Assessing the potential for mangrove oyster aquaculture in an estuarine system of the southeastern coast of Brazil : a geographic information system approach

Assessing the Potential for Mangrove Oyster Aquaculture in an Estuarine System of the Southeastern Coast of Brazil: A Geographic Information System Approach

by

Gilberto Fonseca Barroso

B.Sc. Honours, Santa ~ r s u l a University, 1988 M.Sc., Federal University of Siio Carlos, 1994

A Dissertation Submitted in Partial Fulfillment of the

Requirements for the Degree of DOCTOR OF PHILOSOPHY in the Department of Geography

O Gilberto Fonseca Barroso, 2004 University of Victoria

All rights reserved. This dissertation may not be reproduced in whole or in part, by photocopying or other means, without the permission of the author.

Supervisor: Dr. Mark Flaherty Co-supervisor: Dr. Jack Littlepage

ABSTRACT

Coastal aquaculture is among the fastest growing sectors of the food producing industry. Bivalve farming is a promising activity in low income countries were mollusk could be cultured under low technology and budget, contributing to reduce poverty and assuring food security. Site selection has been considered as a key process in successful aquaculture developments. A suitability model for mangrove oyster farming in the

PiraquC-aqu/PiraquC-mirim estuarine system - PAPMES (Espirito Santo, Brazil) was developed. The suitability model is based on Multi-criteria Evaluation (MCE) which consists of selecting criteria, define their acceptable and optimum ranges, assign their weights of relative importance,

and

combine suitability criteria under a decision rule. A

georeferenced database was created with 8 water quality variables considered related to the habitat requirements of mangrove oyster, with 19 field sampling campaigns on 6 samplings sites embracing an area of 51 lha. Low salinity and dissolved oxygen levels were detected in the upper estuarine sections. Using the geographic information system (GIs) Idrisi32, point data were converted to continuous surface models using second-

order polynomial fit. The normalization process aimed at standardizing the set criteria considering a single scale ranging from low (i.e., 0) to high suitability (i.e., 255). Through pairwise comparison technique weights were assigned to each criteria. Salinity and dissolved oxygen were considered the most important criteria because of their relationship to oyster short-term survival. A weighted linear combination and two

constraints (i.e., fecal coliform > 43 MPN1100mL and navigation channel) were applied as the MCE decision rule. An area of 75ha (14.6% of the PAPMES) was considered

constrained for mangrove oyster fanning. Two suitability models were performed using average and low salinity values. Suitability maps developed onto the 0

-

255 range were reclassified in 4 categories: unsuitable, moderately suitable, suitable, and very suitable. In both models, no area was indicated as unsuitable. Although the low salinity model could

be considered more restrictive, it yielded a very suitable area 26% larger than the average

salinity model. The combination of the two models could bring together risk taking and risk-averse perspectives, respectively. The output of such combination is a map locating 80ha of very suitable areas for mangrove oyster farming, with 9.5ha preferentially designated for intertidal farming using racks. Aquaculture zones are discussed in terms of their interactions with other systems at higher spatial scales, such as the watershed and the coastal zone. GIs can serve as an integrative environment to integrate complex variables in multiple scales. It is only through its integration in multisectoral development plans and programs for the watershed and coastal zone realms that coastal aquaculture will be recognized as sustainable enterprise.

Table of Contents

Abstract Table of Contents List of Table List of Figures List of Acronyms Acknowledgements

. .

11 iv vi ix xvi xviii

1.1 NATURE OF THE PROBLEM

1.2 PURPOSE OF THE STUDY

1.3 OUTLINE OF DISSERTATION

2.1 COASTAL AQUACULTURE OVERVIEW 10

2.1.1 Global Perspective 10

2.1.2 Brazilian Perspective 22

2.2 DECISION SUPPORT APPROACHES FOR AQUACULTURE DEVELOPMENT 34

2.2.1 Site Management 3 6

2.2.2 Regional Planning 37

2.3 COASTAL AQUACULTURE SUITABILITY ANALYSIS: THE ADVANCE OF

GEOGRAPHIC INFORMATION SYSTEMS 43

2.4 SUMMARY 51

CHAPTER 3 BIOPHYSICAL CRITERIA FOR ANGRO GROVE OYSTER FARMING SUITABILITY 3.1 SALINITY 3.2 TEMPERATURE 3.3 TURBIDITY 3.4 PH 3.5 DISSOLVED OXYGEN

3.6 SUSPENDED SEDIMENTS AND ORGANIC SESTON

3.7 CHLOROPHYLL

a

3.8 FECAL COLIFORM BACTERIA 3.9 SUMMARY

4.2 THE ESP~RITO SANTO COAST

4.3 STUDY SITE: THE PIRAQU~~-A~U/~IRAQU~~-MIRIM ESTUARINE SYSTEM 4.3.1 PAPMES Watershed

4.3.2 Human Settlements 4.3.3 Estuarine Resource Uses 4.3.4 Resources Conservation

4.3.5 Coastal Resource Uses and Conflicts

4.4 DATA COLLECTION 4.4.1 The estuarine realm 4.4.2 Water sampling program 4.4.3 Field and laboratory methods

4.5 PAPMES GEOGRAPHIC INFORMATION SYSTEM DEVELOPMENT 4.5.1 Spatial Data

4.6 SUMMARY

CHAPTER 5 BIOPHYSICAL CRITERIA SAMPLING RESULTS

5.1 SALINITY

5.2 TEMPERATURE

5.3 TURBIDITY

5.4 PH

5.5 DISSOLVED OXYGEN

5.6 SUSPENDED SEDIMENTS AND ORGANIC SESTON

5.7 CHLOROPHYLL A

5.8 E"ECAL COLIFORM BACTERIA

5.10 SUMMARY

6.1 CONTINUOUS SURFACE MODEL OF BIOPHYSICAL CRITERIA 132 6.1.1 Classification of PAPMES in salinity zones 159

6.2

Fuzzy

MEMBERSHIPS 161

6.3 MULTIPLE-CRITERIA EVALUATION (RICE) 167

6.4 SUITABILITY EVALUATION UNDER LOW SALINITY CONDITIONS 177

6.5 SUMMARY 185

7.1 IMPLICATIONS

7.2 SUMMARY

CHAPTER 8 CONCLUSIONS REFERENCES

List of Tables

Table 2.1: Relative contributions of fisheries and aquaculture to world fisheries production. Modified from FA0 (2002).

Table 2.2: World aquaculture production (lo6 tons) by group of organisms in 1996, 2000, and 2001, and increment rates for 1996 - 2001 and 2000 -

200 1 .Data from FA0 Fishstat aquaculture production quantities (FAO, 2000). Table 2.3: Bivalve mollusks yield in 1991 and 2001 (lo3 tons) from aquaculture, fisheries and total production. Values in parenthesis in 2001 aquaculture column represent production as percentage of the combined annual yields from world aquaculture and fisheries. Values in parenthesis in 2001 China's aquaculture column represent production as percentage of the combined annual yields from world aquaculture and fisheries. Data from FA0 Fishstat databases (1 950-200 1): aquaculture production quantities, total captures, and total production [FAO, 2000 #178].

Table 2.4: Main species of cupped oysters and their total yields in 1997. Modified from Spencer (2002).

Table 2.5: Brazilian aquaculture production (tons) by group of organisms in 1996,2000, and 200 1. Values in parenthesis in 199 1,2000 and 200 1 columns represent the relative contribution of group of farmed organisms to total production. Increment rates for 1996 - 200 1 and 2000 - 200 1 are presented in

percentage. Data from FA0 Fishstat aquaculture production quantities (FAO, 2000).

Table 2.6: Management strategies to reduce problems with oyster farms

in

Korea (modified from Phillips, 1998).

Table 2.7: Application of computer tools in aquaculture enterprises (Ernst and Nath, 2000).

Table 2.8: Minimum information required for a database on aquaculture at the site level. Compiled from Chua (1997).

Table 2.9: Spatial hierarchies for water resource management in aquaculture operations (LeBlanc, 1989; Muir, 1996).

Table 2.10: Applications of geographic information system in aquaculture planning and management

Table 3.1: Suitable and optimum ranges for selected biophysical criteria of mangrove oyster farming.

PAR and PMR watersheds*. Data from the 2001 demographic census (IBGE www.ibge.~ov.br).

Table 4.2: Special use areas within the PAPMES. Table 4.3 : Sampling sites characteristics.

Table 4.4: Sampling schedule and tidal phases. Table 4.5: Resumed field and laboratory methods.

Table 4.6: Summary of resample transformation of coordinates.

Table 5.1: Descriptive statistics for salinity (%o) at PAPMES sampling sites. Table 5.2: Difference in salinity values between surface and bottom waters

(%d.

Table 5.3: Descriptive statistics for temperature ("C) at PAPMES sampling sites.

Table 5.4: Descriptive statistics for turbidity (NTU) at PAPMES sampling sites.

Table 5.5: Descriptive statistics for pH at PAPMES sampling sites.

Table 5.6: Descriptive statistics for the concentration of dissolved oxygen (mgh) at PAPMES sampling sites.

Table 5.7: Descriptive statistics for suspended sediments (mg/L) at PAPMES sampling sites.

Table 5.8: Descriptive statistics for organic seston (% of suspended sediments) at PAPMES sampling sites.

Table 5.9: Descriptive statistics for Chlorophyll a (pg/L) at PAPMES sampling sites

Table 5.10: Descriptive statistics for fecal coliform (MPNI100mL) at PAPMES sampling sites.

Table 6.1: Interpolation methods for gradual continuous surface models (modified fiom Borrough, 1986).

Table 6.2: Salinity values (%o) extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median salinity of sampling sites (i.e., sites 1,2,3,4,5, and 6).

sites a, b and c) through linear regressions of site distances and actual median temperature of sampling sites (i.e., sites 1,2, 3,4,5, and 6).

Table 6.4: Turbidity values (NTU) extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median turbidity of sampling sites (i.e., sites 1,2,3,4,5, and 6).

Table 6.5: pH values extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median pH of sampling sites (i.e., sites 1,2,3,4,5, and 6).

Table 6.6: Dissolved oxygen values (mg/L) extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median dissolved oxygen of sampling sites (i.e., sites 1, 2, 3, 4, 5, and 6)-

Table 6.7: Suspended sediment values (mg/L) extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median suspended sediments of sampling sites (i.e., sites 1, 2, 3, 4, 5, and 6).

Table 6.8: Chlorophyll a values (pg/L) extrapolated for dummy data points (i.e., sites a, b and c) through linear regressions of site distances and actual median chlorophyll a of sampling sites (i.e., sites 1,2,3,4,5, and 6).

Table 6.9: Fecal coliform values (MPNI100mL) extrapolated for dummy data

points (i.e., sites a, b and c) through linear regressions of site distances and actual median fecal coliform of sampling sites (i.e., sites 1,2,3,4,5, and 6). Table 6.10: Statistics for the linear and quadratic polynomial surface fits of the trend surface analysis on the biophysical attributes in PAPMES.

Table 6.1 1: Median values of measured biophysical criteria, their predicted results from continuous surface models and differences (%) between them. Table 6.12: Continuous rating scale

Table 6.13: Painvise comparison matrix.

Table 6.14 Resulting weights and their relative importance.

Table 6.15: Salinity values (%) values extrapolated for dummy data points (i.e., sites a, b and c) trough linear regressions of sampling site (i.e., sites 1, 2, 3,4,5, and 6) values of 20 December, 2000 and distance between sites.

List of Figures

Figure 2.1 : World aquaculture production per group of organisms in 200 1. Data according to FA0 Fishstat aquaculture production quantities: 1950-2001 (FAO, 2000).

Figure 2.2: Proportion of world aquaculture production by group of organisms and environments in 2001: a) freshwater aquaculture; b) brackish aquaculture; c) marhe aquaculture. Data from to FA0 Fishstat aquaculture production quantities (FAO, 2000): 1950-2001.

Figure 2.3: Trends of world aquaculture production by major groups of organisms. a) freshwater aquaculture; b) brackish aquaculture; c) marine aquaculture. Data from FA0 Fishstat aquaculture production quantities [FAO, 2000 #178]: 1950-2001.

Figure 2.4: Brazil's aquaculture production by groups of organisms in 2001. Data from FA0 Fishstat aquaculture production quantities (FAO, 2000): 1950- 2001.

Figure 2.5: Proportions of Brazil's aquaculture production by group of organisms and environments in 2001: a) freshwater aquaculture; b) marine aquaculture. Data according to FA0 Fishstat aquaculture production quantities (FAO, 2000): 1950-2001.

Figure 2.6: Trends of Brazil's aquaculture production by major groups of organisms. Data according to FA0 Fishstat aquaculture production quantities:

1950-2001 (FAO, 2000).

Figure 2.7: Trend of Brazil's aquaculture production of marine mollusks from 1968 to 1994, and cupped oyster (Crassostrea spp) and the American rock mussel (Perna perna) production from 1995 to 200 1. Data from FA0 Fishstat aquaculture production quantities (FAO, 2000).

Figure 2.8: Flowchart of the leasing process for allocation of public areas for aquaculture purposes. FDA: Federal Delegacy of Agriculture; DFA: Department of Fisheries and Aquaculture; MA means Ministry of Agriculture. Figure 4.1: The Brazilian coast and its characteristics in geomorphology, tides, currents, climate, and vegetation. Modified from Ekau (1999).

Figure 4.2: Espirito Santo coastal sectors according to the state Coastal Zone Management Program (IJSN, 200 1).

Figure 4.3: A set of 10,000 seeds of C. rhizophorae purchased from the IBAMA

Estuarine Aquaculture Station (Sergipe) for farming in Conceiqgo da Barra and

Figure 4.4: Raft method for mangrove oyster farming in Cricare estuary (Conceiqiio da Barra, ES) (Photo courtesy of H. K. de Andrade).

Figure 4.5: Location of the PiraqubaquPiraquC-mirim Estuarine System (Aracruz, ES).

Figure 4.6: Perspective views of the PAPMES. A) landward; b) seaward Figure 4.7: Tide curve of Barra do Riacho Port (Aracruz, ES).

Figure 4.8: PAPMES aquaculture practices (Aracruz, ES).

Figure 4.9: Watersheds of Piraqus-ap and PiraquC-mirim estuarine system. Figure 4.10: Sampling locations and in-water distances.

Figure 4.11: Correlation between tide curve (Barra do Riacho Port, Aracruz, ES) and surface and bottom water temperatures at the lower PiraquC- aqu/P.mirim estuarine system. Red dots show sampling event.

Figure 4.12: Cartographic modeling of airphoto mosaic joining, georeferencing, and contracting procedures.

Figure 4.13: Aerial photographs mosaic georeferenced in UTM coordinates

(datum WGS84).

Figure 4.14: Cartogra hic modeling of shoreline, mangrove, and roads vectors Q

! creation in CartaLinx

.

Figure 5.1: Box-and-whisker plot of salinity (%) of sampling sites in the PAPMES. The inner horizontal line is the median, the top and bottom of the box are the 25' and 75th percentiles (quartiles), and the ends of the whiskers are the 5th and 95' percentiles. N is the number of observations, and the symbols and represent the outlier and the extreme values, respectively.

Figure 5.2: Box-and-whisker plots of salinity (%o) of the sampling sites according to sampling depths: a) site 1, b) site 2, c) site 3, d) site 4, d) site 5, and e) size 6. The inner horizontal is the median, the top and bottom of the box are the 2sth and 75' percentiles (quartiles), and the ends of the whiskers are the 5' and 95th percentiles. N is the number of observations, and the symbols and represent the outlier and the extreme values, respectively. The text in red shows the date and tide cycle in which outlier and extreme values were observed.

Figure 5.3: Box-and-whisker plot of temperature ("C) of sampling sites in the PAPMES. The inner horizontal line is the median, the top and bottom of the

box are the 25th and 75th percentiles (quartiles), and the ends of the whiskers are the 5" and 95" percentiles. N is the number of observations.

Figure 5.4: Relationship between temperature and salinity.

Figure 5.5: Box-and-whisker plots of turbidity (NTU) of the sampling sites. The inner horizontal is the median, the top and bottom of the box are the 25th and

75" percentiles (quartiles), and the ends of the whiskers are the 5th and 95th

percentiles. N is the number of observations, and the symbols and represent the outlier and the extreme values, respectively.

Figure 5.6: A tidal intrusion front at site 2 on 20 December 2000. The salty and less turbid water is flowing in from the right of the foam and the less salty and more turbid water flowing out from the left. Turbidity at 1.5m depth reached

5 1.1 NTU when this picture was taken.

Figure 5.7: Relationship between turbidity and salinity.

Figure 5.8: Box-and-whisker plots of pH at of the sampling sites. The inner horizontal is the median, the top and bottom of the box are the 25" and 75'h percentiles (quartiles), and the ends of the whiskers are the 5th and 95th percentiles. N is the number of observations, and the symbols and represent the outlier and the extreme values, respectively.

Figure 5.9: Box-and-whisker plots of dissolved oxygen (mg/L) at the sampling sites. The inner horizontal is the median, the top and bottom of the box are the

25" and 75" percentiles (quartiles), and the ends of the whiskers are the 5th and

95" percentiles. N is the number of observations, and the symbols represent

the outlier values.

Figure 5.10: Relationship between dissolved oxygen and salinity.

Figure 5.1 1 : Relationship between dissolved oxygen and water temperature. Figure 5.12: Box-and-whisker plots of suspended sediments (mgL) at the sampling sites. The inner horizontal is the median, the top and bottom of the box are the 25th and 75" percentiles (quartiles), and the ends of the whiskers are the 5" and 95" percentiles. N is the number of observations, and the symbols represent the outlier values.

Figure 5.13: Relationship between suspended sediments and turbidity.

Figure 5.14: Box-and-whisker plots of organic seston (%) at the sampling sites. The inner horizontal is the median, the top and bottom of the box are the 25" and 75" percentiles (quartiles), and the ends of the whiskers are the 5" and 95th percentiles. N is the number of observations, and the symbols represent the outlier values.

Figure 5.15: Relationship between suspended sediments and organic seston. 123

Figure 5.16: Box-and-whisker plots of Chlorophyll a (pg/L) at the sampling 125 sites. The inner horizontal is the median, the top and bottom of the box are the

25th and 75th percentiles (quartiles), and the ends of the whiskers are the 5" and

95' percentiles. N is the number of observations, and the symbols represent

the outlier values.

Figure 5.17: Correlation between chlorophyll a and salinity. 125

Figure 5.18: Correlation between chlorophyll a and turbidity. 126

Figure 5.19: Correlation between chlorophyll a and suspended sediments. 126

Figure 5.20: Correlation between chlorophyll a. and organic seston. 126

Figure 5.21: Box-and-whisker plots of fecal coliform (MPNI100mL) at the 128 sampling sites. The inner horizontal is the median, the top and bottom of the box are the 25' and 75' percentiles (quartiles), and the ends of the whiskers are the 5th and 95th percentiles. N is the number of observations, and the symbols represent the outlier values.

Figure 5.22: Correlation between fecal coliform and salinity. 129

Figure 5.23: Correlation between fecal coliform and turbidity. 129

Figure 6.1: In-water distances between adjacent sampling sites and dummy data 138

points. Sites 1 to 6 are the sampling sites and sites a to c are the dummy data

points.

Figure 6.2: Steps for the creation of an ASCII file with salinity values for trend 139 surface analysis

Figure 6.3: Linear regressions applied to derivate salinity values (%o) for 140 dummy data points. a) PAR estuary, and b) PMR estuary

Figure 6.4: Linear regressions applied to derivate dummy data points for 141 temperature values (OC) for dummy data points. a) PAR estuary, and b) PMR estuary.

Figure 6.5: Linear regressions applied to derivate turbidity values (NTU) for 142 dummy data points. a) PAR estuary, and b) PMR estuary.

Figure 6.6: Linear regressions applied to derivate pH values for dummy data 143 points. a) PAR estuary, and b) PMR estuary.

(mg/L) for dummy data points. a) PAR estuary, and b) PMR estuary.

Figure 6.8: Linear regressions applied to derivate suspended sediments values 145 (mg/L) for dummy data points. a) PAR estuary, and b) PMR estuary

Figure 6.9: Linear regressions applied to derivate chlorophyll a values (pg/L) 146 for dummy data points. a) PAR estuary, and b) PMR estuary.

Figure 6.10: Linear regressions applied to derivate fecal coliform values 147 (MPNI100mL) for dummy data points. a) PAR estuary, and b) PMR estuary

Figure 6.11: Sequence of procedures for the creation of the ASCII file 148 associating sampling and dummy data points (dummy sites) coordinates and

salinity values.

Figure 6.12: Cartographic modeling for the generation of continuous surface 149 model of salinity.

Figure 6.13: Cartographic masking of land and ocean the continuous surface 153 model for salinity.

Figure 6.14: Cartographic modeling for fecal coliform (MPNI100mL). 154 Figure 6.15: Continuous surface models for biological criteria attributes: a) 155 salinity (%o); b) temperature PC).

Figure 6.16: Continuous surface models for biological criteria attributes: a) 156 turbidity (NTU); b) pH.

Figure 6.17: Continuous surface models for biological criteria attributes: a) 157 dissolved oxygen (mg/L); b) suspended sediments (mg/L).

Figure 6.18: Continuous surface models for biological criteria attributes: a) 158 chlorophyll a ( p a ) ; b) fecal coliform (MPNIlOOrnL).

Figure 6.19: Cartographic modeling of salinity zones of PAPMES according to 160 the Venice System (Kmmer et al., 1994).

Figure 6.20: Salinity zones in the PAPMES, according to the Venice System. 161 Figures 6.21: Fuzzy set memberships of biophysical criteria attributes: a) 162 Salinity; b) Temperature; c) Turbidity; d) pH; e) Dissolved oxygen; e) Suspended sediments; chlorophyll a; f ) Fecal coliform.

Figure 6.22: Fuzzy membership maps. a) salinity; b) temperature. 163 Figure 6.23: Fuzzy membership maps. a) turbidity; b) pH. 164

xiv

Figure 6.24: Fuzzy membership maps. a) dissolved oxygen; b) suspended sediments.

Figure 6.25: Fuzzy membership maps. a) Chlorophyll a; b) fecal coliform.

Figure 6.26: Development of a final constraint map.

Figure 6.27: Final constraint map for Multi-criteria Evaluation.

Figure 6.28: Cartographic modeling for the Multi-criteria Evaluation procedure. Figure 6.29: Mangrove oyster farming suitability.

Figure 6.30: Mangrove oyster farming suitability per classes.

Figure 6.31: Linear regressions applied to derivate salinity values from December 2000 sampling for dummy data points. a) PAR estuary, and b) PMR

estuary.

Figure 6.32: Cartographic model of low-salinity continuous surface (sampling campaign of 20 December, 2000).

Figure 6.33: Continuous surface of low salinity values (%).

Figure 6.34: Low-salinity zones in the PAPMES, according to the Venice System.

Figure 6.35: Fuzzy membership map of low-salinity.

Figure 6.36: Mangrove oyster farming suitability for low-salinity conditions. Figure 6.37: Classes for mangrove oyster farming suitability in low-salinity conditions.

Figure 7.1: Cartographic modeling of final suitability for mangrove oyster farming at the PAPMES.

Figure 7.2: Selected areas form mangrove oyster fuming in the PAPMES. Figure 7.3: Selected area for mangrove oyster culturing on racks.

Figure 7.4: Sampling sites and tracks (i.e., fishing tracks) of the EEQBA Project in the PAPMES.

Figure 7.5: Mangrove oyster farming in the entrance of the PAPMES. a) location of the experiment in the lower part of the PAPMES; b) oyster seeds

brought from Conceiqiio da Barra; c) fouling of lantern nets before the manual cleanup; d) lantern nets drifting on tidal peak flows; e) measuring oyster shells with a caliper; f) oysters removed form raft floats.

Figure 7.6: Levels of integration of estuarine bivalve farming over a broad 198

AHP ASCII BA BMLP BMP DBOs 20•‹c C. brasiliana C. gigas C. rhizophorae C. virginica CIDA CIRM

c m

CZM CTA DPA dpi DSP EBS EPE EQBA ES GBM GCPs GERCO GIs GPS HEP HES IBAMA ICZM INCAPER IPES IT CNPq LME MA MCM MMA MCT MG MPN List of Acronyms

Analytical Hierarchy Process

American Standard Code for Information Interchange State of Bahia

Brazilian Mariculture Linkage Program Windows Bitmap File

Biochemistry oxygen demand Crassostrea braziliana

Crassostrea gigas Crassostrea rhizophorae Crassostrea virginica

Canadian International Development Agency Interministerial Commission for Ocean Resources

National Council for Scientific and Technological Development Coastal Zone Management

Aquaculture Technological Center Department of Fisheries and Aquaculture dots per inch

Diarrheic Shellfish Poisoning East Brazil Shelf

Estimated Positioning Errors

Environmental Quality and Biodiversity Assessment State of Espirito Santo

Georeferenced Bitmap Metadata Ground Control Points

Coastal Zone Management Program Geographic Information System Global Positioning System Habitat Evaluation Procedure Habitat Evaluation System

Brazilian Institute for the Environment and Renewable Natural Resources

Integrated Coastal Zone Management

ES State Institute of Research, Technical Assistance, Rural and Extension

Institute Jones dos Santos Neves for Support of Research and Development

Information Technology

National Council for Scientific and Technological Development Large Marine Ecosystem

Ministry of Agriculture Multi-criteria Modeling

Ministry for the Environment, Water Resources, and the Amazon Ministry of Science and Technology

State of Minas Gerais

xvii NBS NMEA NGO OWA PAH PAPMES PAR P. perna PMR PSP QHEI RAM RECOS

RJ

RMS RST RDC SDSS SBS SEAMA SEBRAE-ES SHE TIFF USA UFES UTM WLC WGS84 WET

North Brazilian Shelf

National Marine Electronics Association Non-governmental Organization

Order weighted average

Polycyclic aromatic hydrocarbon

Piraque-ap and Piraqu&mirirn Estuarine System Piraque-ac;u River

Perna perna

Piraqu&miritn River

Paralytic Shellfish Poisoning

Qualitative Habitat Evaluation Index Random Access Memory

Use and Appropriation of Coastal Zone Resources Rio de Janeiro

Root Mean Square Idrisi32 raster image file

Idrisi32 raster image documentation file Spatial Decision Support System South Brazil Shelf

Espirito Santo State Environment Secretariat

Brazilian Micro and Small Business Support Service - Espirito

Santo Branch

Simplified Habitat Evaluation Tagged Image File Format United States of America

Federal University of Espirito Santo

Universal Transverse Mercator referencing system Weighted linear combination

World Geodetic System 1984 Wetland Evaluation Technique

xviii

Acknowledgements

Acknowledgement are not that easy when you have to keep in mind everyone that somehow managed to give you some support, suggestions, help, and encouragements. There were so many experiences that became good memories that I'm glad to have had these moments to remember them all.

I must start with Dr. Jack Littlepage, Director of the Brazilian Mariculture Linkage Program (BMLP), and Dr. Mark Flaherty, faculty of the Department of Geography. They opened up the opportunity for me to come to UVic as a Ph.D. candidate of the Geography Graduate Program. I'm also deeply thankful for their guidance in the development of this study as my co-supervisor and supervisor, respectively.

Also, many thanks to Patricia Summers, BMLP Manager, for all the support and help provided during the development of this study. The Canadian International Development Agency (CIDA), through the BMLP, provided much needed financial support as a one year student fellowship and resources. Also, I would like to acknowledge the Brazilian Federal Government for the fellowship provided by CAPES (ref. #2229/98).

Deeply appreciated Dr. Dave Duffus' insights on the interactions in coastal systems and the needs of management, and also for good suggestions regarding the interpolation of point data. The course of Geographic Information System with Dr. Peter Keller was very challenging, particularly in regard to command-driven ArcInfo for UNIX. The motivation and encouragement environment in Dr. Keller's classes made possible to get, at the same time, a good idea about what GIs is all about.

My appreciation also goes to Dr. Jarbas Bonetti Filho for the motivation and insights about the development of this study.

I'm very thankful to Dr. Jean Christophe Joyeux (Department of Ecology and Natural Resources, UFES), who reviewed the text and provided good suggestions during the writing part. Thank you very much my friend!

I would like to express my gratefulness to the Department of Geography of the University of Victoria for accepting me as a student and for the very pleasant

xix environment.

I can not forget all members of the BMLP Board of Directors Iracema do Nascimento, Graco, Francisco Seixas, Teresinha, Edrnar Andreatta, William Pennell, Marco Cutrin, Anthony Dickinson, Miguel Accioly and Angelica for the very pleasant times.

Many thanks to my good friends Veronica Armstrong, Roy Diment, Tristan Armstrong, Ruth Paul, Maycira Costa, and Kevin Kelmer who make life funnier. To the Geography colleagues and friends John Corbert, Laura Johnson, Scott Kelman, Brian Szustzer, Paul Miller, Cimarron Corpe, Robert Braaten, and Dologobinda Pradhan, and Marut Diparos, just to name a few. I'm very pleased to have met you.

For my colleagues of the Department of Ecology and Naturals Resources and especially to Federica Natasha G. Sodrk, Ricardo Gomide, Kamila Perin, Emmanoel Pinheiro, gratitude for all help.

Thanks to my family, for its endless support while I was overseas and during the

development of this study.

And finally, my most profoundly acknowledgements to my wife Alessandra Delazari-Barroso, who always provided me with the motivation, support and inspiration needed and for being so special in my life.

1.1 NATURE OF THE PROBLEM

1.2 PURPOSE OF THE STUDY

1.1 NATURE OF THE PROBLEM

Producing food in a sustainable manner is becoming a major challenge for nations throughout the world. In the fisheries sector, stagnating yields fiom many capture fisheries and increasing demand for fish and fishery products, suggest that production will be inadequate to meet the needs of the world's population without supplementation through aquaculture (Davenport et al., 2003). Aquaculture can be defined as the management of confined species in more-or-less controlled systems (Muir and Nugent, 1997; Barton and Staniford, 1998). It now encompasses a wide range of different species

-

algaelseaweed, molluscs (oysters, mussels, scallops and clams), crustaceans (shrimp and crabs) and fish, and farm systems ranging fiom low intensity, small-holder pond and cage culture, through to large-scale, capital intensive, high technology farms that are owned by multi-national corporations. All of these various cultured products and associated production systems have distinct natural resource use patterns, requirements, and environmental impacts (Odum, 1974; Chua et al., 1989; Chua, 1992; Chua, 1993; Beveridge et al., 1994; Landesman, 1994; Rosenthal, 1994; Hastings and Heinle, 1995; Muir, 1996; Tovar et al., 2000; Black, 2001).

Aquaculture is not a new human enterprise, but is believed to have started perhaps 5,000 years ago when humans learned to entrap fish in areas where they could thrive and be harvested as needed (Beveridge and Little, 2002). The ownership and protection of stocks, through the use of cages, pens, and enclosures, and the supplementation of food, fertilization with inorganic and organic materials, provided a certain degree of independence fkom wild fish stocks. This process was consolidated with the domestication of f m e d species through selective breeding of captive specimens. Historically, small-scale, low technology aquaculture operations were developed as a means to supplement local diets (Gadgil and Berkes, 1991; Beveridge and Little, 2002). Today, intensive aquaculture operations are now viewed as an important means of earning foreign currency through export (Fridley, 1996; Howarth, 1996; Weber, 1 996; Anderson and Fong, 1997; Muir and Nugent, 1997; Eastman, 1999; Bastian, 2000; Brummett, 2003). Aquaculture is now one of the fastest growing food production systems in the world (Howarth, 1996). FA0 statisticians have estimated that world aquacultural output will surpass beef production by 2010 (Bastian, 2000).

Contemporary aquaculture systems are highly productive compared to the traditional ones owing to increasing intensification. The development of modern aquaculture has been described by some analysts as being equivalent to the "green revolution" in agriculture. The term "blue revolution" is used to describe the development of management techniques that increase production with high subsidy inputs (Coull,

1993; Weber, 1996; Costa-Pierce, 2002). Highly subsidized aquaculture operations are

characteristic of systems that are managed to produce economically valuable species such as carnivorous finfish (e.g., salmon) and crustaceans (e.g., prawn and shrimp). With the advent of new technologies for aquaculture intensification and greater access to international markets because of trade liberalization, however, aquaculture is now concentrating on growing organisms fi-om higher trophic levels or "farming up the food web" (Naylor et al., 2000; Davenport et al., 2003).

The growing intensification of aquaculture has raised many concerns over the social and environmental impacts of modern aquaculture, and has called into question the long-term sustainability of these production systems (Folke and Kautsky, 1989; Folke and Kautsky, 1992). The species selected for cultivation and the cultivation intensity largely determine the nature and extent of the environmental impacts associated with aquaculture operations. The dependence on external ecosystems for production and waste assimilation generally decrease from intensive to extensive techniques, and from high trophic level species to species of lower trophic levels (Beveridge et al., 1994; Beveridge et al., 1997; Olsen, 2002). Globally, the growth of intensive production systems for the high valued crops of shrimp and salmon has generated the largest social and environmental criticism against aquaculture. Intensive shrimp farming, for example, has generated a wide range of social and environmental problems including the habitat loss owing to the conversion of mangrove ecosystems to shrimp ponds (Mahamood et al., 1994; Flaherty and Karnjanakesorn, 1995; Dieberg and Kiattisirnkul, 1996; Farnsworth and Ellison, 1997; Bhatta and Bhat, 1998; Primavera, 2000; Carvalho Filho, 2001; Lebel et al., 2002; Ronnback et al., 2002). The production of wastes deteriorate the water quality through the release of uneaten food, fecal and urinary wastes (Chua et al., 1989; Wu et al., 1994; Tsutsurni, 1995; Costa-Pierce, 1996; Panchang et al., 1997; Nunes and Parsons, 1998;

Philippines (Primavera, 1991), Thailand, Ecuador, Honduras and Nicaragua (Currie, 1994). Salmon farming has attracted international attention owing to concerns over the production of organic waste, the release of exotic species, and consumer health (Folke,

1988; Aure and Stigebrandt, 1990; Folke and Kautsky, 1992).

The ongoing controversy over the environmental impacts of aquaculture has raised serious questions about the role that aquaculture can play in enhancing the welfare of coastal people in developing countries. The degradation of coastal ecosystems by intensive production systems and the transfer of the associated social and economic costs to the rural poor, being the main issues (Brzeski and Newkirk, 1997; Pillay, 1997; Karuppiah, 1998; Burbridge et al., 2001; Lebel et al., 2002; Katranidis et al., 2003). There is much more to the aquaculture industry, however, than the culture of shrimp or salmon. Farming of lower trophic level organisms such as marine macroalgae and filter-feeders bivalve mollusks, for example, only requires a marine support area with available inorganic nutrients and primary production, respectively, yet can provide income, employment and food for local communities (Newkirk, 1996; Brzeski and Newkirk, 1997; Williams, 1997; Folke, 1998; Ronnback et al., 2002). In the case of bivalve farming, it relies essentially on good water quality in sheltered areas, with adequate hydrodynamics in order to provide a continuous flow of food (i.e., phytoplankton and suspended organic matter) as well as to dilute and disperse organic material (i.e., feces) produced. Indeed, the farming of benthic herbivore/omnivore animals has been identified as a more sustainable way of nutrient utilization, as it contributes to the closing of biological nutrient cycles (Olsen, 2002). Given the many problems that have emerged with the intensive culture of higher trophic level species, farming lower levels of the food chain may be a good means for coastal communities to produce food for domestic consumption and enhance local incomes over the long-term (Erkom and Griffiths, 1990; Matthiessen, 1991; Brzeski and Newkirk, 1997; Williams, 1997; Phillips, 1998; Olsen, 2002; Ronnback et al., 2002; Small, 2002; Gosling, 2003).

Culture systems for lower trophic level organisms are not entirely problem free. They may cause a local reduction in primary standing crop, and changes in the cycling of the principal nutrients, causing modifications in the structure and functioning of the coastal aquatic ecosystem (Haven and Morales-Alamo, 1966; Odum, 1974; Kautsky and

Evans, 1987; Grenz, 1993; Smaal and Prins, 1993; Haamer, 1996; Kaiser et al., 1998; Crawford, 2003). Biodeposition of organic wastes (feces and pseudo-feces) from bivalve culture in low hydrodynamic environments may also cause local modifications in the benthic community underneath the farming structures. As well, the introduction and transfer of marine molluscs for fisheries and aquaculture brings the risk of transporting competitors, predators, parasites, pests and diseases (Carlton, 1996; Carlton and Mann,

1996; Minchin, 1996). The sustainability of bivalve culture is constrained by many

factors, including the availability of food and spatial and temporal variations in local environmental conditions (Fernhdez-Pato, 1989; Pillay, 1 990; Avault, 1996; Spencer,

2002; Gosling, 2003).

Ecosystems have a limited ability to support aquaculture development. This has lead to growing interest in the application of the concept of carrying capacity to site evaluation and planning (Incze et al., 1981; Dankers, 1993; Heral, 1993; Grant, 1996; Dowd, 1997; Dame and Prins, 1998; Small, 2002; Bacher et al., 2003). According to Heral (1993), carrying capacity models attempt to identify the culture areas and production conditions under which bivalve mollusks can be sustainably farmed to obtain the maximum economic benefit. Overexploitation of the system's capacity to provide dissolved oxygen and to disperse assimilative wastes, for example, is the result of poor management strategies often brought on by stock overcrowding (Mori, 1987). Continual mismanagement of aquaculture systems will promote degradation of rearing grounds (Sugawara and Okoshi, 1991). As a result, reduction of growth rates and even mass mortalities of stock potentially may occur which compromises sustainability (Ferreira et al., 1998). Hence, while exceeding carrying capacity limitations in resource quantity or environmental capacity may be tolerable, or even unnoticed the short-term, it may eventually lead to unacceptable effects and project failure (Beveridge et al., 1994).

In practical terms, bivalve carrying capacity can be understood as the total bivalve biomass supported by a given ecosystem which is a function of the water residence, primary production and bivalve clearance rate (Dame and Prins, 1998). Carrying capacity models must consider the bioenergetics of the cultured species, including filtration rates, particle rejection and selection, food composition and total particulate load as well as absorption efficiency related to food type load (Rosenberg, 1983; Bayne et al., 1989;

Hickman et al., 1991; Navarro et al., 1991; Grant et al., 1993; Smaal and Prins, 1993; Cho et al., 1996; Grant, 1996; Ball et al., 1997; Smaal and Haas, 1997; Grant and Bacher, 1998; Pitcher and Calder, 1998; Smaal et al., 1998). It is extremely difficult to develop generic carrying capacity models owing to inherent spatiotemporal variability in environmental conditions such as hydrographic, meteorologic, and seston parameters. In addition, each target species has its own set of physiological ranges and optimum rates. Thus, site specific information is required for modeling forcing functions.

Inadequate site planning and management can create ecological problems and cause adverse economic impacts (Kautsky and Folke, 1989; Pillay, 1990; Chua, 1992; Avault, 1996). There is a clear need for greater coordination and planning of aquaculture activities in coastal areas. These initiatives, however, must be supported by a scientific knowledge base that can provide policy-makers with a better understanding of ecosystems and their responses to human interventions (Dankers, 1993; GESAMP, 1996; Chua, 1997; Boesch et al., 2000). Of particular importance is the delineation of functional zones where specific economic activities are promoted and regulated by legal measures. Thus, the identification of aquaculture development zones may help to minimize conflict with other coastal resource uses (Barg, 1992; Chua, 1992; Chua, 1993; Chua, 1997; Crosby and Young, 1997; Fernandes and Read, 2000; GESAMP, 2001). Presently, the ability to provide robust predictive relations for management decisions, regarding site selection, expansion of existing operations, and environmental impact is limited (Evans,

1980; Kautsky and Folke, 1989; Barg and Wijkstriim, 1994; Anutha and Johnson, 1996; Dowd, 1997).

A knowledge base with local and regional ecological, sociocultural, and economic conditions should be an integral part of aquaculture development planning. This knowledge base, however, needs to be capable of integrating a wide range of different variables related to local, regional, national and international conditions (Crosby and Young, 1997). Geographic Information Systems have large potential for providing policy- makers with information regarding site capabilities for aquaculture (Kapetsky et al.,

1.2 PURPOSE OF THE

STUDY

This research focuses upon the development of a site capability assessment fi-amework for aquaculture. The overall purpose of the study is to design a system which will integrate the environmental variables which determine the suitability of a site for the culture of native species of molluscs in tropical estuarian environments. The species used for the development of the model is the mangrove oyster, Crassostrea rhizophorae (Guilding, 1828). The study area is the Piraqu6-a~uIPiraqu6-mirim estuarine system (PAPMES) located on the coast of Espirito Santo State (Brazil). The specific study objectives are:

1. To produce a digital cartographic representation of the estuary and adjacent ecosystems, through the digitising of maps, aerial photogrphs and satellite images using the geographical information system Idrisi32;

2. to characterise the overall estuarine environmental quality in terms of biophysical criteria required for mangrove oyster farming based on temperature, salinity, pH, turbidity, suspended inorganic sediments, suspended organic sediments, transparency, phytoplankton biomass (Chlorophyll a), dissolved gases (oxygen) and total and fecal coliforms;

3. to produce continuous surface models from point data of biophysical criteria using interpolation techniques;

4. to develop a multiple criteria evaluation for mangrove oyster farming integrated into a geographic information system.

This dissertation consists of eight chapters including this introduction. Chapter Two sets the context for the study by reviewing recent trends in global aquaculture production, and the emergence of aquaculture in Brazil. Brazilian production is discussed in the view of regulatory measures adopted by the government in order to coordinate the activity. The development of decision support systems (DSS) for site management and regional planning is then discussed, and Multiple-criteria Evaluation (MCE) is presented as an approach to address the inherent complexity of aquaculture sitting. The application of geographic information systems (GIs) to aquaculture planning is then reviewed. Chapter Three presents the biophysical criteria used to evaluate the site suitability for mangrove oyster farming in the Piraqu&apdPiraquC-mirim estuarine system (PAPMES) in the southeastern coast of Brazil. An overview of the Brazilian coast and the study area is presented in Chapter Four. The sampling and analytical design program for the assessment of biophysical criteria for mangrove oyster farming in the PAPMES is also presented. Next, the integration of biophysical criteria using MCE in GIs is presented. The development of GIs methodological procedures for PAPMES mangrove oyster suitability is explained in consideration of the development of a digital cartographic basis, thematic map development, and the integration of biophysical habitat requirements with multiple-criteria evalution. Chapter Five presents the results of biophysical criteria developed from nineteen events of field sampling in the PAPMES. Chapter Six presents the results of the GIs site suitability analysis using the assessment of biophysical criteria for mangrove oyster farming in the PAPMES. The study's implications are presented in Chapter Seven. Chapter Eight presents the main conclusions.

This chapter presents an overview of aquaculture as a valuable activity for food production. World production trends are presented with emphasis on marine bivalve mollusks, especially on oysters. Brazil's trends are also presented with a brief description about recent federal regulation to issue permits for aquaculture leases on coastal waters. Information needs for site management and the regional perspective to integrate coastal aquaculture in watershed and coastal zone dimensions are emphasized. Aquaculture sitting is considered as a key step to optimize resource allocation and to foster sustainability. Applications of geographic information system (GIs) in aquaculture planning and coastal zone management are reviewed.

2.1 COASTAL AQUACULTURE OVERVIEW

2.1.1 Global Perspective 2.1.2 Brazilian Perspective

2.2 DECISION SUPPORT APPROACHES FOR AQUACULTURE DEVELOPMENT

2.2.1 Site Management 2.2.2 Regional Planning

2.3 COASTAL AQUACULTURE SUITABILITY ANALYSIS: THE ADVANCE OF

GEOGRAPHIC INFORMATION SYSTEMS

2.1.1 Global Perspective

According to the United Nations Food and Agriculture Organization (FAO, 1990), aquaculture is the activity of farming aquatic organisms in captivity under managed systems. Farmed species include fish, mollusks, crustaceans, and aquatic plants characteristic of fresh, brackish and marine waters. In 2000, more than 210 species were reported to be farmed (FAO, 2002). Chua (1997) refers to 'coastal aquaculture' as the farming of brackish and marine species in land-based systems (i.e., ponds and raceways) as well as in in-water systems (e.g., rafts, long-lines, cages, and pens) in protected coves, bays, gulfs, and lagoons. The term 'mariculture' is often restricted to the farming of marine species in the open seas, either in the water column or on the seabed. Farming aims to enhance production through many management techniques such as stocking, feeding, protection from predators, and control of undesirable environmental factors, conducted in different types systems. Management level is defined by the stocking density and level of artificial feeding. From the lowest to the highest levels of stocking and feeding, aquaculture can be classified in extensive, semi-intensive, and intensive systems. Higher management levels generally imply in higher profitability, but at the expense of reduced sustainability (Rosenthal, 1994; Nunes and Parsons, 1998; Costa- Pierce, 2002).

Aquaculture contributes to world food security and also promotes economic development by generating jobs and revenues (Howarth, 1996; Muir and Nugent, 1997; FAO, 2002; Brumrnett, 2003). In 2001, aquaculture accounted for 29.1% of the world's fisheries production (128.8 million tons) (Table 2.1). This contribution would be higher if aquatic plant production were considered. The contribution in 2001 was 6.9% higher than 1996. This increasing trend in aquaculture production is due largely to intensification and expansion of aquaculture enterprises worldwide. Aquaculture is the most rapidly growing sector in animal food production. The average annual growth rate for aquaculture is estimated at 9.2% while the growth rate of capture fisheries and meat production systems are 1.4% and 2.8%, respectively for the period 1970 to the present (FAO, 2002).

Inland and marine fish productions differ widely in their relative contributions (Table 2.1). In 2001, aquaculture represented 71 3 % of inland fisheries, but only 15.4% of

marine fisheries (Table 2.1). This difference may be related to the advanced farming techniques and the intensive production of inland aquaculture (FAO, 2002).

The relative contribution of the various farmed groups (e.g., fish, crustaceans, mollusks, plants, and other organisms) and of the type of aquatic environment (i.e., fi-esh, brackish and marine) is influenced by many factors including physiology, ecological conditions in farming areas, and cultural and economics conditions of aquaculture farmers. According to FA0 (2000), the world' aquaculture production in 2001 was 30.3% higher than in 1996 (Table 2.2 and Figure 2.1). The farming of fish accounted for 50% of world aquaculture production. This was followed by mollusks and aquatic plants. In the global scenario, crustaceans contributed only 4.2% by weight of world production.

Although fish farming is the most important contributor to the global aquaculture production in volume, the annual increment rate (2000-2001) for fish farming was only 6.5%, while increment rate for crustaceans reached 12.8% (Table 2.2). This high rate of increase shows the diversifying trend in aquaculture enterprises to produce highly valuable commodities (FAO, 2002).

Table 2.1 : Relative contributions of fisheries and aquaculture to world fisheries production. Modified fiom FA0 (2002). Production Capture Aquaculture Total Aquaculture Capture Aquaculture Total Aquaculture Contribution Contribution (1

o6

tons) (1

o6

tons) (1

o6

tons) (%) (1

o6

tons) (1

o6

tons) (1

o6

tons) (%) Inland 7.4 15.9 23.3 68.2 8.8 22.4 31.2 71.8 Marine 86.1 10.8 96.9 11.1 82.5 15.1 97.6 15,4 Total 93.5 26.7 120.2 22.2 91.3 37.5 128.8 29.1

*

Excluding aquatic plants.

Table 2.2: World aquaculture production (1

o6

tons) by group of organisms in 1996, 2000, and 2001, and increment rates for 1996 - 2001 and 2000 - 2001. Data from FA0 Fishstat aquaculture production quantities (FAO, 2000). Production Increment rate (%) Fish 18.1 (49.3%) 24.5 (49.4%) 26.2 (49.8%) 3 1.3 6.5 Crustaceans 1.2 (3.4%) 1.9 (3.8%) 2.2 (4.2%) 42.9 12.8 Other animals 0.1 (0.2 %) 0.2 (0.3%) 0.2 (0.3%) 62.1 16.7 Mollusks 9.3 (25.5%) 1 1.8 (23.8%) 12.4 (23.6%) 24.7 4.8 Aquatic Plants 7.9 (21.6%) 11.2 (22.6%) 11.6 (22.1%) 32.1 3.6 Total 36.7 (100%) 49.6 (100%) 52.6 (100%) 30.3 5.7

I

Fish 13 Crustaceans El Mollusks IlOther animals Aquatic plants

I

. - . . - . - . .

Figure 2.1 : World aquaculture production per group of organisms in 2001. Data according to FA0 Fishstat aquaculture production quantities: 1950-2001 (FAO, 2000).

Fish farming accounts for 97% of the freshwater aquaculture production (Figure 2.2), with the bulk of the production represented by herbivorous and detritivorous species produced in China (Cen and Zhang, 1998; FAO, 2000; FAO, 2002). Crustaceans are the second most produced freshwater group, with 2% of the production. Brackish water farming (Figure 2.2) is dominated by crustacean production, particularly of highly valuable commodities such as shrimp farmed in coastal ponds, followed by fish, mollusks and aquatic plants.

Marine production is dominated by mollusk farming ( 1 2 . 2 ~ 1 0 ~ tons), followed by aquatic plants, fish, and other aquatic animals (Figure 2.2).

World Freshwater Aquaculture

World Brackish Aquaculture

World Marine

y 10%

Aquaculture

0 Mollusks

Ill Other animals

1

0 Aquatic plants

Figure 2.2: Proportion of world aquaculture production by group of organisms and environments in 2001. Data from to FA0 Fishstat aquaculture production quantities (FAO, 2000): 1950-2001.

Freshwater, brackish, and marine aquaculture has experienced a long-term exponential increase in production since the mid 1980's (Figure 2.3) and is expected to reach more than 83 million tons by 2030 (Bastian, 2000). Freshwater aquaculture (Figure 2.3) is strongly influenced by an increasing production of fish, basically omnivorous and herbivorous species. Also, since 1995 crustaceans in freshwater production and "other animals" have been increasing exponentially. Production of mollusks (i.e., bivalve filter- feeders), crustaceans (i.e., shrimp) and fish in brackish aquaculture can be related to the

high productivity of estuarine systems and the possibility to grow organisms in both land-

based and in-water systems.

The increase in marine aquaculture is due to greater production of mollusks (bivalve filter-feeders), aquatic plants (i.e., macro-algae) and fish (i.e., carnivorous fish such as salmon). More and more developing countries are becoming involved with bivalve farming in coastal ecosystems. According to Gosling (2003), the key factors contributing to the growth of bivalve culture are: recognition of sustainability in food production, assuring income and improving nutrition of rural population; improved hatchery and nursery techniques, as well as improved transportation by road and air.

Bivalve farming is a very diverse activity that is carried out through both on- bottom and off-bottom cultivation techniques, including intertidal and subtidal systems. Essentially, the farming cycle involves the following phases: spat collection, grow out, harvesting, post-harvesting handling, and marketing. Spat can be purchased fiom hatcheries or collected through natural settling using artificial and natural substrates set up in the environment. The grow-out phase refers to the process of fattening, in which bivalves gain weight (shell and meat). Success is directly related to management of stock density, while keeping under control silt, biofouling, disease, and predators (Joseph, 1998). In some cases a post-harvesting handling phase may be required to cleanse bivalves contaminated by pathogenic virus or bacteria in polluted growing waters. Two methods are usually applied: stock transference for 30 days to approved waters (process known as relaying); and the controlled purification (i.e., depuration) where contaminated bivalves are placed in sterilized (ultraviolet light, ozone or chlorine) seawater until they are clean (Huss, 1993; Gosling, 2003).

- .. . .. -

'OD

r

World F r e s W e r ~ q u a c u ~ u r e Production

World Freshwater Aquaculture Productron

World Marine Aquaculture Production

Year

-c- Fish -a- Crustaceans t Other animals

-

Mollusks *Aquatic plants

Figure 2.3: Trends of world aquaculture production by major groups of organisms. Data from FA0 Fishstat aquaculture production quantities (FAO, 2000): 1950-2001.

Products are marketed locally or more widely as whole bivalves or shucked meat, with the latter being consumed fresh, cooled, iced, frozen, steamed, pre-boiled, smoked or canned. Shell by-products can be used on natural culture grounds to provide substrate for spat settlement, thus somewhat promoting a restoration of bivalve beds (Burrell, 1985; Perret and Charty, 1988; Allen and Turner, 1989; Melancon et al., 1997; Smith and Greenhawk, 1998). Oyster shells, which constitute up to 90% of the total animal weight, also serve as raw material for industry (e.g., for the production of calcium hydroxide) and for agriculture liming (Joseph, 1998).

In 2001, aquaculture accounted for 83% of total bivalve mollusk production (Table 2.3). China is the leading producer, with 66% of the world's production. In contrast to fish production, the importance of scallop, clam, mussel, and oyster is higher for aquaculture than for harvestinglfishing (63%, 79%, 84% and 95%, respectively). The annual increase in production rate is 3.2% for bivalve mollusks, 2.8% for scallops, 2.7% for oysters, and 1.9% for mussels. Cupped and flat oysters are the most produced bivalve mollusks accounting for 37% of global mollusk production.

The production of cupped oysters is most significant when compared to flat oysters (Ostrea and Tiostrea species). Flat oyster production in Chile, France, Ireland, Spain, United Kingdom, Greece, Italy, Turkey, Netherlands, USA, and New Zealand was 36% lower in 1997 than in 1987 (Spencer, 2002). The decrease in flat oyster production can be attributed to outbreaks of diseases and the introduction of the more disease- resistant and productive cupped oysters. Cupped oysters are represented by Crassostrea and Saccostrea species, mainly represented by the Pacific or Japanese oyster, C. gigas (Table 2.4). C. gigas is usually farmed in temperate and subtropical waters across continental boundaries, due to intentional and sometimes unintentional introductions. Farming is basically carried out in off-bottom systems (i.e., racks, longlines and rafts), with spat purchased from hatcheries. Production is marketed fresh, shucked, smoked, and canned (Spencer, 2002). In 1997, C. gigas aquaculture produced 92% of the world' cupped oyster production (Table 2.4). World production of

C.

gigas is growing at 6.1% per year considering the production of 1987 and 1997, although it decreased in Korea (-3.9%), Japan (- 1.9%) and in the U.S.A. (-2.7%).

Table 2.3: Bivalve mollusks yield in 1991 and 2001 (lo3 tons) from aquaculture, fisheries and total production. Values in parenthesis in 2001 aquaculture column represent production as percentage of the combined annual yields from world aquaculture and fisheries. Values in parenthesis in 2001 China's aquaculture column represent production as percentage of the combined annual yields from world aquaculture and fisheries. Data from FA0 Fishstat databases (1950-2001): aquaculture production quantities, total captures, and total production (FAO, 2000). Species Global Aquaculture Production Chinese Aquaculture Global Fisheries Total Production Production Oysters 1,289 4,207 (95%) Mussels 1,073 1,3 70 (84%) Clams 770 3,109 (79%) Scallops 379 1,219 (63%) Total 3,513 9,906 (83%) 7,866 (66%) 1,967 1 1,874

Table 2.4: Main species of cupped oysters and their total yields in 1997. Modified fiom Spencer (2002).

1987 1997

Species Common Name Country Tons x 1

o3

Crassostrea gigas PacificIJapanese China 400.5 2,328.6

oyster Korea 303.2 218.1 Japan 258.8 218.0 France 135.8 147.2 USA 42.0 33.2 World total 1,168.2 2,994.0

C. virginica American oyster or Canada 4.7 3.7

Eastern oyster

USA 181.0 171.8

Mexico 50.7 38.5

World total 236.4 220.3

C. rhizophorae Mangrove oyster Caribbean/ Central 2.2 4.8

America

C. angulata Portuguese oyster Portugal <0.01 0.6

Saccostrea commercialis Sydney rock oyster Australia 7.4 5.1

World total (all species) 1,459.6 3,279.4

C. virginica is the second most important species, contributing 16% to the total

cupped oyster production. This species is distributed along the east coast of the USA, Mexico and Canada, tolerating a wide range of salinity (O%O to 42.5%0) and temperature (- 2•‹C to 36OC) (Shurnaway, 1996). It is usually grown through less expensive bottom culture techniques, but is subject to heavy losses caused by disease, predation, and summer anoxic events in bottom waters (Ulanowicz and Tuttle, 1982; Burrell, 1985; Perret and Charty, 1988; Matthiessen, 1991 ; Gottlieb and Schweighofer, 1996; MacKenzie, 1996; Spencer, 2002). Indeed, the production of C. virginica has been decreasing (-0.6% per year) (Table 2.4), and C. gigas is becoming an alternative for

euhaline estuarine farming in the eastern coast of the USA (Gottlieb and Schweighofer, 1996).

Although the contribution of C. rhizophorae to the world's production (0.15% in 1997) is small, farming mangrove oysters is an important small-scale activity that evolved from traditional practices to satisfjr local consumption by coastal communities or small tourism-supported markets (Joseph, 1998). Mangrove oysters are native to the Caribbean, Central America and tropical eastern South America. The natural habitat of this estuarine oyster is the root system of mangrove trees, primarily the red mangrove, Rhizophora mangle (Littlewood, 1988).

Mangrove oysters are usually cultivated on off-bottom systems, including intertidal techniques (e.g., stake and rack methods) and subtidal methods (e.g., longlines, rafts). The type of culture method depends upon the local hydrology, topography of culture sites, growth and reproduction rates, biological characteristics, requirements of the oysters and the end-product in view. Subtidal techniques are advantageous because the constant immersion in the estuarine waters results in a better growth rate. Also, the suitability of the growing area is independent of the nature of the bottom. Husbandry management also provides ease of cleaning, control of biofouling, and harvesting (Joseph, 1998). Conversely, intertidal growth allows the natural control of biofouling exposing competing invertebrate and macro-algae communities to the air and direct insulation. In Cuba, C. rhizophorae is farmed attached to red mangrove branches used to collect oyster spats. The mangrove branches are tied together to provide intertidal suspended collectors, and after the fifth month selective monthly harvests are carried out until the eight month (Nikolic et al., 1976). In Jamaica, subtidal cultivation produces equal or higher numbers of large oysters than does subtidal cultivation (Littlewood, 1988). Products of C. rhizophorae can be marketed as whole, freshly shucked, steam-cooked, boiled or frozen (Joseph, 1998).

Mussels are the second most farmed bivalve mollusks and besides the characteristics of high fecundity and free-swimming larvae, which are common to most other bivalve species, their rapid growth rates and relative resistance to pest and diseases contribute to their potential for substantial increases in production. The bulk of world production is formed by the blue mussel Mytilus edulis and the Mediterranean mussel M.

galloprovincialis. The first, although it is relatively widespread, is typically found in temperate waters of north (north and east Europe, east coast of North America) and south hemispheres (Chile, Argentina and Falklands Islands). M. galloprovincialis is also found in temperate waters but its range extends in warmer waters of the Mediterranean (Pillay, 1990; Spencer, 2002). The tropical/subtropical genus Perna has also a great potential for cultivation. P. virdis has been cultivated in southeastern Asia, P. perna in the Caribbean, Brazil and in South Africa, but the most impressive species in term of production statistics is P. canaliculu, the New Zealand green shell mussel (Spencer, 2002). Although, mussels can be cultivated on poles, rafts and loglines are the most suitable techniques for intensive farming with mechanized methods, particularly in the rias (drowned river valleys) of Galicia in Spain (Small, 2002) and fjords of New Zealand (Suplicy, 2001).

Compared to other bivalve species farming scallops is a relatively new activity, beginning in the 1930s in Japan as an alternative to the severely depleted wild stocks owing to overfishing (Spencer, 2002). At present, Patinopecten yessoensis, the Japanese scallop, produced in China and Japan contributes 86% of the world's scallop production (1,746 thousand tons in 1 997). Usually, scallops are farmed on suspended longlines using pearl and lantern nets but bottom culture has been applied in Japan. The major constraints to scallop aquaculture industry are related to high mortalities of larvae in hatcheries due to disease and also because of seed handling and transport, predation, and low harvesting efficiency (Spencer, 2002).

2.1.2 Brazilian Perspective

Brazilian aquaculture is a diversified activity producing commercially and experimentally at least 64 different species, including 51 fishes (excluding ornamental species), 5 crustaceans, 4 mollusks, 2 reptiles, 1 amphibian, and 1 algae (Ostrensky et al., 2000), by far the majority of these are freshwater species. The animals most commonly farmed in Brazil are freshwater fish, such as tilapia (Oreochromis spp), common carp (Cyprinus carpio), pacu (Piaractus mesopotamicus), tambaqui (Colossoma macropomum), catfish (Pseudoplatystoma sp.), marine shrimp (Litopenaeus vannamei),