mesoscale features and micronekton in the Mozambique Channel: An acoustic approach. Deep-Sea Res Pt Ii 100:164-173
Bejarano AC, VanDola FM, Gulland FM, Rowles TK, Schwacke LH (2008) Production and Toxicity of the Marine Biotoxin Domoic Acid and Its Effects on Wildlife: A Review. Human and Ecological Risk Assessment: An International Journal 14:544-567
Berdalet E, Fleming LE, Gowen R, Davidson K, Hess P, Backer LC, Moore SK, Hoagland P, Enevoldsen H (2015) Marine harmful algal blooms, human health and wellbeing: challenges and opportunities in the 21st century. J Mar Biol Assoc UK 2015
Berman FW, Murray TF (1997) Domoic acid neurotoxicity in cultured cerebellargranule neurons is mediated predominantly by NMDA receptors that are acti-vated as a consequence of excitatory amino acid release. Journal of Neurochemistry 69:693-703
Bianchi TS, Bauer JE (2011) Particulate Organic Carbon Cycling and Transformation. Treatise on Estuarine and Coastal Science
Bianchi TS, Filley T, Dria K, Hatcher PG (2004) Temporal variability in sources of dissolved organic carbon in the lower Mississippi river. Geochimica et Cosmochimica Acta 68:959- 967
Blanco J, Livramento F, Rangel IM (2010) Amnesic shellfish poisoning (ASP) toxins in plankton and molluscs from Luanda Bay, Angola. Toxicon 55:541-546
Blondeau-Patissier D, Gower JFR, Dekker AG, Phinn SR, Brando VE (2014) A review of ocean color remote sensing methods and statistical techniques for the detection, mapping and analysis of phytoplankton blooms in coastal and open oceans. Progress in Oceanography 123:123-144
Blum M, Subba Rao DV, Pan Y, Swaminathan S (2006) Development of statistical models for prediction of the neurotoxin domoic acid levels in the pennate diatom Pseudo-nitzschia pungens f. multiseries utilizing data from cultures and natural blooms. In: Rao DVS (ed) Algal Cultures: Analogues of Blooms and Applications, Book 2. Science Publishers, Enfield, NH
Boisier P, Ranaivoson G, Rasolofonirina N, Andriamahefazafy B, Roux J, Chanteau S, Satake M, Yasumoto T (1995) Fatal mass poisoning in Madagascar following ingestion of a shark (Carcharhinus leucas): clinical and epidemiological aspects and isolation of toxins. Toxicon 33:1359-1364
113 Bonachela JA, Raghib M, Levin SA (2011) Dynamic model of flexible phytoplankton nutrient uptake. Proc Natl Acad Sci U S A 108:20633-20638
Bosse KR, Sayers MJ, Shuchman RA, Fahnenstiel GL, Ruberg SA, Fanslow DL, Stuart DG, Johengen TH, Burtner AM (2019) Spatial-temporal variability of in situ cyanobacteria vertical structure in Western Lake Erie: Implications for remote sensing observations. Journal of Great Lakes Research 45:480-489
Bouaïcha N, Chézeau A, Turquet J, Quod JP, Puiseux-Dao S (2001) Morphological and toxicological variability of Prorocentrum lima clones isolated from four locations in the south-west Indian Ocean. Toxicon 39:1195-1202
Boyce DG, Lewis MR, Worm B (2010) Global phytoplankton decline over the past century. Nature 466:591-596
Bradshaw CJA, Fitzpatrick BM, Steinberg CC, Brook BW, Meekan MG (2008) Decline in whale shark size and abundance at Ningaloo Reef over the past decade: The world’s largest fish is getting smaller. Biol Conserv 141:1894-1905
Bristow LA, Mohr W, Ahmerkamp S, Kuypers MMM (2017) Nutrients that limit growth in the ocean. Curr Biol 27:R474-R478
Bryden HL, Beal LM, Duncan LM (2005) Structure and Transport of the Agulhas Current and Its Temporal Variability. Journal of Oceanography 61:479-492
Buitenhuis ET, Li WKW, Vaulot D, Lomas MW, Landry MR, Partensky F, Karl DM, Ulloa O, Campbell L, Jacquet S, Lantoine F, Chavez F, Macias D, Gosselin M, McManus GB (2012) Picophytoplankton biomass distribution in the global ocean. Earth System Science Data 4:37-46
Bužančić M, Ninčević Gladan Ž, Marasović I, Kušpilić G, Grbec B (2016) Eutrophication influence on phytoplankton community composition in three bays on the eastern Adriatic coast. Oceanologia 58:302-316
Cagua EF, Cochran JEM, Rohner CA, Prebble CEM, Sinclair-Taylor TH, Pierce SJ, Berumen ML (2015) Acoustic telemetry reveals cryptic residency of whale sharks. Biol Letters 11
Capietto A, Escalle L, Chavance P, Dubroca L, Delgado de Molina A, Murua H, Floch L, Damiano A, Rowat D, Merigot B (2014) Mortality of marine megafauna induced by fisheries: Insights from the whale shark, the world’s largest fish. Biol Conserv 174:147- 151
Carpenter SR, Caraco NF, Correll DL, Howarth RW, Sharpley AN, Smith VH (1998) Nonpoint Pollution of Surface Waters with Phosphorus and Nitrogen. Ecological Applications 8:559-568
114 Casteleyn G, Chepurnov VA, Leliaert F, Mann DG, Bates SS, Lundholm N, Rhodes L, Sabbe K, Vyverman W (2008) Pseudo-nitzschia pungens (Bacillariophyceae): A cosmopolitan diatom species? Harmful Algae 7:241-257
Cermeño P, Marañón E, Rodríguez J, Fernández E (2005) Large-sized phytoplankton sustain higher carbon-specific photosynthesis than smaller cells in a coastal eutrophic ecosystem. Marine Ecology Progress Series 297:51-60
Chassot E, Bonhommeau S, Dulvy NK, Melin F, Watson R, Gascuel D, Le Pape O (2010) Global marine primary production constrains fisheries catches. Ecology Letters 13:495- 505
Chen Y, Zhang F (2013) Dynamics of a delayed predator–prey model with predator migration. Applied Mathematical Modelling 37:1400-1412
Coale KH, Johnson KS, Fitzwater SE, Gordon RM, Tanner S, Chavez FP, Ferioli L, Sakamoto C, Rogers P, Millero F, Steinberg P, Nightingale P, Cooper D, Cochlan WP, Landry MR, Constantinou J, Rollwagen G, Trasvina A, Kudela R (1996) A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383:495-501
Cochlan WP, Herndon J, Kudela RM (2008) Inorganic and organic nitrogen uptake by the toxigenic diatom Pseudo-nitzschia australis (Bacillariophyceae). Harmful Algae 8:111- 118
Colman JG (1997) A review of the biology and ecology of the whale shark. J Fish Biol 51:1219- 1234
Compton JC (2011) Diatoms: Ecology and Life Cycle. Nova Science Publishers, Inc.
Couturier LI, Marshall AD, Jaine FR, Kashiwagi T, Pierce SJ, Townsend KA, Weeks SJ, Bennett MB, Richardson AJ (2012) Biology, ecology and conservation of the Mobulidae. J Fish Biol 80:1075-1119
Couturier LI, Rohner CA, Richardson AJ, Marshall AD, Jaine FR, Bennett MB, Townsend KA, Weeks SJ, Nichols PD (2013) Stable isotope and signature fatty acid analyses suggest reef manta rays feed on demersal zooplankton. Plos One 8:e77152
Cusack C, Mouriño H, Moita MT, Silke J (2015) Modelling Pseudo-nitzschia events off southwest Ireland. J Sea Res 105:30-41
Cushing DH (1975) Marine ecology and fisheries. Cambridge University Press, New York
Cushing DH (1990) Plankton Production and Year-Class Strength in Fish Populations - an Update of the Match Mismatch Hypothesis. Adv Mar Biol 26:249-293
D'Agostino VC, Degrati M, Sastre V, Santinelli N, Krock B, Krohn T, Dans SL, Hoffmeyer MS (2017) Domoic acid in a marine pelagic food web: Exposure of southern right whales
115 Eubalaena australis to domoic acid on the Peninsula Valdes calving ground, Argentina. Harmful Algae 68:248-257
DAFF (2016) South African Live Molluscan Shellfish Monitoring And Control Programme. Republic of South Africa Department of Agriculture, Forestry and Fisheries, Cape Town
Danger M, Daufresne T, Lucas F, Pissard S, Lacroix G (2008) Does Liebig's law of the minimum scale up from species to communities? Oikos 117:1741-1751
Davidson K, Gowen RJ, Harrison PJ, Fleming LE, Hoagland P, Moschonas G (2014) Anthropogenic nutrients and harmful algae in coastal waters. J Environ Manage 146:206- 216
De Monte S, Soccodato A, Alvain S, d'Ovidio F (2013) Can we detect oceanic biodiversity hotspots from space? Isme J 7:2054-2056
Del Rio R, Bargu S, Baltz D, Fire S, Peterson G, Wang Z (2010) Gulf menhaden (Brevoortia patronus): A potential vector of domoic acid in coastal Louisiana food webs. Harmful Algae 10:19-29
Delegrange A, Lefebvre A, Gohin F, Courcot L, Vincent D (2018) Pseudo-nitzschia sp. diversity and seasonality in the southern North Sea, domoic acid levels and associated phytoplankton communities. Estuarine, Coastal and Shelf Science 214:194-206
Dial R, Roughgarden J (1998) Theory Of Marine Communities: The Intermediate Disturbance Hypothesis. Ecology 79:1412-1424
Diogene J, Reverte L, Rambla-Alegre M, Del Rio V, de la Iglesia P, Campas M, Palacios O, Flores C, Caixach J, Ralijaona C, Razanajatovo I, Pirog A, Magalon H, Arnich N, Turquet J (2017) Identification of ciguatoxins in a shark involved in a fatal food poisoning in the Indian Ocean. Sci Rep 7:8240
Doucette GJ, Mikulski CM, Jones KL, King KL, Greenfield DI, Marin R, Jensen S, Roman B, Elliott CT, Scholin CA (2009) Remote, subsurface detection of the algal toxin domoic acid onboard the Environmental Sample Processor: Assay development and field trials. Harmful Algae 8:880-888
Downes-Tettmar N, Rowland S, Widdicombe C, Woodward M, Llewellyn C (2013) Seasonal variation in Pseudo-nitzschia spp. and domoic acid in the Western English Channel. Continental Shelf Research 53:40-49
Duffy CAJ (2002) Distribution, seasonality, lengths, and feeding behaviour of whale sharks (Rhincodon typus) observed in New Zealand waters. New Zeal J Mar Fresh 36:565-570
Dwernychuk LW, Boag DA (1971) Ducks nesting in association with gulls-an ecological trap? Canadian Journal of Zoology 50:559-563
116 Eckert SA, Stewart BS (2001) Telemetry and satellite tracking of whale sharks, Rhincodon typus, in the Sea of Cortez, Mexico, and the north Pacific Ocean. Environmental Biology of Fishes 60:299-308
EFSA (2009) Scientific Opinion of the Panel on Contaminants in the Food Chain on a request from the European Commission on marine biotoxins in shellfish – domoic acid. In: J. EFSA (ed)
Estrada M, Berdalet E (1997) Phytoplankton in a turbulent world. Scientia Marina 61:125-140
FAO (2005) National Aquaculture Sector Overview. Mozambique., Rome
FAO (2017) The future of food and agriculture – Trends and challenges. Rome
FAO (2018a) The future of food and agriculture – Alternative pathways to 2050. Rome
FAO (2018b) The State of World Fisheries and Aquaculture 2018 - Meeting the sustainable development goals. Rome
Fawcett A, Pitcher GC, Bernard S, Cembella AD, Kudela RM (2007) Contrasting wind patterns and toxigenic phytoplankton in the southern Benguela upwelling system. Marine Ecology Progress Series 348:19-31
Fehling J, Davidson K, Bolch C, Tett P (2006) Seasonality of Pseudo-nitzschia spp. (Bacillariophyceae) in western Scottish waters. Marine Ecology Progress Series 323:91- 105
Fehling J, Davidson K, Bolch CJ, Bates SS (2004) Growth and Domoic Acid Production by Pseudo-nitzschia seriata (Bacillariophyceae) under Phosphate and Silicate Limitation. J Phycol 40:674-683
Fernandes LF, Brandini F (2010) The potentially toxic diatom Pseudo-nitzschia H. Peragallo in the Paraná and Santa Catarina States, Southern Brazil. Ineringia Serie Botanica 65:47-62
Fernandes LF, Hubbard KA, Richlen ML, Smith J, Bates SS, Ehrman J, Leger C, Mafra LL, Jr., Kulis D, Quilliam M, Libera K, McCauley L, Anderson DM (2014) Diversity and toxicity of the diatom Pseudo-nitzschia peragallo in the Gulf of Maine, Northwestern Atlantic Ocean. Deep Sea Res Part 2 Top Stud Oceanogr 103:139-162
Ferriss BE, Marcinek DJ, Ayres D, Borchert J, Lefebvre KA (2017) Acute and chronic dietary exposure to domoic acid in recreational harvesters: A survey of shellfish consumption behavior. Environ Int 101:70-79
Field CB, Behrenfeld MJ, Randerson JT, Falkowski P (1998) Primary production of the biosphere: Integrating terrestrial and oceanic components. Science 281:237-240
Fire SE, Wang Z, Berman M, Langlois GW, Morton SL, Sekula-Wood E, Benitez-Nelson CR (2010) Trophic Transfer of the Harmful Algal Toxin Domoic Acid as a Cause of Death in
117 a Minke Whale (Balaenoptera acutorostrata) Stranding in Southern California. Aquatic Mammals 36:342-350
Fordyce A (2018) Reef fishes of praia do Tofo and praia da Barra, Inhambane, Mozambique. Western Indian Ocean Journal of Marine Science 17:71-91
Fowler SL, Reed TM, Dipper FA (2002) Elasmobranch Biodiversity, Conservation and Management: Proceedings of the International Seminar and Workshop, Sabah, Malaysia. IUCN SSC Shark Specialist Group. IUCN, Gland, Switzerland and Cambridge, UK
Fry B, Wainright SC (1991) Diatom sources of 13C-rich carbon in marine food webs. Marine Ecology Progress Series 76:149-157
Garcia NS, Sexton J, Riggins T, Brown J, Lomas MW, Martiny AC (2018) High Variability in Cellular Stoichiometry of Carbon, Nitrogen, and Phosphorus Within Classes of Marine Eukaryotic Phytoplankton Under Sufficient Nutrient Conditions. Front Microbiol 9:543
Gifford A, Compagno LJV, Levine M (2007a) Aerial survey of whale sharks (Rhincodon typus) off the East Coast of Southern Africa. In: Keesing J, Irvine T (eds) First International Whale shark Conference: promoting international collaboration in whale shark conservation, science and management Conference overview, Abstracts and Supplementary Proceedings CSIRO Marine and Atmospheric research, Australia
Gifford A, Compagno LJV, Levine M, Antoniou A (2007b) Satellite tracking of whale sharks using tethered tags. Fish Res 84:17-24
Gili JM, Coma R (1998) Benthic suspension feeders: their paramount role in littoral marine food webs. Trends Ecol Evol 13:316-321
Glibert P, Seitzinger S, Heil CA, Burkholder JM, Parrow MW, Codispoti LA, Kelly V (2005) The role of eutrophication in the global proliferation of harmful algal blooms. Oceanography 18:198-209
Goldstein T, Mazet JA, Zabka TS, Langlois G, Colegrove KM, Silver M, Bargu S, Van Dolah F, Leighfield T, Conrad PA, Barakos J, Williams DC, Dennison S, Haulena M, Gulland FM (2008) Novel symptomatology and changing epidemiology of domoic acid toxicosis in California sea lions (Zalophus californianus): an increasing risk to marine mammal health. Proc Biol Sci 275:267-276
Graham RT, Roberts CM (2007) Assessing the size, growth rate and structure of a seasonal population of whale sharks (Rhincodon typus Smith 1828) using conventional tagging and photo identification. Fish Res 84:71-80
Graham RT, Witt MJ, Castellanos DW, Remolina F, Maxwell S, Godley BJ, Hawkes LA (2012) Satellite tracking of manta rays highlights challenges to their conservation. Plos One 7:e36834
Graneli E, Turner JT (2006) Ecology of Harmful Algae. Springer
118 Grattan LM, Holobaugh S, Morris JG (2016) Harmful algal blooms and public health. Harmful Algae 57:2-8
Hallegraeff GM (2019) A review of harmful algal blooms and their apparent global increase. Phycologia 32:79-99
Hamisi MI, Mamboya FA (2014) Nutrient and Phytoplankton Dynamics along the Ocean Road Sewage Discharge Channel, Dar es Salaam, Tanzania. Journal of Ecosystems 2014:1-8
Hampson DR, Manalo JL (1998) The activation of glutamate receptors by kainic acid and domoic acid. Nat Toxins 6:153-158
Hanson CE, Pattiaratchi CB, Waite AM (2005) Sporadic upwelling on a downwelling coast: Phytoplankton responses to spatially variable nutrient dynamics off the Gascoyne region of Western Australia. Continental Shelf Research 25:1561-1582
Harethardottir S, Pancic M, Tammilehto A, Krock B, Moller EF, Nielsen TG, Lundholm N (2015) Dangerous Relations in the Arctic Marine Food Web: Interactions between Toxin Producing Pseudo-nitzschia Diatoms and Calanus Copepodites. Mar Drugs 13:3809- 3835
Harris GP (1986) Phytoplankton Ecology: Structure, Function, and Fluctuations. Chapman and Hall Ltd., USA
Haskell PJ, McGowan A, Westling A, Mendez-Jimenez A, Rohner CA, Collins K, Rosero- Caicedo M, Salmond J, Monadjem A, Marshall AD, Pierce SJ (2015) Monitoring the effects of tourism on whale shark Rhincodon typus behaviour in Mozambique. Oryx 49:492-499
Hasle GR (1995) Pseudo-nitzschia pungens and P. multiseries (Bacillariophyceae) - Nomenclatural History, Morphology, and Distribution. J Phycol 31:428-435
Hasle GR (2002) Are most of the domoic acid-producing species of the diatom genus Pseudo- nitzschia cosmopolites? Harmful Algae 1:137-146
Heisler J, Glibert P, Burkholder J, Anderson D, Cochlan W, Dennison W, Gobler C, Dortch Q, Heil C, Humphries E, Lewitus A, Magnien R, Marshall H, Sellner K, Stockwell D, Stoecker D, Suddleson M (2008) Eutrophication and Harmful Algal Blooms: A Scientific Consensus. Harmful Algae 8:3-13
Hermes JC, Masumoto Y, Beal LM, Roxy MK, Vialard J, Andres M, Annamalai H, Behera S, D’Adamo N, Doi T, Feng M, Han W, Hardman-Mountford N, Hendon H, Hood R, Kido S, Lee C, Lee T, Lengaigne M, Li J, Lumpkin R, Navaneeth KN, Milligan B, McPhaden MJ, Ravichandran M, Shinoda T, Singh A, Sloyan B, Strutton PG, Subramanian AC, Thurston S, Tozuka T, Ummenhofer CC, Unnikrishnan AS, Venkatesan R, Wang D, Wiggert J, Yu L, Yu W (2019) A Sustained Ocean Observing System in the Indian Ocean for Climate Related Scientific Knowledge and Societal Needs. Frontiers in Marine Science 6
119 Hessen DO (1992) Nutrient Element Limitation of Zooplankton Production. The American Naturalist 140:799-814
Higgins HW, Wright SW, Schluter L (2011) Quantitative Interpretation of Chemotaxonomic Pigment Data. In: Roy S, Llewellyn CA (eds) Phytoplankton pigments : characterization, chemotaxonomy, and applications in oceanography, Book 257-313. Cambridge University Press, New York
Hiolski EM, Kendrick PS, Frame ER, Myers MS, Bammler TK, Beyer RP, Farin FM, Wilkerson HW, Smith DR, Marcinek DJ, Lefebvre KA (2014) Chronic low-level domoic acid exposure alters gene transcription and impairs mitochondrial function in the CNS. Aquat Toxicol 155:151-159
Hoagland P, Scatasta S (2006) The Economic Effect of Harmful Algal Blooms. In: Graneli E, Turner JT (eds) Ecology of Harmful Algae. Springer
Hodgkiss IJ, Ho KC (1997) Are changes in N:P ratios in coastal waters the key to increased red tide blooms? Hydrobiologia 352:141-147
Hoffmayer ER, Franks JS, Driggers WB, Oswald KJ, Quattro JM (2007) Observations of a Feeding Aggregation of Whale Sharks, Rhincodon typus, in the North Central Gulf of Mexico. Gulf and Caribbean Research 19
Howarth RB (2001) Intertemporal Social Choice and Climate Stabilization. International Journal of Environment and Pollution 15:386-405
Howarth RW (1988) Nutrient Limitation of Net Primary Production in Marine Ecosystems. Annu Rev Ecol Syst 19:89-110
Hsu H-H, Joung S-J, Liao Y-Y, Liu K-M (2007) Satellite tracking of juvenile whale sharks, Rhincodon typus, in the Northwestern Pacific. Fish Res 84:25-31
Hu CM (2009) A novel ocean color index to detect floating algae in the global oceans. Remote Sens Environ 113:2118-2129
Hu CM, Cannizzaro J, Carder KL, Muller-Karger FE, Hardy R (2010a) Remote detection of Trichodesmium blooms in optically complex coastal waters: Examples with MODIS full- spectral data. Remote Sens Environ 114:2048-2058
Hu CM, Lee ZP, Ma RH, Yu K, Li DQ, Shang SL (2010b) Moderate Resolution Imaging Spectroradiometer (MODIS) observations of cyanobacteria blooms in Taihu Lake, China. J Geophys Res-Oceans 115
Hu Z-X, Xu N, Aifen L, Duan S (2008) Effects of different N:P ratios on the growth of Pseudo- nitzschia pungens, Prorocentrum donghaiense and Phaeocystis globosa. Acta Hydrobiologica Sinica 32:482-487
120 Huang Y, Diekmann O (2001) Predator migration in response to prey density: What are the consequences? Journal of Mathematical Biology 43:561-581
Humphries S (2009) Filter feeders and plankton increase particle encounter rates through flow regime control. P Natl Acad Sci USA 106:7882-7887
Husson B, Hernandez-Farinas T, Le Gendre R, Schapira M, Chapelle A (2016) Two decades of Pseudo-nitzschia spp. blooms and king scallop (Pecten maximus) contamination by domoic acid along the French Atlantic and English Channel coasts: Seasonal dynamics, spatial heterogeneity and interannual variability. Harmful Algae 51:26-39
Hutchinson GE (1961) The Paradox of the Plankton. The American Naturalist 95:137-145
Jeffery B, Barlow T, Moizer K, Paul S, Boyle C (2004) Amnesic shellfish poison. Food Chem Toxicol 42:545-557
Jiang R, Wang Y-S (2018) Modeling the ecosystem response to summer coastal upwelling in the northern South China Sea. Oceanologia 60:32-51
Jones RH, Flynn KJ (2005) Nutritional status and diet composition affect the value of diatoms as copepod prey. Science 307:1457-1459
Jury M (1981) Coastal Winds and Upwelling. Transactions of the Royal Society of South Africa 44:299-302
Kenitz K, Williams RG, Sharples J, Selsil Ö, Biktashev VN (2013) The paradox of the plankton: species competition and nutrient feedback sustain phytoplankton diversity. Marine Ecology Progress Series 490:107-119
Kilham P, Hecky RE (1988) Comparative Ecology of Marine and Fresh-Water Phytoplankton. Limnology and Oceanography 33:776-795
Klemas V (2011) Remote Sensing Techniques for Studying Coastal Ecosystems: An Overview. J Coastal Res 27:2-17
Kotut K, Ballot A, Krienitz L (2006) Toxic cyanobacteria and their toxins in standing waters of Kenya: implications for water resource use. Journal of Water and Health 4:233-245
Kvitek RG, Goldberg JD, Smith GJ, Doucette GJ, Silver MW (2008) Domoic acid contamination within eight representative species from the benthic food web of Monterey Bay, California, USA. Marine Ecology Progress Series 367:35-47
Lamont T, Barlow RG, Morris T, van den Berg MA (2014) Characterisation of mesoscale features and phytoplankton variability in the Mozambique Channel. Deep Sea Research Part II: Topical Studies in Oceanography 100:94-105
121 Lamont T, Roberts MJ, Barlow RG, Morris T, van den Berg MA (2010) Circulation patterns in the Delagoa Bight, Mozambique, and the influence of deep ocean eddies. Afr J Mar Sci 32:553-562
Landsberg JH, Flewelling LJ, Naar J (2009) Karenia brevis red tides, brevetoxins in the food web, and impacts on natural resources: Decadal advancements. Harmful Algae 8:598-607
Lane JQ, Raimondi PT, Kudela RM (2009) Development of a logistic regression model for the prediction of toxigenic Pseudo-nitzschia blooms in Monterey Bay, California. Marine Ecology Progress Series 383:37-51
Latasa M (2007) Improving estimations of phytoplankton class abundances using CHEMTAX. Marine Ecology Progress Series 329:13-21
Leandro LF, Teegarden GJ, Roth PB, Wang Z, Doucette GJ (2010) The copepod Calanus finmarchicus: A potential vector for trophic transfer of the marine algal biotoxin, domoic acid. J Exp Mar Biol Ecol 382:88-95
Lefebvre KA, Bargu S, Kieckhefer T, Silver MW (2002) From sanddabs to blue whales: the pervasiveness of domoic acid. Toxicon 40:971-977
Lefebvre KA, Kendrick PS, Ladiges W, Hiolski EM, Ferriss BE, Smith DR, Marcinek DJ (2017) Chronic low-level exposure to the common seafood toxin domoic acid causes cognitive deficits in mice. Harmful Algae 64:20-29
Lefebvre KA, Robertson A (2010) Domoic acid and human exposure risks: a review. Toxicon 56:218-230
Lehto T, Gonçalves R (2008) Mineral Resources Potential In Mozambique. Geological Survey of Finland 48:307-321
Lelong A, Hegaret H, Soudant P, Bates SS (2012) Pseudo-nitzschia (Bacillariophyceae) species, domoic acid and amnesic shellfish poisoning: revisiting previous paradigms. Phycologia 51:168-216
Lema KA, Latimier M, Nezan E, Fauchot J, Le Gac M (2017) Inter and intra-specific growth and domoic acid production in relation to nutrient ratios and concentrations in Pseudo- nitzschia: phosphate an important factor. Harmful Algae 64:11-19
Lincoln JA, Turner JT, Bates SS, Leger C, Gauthier DA (2001) Feeding, egg production, and egg hatching success of the copepods Acartia tonsa and Temora longicornis on diets of the toxic diatom Pseudo-nitzschia multiseries and the non-toxic diatom Pseudo-nitzschia pungens. Hydrobiologia 453:107-120
Litchman E, Klausmeier CA (2008) Trait-Based Community Ecology of Phytoplankton. Annu Rev Ecol Evol S 39:615-639
122 Lugomela C (2013) Population Dynamics Of Pseudo-nitzschia Species (Bacillariophyceae) In The Near Shore Waters Of Dar Es Salaam, Tanzania. Tanzania Journal of Science 39:38- 48
Lugomela C, Lyimo TJ, Bryceson I, Semesi AK, Bergman B (2002) Trichodesmium in coastal waters of Tanzania: diversity, seasonality, nitrogen and carbon fixation. Hydrobiologia 477:1-13
Lugomela C, Pratap HB, Mgaya YD (2006) Cyanobacteria blooms—A possible cause of mass mortality of Lesser Flamingos in Lake Manyara and Lake Big Momela, Tanzania. Harmful Algae 5:534-541
Lund DC, Lynch-Stieglitz J, Curry WB (2006) Gulf Stream density structure and transport during the past millennium. Nature 444:601-604
Lundholm N, Clarke A, Ellegaard M (2010) A 100-year record of changing Pseudo-nitzschia species in a sill-fjord in Denmark related to nitrogen loading and temperature. Harmful Algae 9:449-457
Lundholm N, Krock B, John U, Skov J, Cheng J, Pancic M, Wohlrab S, Rigby K, Nielsen TG, Selander E, Harethardottir S (2018) Induction of domoic acid production in diatoms- Types of grazers and diatoms are important. Harmful Algae 79:64-73
Lutjeharms JRE (2006) The ocean environment off southeastern Africa: a review. S Afr J Sci 102:419-426
Lutjeharms JRE, Stockton PL (1991) Aspects of the upwelling regime between Cape Point and Cape Agulhas, South Africa. South African Journal of Marine Science 10:91-102
Mackas DL, Denman KL, Abbott M (1985) Plankton patchiness: Biology in the physical vernacular. B Mar Sci 37:652-674
MacKenzie L, Beuzenberg V, Holland P, McNabb P, Selwood A (2004) Solid phase adsorption toxin tracking (SPATT): a new monitoring tool that simulates the biotoxin contamination of filter feeding bivalves. Toxicon 44:901-918
Mackey MD, Mackey DJ, Higgens HW, Wright SW (1996) CHEMTAX- a program for estimating class abundances from chemical markers: application to HPLC measurements of phytoplankton. Marine Ecology Progress Series 144:265-283
Madhu NV, Maheswaran PA, Jyothibabu R, Sunil V, Revichandran C, Balasubramanian T, Gopalakrishnan TC, Nair KKC (2002) Enhanced biological production off Chennai triggered by October 1999 super cyclone (Orissa). Current Science 82:1472-1479
Margalef R (1978) Life-forms of phytoplankton as survival alternatives in an unstable environment. Oceanologica Acta 1: 493-509
123 Martin-Jezequel V, Calu G, Candela L, Amzil Z, Jauffrais T, Sechet V, Weigel P (2015) Effects of Organic and Inorganic Nitrogen on the Growth and Production of Domoic Acid by Pseudo-nitzschia multiseries and P. australis (Bacillariophyceae) in Culture. Mar Drugs 13:7067-7086
Matthews M, Bernard S (2015) Eutrophication and cyanobacteria in South Africa’s standing water bodies: A view from space. S Afr J Sci 111
McCabe RM, Hickey BM, Kudela R, Lefebvre K, Adams NG, Bill BD, Fulland FMD, Thomson RE, Cochlan W, Trainer V (2016) An unprecedented coastwide toxic algal bloomlinked to anomalous ocean conditions. Geophys Res Lett 43:366-376
McKibben SM, Peterson W, Wood AM, Trainer VL, Hunter M, White AE (2017) Climatic regulation of the neurotoxin domoic acid. Proc Natl Acad Sci U S A 114:239-244
McKibben SM, Watkins-Brandt KS, Wood AM, Hunter M, Forster Z, Hopkins A, Du X, Eberhart B-T, Peterson WT, White AE (2015) Monitoring Oregon Coastal Harmful Algae: Observations and implications of a harmful algal bloom-monitoring project. Harmful Algae 50:32-44
Méndez SM, Ferrario M, Cefarelli AO (2012) Description of toxigenic species of the genus Pseudo-nitzschia in coastal waters of Uruguay: Morphology and distribution. Harmful Algae 19:53-60
Menezes A, Eide A, Raakjær J (2011a) Moving Out of Poverty: Conditions for Wealth Creation in Small-Scale Fisheries in Mozambique. Poverty Mosaics: Realities and Prospects in Small-Scale Fisheries:407-425
Menezes A, Eide A, Raakjær J (2011b) Moving Out of Poverty: Conditions for Wealth Creation in Small-Scale Fisheries in Mozambique. . Poverty Mosaics: Realities and Prospects in Small-Scale Fisheries
Michalak AM, Anderson EJ, Beletsky D, Boland S, Bosch NS, Bridgeman TB, Chaffin JD, Cho K, Confesor R, Daloglu I, Depinto JV, Evans MA, Fahnenstiel GL, He L, Ho JC, Jenkins L, Johengen TH, Kuo KC, Laporte E, Liu X, McWilliams MR, Moore MR, Posselt DJ, Richards RP, Scavia D, Steiner AL, Verhamme E, Wright DM, Zagorski MA (2013) Record-setting algal bloom in Lake Erie caused by agricultural and meteorological trends consistent with expected future conditions. Proc Natl Acad Sci U S A 110:6448-6452
Morgan KL, Larkin SL, Adams CM (2009) Firm-level economic effects of HABS: A tool for business loss assessment. Harmful Algae 8:212-218
Muscatine L, Porter JW (1977) Reef Corals - Mutualistic Symbioses Adapted to Nutrient-Poor Environments. Bioscience 27:454-460
Nairn AEM, Lerche I, Iliffe JE (1991) Geology, basin analysis, and hydrocarbon potential of Mozambique and the Mozambique Channel. Earth-Science Reviews 30:81-123
124 Negri RM, Montoya NG, Carreto JI, Akselman R, Inza D (2004) Pseudo-nitzschia australis, Mytilus edulis, Engraulis anchoita, and Domoic Acid in the Argentine Sea. In: Reguera B, Blanco J, Fernandes ML, Wyatt T (eds) Harmful Algae. Intergovernmental Oceanographic Commission of UNESCO, Paris
Olofsson M, Karlberg M, Lage S, Ploug H (2017) Phytoplankton community composition and primary production in the tropical tidal ecosystem, Maputo Bay (the Indian Ocean). J Sea Res 125:18-25
Olson MB, Lessard EJ, Cochlan WP, Trainer VL (2008) Intrinsic growth and microzooplankton grazing on toxigenic Pseudo-nitzschia spp. diatoms from the coastal northeast Pacific. Limnology and Oceanography 53:1352-1368
Olson MB, Lessard EJ, Wong CHJ, Bernhardt MJ (2006) Copepod feeding selectivity on microplankton, including the toxigenic diatoms Pseudo-nitzschia spp., in the coastal Pacific Northwest. Marine Ecology Progress Series 326:207-220
Paerl HW (1997) Coastal eutrophication and harmful algal blooms: Importance of atmospheric deposition and groundwater as ''new'' nitrogen and other nutrient sources. Limnology and Oceanography 42:1154-1165
Paerl HW, Paul VJ (2012) Climate change: Links to global expansion of harmful cyanobacteria. Water Res 46:1349-1363
Pan Y, Bates SS, Cembella AD (1998) Evironmental Stress and Domoic Acid Production by Pseudo-nitzschia: a Physiological Perspective. Natural Toxins 6:127-135
Pan YL, Rao DVS, Mann KH (1996) Changes in domoic acid production and cellular chemical composition of the toxigenic diatom Pseudo-nitzschia multiseries under phosphate limitation. J Phycol 32:371-381
Parsons ML, Dortch Q, Turner RE (2002) Sedimentological evidence of an increase in Pseudo- nitzschia (Bacillariophyceae) abundance in response to coastal eutrophication. Limnology and Oceanography 47:551-558
Pauly D, Christensen V (1995) Primary Production Required to Sustain Global Fisheries Nature 374:279-279
Pereira MAM, Litulo C, Santos R, Leal MC, Fernandes RS, Tibiriçá Y, Williams J, Atanassov B, Carreira F, Massingue A, Marques da Silva I (2014) Mozambique Marine Ecosystem Review. In: Costa H (ed) Biodinâmica, Maputo
Pierce SJ, Mendez-Jimenez A, Collins K, Rosero-Caicedo M, Monadjem A (2010) Developing a Code of Conduct for whale shark interactions in Mozambique. Aquat Conserv 20:782- 788
Pierce SJ, Norman B (2016) Rhincodon typus. The IUCN Red List of Threatened Species
125 Pierce SJ, White WT, Marshall AD (2008) New record of the smalleye stingray, Dasyatis microps (Myliobatiformes: Dasyatidae), from the western Indian Ocean. Zootaxa 1734
Pinckney JL, Millie DF, Howe KE, Paerl HW, Hurley JP (1996) Flow scintillation counting of 14C-labeled microalgal photosynthetic pigments. Journal of Plankton Research 18:1867- 1880
Pitcher G, Pillar S (2010) Harmful algal blooms in eastern boundary upwelling systems. Progress in Oceanography 85:1-4
Pitcher GC, Calder D (2000) Harmful algal blooms of the southern Benguela Current: a review and appraisal of monitoring from 1989 to 1997. S Afr J Marine Sci 22:255-271
Pitcher GC, Cembella AD, Krock B, Macey BM, Mansfield L, Probyn TA (2014) Identification of the marine diatom Pseudo-nitzschia multiseries (Bacillariophyceae) as a source of the toxin domoic acid in Algoa Bay, South Africa. Afr J Mar Sci 36:523-528
Pitcher GC, Nelson G (2006) Characteristics of the surface boundary layer important to the development of red tide on the southern Namaqua shelf of the Benguela upwelling system. Limnology and Oceanography 51:2660-2674
Porter A, Bohl D, Kwasi S, Donnenfeld Z, Cilliers J (2017) Prospects and Challenges: Mozambique’s Growth and Human Development Outlook to 2040.
Prebble CEM, Rohner CA, Pierce SJ, Robinson DP, Jaidah MY, Bach SS, Trueman CN (2018) Limited latitudinal ranging of juvenile whale sharks in the Western Indian Ocean suggests the existence of regional management units. Marine Ecology Progress Series 601:167-183
Preen AR, Marsh H, Lawler IR, Prince IT, Shepherd R (1997) Distribution and Abundance of Dugongs, Turtles, Dolphins and other Megafauna in Shark Bay, Ningaloo Reef and Exmouth Gulf, Western Australia. Wildlife Research 24:185-208
Probyn TA (1992) The inorganic nitrogen nutrition of phytoplankton in the southern Benguela: new production, phytoplankton size and implications for pelagic foodwebs. South African Journal of Marine Science 12:411-420
Ptacnik R, Solimini AG, Andersen T, Tamminen T, Brettum P, Lepisto L, Willen E, Rekolainen S (2008) Diversity predicts stability and resource use efficiency in natural phytoplankton communities. P Natl Acad Sci USA 105:5134-5138
Puech B, Batsalle B, Roget P, Turquet J, Quod JP, Allyn J, Idoumbin JP, Chane-Ming J, Villefranque J, Mougin-Damour K, Vandroux D, Gauzere BA (2014) Family tetrodotoxin poisoning in Reunion Island (Southwest Indian Ocean) following the consumption of Lagocephalus sceleratus (Pufferfish). Bull Soc Pathol Exot 107:79-84
126 Putra MIH, Lewis SA, Kurniasih EM, Prabuning D, Faiqoh E (2016) Plankton Biomass Models Based on GIS and Remote Sensing Technique for Predicting Marine Megafauna Hotspots in the Solor Waters. IOP Conference Series: Earth and Environmental Science 47
Quartly GD, Srokosz MA (2004) Eddies in the southern Mozambique Channel. Deep-Sea Res Pt Ii 51:69-83
Quilliam MA, Sim PG, McCulloch AW, McInnes AG (2006) High-Performance Liquid Chromatography of Domoic Acid, a Marine Neurotoxin, with Application to Shellfish and Plankton. International Journal of Environmental Analytical Chemistry 36:139-154
Radan RL, Cochlan WP (2018) Differential toxin response of Pseudo-nitzschia multiseries as a function of nitrogen speciation in batch and continuous cultures, and during a natural assemblage experiment. Harmful Algae 73:12-29
Raj RP, Peter BN, Pushpadas D (2010) Oceanic and atmospheric influences on the variability of phytoplankton bloom in the Southwestern Indian Ocean. Journal of Marine Systems 82:217-229
Ratti JT, Reese KP (1988) Preliminary Test of the Ecological Trap Hypothesis. The Journal of Wildlife Management 52:448-491
Raven JA (1998) The twelfth Tansley Lecture. Small is beautiful: the picophytoplankton. Functional Ecology 12:503-513
Redfield AC (1958) Biological Control of Chemical Factors in the Environment. American Scientist 46:204-221
Reynolds CS (2006) Ecology of Phytoplankton. Cambridge University Press
Rhodes L, Jiang W, Knight B, Adamson J, Smith K, Langi V, Edgar M (2013) The genus Pseudo-nitzschia (Bacillariophyceae) in New Zealand: analysis of the last decade's monitoring data. New Zeal J Mar Fresh 47:490-503
Rhodes L, Scholin C, Garthwaite I (1998) Pseudo-nitzschia in New Zealand and the role of DNA probes and immunoassays in refining marine biotoxin monitoring programmes. Nat Toxins 6:105-111
Ritzman J, Brodbeck A, Brostrom S, McGrew S, Dreyer S, Klinger T, Moore SK (2018) Economic and sociocultural impacts of fisheries closures in two fishing-dependent communities following the massive 2015 U.S. West Coast harmful algal bloom. Harmful Algae 80:35-45
Rohner CA, Armstrong AJ, Pierce SJ, Prebble CE, Cagua EF, Cochran JE, Berumen ML, Richardson AJ (2015) Whale sharks target dense prey patches of sergestid shrimp off Tanzania. J Plankton Res 37:352-362
127 Rohner CA, Couturier LIE, Richardson AJ, Pierce SJ, Prebble CEM, Gibbons MJ, Nichols PD (2013a) Diet of whale sharks Rhincodon typus inferred from stomach content and signature fatty acid analyses. Marine Ecology Progress Series 493:219-235
Rohner CA, Pierce SJ, Marshall AD, Weeks SJ, Bennett MB, Richardson AJ (2013b) Trends in sightings and environmental influences on a coastal aggregation of manta rays and whale sharks. Marine Ecology Progress Series 482:153-168
Rohner CA, Richardson AJ, Jaine FRA, Bennett MB, Weeks SJ, Cliff G, Robinson DP, Reeve- Arnold KE, Pierce SJ (2018) Satellite tagging highlights the importance of productive Mozambican coastal waters to the ecology and conservation of whale sharks. PeerJ 6:e4161
Rowat D, Gore M, Meekan MG, Lawler IR, Bradshaw CJA (2009a) Aerial survey as a tool to estimate whale shark abundance trends. J Exp Mar Biol Ecol 368:1-8
Rowat D, Speed CW, Meekan MG, Gore MA, Bradshaw CJA (2009b) Population abundance and apparent survival of the Vulnerable whale shark Rhincodon typus in the Seychelles aggregation. Oryx 43:591-598
Runge JA (1988) Should we expect a relationship between primary production and fisheries? The role of copepod dynamics as a filter of trophic variability. Hydrobiologia 167:61-71
Sá C, Leal MC, Silva A, Nordez S, André E, Paula J, Brotas V (2013) Variation of phytoplankton assemblages along the Mozambique coast as revealed by HPLC and microscopy. J Sea Res 79:1-11
Saeed AF, Awan SA, Ling S, Wang R, Wang S (2017) Domoic acid: Attributes, exposure risks, innovative detection techniques and therapeutics. Algal Research 24:97-110
Saji NH, Goswami BN, Vinayachandran PN, Yamagata T (1999) A dipole mode in the tropical Indian Ocean. Nature 401:360–363
Schouten MW, de Ruijter WPM, van Leeuwen PJ, Ridderinkhof H (2003) Eddies and variability in the Mozambique Channel. Deep Sea Research Part II: Topical Studies in Oceanography 50:1987-2003
Schultz IR, Skillman A, Woodruff D (2008) Domoic acid excretion in dungeness crabs, razor clams and mussels. Mar Environ Res 66:21-23
Schumann EH, Perrins L-A, Hunter IT (1982) Upwelling along the South Coast of the Cape Province, South Africa. S Afr J Sci 78:238-242
Seeyave S, Probyn TA, Pitcher GC, Lucas MI, Purdie DA (2009) Nitrogen nutrition in assemblages dominated by Pseudo-nitzschia spp., Alexandrium catenella and Dinophysis acuminata off the west coast of South Africa. Marine Ecology Progress Series 379:91- 107
128 Sekula-Wood E, Benitez-Nelson C, Morton S, Anderson C, Burrell C, Thunell R (2011) Pseudo- nitzschia and domoic acid fluxes in Santa Barbara Basin (CA) from 1993 to 2008. Harmful Algae 10:567-575
Sequeira AMM, Mellin C, Delean S, Meekan MG, Bradshaw CJA (2013) Spatial and temporal predictions of inter-decadal trends in Indian Ocean whale sharks. Marine Ecology Progress Series 478:185-195
Shanks AL, Morgan SG, MacMahan J, Reniers AJ, Kudela R, Jarvis M, Brown J, Fujimura A, Ziccarelli L, Griesemer C (2016) Variation in the abundance of Pseudo-nitzschia and domoic acid with surf zone type. Harmful Algae 55:172-178
Sidharta R (2005) The Current Status Of Research On Harmful Algal Bloom (HAB) In Indonesia. Journal of Coastal Development 8:75-86
Sigman D, Hain M (2012) The Biological Productivity of the Ocean. Nature Education 3
Simon N, Cras AL, Foulon E, Lemee R (2009) Diversity and evolution of marine phytoplankton. Cr Biol 332:159-170
Sleeman JC, Meekan MG, Wilson SG, Jenner CKS, Jenner MN, Boggs GS, Steinberg CC, Bradshaw CJA (2007) Biophysical correlates of relative abundances of marine megafauna at Ningaloo Reef, Western Australia. Mar Freshwater Res 58:608-623
Smayda TJ (1990) Novel and nuisance phytoplankton blooms in the sea: evidence for a global epidemic. In: Graneli E, Sundstorm B, Edler I, Anderson DM (eds) Toxic Marine Phytoplankton. Elsevier, New York
Smayda TJ (2008) Complexity in the eutrophication-harmful algal bloom relationship, with comment on the importance of grazing. Harmful Algae 8:140-151
Smith J, Connell P, Evans RH, Gellene AG, Howard MDA, Jones BH, Kaveggia S, Palmer L, Schnetzer A, Seegers BN, Seubert EL, Tatters AO, Caron DA (2018) A decade and a half of Pseudo-nitzschia spp. and domoic acid along the coast of southern California. Harmful Algae 79:87-104
Speed CW, Meekan MG, Rowat D, Pierce SJ, Marshall AD, Bradshaw CJA (2008) Scarring patterns and relative mortality rates of Indian Ocean whale sharks. J Fish Biol 72:1488- 1503
Sterner RW, Schultz KL (1998) Zooplankton nutrition: recent progress and a reality check. Aquatic Ecology 32:261-279
Stevens JD (2007) Whale shark (Rhincodon typus) biology and ecology: A review of the primary literature. Fish Res 84:4-9
Swart NC, Lutjeharms JRE, Ridderinkhof H, de Ruijter WPM (2010) Observed characteristics of Mozambique Channel eddies. Journal of Geophysical Research 115
129 Tamele IJ, Silva M, Vasconcelos V (2019) The Incidence of Marine Toxins and the Associated Seafood Poisoning Episodes in the African Countries of the Indian Ocean and the Red Sea. Toxins (Basel) 11
Tammilehto A, Nielsen TG, Krock B, Moller EF, Lundholm N (2015) Induction of domoic acid production in the toxic diatom Pseudo-nitzschia seriata by Calanoid copepods. Aquat Toxicol 159:52-61
Tammilehto A, Nielsen TG, Krock B, Møller EF, Lundholm N (2012) Calanus spp.—Vectors for the biotoxin, domoic acid, in the Arctic marine ecosystem? Harmful Algae 20:165- 174
Taylor JG, Pearce AF (1999) Ningaloo Reef currents: implications for coral spawn dispersal, zooplankton and whale shark abundance. Journal of the Royal Society of Western Australia 82:57-65
Ternon JF, Bach P, Barlow R, Huggett J, Jaquemet S, Marsac F, Ménard F, Penven P, Potier M, Roberts MJ (2014) The Mozambique Channel: From physics to upper trophic levels. Deep Sea Research Part II: Topical Studies in Oceanography 100:1-9
Terseleer N, Gypens N, Lancelot C (2013) Factors controlling the production of domoic acid by Pseudo-nitzschia (Bacillariophyceae): A model study. Harmful Algae 24:45-53
Tew Kai E, Marsac F (2010) Influence of mesoscale eddies on spatial structuring of top predators’ communities in the Mozambique Channel. Progress in Oceanography 86:214- 223
Tew-Kai E, Marsac F (2009) Patterns of variability of sea surface chlorophyll in the Mozambique Channel: A quantitative approach. Journal of Marine Systems 77:77-88
Theberge MM, Dearden P (2006) Detecting a decline in whale shark Rhincodon typus sightings in the Andaman Sea, Thailand, using ecotourist operator-collected data. Oryx 40:337-342
Thessen AE (2007) Taxonomy and Ecophysiology of Pseudo-nitzschia in the Chesapeake BAY. Doctor of Philosophy Dissertation, University of Maryland, Digital Repository at the University of Maryland
Thessen AE, Bowers HA, Stoecker DK (2009) Intra- and interspecies differences in growth and toxicity of Pseudo-nitzschia while using different nitrogen sources. Harmful Algae 8:792-810
Thessen AE, Soniat TM, Dortch Q, Doucette GJ (2010) Crassostrea virginica grazing on toxic and non-toxic diatoms. Toxicon 55:570-579
Thorel M, Claquin P, Schapira M, Le Gendre R, Riou P, Goux D, Le Roy B, Raimbault V, Deton-Cabanillas AF, Bazin P, Kientz-Bouchart V, Fauchot J (2017) Nutrient ratios influence variability in Pseudo-nitzschia species diversity and particulate domoic acid production in the Bay of Seine (France). Harmful Algae 68:192-205
130 Tibiriçá Y, Birtles A, Valentine P, Miller DK (2011) Diving Tourism in Mozambique: An Opportunity at Risk? Tourism in Marine Environments 7:141-151
Tilman D, Kilham SS, Kilham P (1982) Phytoplankton Community Ecology - the Role of Limiting Nutrients. Annu Rev Ecol Syst 13:349-372
Tilstone GH, Lotliker AA, Miller PI, Ashraf PM, Kumar TS, Suresh T, Ragavan BR, Menon HB (2013) Assessment of MODIS-Aqua chlorophyll-a algorithms in coastal and shelf waters of the eastern Arabian Sea. Continental Shelf Research 65:14-26
Trainer V, Eberhart B-T, Wekell JC, Adams NG, Hanson L, Cox F, Dowell J (2003) Paralytic shellfish toxins in Puget Sound, Washington. Journal of Shellfish Research 22:213-233
Trainer V, Hickey BM, Bates SS (2008) Toxic Diatoms. In: Walsh P, Smith S, Fleming LE, Solo-Gabriele H, Gerwick W (eds) Oceans and Human Health. Academic Press
Trainer VL, Adams NG, Bill BD, Stehr CM, Wekell JC, Moeller P, Busman M, Woodruff D (2000) Domoic acid production near California coastal upwelling zones, June 1998. Limnology and Oceanography 45:1818-1833
Trainer VL, Bates SS, Lundholm N, Thessen AE, Cochlan WP, Adams NG, Trick CG (2012) Pseudo-nitzschia physiological ecology, phylogeny, toxicity, monitoring and impacts on ecosystem health. Harmful Algae 14:271-300
Trainer VL, Hickey BM, Homer RA (2002) Biological and physical dynamics of domoic acid production off the Washington coast. Limnology and Oceanography 47:1438-1446
Trainer VL, Moore SK, Hallegraeff G, Kudela RM, Clement A, Mardones JI, Cochlan WP (2019) Pelagic harmful algal blooms and climate change: Lessons from nature’s experiments with extremes. Harmful Algae
Trainer VL, Pitcher GC, Reguera B, Smayda TJ (2010) The distribution and impacts of harmful algal bloom species in eastern boundary upwelling systems. Progress in Oceanography 85:33-52
Trainer VL, Wells ML, Cochlan WP, Trick CG, Bill BD, Baugh KA, Beall BF, Herndon J, Lundholmf N (2009) An ecological study of a massive bloom of toxigenic Pseudo- nitzschia cuspidata off the Washington State coast. Limnology and Oceanography 54:1461-1474
Turner JT (2004) The importance of small planktonic copepods and their roles in pelagic marine food webs. Zoological Studies 43:255-266
Turner RE, Dortch Q, Justic D, Swenson EM (2002) Nitrogen loading into an urban estuary: Lake Pontchartrain (Louisiana, U.S.A.). Hydrobiologia 487:137-152
Turner RE, Rabalais NN (1991) Changes in Mississippi River Water Quality This Century. Bioscience 41:140-147
131 Usman MT, Reason CJC (2004) Dry spell frequencies and their variability over southern Africa. Climate Research 26:199-211
Utermöhl H (1958) Toward the improvement of the quantitative phytoplankton method. Mitteilungen International Verieningung fur Limnologie 9:1-39
Uz BM (2007) What causes the sporadic phytoplankton bloom southeast of Madagascar? Journal of Geophysical Research 112 van der Molen JS, Scharler UM, Muir D (2016) Species composition, abundance and biomass of microphytoplankton in the KwaZulu-Natal Bight on the east coast of South Africa. Afr J Mar Sci 38:S139-S153
Van Dolah FM (2000) Marine Algal Toxins: Origins, Health Effects, and Their Increased Occurrence. Environ Health Persp 108:133-141
Van Heukelem L, Lewitus AJ, Kana TM, Craft NE (1994) Improved separations of phytoplankton pigments using temperature-controlled high performance liquid chromatography. Marine Ecology Progress Series 114:303-313
Van Meerssche E, Greenfield DI, Pinckney JL (2018) Coastal eutrophication and freshening: Impacts on Pseudo-nitzschia abundance and domoic acid allelopathy. Estuarine, Coastal and Shelf Science 209:70-79
Van Meerssche E, Pinckney JL (2017) The influence of salinity in the domoic acid effect on estuarine phytoplankton communities. Harmful Algae 69:65-74
Vanni MJ (1987) Effects of Nutrients and Zooplankton Size on the Structure of a Phytoplankton Community. Ecology 68:624-635
Varela R, Lima FP, Seabra R, Meneghesso C, Gomez-Gesteira M (2018) Coastal warming and wind-driven upwelling: A global analysis. Sci Total Environ 639:1501-1511
Venables S, Winstanley G, Bowles L, Marshall AD (2016) A Giant Opportunity: the economic impact of manta rays on the Mozambican tourism industry - an incentive for increased management and protection. Tourism in Marine Environments 12:51-68
Veschasit O, Meksumpun S, Thawonsode N, Lirdwitayaprasit T (2017) Accumulation of domoic acid in marine organisms from Sriracha bay, Chonburi province, Thailand. ScienceAsia 43
Walker ND (1986) Satellite-Observations of the Agulhas Current and Episodic Upwelling South of Africa. Deep-Sea Res 33:1083-1106
Wells ML, Trainer VL, Smayda TJ, Karlson BS, Trick CG, Kudela RM, Ishikawa A, Bernard S, Wulff A, Anderson DM, Cochlan WP (2015) Harmful algal blooms and climate change: Learning from the past and present to forecast the future. Harmful Algae 49:68-93
132 Wells ML, Trick CG, Cochlan W, Hughes MP, Trainer V (2005) Domoic acid: The synergy of iron, copper, and the toxicity of diatoms. Limnology and Oceanography 50:1908–1917
Williams JL, Pierce SJ, Rohner CA, Fuentes MMPB, Hamann M (2017) Spatial Distribution and Residency of Green and Loggerhead Sea Turtles Using Coastal Reef Habitats in Southern Mozambique. Frontiers in Marine Science 3
Willis C, Papathanasopoulou E, Russel D, Artioli Y (2018) Harmful algal blooms: the impacts on cultural ecosystem services and human well-being in a case study setting, Cornwall, UK. Mar Policy 97:232-238
Wilson SG, Taylor JG, Pearce AF (2001) The seasonal aggregation of whale sharks at Ningaloo Reef, Western Australia: currents, migrations and the El Niño/Southern Oscillation. Environmental Biology of Fishes 61:1-11
Wohlgeschaffen GD, Mann KH, Subba Rao DV, Pocklington R (1992) Dynamics of the phycotoxin domoic acid: accumulation and excretion in two commercially important bivalves. J Appl Phycol 4:297-310
Work TM, Barr B, Beale AM, Fritz L, Quilliam MA, Wright JLC (1993) Epidemiology of Domoic Acid Poisoning in Brown Pelicans (Pelecanus occidentalis) and Brandt's Cormorants (Phalacrocorax penicillatus) in California. Journal of Zoo and Wildlife Medicine 24:54-62
World Bank (2018a) Communities livelihoods fisheries : fisheries governance and shared growth in Mozambique. Washington D. C.
World Bank (2018b) Strong but not Broadly Shared Growth—Mozambique Poverty Assessment. Poverty and Equity Global Practice Africa Region
World Bank IFC (2006) The Tourism Sector in Mozambique: A Value Chain Analysis, Vol 1. World Bank Group International Finance Corporation
Wright JLC, Boyd RK, Defreitas ASW, Falk M, Foxall RA, Jamieson WD, Laycock MV, Mcculloch AW, Mcinnes AG, Odense P, Pathak VP, Quilliam MA, Ragan MA, Sim PG, Thibault P, Walter JA, Gilgan M, Richard DJA, Dewar D (1989) Identification of Domoic Acid, a Neuroexcitatory Amino-Acid, in Toxic Mussels from Eastern Prince- Edward-Island. Canadian Journal of Chemistry-Revue Canadienne De Chimie 67:481- 490
Wyatt ASJ, Lowe RJ, Humphries S, Waite AM (2013) Particulate nutrient fluxes over a fringing coral reef: Source-sink dynamics inferred from carbon to nitrogen ratios and stable isotopes. Limnology and Oceanography 58:409-427
Yager T (2015) The Mineral Industry of Mozambique. Minerals Yearbook. USGS
Zabaglo (2016) Environmental roles and biological activity of domoic acid: A review.
133 Zhang YC, Ma RH, Duan HT, Loiselle SA, Xu JD, Ma MX (2014) A Novel Algorithm to Estimate Algal Bloom Coverage to Subpixel Resolution in Lake Taihu. Ieee J-Stars 7:3060-3068
Zhou Z, Tao Y, Li S, Ding W (2013) Hydrocarbon potential in the key basins in the East Coast of Africa. Petroleum Exploration and Development 40:582-591
134 Vita
Holly Ann Kelchner was born and raised in Minnesota. Frequent family visits to the Outer
Banks, NC inspired her love for the ocean at a young age. She received her Bachelor of Science degree in Marine Biology from the University of Alaska Southeast in May 2017. While studying at the Juneau campus, Holly participated in various research topics such as seaweed mariculture, shrimp endocrinology, marine debris, and ecotourism impacts on marine mammals in Southeast
Alaska. Following graduation, she joined the Minnesota Department of Natural Resources as an
Interpretive Naturalist at Mystery Cave State Park. Holly began graduate school at Louisiana
State University and Agricultural and Mechanical College in the fall of 2017 to further develop research skills and techniques studying coastal plankton dynamics, specifically in harmful algal bloom ecology. In June 2019, she began applying the laboratory techniques learned during her master’s research analyzing cyanobacteria toxin samples from the Lake Erie as a Research
Technician for the Cooperative Institute for Great Lakes Research (CIGLR). She expects to graduate in May 2020 and intends to continue working for CIGLR following graduation and continue to develop her career in the marine sciences research.
135