Reproductive biology and recruitment of Xiphopenaeus kroyeri in Suriname: Implications for resource management

Pieter Torrez – Thesis Supervisor: T.Willems Promotor: Prof.Dr.M.Vincx Co-Promotors: Dr.K.Hostens & Dr A.De Backer

June 2015

Order: Decapoda Famly: Penaeidae Thesis Species: Xiphopenaeus kroyeri Common name: seabob Distribution: On the continental shelf Infographic from North Carolina (USA) to Santa Catarina(Brazil) Recruitment of the seabob shrimp in Suriname Suriname MSc fishery VMS (Vessel Monitoring By Satellite) Harvest Control Rule (HCR) Limited Licenses

CodeTED (ofTurtle Practice Excluder Device) TED (Turtle Excluder Device) BRD (Bycatch Reduction Device) Brazil

I Designated Seabob trawl zone N Mangrove s + Artisanal zone = nursery ground s T

R VMS Florida Double Twin rig Demersal Trawl O = D 500 U 20 Side C I I O O T Mysis Nauplius zoae Ocean Top I Egg O Postlarv e Juvenile N Estuary Adult & O B J  Examine spatio-temporal patterns of the early stages of E the seabob shrimp along an estuary-offshore gradient C T  Relate environmental variables to the abundance of Methods: I seabob shrimp V 4 sampling stations (I = Inshore, O = Offshore) E  Draw conclusions from the results that relate to the 8 campaigns (every 6 weeks, Feb 2014-Feb 2015) S management of the seabob fishery in Suriname + 30min otter trawl Juveniles, Adults R 5 min hyperbenthic trawl Postlarvae Secchi Depth, Turbidity E Results: Average Postlarvae Density 2014 - 2015

S Salinity, Temperature, Environmental

U ) Grain Size, Mud content Variables ² L Continious “Background”recruitment T S  2 recruitment events inshore  Bottom water temperature peaked during main recruitment event  inshore stations were characterized by a lower salinity, a higher temperature, lower clarity, higher turbidity  Postlarvae and small juveniles were more Postlarvae Density (count/m Density Postlarvae present in the shallow inshore zone, large Wet season Time Dry season juveniles and adults in the deep offshore zone To protect the reproductive females of the seabob during the main recruitment event a closed fishing season would be applicable between June - August Ghent University

Faculty of Sciences

Department of Biology

Academic Year 2014-2015

Reproductive biology and recruitment of Xiphopenaeus kroyeri in Suriname: Implications for resource management

Pieter Torrez

Supervisor: T.Willems

Co-promotors: Dr.K.Hostens & Dr.A.De Backer

Promotor: Prof.Dr.M.Vincx

Master thesis submitted for the partial fulfillment of the title of

Master of Science in Marine Biodiversity and Conservation

Within the ERASMUS MUNDUS Master Programme EMBC

Preface

This thesis fits in the framework of EMBC, an International Master of Science in Marine Biodiversity and Conservation. The two-year master program is offered by a university consortium consisting of 6 university partners and a wide global network of associate members coordinated by Ghent University, Belgium. The intention of this thesis fits in the PhD research of Tomas Willems: "Towards a sustainable fisheries management on the seabob shrimp along the coastal zone of Suriname". In 2011 a PhD scholarship was granted to Tomas Willems (ILVO, Flemish Institute for Agriculture and Fisheries Research), to conduct research on a MSC-granted (2011) tropical shrimp fishery in Suriname, South America. The main targets of his research are i.a. to identify the ecological role of the seabob in the marine foodweb and to assess the secondary impacts of the fishery on the ecosystem. The results from the research will be translated into scientifc advice to adjust the management of the seabob fishery ensuring its sustainability and economic profitability on the long-term (ILVO 2015).

This thesis elaborates on the reproductive biology and recruitment of the seabob on a spatio-temporal scale along an estuary-ocean gradient and the environmental factors influencing the abundance of the different life stages through time and space. Eight sampling campaigns were implemented from February 2014 until February 2015. I joined the sampling campaign at the end of August where I spent 5 days on the seabob trawler Cobia to get used to the sampling techniques used in the marine field in a developing country, which is not always a matter of course. From February 2015 until May 2015, hyperbenthic samples and onboard gathered data were analyzed at the lab in ILVO, Ostend, Belgium. The structure of the thesis follows the structure of a scientific paper, the introduction covers 3 sections; the first section gives a brief introduction to ecosystem based fisheries management and the status of global shrimp fisheries these days, the second part highlights the distribution and catch of seabob in general and the management strategy of the seabob fishery in Suriname. The closing part of the introduction covers the objectives of the thesis. The methods and materials section comprehends the used samples techniques to obtain the seabob and environmental data throughout the year and the used statistics to identify the patterns. The methods and materials part is followed by the results. The discussion compares the obtained results with recruitment studies previously done in Brazil and gives recommendations for management decisions and further research, encouraging the sustainability of the fishery.

There are a lot of people I would like to acknowledge for their support during the progress of my study and the master thesis. First of all I would like to thank my parents for all the years of study that preceded this master thesis. They supported me both financially and mentally. I also would like to thank Tomas, for guiding me in Suriname and the assistance with the elaboration of the samples in the lab, the data analysis and the discussion of the obtained results. I would also like to thank Dr.K.Hostens and Dr.A.De Backer for reading and adjusting my thesis and I would like to acknowledge Prof.Dr.M.Vincx to be the promotor of this thesis and guiding Tomas through his PhD research. At last I would like to express my gratitude for friends, family, the EMBC family, professors from Ghent University and GMIT, ILVO, the people I met in Suriname (Fred, I hope you can build your eco-lodge in of the most pristine and fascinating places in the heart of the Surinam’ rainforest) and the crew of the seabob trawler Cobia and Neptune 6 (yes, you Steve, I’ll never forget the chicken feet).

CONTENTS

ABSTRACT ...... 1 EXECUTIVE SUMMARY ...... 2 1. INTRODUCTION ...... 3 2. METHODS AND MATERIALS ...... 7

2.1 STUDY AREA ...... 7 2.2 SAMPLING, SAMPLING DESIGN AND DATA ORIGIN ...... 8 2.2.1 Abiotic parameters ...... 8 2.2.2 Life stages of X.kroyeri ...... 9 2.3 DATA ANALYSIS...... 9 2.3.1 Abiotic parameters ...... 9 2.3.2 Life stages of X.kroyeri ...... 10 2.3.3 Statistics ...... 10 2.4 VON BERTALANFFY GROWTH MODEL (VBGM) & SIZE AT FIRST MATURITY ...... 11 3. RESULTS ...... 12

3.1 ABIOTIC PARAMETERS ...... 12 3.1.1 Salinity and temperature ...... 12 3.1.2 Secchi and suspended matter...... 13 3.1.3 Sediment analysis ...... 14 3.2 LIFE STAGES OF X.KROYERI ...... 15 3.2.1 Densities of postlarvae, juveniles and adults ...... 15 3.2.2 Cohort analysis ...... 17 3.3 THE LIFE STAGES OF X.KROYERI AND THE ENVIRONMENT ...... 18 3.3.1 Postlarvae and the abiotic parameters ...... 18 3.3.2 CCA ...... 19 3.4 VON BERTALANFFY GROWTH MODEL (VBGM) & SIZE AT FIRST MATURITY ...... 19 4. DISCUSSION ...... 20 5. CONCLUSION ...... 23 6. APPENDIX ...... 23 7. R-CODE ...... 33 8. REFERENCES ...... 35 Abstract

The Atlantic seabob shrimp Xiphopenaeus kroyeri occurs abundantly in the shallow coastal waters of Suriname, where it is an important natural fishing resource for both industrial and artisanal fisheries. Studies on the species are very limited on the northern coast of South America, although information on the biology of exploited species is crucial for a sustainable and ecosystem-based fisheries management. In 2014, a year-round sampling was conducted to reveal the life cycle dynamics of X.kroyeri on a spatio-temporal scale along an estuary-ocean gradient. Hyper – and epibenthic samples were taken to sample the postlarval, juvenile and adult stages of X. kroyeri. Five abiotic parameters were recorded: temperature, salinity, Secchi depth, suspended matter and sediment. Results have shown that recruitment occurred continuously in the in- and offshore zone. A main and a smaller recruitment event (increased postlarvae density) occurred respectively in July and October in the inshore zone. The bottom temperature peaked during the main recruitment event in the in- and offshore zone. Postlarvae and small juveniles were more present in the shallow inshore zone, large juveniles and adults in the deep offshore zone. The sustainability of the fishery could be enhanced by a closed fishing season from June through August to protect the reproductive females during the main recruitment event.

Keywords: Atlantic seabob shrimp, Xiphopenaeus kroyeri, sustainable fishery, MSC, shrimp, recruitment, postlarvae, spatio-temporal, Suriname, South America

Postlarve of Xiphopenaeus kroyeri © Hans Hillewaert Adult of Xiphopenaeus kroyeri © Tomas Willems

Research was performed at the University of Surinam (Anton de Kom) and the Flemish Institute for Agriculture and Fisheries Research (ILVO ) in Ostend, Belgium.

Torrez P., Willems T., Backer D. A. ,Hostens K. ,Vincx M. 2015. Reproductive biology and recruitment of Xiphopenaeus kroyeri in Suriname: Implications for resource management. Master thesis, 1-39.

No data can be taken out of this work without prior approval of the thesis-promoter

I hereby confirm that I have independently composed this Master thesis and that no other than the indicated aid and sources have been used. This work has not been presented to any other examination board.

05 June 2015 Signature

1

Executive Summary

The Atlantic seabob shrimp Xiphopenaeus kroyeri occurs abundantly in the shallow coastal waters of Suriname, where it is an important fishing resource for both industrial and artisanal fisheries. Studies on the species are very limited on the northern coast of South America, although information on the biology of exploited species is crucial for a sustainable and ecosystem-based fisheries management. In 2014, a year-round sampling (8 campaigns, every 6 weeks, Feb 2014 – Feb 2015) was conducted to reveal the recruitment of X.kroyeri on a spatio-temporal scale along an estuary-ocean gradient with 4 stations. The objectives of this thesis were to: 1) examine spatio-temporal patterns of the early life stages of X. kroyeri along an estuary-ocean gradient, 2) relate environmental variables to the abundance of the early life stages of X. kroyeri , 3) draw conclusions from the results that relate to the management of the seabob fishery in Suriname. Scientific samples were taken with the seabob trawlers Neptune 6 and Cobia. Adult and juvenile X. kroyeri were caught by dragging the onboard try- net for 30 minutes across the bottom. All X. kroyeri individuals were onboard sorted by sex and stage of maturity, counted and measured with a caliper (1 mm). Thereafter a hyperbenthic trawl was dragged for ca. 5 min. right along the hyperbenthic zone of the water column to catch the postlarvae of X. kroyeri. Five environmental variables were measured at each sampling event: temperature and salinity (CTD, no data from July), suspended matter (Niskin bottle), visibility (Secchi disk), sediment composition (Van Veen grab, median grain size & mud content). The life stages of X. kroyeri were expressed in densities (m-²) and subjected to a length-frequency cohort analysis to identify the periods of recruitment. Results have shown that recruitment occurred continuously in the in- and offshore zone. A main and a smaller recruitment event (increased postlarvae density) occurred respectively in July and October in the inshore zone. The bottom temperature peaked during the main recruitment event in the in- and offshore zone. The inshore stations were characterized by a lower salinity, a higher temperature, lower clarity, higher amount of suspended matter then the offshore stations. Postlarvae and small juveniles were more present in the shallow inshore zone, large juveniles and adults in the deeper offshore zone. The sustainability of the fishery could be enhanced by a closed fishing season from June through August to protect the reproductive females during the main recruitment event.

The dynamic coast of Suriname (The Terra/MODIS image by NASA, Naipal 2010).

2

1. Introduction

Fisheries and aquaculture represent one of the largest sources of proteins, and provide a livelihood for more than 140 million people across the globe (FAO 2014, Garcia & Cochrane 2005). With an annual catch of 3,4 million tons, wild caught shrimp contribute significantly to the global fisheries catch (4-5%). At present, the worldwide shrimp production (fisheries and aquaculture combined) attains unseen levels of 7-8 million tons a year (FAO 2014). Although tropical shrimp fisheries in developing countries often form the most valuable fishery export product, they are notorious for their high bycatch and discard levels (Davies et al. 2009, Gillet 2008, ICES/FAO 2005). Recent assessments estimated that tropical shrimp fisheries generate 27% (1,86 million tons) of the total annual global bycatch (Gillet 2008, ICES/FAO 2005). The tropical shrimp industry grew promptly from the 1960s onwards; nowadays the number of trawlers has decreased due to the shift to aquaculture (FAO 2014, Gillet 2008). However fishing effort remains high and overfishing is fragrant; especially in tropical areas were overcapacity of the fleet predominates (Gillet 2008). Lately, scientific research has aimed its attention to the reduction of the bycatch levels and physical impacts of the trawling gear (Broadhurst 2000, Eayrs 2005, Gillet 2008, Macfadyen et al. 2013, Southall et al. 2011, Sudhakara 2011). Tropical shrimp fisheries these days are characterized by low profitability due to rising fuel costs and competition with farmed shrimp (Globefish 2014 Highlights, Gillet 2008, Hempeel et al. 2002, Szuster 2006, appendix 1.1).

Marine ecosystems are next to overfishing also affected by an increasing stress of non-fishing hazards like eutrophication, offshore mining, oil and gas extraction and human settlement along the coasts. Now that it becomes clear that also remote areas like the deep sea and polar seas are becoming affected by human activity, a new strategy has been developed to exploit marine resources in a more sustainable way: Ecosystem Based Fishery Management (EBFM) (Cook et al. 2014, Levin and Möllmann 2015, Garcia & Cochrane 2005, Voss et al. 2014, UNEP 2014, Zhou et al. 2010). EFBM has been proposed as a more effective and holistic approach for managing fisheries in a sustainable way with the focus on reducing habitat destruction, incidental mortality of non-target species (bycatch) and changes in the structure and functioning of ecosystems (Cook et al. 2014, Garcia & Cochrane 2005, Levin & Möllmann 2015, Zhou et al. 2010). It acts on a precautionary approach reducing irreversible impacts like regime shifts until more scientific and local ecological knowledge is known (Cook et al. 2014, Garcia & Cochrane 2005, Levin & Möllmann 2015, Zhou et al. 2010). By understanding the complexity, interaction and the variability of ecosystems on a spatio- temporal scale awareness can be raised to promote an economic comprehension of maintaining healthy ecosystems and exploiting them in a sustainable way to assure long-term socio-economic benefits (Cook et al. 2014, Garcia & Cochrane 2005, Levin & Möllmann 2015, Voss et al. 2014, Zhou et al. 2010). Finally, due to the complexity of marine ecosystems, we need to identify and prioritize the key components and these should be the focus of the management actions (Cook et al. 2014, Garcia & Cochrane, Levin & Möllmann 2015, Voss et al. 2014, Zhou et al. 2010). In spite of all the efforts that have been made to implement ecosystem approaches in fisheries management there has been a decelerated shift from single species management towards EFBM (Cook et al. 2014, Levin and Möllmann 2015, Voss et al. 2014).

According to Tavares (2002) shrimps from the family penaeidae contribute to 95% of the total commercial shrimp production in the Western Central Atlantic. One of the commercial penaeid species found in this area is the seabob shrimp (Xiphopenaeus kroyeri). This shrimp is found along the continental shelf from North Carolina (USA) to Santa Catarina (Brazil) (Almeida et al. 2012, Fig. 1). The species was first described in 1862 by Heller as Penaeus kroyeri and renamed to Xiphopenaeus kroyeri in 1869 by Sidney Irving Smith (Holthuis 1980). This species is of commercially importance in USA, Mexico, Trinidad and Tobago, Guyana, Suriname, French Guyana and Brazil (Fig. 1).

3

In the last 5 decades there has been a significant increase in catch from 6000 t in the 1960s to 52007 t in 2012 (Almeida et al. 2012). In North Brazil fishing for X.kroyeri became popular as stocks of Penaeus subtilis, P. brasiliensis and P. paulensis were decreasing. The largest fishery for X.kroyeri shrimp operates in Guyana (24 833 t in 2012) where the species is suggested to be overexploited (FAO FISHSTAT 2015). Due to its economic importance, research on its biology, ecology and reproduction has steadily increased over the last years (Almeida et al. 2012, Branco 2005, Castro et al. 2005, Campos et al. 2009, Heckler et al. 2014, Santos et al. 2013).

Fig.1: Distribution (darkgrey) and catch of Xiphopenaus kroyeri (red) in 2012. Zero values were observed for French Guiana and Trinidad & Tobago. X.kroyeri is only fished by small artisanal fisheries in these countries (data from FAO FISHSTAT 2015). The seabob fishery in Suriname began developing in the mid-1990s and it is currently harvested by 20 ‘Florida’ double twin rig demersal shrimp trawlers owned by 2 companies: Heiploeg Suriname and SAIL. The number of licenses decreased 30% since 2008. The trawlers operate in a designated trawl zone (Fig. 2, 18-35 m, FAO Statistical area 31) to assure the protection of the artisanal fisheries and the nursery zones (LVV 2012, Southall et al. 2011). The artisanal fishery contains 500 vessels with Chinese seines aiming for seabob (Southall et al. 2011). Beyond 35 m depth heavier trawling gear is used to target finfish or Penaeus sp. The ‘Florida’ double twin rig demersal trawl has 4 otter boards, a low net opening (2 m), a mesh size of 57 mm in body and wings, 45 mm in the cod end and no rock hopper bobbins. The shrimps live in soft sediment burrows in the bottom and light tickler chains on the trawl net stimulate seabob to flee their burrows (Almeida et al. 2012, Castro et al. 2005, Freire et al. 2011, Heckler et al. 2014, Simoes et al. 2010). Fig. 2: EEZ of Suriname and the zonation (Southall et al. 2011).

4

Seabob fishing trips normally last for 6-8 days. Each time 2 days are spent to reach and depart from the designed trawl zone, 2 days between two trips are devoted to rest. The boat is mostly filled with around 10 t of seabob after approximately 40 hauls. A trawler has an average of 30 trips a year with a maximum of 220 days. Shrimps are processed in two factories, where they are sorted, cleaned, peeled, frozen and exported to the USA and Europe (Southall et al. 2011). The fishery is managed by the harvest control rule (HCR) that aims at maintaining the biomass 10% higher than the MSY to ensure the sustainability of the stock. During the stock assessment in 2009 a logistic surplus production model was implemented using catch data from 1989-2008 and effort data from 1998- 2008. The results indicated that the Suriname seabob is not overfished, with a B/BMSY (current biomass relative to biomass at maximum sustainable yield) value of 1,22 and a median value F/FMSY (current fishing mortality relative to fishing mortality) of 0,82. Reference points for the fishery are set to be B/BMSY = 1,10 (biomass at 10% higher than MSY level) and B/BMSY =0,60 (biomass at 60% of level required to produce MSY). These reference levels assure the precautionary approach of the fishery, taking into account recruitment failure and the ecological role of X.kroyeri until more research is done. In the Suriname seabob fishery CPUE (catch per unit of effort) is used as a primary management tool, it is measured on a daily basis and monthly reported to the authorities. The threshold for the CPUE is 1,48 t/day, which has so far not been surpassed. The CPUE level has steadily declined over the last 10 years, but the B/BMSY levels remains above the reference point. If CPUE falls below the threshold, the allowed catch will be adjusted to TAD = (Current Catch Rate – Limit Catch Rate) * 8625; 0,89 t/day is the ultimate brink to close the fishery, at this CPUE rate; the stock biomass would be 60% of the MSY (LVV 2012, MSC fishery Fact Sheet 2012, Southall et al. 2011). The bycatch of the fishery consists mainly of demersal roundfish, flatfish, elasmobranchs, brackish water fish and invertebrates. An assessment on the average composition of the hauls was estimated on: 69% seabob, 19% retained valuable species and 12% discard (Southall et al. 2011). Vulnerable turtles like the leatherback turtle (Dermochelys coriacea) are regulary seen in Suriname waters as these species use the beaches of the Guianas as nesting sites. They use the coastal waters of Suriname to move between the breeding and feeding grounds (IUCN 2015, LVV 2012, Reichart & Fretey 1993, Southall et al. 2011, Turtle Expert Working Group 2007). The fishery invested severely in improving their sustainability. Their efforts in applying BRD’s (Bycatch Reduction Device), TED’s (Turtle Excluder Device), limited licenses, VMS (Vessel Monitoring by Satellite), stock assessment and the harvest control rule (HCR) granted them the first ecolabel for tropical shrimp fisheries by the Marine Stewardship Council (MSC) in 2011. Though the recognition of improved sustainability of the fishery does not imply that the fishery is optimal managed, further improvements of the management and research of the area are necessary to enhance the single- species management to a multispecies ecosystem-based fishery management (ILVO 2015, Southall et al. 2011).

The study of recruitment has been a consistent focus of research in fisheries science (Rice and Browman 2014). According to Allen and Hightower (2010) recruitment can be defined as the number of fish/shrimp that survive to a specific age or size in a given year. It is hereby important to indicate at which life stage and size we can consider the recruits; this appears at a particular age and/or size when very high level naturally mortality rates have diminished. For shrimps, maturity is estimated based on the length of the carapace (Almeida et al. 2012, Castro et al. 2005, Fernandes et al. 2011). An important aspect from recruitment is to detect the variability among population size and age structure (ratio of large shrimp, reproductive individuals,..). This variation in recruitment in and among years is a necessary consideration when evaluating different harvest policies. Among the different kinds of overfishing recruitment overfishing can be defined as a steep reduction in the spawning stock to a level failure in recruitment is drastically reducing population size. A further failure in recruitment will eventually cause the fishery to collapse (Allen & Hightower 2010, Gillet 2008, Pauly et al. 1984). Xiphopeneaus kroyeri occurs abundantly around depths of 27 m (Almeida et al. 2012). The life cycle of this shrimp is a 12-stage cycle where eggs hatch and form a nauplius which develops after 6 moults and 50 hours in a zoea. The zoea undergoes 3 moults in 4-6 days before transformation to the mysis stage, which in its turn converts to a postlarve after 3 mysis moults and approximate 10-12 days (Colvin & Brand 1977, Gillet 2008). Spawning of penaeids occurs offshore and the early stages move inshore (Tavares 2002). Subsequently the postlarvae settle on the benthic 5

substrate and grow to a total length of 25 mm where they are considered to be juveniles (Cook 1964). Juveniles move gradually towards deeper water as they grow. When juveniles reach sexual maturity they couple and the females lay eggs (Colvin & Brand 1977, Gillet 2008). In contrast to other penaeids the seabob is not relying on estuaries for its development (Castro et al. 2005).

Data from stock assessment of the Suriname seabob stock is incomplete, according to Medley (2009) seabob populations are healthy and not overfished but ecological knowledge is lacking. However the assessment of the separation of the two seabob stocks in Guyana and Suriname was one of the conditions to acquire the MSC label. During the CRFM 2009 scientific meeting morphometric data, spawning and recruitment patterns were examined. Results have shown slight differences in observed mean size data and recruitment patterns but data limitations and inconsistencies in data could not guarantee the full insulation of the seabob stock in Suriname. Therefore the 2009 assessment approved the precautionary approach of treating the stocks from Guyana and Suriname separately for assessment purposes and to protect local genetic differences until further scientific research is done (Southall et al. 2011). Three years later, in 2011 no further progress has been made on the issue of stock assessment (CRFM 2011). Xiphopenaeus kroyeri has shown to be a species with high rates of natural mortality with a rapid growth, and development which is resilient to heavy fishing (Lopes et al. 2014), nevertheless the processes that regulate juvenile growth and the variability in juvenile survival (natural mortality, influence of artisanal communities on survival and mangrove destruction for coastal development) need to be properly understood for effective management and decision-making (Baker et al. 2014). The aim of the current study was to reveal the periods of recruitment and the spatio-temporal distribution of the different life stages of X.kroyeri and their environmental preferences. Postlarvae have been found in the estuary of the Suriname river before (Kerkhove et al. 2014) but no research has been done to link the variability of the environmental parameters through time and space to the distribution of the different life stages. Therefore an estuary-ocean transect with 4 stations was sampled for one year through 8 campaigns. This ecological knowledge could contribute to the implementation of the ecosystem based fisheries management to the fishery and guide fishery officers in defining a closed fishing season during large recruitment events to ensure the long-term profit of the resource. Hereto, the following research objectives were identified:

o Examine spatio-temporal patterns of the early life stages of X.kroyeri along an estuary-ocean gradient. o Relate environmental variables to the abundance of X.kroyeri o Draw conclusions from the results that relate to the management of the seabob fishery in Suriname

6

2. Methods And Materials

2.1 Study area

Suriname is a former Dutch colony situated at the northeastern coast of South-America bordering Brazil in the south and Guyana and French Guyana in the west and east respectively. Suriname is situated close to the equator (2° and 6° N, 54° and 58° W) and subjected to two dry and - rainy seasons (Government of Suriname 2001, Sakimin 2009). From December until the end of January there is a short rainy season, followed by a small dry season until the end of April, at the end of April the long rainy season starts until mid-August, a long dry season fulfills the annual cycle. The climate is characterized by a high annual relatively constant temperature (22-32°) with a lot of sun hours. The average rainfall is high and occurs under the form of heavy rain showers (FAO 2008, Government of Suriname 2014, World Weather and Climate Information 2015).

Along the 380 km long coastline of Suriname four rivers disembogue in the Atlantic Ocean: the Maroni river (at the border with French Guyana), the Suriname river, the Coppename river and the Corantijn river (at the border with Guyana). The coastal zone consists mainly of pristine mangrove forests that form hibernating grounds for migratory shorebirds from North America and there are also nesting sites for turtles in the East (Galibi, Matapica)(FAO 2008, Sakhimin 2009, Winterwerp 2009).The coastline of the Guianas (Guyana, Suriname and French Guiana) is bordered by the Guiana continental shelf and influenced by muddy waters originating from the largest river in the world: the Amazon. This river releases around 173,000 m3 freshwater per second and around 1 billion tons of sediment a year which consists mainly out of silt and clay (Kuehl et al. 1985, Wells and Coleman 1981, Nittrouer et al. 1986, Anthony et al. 2010). This huge sediment supply is transported northwest by the Guiana Current that is formed as a consequence of the bifurification of the North Equatorial current off the South American coast and the subsequent retroflection of the South Equatorial Current in the north-westward direction (Fig. 3). About half of the sediment remains in the water column due to turbulence Fig. 3: The Guiana Current (GC=guyana current; NBC=north of waves and tidal currents. The other half brazil current; NECC=north equatorial counter current). precipitates under the form of migrating mud banks that are characterized by bank - and interbank phases (Nittrouer et al. 1986). According to Wells and Coleman (1981) the mud banks in Suriname become indistinct at 20-40 km from the coastline at depths of about 20 m. The annual outflow of the Amazon is defined by seasonality with peak discharge between January and April correlated with the annual peak in trade winds and wave activity affecting the Amapa-Guyana’s coast (Anthony et al. 2010, Anthony et al. 2013, Froiefond et al. 1988, Hu et al. 2011, Park and Latrubesse 2014). The coastal zone of Suriname can be divided in three zones based on a color difference of the surface water, a 'brown', 'green' and 'blue' zone can be observed (see NASA image executive summary). Samples were taken in the 'brown' and 'green' zone (appendix 2.1).

7

2.2 Sampling, sampling design and data origin TK24 TK17 To study the spatio-temporal patterns of the different TK10 life stages of X. kroyeri, four stations (TK3-10-17-24) were sampled along an inshore-offshore gradient in TK03 front of the Suriname river on 8 sampling occasions during one year (Feb 2014 till Feb 2015 approximately every 6 weeks, Fig. 4). Depth increased gradually when moving further offshore from resp. 6, 8, 18 to 27 m. Each station was sampled for abiotic parameters and Fig. 4: Map of the coastal zone of Suriname with the 4 for the life stages of X. kroyeri. Scientific samples were sampled stations, Kerkhove et al. (2014) taken with the seabob trawlers Neptune 6 and Cobia. Each sampling campaign lasted for 5-6 days, of which one day was spent each time to do the research. The sampling process followed a consistent method where data and remarks were recorded on an onboard form (appendix 2.3).

Seabob (Xiphopenaus kroyeri) 2.2.1 Abiotic parameters

At each sampling event, five environmental parameters were measured per station. Water salinity (0,01 psu), temperature (0,001 °C) and depth (m) were measured (CTD; SAIV SD200) at all occasions, except for July due to a technical error. Water clarity was measured with a Secchi-disk. Sub-surface total suspended matter Fig. 5: From left to right top: hyberbenthic trawl, CTD deployment, Van Veen grab, concentrations (SS-TMS, g/ml) from left to right bottom: Niskin bottle, Secchi disk, pregnant X.kroyeri females were assessed from water with olive green ovaries. collected at 5 m depth with a Niskin bottle. Onboard, a subsample was filtered on pre-washed, pre-weighted GF/F filters and stored at -20°C. Filters were subsequently dried in the lab (48h at 70°C) and re-weighted (0,0001 g) to calculate SS-TSM. A Van Veen Grab (0,1 m²) was used to collect sediment at every station. A sediment subsample was dried in the lab (48 h at 70 °C), and analyzed for grain size composition (Malvern Mastersizer 2000G hydro version 5.40, Malvern 1999). Median grain size, median grain size of the sand fraction (63-2000 µm; MEDSAND) and the percentage of mud (<63 µm; MUD) were used as granulometric parameters.

8

2.2.2 Life stages of X.kroyeri

Adult and juvenile X.kroyeri were sampled by dragging a small otter trawl (2,6 m width; the ‘try-net’) with a mesh size of 45 mm for 30 minutes across the bottom at a speed of 2,5 knots. All shrimps were onboard sorted by sex and stage of maturity, counted and measured. Sexes were distinguished based on the presence of petasma in males and thelycum in females. In males, the development of the petasma was used to separate juveniles from adults while in females the color of the abdominal cavity (olive green in mature females) was used to determine maturity (Almeida et al. 2012, Oliveira 1991). The carapace length was measured to the nearest 1 mm using a caliper.

To collect the postlarval stages of X.kroyeri, a hyperbenthic trawl (mesh size 1 mm) was deployed for ca. 5 min at a speed of 2,5 knots. The device had a lower and upper net, sampling the water column between 20-50 cm and 50-100 cm above the seabed respectively (appendix 2.4). Hyperbenthic samples were washed out of the net in a bucket and sieved to a smaller sample. Samples were stored in 4% formaldehyde onboard, which was replaced in the lab by 70% ethanol after at least 48h fixation, hyperbenthic samples were examined under a stereomicroscope and all postlarvae and larger stages (juvenile and adult) of X.kroyeri were sorted out and counted. The first life stages (nauplius, zoea and mysis) were not identifiable and could therefore not be sorted from the samples. The identification of the postlarvae was based on morphological characteristics described by Cook (1964) and Oliveira & LHomme (1993), i.e. the presence of an anteriorly extension of the rostrum between the eyes and 5 pairs of functional pleopods (appendix 2.5 & 2.6). Further distinction between the postlarvae of the two most common species Xiphopenaeus kroyeri and Penaeus subtilis was made, based on size difference characteristics (Oliveira & LHomme 1993, Cook 1964) such as the length of the 6th abdominal segment, the pleopods and the length of 4th and 5th pair of pereopods. A distinction between juveniles and postlarvae was made based on Cook’s (1964) total standard length of 25 mm (measured from the tip of the rostrum until the tip of the telson). Postlarvae of X.kroyeri were counted and measured indirectly by drawing the dorsal contour of the body with a pencil and an integrated binocular mirror. Images were scanned and measured with ImageJ (a public domain Java-based program from the National Institutes of Health, version 1.6.0_20) till the nearest 0,01 cm. Lengths of the contours were rescaled by using a predefined and drawn line of known length (1 cm).

2.3 Data analysis

2.3.1 Abiotic parameters

Data from the CTD device was extracted in the software Minisoft SD200W (Morten Hammersland Programvare, version 3.8.5.122). Three values of each upcast measurement were selected: the surface (0,5 m depth), 5 m (depth of Niskin bottle deployment) and bottom water values of salinity and temperature. The 5m-values seemed to be highly correlated with the surface values and were therefore omitted for further data analysis (rs,Sal= 0,957; rs,Temp= 0,911, Spearman). The sediment consisted mainly out of mud (<63 µm), for this reason data of the median grain size of the sand fraction was omitted for further analysis.

9

2.3.2 Life stages of X.kroyeri

Numbers of postlarvae, juveniles and adults were expressed as individuals per m² by dividing the count of individuals by the trawled area (m-², appendix 2.7). Postlarvae from the lower and upper net were summed. The equation TL = (CL + 0,0931)/0,2174 (Lopes et al. 2014) with TL as the total length and CL as the length of the carapace was used to calculate the total length from the carapace length. Length-frequency tables were drafted for postlarvae, juveniles and adults with size classes of 1 mm and processed in FISAT II for cohort analysis to identify recruitment events. The Bhattacharya (1967) method was applied to split the composite length-frequency distributions into separate normal distributions (Sparre & Venema 1998). The normal components (mean, theoretical number per group, standard deviation) of frequency size distributions were identified for each date with an adequate number of individuals. The means and standard deviations of the normal distributions for all sampling dates were then linked to trace the modal length progression of the cohorts by applying the software NORMSEP also available in FISAT II (Pauly et al. 1984, Pauly et al. 1987, Van Hoey et al. 2006).

2.3.3 Statistics

Visualization of the obtained data was done in Rstudio (R Development Core Team 2014, version 0.98.1062), and the R package ggplot2 (Wickham 2015, version 1.0.1). A Loess smoother was used to fluently connect the eight sampled environmental data points by station on a real-time x-axis (date of sampling). The method uses locally weighted lineair regression to smooth data and was used to better illustrate trends in the changing abiotic parameters (Cleveland 1979, Cleveland & Devlin 1988, appendix 2.8). Missing CTD data from July was estimated based on this Loess smoother. Means and standard deviations were calculated for the abiotic parameters. Correlations between the abiotic parameters were calculated using the ‘Spearman rank’ correlation. Annual changes in the densities of the life stages were visualized by connecting the density data points by straight lines and by station on a real-time x-axis (date of sampling). To examine the spatio-temporal patterns of the abiotic parameters and the life stages of X.kroyeri, months and stations were combined to the factors ‘season’ and ‘shore’ respectively. The months February 2014 until August 2014 were considered to be the wet season, while the months August 2014 till February 2015 were considered to be the dry season (World Weather and Climate Information 2015). The stations TK03 and TK10, who occurred in the ‘brown’ zone, had similar depths and were defined as the inshore stations, while TK17 and TK24 occurred in the ‘green’ zone and were defined as the offshore stations. Means and standard deviations of the abiotic parameters and the life stages of X.kroyeri were calculated for each factor or the interaction between the factors (Dry-Wet, Inshore-Offshore). Relative densities of the life stages of X.kroyeri were calculated for each factor (Dry-Wet, Inshore-Offshore). Statistical tests were done in PRIMER with the PERMANOVA add-on (Clarke et al. 2006, Version 6.1.16). Two- factor Permanova with factors ‘season’ and ‘shore’ was performed to test for differences in environmental parameters and the life stages of X.kroyeri. When a significant ‘season x shore’ interaction was found, pairwise tests within ‘season’ and within ‘shore’ were performed. Correlations between the density of postlarvae and the abiotic parameters were calculated using the ‘Spearman rank’ correlation and visualized by straight line graphs by the factor shore (average values of the stations). A CCA (canonical correspondence analysis, Almeida et al. 2012, Braak & Verdonshot 1995, McGarigal et al. 2000) was used to explain the relationships between the biological distribution of all the life stages of X.kroyeri and the abiotic parameters. The adonis function from the vegan package (Oksanen et al. 2015, Version 2.2-1) was used to find significant environmental variables. A CCA plot was drawn with the environmental parameters with the lowest p-values explaining the distribution of the life stages the best (package ggvegan, Simpson 2013, version 0.0-1.). 10

2.4 Von Bertalanffy growth model (VBGM) & size at first maturity

The parameters from a Von Bertalanffy growth model (VBGM) and size at first maturity were also estimated. Several studies along the Brazilian coast have focused on the estimation of these growth and reproductive parameters (Almeida et al. 2012, Branco et al. 1999, Branco 2005, Castro et al. 2005, Coelho & Santos 1993, Fernandes et al. 2011, Nakagaki & Negreiros-Fransozo 1998). Determining growth parameters and size at first maturity can be useful in estimating the sustainability of the fishery and indicate if overexploitation of the resource occurs; even if recruits are protected (overfishing will lead to maturity at a smaller size). A Von Bertalanffy (1938) growth model is a common used tool in applying growth of relatively long-living species like fish but the equation has been used in the growth estimation of short-living penaieds as well (Lewis and Fontoura 2005, Oh et al. 1999). The formula of the growth function can be ascribed as: TL(t) = TL∞(1 − exp−k(t−t0)), where TL(t) is the total length (mm) in time t, TL∞ is the asymptotic length (mm), k is the growth coefficient (growth/year), and t0 is the hypothesized age at zero length. The theoretical parameter t0 was drawn from the equation by using the length at age zero as the start of growth on 2 mm (Oh et al. 1999). The asymptotic length and the growth coefficient were estimated in the program FISAT II (Version 1.2.2, Gayanilo et al. 2005) by applying the ELEFAN I (Electronic Length- Frequency Analysis) method and estimating the best combination of parameters based on the highest Rn (Enin 1995, Pauly et al. 1981, Pauly et al. 1984). This was done for male and female X.kroyeri of the epibenthic net separately and together to examine sexual growth differences. Data from the hyperbenthic samples was not used as small juveniles were underrepresented and the program could not handle a bimodal length-frequency distribution. Longevity was estimated based on the time for reaching 95% of the maximum average length (Lewis and Fontoura 2005, Taylor 1962). Size at first maturity for females could not be estimated as females were only considered to be mature onboard if the ovaries were green. To estimate size at first maturity (CL50), proportions of mature to juvenile mature males were considered for each size class (1 mm) by fitting the logistical model y = 1 / (1 + e[–r (CL – CL50)]), where y is the estimated proportion of adult shrimps, CL50 is the carapace size at the onset of sexual maturity, and r is the coefficient for the slope of the logistic curve (Almeida et al. 2012, Campos et al. 2009, Fernandes et al. 2011).

11

3. Results

3.1 Abiotic parameters

3.1.1 Salinity and temperature

Wet Dry Wet Dry S

Fig. 6-7: Surface and bottom salinity and temperature over one year (2014-2015) with the inshore stations (red and green) and the offshore stations (blue and yellow). A loess smoother was applied to the sampled data points who were scaled on a real-time x-axis.

Salinity of the surface and bottom water varied respectively around 34,4 ± 1,9 psu and 35,2 ± 1,9 psu. Temperature of the surface and bottom water varied respectively around 27,4 ± 0,8 °C and 26,3 ± 1,5 °C. The salinity of both surface and bottom water attained a maximum in November (resp. 36,2 and 36,5 psu) and a minimum in February-March (resp. 28,7 and 29,7 psu) (Fig. 6-7). Two-factor Permanova detected a significant ‘season’ x ‘shore’ interaction for both surface and bottom water salinity (resp. Pseudo-F = 7,4039; p <0,01; Fig. 8 and Pseudo-F = 24,983; p <0,001; Fig. 9; appendix 3.1). Pairwise tests showed that in both the wet and the dry season, surface and bottom water salinity were significantly lower inshore than offshore (resp. p <0,001 & p <0,05 and p <0,001 & p <0,01, appendix 3.1), with a lower average surface and bottom salinity inshore (Fig. 8-9). Further, both inshore (p <0,001) and offshore (p <0,001) surface water salinity was significantly lower in the wet compared to the dry season. Bottom water salinity was only significantly lower in the inshore

S S

S NS NS S

Fig. 8-9: Mean surface and bottom water salinity and temperature (± SD) by season (Dry-Wet) and shore (Inshore-Offshore) or the interaction. S =significant, NS = non-significant.

12

zone in the wet season (p < 0,001). Surface water temperature reached a maximum and a minimum respectively in October-November (28,7 and 28,3 °C) and March (25,7 and 23,2 °C). The bottom water temperature at the inshore stations (TK03 & TK10) reached a maximum in June-July (28,2°C) and August-September (27,5 °C) at the offshore stations (TK17 & TK24, Fig. 6-7). Surface water temperature was significantly influenced by the factor ‘season’, not by ‘shore’ or the interaction between them (Pseudo-F = 4,2796; p < 0,05; Fig. 8, appendix 3.1). Surface water temperature was lower in the wet than the dry season (Fig. 8-9). Bottom water temperature was only significantly influenced by the factor ‘shore’ with a higher bottom temperature inshore than offshore (Pseudo-F = 19,681; p <0,001; Fig. 9, appendix 3.1). The highest found correlation was negative and occurred between bottom water salinity and temperature (r = - 0,68; Spearman; p < 0,001).

3.1.2 Secchi and suspended matter

Secchi depth and amount of suspended matter varied respectively around 1,2 ± 1,1 m and 91,1 ± 57,9 mg/l. The clarity of the water column of the offshore stations (TK 17 & TK24) reached its maximum around August-September (5,4 m) and its minimum around February – May (1 m, Fig. 10). For the inshore stations (TK03 & TK10), no huge variations in water clarity through the year were observed. Two maximum peaks in the amount of suspended matter around May-June (228,9 mg/l) and February (187,6 mg/l) can be observed at the inshore stations, and as well two minima were observed around March (44,0 mg/l) and November (56,6 mg/l, Fig. 10). Wet Dry The average amount of suspended matter at Wet Dry the offshore stations remained low the whole S year round. Secchi depth was significantly influenced by the factors ‘season’ and ‘shore’ but not by their interaction. In the wet season and in the inshore stations, Secchi depth was significantly lower compared to resp. the dry season (Pseudo-F = 5,531 ; p < 0,05; Fig. 11, appendix 3.2) and the offshore area (Pseudo-

Fig. 10: Secchi depth and the amount of suspended matter over F=29,348; p <0,001; Fig. 11; appendix 3.2). one year (2014-2015) with inshore stations (red and green) and Suspended matter was only significantly offshore stations (blue and yellow). A loess smoother was applied influenced by the factor ‘shore’ with higher to the data points who were scaled on a real-time axis. concentrations of SPM inshore than offshore (Pseudo-F=28,538; p <0,001; Fig. 12; appendix 3.2). The variables Secchi depth and suspended matter were inverse correlated (R=-0,724; Spearman, p < 0,001).

Fig. 11-12: Mean suspended matter and Secchi depth by season (Dry- Wet) and shore (Inshore-Offshore) (± SD). S = significant, NS = non-significant S S NS S

13

3.1.3 Sediment analysis

The values of the median grain size and the percentage mud content varied respectively around 9,6 ± 0,9 μm and 98,34 ± 3,4 %. Fluctuations of the median grain size were mainly seen in station TK24 Table 9: p-values posthoc station, *<0.05,**<0.01,***<0.001 where it reached a maximum in August-September (12,9 μm) and minimum at the end of January (8,9 μm, Fig. 13). The median grain size of station TK24 was higher compared to the other stations and fluctuated and dropped heavily around August- September. Mud content in station TK03 dropped in July and increased again around September. The percentage mud in station TK24 peaked in May-June (99,7%) and January-February (99,7%) and dropped in July-August (93,5 %, Fig. 13). The median grain size was not significantly influenced by the factors ‘season’ and ‘shore’ or the interaction between Wet Dry them (Fig. 14). A significant difference Wet Dry of the mud content was only found for S the factor “shore”. The percentage Fig.13: Median grain size and the percentage mud over one year mud was significantly higher inshore (2014-2015) with the inshore stations (red and green) and the offshore stations (blue and yellow). A loess smoother was applied than offshore (Pseudo-F=; 3,8128; p < to the data points who were scaled on a real-time axis. 0,05; Fig. 15; appendix 3.3)

NS NS NS S

Fig. 14-15: Mean median grain size and mud content by season (Dry-Wet) and shore (Inshore-Offshore) (± SD). S =significant, NS = non-significant.

14

3.2 Life stages of X.kroyeri

3.2.1 Densities of postlarvae, juveniles and adults

A total number of 4822 postlarvae and 156 juveniles were found in the hyperbenthic samples and 1659 individuals (962 juveniles and 697 adults) were sampled by the epibenthic net. Figure 16 represents the densities of postlarvae, juveniles from the hyper - and epibenthic net and adults through time and space. Postlarvae showed a peak density in July in both

inshore stations (TK03 & TK10) followed by

a smaller peak in TK10 in October (Fig. 16).

Juveniles sampled with the hyperbenthic net peaked in March-April at station TK17

and in August-September at station TK03. Postlarvae Juveniles from the epibenthic net showed an inshore-offshore migration with juvenile

density peaks at stations TK03-TK10 in August - September and stations TK10, Wet Dry TK17, TK24 towards the end of the year. A Wet Dry similar pattern was observed for the S densities of adults with highest densities

Juveniles (Hyper) Juveniles towards the end of the year at the offshore stations (TK17 & TK24). Fig. 17 illustrates

the relative densities of postlarvae,

juveniles from the hyper - en epibenthic net and adults by season and shore. The large densities of postlarvae relative to juveniles and adults are dominating both graphs. The

Juveniles (Epi) Juveniles wet season was completely dominated by

postlarvae, whereas the dry season was

characterized by a relative higher

percentage of juveniles and adults. When

looking at the spatial pattern, inshore

stations were entirely dominated by

postlarvae, while in the offshore stations Adults Adults

relative higher percentages of juveniles and adults occurred (Fig. 17). The density of postlarvae and the juveniles from the hyperbenthic net was not significantly influenced by the factors ‘season’ and Fig. 16: Densities of postlarvae juveniles from the epi- and ‘shore’ or the interaction between them hyperbentic net and adults expressed as Ind. m-² over one year (2014-2015) with the inshore stations (red and green) and the (Fig. 18-19, appendix 3.4). Density of the offshore stations (blue and yellow). Data points were scaled to epibenthic juveniles and the adults were a real-time axis. significantly influenced by the factor ‘season’, not by ‘shore’ or the interaction between them (resp. Pseudo-F = 6,0528 & Pseudo- F=4,4695) with lower densities during the wet season (both p <0,05; Fig. 20-21 ; appendix 3.4)

15

Fig. 17: Relative densities of postlarvae, juveniles and adults by season (Wet-Dry) and Shore (Inshore-Offshore).

NS NS NS NS

S NS S NS

Fig. 18-21: Mean density of postlarvae, juveniles, adult by season (Dry-Wet) and shore (Inshore-Offshore) (± SD). S =significant, NS = non-significant

16

3.2.2 Cohort analysis

Total lenght (mm) of postlarvae, juveniles and adults by month

Fig. 22-23: Length-frequency data for the postlarvae (left), juveniles (hyper and epi) and adults (right).

Figures 22-23 represent the length-frequency data of postlarvae, juveniles and adults by month. From figure 22 one large recruitment event can be seen in July with a smaller one in October. Though postlarvae were observed in every month during analysis of the hyperbenthic samples. The lowest number of postlarvae were found in the month November (nallstations = 4). On figures 22-23 several cohorts can be followed through time but it becomes harder to separate cohorts as they grow in time due to lack of data. Fig. 24 gives the mean (black) and standard deviation (red) of the cohorts identified by NORMSEP. For the months of November and May no cohorts could be identified. Four cohorts were identified during the main recruitment event. In October and February two cohorts were identified. In February-March one cohort was identified surpassing the 25 mm border which probably was recruited in January. No Fig. 24: Result of cohort analysis by NORMSEP, mean (black), standard deviation (red) through time on a real-time x-axis. Cohorts are connected cohorts could be identified beyond with a dotted line. 40 mm due to lack of data.

17

3.3 The life stages of X.kroyeri and the environment

3.3.1 Postlarvae and the abiotic parameters

Figure 25 represents the density of postlarvae and the abiotic parameters (excluding the 2 parameters from the sediment as postlarvae were caught in the hyperbenthic zone). As mentioned before the main peak of recruitment occurred at the end of the wet season in July at the inshore stations. Wet Dry Surface and bottom water salinity Wet Dry were high during the main S recruitment in contrast with the low values during the wet season. During this event salinity was lower in the inshore zone then the offshore zone. Surface and bottom water temperature were high at the in - and offshore stations during the main recruitment event. Bottom water temperature peaked around the main recruitment event and was higher at the inshore stations. Secchi depth was low and suspended matter was high at the inshore stations during the main recruitment event (Fig. 25). The highest correlation found was negative and occurred between the densities of postlarvae and the abiotic parameter surface water salinity (R = -0,59; Spearman, p < 0,05).

Fig. 25: Postlarvae density, surface salinity, surface temperature, bottom salinity, bottom temperature, Secchi depth, suspended matter over one year (2014-2015). The sampled data points were scaled on a real-time x-axis.

18

3.3.2 CCA

Only one significant variable was found (table 1), though surface water salinity and Secchi were used to draw the CCA (Fig. 26). The CCA showed two groups of life stages: postlarvae (PL) & small juveniles from the hyperbenthic net (JH) and large juveniles (JE) & adults from the epibenthic net (A) indicating a similar distribution pattern. The densities of adults and juveniles from the Small recruitment epibenthic net were positively correlated event with surface water salinity and Secchi

Main recruitment depth. The densities of juveniles from the

event hyperbenthos net and postlarvae were negatively correlated with surface water

salinity and secchi depth. Axis 2: 28 2: Axis %

Table 1. Significant environmental variables calculated by the Adonis function (9999 permutations) Environmental Variable R² p-value

Surface_Salinity 0,27274 0,0001*** Surface_Temperature 0,03575 0,2619 Bottom_Salinity 0,00763 0,8008 Fig. 26: Canonical correspondence analysis of the Bottom_Temperature 0,01467 0,6023 Axis 1: 70% environmental variables (S_Sal: surface salinity, Secchi: Secchi depth, and the density of the demographic categories Secchi depth 0,06079 0,1015 (postlarvae (PL), juveniles hyper (JH), juveniles epi (JE), adults Suspended_Matter 0,01850 0,5118 (A)). Data from August_TK24 was excluded as the densities of Depth 0,00971 0,7360 the 4 demographic categories were zero. Median Grain Size 0,01481 0,5966 % Mud 0,00554 0,8678

3.4 Von Bertalanffy growth model (VBGM) & size at first maturity

A Von Bertalanffy Growth Curve was applied to the total length (mm) length-frequency data of males and females from the epibenthic net (Fig. 27). Parameters of the Von Bertalanffy growth model were estimated on L∞ = 131,25 (mm), K = 0,70 (year); t0 = -0,0095268 (Rn = 0,216) for males and for females L∞ = 147(mm); K = 0,55 (year); t0 = -0,01082 (Rn = 0,204) with an estimated longevity of 79 and 122 weeks for males and females respectively at 95% of the asymptotic length. The combined length frequency data of males and females from the epibenthic net gave the following growth parameters: L∞ = 144 (mm), K = 0,58 (year) ; t0 = - CL50 = 14,4 mm 0,01053 (Rn = 0,230). Male carapace size at first maturity (CL50) was estimated to be 14,4 mm (Fig. 27).

Fig. 27: VGBM for male and female X.kroyeri & estimated carapace length at first maturity for male X.kroyeri. 19

4. Discussion

In the present study population dynamics of Xiphopenaeus kroyeri in Suriname have been investigated for the first time along an inshore-offshore transect over one year by assessing the density of the different life stages of the X.kroyeri. Results have shown that recruitment occurred continuously in the in - and offshore zone. A main and a smaller recruitment event (increased postlarvae density) occurred in July and October in the inshore zone. The bottom temperature peaked during the main recruitment event in the in- and offshore stations. The inshore stations were characterized by a lower salinity, a higher temperature, lower clarity and a higher amount of suspended matter then the offshore stations. Postlarvae and small juveniles were more present in the shallow inshore zone, large juveniles and adults in the deep offshore zone.

Spatio-temporal patterns of the life stages of X.kroyeri and the environment

Recruitment occurred continuously as postlarvae were found in every sampling campaign. However density of the postlarvae increased abruptly in the months July and October. Results from cohort analysis found the largest number of cohorts during the main recruitment event. This analysis could estimate time of recruitment/spawning of the newly recruited cohorts more precisely as the first larval stages of X.kroyeri become postlarvae after approximately 16 days (Almeida et al. 2012, Colvin & Brand 1977, Gillet 2008). Based on this knowledge, reproductive females spawned around the end of June and the 20th of September for the main and small recruitment event respectively. In a study in French Guiana by Oliveira (1991) a similar recruitment pattern occurs with peak recruitment in July-August and November (Oliveira 1991). Bottom water temperature peaked during the main recruitment event in the in - and offshore stations. According to Almeida et al. (2012), Heckler et al. (2014) temperature is an essential factor for reproductive females along the Brazilian coast and has an influence on larval development and survival. No significant difference in postlarvae density was found along the inshore-offshore gradient but postlarvae were in general more found in the inshore stations. Fluctuations in postlarvae densities were mainly caused by the seasonal recruitment (main and small recruitment event). The main recruitment event occurred at the end of the wet season in the inshore stations. This season was marked by a low surface and bottom water salinity due to large fresh water inflows originating from the discharge of the Suriname River and the Amazon which has its peak efflux between January and April (Anthony et al. 2010, Anthony et al. 2013, Froiefond et al. 1988, Hu et al. 2011, Park and Latrubesse 2014). The inshore stations were marked by a lower salinity then the offshore stations. Salinity of the surface water showed a higher variability then the salinity of the bottom water. The bottom water salinity of the offshore stations did not differ between the seasons offering a more stable environment. Oliveira (1991) found very low densities of postlarvae at salinity levels of 5 psu in the estuary of Cayenne, French Guiana (Oliveira 1991). Although X.kroyeri does not need the estuary to fulfill their life cycle (Almeida et al. 2012, Branco 2005, Branco et al. 2013, Castro 2005, Heckler et al. 2014, Simoes et al. 2010, Santos et al. 2013). The inshore zone was further characterized by a significant lower clarity and higher amount of suspended matter then the offshore zone. During the main recruitment event suspended matter was high in the inshore stations. The increased freshwater outflow from the Suriname River and the Amazon during the wet season probably brings a lot of food with the currents and may influence the recruitment events. A trophic ecology study of X.kroyeri in Suriname by Kerkhove et al. (2014) illustrates the omnivorous diet of postlarvae. Kerkhove (2014) suggests that postlarvae mainly feed on organic detritus and small prey like copepods. Lucifer faxoni also belongs to the diet (Kerkhove et al. 2014) and this species was found in densely aggregations both in the in – and offshore zone of the present study. Postlarval lengths of 20-25 mm were underrepresented in the present study and may be the cause of a more rapid recent settlement of postlarvae from the pelagic environment to the 20

benthic substrate (Oliveira 1991). Also length measurement of individuals of 2-3 mm total length were morphologically observed to be postlarvae by a stereomicroscope, in contrast with the recruitment study by Oliveira (1991) who considered the range of postlarval length to be between 6 and 25 mm. No significant spatial or seasonal difference was found for the densities of the small juveniles caught by the hyperbenthic net (appendix 3.5). However most juveniles from the hyperbenthic net were found in the inshore zone in August right after the main recruitment event. The inshore zone in August was marked by an increased surface and bottom water salinity and surface water temperature. Bottom water temperature and the amount of suspended matter in the inshore zone were still high and may have enhanced the development of the juveniles. A high temperature and large amounts of detritus are needed for their high energy demand for growth (Almeida et al. 2012, Branco et al. 2013, Heckler et al. 2014). The inshore zone was characterized by a low clarity the whole year around and could offer protection against predators as small juveniles are characterized by a bad burrowing capacity (Almeida et al. 2012, Heckler et al. 2014). A significant seasonal pattern was found for the juveniles and adults from the epibenthic net, representing more individuals were present at the end of the dry season as a consequence of the main and small recruitment events. Length-frequency analysis by month showed the growth of the postlarvae at the main recruitment event in July with small juveniles in August and large sized juveniles and adults at the end of the dry season (November-January). Juveniles and adults from the epibenthic net were more frequently found in the deep offshore stations where conditions are more stable for spawning (Grabowski et al. 2014). Salinity of the bottom water at the offshore stations did not differ significantly during both seasons. Several authors suggest that a low salinity prevents maturing juveniles and adults to enter the inshore zone, especially in the wet season when salinity is low (Heckler et al. 2014a, Lopes et al. 2014, Oliveira 1991). Surface temperature increased to a maximum towards the mid of the dry season and may have triggered the second recruitment event. Bottom water temperature was lower offshore but peaked during the main recruitment event. According to Heckler et al. (2014) and Almeida et al. (2012) temperature stimulates reproductive females to the growth and development of the ovaries. Bottom temperature serves also as a biological barrier and X.kroyeri perish beneath 21,8 °C (Bissaro et al. 2013). The offshore zone was also characterized by a significant lower amount of suspended matter and an increased clarity offering more stable conditions. Sediment analysis has shown that the four stations in general were characterized by a high mud content; originating from the high silt and clay discharge coming from the Surinam River and the Amazon River (Anthony et al. 2013, Hu et al. 2004, Kuehl et al. 1985, Nittrouer et al. 1986, Park and Latrubesse 2014). The furthest station TK24 was marked by a lower mud content and a larger median grain size. Freire et al. (2011) suggests that X.kroyeri prefers fine sand/very fine sand as a benthic substrate. This substrate requires less energy for excavation and X.kroyeri is able to flee from predators more rapidly (Dall et al. 1990, Freire et al. 2011). Though mud content in the coastal area of Suriname is high and X.kroyeri is found in dense aggregations (Pérez et al. 2014). According to Pérez et al. (2014) higher fishing activity occurs in the east of Suriname, where fishing areas are deeper and X.kroyeri catches are known to be bigger. Most pregnant females are also found in the east (Pérez et al. 2014). It is possible that X.kroyeri migrate eastwards to deeper grounds where the proportion sand/mud is higher.

Size at first maturity

The maturity of females was defined by the color of the abdominal cavity onboard. Size at first maturity could not be determined as no clear distinction was made between the juvenile-adult stage. Almeida et al. (2012) estimated the maturity of females based on a study by Fransozo et al. (2011)

21

that focused on the presence of a space between the aperture and the thelycum of females to determine maturity. According to Campos et al. (2009) macroscopic observation of the ovaries of females may lead to misclassification and under- or over L∞ K Study 131,3 mm ♂ 0,70 /year ♂ estimation of the size at first maturity (CL50) (Almeida et al. Present study 147,0 mm ♀ 0,55 /year ♀ 2012, Branco et al. 1999, Branco 2005, Castro et al. 2005, 134,4 mm ♂ 1,93 /year ♂ Fernandes et al .2011 Coelho & Santos 1993, Fernandes et al. 2011, Nakagaki & 148,8 mm ♀ 1,65 /year ♀ 128,0 mm ♂ 3,14 /year ♂ Lopes et al. 2014 Negreiros-Fransozo 1998). Campos et al. (2009) emphasized 145,0 mm ♀ 2,50 /year ♀ 133,0 mm ♂ 0,30 /year ♂ Branco et al. 2005 the observation error by a microscopic visualization of the 154,0 mm ♀ 0,26 /year ♀ 135,0 mm ♂ 0,61 /year ♂ development of the ovaries which undergo four maturation Branco et al. 1994 150,0 mm ♀ 0,53 /year ♀ 128,0 mm ♂ 3,14 /year ♂ stages. The author suggests that juvenile females become Grabowski et al. 2014 145,0 mm ♀ 2.65 /year ♀ adults around a mean carapace length of 24 mm and supports Table 2: Growth parameters of male and female this theory by the slower ontogenetic growth of females and a X.kroyeri. delay in ovary maturation. This delay in growth has been illustrated by the Von Bertalanffy growth model; a lower growth coefficient for females was observed in the present study and has been the case in studies along the Brazilian coast (Table 2). Females have shown to grow to a larger size (table 2), and could offer a larger space for ovary development and oocyte production (Bissaro et al. 2013). Determining the CL50 of female X.kroyeri remains a problematic issue, but the parameter can be useful in estimating the sustainability of the fishery and indicate if overexploitation of the resource occurs even if recruits are protected. However the parameter is also subjected to the variability of environmental parameters, seasonality and latitude (Almeida et al. 2012, Lopes et al. 2014). Failure in the true assessment of this parameter can result in assuming wrong premises and implementing inadequate management decisions.

Implications for resource management

The observed recruitment pattern (continuous recruitment with a main and smaller recruitment event in July and October respectively) and the inshore-offshore migration have been illustrated in several Brazilian studies. Along the Brazilian coast the main and smaller recruitment event occurs around April and November resp. (Almeida et al. 2012, Branco et al. 2005, Branco et al. 2013, Castro 2005, Heckler et al. 2014, Simoes et al. 2010, Santos et al. 2013). A non-fishing season (March through May) is implemented in the overexploited stocks of south and south-eastern coast of Brazil to protect juveniles and reproductive females during the main recruitment event (Almeida et al. 2012, Branco et al. 2013, Castro 2005, FAO 2002, Heckler et al. 2014, Grabowski et al. 2014). According to Simoes et al. (2010) and Heckler et al. (2014) larger densities of X.kroyeri occur after the closing season. A similar management decision could be implemented in the seabob fishery of Suriname, a closing season from June through August to protect the reproductive females. Juvenile are already protected as no trawling is allowed in the shallow inshore zone (Southall et al. 2011). Though in a study by Perez et al. (2014) CPUE levels of the seabob fishery in Suriname from the last years seem to stay within the safe biological limits and no overfishing is currently occurring. Future research should focus on the existence of metapopulations along the coasts of the Guiana’s as recruitment patterns have shown to be similar between Suriname and French Guiana (Oliveira 1991). It is plausible that several seabob populations along the coasts of the Guiana's are connected with each other as larves last for 16 days and a strong north-west current (Guinea current) exists. This phylogeographic research is especially important for further stock assessment that needs to consider the connectivity between the seabob stocks of Guyana and Suriname. If connectivity is significant, overfishing of the Suriname seabob stock could further diminish the overfished Guyana stock (FAO FISHSTAT2015).

22

5. Conclusion

In the present study population dynamics of the X.kroyeri shrimp in Suriname has been investigated for the first time along an inshore-offshore transect over one year by assessing the density of the different life stages of the X.kroyeri. Results have shown that recruitment occurred continuously in the in- and offshore zone. A main and a smaller recruitment event (increased postlarvae density) occurred in July and October in the inshore zone. The bottom temperature peaked during the main recruitment event in the in- and offshore zone. The inshore stations were characterized by a lower salinity, a higher temperature, lower clarity and a higher amount of suspended matter then the offshore stations. Postlarvae and small juveniles were more present in the shallow inshore zone, large juveniles and adults in the deep offshore zone. The sustainability of the fishery could be enhanced by a closed fishing season from June through August to protect the reproductive females during the main recruitment event.

6. Appendix

1.Introduction

1.1. Table 3 groups (Gillet 2008)

Species Penaeids Carideans Sergastids Group/Character

Habitat zone Mainly tropical and Mainly temperate and Mainly tropical and temperate temperate arctic

Protandrous Separate Separate Sex hermaphrodites

Female hold fertilized Reproduction Eggs released directly eggs to her abdominal Eggs released directly into into water appendages until water hatching

Clutch Size 500 000 -1 000 000 100-4000 7700-8700

Life cycle Short(<3 years) Long(3-8 years) Very Short(one year or less)

Body size Small to large Very small to large Small to microscopic

23

2. Methods

2.1. Salinity-Temperature diagrams (Strømme et al. 1989)

Up to 3 different water masses can be found along the continental shelf of Suriname: deep Antarctic Intermediate water (AIW), high saline North Atlantic Central Water (NACW) around a depth of 100m and an upper water mass with high spatio-temporal variability that is characterized by a lower salinity due to influence of fresh water from the Amazon, local rivers and/or precipitation. Primary production is influenced by nutrient rich North Central Atlantic water welling up from 100m depth to levels of 50m. Temperature of the shelf water ranges between 25°C and 29°C, high values are found in the dry seasons. This is also the period when the Guinea current diminishes which leads to a weaker upwelling of the subsurface water and a lower primary productivity

2.2. Coordinates (Kerkhove et al. 2014) and sampling dates

Station Coordinates TK03 6°01’36” N, 55°12’40”W TK10 6°09’40” N, 55°12’40”W; TK17 6°14’00” N, 55°12’40”W TK24 6°18’08” N, 55°12’40”W Sampling Campaign nr. Date

1 23/02/2014

2 30/03/2014

3 21/05/2014

4 12/07/2014

5 22/08/2014

6 05/10/2014

7 16/11/2014

8 22/01/2015

24

2.3 Onboard form

25

2.4 Hyperbenthosnet (Mees 1994)

2.5 Adults characteristics (Oliveira 1991)

26

2.6 postlarvae characteristics (Oliveira & L’Homme 1993)

27

2.7 Calculation of the trawled area of the hyper- and epibenthic net

The trawled area of the epibenthos net was calculated by the formula: (3,85m x 2/3) (opening of the net) X (2,5 x 0,514 m/s) (speed; average of 2,5 knots) X (30 x 60 s) (time) = 5936,7 m ². In the same way, the dragged area of the hyperbenthosnet was estimated at 375 m² (1m net opening, 5 min trawling; 2,5 knots).

2.8 Loess Smoother (IOWA State University 2015)

28

3. Results

3.1 Salinity & Temperature significant p-values ( - = p >0.05* = p<0.05,** = p<0.01,*** =p<0.001)

Variable Parameter Mean St dev Test P-value Pattern Surface Salinity (psu) Season Dry 35,5 0,6 Wet 33,3 2,0 Shore Inshore 33,6 2,2 Offshore 35,1 0,9 Shore X Season PERMANOVA ** temporal-spatial Shore “ *** temporal- spatial Season “ *** temporal-spatial Inshore:Wet – Inshore:Dry “ *** temporal- spatial Offshore:Wet– Offshore:Dry “ *** temporal- spatial Wet:Inshore – Wet:Offshore “ *** temporal- spatial Dry-Inshore – Dry-Offshore “ * temporal- spatial Bottom Salinity (psu) Season Dry 35,5 0,6 Wet 33,3 2,0 Shore Inshore 34,0 1,9 Offshore 36,4 0,1 Shore X Season PERMANOVA *** temporal-spatial Shore “ *** temporal- spatial Season “ *** temporal-spatial Inshore:Wet – Inshore:Dry “ *** temporal- spatial Offshore:Wet– Offshore:Dry “ - Wet:Inshore – Wet:Offshore “ *** temporal- spatial Dry-Inshore – Dry-Offshore “ ** temporal- spatial Surface Temperature (°) Season Dry 27,8 0,7 Wet 27,0 0,7 Shore Inshore 27,5 0,8 Offshore 27,3 0,8 Shore X Season PERMANOVA - Shore “ - Season “ * temporal Bottom Temperature (°) Season Dry 26,4 1,5 Wet 26,2 1,5 Shore Inshore 27,3 1,0 Offshore 25,4 1,3 Shore X Season PERMANOVA - Shore “ *** spatial Season “ -

29

3.2 Secchi & Suspended Matter significant p-values ( - p>0,05,* = p<0.05,** = p<0.01,*** =p<0.001)

Variable Parameter Mean St dev Test P-value Pattern Secchi depth (m) Season Dry 1,6 1,3 Wet 0,9 0,7 Shore Inshore 0,5 0,3 Offshore 2,0 1,1 Shore X Season PERMANOVA - Shore “ *** spatial Season “ * temporal Suspended matter (mg/l) Season Dry 82,9 46,0 Wet 99,3 80,0 Shore Inshore 134,8 64,2 Offshore 47,4 21,7 Shore X Season PERMANOVA - Shore “ *** spatial Season “ -

3.3 Median Grain Size, Median Grain size of the sand fraction, percentage mud significant p-values (* = p<0.05,** = p<0.01,*** =p<0.001)

Variable Parameter Mean St dev Test P-value Pattern Median Grain Size (µm) Season Dry 9.5 1.0 Wet 9.7 0.8 Shore Inshore 9.3 0.6 Offshore 9.9 1.1 Shore X Season PERMANOVA - Shore “ - Season “ -

Median Grain Size of the Season Dry 99.0 58.3 Sand Fraction (µm) Wet 105.3 55.0 Shore Inshore 80.6 32.4 Offshore 123.7 66.4

Percentage Mud (%) Season Dry 98.7 3.5 Wet 98.0 3.0 Shore Inshore 99.4 0.8 Offshore 97.3 4.3 Shore X Season PERMANOVA - Shore “ * spatial Season “ -

30

3.4 Density of Postlarvae, Juvenile (epi and hyper) and adult p-values((* = p<0.05,** = p<0.01,*** =p<0.001)

Variable Parameter Mean St dev Test P-value Pattern Postlarvae Density Season Dry 0.1243515 0.4555835 (count/m²) Wet 0.6395335 1.60714 Shore Inshore 0.7423192 1.62906704 Offshore 0.0215658 0.03303751 Shore X Season PERMANOVA - Shore “ - Season “ - Hyper Juvenile Density Season Dry 0.01459144 0.0496618 (count/m²) Wet 0.01019664 0.02137453 Shore Inshore 0.016820688 0.04940622 Offshore 0.007967389 0.02122942 Shore X Season PERMANOVA - Shore “ - Season “ - Epi Juvenile Density Season Dry 0.00799055 0.0090945 (count/m²) Wet 0.002179241 0.002497265 Shore Inshore 0.003463624 0.004903601 Offshore 0.006706167 0.008792316 Shore X Season PERMANOVA - Shore “ - Season “ * Temporal Adult Density Season Dry 0.005737615 0.007662856 (count/m²) Wet 0.001684437 0.00151911 Shore Inshore 0.002147658 0.002290257 Offshore 0.005274395 0.007702557 Shore X Season PERMANOVA - Shore “ - Season “ * Temporal

31

3.5 Carapace lengths of juvenile (hyper and epi) and adult seabob

Method:

Carapace lengths of juveniles (hyper - and epi) and adults were tested for size differences by conducting a Kruskal test and a posthoc pairwise Wilcox test with Bonferroni correction. A Chi square and Wilcox test were used to test for differences in the monthly sex-ratio and carapace length between males and females of the epibenthic net.

Results:

156 and 962 juveniles from resp. the hyper – and epibenthic net, and 697 adults had average carapace lengths of 9,4 ± 3,2 mm; 16,9 ±4,1 mm and 19,8 ± 3,2 mm respectively. A significant size difference was found between all 3 groups (Kruskal-Wallis test, p< 0,001, Fig. 24). Based on the individuals from the epibenthic net, the study was in general female-biased, 900 females (54,6%) and 749 males (45,4%) were found. A Chi-square test Pairwise Wilcox test Bonf. corr indicated that the sex-ratio differed from 1:1 Juvenile Hyper – Adult during the 8 month sampling campaign (X² = Juvenile Epi – Adult p<2.2e-16 9,9576, p > 0,05). The average carapace Juvenile Hyper - Epi length of the females was 19,0 ± 4,2 mm and significantly higher (Wilcox, p <0,001) then the average male carapace length 17,1 ± 3,4 Fig. 23: Average carapace length for juveniles from the epi- mm (Fig 30). Figure 24 represents the and hyper – benthicnet and adults with mean (black) and frequency distribution of the carapace length standard deviation (red) and posthoc p-values. by sex and stage (I= Immature, M = Mature). An abrupt distinction in female juvenile-adult at a carapace length of 24 mm was observed at Fig. 25, as female juveniles exceeding 24 mm were considered to be mature. Male juveniles began to mature at a carapace length of 12 mm and all juveniles were mature at a carapace length of 24 mm. Juvenile females began to mature at a carapace length of 15 mm (Fig. 26).

Fig. 25: Frequency distribution of the carapace Fig. 24: Frequency distribution of the carapace length length of caught juveniles and adults (males top, of found juveniles and adults of the epibentic net by females bottom) of the epibenthic net by sex and sex. dashed line = average, dotted line : standard stage (I = Immature, M = Mature). Longdashed line = deviation average, dotted line: standard deviation 32

7. R-code

6.1 Proportion juvenile-mature males

M <- subset(Seabob_CL,Sex =="M") require(plyr) results.by.CL <- ddply(.data = M,.var = c("CL.mm."), .fun = function(x) {

data.frame(n = nrow(x),

An = nrow(subset(x, Stage %in%

c("M"))),

Proportion_Adult = nrow(subset(x, Stage %in%

c("M"))) / nrow(x))})

6.2 CCA Graph library(vegan) library(ggvegan)

Seabob<-Seabob[c("A","JE","PL","JH")]

Env<-Env[c("S_Sal","Secchi")]

Seabob <-scale(Seabob)

Seabob <-(Seabob)^2

Env <-scale(Env)

Env <-(Env)^2 ord<-cca(Seabob, Env) obj<-fortify(ord)

Label.1 <- c("A","JE","PL","JH","Fe03","Feb10","Feb17","Feb24","Mar03","Mar10","Mar17"

,"Mar24","May03","May10","May17","May24","Aug03","Aug10","Aug17","Aug24","Oct03",

"Oct10","Oct17","Oct24","Nov03","Nov10","Nov17","Nov24","Jan03","Jan10","Jan17",

"Jan24","con1","con2","con3","con4","con5","con6"

,"con7","con8","con9","con10","con11","con12","con13","con14",

"con15","con16","con17","con18","con19","con20","con21","con22"

,"con23" ,"con24","con25","con26","con27","con28" ,"S_Sal"

,"Secchi" ) obj <- data.frame(obj,Label.1) layers<- levels(obj$Score)

33 dimlabels <- attr(obj, "dimlabels") obj <- obj[obj$Score %in% layers, , drop = FALSE] plt <- ggplot() want <- obj$Score %in% c("species")

plt <- plt +

geom_text(data = obj[want, , drop = FALSE ],

aes(x = Dim1, y = Dim2, label = Label.1,

colour = Score),colour = "#7C1919",fontface = "bold",size =5) want <- obj$Score %in% c("sites") plt <- plt +

geom_text(data = obj[want, , drop = FALSE ],

aes(x = Dim1, y = Dim2, label = Label.1,

colour = Score),colour = "#2D2D2D",size = 2) want <- obj$Score == "biplot" tmp <- obj[want, ] obj <- obj[!want, ] bnam <- tmp[, "Label.1"] cnam <- obj[obj$Score == "centroids", "Label.1"] obj <- rbind(obj, tmp[!bnam %in% cnam, , drop = FALSE]) mul <- vegan:::ordiArrowMul(obj[want, , drop = FALSE ]) obj[want, c("Dim1","Dim2")] <- mul * obj[want, c("Dim1","Dim2")] plt <- plt + geom_segment(data = obj[want, , drop = FALSE ],

aes(x = 0, y = 0, xend = Dim1, yend = Dim2),

arrow = arrow(length = unit(0.2, "cm")),

colour = col) plt <- plt + geom_text(data = obj[want, , drop = FALSE ],

aes(x = Dim1, y = Dim2, label = Label.1),

angle = 0.5, colour = "#2C422D",fontface = "bold", size = 4.5) xlab <- dimlabels[1] ylab <- dimlabels[2] plt <- plt + xlab(xlab) + ylab(ylab)

#plt <- plt + + coord_fixed() plt<- plt + ggtitle("CCA") + theme(plot.title = element_text(size=20, face="bold", colour = "#2D2D2D", vjust=2)) plt <- plt + theme(axis.text = element_text(face = "bold", size = 10))

## change colour x-axe 34 plt<-plt + theme(

axis.title.x = element_text(face="bold",color="#2D2D2D", size = 15),

axis.title.y = element_text(face="bold",color="#2D2D2D" , vjust=1, size=16))

## change background colour plot plt <-plt+ theme(plot.background = element_rect(fill = 'white')) plt <-plt + theme(panel.background = element_rect(fill = '#EAE5BE'))

8. References

Afifi A., Clark V., May S. 2004. Computer-Aided Multivariate Analysis. 4th ed. Boca Raton, Fl: Chapman & Hall/CRC.

Allen, M. S., Hightower J. E.. 2010. Fish population dynamics: mortality, growth, and recruitment. Inland Fisheries Management in North America, 43-79.

Almeida A. C. J., Baeza A. , Fransozo V., Castilho A. L., Fransozo A. 2012. Reproductive biology and recruitment of Xiphopenaeus kroyeri in a marine protected area in the Western Atlantic: implications for resource management. Aquatic Biology, Vol. 17: 57–69.

Anthony E.J., Gardel A., Gratiot N. ,Proisy C., Allison M.A., Dolique F., Fromard F.2010. The Amazon-influenced muddy coast of South America: A review of mud-bank–shoreline interactions. Earth-Science Reviews 103, 99–121.

Anthony J.E., Gardel A., Proisy C., Fromard F., Gensac.E., Peron C., Walcker R., Lesourd S.2013. The role of fluvial sediment supply and river-mouth hydrology in the dynamics of the muddy, Amazon-dominated AmapáeGuianas coast, South America: A three-point research agenda. Journal of south American earth Sciences. 18-24.

Baker R., Fujiwara M, Minello J.T. 2014.Juvenile growth and mortality effects on white shrimp Litopenaeus setiferus population dynamics in the northern Gulf of Mexico. Fisheries Research 155 ,74–82.

Braak C.J., Verdonshot F.M.P. 1995. Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquatic Sciences 57/3.

Branco J., 2005. Fishery and biology of the seabob shrimp Xiphopenaeus kroyeri(Heller)(Crustacea, Penaeidae) at Armacão do Itapocoroy, Penha, Santa Catarina, Brazil, Revista Brasileira de Zoologia 22(4), 1050-1062.

Branco J., Junior Cassiano Moritz H. 2001.Alimentacao natural do camarao sete-barbas, Xiphopenaeus kroyeri (Heller) (Crustacea, Decapoda) na Armacao do Itapocoroy, Penha, Santa Catarina, Revta bras. Zool. 18 (1): 53 – 61.

Branco J., Lunardon-Branco M. J., De Finis A, 1994, the growth of Xiphopenaeus kroyeri (Heller,1862)(Crustacea: Natantia:Penaeidae) from the Matinhos region, Parana state, Brazil. Arq. Biol. Tecnol. 37 (1): 1-8.

Branco J. Santos L.R., Barbieri E., Santos M.D.C.F., Rodrigues Filho J.L., 2013.Bol. Inst. Pesca, São Paulo, 39(3): 237 – 250.

Broadhurst, M.K. 2000. Modifications to reduce bycatch in prawn trawls: a review and framework for development. Reviews in Fish Biology and Fisheries, 10: 27–60.

Bulgakov N.P., Bulgakov S.N., Eremeev V.N. 1998.Guiana Current over North Brazil and cintinental slope. Physical Oceanography., vol 9,55-70.

Campos B. D.R, Luiz F.D.C, D'Incao F., Branco J .2009. Ovarian development and length at first maturity of the sea‑bob‑ shrimp Xiphopenaeus kroyeri (Heller) based on histological analysis, Nauplius 17(1): 9-12.

Castro R. H., Costa R. C., Frasozo A., Mantelatto L.M. F.2005. Population structure of the seabob shrimp Xiphopenaeus kroyeri(Heller,1862)(Crustacea: Penaeoidea) in the littoral of São Paulo, Brazil, Sci Mar, 69(1), 105-112.

Clarke, KR.1993. Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18: 117-143.

Clarke, KR, Gorley, RN, 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth.

Cleveland, W.S. 1979., Robust Locally Weighted Regression and Smoothing Scatterplots. Journal of the American Statistical Association, Vol. 74, pp. 829-836. 35

Cleveland, W.S. and Devlin, S.J. 1988, Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting,Journal of the American Statistical Association, Vol. 83, pp. 596-610.

Colvin L.B., Brand C.W.1977. The protein requirement of penaeid shrimp at various life-cycle stages in controlled environment systems. Proceedings of the annual meeting - World Mariculture Society, Volume 8, Issue 1-4, pages 821–840.

Cook H.L. 1964, A generic key to the Protozoean, mysis, and postlarval stages of the littoral Penaeidae of the northwestern Gulf of Mexico. Fishery Bulletin: Vol 65, no. 2.

Cook G.S., Fletcher P.J., Kelble C.R. 2014. Towards marine ecosystem based management in South Florida:Investigating the connections among ecosystem pressures, states,and services in a complex coastal system. Ecological Indicators 44 (2014) 26– 39.

Cha.B.J,Yoon S. P., Lee J.S, Kim S.G., Jung.R.H., Yoon, W.D, Shin J.K.J.2009.A primary study on the effect of artificial disturbance on a fishing area by shrimp beam trawl Kor. Soc. Fish. Tech., 45(4), 223 233.

Dall, W., Hill, B.J., Rothlisberg, P.C., Staples, D.J., 1990. Biology of the Penaeidae. In: Blaxter, J.H.S., Southward, A.J. (Eds.), Advances in Marine Biology, vol. 27. Academic Press, London, UK. 489 pp

Dayton P.K, Thrush S.F., Agardy M.T., Hofman R.J. 1995. Environmental effects of marine fishing, Aquatic conservation: marine and freshwater ecosystems, vol 5, 205-232.

Davies R.W.D. Cripps S.J., Nickson A., Porter G.2009. Defining and estimating global marine fisheries bycatch. 1-12

Derrell, C., Ferreira, L., Medley, P. Soekradj, R. 2009. Report of the Shrimp and Groundfish Resource Working Group (SGWG). CRFM 2009. In CRFM Fishery Report - 2009. Volume 1. Report of Fifth Annual Scientific Meeting – Kingstown, St. Vincent and the Grenadines, 09-18 June 2009. 167p.

Enin Udeme I.1995.First estimates of growth, mortality and recruitment parameters of Macrobrachium macrobrachion Herklots, 1851 in the Cross River estuary, Nigeria, . Dana, vol. 11(1), pp. 29-38.

Eayrs, S. 2005 (revised 2007). A guide to bycatch reduction in tropical shrimp-trawl fisheries.1-124.

FAO. 2002 Report of the first Regional Conference on the Sustainability of Fisheries Resources in the Brazil Guianas Shelf. Paramaribo, Suriname, 5-7 FAO Fisheries Report. No. 676, 27p.

FAOa. 2013. Case study on shared stocks of the shrimp and groundfish fishery og the Guinas-Brazil shelf CLME Case Study on Shrimp and Groundfish - Report. No. 9, 99 p.FAO, 2008, The Republic of suriname, 1-21

FAOb. 2013. DATA.FAO.ORG PORTAL. Atlantic seabob (UserTag),[Conseiled on 27 Apr 2015]. URI: http://data.fao.org/ref/37e38dc5-c7c7-4105-955d-4ef31db81245.html?version=1.0

FAO. 2014. The State of World Fisheries and Aquaculture 2014. Rome. 223 p.

Fernandes L.P., Silva A.C., Jardim L.P., Keunecke K.A., Bendeditto A.P.M. 2011.Growth and recruitment of the Atlantic seabob shrimp Xiphopenaeus kroyeri (Heller, 1862)(Decapoda, Penaeidae) on the coast of Rio de Janeiro, southeastern Brazil. Crustaceana 84 (12-13), 1465-1480.

Fransozo V, Santos DC, López-Greco LS, Bolla EA Jr. 2011. Development of secondary sexual characters in the seabob shrimp Xiphopenaeus kroyeri (Heller, 1862) (Crustacea, Decapoda, Penaeidae): a scanning electron microscope study. Invertebr Reprod Dev 55: 6-15.

Freire A.M. F.,Luchiari A.C., Fransozo V. 2011. Environmental substrate selection and daily habitual activity in Xiphopenaeus kroyeri shrimp (Heller, 1862)(Crustacea:Penaeioidea), Indian Journal of Geo-Marine Science Vol. 40(3), p 325-330.

FroideFond J.M.,Pujos M., Andre X.1988. Migration of mud banks and changing coastline in French Guiana. Marine Geology 84, 19-30.

IUCN. 2015. The IUCN Red List of Threatened Species. Version 2014.3. Downloaded on 24 April 2015.

Garcia S.M., Cochrane K.L 2005. Fisheries and Aquaculture topics. The ecosystem approach to fisheries management. Topics Fact Sheets. In: FAO Fisheries and Aquaculture Department [online] [conseiled24 april 2015]. http://www.fao.org/fishery/topic/13261/en

Gusmão J., Piergiorge R.M. , Tavares C.2013. The contribution of genetics in the study of the seabob shrimp populations from the Brazilian coast . Bol. Inst. Pesca, São Paulo, 39(3): 323 – 338.

GLOBEFISH Highlights. 2014. FAO/GLOBEFISH Highlights. p. 64

Gillett, R. 2008.Global study of shrimp fisheries.FAO Fisheries Technical Paper. No. 475,331p.7

GIWA. 2006. Challenges to International Waters; Regional Assessments in a Global Perspective. United Nations Environment Programme. 1 - 120. 36

Grabowski C.R., Simões S. M., Castilho A. L. 2014. Population structure, sex ratio and growth of the seabob shrimp Xiphopenaeus kroyeri (Decapoda, Penaeidae) from coastal waters of southern Brazil, ZooKeys 457: 253-269.

Heckler G.S., Lopes M., Simões S.M. Shimizu R.M., Da Costa R.C. 2014. Annual, seasonal and spatial abundance of the seabob shrimp Xiphopenaeus kroyeri (Decapoda, Penaeidae) off the Southeastern coast of Brazil. Annals of the Brazilian Academy of Sciences. 83(3), 1337-1346.

Hempel, E., U. Winther & J. Hambrey 2002. Can Shrimp Farming Be Undertaken Sustainably?.World Bank/NACA/FAO/WWF Consortium Program on Shrimp Farming and the Environment. 1-112.

Hu C., Montgomery E.T., Schmitt R.W., Muller-Karger F.E. 2004. The dispersal of the Amazon and Orinoco River water in the tropical Atlantic and Caribbean Sea: Observation from space and S-PALACE floats. Deep-Sea Research II. 1151-1171.

ICES/FAO. 2005. Indonesia. Review of bycatch in world shrimp trawl fisheries. Report of the ICES/FAO Working Group on Fishing Technology and Fish Behaviour, 18–22.

ILVO. 2012. Over Seabobs en Suriname..ILVO-onderzoek naar de rol van de seabobgarnaal in het ecosysteem. [Conseiled on 4 april 2015]. http://www.ilvo.vlaanderen.be/NL/Pers-en-media/Alle-media/articleType/ArticleView/articleId/411/Over-Seabobs- en-SurinamebrILVO-onderzoek-naar-de-rol-van-de-seabobgarnaal-in-het-ecosyteem

IOWA State University. 2015. Deparmtent of Statistics.Handout for Statistics 416: Statistical Design and Analysis of Microarray and RNA-seq Experiments [conseiled on 15 April 2015]. http://streaming.stat.iastate.edu/~stat416/LectureNotes/handout_LOWESS.pdf

Kovacs C., Cox T. 2014. Quantification of Interactions Between Common Bottlenose Dolphins (Tursiops truncatus) and a commercial shrimp trawler near Savannah, Gerogia.Aquatic Mammals, 40, 81-94.

Kuehl S., Nitrrouer C.A., DeMaster D.J.1985. Distribution of sedimentary structures in the Amazon subaqueous delta. Continental Shelf Research, Vol 6, 311-336.

Levin P.S., Möllmann C. 2015. Marine ecosystem regime shifts: challenges and opportunities for ecosystem-based management. Phil. Trans. R. Soc. B 370: 20130275

Leung P.S., Engleeds C. 2006. An overview of global shrimp markets and trade. Shrimp culture: economics, market and trade. Blackwell Publishing,. 1-335.

Levin PS, Möllmann C. 2015. Marine ecosystem regime shifts: challenges and opportunities for ecosystem-based management. Phil. Trans. R. Soc. B 370: 20130275.

Lewis D.D S., Fontoura N.F. 2005. Maturity and growth of Paralonchurus brasiliensis females in southern Brazil (Teleostei, Perciformes, Sciaenidae). J. Appl. Ichthyol. 21, 94–100.

Lopes D.F.C., Peixofo S.R.M, Fredau L F., Silva D E. F. B. 2014. Population biology of seabob shrimp Xiphopenaeus kroyeri (Heller, 1862) captured on the south coast of Pernambuco stat, Northeastern Brazil. Brazilian Journal of Oceanography, 62(4):331-340.

LVV. 2012. Visserij management plan voor suriname de seabob garnalen(Xiphopenaeis kroyeri) visserij. Ministerie van Landbouw, veeteelt en visserij. 1 - 27.

Naipal S. 2010. Coastal Morhopdynamic of Suriname. KULeuven [conseiled on 10 May 2015]. http://www.kuleuven.be/hydr/SurinameCoast/SurinameCoast.htm

Neiva G.D.S., Wise P.J. 1963. The Biology and fishery of the sea bob shrimp of Santos Bay, Brazil, Caribbean Fisheries, 1-9.

Macfadyen, Banks R., Davies R. 2013. Tropical shrimp trawling:Developing a management blueprint and adapting and implementing it in specific countries and fisheries. Marine Policy, 40 25–33.

MALVERN. 1999. Operators Guide Worcestershire, UK., Malvern

Mees, J.1994. The hyperbenthos of shallow coastal waters and estuaries: community structure and biology of the dominant species = Het hyperbenthos van ondiepe kustgebieden en estuaria: gemeenschapsstruktuur en biologie van de dominante soorten. PhD Thesis. Universiteit Gent. Mariene Biologie. Instituut voor Dierkunde. Vakgroep Morfologie, Systematiek en Ecologie: Gent. 212 pp.

MSC fishery Fact Sheet. 2012. Suriname Atlantic seabob shrimp.1-2.

Nuengsigkapian P., Spence C. 2001, Summary of the Reykjavik conference on responsible fisheries in the marine ecosystem, International Institute for Sustainable Development, Vol 61 (1), 1 - 11.

Nitrouer C.A.,Curtin T.B., Master D.J.1986. Concentration and flux of suspended sediment on the Amazon continental shelf. Continental Shelf Research, Vol. 6, 151 to 174.

37

Nittrouer C.A., DeMaster D J. 1986.Sedimentary processes on the Amazon continental shelf: past, present and future research. Continental Shelf Research, Vol 6. 5-30.

Oliveira L J.1991. Biologie et dynamique des populations de la crevette Xiphopenaeus kroyeri (Heller,1862) en Guyane francaise. Phd, 1-213.

Oliveira J.L.,LHomme F.1993.Les caractères différenciateurs des postlarves des espèces Xiphopenaeus kroyeri et Penaeus subtilis. Rev. Hydrobiol.trop. 26(1):71:77.

Oh, C.-W., Hartnoll, R. G. Nash, R. D. M. 1999. Population dynamics of the common shrimp, Crangon crangon (L.), in Port Erin Bay, Isle of Man, Irish Sea. – ICES Journal of Marine Science, 56: 718–733.

Park E., M.Latrubesse E. 2014. Modeling suspended sediment distribution patterns of the Amazon River using MODIS data. Remote Sensing of Environment 147. 232-242.

Pauly, D., Ingles J., Neal R. 1984.Application of shrimp stocks of objective methods for the estimation of growth, mortality and recruitment-related parameters from lenght-frequency (ELEFAN I and II), p 220-234.

Pauly, D and G.R. Morgan.1987. Length Based methods in fisheries research. ICLARM Conference Proceedings, 13 - 468.

Pauly D., David N.1981. ELEFAN I, a basic program for the objective extraction of growth parameters from lenght-frequency data.Berichte der Deutshcen Wissenschaftlichen Kommission für Meeresforschung 28(4):205-211.

Pauly, D. 1984. Fish population dynamics in tropical waters: a manual for use with programmable calculators. ICLARM Studies and Reviews 8, 325 p. International Center for Living Aquatic Resources Management, Manila, Philippines.

Pérez R.A. Willems T., Backer D. A. ,Hostens K. ,Vincx M. 2014. Ecology and dynamics of the Atlantic seabob shrimp in Suriname based on fisheries data and local ecological knowledge. 1-68.

Regional Fishery Bodies summary descriptions. Caribbean Regional Fisheries Mechanism (CRFM). 2013. Fishery Governance Fact Sheets. In: FAO Fisheries and Aquaculture Department [Conseiled on 24 April 2015]. http://www.fao.org/fishery/rfb/crfm/en

R Development Core Team. 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing,Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.orgversion 0.98.1062

Rice. J, Browman I.H. 2014. Where has all the recruitment research gone, long time passing? ICES Journal of Marine Science 71(8), 2293–2299.

Reichart, H. A., FreteyJ. 1993. WIDECAST Sea Turtle Recovery Action Plan for Suriname. UNEP Caribbean Environment Programme, CEP Technical Report No. 24 xiv + 65 pp.

Sakimin C. 2009. First Caribbean Initiative Meeting for the Implementation of the Ramsar Convention on Wetlands, Nature Conservation Division, 1-17.

Sairras C, Bruce E. 2008. Agrarisch potentieel: visserij & aquacultuur. Stichting Planbureau Suriname. 1-47.

Southall T., Pfeiffer N., Singh-Renton S. 2011. MSC sustainable fisheries certification Suriname Atlantic Seabob shrimp. FOOD CERTIFICATION INTERNATIONAL LTD, 1-273.

Strømme T., Sætersdal G. , Gade G., Bianchi G., Zaera D. Alvheim O. and Ullebust H. 1989. Final Report - Surveys of the Fish Resources in the Shelf areas between Suriname and Colombia .1-119.

Szuster B. 2006. Environment and Livelihoods in Tropical 86 Coastal Zones. 1- 13.

Santos C.F. M., Branco J.O., Barbieri E. 2013, Biology and fishery of seabob shrimp in the northeastern states of Brazil without a regulatory period of closure.. Bol. Inst. Pesca, São Paulo, 39(3): 217 – 235.

Simpson G.L.2013. ggvegan: ggplot2 plots for vegan functions. R package version 0.0-1.

Simoes M. S., Costa R.C., Fransozo A;, Castilho A. L, 2010, Diel variation in abundance and size of the seabob shrimp Xiphopenaeus kroyeri (Crustacea, Penaeoidea) in the Ubatuba region, Southeastern Brazil, Anais da Academia Brasileira de Ciências, 82(2): 369-378.

Somma A. 2003. The environmental consequences and economic costs of depleting the world’s oceans. Economic Perspectives. An Electrix Journal of the U.S. Department of State.

Sparre, P., Venema, S.C.1998. Introduction to tropical fish stock assessment. Part 1. Manual. Fao Fisheries Technical Paper. No 306.1,1,Rev 2. 407p.

Sudhakara Rao G. 2011. Turtle excluder device (TED) in trawl nets: applicability in Indian trawl fishery. Indian J. Fish., 58(4) : 115-124.

38

Taylor C. 1962: Growth equations with metabolic parameters. J. Cons. Perm. Internat. l’Explor. de la Mer. 27, 270–286.

TAVARES M. 2002. Shrimps. In: W. Fischer, editor. FAO species identification sheets for fishery purposes: Western Central Atlantic (Fishing Area 31). 251-291 p..

Turtle Expert Working Group. 2007. An assessment of the leatherback turtle in the the Atlantic Ocean. NOAA. Technical Memorandum NMFS-SEFSC-555,116p.

UNEP. 2004..Overfishing, a major threat to the global marine ecology,Environmental Alert Bulletin.4, 1-4.

VON BERTALANFFY, L.1938. A quantitative theory of organic growth (inquiries on growth laws. II. Hum. Biol., v. 10, p. 181- 213.

Voss R., Quaas M.F., Schmidt J.O., Tahvonen O., Lindegren M., Möllmann C. 2014. Assessing Social – Ecological Trade- Offs to Advance Ecosystem-Based Fisheries Management. PLoS ONE 9(9): e107811. doi:10.1371/journal.pone.0107811

Wells, J.T., Coleman, J.M. 1981. Periodic mudflat progradation, northeastern coast of South America: a hypothesis. Mar. Geol. 95, 111–130.

Wickham. H. 2009.ggplot2: elegant graphics for data analysis

Willems T., Depestele J., De Backer A., Hostens K.2013. By-catch of rays in the trawl fishery for atlantic seabob shrimp Xiphopenaeus kroyeri in Suriname: How effective are TEDs and RDs? Ilvo Mededeling, nr 139.1-22.

Winterwerp JC. 2009. Integrated Coastal Zone Management.Deltares.1-4

Zhou S.,Smith A.D.M, Punt A.E., richardson A.J., Gibbs M., elizabeth F.A., Pascoe S., Bulman C., Bayliss P, Sainsbury K.2010 Ecosystem-based fisheries management requires a change to the selective fishing philosophy. PNAS., Vol. 107 | no. 21 | 9485–9489

39