CONNECTIVITY OF MARINE BIVALVE SPECIES IN THE NORTHERN GULF OF

CONNECTIVITY OF MARINE BIVALVE SPECIES IN THE NORTHERN GULF OF
CONNECTIVITY OF MARINE BIVALVE SPECIES IN THE NORTHERN GULF OF
CALIFORNIA: IMPLICATIONS FOR FISHERIES MANAGEMENT AND
CONSERVATION
by
Gaspar Soria
_____________________
A Dissertation Submitted to the Faculty of the
SCHOOL OF NATURAL RESOURCES AND THE ENVIRONMENT
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
WITH A MAJOR IN NATURAL RESOURCES
In the Graduate College
THE UNIVERSITY OF ARIZONA
2010
2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the dissertation
prepared by Gaspar Soria
entitled:
Connectivity of Marine Bivalve Species in the Northern Gulf of California: Implications
for Fisheries Management and Conservation
and recommend that it be accepted as fulfilling the dissertation requirement for the
Degree of Doctor of Philosophy
_______________________________________________________________________
Date: 11/08/10
Dr. William W. Shaw
_______________________________________________________________________
Date: 11/08/10
Dr. Phillip Guertin
_______________________________________________________________________
Date: 11/08/10
Dr. Peter T. Raimondi
_______________________________________________________________________
Date: 11/08/10
Dr. Richard Cudney-Bueno
_______________________________________________________________________
Date:
Final approval and acceptance of this dissertation is contingent upon the candidate’s
submission of the final copies of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and
recommend that it be accepted as fulfilling the dissertation requirement.
________________________________________________ Date: 11/08/10
Dissertation Director: Dr. William W. Shaw
3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an
advanced degree at the University of Arizona and is deposited in the University Library
to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided
that accurate acknowledgment of source is made. Requests for permission for extended
quotation from or reproduction of this manuscript in whole or in part may be granted by
the head of the major department or the Dean of the Graduate College when in his or her
judgment the proposed use of the material is in the interests of scholarship. In all other
instances, however, permission must be obtained from the author.
SIGNED: Gaspar Soria
4
ACKNOWLEDGEMENT
First, I would like to thank Dr. William Shaw for accepting me as his graduate student. Bill you
were always there every time I needed help, filling out forms, signing IMOAs, and most
importantly, providing feedback, comments, and suggestion on every manuscript and beyond.
Thanks to you and Darcy for having us. To Richard Cudney-Bueno, I am deeply thankful for the
opportunity to be part of the PANGAS project and for your support and helpful feedback during
this research. I am also very grateful to my other committee members, Phil Guertin and Pete
Raimondi, for their valuable comments and feedback. I am thankful to The David and Lucile
Packard Foundation, The Nature Conservancy and the TNC RJ KOSE Grant Program, and the
Wallace Research Foundation for financial support provided throughout these years. I thank Jorge
Tordecillas-Guillén, Eduardo Araiza-Zamora, and Francisco Hoyos-Chairez from the Centro
Reproductor de Especies Marinas del Estado de Sonora (CREMES) and Maria Lourdes JuárezRomero from the Instituto de Acuicultura del Estado de Sonora (IAES), Mexico for their kind
support. I also thank Community and Biodiversity A.C., particularly Jorge Torre-Cosio and Mario
Rojo-Amaya, for their support. I thank Peggy Turk-Boyer, Ivan Martínez-Tovar, RenéLoaizaVillanueva, Angeles Sanchez-Cruz, from the Intercultural Center for the Study of Deserts and
Oceans (CEDO). Ivan, I will never forget your support before, during, and after every field trip
throughout the Puerto Peñasco Corridor. I appreciate the logistical support and friendship of the
fishing cooperative Buzos de Puerto Punta Peñasco, especially to Cuco Salazar, Valentín León,
Angelillo Mendoza, Adolfo Ramos, and Leopoldo Encinas from Bahía de Kino. Thanks also to all
Pangueros, particularly to my friend Tad Pfister. I enjoyed every epic journey that we shared . It
was sad that I had finished my field work right before you came to work in PANGAS. Otherwise,
I certainly would have grasped the soul of the Gulf of California much better. To Adrián
Munguía-Vega, it was a pleasure to work with you and share ideas, approaches, perspectives, etc.
You were always thinking in the right direction. Thanks to Miguel Lavin and Guido Marinone for
your support and willingness to collaborate on this research. To our friends: Marcia Moreno-Báez
and James Collins, Gabriela Wlasiuk and Leo Alonso, Martin Pessah and Paula Turco, Camilo
Villegas and Paula, Jennie Duberstein (special thanks for editing help), Karla Pelz, Sonya
Stecker, Alyssa and Dennis Rosemartin, Alberto Macías-Duarte and family, Yamillet Carrillo,
Mark Ogonowski, Susie Qashu, Erick Wallace, and Andy Honaman (you rock). To my field
crew: William Ludt, Diana Manjon, and John Hall: you guys were amazing. To Alba and Jorge
Moreno, our ad-hoc parents in Mexico. To the people supporting me and my family from far
away but very close at the same time: my parents, Marta and Guillermo, Ana’s parents, Ana
María and Carlos, and to the rest of our families (hermanos, primos, sobrinos); for their true
support and long patience waiting for us to come back for the last 8 years! Bienvenidas Juanita y
Bianca Sofía! We love you and missed you so much. To my wife Ana for your love, patience, and
for encouraging me to keep going. To our son Nahuel Soria-Cinti, who has become the leitmotif
of our lives and the new source of inspiration. This dissertation is the result of many people
dispersing, settling, and working together. To all of them thank you. Now, it seems it is time to
drift, disperse and settle again somewhere in Patagonia, Argentina.
5
DEDICATION
For Nahuel Soria-Cinti
6
TABLE OF CONTENTS
LIST OF FIGURES ......................................................................................................................... 7
LIST OF TABLES ......................................................................................................................... 12
ABSTRACT................................................................................................................................... 13
INTRODUCTION ......................................................................................................................... 15
The problem and its global context .......................................................................................... 15
This Dissertation ........................................................................................................................ 21
Explanation of the dissertation format .................................................................................... 22
PRESENT STUDY ........................................................................................................................ 26
Study area ................................................................................................................................... 26
Early life history of rock scallop Spondylus calcifer ............................................................... 32
Assessing biological connectivity through marine bivalve larvae dispersal ......................... 33
An overlooked fishery resource: the catarina scallop, Argopecten ventricosus .................... 35
FIGURES ....................................................................................................................................... 38
REFERENCES .............................................................................................................................. 39
APPENDIX A: SPAWNING INDUCTION, FECUNDITY ESTIMATION, AND LARVAL
CULTURE OF SPONDYLUS CALCIFER (CARPENTER, 1857) (BIVALVIA:
SPONDYLIDAE) .......................................................................................................................... 48
APPENDIX B: LINKING BIO-OCEANOGRAPHY AND POPULATION GENETICS TO
ASSESS LARVAL CONNECTIVITY ......................................................................................... 77
APPENDIX C: RECRUITMENT OF CATARINA SCALLOP, ARGOPECTEN
VENTRICOSUS, LARVAE ON ARTIFICIAL COLLECTORS ALONG A GEOGRAPHICAL
AREA IN THE NORTHERN GULF OF CALIFORNIA ........................................................... 140
APPENDIX D: CAPTACIÓN DE SEMILLAS DE BIVALVOS (POSTLARVAS) EN LA
ZONA DE BAHÍA DE KINO, SONORA, MÉXICO ................................................................. 180
7
LIST OF FIGURES
Figure 1: a) The Gulf of California and b) the main study area of Puerto Peñasco (PP) corridor and
former marine reserves (blue areas)......................................................................................................... 38
APPENDIX A: SPAWNING INDUCTION, FECUNDITY ESTIMATION, AND
LARVAL CULTURE OF SPONDYLUS CALCIFER (CARPENTER, 1857)
(BIVALVIA: SPONDYLIDAE)
Figure A.1. Location of sampling site, southern region of Tiburón Island, in the Gulf of California,
Mexico. .................................................................................................................................................... 71
Figure A.2. Mean number of oocytes spawned by each induced-to-spawn female S. calcifer under
hatchery conditions .................................................................................................................................. 72
Figure A.3. Mean water temperature values throughout S. calcifer larvae rearing. Vertical lines indicate
the standard deviation. ............................................................................................................................. 73
Figure A.4. Shell height of S. calcifer larvae reared with three diet treatments: (–x–) 30 cells µl-1, (–■–)
50 cells µl-1, and (–▲–) 70 cells µl-1. Each point represents the mean shell height value for each
treatment. ................................................................................................................................................. 74
Figure A.5. Mean growth rates and 95% confidence intervals (CI) of S. calcifer larvae reared with three
diet treatments.......................................................................................................................................... 75
Figure A.6. Mean survival of S. calcifer larvae reared with three diet treatments: (–x–) 30 cells µl-1, (–
■–) 50 cells µl-1, and (–▲–) 70 cells µl-1. On day 14th, the standard deviations were 29.4, 32.1, and 11.1
at 30, 50 and 70 cells µl-1, respectively. ................................................................................................... 76
APPENDIX B: LINKING BIO-OCEANOGRAPHY AND POPULATION GENETICS
TO ASSESS LARVAL CONNECTIVITY
Fig. B.1. The Northern Gulf of California a), spatial units of analysis (gray solid lines), release sites (red
cruxes), genetic sample collection sites (arrows), and fishing beds (green zones). SLI: San Lorenzo
Island and SEI: San Esteban Island. Panel b) the main study area of Puerto Peñasco (PP) corridor and
former marine reserves (blue areas), and selected spatial units of analysis (gray solid lines) EBO: EL
Borrascoso, LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques. SJO: San
8
Jorge Island, and SFR: San Francisquito. Larvae collection sites are depicted with black stars. Panel c)
southern fishing beds. ............................................................................................................................ 129
Fig. B.2. Final position of particles from the coupled biological-oceanographic model for Spondylus
calcifer a) higher dispersion case (position at high tide of passive particles released at spring tide):
outputs at 1, 2, and 3 weeks, b) output at 2 weeks for higher dispersion case, and c) lower dispersion
case (position at low tide of active particles released at neap tide): output at 2 weeks. Color at each
release site matches its particle’s colors. EBO: El Borrascoso, LCH: La Cholla, SBE: Sandy Beach,
LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, SFR: San Francisquito, PLO: Puerto
Lobos, PLI: Puerto Libertad, LCU: Las Cuevitas, and DDS: Desemboque de los Seris. ...................... 130
Fig. B.3: Relative abundance of particles for each spatial unit of analysis at 1, 2, and 3 weeks. Gray
bars: lower dispersion case (position at low tide of active particles released at neap tide). Black bars:
higher dispersion case (position at high tide of passive particles released at spring tide). LCH: La
Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, SFR: San
Francisquito, PLO: Puerto Lobos, PLI: Puerto Libertad, LCU: Las Cuevitas, and DDS: Desemboque de
los Seris. ................................................................................................................................................ 131
Fig. B.4: Relative abundance of particles for each area at 1, 2, and 3 weeks for the downstream areas of
Vaquita Refuge and El Borrascoso. Gray bars: lower dispersion case (position at low tide of active
particles released at neap tide). Black bars: higher dispersion case (position at high tide of passive
particles released at spring tide). VAQ: Vaquita Refuge, EBO: EL Borrascoso, LCH: La Cholla, SBE:
Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, SFR: San Francisquito,
PLO: Puerto Lobos, PLI: Puerto Libertad, and LCU: Las Cuevitas. ..................................................... 132
Fig. B.5: Relative abundance of particles for each area at 1, 2, and 3 weeks for the upstream areas of
Puerto Lobos and Puerto Libertad. Gray bars: lower dispersion case (position at low tide of active
particles released at neap tide). Black bars: higher dispersion case (position at high tide of passive
particles released at spring tide). PLO: Puerto Lobos, PLI: Puerto Libertad, LCU: Las Cuevitas, ISE:
San Esteban Island, DDE: Desemboque de los Seris, IPA: Patos Island, ITIn: Tiburón Island (north),
IDA: El Dátil Island. .............................................................................................................................. 133
Fig. B.6. Spatial autocorrelation coefficient (r) among individuals of Spondylus calcifer. The genetic
similarity between pairs of individuals within each distance class is measured by r. Positive values
indicate individuals are genetically more similar than expected by random. Bars represent 95% CI.
Dashed lines represent upper (U) and lower (L) confidence limits bound the 95% CI about the null
hypothesis of no spatial structure for the combined data set as determined by permutation. When r = 0,
distance class length = 88.1 km. ............................................................................................................ 134
9
Fig. B.7. GENELAND’s clustering algorithm: a) posterior distribution of the number of distinct genetic
clusters, b) Sampling localities (red dots) and the assignments to the two genetic clusters (green and
gray, respectively). LCH: La Cholla, LTA: Los Tanques, SJO: San Jorge Island, PLO: Puerto Lobos,
and DDS: Desemboque de los Seris. ..................................................................................................... 135
Fig. B.8. STRUCTURE’ clustering algorithm: a) Mean and standard deviation of ln probability of data
for K 1-6, b) Bar plot showing the mean individual assignment probabilities among 10 independent
replicates of K = 2 (red: southern cluster and green: northern cluster). DDS: Desemboque de los Seris,
PLO: Puerto Lobos, SJO: San Jorge Island, SFR: San Francisquito, LTA: Los Tanques, and LCH: La
Cholla. ................................................................................................................................................... 136
Fig. B.9. Mean number (SD) of Spondylus calcifer spat recruited on artificial collectors pooled per site
and year. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge
Island and SFR: San Francisquito. ......................................................................................................... 137
Fig. B.10. Mean number (SD) of Spondylus calcifer spat recruited per collector at different depth and
year. Different letters indicate significant different values (Tukey’s p < 0.05) between depths after oneway ANOVAs for each site and year. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA:
Los Tanques, SJO: San Jorge Island, and SFR: San Francisquito. ........................................................ 138
Fig. B.11. Correlation between relative abundances of predicted values (Wi) and observed relative
values of Spondylus calcifer spat recruited on artificial collectors (Oi) at a) 1 week, b) 2 weeks, and c) 3
weeks for lower dispersion case (position at low tide of active particles released at neap tide) and higher
dispersion case (position at high tide of passive particles released at spring tide). LCH: La Cholla, SBE:
Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR: San
Francisquito. .......................................................................................................................................... 139
APPENDIX C: RECRUITMENT OF CATARINA SCALLOP, ARGOPECTEN
VENTRICOSUS, LARVAE ON ARTIFICIAL COLLECTORS ALONG A
GEOGRAPHICAL AREA IN THE NORTHERN GULF OF CALIFORNIA
Figure C.1. a) The Gulf California and the Baja California Peninsula, Mexico. b) The study area of
Puerto Peñasco showing spat collection sites (black) for catarina scallop, Argopecten ventricosus. LCH:
La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and
SFR: San Francisquito. .......................................................................................................................... 170
Figure C.2. Catarina scallop Argopecten ventricosus. a) Annual fishery’s captures for Northwest
Mexico (solid line) (Source FAO, 2010), and Official captures declared at the regional office of
10
CONAPESCA in Puerto Peñasco (dashed line), Mexico. b) Annual aquaculture production for
Northwest Mexico (Source FAO, 2010). Values are expressed in Mt of whole animal weight. ........... 171
Figure C.3. Diagram of vertical line collecting unit used at each spat collection site. .......................... 172
Figure C.4. Sea bottom water temperature recorded every 4 hours at each spat collection site along the
Puerto Peñasco area from July 23rd, 2007 to August 24th, 2008. The set of arrows in San Jorge Island
panel represent the moment when collectors were replaced, which is also representative for the other
sites. We ended the field collection of spat on August 24th, 2008. *Temperature loggers were lost for the
periods were temperature data is missing. ............................................................................................. 173
Figure C.5. Monthly spat recruitment per collector (black bars) and monthly mean bottom temperature
(oC) (gray lines) at each collection site. Vertical lines represent standard deviation. ............................ 174
Figure C.6. Mean number of spat recruited per collector at different depths 1 m ( ), 3 m ( ), 5 m ( ),
and 7 m ( ) at each collection site at different months. Different letters indicate significantly different
values (Tukeys’s p < 0.05) between depths after one-way ANOVAs for each site and month. Vertical
lines represent standard deviation. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los
Tanques, SJO: San Jorge Island, and SFR: San Francisquito. ............................................................... 175
Figure C.7. Modal analysis of size frequency distributions of Argopecten ventricosus spat recruited on
artificial collectors in December 2007 at different sites in Puerto Peñasco. Cohorts were fitting mixture
distributions to the size dataset. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los
Tanques, SJO: San Jorge Island, and SFR: San Francisquito. * In these cases the analysis did not
differentiate the smaller spat (<3 mm in shell height size) as an independent cohort, which may
represent a recently recruited cohort. ..................................................................................................... 176
Figure C.8. Modal analysis of size frequency distributions of Argopecten ventricosus spat recruited on
artificial collectors in February 2007 at different sites in Puerto Peñasco. Cohorts were fitting mixture
distributions to the size dataset. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los
Tanques, SJO: San Jorge Island, and SFR: San Francisquito. * In these cases the analysis did not
differentiate the smaller spat (<3 mm in shell height size) as an independent cohort, which may
represent a recently recruited cohort. ..................................................................................................... 177
Figure C.9. Modal analysis of size frequency distributions of Argopecten ventricosus spat recruited on
artificial collectors in April 2007 at different sites in Puerto Peñasco. Cohorts were fitting mixture
distributions to the size dataset. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los
Tanques, SJO: San Jorge Island, and SFR: San Francisquito. * In these cases the analysis did not
differentiate the smaller spat (<3 mm in shell height size) as an independent cohort, which may
represent a recently recruited cohort. ..................................................................................................... 178
11
Figure C.10. Modal analysis of size frequency distributions of Argopecten ventricosus spat recruited on
artificial collectors in June 2007 at different sites in Puerto Peñasco. Cohorts were fitting mixture
distributions to the size dataset. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los
Tanques, SJO: San Jorge Island, and SFR: San Francisquito. * In these cases the analysis did not
differentiate the smaller spat (<3 mm in shell height size) as an independent cohort, which may
represent a recently recruited cohort. ..................................................................................................... 179
Figura E.1. Ciclo de vida de bivalvos comerciales como el callo de escarlopa, callo de riñón y
madreperla. ............................................................................................................................................ 183
Figura E.2. Diagrama de una línea de colectores con 8 redes colectoras de netlon. ............................. 184
Figura E.3: Diferentes etapas de armado de colectores y vista parcial de los colectores instalados. a)
Boyas, b) Redes colectoras, c) redes colectoras dentro de bolsas, d) líneas colectoras, e) amarre de una
línea a un ancla de hierro y f) vista parcial desde debajo de una línea colectora. .................................. 186
Figura E.4. Red colectora con semillas de hachas. ................................................................................ 187
Figura E.5: Procesado para extracción de las semillas pegadas a la bolsa (verde) y a la red colectora
(azul). ..................................................................................................................................................... 187
Figure E.6. Diferentes especies de bivalvos de importancia comercial que puede ser encontrados en
colectores artificiales. a) Madre perla (P. mazatlanica) y callo de árbol (Pteria sterna), b) Almeja
mechuda (S. princeps) y callo de escarlopa (S. calcifer), c) almeja voladora (E. vogdesi) y almeja
catarina (A. ventricosus), y d) hachas (Pinnidae). ................................................................................. 188
12
LIST OF TABLES
APPENDIX B: LINKING BIO-OCEANOGRAPHY AND POPULATION GENETICS
TO ASSESS LARVAL CONNECTIVITY
Table B.1: Lineal distance (km) traveled by particles released at different sites. Higher dispersion case
(HD): position at high tide of passive particles released at spring tide, and lower dispersion case (LD):
position at low tide of active particles released at neap tide. Mean (SD) values ................................... 126
Table B.2. Genetic variation among sampled localities. Sample size (N), mean (+SE) numbers of alleles
(NA), effective alleles (NE), and observed (HO) and expected (HE) heterozygosities. Last two columns
show the p values from BOTTLENECK according to a Sign test and a Wilcoxon sign rank test for the
stepwise mutation model (above) and the two-phase mutation model (below). .................................... 127
Table B.3. Genetic differentiation between pairs of localities: Mean (95% CI) Fst (above diagonal) and
Gst' (below diagonal). LCH: La Cholla, LTA: Los Tanques, SJO: San Jorge Island, PLO: Puerto Lobos,
and DDS: Desemboque de los Seris. ..................................................................................................... 128
APPENDIX C: RECRUITMENT OF CATARINA SCALLOP, ARGOPECTEN
VENTRICOSUS, LARVAE ON ARTIFICIAL COLLECTORS ALONG A
GEOGRAPHICAL AREA IN THE NORTHERN GULF OF CALIFORNIA
Table C.1: Sea water salinity, dissolved oxygen and temperature recorded at each collection site in the
study area of Puerto Peñasco. ................................................................................................................ 168
Table C.2: Argopecten ventricosus. Spat shell size frequency distributions at each site and month from
December 2007 to June 2008. Cohorts were identified fitting mixture distributions to the size dataset.
Mean ( µ ), standard deviation (SD), and proportion (%) of data explained by each component of the
modeled mixture distributions. .............................................................................................................. 169
13
ABSTRACT
Understanding the level of biological connectivity among populations of
harvested species is an important step towards establishing fisheries management and
conservation guidelines. Many marine benthic resources present a complex
metapopulation structure in which separate subpopulations of sessile post-larval
individuals are connected through larval dispersal. The extent to which these
subpopulations are linked is termed connectivity and can have different patterns and
implications. Therefore, good management practices require tools that explicitly
acknowledge this complexity across scales.
I investigated the level of connectivity in a commercially important benthic
species, the rock scallop (Spondylus calcifer), in an ecologically sensitive region in the
NE margin of the Gulf of California, Mexico. My approach involved the development of
a predictive coupled biological-oceanographic model (CBOM), which simultaneously
incorporated key oceanographic and biological features. I validated CBOM outputs by
means of two different techniques: population genetics analysis and measurements of
spat abundance on artificial collectors.
In order to infer the planktonic period of S. calcifer larvae to be used as an input
for the model, I studied the early life history of the species under laboratory conditions. I
estimated that the minimum period for larvae of S. calcifer to reach the settlement is
approximately 15 days after fertilization. In addition to providing information useful for
14
the model, this study produced information about the experimental conditions under
which spawning induction and rearing of the species can be successful.
I found strong connectivity along the study region (covering approximately 300
km of coastline). Sampled localities showed low levels of genetic structure, suggesting
the existence of two subtly differentiated genetic populations. Both genetic and CBOM
spatial scales of connectivity are in agreement suggesting that, on average, connectivity
between subpopulation decreases when the geographic distance between them is >100
km.
This study provides a multidisciplinary approach to evaluate the direction,
magnitude and spatial scale of larval dispersal and connectivity, with implications for
fisheries management and conservation in the study region. More broadly, it provides a
baseline for future studies on coastal connectivity at various spatial scales of interest in
the Gulf of California and beyond.
15
INTRODUCTION
The problem and its global context
One of the major goals of fisheries management worldwide is to secure
sustainability of resource use in a world where the demand for seafood is scaling up
rapidly (FAO 2002, Hilborn et al. 2003, Norse & Crowder 2005). In 2000, total landings
of fisheries (including marine, coastal, freshwater, and brackish waters) averaged 95
million Mt and 4 million Mt less (91 million as reported) in 2008. During the same
period, total aquaculture production increased from 42 million Mt in 2000 to 70 million
Mt in 2008, which represents a 65% increase in production (FAO 2010). In 1995, the
United Nations Food and Agriculture Organization declared that 65% of the world’s
fishery resources were fully exploited, overexploited, or depleted, a trend that has not
improved in recent (Hilborn 2005, FAO 2009). In addition, because of the impact of
fisheries on marine ecosystem, higher trophic level fishes have been replaced by lower
trophic level fishes and invertebrates, and commercial fleets are fishing deeper and
further away from their home communities and markets (Pauly & Palomares 2005, Worm
et al. 2006). Further, despite the declining trend of most fisheries landings, we are facing
deterioration of ecological habitats and serious impacts on biodiversity (Heppel et al.
2005, Hilborn 2005, Watling 2005). Because of this decline in fisheries captures
worldwide, it has been suggested that aquaculture production could be an appropriate
strategy to complement or even increase the natural production of at least some species
(González-Anativia 2001, Spencer 2002, Kapetsky & Aguilar-Manjarrez 2007).
16
A major fisheries sector worldwide is constituted by small-scale fisheries (SSFs).
According to Berkes at al., (2001), out of a total of 51 million fishers worldwide (smallscale and large-scale/industrial, this sector comprises approximately 50 million fishers.
SSFs, also known as artisanal, inshore, or coastal fisheries, are socially and economically
very important (FAO 2002, 2009). This type of fishery is characterized by the use of
small boats (though usually in high numbers) and small captures (relative to large-scale
fisheries), where resources are extracted manually with less mechanized gear compared
to large-scale fishing (Berkes et al. 2001). The management of SSFs is challenging due to
high levels of informality, social complexity, and heterogeneity of captures and fishing
gears, and by the fact that many SSFs are based in relatively small and isolated
communities (Berkes et al. 2001).
Marine bivalve species such as scallops, clams, oysters, and blue mussels are the
target of both artisanal and industrial fisheries worldwide and they have not escaped the
overexploitation trends observed globally (Stotz 2000, Ciocco et al. 2006, Félix-Pico
2006). These declines (and in some cases collapses) frequently involved social and
ecological consequences. Within the fisheries management realm there is an ample
variety of instruments intended to achieve sustainable or rational use of resources.
Fisheries management is mainly focused on two strategies: i) the regulation of access to a
fishery and ii) the regulation of the fishing process through implementation of fishing
tactics or measures to constrain what and/or how much is being harvested by those
entering the fishery. In the first group, common strategies for managing benthic
organisms (though not restricted to them) include: fishing licenses (with limited entry
17
systems being the most restrictive type); allocation of property or use-rights over an area
such as Territorial User Rights in Fisheries (TURFs); limiting effort (the use of certain
fishing gear) or quota units (such as Individual Transferable Quotas, or ITQs); and
Marine Protected Areas (MPAs) where access is banned completely or granted with
various levels of restriction to specific groups, communities, or individuals (Jamieson &
Campbell 1998, Berkes et al. 2001, Pollnac et al. 2010). All of these strategies control
access and any management scheme includes tactics to control what is being harvested in
various combinations. Such tactics may involve: size limits, quotas, escapement and
direct effort regulation such as controlling the efficiency and selectivity of a fishing gear,
seasonal closures, rotation of fishing areas, banning of highly destructive gears, etc.
(Jamieson & Campbell 1998, Norse & Crowder 2005, Watling 2005, Orensanz et al.
2006). In addition to these tactics, in many benthic bivalve fisheries direct interventions
to enhance productivity have been implemented, including habitat restoration, stock
enhancement, control of predators, etc. (Salm et al. 2000, Orensanz et al. 2006, Phillips et
al. 2007, White et al. 2007).
Marine Reserves (a specific type of MPA) are explicit locations closed to fishing
and other extractive activities (Agardy et al. 2003, Hilborn et al. 2004, Lester et al. 2009).
Marine reserves are intended to provide refuge to exploited species and the ecosystems
they inhabit so that they can recover from fishing pressure. In addition, it is expected that
adjacent fished areas will benefit from density-dependent spillover and enhanced larval
18
dispersal (Gell & Roberts 2003, Hilborn et al. 2004, Lester et al. 2009). Given that
benthic bivalve species are sessile1 and broadcast spawners and the nature of the fishing
process is often spatially heterogeneous (fishing effort is not distributed homogeneously
over space), management measures that explicitly acknowledge spatial structure, such
marine reserves, are most suitable (Jamieson & Campbell 1998, Orensanz et al. 2006).
Studies have demonstrated the positive effects of spatially explicit management tools on
species recovery in different fisheries worldwide (Gell & Roberts 2003, Lester et al.
2009). For benthic bivalve species, researchers have observed increased densities and
biomass within the reserve and positive effects on larval export from the reserve to
adjacent areas (Gell & Roberts 2003, Beukers-Stewart et al. 2005, Cudney-Bueno et al.
2009b).
Although, the establishment of marine reserves is a promising tool for fisheries
management and conservation of ecosystems and biodiversity, siting marine reserves is a
complex task which involves biological, physical, and social factors (Crowder et al. 2000,
Pollnac et al. 2010). While there is no doubt that the creation of marine reserves are
needed for the conservation of many benthic species, information relevant for their
design is often lacking. Common criteria prescribed for the design and siting of marine
1
In fisheries management, the term sessile is used also to indicate that the movements of post-
settlers (benthic juvenile and adult scallops) have a small spatial scale compared to processes relevant to
large-scale dynamics (e.g., larval dispersal and fleet behavior) (Orensanz et al. 2006).
19
reserves require knowledge about the location of potential subpopulations relative to the
direction of marine currents, the spatial pattern of the beds within reserves, the location of
areas of high densities where recruitment has been successful over the years, and
population connectivity (Orensanz et al. 2006). Therefore, one major piece of information
relies on our ability to understand the spatial scale of the process involved.
Benthic bivalve species are structured in metapopulations in which separate
subpopulations (beds) of sessile post-larval individuals are connected to each other only
through larvae dispersal (Lipcius et al. 2005, Orensanz et al. 2006). The extent to which
these beds are linked by the exchange of larvae is termed connectivity and can have
multiple different patterns (Palumbi 2003, Orensanz et al. 2006). For instance, depending
on how connectivity is measured (e.g. demography, genetics) it can have different
meanings and implications (e. g. for fisheries management, conservation of species and
their evolutionary potential) (Lowe & Allendorf 2010). Both direct and indirect methods
to evaluate the extent of connectivity exist (Hedgecock et al. 2007, Werner et al. 2007).
One such indirect approach to evaluate connectivity is through the development
of explicit coupled biological-oceanographic models (CBOMs). CBOMs can inform us
about the direction, magnitude, and spatial scale of larval dispersion (Gilg & Hilbish
2003, Werner et al. 2007, Carr et al. 2008, Watson et al. 2010). CBOMs simultaneously
include key oceanographic and biological factors to provide insights regarding
connectivity between specific regions, including marine reserves, and provide the basis to
create hypotheses about population dynamics. Although CBOMs have been applied for
assessing connectivity among marine species, validation of the models’ outputs (i.e.
20
coupling theoretical predictions with field data regarding the direction, spatial scale and
magnitude of the larval dispersion) is only rarely performed either directly or indirectly
(Palumbi 2003, Pelc et al. 2010, Selkoe et al. 2010). This is in spite of the importance of
this step for CBOMs to be considered a reliable management tool (Pelc et al. 2010).
Attempts to implement MPAs primarily as a conservation tool in the Gulf of
California, Mexico (Fig. 1a), began as early as 1993 with the creation of the Upper Gulf
of California and Colorado River Delta Biosphere Reserve, followed by many others
(Aburto-Oropeza & Lopez-Sagastegui 2006). My dissertation focuses on a “distinctive”
marine reserve; the Puerto Peñasco (PP) marine reserve network located in the NE
margin of the Northern Gulf of California (Fig. 1b), which was informally and
temporarily established for three fishing seasons (from 2002 to 2005) as a communitybased effort led by a local fishing cooperative (Cudney-Bueno et al. 2009a). It was
informal because it was not formally recognized by the government and temporary
because the lack of support from the government coupled with the intrusion of fishers
from other communities undermined this management effort. The intention of this
network was to protect the declining stocks of rock scallop, Spondylus calcifer, within the
area where PP fishers usually work. Following this ad-hoc initiative, in summer 2006 the
Mexican government granted exclusive access rights to the aforementioned fishing
cooperative to extract S. calcifer in that region and required fishers to prepare a regional
management plan. This plan is a clear example of the implementation of a combination of
tools to regulate access and the fishing process of an important local fishery. It was
required to include annual stock assessments, summer season closures, limits on total
21
allowable catches, minimum harvestable sizes, and the establishment of marine reserves
(Cudney-Bueno et al. 2009a). Therefore, in order to locate effective marine reserves to
support this management plan, improving and validating existing connectivity studies
(Marinone et al. 2008, Cudney-Bueno et al. 2009a) are critical.
This Dissertation
The goal of this dissertation is to understand the extent of biological connectivity
between spatially separate beds in a commercially important marine benthic bivalve
species, Spondylus calcifer, in the Puerto Peñasco corridor, Northern Gulf of California
(NGC), Mexico. By addressing this challenging topic this dissertation aims to provide a
multidisciplinary view of the direction, magnitude, and spatial scale of larval dispersal,
and discuss implications for biological connectivity, fisheries management, and
conservation of the species in the study region.
The specific goals of this dissertation are as follows:
i. To determine the early life history of the rock scallop, S. calcifer, under
laboratory conditions (to be used as input for step ii);
ii. To develop a CBOM for S. calcifer to facilitate future management of this
species in the PP corridor area;
iii. To validate CBOM’s outputs by means of two different techniques: the
estimation of population genetic structure and the measurement of spat
abundance on artificial collectors; and
iv. To develop guidelines for the sustainable management of S. calcifer,
including the establishment of marine reserves in the study region.
22
Explanation of the dissertation format
The results of this dissertation are presented as three separate appended
manuscripts (Appendices A, B, and C). The manuscripts present in-depth details of
specific research questions addressed, methodology, results, and discussion. Various
colleagues appear as co-authors based on our mutual collaboration to conduct specific
research needed to accomplish this research. However, the research design, analysis,
writing, and the majority of the data collected for this research are entirely my own and
the dissertation as a whole represents my original and independent work.
APPENDIX A: “Spawning Induction, Fecundity Estimation, and Larval Culture
of Spondylus calcifer (Carpenter, 1857) (Bivalvia: Spondylidae)” is an article published
in January 2010 in the Journal of Shellfish Research, Volume 28 (1), pages 143-149. In
addition, some of the results of this research were presented at the XII Congress of the
Researchers Association of the Sea of Cortés, in Guaymas, Mexico, in March 2010. This
article describes the early life history of the rock scallop S. calcifer and provides useful
information about the experimental conditions under which spawning induction and
rearing of larvae can be successful. The results of this study were used as input for the
development of CBOM for this species for the NGC (see APPENDIX B). I wrote this
article in collaboration with Jorge Tordecillas-Guillen, William Shaw, and Richard
Cudney-Bueno. Jorge Tordecillas-Guillen, an expert in the cultivation of bivalve’s larvae
working at Centro Reproductor de Especies Marinas del Estado de Sonora (CREMES),
Mexico, was instrumental for the management of the species under controlled conditions.
23
William Shaw (my dissertation director) and Richard Cudney-Bueno (University of
California Santa Cruz and Packard Foundation) acted as my advisors and provided
important feedback throughout the development of the manuscript.
APPENDIX B: “Linking Bio-Oceanography and Population Genetics to Assess
Larval Connectivity” is a manuscript to be submitted to the journal Marine Ecology
Progress Series. Partial results of this research have been presented at the 7th Mote
International Symposium in Fisheries Ecology “The spatial dimensions of fisheries:
putting it all in place” in Sarasota, Florida, USA, in November 2008. This study
investigates the degree of biological connectivity that exists between spatially separated
bivalve beds of one of the most important benthic resources in the NGC, the rock scallop
S. calcifer. We developed an explicit coupled biological-oceanographic model (CBOM)
and validated CBOM’s outputs through two different techniques: population genetics
with nine microsatellite loci and measurements of spat abundance on artificial collectors.
Our analysis provides a multidisciplinary view of the direction, magnitude and spatial
scale of larval dispersal, and its implication for biological connectivity and fishery
management in the study area.
Participating as co-authors are Adrián Munguía-Vega, Guido Marinone, Marcia
Moreno-Báez, Iván Martínez-Tovar, and Richard Cudney-Bueno. A. Munguía-Vega, a
PhD student from the School of Natural Resources and the Environment at the University
of Arizona, performed the genetic analyses of this species’ samples and contributed with
the writing of the sections on genetics in the manuscript. G. Marinone, from the
24
Departamento de Oceanografía Física of the Centro de Investigación Científica y de
Educación Superior de Ensenada (CISECE), Mexico, was instrumental in the
development of the coupled bio-oceanographic model used to predict larval dispersion
flows for the study area. M. Moreno-Báez, a former PhD student from the School of
Natural Resources and the Environment at the University of Arizona and now Post-Doc
fellow at SCRIPPS Institution of Oceanography, collaborated with me in the
development and analysis of GIS layers. Iván Martínez-Tovar from the Centro
Intercultural de Estudios de Desiertos y Océanos (CEDO), Mexico, collaborated with me
throughout the intensive field work to collect and process S. calcifer and other associated
species’ samples. Richard Cudney-Bueno provided important feedback through the
development of the manuscript and helped obtain funding to conduct this research.
APPENDIX C; “Recruitment of Catarina Scallop, Argopecten ventricosus, Larvae
on Artificial Collectors Along a Geographical Area in Puerto Peñasco, Northern Gulf of
California, Mexico” will be submitted to the Journal of Experimental Marine Biology
and Ecology. Although S. calcifer was the primary focus of this dissertation, through the
use of artificial collectors I was able to gather information on the spatial and temporal
availability of larvae of several other species of commercial importance for the PP area.
The catarina scallop is one of the most prominent commercial fishery species for PP
fishers, and therefore I decided to include the main results for the species as a separate
chapter to provide useful information for management. The most relevant results of this
research have been presented at the XII Congress of the Researchers Association of the
25
Sea of Cortés, in Guaymas, Mexico, in March 2010. This paper describes the spatial and
temporal availability of catarina scallop, A. ventricosus, larvae along an environmental
gradient in Puerto Peñasco, Sonora. The paper provides useful information needed for the
development of aquaculture and conservation initiatives in an important fishing
community in the NGC.
Participating co-authors are Iván Martínez-Tovar and Alberto Macías-Duarte.
Iván Martínez-Tovar from CEDO collaborated in this research throughout the intensive
field work to collect and process A. ventricosus samples. Alberto Macías-Duarte, a PhD
candidate from the School of Natural Resources and the Environment at the University of
Arizona, collaborated with me in field experiment design and data analysis and provided
important feedback on the manuscript.
In addition to these research articles, in Appendix D I have included a technical
report that could be useful to anyone interested in the collection of commercial bivalve
spats. I wrote this report as the request of the NGO Comunidad y Biodiversidad A.C.,
Guaymas, Mexico. Using this technical report as a guide, this NGO intends to duplicate
similar studies in the fishing community of Bahía de Kino, Sonora.
26
PRESENT STUDY
The methods, results, and conclusions of this study are presented in the papers
appended to this dissertation. The following is a summary of the most important findings
of my research carried.
Study area
Gulf of California
The Gulf of California, Mexico (Fig. 1a) is characterized by its natural richness,
biodiversity, and endemic species (Brusca 2010). Within this context and throughout both
margins of the unique environment of the Gulf, large-scale and small-scale fisheries and
aquaculture activities take place (Cisneros-Mata 2010). Overall, the production of the
region contributes most of Mexico's fishery resources and is a major supplier of fish and
shellfish products in the southwestern United States and eastern Asia (Carvajal et al.
2010, Cisneros-Mata 2010). Throughout the Gulf of California, the large-scale fishing
sector is estimated to comprise about 10,000 fishers and 1,300 boats, while the smallscale sector involves approximately 50,000 fishers and 25,000 boats (Cisneros-Mata
2010). In addition, aquaculture initiatives are increasing rapidly, fueled predominantly by
the development of shrimp farming (Cisneros-Mata 2010). Nonetheless, even though
fisheries and aquaculture activities are socially and economically relevant for the region,
concerns about sustainability of these activities are increasing among managers, scholars,
and fishers (Ulloa et al. 2006, Cudney-Bueno et al. 2009a, Carvajal et al. 2010, CisnerosMata 2010). These concerns and the desire of several sectors of society to preserve the
27
natural richness of the region led to the creation of a number of marine protected areas
including Biosphere Reserves, National Marine Parks, and Zones for the Protection of
Flora and Fauna (Aburto-Oropeza & Lopez-Sagastegui 2006, Cudney-Bueno et al.
2009a).
Puerto Peñasco and its marine reserve network
Puerto Peñasco is an important fishing community on the NE side of the Gulf of
California (Fig. 1b), located south of the Upper Gulf of California and Colorado River
Delta Biosphere Reserve. One major sector of this community consists of small-scale
fishers, involving approximately 250 boats. Puerto Peñasco fishers use gillnets (for fish
and shrimp), longlines (for fish), traps (for crabs), and commercial diving gear (for
mollusks and fish). Commercial diving (locally referred as “buceo”) is an activity that
brings significant economic returns to local fishers (compared to other available
employment sources) and employs about 12 boats dedicated to this activity year round2.
It includes the harvest of clams (Dosinia sp.), scallops (Argopecten ventricosus), rock
scallop (Spondylus calcifer), pen shells (mostly Atrina tuberculosa), fish (mainly
snappers and groupers), octopus (Octopus spp.), and black and pink murex snails
2
In addition to these 12 boats, another 50-100 boats from PP and other communities in the Gulf of
California may also participate in commercial diving during the fishing seasons of some species like the
black murex snail or the catarina scallop (Martínez-Tovar, I., personal communication), taking advantage
of booms of these species.
28
(Hexaplex nigritus and Phyllonotus erythrostoma, respectively). The main fishing area of
these commercial divers extends southeastward to San Francisquito-San Jorge Island and
northwestward to El Borrascoso (Fig. 1b). The habitats that commercial diving fishers
use to fish have a maximum depth of 20 m and are characterized mainly by rocky reefs
and sandy areas. The tidal regimen is semidiurnal (range = 2.9-8.7 m, mean neap and
spring tides respectively) and water temperature varies markedly seasonally, from 12ºC
in winter to 32°C in summer. The divers are supplied with air from a compressor placed
on the boat through a hose (50-100 m in length) with an air regulator at the end. Divers
fish at different depths and spent a variable amount of time (commonly 3-5 hours)
extracting resources (Cudney-Bueno & Turk-Boyer 1998).
The reserve network established by the divers in 2002 included an offshore
reserve surrounding San Jorge Island (located <50 km southeast of PP) and two coastal
reserves; Sandy Beach in the northwest limit, and Las Conchas in the center (CudneyBueno et al. 2009b) (Fig. 1b). The PP corridor also includes the fishing beds (for
commercial divers) of La Cholla (northwest), Los Tanques (center), the southernmost bed
of San Francisquito, and the western bed of El Borrascoso (located ~75 km west of PP)
(Fig. 1b).
Two scallop species: the rock and catarina scallops
Two scallop species, the rock scallop Spondylus calcifer and the catarina scallop,
Argopecten ventricosus, are fished in the PP corridor. Although both species are
29
primarily fished for its adductor muscle (locally know as “callo”), the fishing processes
for each are different.
Spondylus calcifer is a significant species because of its threatened status. The
legal harvest of the species requires a special permit issued by SEMARNAT3 upon
compliance with strict restrictions; otherwise the harvest is permanently banned
throughout Mexico (NOM-059-ECOL-1994) (SEMARNAT 2001). The species inhabits
intertidal and subtidal areas from the Peruvian coast to the Gulf of California (Poutiers
1996). Within the NGC, the PP corridor presents the most important fishing grounds
(Monsalvo-Spencer et al. 1997, Cudney-Bueno & Rowell 2008). The fishery involves
approximately 45 fishers (~12 boats) that extract rock scallop from at least eight spatially
separated beds. The captureis mainly commercialized at the local market. Given its
conservation status and the fact that local fishers have long extracted this resource
without legal permits, fisheries landings have often been unreported or reported as other
species, such as penshells (also harvested for their adductor muscle or “callo”).
Throughout its range of distribution, S. calcifer inhabits inter-tidal and sub-tidal
rocky reef areas up to 50 m in deep. The species is gonochoric, with an annual
reproductive cycle. It spawns over a short period from July to October, followed by a
long inactive period during the winter (Villalejo-Fuerte et al. 2002, Cudney-Bueno &
Rowell 2008). Its size at first spawning is 86-113 mm in shell height (Villalejo-Fuerte et
3
The Secretary of the Environment and Natural Resources
30
al. 2002), estimated to be between 2.5-4 years of age (Cudney-Bueno and Rowell 2008).
S. calcifer can attain a longevity of at least 10-12 years (Cudney-Bueno and Rowell
2008) and obtaining a maximum shell size of 250 mm (Poutiers 1995, Skoglund &
Mulliner 1996). Despite their ecological and socio-economic importance, little is known
about the early life history of S. calcifer.
On the other hand, A. ventricosus is an economically important resource
throughout the entire Gulf of California and the Pacific side of the Baja California
Peninsula, Northwest Mexico (González-Anativia 2001, Félix-Pico 2006). Of
approximately 20 bivalve species harvested in the region, A. ventricosus landings
accounted for 50% of the total captures from 1986-2001 (Carta Nacional Pesquera 2004).
Unlike S. calcifer, most of the A. ventricosus captures are exported to the United
States(Félix-Pico 2006). The species is also cultured, but in a much lower proportion than
is harvested from the wild. For instance, the largest record of landings of A. ventricosus
totaled 15,800 Mt in 2008 (FAO 2010), while aquaculture production peaked in 2001
with 127 Mt (Carta Nacional Pesquera 2006, FAO 2010).
Argopecten ventricosus inhabits fine and coarse sandy bottoms up to 180 m deep.
It is a functional hermaphrodite species and broadcast spawner that can reach 90 mm of
maximum shell height (Peña 2001). Sexual maturity (when 50% of scallops exhibit
mature gonads) can occur at <4 months (mean shell height = 20 mm) (Cruz et al. 2000).
Scallops at maturing and spawning stages can be present year-round; however the
reproductive pattern vary according to local environmental conditions (BaqueiroCárdenas & Aranda 2000, Luna-González et al. 2000). Several techniques have been
31
applied to culture A. ventricosus to commercial sizes. Depending upon the culture
technique, catarina scallops can reach the commercial size (mean shell height = 60 mm)
in <1 year (Maeda-Martínez et al. 2000, Avendaño et al. 2001, Félix-Pico 2006). In
Puerto Peñasco the species represents an important economic influx for local fishers
when it is available in the natural environment. Nonetheless, in this region landings have
fluctuated markedly over recent years, with the most recent and intense fishing pulses
observed in 2002 and 2009 (Martínez-Tovar, pers. obs). The 2009 fishing pulse lasted
about 7 months, involved approximately 100 fishers, and was localized over a single bed,
accounting for ∼7000 Mt (Martínez-Tovar, pers. obs). In spite of its importance, there are
no records of aquaculture initiatives for the species in the Puerto Peñasco area and the
natural availability of scallop spat (and the factors affecting it) has not yet been
described.
Despite such contrasting differences, there is a point in common between these
two scallop fisheries. When marine reserves were established in the PP region to protect
S. calcifer stocks, the fishing effort entirely shifted to harvest a single, dense bed of A.
ventricosus located in front of PP, which was harvested to depletion. It is in this regard
that the management plan for S. calcifer should incorporate catarina scallop as an
alternative resource to be sustainably fished during the summer closure of S. calcifer and
thus help mitigate detrimental fishing practices for both resources.
32
Early life history of rock scallop Spondylus calcifer
In this study we describe the early life history of rock scallop S. calcifer under
laboratory conditions. In order to improve the coupled biological-oceanographic models
(CBOM) developed for the region for this species, we estimated the period that S. calcifer
larvae need to reach the pediveliger stage after fertilization. This study allowed us to infer
the planktonic period of S. calcifer larvae, during which they can be subjected to
dispersion by currents. In addition, this study provided information about the
experimental conditions under which spawning induction and rearing of S. calcifer larvae
can be successful.
We conducted this study in collaboration with the Centro Reproductor de
Especies Marinas del Estado de Sonora, Mexico, which is a state-owned institute of
marine aquaculture research based in Bahía de Kino, Sonora. We collected adults of S.
calcifer from a natural stock and approached spawn induction and culture of the species’
larvae under hatchery conditions using methodologies commonly used in the production
of scallops. We evaluated the effects on growth rate and final survival of larvae reared
with three dietary treatments: 1) 30 cells µl-1; 2) 50 cells µl-1; and 3) 75 cells µl-1.
Both females and males responded positively to thermal shock induction. Two
weeks after fertilization, larvae reached the pediveliger stage and we ended the
experiment. We found that S. calcifer larval growth rates were significantly different
between diet treatments, with larvae reared at 50 cells µl-1 exhibiting the highest growth
rate (12.42 µm day-1) of all treatments. At the end of the experiment, larvae fed at 50 cell
33
µl-1 attained a larger size (mean height = 234.01 µm; SD 28.03; n = 115) than larvae from
the other two treatments. We did not find significant differences in larval survival
between diet treatments at the end of the experiment (F2,6 = 0.63; p = 0.56).
Our findings suggest that the minimum period for larvae of S. calcifer to begin
settlement is approximately 15 days after fertilization under the experimental conditions
assessed. The first appearance and the extension of the planktonic stage represent the
minimum extension that larvae can be subject to dispersion by oceanographic currents.
This specific result was used as an input for the development of a CBOM (see
APPENDIX B). In addition, the implications of the information provided through this
study extend beyond the development of CBOMs in the sense that this work constitutes
the first attempt to cultivate the species in captivity and provides the basis for future
research on S. calcifer under controlled conditions.
Assessing biological connectivity through marine bivalve larvae dispersal
This paper analyzed the degree of biological connectivity that exists between
spatially separate bivalve beds of one of the most important benthic resources in the
Northern Gulf of California, the rock scallop S. calcifer. Benthic bivalve species like S.
calcifer are structured in metapopulations in which separate subpopulations of sessile
post-larval individuals are connected to each other only through larvae dispersal. The
extent to which these subpopulations are linked by the exchange of larvae is termed
connectivity and can have multiple and different patterns. In addition, the proper
implementation of marine reserves requires knowledge about the spatial and temporal
34
dynamic of biological connectivity between subpopulations, particularly those that are
sources of larvae that can populate other habitats. In order to locate effective reserves to
support the creation of a management plan for the species, particularly regarding the
establishment of marine reserves, improving and validating existing connectivity studies
is critical.
To better understand S. calcifer population dynamics, we developed and
enhanced a coupled biological and oceanographic model (CBOM) for the species with a
specific interest in the area of the PP corridor. CBOMs simultaneously include key
oceanographic and biological factors to provide insights regarding demographic
connectivity between specific regions, including marine reserves and provide the basis to
develop hypotheses about population dynamics. In our particular study, the purpose of
this CBOM was to assess whether San Jorge Island is the only source that provides larvae
to the PP corridor or if there are other potential important sources to the south. We
validated CBOM’s outputs by means of two different techniques: the estimation of
population genetic structure with nine microsatellite loci and the measurement of spat
abundance on artificial collectors.
We found strong demographic connectivity between the corridor and southern
sources such as Puerto Lobos (located 150 km south), which was supported by genetic
data. Sampled localities showed low levels of genetic structure, suggesting the existence
of two subtly differentiated genetic clusters throughout the sampled area along the eastern
side of the NGC. On average, the spatial scale of demographic and genetic connectivity is
in agreement, suggesting that connectivity decreases when the spatial scale is >100 km.
35
We observed high correlation between CBOM’s outputs and spat recruitment on artificial
collectors. Larvae recruitment within the corridor is linked to a large spatial scale of
larval inputs, including local sources and subpopulations further south.
The absence of a strong barrier to migration suggests that the siting of marine
reserves along upstream sites would likely benefit downstream subpopulations. The
spatial scale of connectivity (~100 km) should be used as a reference for the strategic
siting of both upstream and downstream marine reserves in the study area.
In summary, since we estimated low levels of genetic structure among sampled
localities, without the development of a spatially explicit CBOM we would have been
unable to estimate a relative degree of importance for sources contributing into the
corridor. It is in these situations when CBOMs become an important tool to complement
population genetic studies and vice versa, helping to visualize the relative strength of
connectivity among sites, as in this study. CBOMs and population genetics are powerful
complementary tools to assess the relative strength of connectivity between sites. This
study provides novel information for the design and siting of marine reserves as a fishery
management tool in the NGC region.
An overlooked fishery resource: the catarina scallop, Argopecten ventricosus
This work constitutes and extension and complement the research conducted to
estimate spatial and temporal availability of bivalve larvae throughout the PP corridor. In
this sense, we evaluated the availability of A. ventricosus from June 2007 to August
2008. The aim of this study was to provide information about the natural availability of
36
A. ventricosus spat (optimum site locations, timing, depth variances, and intensity of
settlement), needed for the development of aquaculture and conservation initiatives for
the region of PP, where this information is unknown. For that purpose, we estimated
recruitment of A. ventricosus spat on artificial collectors deployed in 6 sites along the PP
corridor. We also placed temperature loggers to record bottom temperature, and
measured sea surface water salinity, temperature, and dissolved oxygen.
In this study we demonstrated that the natural collection of A. ventricosus spat on
artificial collectors in the corridor can be successfully performed over a protracted period
(December-June), with an intense activity throughout cold and temperate months. Also,
spat recruitment varied markedly between seasons and we observed highly significant
differences per site. The presence of diverse recruitment patterns suggest that large scale
processes (seasonal currents, proximity to parental stock, etc.) and local site conditions
(e.g. sea water parameters) may play a significant role as drivers of the observed patterns.
We also found that spat recruitment in the PP corridor was negatively correlated
with seawater temperature. For most sites, we observed higher spat recruitment
abundances in February when sea water temperature was ∼15 ºC. In addition, these spat
recruitment peaks may take place when primary productivity values are relatively higher
in comparison to summer months. It is noteworthy that spat recruitment is characterized
by the presence of multimodal distributions of shell size frequencies and strong peaks of
small sized scallop spat in most cases.
37
Because of the stochastic nature of captures of A. ventricosus (a common
phenomenon also seen in other scallops fisheries elsewhere), it has been suggested that
aquaculture production of this species could be an appropriate strategy to complement
fisheries landings or even increase the natural production of scallops. Furthermore, the
development of aquaculture and conservation initiatives (e.g. stock enhancement and
repopulation programs), relies greatly on the availability of larvae being produced either
under hatchery conditions or collected from natural environments. Thus, we argue that
the information gathered through this study can be used to promote the development of
aquaculture, sustainable management, and conservation initiatives.
38
FIGURES
Figure 1: a) The Gulf of California and b) the main study area of Puerto Peñasco (PP)
corridor and former marine reserves (blue areas).
39
REFERENCES
Aburto-Oropeza O, Lopez-Sagastegui C (2006) Red de Reservas Marínas del Golfo de
California: Una compilación de los esfuerzos de conservación, Greenpeace,
Mexico, DF
Agardy T, Bridgewater P, Crosby MP, Day J and others (2003) Dangerous targets?
Unresolved issues and ideological clashes around marine protected areas. Aquat
Conserv Mar Freshw Ecosystems 13:353-367
Avendaño M, Cantillanez M, Le Pennec M, Lodeiros C, Freites L (2001) Cultivo de
pectinidos iberoamericanos en suspensión. In: Maeda-Martínez AN (ed) Los
Moluscos Pectínidos de Iberoamérica: Ciencia y Acuicultura. Editorial Limunsa,
México, p 193-211
Baqueiro-Cárdenas E, Aranda DA (2000) A review of reproductive patterns of bivalve
mollusks from Mexico. Bull Mar Sci 66:13-27
Berkes F, Mahon R, McConney P, Pollnac RB, Pomeroy RS (2001) Managing smallscale fisheries, alternative directions and methods. International Development
Research Centre, Ottawa
Beukers-Stewart BD, Vause BJ, Mosley MWJ, Rossetti HL, Brand AR (2005) Benefits
of closed area protection for a population of scallops. Mar Ecol Prog Ser 298:189204
Brusca R (2010) The Gulf of California: Biodiversity and Conservation. The University
of Arizona Press, Tucson
Carr S, Capet X, McWilliams J, Pennington T, Chavez F (2008) The influence of diel
vertical migration on zooplankton transport and recruitment in an upwelling
40
region: estimates from a coupled behavioral-physical model. Fish Oceanogr 17:115
Carta Nacional Pesquera (2004) Secretaria de Agricultura, Ganadería, Desarrollo Rural,
Pesca y Alimentación (SAGARPA). Diario Oficial de la Federación, México, p
76-189
Carta Nacional Pesquera (2006) Secretaria de Agricultura, Ganadería, Desarrollo Rural,
Pesca y Alimentación (SAGARPA). Diario Oficial de la Federación, México, p 1128
Carvajal MA, Robles A, Ezcurra E (2010) Ecological Conservation in the Gulf of
California. In: Brusca R (ed) The Gulf of California: Biodiversity and
conservation. The University of Arizona Press, Tucson, p 219-251
Ciocco NF, Lasta M, Narvarte MA, Bremec C, Bogazzi E, Valero J, Orensanz JM (2006)
Scallops of Argentina. In: Shumway S, Parson J (eds) Scallops: Biology, Ecology
and Aquaculture, Vol 26. Elsevier, Amsterdam, p 1251-1292
Cisneros-Mata MA (2010) The importance of fisheries in the Gulf of California and
ecosystem-based sustainable co-management for conservation. In: Brusca R (ed)
The Gulf of California biodiversity and conservation. The University of Arizona
Press, Tucson, p 119-134
Crowder LB, Lyman SJ, Figueira WF, Priddy J (2000) Source-sink population dynamics
and the problem of siting marine reserves. Bull Mar Sci 66:799-820
Cruz P, Rodriguez-Jaramilllo C, Ibarra AM (2000) Environment and population origin
effects on first sexual maturity of catarina scallop, Argopecten ventricosus
(Sowerby II, 1842). J Shellfish Res 19:89-93
41
Cudney-Bueno JR, Turk-Boyer PJ (1998) Pescando entre mareas del alto Golfo de
California. Report No. 1, Centro Intercultural de Estudios de Desiertos y Océanos,
Puerto Peñasco
Cudney-Bueno R, Bourillón L, Sáenz-Arroyo A, Torre-Cosío J, Turk-Boyer P, Shaw
WW (2009a) Governance and effects of marine reserves in the Gulf of California,
Mexico. Ocean Coast Manage 52:207-218
Cudney-Bueno R, Lavín MF, Marinone SG, Raimondi PT, Shaw WW (2009b) Rapid
effects of marine reserves via larval dispersal. PLoS ONE 4(1):e4140.
doi:4110.1371/journal.pone.0004140
Cudney-Bueno R, Rowell K (2008) Establishing a baseline for management of the rock
scallop Spondylus calcifer (Carpenter 1857): Growth and reproduction in the
Upper Gulf of California, Mexico. J Shellfish Res 27:625-632
FAO (2002) Understanding the cultures of fishing communities: A key to fisheries
management and food security. Report No. Fisheries Technical Paper 401, FAO
FAO (2009) The state of world fisheries and aquaculture, Food and Agriculture
Organization of the United Nations, Rome, Italy
FAO (2010) Global production statistics 1950-2008. April 22, 2010.
http://www.fao.org/fishery/topic/16140/en
Félix-Pico EF (2006) Mexico. In: Shumway SE, Parsons GJ (eds) Scallops: Biology,
Ecology and Aquaculture, Vol 35. Elsevier, Amsterdam, p 1337-1390
Gell FR, Roberts CM (2003) Benefits beyond boundaries: the fishery effects of marine
reserves. Trends Ecol Evol 18:448-455
42
Gilg MR, Hilbish TJ (2003) The geography of marine dispersal: coupling genetics with
fine-scale physical oceanography. Ecology 84:2989-2998
González-Anativia CR (2001) Mercados y Comercialización de Pectínidos. In: MaedaMartínez AN (ed) Los Moluscos Pectínidos de Iberoamérica: Ciencia y
Acuicultura. Editorial Limusa, México, p 451-468
Hedgecock D, Barber PH, Edmands S (2007) Genetic approaches to measuring
connectivity. Oceanography 20:70-79
Heppel SS, Heppel SA, Read AJ, Crowder L (2005) Effects of fishing on long-lived
marine organisms. In: Norse E, Crowder L (eds) Marine Conservation Biology:
The Science of Maintaining the Sea's Biodiversity, Vol 13, Island Press, p 211231
Hilborn R (2005) Are sustainable fisheries achievable? In: Norse E, Crowder L (eds)
Marine Conservation Biology: The Science of Maintaining the Sea's Biodiversity,
Vol 15. Island Press, p 247-261
Hilborn R, Branch TA, Ernst B, Magnusson A, Minte-Vera CV, Scheuerell MD, Valero
JL (2003) State of the world's fisheries. Annual Review of Environment and
Resources 28:359-399
Hilborn R, Stokes K, Maguire J-J, Smith T and others (2004) When can marine reserves
improve fisheries management? Ocean Coast Manage 47:197-205
Jamieson GS, Campbell A (1998) Proceeding of the North Pacific symposium on
invertebrate stock assesment and management, Vol 125. NRC-CNRC, Otawa
43
Kapetsky JM, Aguilar-Manjarrez J (2007) Geographic information systems, remote
sensing and mapping for the development and management of marine
aquaculture, Vol 458. Food and Agriculture Organization of the United Nations,
Rome
Lester SE, Halpern BS, Grorud-Colvert K, Lubchenco J and others (2009) Biological
effects within no-take marine reserves: a global synthesis. Mar Ecol Prog Ser
384:33-46
Lipcius RN, Crowder LB, Morgan LE (2005) Metapopulation structure and marine
reserves. In: Norse E, Crowder L (eds) Marine Conservation Biology: The
Science of Maintaining the Sea's Biodiversity, Vol 19. Island Press, p 328-345
Lowe WH, Allendorf FW (2010) What can genetic tell us about population connectivity?
Mol Ecol 19:3038-3051
Luna-González A, Cáceres-Martínez J, Zúñiga-Pacheco C, López-López S, CeballosVázquez BP (2000) Reproductive cycle of Argopecten ventricosus (Sowerby
1842) (Bivalvia: pectinidae) in the Rada del Puerto de Phichilingue, B. C. S.,
México and its relation to temperature, salinity and food. J Shellfish Res 19:107112
Maeda-Martínez AN, Omart P, Mendez L, Acosta B, Sicard MT (2000) Scallop growout
using a new bottom-culture system. Aquaculture 189:73-84
Marinone SG, Ulloa MJ, Pares-Sierra A, Lavin MF, Cudney-Bueno R (2008)
Connectivity in the northern Gulf of California from particle tracking in a threedimensional numerical model. J Mar Sys 71:149-158
44
Monsalvo-Spencer P, Maeda-Martínez AN, Reynoso-Granados T (1997) Reproductive
maturity and spawning induction in the Catarina scallop Argopecten ventricosus
(=circularis) (Sowerby II, 1842). J Shellfish Res 16:67-70
Norse EA, Crowder LB (2005) Marine Conservation Biology: The Science of
Maintaining the Sea’s Biodiversity. Island Press, Washington, Covelo, London
Orensanz JM, Parma AM, Turk T, Valero J (2006) Population, Dynamics and
Management of Natural Scallops. In: Shumway SE, Parsons GJ (eds) Scallops:
Biology, Ecology and Aquaculture, Vol 35. Elsevier, Amsterdam, p 765-868
Palumbi SR (2003) Population genetics, demographic connectivity, and the design of
marine reserves. Ecol Applications 13:146-158
Pauly D, Palomares ML (2005) Fishing down marine food web: It is far more pervasive
than we thought. Bull Mar Sci 76:197-211
Pelc RA, Warner RR, Gaines SD, Paris CB (2010) Detecting larval export from marine
reserves. PNAS doi:10.1073/pnas.0907368107
Peña JB (2001) Taxonomía, morfología, distribución y hábitat de los pectínidos
iberoamericanos. In: Maeda-Martínez AN (ed) Los Moluscos Pectínidos de
Iberoamérica: Ciencia y acuicultura. Editorial Limusa, México, p 1-23
Phillips BF, Melville-Smith R, Caputi N (2007) The western rock lobster in western
Australia. In: McClanahan TR, Castilla JC (eds) Fisheries Management: Progress
towards sustainability, Vol 11. Blackwell Publishing, p 231-271
Pollnac R, Christie P, Cinner JE, Dalton T and others (2010) Marine reserves as linked
social-ecological systems. PNAS doi:10.1073/pnas.0908266107
45
Poutiers JM (1995) Bivalvos. In: Fischer W, Krupp F, Schneider W, Sommer C,
Carpenter KE, Niem VH (eds) Guía FAO para la identificación de especies para
los fines de la pesca: Pacífico centro oriental, Vol 1: Plantas e Invertebrados.
Organizaciones de las Naciones Unidas para la Agricultura y la Alimentación,
Roma, p 646
Poutiers JM (1996) Guía FAO para la identificación de especies para los fines de la
pesca: Pacífico centro oriental. Report No. 1: invertebrados, Food and Agriculture
Organization, Roma
Salm RV, Clarck J, Siirila E (2000) The Role of Protected Areas. In: Marine and Coastal
Protected Areas: A guide for planners and managers, Vol 1. IUCN, Washington
DC, p 14-35
Selkoe KA, Watson JR, White C, Horin TB and others (2010) Taking the chaos out of
genetic patchiness: seascape genetics reveals ecological and oceanographic
drivers of genetic patterns in three temperate reef species. Mol Ecol 19:3708-3726
SEMARNAT (2001) Norma Oficial Mexicana NOM-059-ECOL-2001. Protección
ambiental -Especies nativas de México de flora y fauna silvestres- Categorias de
riesgo y especificaciones para su inclusión, exclusión o cambio. Lista de especies
en Riesgo. Accessed 22 April 2010.
http://www.semarnat.gob.mx/leyesynormas/normas/Pages/normasoficialesmexica
nasvigentes.aspx
Skoglund C, Mulliner DK (1996) The genus Spondylus (Bivalvia: Spondylidae) of the
Panamic Province. The Festivus 28:93-107
Spencer BE (2002) Molluscan Shellfish Farming. Blackwell Publishing
46
Stotz W (2000) When aquaculture restores and replaces an overfished stock: Is the
conservation of the species assured? The case of the scallop Argopecten
purpuratus in northern Chile. Aqua Int 8:237-247
Ulloa R, Torre J, Bourillón L, Gondor A, Alcantar N (2006) Planeación ecorregional para
la conservación marina: Golfo de California y costa occidental de Baja California
Sur, Informe final a The Nature Conservancy. Comunidad y Biodiversidad, A.C.,
Guaymas, México
Villalejo-Fuerte M, Arellano-Martínez M, Ceballos-Vázquez BP, García-Domínguez F
(2002) Reproductive cycle of Spondylus calcifer Carpenter, 1857 (Bivalvia:
Spondylidae) in the "Bahia de Loreto" National Park, Gulf of California, Mexico.
J Shellfish Res 21:103-108
Watling L (2005) The global destruction of bottom habitats by mobile fishing gear. In:
Norse E, Crowder L (eds) Marine Conservation Biology, Vol 12. Island press, p
198-209
Watson JR, Mitarai S, Siegel DA, Caselle JE, Dong C, McWilliams JC (2010) Realized
and potential larval connectivity in the Southern California Bight. Mar Ecol Prog
Ser 401:31-48
Werner FE, Cowen RK, Pars CB (2007) Coupled biological and physical models: Present
capabilities and necessary developments for future studies of population
connectivity. Oceanography 20:54-69
White AT, Gomez E, Alcala A, Russ G (2007) Evolution and lessons from fisheries and
coastal management in the Philippines. In: McClanahan TR, Castilla JC (eds)
Fisheries Management: Progress towards sustainability, Vol 5. Blackwell
Publishing, p 88-111
47
Worm B, Barbier E, Beaumont N, Duffy E and others (2006) Impacts of biodiversity loss
on ocean ecosystem services. Science 314
48
APPENDIX A: SPAWNING INDUCTION, FECUNDITY ESTIMATION, AND
LARVAL CULTURE OF SPONDYLUS CALCIFER (CARPENTER, 1857)
(BIVALVIA: SPONDYLIDAE)
PUBLISHED IN JOURNAL OF SHELLFISH RESEARCH, vol 28 (1) 143-149. 2010
Gaspar Soria1*, Jorge Tordecillas-Guillen2, Richard Cudney-Bueno1,3, and William
Shaw1
1
School of Natural Resources and the Environment, University of Arizona. Biosciences
East 325D. 1311 E 4th street Tucson, AZ. USA. (85721). [email protected]
2
Centro Reproductor de Especies Marinas del Estado de Sonora, IAES. Comonfort y
Paseo del Canal, Edificio Sonora. Hermosillo, Mexico (83280).
3
Institute of Marine Sciences, University of California Santa Cruz. 100 Schaffer Rd.
Santa Cruz, CA, USA (95060).
*Corresponding author: Gaspar Soria, The University of Arizona, School of Natural
Resources, Biological Sciences East 325, Tucson, AZ 85721, USA. Email address:
[email protected] Tel. +1 520 621 5568/626 5607, Fax. 520 621 8801
Keywords: Rock scallop, spiny oyster, Spondylus calcifer, spawning induction, fecundity,
larvae culture, Gulf of California
49
Spawning induction, fecundity estimation, and larval culture of Spondylus calcifer
(Carpenter, 1857) (Bivalvia: Spondylidae)
Gaspar Soria, Jorge Tordecillas-Guillen, Richard Cudney-Bueno, and William Shaw
Abstract
In this study we describe spawning induction, fecundity estimation and the early
life history of the rock scallop Spondylus calcifer under laboratory conditions. We
collected adults of S. calcifer from a natural stock in the Gulf of California, Mexico
(28°37’N; 112°15’W). We induced spawning by means of thermal shocks (10°C
magnitude) and estimated the species’ fecundity as a function of size. We evaluated the
effects on growth rate and final survival of larvae reared with three dietary treatments: 1)
30 cells µl-1; 2) 50 cells µl-1; and 3) 75 cells µl-1. We reared larvae in 15 l containers at a
density of 3 larvae ml-1 and 30ºC, and renewed 100% of the water culture every 48 h. The
diet comprised a combination (1:1) of Isochrysis galbana (clone T-ISO) and Chaetoceros
calcitrans (clone C-CAL). Both females and males responded positively to thermal shock
induction. Mean size of oocytes was 56.0 µm (SD= 4.2; n= 30). Mean number of oocytes
spawned was 48.9 x 106 (SD= 13.2 x 106) with the smallest female (shell height = 110.5
mm) spawning 28.95 x 106 oocytes and the largest (shell height= 180 mm) 71.76 x 106
oocytes. We observed the first group of throchophore and D-larvae at 7 h and 17 h after
fertilization, respectively. At the beginning of the experiment, the mean shell height of
50
D-larvae was 79.8 µm (SD= 8.54; n = 35). Two weeks after fertilization, larvae reached
the pediveliger stage and we ended the experiment. We found that S. calcifer larval
growth rates were significantly different between diet treatments (F2, 2703 = 24.65; p <
0.001), with larvae reared at 50 cells µl-1 exhibiting the highest growth rate (12.42 µm
day-1) of all treatments. At the end of the experiment, larvae fed at 50 cell µl-1 attained a
larger size (mean height = 234.01 µm; SD 28.03; n = 115) than larvae from the other two
treatments (30 cells µl-1: mean height = 210.48 µm, SD 30.81, n = 107; and 70 cells µl-1:
mean height = 221.81 µm, SD 29.81, n = 104). We did not find significant differences in
larval survival between diet treatments at the end of the experiment (F2,6 = 0.63; p =
0.56). Our findings suggest that the minimum period for larvae of S. calcifer to begin
settlement is approximately 15 days after fertilization under the experimental conditions
assessed. The first appearance and the extension of the planktonic stage represent the
minimum extension that the larvae can be subject to dispersion by oceanographic
currents. Whether S. calcifer can delay settlement if no suitable substrate is found was not
addressed in this study. These results will be used as an input for the development of a
coupled biological-oceanographic model (CBOM) that can assist in management of the
rock scallop fishery in the Gulf of California by predicting the species’ larval dispersion
patterns from known reproductive sources.
Introduction
Examining the level of biological connectivity among populations of harvested
species is an important step towards establishing management and conservation
51
guidelines, including the design of networks of marine reserves. For species which
distribution is largely dependent on larval dispersal (such as most shellfish), the
development of predictive coupled biological-oceanographic models (CBOMs) (Carr et
al. 2008, Rasmussen et al. 2008, Siegel et al. 2008) can be used to explore potential
outcomes of connectivity among populations (Tremblay et al. 1994, Marinone et al.
2008). However, in order to produce more realistic outputs, these models require input of
early life history data such as the duration of planktonic larval stages (Marinone et al.
2008, Cudney-Bueno et al. 2009), information often lacking for commercial species.
The rock scallop, Spondylus calcifer (Carpenter, 1857), (also commonly referred
to as “spiny oyster” or “donkey thorny oyster”), is commercially harvested through most
of its geographic distribution from the coast of Perú to the Gulf of California (GC),
México (Poutiers 1995). Within the GC, the species is harvested by small-scale fishermen
along the coast of the states of Sonora, Baja California, Baja California Sur, and in the
Midriff Archipelago Region (MAR) (Cudney-Bueno and Turk-Boyer 1998, VillalejoFuerte and Muñetón-Gómez 2002, Cudney-Bueno and Rowell 2008, Moreno et al. 2008)
(Fig. 1). In the Northern Gulf of California (NGC), the rock scallop is commonly known
as callo de escarlopa. Fishermen harvest it for its adductor muscle (locally known as
callo), which is sold at local markets (Cudney-Bueno and Turk-Boyer 1998). Even
though rock scallops rarely appear in official landings because their landings are often
misreported as other species’ such as penshells (Atrina spp. and Pinna spp) (Moreno et
al. 2005), the NGC and MAR are likely the most intensively fished areas for this species
in the GC (Cudney-Bueno and Rowell 2008, Moreno et al. 2008).
52
S. calcifer inhabits rocky reef areas from the inter-tidal to 55m in depth (Poutiers
1995). It is a gonochoric species with and estimated sex ratio of 1:1 (Villalejo-Fuerte et
al. 2002, Cudney-Bueno and Rowell 2008). Its first size at spawning is 86-113 mm in
shell height (Villalejo-Fuerte et al. 2002), estimated to be between 2.5-4 years of age
(Cudney-Bueno and Rowell 2008). S. calcifer can attain a longevity of at least 10-12
years (Cudney-Bueno and Rowell 2008) and 250 mm in shell size (Poutiers 1995,
Skoglund and Mulliner 1996). In the GC, the species exhibits latitudinal differences
throughout the spawning period from June to October, followed by a protracted period of
gonad recovery during autumn and winter seasons (Villalejo-Fuerte et al. 2002, CudneyBueno and Rowell 2008). Fertilization is external and planktonic stages are followed by
the settlement of sessile individuals.
Because of its commercial importance and threatened status within Mexican
waters (the species is protected by Mexican laws; unless authorized by the ‘Secretaría de
Medio Ambiente y Recursos Naturales’ or SEMARNAT (the Secretary of the
Environment) through a special permit with strict restrictions, its harvest is banned
throughout Mexico (NOM-059-ECOL-1994), a number of connectivity-related studies
are underway in the NGC, including population genetics, larval abundance and
distribution, collection of fishery data, oceanographic surveys and the development of
larval connectivity models. One such study determined the oceanographic circulation
patterns in the NGC, suggesting a counterclockwise connectivity via advection of larvae
during summer (Marinone et al. 2008), when rock scallops reproduce. Based on this
oceanographic model of dispersion, several hypotheses addressing connectivity matrices
53
were developed for the region. However, these hypotheses were formulated assuming
hypothetical durations of planktonic stages, as the species’ larval development had not
yet been described.
In order to validate and improve the connectivity matrices developed for the
region, through this study we estimated the period that S. calcifer larvae need to reach the
pediveliger stage after fertilization. This will allow us to infer the planktonic period of S.
calcifer larvae, during which they can be subjected to dispersion by currents. In addition,
information on rearing species belonging to the genus Spondylus is scarce, limited to a
study conducted for Spondylus tenebrosus in the coast of Hawaii Parnell (2002). This
study provides information about the experimental conditions under which spawning
induction and rearing of S. calcifer larvae can be successful.
Material and Methods
We approached spawn induction and culture of S. calcifer larvae under hatchery
conditions using methodologies commonly used in the production of pectinids species
(Delaunay et al. 1993, Ibarra et al. 1997, Laing 2002, Christophersen and Strand 2003,
Martínez and Pérez 2003, Narvarte and Pascual 2003).
Parental stock conditioning
We collected a first group of adult S. calcifer from a natural stock at Tiburón
Island, Gulf of California (28°37’N; 112°15’W) on May 23rd (Fig. 1). Because these
animals spawned unattended during the night in mid June, and larvae were not recovered,
54
we brought a second group of adults (n = 46; shell height range: 115 - 210 mm) on June
20th, 2008 from the same locality and dismissed the first group of animals. We dislodged
the animals from the rocks by means of scuba-diving, kept them in an insulated box and
brought them to a laboratory (Centro Reproductor de Especies Marinas del Estado de
Sonora, Mexico) in the fishing town of Bahía de Kino (located ~2 h away from the
collection site), where the experiment was carried out. At the laboratory we scrubbed the
shells of each individual to remove epibionts. In order to keep the animals under
appropriate water quality parameters we placed the parental stock in two 450 l raceways
containers supplied with flow-through of seawater at an exchange rate of 2 l min-1 (partial
exchange of 6.4 times per day). The seawater was stored in a 45 m3 tank and enriched
with Isochrysis galbana (clone T-ISO) and Chaetoceros calcitrans (clone C-CAL) at an
average algae concentration of 150 cells µl-1 (ratio 1:1). Additionally, we fully renewed
the water culture every day to prevent the accumulation of debris and feces.
We kept the parental stock at 30 ± 1°C until individuals were induced to spawn.
We acclimatized the adults by increasing the temperature by 1°C every 48 h, starting at
24°C, the sea water temperature at the moment of the collection. Once the experiment
ended, we returned the individuals to their natural environment.
We provided aeration through a glass-bonded silica diffuser in order to maintain a
dissolved oxygen (DO) level above 80% of saturation. We used a hand-sensor (YSI 85;
Yellow Spring Inc, Yellow Spring, OH, USA) to measured salinity (g/l), temperature
(°C) and DO (%) levels daily. We calibrated the oxygen-meter at room atmosphere
55
before use. In addition, we programmed and placed a temperature logger (Maxim-Dallas;
ibutton 1-Wire model DS1922L) to record temperature values every hour.
We produced monospecific mass algal cultures in 20 l plastic carboys at 21 ± 1°C
in sterilized seawater enriched with nutrients and F/2 medium, and continuously
illuminated (cool-white fluorescent lights) and aerated (air enriched with 0.5% carbon
dioxide).
Spawning induction and fecundity estimation
On July 9th 2008, we induced the entire group of rock scallops to spawn by means
of a series of thermal shocks. Since the gonads are not normally visible, we were unable
to determine the maturity stage of each individual through visual examination. We
selected a thermal variation of 10°C decrease departing from the rearing condition of
30°C. We exposed the batch of animals to a rearing condition of 20°C for one hour, and
supplied them with I. galbana (200 cells µl-1). Then, we transferred the group back to a
rearing condition of 30°C for an hour, and place them back again at 20°C for another
hour. After the last thermal shock we divided the group in two 450 l tanks for visual
inspection of any signs of spawning activity.
After the first expulsion of oocytes we transferred each spawning female to an
individual flat bottom (18 l) bucket filled with 1 µm filtered and UV treated seawater,
and left them to spawn freely. We filtered the volume of oocytes spawned by each female
through a 100 µm nylon-mesh to eliminate debris and transferred it to a plastic tank (18
l). We estimated the total number of oocytes released by each female taking counts of 25
56
µl (5 replicates) from the oocytes’ suspension and extrapolated this count to the total
volume. After that, we pooled oocytes from different females and placed the spawning
males in a common container to obtain fresh spermatozoa. We fertilized the oocytes by
adding sperm at an approximate ratio of 10 to 1 (spermatozoa-oocyte). We incubated the
zygotes in a flat bottom tank supplied with subtle aeration at all times.
Larval experiment
We studied the effects on the growth rate (measured by the increase in shell
height) and final survival of S. calcifer larvae reared with three dietary treatments: 1) 30
cells µl-1; 2) 50 cells µl-1; 3) 75 cells µl-1 (control 0 cells µl-1). When larvae reached the
straight hinge stage (D-larvae) we sifted them through a 35 µm nylon-mesh to retain the
larvae, placing them in several 18 l buckets to count them before beginning the
experiment. We randomly allocated each dietary treatment and replicate (3 replicates per
treatment and control). We cultured the larvae in 5 µm-filtered seawater in 15 l containers
at a density of 3 larvae ml-1, and held at 30ºC. We chose this rearing temperature because
it is the average temperature in the natural environment throughout both the spawning
period and larval development (Villalejo-Fuerte et al. 2002, Cudney-Bueno and Rowell
2008, Soria et al. 2008). T he diet was a combination of two algal species commonly used
in pectinids hatcheries, I. galbana (clone T-ISO) and C. calcitrans (clone C-CAL)
(Martínez et al. 1995, Narvarte and Pascual 2003). We divided the daily food allotment
into thirds, and provided a third at morning, noon, and evening. We renewed 100% of the
water culture every 48 h, retaining the larvae in a 35 µm nylon-mesh. We ended the
57
experiment when larvae reached the pre-metamorphic stage (beginning of the settlement),
which we identified by the presence of a foot and the eye spot.
At the moment of water exchange, we gently homogenized the larval suspension
to take a 0.1 ml sample from each container. We used the software Image ProPlus 4.0,
Media Cybernetics, to measure shell height of a sample of approximately 45 larvae based
on digital images. We fitted daily growth rates to a linear growth model and tested
differences among growth rates following an ANCOVA analysis (Sokal and Rohlf 1995).
We estimated larval densities every 4 days, and tested differences in final survival among
treatments with a one-way ANOVA test (single factor, diet) (Sokal and Rohlf 1995).
Results
Spawning induction and fecundity estimation
Thirteen female and eight male individuals (shell-height range 115 – 190.0 mm)
responded positively to the second day (consecutive) of thermal shock induction. The
spawning began approximately 1 h after the second thermal shock, when animals were
returned to the rearing tanks at 30°C. We noticed that oocytes were released after either a
series of strong “shell clapping” events or a significant and protracted opening of the
valves, when clouds of oocytes were suddenly spawned. Females repeated the expulsion
of gametes 3-5 times. After spawning, oocytes had a rounded shape and orange
coloration. The mean number of oocytes released was 48.9 x 106 (SD= 13.2 x 106), and
numbers varied positively in proportion to the size of the female. The smallest female
(shell height 115 mm) spawned 28.95 x 106 oocytes, while the highest amount of oocytes
58
(71.76 x 106) was spawned by one of the largest individuals (shell height 180 mm) (Fig.
2). The mean size of oocytes was 56.0 µm (SD= 4.2; n= 30). As for the males, their
spawning activity followed a similar pattern than that of the females. The shell height of
these individuals ranged between 120 and 180 mm. We did not determine the sex of those
individuals that did not spawn. We observed the first group of throchophore larvae 7 h
after fertilization, and D-larvae 17 h after fertilization. Mean temperature throughout
embryonic development was 29.52°C (SD = 0.83).
Larval experiment
The mean water temperature throughout the experiment was 30.3°C (SD= 0.82)
(Fig. 3). Salinity ranged between 34.9 and 35.1 g l-1, and DO concentration was higher
than 95% at all times.
At the beginning of the experiment the mean shell height of D-larvae was 79.8
µm (SD= 8.54; n = 35). On July 26th, 2 weeks after fertilization, larvae reached the
pediveliger stage and we ended the experiment. A week after the beginning of the
experiment we observed that larvae reared under the control treatment (0 cells µl-1)
increased their shell size by 23 µm on average ( 3.25 µm day-1) and mortality was higher
than 95%. For these reasons we exclude results from the control treatment from our
statistical analysis. We found that S. calcifer larval growth rates were significantly
different between diet treatments (F2, 2703 = 24.65; p < 0.001), with larvae reared at 50
cells µl-1 exhibiting the highest growth rate of all treatments (12.42 µm day-1) (Fig. 4 and
Fig. 5). At the end of the experiment, larvae fed at 50 cell µl-1 attained a larger size (mean
59
height = 234.01 µm; SD 28.03; n = 115) than larvae from the other two treatments (30
cells µl-1: mean height = 210.48 µm, SD 30.81, n = 107; and 70 cells µl-1: mean height =
221.81 µm, SD 29.81, n = 104). We did not find significant differences in larval survival
among diet treatments at the end of the experiment (F2,6 = 0.63; p = 0.56) (Fig. 6).
Discussion
Spawning induction and fecundity estimation
It has been argued that temperature and primary productivity are the most
important exogenous factors regulating and synchronizing the reproduction cycle of
bivalve species (Barber and Blake 2006). For the rock scallop, S. calcifer, it was
suggested that the reproductive cycle is mainly driven by these parameter, where
energetic resources are allocated on the adductor muscle throughout the colder months
(winter and spring) and the spawning taking place when sea water temperature is at least
29°C (Villalejo-Fuerte et al. 2002, Cudney-Bueno and Rowell 2008). On the other hand,
under laboratory conditions, once the adults are mature, either physical or chemical
stimulus might trigger the spawning of gametes. In the case of S. calcifer, the spawning
cue that induced adults to spawn was a sharp decrease in water temperature followed by a
combination of warming and cooling steps, a technique that has also been documented
for other pectinacean species (Narvarte and Pascual 2003, Parsons 2006, von Brand et al.
2006). There are other methods to induce spawning in bivalves. For instance, Parnell
(2002) suc cessfully applied a combination of intramuscular serotonin injection (a
60
neurotransmitter) and warm water shocks (from 26°C to 35°C) to induce spawning in
adults of S. tenebrosus.
It is unlikely that the temperature variation applied in the laboratory in our study,
as a physical stimulus, somewhat mimics temperature variations in nature. During
summer there are no upwelling events along the east border of the Northern Gulf of
California (NGC) and Tiburón Island. However, colder and enriched sea water masses
from the Peninsula side (Angel de la Guarda Island) are exported towards the east side of
the gulf throughout the spawning season of S. calcifer (Sánchez-Velasco et al. 2009).
Further studies should be attempted in order to understand the role environmental factors
plays in the reproductive behavior and to fine tune the spawning induction techniques
under laboratory conditions for this commercial species.
The reproductive cycle of S. calcifer is scarcely known throughout its range of
distribution within the Gulf of California (GC). According to a histological analysis
performed by Villalejo-Fuerte et al. (2002), the spawning season of individuals of S.
calcifer from Bahia de Loreto, Baja California Sur (Fig. 1), occurs in August (~40% of
individuals), September (~35%), and October (< 10%). Based on visual examination of
the gonads of individuals from Puerto Peñasco in the NGC, the species spawns earlier
and during a shorter period (July and August) than in Bahia de Loreto, (Cudney-Bueno
and Rowell 2008). Though under laboratory conditions, in this study the first group of
animals collected spawned in June (% of animals was not determined), and from the
second group, 45% of the individuals (males and females combined) induced to spawn
responded positively to the thermal shock at the beginning of July. Given that within the
61
GC S. calcifer is found throughout sites with marked variation in biological and physical
attributes (e. g. the Midriff Archipelago Region and both margins of the gulf), further
studies addressing the reproductive cycle of the species throughout its range of
distribution should be performed in order to infer the reproductive timing and availability
of mature adults to be induce to spawning under laboratory conditions. Additionally, the
existence of latitudinal differences in the timing of reproduction and spawning would
define the availability of larvae in the water column and the CBOMs’ output.
The total number of oocytes spawned was positively related with shell size. The
smallest female sampled in our study (height= 115.0 mm) that positively responded to
induction coincided with the size at which 50% of the population is spawning as reported
by Villalejo-Fuerte et al. (2002). However, S. calcifer can begin spawning at 86 mm
(Villalejo-Fuerte et al. 2002). Unfortunately, because our sampling design involved only
collecting large individuals, we cannot provide inferences about the spawning behavior of
smaller individuals. However, although biased towards the collection of large individuals,
it is noteworthy that the majority of the specimens collected for this study were larger
than those reported for other areas of the GC even after extensive sampling efforts. For
instance, in a study to estimate age and growth of S. calcifer in the area of Puerto Peñasco
(where much of the harvest currently takes place) the largest individual collected (n =
347) was 186.6 mm but it was so badly burrowed by epibionts that longevity was
estimated for the next largest individual collected (160.7 mm) at 10-12 y (Cudney-Bueno
and Rowell 2008). Brusca (1980) and Keen (1971) report the species’ maximum size at
150 mm in shell height. Villalejo-Fuerte et al. (2002) reported sizes up to 160.5 mm (n =
62
220) for Bahía de Loreto in the southern GC. However, the largest individual collected in
our study was 210 mm in shell height, and almost 50% of the rock scallops sampled
(males and females combined) were larger than 160 mm, the largest size for which age
estimations have been made. These differences in sizes could be explained by several
factors including; genetic variability throughout its range of distribution, differences in
bio-physical attributes, or perhaps due to fishing pressure, with areas having a lower
fishing pressure allowing individuals to reach larger sizes than in more heavily fished
sites. Nevertheless, our results show that the total number of oocytes produced is
positively related with the size of reproductive individuals, suggesting that overall site
specific reproductive output will be higher where larger individuals are found. Therefore,
management and conservation efforts should pay particular attention to locate rock
scallop beds that have larger, reproductive individuals that can act as important larval
source points.
Larval development
In this study, all groups of larvae reached the pediveliger stage in 15 d, suggesting
that the minimum period for S. calcifer larvae to reach the settlement period is two weeks
after fertilization. This finding is in accordance with published reports for other
pectinaceans species (Merino et al. 2009).
Larvae reared under different conditions showed significantly different growth
rates, with the group reared at 50 cells µl-1 attaining the largest shell size. However,
culturing larvae at high temperatures as we assessed in this study, might have
63
disadvantages such as significant increments of lethal pathogens for the larvae (Jorquera
et al. 2001, Uriarte et al. 2001a). The higher larval mortality observed at the 70 cells µl-1
diet treatment could be related to a higher input of pathogens because of a higher amount
of algae supplied. On the other hand, the depressed growth rate exhibited by larvae reared
at 30 cells µl-1 might be related to an insufficient food supply as has been observed in
other pectinacean species (Uriarte et al. 2001b, MacDonald et al. 2006, Merino et al.
2009).
The development of larval stages is influenced mainly by temperature, food
quality and supply (Navarro 2001, Cragg 2006, Farias and Uriarte 2006, MacDonald et
al. 2006). Our finding suggesting that the minimum period for larvae of S. calcifer to
begin settlement in 15 d after fertilization might vary according to the culture conditions.
In this study, we cultivated S. calcifer larvae at a temperature close to the average water
temperature larvae will likely be exposed to in the region of Puerto Peñasco in the NGC,
where most of the fishery takes place. Also, the planktonic phase might be protracted at
lower temperatures (Cragg 2006, MacDonald et al. 2006). Fo r instance, for the cogeneric species S. tenebrosus, Parnell (2002) observed throchophore larvae and D-larvae
within 11 h and 21 h respectively after fertilization, when larvae were reared at 22°C.
The minimum period to reach the pediveliger stage for S. tenebrosus larvae was 12 d and
first post-larvae were observed 3 weeks after fertilization (Parnell 2002). In our case, the
rearing temperature was higher and the first stages of trochophore and D-larvae of S.
calcifer were first observed as early as 7 and 17 h after fertilization, respectively. On the
other hand, the planktonic phase is likely to be shorter under optimum conditions of both
64
availability and quality of algae (Navarro 2001, Farias and Uriarte 2006), and it is likely
that larvae in its natural environment would show different growth rates according to the
food quality and availability in the region. As an input for coupled biologicaloceanographic models, the first appearance of the trochophore larvae and the extension of
the planktonic phase estimated in this work would represent the beginning and the end of
the planktonic stage, respectively, and therefore, the minimum period of time for larval
dispersion by oceanographic currents in the water column. Nevertheless, whether S.
calcifer can delay settlement if no suitable substrate is found was not addressed in this
study. For S. tenebrosus, Parnell (2002) recorded that the species can delay settlement
and remains planktonic for at least 2 months when substrate is not provided.
Understanding the capability of the species to delay the settlement when no suitable
substrate is available is a key factor for the improvement of connectivity models and
requires further studies.
Given that S. calcifer is distributed throughout an extensive range of coastal
habitats, specific studies should be conducted in order to understand the effect of
environmental variables on the development of larval stages and its implications for the
improvement of connectivity models. Finally, further research is needed in order to
optimize the production of S. calcifer larvae under controlled conditions and to develop
optimal reproductive conditioning techniques.
65
Acknowledgements
We acknowledge the financial support provided from The David and Lucile
Packard Foundation, The Nature Conservancy and the Conservancy’s RJ KOSE Grant
Program, and the Wallace Research Foundation. We also thank E. Araiza-Zamora, and F.
Hoyos-Chairez from the Centro Reproductor de Especies Marinas del Estado de Sonora
who greatly contributed with the rock scallop larvae culture. We appreciate the logistic
support provided by L. Encinas for his participation on rock scallop sampling and M. L.
Juárez-Romero from Instituto de Acuicultura del Estado de Sonora, Mexico. We wish to
thank A. Cinti, T. Pfister, and I. Martínez-Tovar for reviewing and providing insightful
comments on early versions of the manuscript. Collection of S. calcifer individuals was
made under permit #SGPA/DGVS 01349/08 provided by Mexico’s Secretary of the
Environment. This is a scientific contribution of the PANGAS Project
(www.pangas.arizona.edu).
66
Literature cited
Barber, B. J. & N. J. Blake. 2006. Reproductive physiology. In: S. E. Shumway & G. J.
Parsons, editors. Scallops: Biology, ecology and aquaculture: Elsevier. pp. 357416.
Brusca, R. 1980. Common intertidal invertebrates of the Gulf of California. Tucson, The
University of Arizona Press. pp. 513.
Carr, S., X. Capet, J. McWilliams, T. Pennington & F. Chavez. 2008. The influence of
diel vertical migration on zooplankton transport and recruitment in an upwelling
region: Estimates from a coupled behavioral-physical model. Fish. Oceanogr.
17:1-15.
Christophersen, G. & O. Strand. 2003. Effect of reduced salinity on the great scallop,
Pecten maximus spat at two rearing temperatures. Aquaculture 215:79-92.
Cragg, S. M. 2006. Development, physiology, behavior and ecology of scallop larvae. In:
S. E. Shumway & G. J. Parsons, editors. Scallops: Biology, ecology and
aquaculture. Amsterdam: Elsevier. pp. 45-122.
Cudney-Bueno, J. R. & P. J. Turk-Boyer. 1998. Pescando entre mareas del alto Golfo de
California.Centro Intercultural de Estudios de Desiertos y Océanos. Puerto
Peñasco 164 pp.
Cudney-Bueno, R., M. F. Lavín, S. G. Marinone, P. T. Raimondi & W. W. Shaw. 2009.
Rapid effects of marine reserves via larval dispersal. PLoS ONE 4(1):e4140.
doi:4110.1371/journal.pone.0004140.
Cudney-Bueno, R. & K. Rowell. 2008. Establishing a baseline for management of the
rock scallop Spondylus calcifer (carpenter 1857): Growth and reproduction in the
upper Gulf of California, Mexico. J. Shellfish Res. 27:625-632.
67
Delaunay, F., Y. Marty, J. Moal & J. F. Samin. 1993. The effect of monospecific algal
diets on growth and fatty acid composition of Pecten maximus (l.) larvae. J. Exp.
Mar. Biol. Ecol. 173:163-179.
Farias, A. & I. Uriarte. 2006. Nutrition in pectinids. In: S. E. Shumway & G. J. Parsons,
editors. Scallops: Biology, ecology and aquaculture. Amsterdam: Elsevier. pp.
521-542.
Ibarra, A. M., J. L. Ramírez & G. García. 1997. Stocking density on larval growth and
survival of two catarina scallop (Argopecten ventricosus=circularis) (sowerby ii,
1842) populations. Aqua. Res. 28:443-451.
Jorquera, M. A., F. R. Silva & C. E. Riquelme. 2001. Bacteria in the culture of the
scallop Argopecten purpuratus (Lamarck, 1819). Aqua. Int. 9:285-303.
Keen, A. M. 1971. Sea shells of tropical west America. Stanford University Press.
Stanford. pp. 1064.
Laing, I. 2002. Effect of salinity on growth and survival of king scallop spat (Pecten
maximus). Aquaculture 205:171-181.
MacDonald, B. A., V. M. Bricelej & S. E. Shumway. 2006. Physiology: Energy
acquisition and utilization. In: S. E. Shumway & J. Parson, editors. Scallops:
Biology, ecology and aquaculture. Amsterdam: Elsevier. pp. 417-492.
Marinone, S. G., M. J. Ulloa, A. Pares-Sierra, M. F. Lavin & R. Cudney-Bueno. 2008.
Connectivity in the northern Gulf of California from particle tracking in a threedimensional numerical model. J. Mar. Sys. 71:149-158.
Martínez, G., L. A. Caceres, E. Uribe & M. A. Díaz. 1995. Effect of the different feeding
regimens on larval growth and the energy budget of the juvenile Chilean scallops,
Argopecten purpuratus Lamarck. Aquaculture 132:313-323.
Martínez, G. & H. Pérez. 2003. Effect of different temperature regimes on reproductive
conditioning in the scallop Argopecten purpuratus. Aquaculture 228:153-167.
68
Merino, G., E. Uribe, G. Soria & E. von Brand. 2009. A comparison of larval production
of the northern scallop, Argopecten purpuratus, in close and recirculating culture
systems. Aquacul. Eng. 40:95-103.
Moreno, C., M. Rojo & J. Torre. 2008. Diagnóstico socioeconómico de la pesca artesanal
en la región del norte del Golfo de California. Reporte interno para proyecto
Pangas. Comunidad y Biodiversiad, A.C. Guaymas. 36 pp.
Moreno, C., J. Torre, L. Bourillón, M. Durazo, A. H. Weaver, R. Barraza & R. Castro.
2005. Estudio y evaluación de la pesquería de callo de hacha (Atrina tuberculosa)
en la región de Bahía de Kino, sonora y recomendaciones para su manejo.
Comunidad y Biodiversiad, A.C. Guaymas. 27 pp.
Narvarte, M. A. & M. S. Pascual. 2003. Fertilization, larval rearing and post-larval
growth of the tehuelche scallop Aequipecten tehuelchus d´Orb., 1846.
Aquaculture 217:259-274.
Navarro, J. M. 2001. Fisiología energética de pectínidos. In: A. N. Maeda-Martínez,
editor editors. Los moluscos pectínidos de Iberoamérica: Ciencia y acuicultura.
México: Editorial Limusa. pp. 61-76.
Parnell, P. E. 2002. Larval development, precompetent period, and a natural spawning
event of the petinacean bivalve Spondylus tenebrosus (reeve, 1856). The Veliger
45:58-64.
Parsons, G. J. 2006. Sea scallop aquaculture in the northwest Atlantic. In: S. E. Shumway
& G. J. Parsons, editors. Scallops: Biology, ecology and aquaculture. Amsterdam:
Elsevier. pp. 907-930.
Poutiers, J. M. 1995. Bivalvos. In: W. Fischer, F. Krupp, W. Schneider, C. Sommer, K.
E. Carpenter & V. H. Niem, editors. Guía FAO para la identificación de especies
para los fines de la pesca: Pacífico centro oriental. Roma: Organizaciones de las
Naciones Unidas para la Agricultura y la Alimentación. 1:192-195.
69
Rasmussen, L. L., B. D. Cornuelle, L. A. Levin, J. L. Largier & E. Di Lorenzo. 2008.
Effects of small-scale features and local wind forcing on tracer dispersion and
estimates of population connectivity in a regional scale circulation model. J.
Geophys. Res. 114:C01012.
Sánchez-Velasco, L., M. F. Lavín, M. Peguero-Icaza, C. A. León-Chávez, F. ContrerasCatala, S. G. Marinone, I. V. Gutiérrez-Palacios & V. M. Godínez. 2009.
Seasonal changes in larval fish assemblages in a semi-enclosed sea (Gulf of
California). Cont. Shelf Res. 29:1697-1710.
Siegel, D. A., S. Mitarai, C. J. Costello, S. D. Gaines, B. E. Kendall, R. R. Warner & K.
B. Winters. 2008. The stochastic nature of larval connectivity among nearshore
marine populations. Proc. Natl. Acad. Sci. USA 105:8974-8979.
Skoglund, C. & D. K. Mulliner. 1996. The genus Spondylus (Bivalvia: Spondylidae) of
the panamic province. The Festivus 28:93-107.
Sokal, R. & J. Rohlf 1995. Biometry: The principles and practice of statistics in
biological research. New York, Freeman and Co. Pages pp.
Soria, G., M. Moreno-Báez, A. Munguía-Vega, T. Pfister, S. G. Marinone, M. F. Lavín, I.
Martínez, W. Ludt, D. Manjon, J. Hall & J. R. Cudney-Bueno. 2008. Field testing
of Gulf California oceanographic connectivity models. Final Report to The Nature
Conservancy. Pangas project, Tucson. 24 pp.
Tremblay, J. M., J. W. Loder, F. E. Werner, C. E. Naimie, F. H. Page & M. M. Sinclair.
1994. Drift of sea scallop larvae Placopecten magellanicus on Georges bank: A
model study of the roles of mean advection, larval behavior and larval origin.
Deep Sea Research Part II: Topical Studies in Oceanography 41:7-49.
Uriarte, I., A. Farías & J. C. Castilla. 2001. Effect of the antibiotic treatment during larval
development of the Chilean scallop Argopecten pupuratus. Aquacul. Eng. 25:139147.
70
Uriarte, I., G. Rupp & A. Abarca. 2001. Producción de juveniles de pectínidos
Iberoamericanos bajo condiciones controladas. In: A. N. Maeda-Martínez, editor
editors. Los moluscos pectínidos de Iberoamérica: Ciencia y acuicultura. México:
Editorial Limusa. pp. 147-171.
Villalejo-Fuerte, M., M. Arellano-Martínez, B. P. Ceballos-Vázquez & F. GarcíaDomínguez. 2002. Reproductive cycle of Spondylus calcifer carpenter, 1857
(Bivalvia: Spondylidae) in the "Bahia de Loreto" National park, Gulf of
California, Mexico. J. Shellfish Res. 21:103-108.
Villalejo-Fuerte, M. & M. S. Muñetón-Gómez. 2002. Tópicos sobre la biología de la
almeja burra Spondylus calcifer (carpenter, 1857). Hidrobiológica 12:79-81.
von Brand, E., G. Merino, A. Abarca & W. Stotz. 2006. Scallop fishery and aquaculture
in Chile. In: S. Shumway & G. J. Parsons, editors. Scallops: Biology, ecology and
aquaculture. Amsterdam: Elsevier. pp. 1293-1311.
71
Figures
Figure A.1. Location of sampling site, southern region of Tiburón Island, in the Gulf of
California, Mexico.
72
6
Number of oocytes (x10 )
80
y = 0.4392x - 19.658
70
2
R = 0.5473
60
50
40
30
20
10
100
120
140
160
Shell height (mm)
180
200
Figure A.2. Mean number of oocytes spawned by each induced-to-spawn female S.
calcifer under hatchery conditions
73
33
o
Temperature ( C)
35
31
29
27
25
1
3
5
7
9
Days
11
13
15
Figure A.3. Mean water temperature values throughout S. calcifer larvae rearing. Vertical
lines indicate the standard deviation.
74
Shell height (µm)
240
210
180
150
120
90
60
1
3
5
7
9
11
Days after fertilization
13
15
Figure A.4. Shell height of S. calcifer larvae reared with three diet treatments: (–x–) 30
cells µl-1, (–■–) 50 cells µl-1, and (–▲–) 70 cells µl-1. Each point represents the mean shell
height value for each treatment.
13.0
-1
Growth rate (µm day )
75
12.5
12.0
11.5
11.0
10.5
10.0
30
50
-1
Treatment (cell ml )
70
Figure A.5. Mean growth rates and 95% confidence intervals (CI) of S. calcifer larvae
reared with three diet treatments.
76
Survival (%)
100
80
60
40
20
0
0
2
4
6
8
10 12
Days after fertilization
14
16
Figure A.6. Mean survival of S. calcifer larvae reared with three diet treatments: (–x–) 30
cells µl-1, (–■–) 50 cells µl-1, and (–▲–) 70 cells µl-1. On day 14th, the standard deviations
were 29.4, 32.1, and 11.1 at 30, 50 and 70 cells µl-1, respectively.
77
APPENDIX B: LINKING BIO-OCEANOGRAPHY AND POPULATION GENETICS
TO ASSESS LARVAL CONNECTIVITY
TO BE SUBMITED TO THE JOUORNAL MARINE ECOLOGY PROGRESS SERIES.
G. Soria14, A. Munguía-Vega1, G. Marinone2, M. Moreno-Báez1, I. Martínez-Tovar3, and
R. Cudney-Bueno1,4
1
School of Natural Resources and the Environment, University of Arizona. Tucson,
Arizona 85721, USA
2
Departamento de Oceanografía Física, Centro de Investigación Científica y de
Educación Superior de Ensenada. Ensenada, Baja California, México
3
Centro Intercultural de Estudios de Desiertos y Océanos. Puerto Peñasco, Sonora 83550,
México
4
Institute of Marine Sciences, University of California Santa Cruz. Santa Cruz, California
95060, USA
Keywords: Spondylus calcifer, Connectivity, Larval dispersal, Microsatellites, Genetic
structure, Spat collectors, Marine reserves, Gulf of California
4
Corresponding author: [email protected]
78
Linking bio-oceanography and population genetics to assess larval connectivity
G. Soria, A. Munguía-Vega, G. Marinone, M. Moreno-Baez, I. Martínez-Tovar, and R.
Cudney-Bueno
Abstract
Marine reserves (areas closed to fishing) have been advocated for the
management of many species, including the rock scallop, Spondylus calcifer, in the
Northern Gulf of California (NGC), Mexico. We developed an explicit coupled
biological-oceanographic model (CBOM) to assess demographic connectivity among
fished subpopulations of S. calcifer. We focused on the Puerto Peñasco corridor, located
in the northeastern portion of the NGC. We validated CBOM’s outputs through two
different techniques: population genetics with nine microsatellite loci and measurements
of spat abundance on artificial collectors. We found strong demographic connectivity
between the corridor and southern sources. Sampled localities showed low levels of
genetic structure, suggesting the existence of two subtly differentiated genetic clusters.
On average, the spatial scale of demographic and genetic connectivity is in agreement,
suggesting that connectivity decreases when the spatial scale is >100 km. We observed
high correlation between CBOM’s outputs and spat recruitment on artificial collectors.
Larvae recruitment within the corridor is linked to a large spatial scale of larval inputs,
including local sources and subpopulations further south. The absence of a strong barrier
to migration suggests that the siting of marine reserves along upstream sites would likely
benefit downstream subpopulations. The spatial scale of connectivity (~100 km) should
79
be used as a reference for the strategic siting of both upstream and downstream marine
reserves in the study area. CBOMs and population genetics are powerful complementary
tools to assess the relative strength of connectivity among sites. This study provides novel
information for the design and siting of marine reserves as a fishery tool in the NGC.
80
Introduction
Marine reserves and fisheries management
The establishment of marine reserves (areas closed to fishing) is a promising tool
for fisheries management and conservation of ecosystems and biodiversity (Crowder et
al. 2000, Hilborn et al. 2004, Jones et al. 2007). However, siting marine reserves is a
complex task which involves biological, physical and anthropogenic factors (e.g.
fisheries) (Crowder et al. 2000, Pollnac et al. 2010). The positive effects of explicit
spatial management tools, such as marine reserves, on fisheries’ recovery has been seen
in different fisheries world-wide (Gell & Roberts 2003, Lester et al. 2009), particularly in
those fisheries targeting sessile or sedentary stocks (Hilborn et al. 2004). For mobile
species, adjacent fished areas might benefit from density-dependent spillover and
enhanced larval dispersal (Gell & Roberts 2003, Hilborn et al. 2004). The later effect, on
the other hand, is the primary response for the case of benthic and sessile species, such as
scallops, clams or mussels. For this group of species, increased densities and biomass
after the establishment of marine reserves have been observed within the reserve (Gell &
Roberts 2003, Beukers-Stewart et al. 2005). More recently, positive effects on larval
export from the reserve to the surrounding fishing beds have been suggested as well
(Cudney-Bueno et al. 2009, Pelc et al. 2009).
81
Measuring connectivity
Benthic bivalve species are structured in metapopulations in which separate
subpopulations (beds) of sessile post-larval individuals are connected to each other only
through larval dispersal (Lipcius et al. 2005, Orensanz et al. 2006). The extent to which
these subpopulations are linked by the exchange of larvae is termed connectivity and can
have multiple and different patterns (Palumbi 2003, Orensanz et al. 2006). Nevertheless,
the proper implementation of marine reserves requires knowledge about the spatial and
temporal dynamic of biological connectivity between subpopulations, particularly those
that are sources of larvae that can populate other habitats (Kaplan 2006, Fogarty &
Botsford 2007, Pelc et al. 2010).
Depending on how connectivity is measured it can have different meanings and
implications (e. g. for fisheries management, conservation of species and their
evolutionary potential) (Lowe & Allendorf 2010). Demographic connectivity is the
degree to which subpopulation growth rate is affected by dispersal, and is an important
focus for fisheries’ management. Likewise, genetic connectivity is the degree to which
gene flow affects evolutionary processes such as gene frequency and genetic diversity
within subpopulations (Lowe & Allendorf 2010). For both approaches, direct and indirect
methods of measurement exist. Direct methods for measuring demographic connectivity,
such as capture-mark-recapture, are not feasible for marine invertebrate larvae for several
reasons (Lowe & Allendorf 2010, Pelc et al. 2010). Thus, indirect methods such as
explicit coupled biological-oceanographic models (CBOMs) can inform us about the
direction, spatial scale and magnitude of larval dispersion (Gilg & Hilbish 2003, Werner
82
et al. 2007, Carr et al. 2008, Watson et al. 2010). On the other hand, direct methods are
available for estimating genetic connectivity including the use of multiple-locus
genotypes to either assign individuals to their subpopulation of origin (Pritchard et al.
2000; Guillot et al. 2005) or conduct parentage analysis (Hedecock et al. 2007; Lowe &
Allendorf 2010). However, this last approach requires sampling almost all of the potential
parents. Indirect methods estimate the degree of genetic differentiation between
subpopulations based on the assumption that those subpopulations have reached
equilibrium and share the same alleles at the same frequencies. Such indexes of
differentiation include Fst and other similar metrics (Gst’, D, private alleles, etc). Other
indirect methods include the analyses of the spatial scale over which genetic
differentiation occurs (isolation by distance and spatial autocorrelation) (Hedgecock et al.
2007, Lowe & Allendorf 2010).
Coupled biological-oceanographic model (CBOM)
CBOMs simultaneously include key oceanographic and biological factors to
provide insights regarding demographic connectivity between specific regions, including
marine reserves, and provide the basis to elaborate hypotheses about population
dynamics. These hypotheses can be further validated, for instance, through collection of
individual in-situ and population genetic analyses (Werner et al. 2007).
Integrating biological variables into physical models can significantly enhance
our understanding of biological connectivity (Aiken et al. 2007, Marinone et al. 2008,
Rasmussen et al. 2008, Siegel et al. 2008). Larval dispersion by marine currents is
83
strongly influenced by the pre-competency period (the length of time larvae spend in the
plankton), the competency period (when larvae are capable to settle), ontogenic changes,
larval swimming capacities, and the timing of spawning (Siegel et al. 2003, Aiken et al.
2007, Watson et al. 2010). Thus, predictive models that integrate biological features,
spatial structure of the metapopulation, and oceanographic factors are valuable tools for
understanding connectivity among fished subpopulations. In fisheries, for instance, if two
subpopulations show symmetric and balanced connectivity, the potential for them to
repopulate each other is likely high (Lipcius et al. 2005). However, symmetrical
connectivity has been rarely or never seen in coastal marine environments (Gilg &
Hilbish 2003, Cudney-Bueno et al. 2009, Pelc et al. 2009). Conversely, when
asymmetrical and directional connectivity is present, identifying source populations and
understanding connectivity are crucial steps to avoid over-fishing, particularly in those
isolated beds that might be acting as a sole source for downstream areas (Lipcius et al.
2005, Orensanz et al. 2006, Watson et al. 2010).
Surprisingly, although CBOMs have been applied to assess connectivity among
marine species, validation of the model's outputs (i.e. coupling theoretical predictions
with field data regarding the direction, spatial scale and magnitude of the larval
dispersion) is only rarely performed either directly or indirectly (Palumbi 2003, Pelc et al.
2010, Selkoe et al. 2010). This is in spite of the importance of this step for CBOMs to be
considered a reliable management tool (Cudney-Bueno et al. 2009, Pelc et al. 2010).
Validation has the potential to help determine the model parameter values that are closer
to reality. In this regard, determining genetic connectivity through population genetic
84
studies between the areas incorporated into the model is a powerful tool for corroborating
the accuracy of CBOM’s outputs since larval dispersal would have measurable and long
lasting effects on gene frequencies and population genetic structure. Genetic markers
such as microsatellites are ideal for this task given their high polymorphism, fast
mutation rate and codominance (Hellberg et al. 2002, Gilg & Hilbish 2003, Beaumont
2006, Hedgecock et al. 2007). In addition, sampling larval abundance is often unfeasible
in marine bivalves because larvae of many species look alike and thus their identification
is difficult (Arnold et al. 1998, Beukers-Stewart et al. 2005, Aiken et al. 2007, Watson et
al. 2010). However, the recruitment of spat (post-settled larvae) can be used as a proxy of
patterns of larval abundance (Arnold et al. 1998, Pelc et al. 2009).
The Northern Gulf of California: marine reserves and connectivity studies
Marine reserves have been advocated as a fishery tool for the management of the
rock scallop, Spondylus calcifer, in the Northern Gulf of California (NGC), Mexico
(Cudney-Bueno et al. 2009). Locally known as escarlopa, the species is fished for its
adductor muscle by small-scale hookah-diving fishers along both sides of the NGC, and
around the Midriff Archipelago Region (Villalejo-Fuerte & Muñetón-Gómez 2002,
Cudney-Bueno & Rowell 2008, Moreno et al. 2008) (Fig. 1). In 2002, a communitybased effort led by a fishing cooperative established and enforced an informal marine
reserve network (not formally recognized by the government) to protect the declining
stocks of S. calcifer within the area they usually fish, the Puerto Peñasco (PP) corridor
located in the northeastern portion of the NGC (Fig. 1b) (Cudney-Bueno et al. 2009).
85
This network included an offshore reserve surrounding San Jorge Island (located <50 km
southeast of PP) and two coastal reserves; Sandy Beach in the northwest limit, and Las
Conchas in the center (Cudney-Bueno et al. 2009). The PP corridor also includes the
fishing beds of La Cholla (northwest), Los Tanques (center), the southernmost bed of San
Francisquito and the western bed of El Borrascoso (located ~75 km west of PP) (Fig. 1b).
This network was monitored and locally enforced by the fishing cooperative. However,
after a brief period, this community-based initiative was discontinued due to lack of
support from the Government and uncontrolled fishing pressure from both inside and
outside the cooperative (Cudney-Bueno & Basurto 2009).
As part of a separate effort, an oceanographic larval dispersion model was
developed to estimate connectivity in the NGC (Marinone et al. 2008). This former
model is considered an order zero connectivity as particles were passive and is the result
of advection and a random walk process to simulate turbulent motion from a 3-D
baroclinic numerical model. With this model, 2000 particles were randomly released,
between 0 and 60 m, within 21 equally sized areas, and tracked for up to 4 months. The
domain of the model has a mesh size of ~3.9 x 4.6 km in the horizontal and 12 layers in
the vertical. This model predicted two main dispersal patterns for the summer, when
S.calcifer and other commercial mollusk reproduce: 1) the dispersion of particles is
always cyclonic, and 2) there is no influence from the Baja California Peninsula towards
the east margin of the gulf, even after 4 months.
Cudney-Bueno et al. (2009) used this model to evaluate the likely effects of these
reserves on adjacent areas. Based on the model outputs, Cudney-Bueno et al. (2009)
86
hypothesized that observed increment on S. calcifer densities in the northernmost sites of
the reserve network could be attributed to the San Jorge Island acting as a key source for
larval export to adjacent areas. On the other hand, the model suggested that it was highly
unlikely to find any substantial direct influence from reefs located south of the PP reserve
network such as Puerto Libertad and Desemboque de los Seris (~150 and ~200 km
southeast of PP respectively).
Although most fisheries in Mexico are managed by SAGARPA5, the access to
harvest S. calcifer is regulated by the SEMARNAT6. This agency requires a special
permit for this species because it is considerate a threatened species (SEMARNAT 2001).
Under this scenario, in summer 2006, the Mexican government granted exclusive access
rights to the aforementioned fishing cooperative to fish S. calcifer within the PP corridor
and required fishers to prepare a regional management plan. This plan calls for summer
season closures, limits on total allowable catches, annual stock assessments, and the
establishment of marine reserves (Cudney-Bueno & Basurto 2009). In order to locate
effective reserves to support this management plan, particularly regarding the
establishment of marine reserves, improving and validating existing connectivity studies
is critical. As Cudney-Bueno et al. (2009) pointed out, the closure of San Jorge Island
might have benefitted downstream sites but did not result in a rebound of scallops within
5
The Secretary of Fisheries and Agriculture
6
The Secretary of the Environment and Natural Resources
87
the island reserve. It is unknown whether this lack of observed recruitment benefits is
related to local density dependent factors or due to a larger scale spatial process which
could involve other beds south of the PP corridor not considered in their study. Thus, it is
likely that the observed increased recruitment could be attributed at least in part to larval
dispersal from areas other than the monitored marine reserves. Therefore, corroborating if
observed positive effects of marine reserves (i.e. increments on individual densities) are a
direct consequence of the establishment of the reserves or just a correlation that does not
imply causation can be explained by improving existing connectivity studies and by
means of validation techniques.
Goal
To better understand S. calcifer population dynamics, we developed and enhanced
a CBOM for the species with a specific interest in the area of the PP corridor. The
purpose of this CBOM was to assess whether San Jorge Island is the only source that
provides larvae to the PP corridor or if there are other potential important sources to the
south. Since the horizontal resolution of the former model (Marinone et al. 2008) was
somewhat large for detailed local connectivity studies, we used a higher resolution model
which incorporates new oceanographic features (Marinone 2008). Also, we improved the
model by merging new features into the analysis such as species-specific pre-competency
period, larvae motility, and new particle releasing sites based on the metapopulation
structure of the species. We validated CBOM’s outputs by means of two different
techniques: the estimation of population genetic structure with nine microsatellite loci
88
and the measurement of spat abundance on artificial collectors. This study provides a
multidisciplinary approach to evaluate the direction, magnitude and spatial scale of larval
dispersal and connectivity, with implications for fisheries management and conservation
in the study region.
Materials and Methods
Biology of Spondylus calcifer
Adults of S. calcifer are sessile and inhabit rocky reef areas from the inter-tidal to
55m in depth (Poutiers 1995). The species is a gonochoric and broadcast spawner, with
an estimated pre-competency period of approximately 2 weeks after fertilization under
laboratory conditions (Soria et al. 2010). In the NGC spawning takes place mostly in
July, followed by a protracted period of gonad recovery during autumn and winter
(Cudney-Bueno & Rowell 2008). Thus, the dispersion of larvae by oceanographic
currents is the only mechanism for connectivity between subpopulations. Adults can live
at least 10-12 years (Cudney-Bueno & Rowell 2008) and may grow to 250 mm in shell
size (Skoglund & Mulliner 1996). Spawning has been reported to begin at 86-113 mm in
shell height (Villalejo-Fuerte et al. 2002) and 2.5-4 years of age (Cudney-Bueno &
Rowell 2008); and spawn quantity was estimated between 30 and 72 x 106 oocytes,
varying according to the size of the animal (Soria et al. 2010).
89
Spatial units of analysis and selection of particle release sites
To evaluate connectivity, we developed a spatially explicit cartographic model to
define the spatial units of analysis that are appropriate or could be adapted for the
management of the species. We established these spatial units of analysis by combining
physical and political domains, and local knowledge from fishers about S. calcifer fishing
beds. The physical domain of the study area was defined following the coast line
developed by the Instituto Nacional de Estadística, Geografía e Informática, México
(www.inegi.org.mx); the 30 m isobath, which represent the limit of the S. calcifer fishery;
and the 60 m isobath as the deepest boundary for the analysis as implemented by
Marinone et al. (2008). We incorporated pre-existent political boundaries, such as state
and municipal borders, and marine protected areas (CONANP 2009). We relied on
fishers’ knowledge to delimit and select key fishing beds from which to release particles.
Fishers have been shown to have reliable knowledge of high density (fishable) beds in
most scallop fisheries worldwide (Orensanz et al. 2006). Local knowledge was obtained
by asking fishers to map scallop beds and then digitizing these maps as part of an effort
to map the spatial distribution of small-scale fisheries in the NGC (Moreno-Baez et al.
2010). We processed, integrated, and analyzed all the digital layers under ArcGIS 9.3
with the Spatial Analyst Extension and Model Builder tools.
CBOM development and data analysis
We used the numerical model of Marinone (2008) which is three-dimensional
baroclinic numerical Hamburg Shelf Ocean Model (HAMSOM) developed by Backhaus
90
(1985) and adapted to the NGC by Marinone (2003). Briefly, the hydrodynamical model
of Marinone (2008) has a higher resolution with mesh size of ~1.31 × 1.54 km in the
horizontal, and 12 layers in the vertical with the lower levels fixed at 10, 20, 30, 60, 100,
150, 200, 250, 350, 600, 1000, and 4000 m. The forcing includes; a) at the open boundary
model tide components (M2, S2, N2, K2, K1, O1, P1, Ssa, and the Sa), climatological
hydrography historical data base, and b) at the sea surface climatological heat and fresh
water fluxes. For wind, we used as forcing the seasonal climatology constructed from
QUICKSCAT data. The model equations are solved semi-implicitly with fully prognostic
temperature and salinity fields, thus allowing time-dependent baroclinic motions. For
more details see Marinone (2008) and references cited therein. Also we improved the
approach used in Marinone et al. (2008) and Cudney-Bueno et al. (2009) by
incorporating new biological features as described below.
We incorporated species-specific biological data from Soria et al. (2010) to
model the pre-competency and competency periods when larvae are subjected to
dispersion by currents. We seeded 400 particles (i.e. larvae) from specific sites matching
important locations of high density beds identified by the fishers (described in the section
above) and from sites within established marine protected areas in the NGC (Fig. 1a).
From the output of the model, we calculated particles’ trajectories following the
advection/diffusion scheme described in Marinone et al. (2008). We conducted four
different experiments starting the advection at spring (July 15, 2007) and neap (July 22,
2007) tides, and with (active) and without (passive) motility of the particles. We
differentiated the motility capacity of the larvae (simulated by particles) in three stages:
91
(1) embryonic phase in which the larvae stay at the bottom for one day while the larvae
changes from fertilized egg to straight-hinge larvae, (2) veliger phase when larvae are
completely passive and reach the competency period after fourteen days and, (3)
pediveliger phase in which larvae have a tendency to move toward the sea-bottom. These
behaviors were simulated in the model by switching the vertical velocity to zero in the
first stage, and for the third stage the particles were freely advected when the vertical
velocity is negative (towards the bottom) and set to zero when the velocity is positive
(towards the surface).
We tracked particle positions (latitude and longitude) every hour after the
releasing time and queried the data by means of ArcGIS 9.3. to obtain the abundance and
origin of the particles reaching every spatial unit of analysis. For each experiment, we
obtained the final position of particles at both low and high tides. From the different
CBOM’s outputs we selected the two extreme cases: (1) lowest dispersion case (position
at low tide of active particles released at neap tide), and (2) highest higher dispersion case
(position at high tide of passive particles released at spring tide). Based on laboratory
experiments, S. calcifer’s larvae reach the competency period at 2 weeks (Soria et al.
2010). We selected a 2 week period as the settlement time period and made comparisons
with 1 and 3 weeks (1 week means larvae could settle earlier and 3 weeks means larvae
can delay settlement if suitable substrate is not available). In addition, for each spatial
unit of analysis within the PP corridor, we obtained the relative abundance of particles
(Wi) to the total of particles arriving in the corridor and to area of the spatial unit of
analysis (Ai),
92


nP
Wi =  n = 6 i

 ∑ nPi
 i =1


 * 100 * A −1
i



I = {1.2...,6}
(1)
where n is the number of particles at each spatial unit, and I is the set of 6 spatial
units of analysis at the PP corridor: La Cholla, Sandy Beach, Las Conchas, Los Tanques,
San Jorge Island, and San Francisquito (Fig. 1b). We identified the sources contributing
the most into each spatial unit.
Population genetic structure: Microsatellite markers analysis
We collected and preserved in 70% ethanol ~1 g of adductor muscle from S.
calcifer individuals from 6 localities (populations) where particles were released,
including four localities within the PP corridor (La Cholla, Los Tanques, San Jorge
Island, San Francisquito) and two sites located southwards (Puerto Lobos and
Desemboque de los Seris) (Fig. 1a). We extracted genomic DNA using the DNeasy blood
and tissue kit (Qiagen). Following Munguia-Vega et al. (2010), we amplified all samples
by the polymerase chain reaction (PCR) and genotyped every individual at 10
polymorphic nuclear microsatellite DNA loci (Spca4, Spca9, Spca16, Spca34, Spca61,
Spca1A, Spca9B, Spca12, Spca24 and Spca39). We performed genotyping on an ABI
PRISM 3730XL Genetic Analyzer (Applied Biosystems), estimated allele sizes using
GENOTYPER 3.7 (Applied Biosystems) and classified them into bins with FLEXIBIN
(Amos et al. 2007).
93
We estimated Deviations from Hardy-Weinberg equilibrium per locus and
population, and linkage equilibrium between pairs of loci by means of FSTAT 2.9.3.2
(Goudet 1995). We obtained adjusted p values using a sequential Bonferroni test for
multiple comparisons (Rice 1989) with a α = 0.05. We calculated the number of alleles
(NA), the effective number of alleles (NE) (which corrects for differences in sample size
among samples), and observed (HO) and expected (HE) heterozygosities, in GENALEX
(Peakall & Smouse 2006). Then, we calculated levels of genetic differentiation (i.e.
differences in allele frequencies) among pairs of sampled localities with the fixation
index (Fst) and the standardized measure Gst’ (Hedrick 2005) that measures true
differentiation for highly polymorphic loci such as microsatellites (Jost 2008). We
calculated mean Fst and Gst values and 95% confidence intervals (CI) using the software
of Neff and Fraser (2010), resampling individuals and loci 1,000 times.
We estimated the spatial scale of genetic structure with a spatial autocorrelation
analysis implemented in GENALEX. This method uses both pairwise geographical and
squared genetic distance matrices from individual samples to generate a spatial
autocorrelation coefficient (r) among individuals within five even-distance bins (50 km
each). We selected the number and size of bins to ensure that at least 30 data points
occurred in each bin (Rossi et al. 1992). The coefficient provides a measure of the genetic
similarity between pairs of individuals within each bin, allowing the identification of
critical distances beyond which dispersal effects are weak (i.e. genetic neighborhood
size) (Wright 1978). Statistical significance was achieved by 1,000 permutations and
1,000 bootstrap replicates to estimate 95% CI. The method employs a multivariate
94
approach that simultaneously assesses alleles and loci, strengthening the spatial signal by
reducing stochastic noise. This method is more powerful for detecting genetic structure
over space than the traditional analysis of isolation-by-distance (Smouse & Peakall
1999).
We employed the Bayesian clustering method implemented in GENELAND
(Guillot et al. 2005, Guillot et al. 2008) to estimate the number of genetic clusters (K)
present throughout the geographical range of the sampled localities. GENELAND uses
the geographical coordinates of each sampled individual to calculate geographic distances
and modify the probability that any two individuals are assigned to the same cluster; as
the geographic distance between them decreases the probability of assignment to the
same cluster also diminishes. We generated 10 independent runs of 1 x 106 Markov chain
Monte Carlo (MCMC) iterations, sampling every 1,000th, with a number of predefined
genetic clusters (K) ranging from 1 to 6 (i.e. the number of localities sampled). This was
performed by means of the spatial model with null alleles, with a maximum of 300
nuclei, without uncertainty on coordinates, and correlated allele frequencies to improve
the detection of subtle genetic structure (Guillot 2008). We discarded the first 25% of the
run as the burn-in period, and sampled the posterior distribution of K values to estimate K
mode.
For comparison with the GENELAND model, we used the software
STRUCTURE 2.3.1 (Pritchard et al. 2000, Hubisz et al. 2009). We used a categorical
variable for defining sampling location to modify the prior distribution for each
individual's population assignment. This model allows for structure to be detected at the
95
low levels of divergence that characterize many marine populations (Palumbi 2003, see
Results Section) without biasing towards detecting structure when it is not present
(Hubisz et al. 2009). We performed 10 independent STRUCTURE runs for a number of
K ranging from 1 to 6. In each run, we included a burn-in period of 1 x 105 MCMC
repetitions followed by 1 x 105 repetitions to sample from the posterior distribution. We
assumed admixture and correlated allele frequencies as recommended for cases of low
differentiation (Falush et al. 2003). To estimate the most likely number of K, we used the
∆K method (Evanno et al. 2005) as implemented in STRUCTURE HARVESTER
(http://taylor0.biology.ucla.edu/struct_harvest). Because independent runs of the most
likely K value may obtain similar membership coefficient estimates but with a different
cluster label (i.e. "label switching"), we used the "greedy" search in CLUMPP 1.1
(Jakobsson & Rosenberg 2007) to permute the coefficient matrices so that all replicates
had as close a match as possible, and calculated a mean value across the 10 runs. We
entered the CLUMPP’s output into DISTRUCT 1.1 (Rosenberg 2004) to obtain the
display of each individual membership coefficient in the form of bar plots. We
considered individuals with an assignment probability < 50% to the population where
they were sampled as misassigned and with a predominantly migrant ancestry.
Finally, we tested for evidence of recent reductions in effective population size
using the software BOTTLENECK (Cornuet & Luikart 1996). In a recently bottlenecked
population, such as those subject to strong fishing pressure, the observed heterozygosity
is higher than the expected equilibrium heterozygosity computed from the observed
number of alleles. We used both the step-wise and the two-phase mutation models with
96
1,000 replicates each, and tested for statistical significance with a Sign test and a
Wilcoxon sign rank test.
Temporal and spatial variation of spat abundance
In 2007 and 2008, we estimated recruitment of S. calcifer spat in the area of PP
throughout the spawning season. We evaluated recruitment on artificial collectors
deployed on each spatial unit covering ~75 km of coastline within the PP corridor (Fig.
1b). These sites corresponded with the location of particle releasing sites previously
described. At each site, we deployed three lines (Ø= 0.8 cm) moored by one end to an
anchor and by the other end to a plastic-buoy (buoyancy = 5 kg) to maintain each line in a
vertical position. We enclosed collectors (polyethylene Netlon®: 200 x 40 cm; mesh
opening = 7x12 mm) into a plastic-bag (60 x 40 cm; mesh opening = 0.8x1 mm) and tied
each bag to the main rope at 1, 3, 5, and 7 m from the bottom. This outer bag retained any
detached spat, most likely to occur in sites with strong currents. These Netlon® nets and
bags are commonly used for collecting commercial pectinids’ larvae worldwide
(Shumway & Parsons 2006).
We deployed collectors on June 22nd 2007, and June 27th 2008, and retrieved them
after 2 months by means of SCUBA-diving. We detached the spat by washing the
collectors and sieving each sample through a 250-µm nylon mesh, and transferred it to a
plastic pan to be counted under a dissecting scope. We used a repeated measures analysis
of variance (longitudinal study) to test significant differences in the number of recruited
spat between years, sites and depth as well as combined factors. We applied a natural
97
logarithmic transformation [i.e. ln (x+1)] to the numbers of spat per collector to meet
normality and homogeneity of variances. For 2007, we obtained the abundance of settled
spat (Oi) on each collection site relative to the total number of spat settled within the PP
corridor and correlated this estimation with the values of relative abundance of particles
(Wi) as predicted by the model (Eq. 1).
Results
Spatial units of analysis and selection of particle release sites
For the whole NGC we generated 56 spatial units of analysis with estimated areas
ranging from 10 to 361 km2, of which 49 were adjacent to the coastline (Fig. 1a). Based
on fishers’ knowledge from four fishing communities (Bahía de los Ángeles, Puerto
Libertad, Puerto Lobos, and Puerto Peñasco) we selected key fishing beds as sites to
release particles. Most of these fishing beds were located along the coast of the states of
Sonora and Baja California (Fig. 1a-c).
CBOM outputs for the PP corridor
Overall, the cloud of particles showed a cyclonic dispersion (Fig. 2). At 2 weeks,
for the lower dispersal case (position at low tide of active particles released at neap tide),
we observed for the PP corridor a significant influence of particles from southern remote
sources such as Puerto Lobos (15-93% of the particles arriving to each spatial unit) and
Puerto Libertad (<5-72%), located ~100 and ~150 km south of the corridor, respectively
(Fig. 2). For the higher dispersal case (position at high tide of passive particles released at
98
spring tide), we observed an even higher incidence (about 5% more) of particles arriving
to the corridor from these distant sources. The output for the higher dispersal case
resulted in the occurrence of particles from more remote areas ( X higher dispersion case
= 82.8 km,
X
lower dispersion case = 78.8 km at 2 weeks, range 1-18 km more, except
for SJO), reaching the PP corridor as the dispersion allowed particles to travel longer
distances from southern sources (Table 1, Fig. 2).
For those sites contributing larvae into the PP corridor, the estimated mean lineal
distance of dispersal at 2 weeks ranged between 50 and 135 km depending on the site of
origin (Table 1). Two sites located in the northwest limit of the PP corridor (La Cholla
and Sandy Beach) showed a lower mean larval dispersal (66-72 km) compared to central
sites (77-93 Km, Las Conchas, Los Tanques). San Jorge Island showed the lowest mean
larval dispersal distance (47-56 km) at all times. Puerto Lobos and Puerto Libertad
showed the highest mean lineal distance (109-135 km) (Table 1).
At 2 weeks, we observed for both dispersal cases that San Jorge and Puerto Lobos
were the two main sources of particles reaching La Cholla, accounting on average for
45% of the particles from each source, and <5% from San Francisquito and Puerto
Libertad, respectively (Fig. 3b). For Sandy Beach, Las Conchas, and Los Tanques, the
remote site of Puerto Lobos was the main source of particles (68-93%), with minor inputs
from San Jorge Island (4-22%, except for los Tanques), San Francisquito (10-18%), and
Puerto Libertad (1-25%). Particles released from Puerto Libertad greatly contributed to
San Jorge Island (up to 85%) and San Francisquito (72%), followed by particles released
99
from Puerto Lobos (~25%) (Fig. 2 & 3b). San Francisquito was the only spatial unit
receiving particles from Las Cuevitas and only for the higher dispersion case only.
Assuming that larvae are capable of reaching the competency period in 1 week,
our model suggests that San Francisquito and San Jorge Island (in that order) were the
main sources of particles seeding La Cholla, Sandy Beach, and Las Conchas (70-80%
San Francisquito and 35-50% San Jorge Island, Fig. 3a). Particles released from the
southeastern side of the PP corridor showed a flow towards the coast while those released
from San Jorge Island showed an off-shore pathway (Fig. 2). For Los Tanques, San
Francisquito was the main source, accounting for ~75% of the particles. More than 90%
of the particles arriving at San Jorge Island originated in Puerto Lobos, particles from this
former site was the sole source of particles reaching San Francisquito (Table 1 & Fig.
3a).
If we assume that larvae can delay settlement until reaching the 3rd week, San
Jorge Island, Puerto Lobos, and Puerto Libertad would be contributing equally to
northwestern areas of the PP corridor. The southeastern limit of the corridor receives
particles mainly from Puerto Libertad and Las Cuevitas (Fig. 2 & 3c). We observed that
the central area (Las Conchas and Los Tanques) showed characteristics of a transition
zone, primarily receiving particles from southern areas (Fig. 3c). During the 1-3 weeks
modeled, particles from the southernmost site (Desemboque de los Seris) never reached
the PP corridor in any detectable frequency.
100
With the exception of San Jorge Island and San Francisquito, the model generally
showed that particles released from all sites within the PP corridor were exported to
downstream sites, outside the former reserve network, and towards northwest regions
such as El Borrascoso, the Upper Gulf of California reserve, and the Vaquita Refuge
(Fig. 2-4). At 2 weeks we observed that El Borrascoso is mainly seeded by larvae
released from sites within the PP corridor and from a distant source at Puerto Lobos,
while El Borrascoso is the primary source of larvae for the Vaquita Refuge (Fig. 2 & 4).
In terms of the potential sources for Puerto Lobos and Puerto Libertad, at 1 week
we observed that Puerto Libertad is the only source for Puerto Lobos with minor
contribution from Las Cuevitas (<5%), while particles released from southern reefs such
as Las Cuevitas and Desemboque de los Seris are the main sources (70 and 30%,
respectively) arriving at Puerto Libertad. Minor contributions from the northern region of
Tiburon Island were also observed. At 2 and 3 weeks, the model predicts that particles
from Las Cuevitas (80 and 50%, respectively) and other southern sites including
Desemboque de los Seris (10-45%, respectively), will reach Puerto Lobos (Fig. 2 & 5).
Population genetic structure: Microsatellite markers analysis
We genotyped 176 individuals from 6 localities at each of 10 microsatellite loci.
Because locus Spca24 was duplicated in some of the sampled localities making
impossible to accurately assign alleles to a particular locus, we excluded it from the
analysis. From the remaining 9 loci, we did not find any evidence of significant
deviations from Hardy-Weinberg equilibrium at 54 tests performed for each locus in each
101
population (all p values > 0.0009). From a total of 204 tests of departures from linkage
equilibrium covering all pairs of loci and localities, we did not find any instance of
significant disequilibrium (p > 0.0002). Therefore, we assumed loci to be independent in
all subsequent analyses.
We observed moderate levels of genetic variation among sampled localities. The
mean number of alleles (NA) varied from 8.3 to 9.7 while the effective mean number of
alleles (NE) ranged from 4.2 to 4.8. Mean observed (HO) and expected (HE)
heterozygosites ranged from 0.504 to 0.593 and 0.579 to 0.668, respectively (Table 2).
Over all sampled populations, we detected very low levels of population genetic structure
(Fst = 0.008; 95% CI -0.001 to 0.021). Among pairs of localities, San Francisquito-Los
Tanques showed the lowest mean levels of population genetic structure (Fst = 0.020;
95% CI = -0.017, 0.080 and Gst' = 0.039; 95% CI = -0.025, 0.138), while San
Francisquito-San Jorge Island had the highest average levels of differentiation (Fst =
0.044; 95% CI = -0.003, 0.129 and Gst' = 0.086; 95% CI = -0.004, 0.228) (Table 3).
Based on the analysis of spatial autocorrelation (r), we observed that only
individuals within the first distance bin (0-50 km) had a significant positive correlation,
suggesting that S. calcifer individuals were genetically more similar than expected by
random (r = 0.008, p = 0.01) (Fig. 6). We did not find significant genetic spatial structure
for bins 50-100, 100-150, and 150-200 km. The last bin (200-250 km) showed significant
negative correlation (r -0.008, p = 0.02). The distance at which the positive
autocorrelation broke down was estimated at 88.1 km (i.e. r = 0).
102
We consistently estimated a mode of two genetic clusters after performing 10
independent GENELAND runs (Fig. 7a). One genetic cluster was composed of La
Cholla, Los Tanques, and San Francisquito, while the second cluster included San Jorge
Island and the two southern localities of Puerto Lobos and Desemboque de los Seris (Fig.
7b). However, according to GENELAND, very low levels of genetic differentiation (Fst
= 0.007) were present between these two clusters. On the other hand, the standard
STRUCTURE model (Pritchard et al. 2000, Falush et al. 2003) was unable to detect
population structure between the samples because of the low levels of differentiation
present in the data (pairwise Fst < 0.044, Table 3) (Latch et al. 2006, Hubisz et al. 2009).
Using information about sampling location to modify the probability that any two
individuals are assigned to the same cluster (Hubisz et al. 2009), we estimated the highest
mean value of ln probability of data (and lowest variation around the estimate) for K = 1
(average ln[K] = -4,966.21, Fig.8a), suggesting the absence of population structure
throughout the sampled localities. The ∆K method suggested that K = 3 (average ln[K] =
-5,001.88) was the most likely value. However, it should be noted that this method is
known to be unable to find the best K if the true K = 1 (Evanno et al. 2005), as suggested
by the ln probability of the data. For comparison with the GENELAND output, the mean
individual assignment probabilities are shown when K = 2 (average ln[K] = -4992.23,
Fig. 8b). Here, we observed a subtle transition from one predominantly southern genetic
cluster (shown in red) in Desemboque de los Seris (93.7% of individuals assigned to the
southern cluster) and Puerto Lobos (96.7%), to a relative lower proportion of individuals
103
assigned to this southern cluster in the rest of the sampled populations (San Jorge Island
72.2%; San Francisquito 65.65%; Los Tanques 87.5%; and La Cholla 81.2%).
According to BOTTLENECK, we estimated that La Cholla, Los Tanques, and
San Francisquito showed strong signatures of a recent reduction in population size (all p
values < 0.05, Sign test and a Wilcoxon sign rank test under both mutation models)
(Table 2). We also estimated some evidence of a bottleneck for the other three localities,
but they were not consistent among statistical tests and/or mutation models (Table 2).
Temporal and spatial variation of spat abundance
Overall, recruitment of S. calcifer spat did not differ between spawning seasons
(repeated measures 3-way ANOVA; F1, 65 = 2.22, p = 0.14). We found that spat
recruitment for each year was significantly influenced by site location (repeated measures
3-way ANOVA; F1, 5 = 7.42, p < 0.001) (Fig. 9). At each depth (considering all localities)
we did not find significant differences between years (repeated measures 3-way
ANOVA; F1, 65 = 0.32, p < 0.86). In 2007, 70% of all recruited spat for the whole corridor
corresponded to spat recruited at San Jorge Island and San Francisquito (both sites
pooled). However, in 2008 this value was reduced to 55% for these localities taken
together, and spat recruitment at La Cholla represented 25% (Fig. 9). Both years, the
mean number of spat recruited per collector at San Jorge Island and San Francisquito
decreased markedly near the bottom (Fig. 10). In 2007, San Jorge Island showed on
average 650.7 spat collector-1 (SD = 462.1) and San Francisquito 696.4 spat collector-1
(SD = 89.1) (Fig. 10). In 2008, we observed similar values for San Jorge Island (x = 730,
104
SD = 54.6), but San Francisquito showed a 50% decrease in the mean number of spat
recruited per collector (x = 243.7; SD = 87.8). This inverse pattern was also found at La
Cholla and Los Tanques in 2007 and at Sandy Beach in 2008 (Fig. 10). Both years, we
observed the lower recruitment values (10-250 spat collector-1) at Sandy Beach, Las
Conchas, and Los Tanques (Fig. 10).
For summer 2007, at 1 week of pre-competency period, both lower and higher
dispersion cases showed a lack of correlation between the relative abundance of recruited
spat (Oi) and the relative abundance of predicted particles (Wi) (higher dispersion case: r
= 0.360, p = 0.484; lower dispersion case: r = 0.361, p = 0.482). At 2 weeks, Oi and Wi
showed a significant correlation (higher dispersion case: r = 0.920, p = 0.009; lower
dispersion case: r = 0.894, p = 0.016). At 3 weeks, we observed a similar significant
positive correlation between Oi and Wi for the higher dispersion case only (r = 0.886, p =
0.019) (Fig. 11).
Discussion
Connectivity patterns in the NGC
Understanding dispersal pathways and the spatial scale of connectivity between
sites are important steps for the appropriate siting of marine reserves (Halpern 2003,
Palumbi 2003). Along the eastern side of the NGC, larvae dispersion pathways during
summer are driven by the borders of the basin-wide seasonally-reversing eddy that
dominates the large-scale circulation, where in this area can be described by an axis
105
oriented from southeast sources towards northwest areas. Alongshore, coastal currents
promote a complete asymmetry in this directional transport of larvae.
In this study, the indirect measures of genetic connectivity suggested very low
levels of genetic structure among pairwise comparisons of sampled localities (Fst <0.05
and Gst’ <0.09, Table 3). Fst and Gst’ estimates were approximately proportional to each
other in each comparison, suggesting that the low Fst values observed are real and not an
artifact of highly variable loci (Jost 2008). Although Fst and Gst’ estimates depend on the
absolute number of immigrants (Nm) and do not inform about their proportion (m)
relative to the entire population size (Hedgecock et al. 2007, Lowe & Allendorf 2010),
the low genetic differentiation observed suggest that at least a few larvae disperse, settle
and reproduce across the entire study area once in a while. In theory, one effective
migrant per generation between distant localities is enough to ensure the same alleles are
shared among populations over long periods of evolutionary time (Wright 1978). This
would preserve the adaptive potential of the populations and avoid the harmful effects of
genetic drift and inbreeding. Thus, levels of genetic differentiation are heavily influenced
by rare long-distance dispersal rather than average levels of demographic connectivity.
Our results imply the presence of adaptive and inbreeding genetic connectivity (see Lowe
& Allendorf 2010) between the PP corridor and the reefs located ∼200 km south, but do
not distinguish if moderate levels of gene flow or even random mating are present
(Palumbi 2003, Hedgecock et al. 2007). High levels of genetic differentiation are
expected only in those cases where gene flow is severely reduced during multiple
continuous generations, a case not observed in S. calcifer at the spatial scale studied.
106
Predicted values from CBOM’s outputs suggests strong demographic connectivity
between the PP corridor and southern sources such as Puerto Lobos (located 150 km
south), which was supported by genetic data. CBOM’s outputs also suggest that sources
located further south, such as Puerto Libertad and Las Cuevitas (~175 km away), likely
have weak demographic links with the corridor; while Desemboque de los Seris (~200
km away) is highly unlikely as a source to the corridor even after 3 weeks (Fig 3). This
apparent contradiction between CBOM and genetic outputs for Desemboque de los Seris
could be due to the limitation of the model to represent stochastic events (e.g. El Nino
Southern Oscillation events) that might disperse potential larvae from remote sources
under such extreme conditions. Another important issue might be related to the limited
number of particles released for each site (400) compared to the real number of larvae
produced by all the individuals within a site. Attention to this complexity require further
studies because a weak representation could lead to an underestimation of the likely small
frequency of long distance dispersers that drive the Fst and Gst values. If we consider a
normal distribution of the distance traveled by the particles released in Desemboque de
los Seris after 3 weeks (Table 1), the 99.0% CI reaches 159.93 km, while the 99.9% CI
includes 187.21 km. Thus, although the contribution of Desemboque de los Seris to the
corridor is probably of small demographic importance as suggested by the CBOM, it is
certainly biologically and oceanographically plausible that a few larvae from that locality
reach the corridor.
Genetic data also suggests that the pre-competency period for S. calcifer larvae in
the field is likely similar to that observed in laboratory (2 weeks, Soria et al. 2010) or
107
larger (3 weeks). On the other hand CBOM’s outputs for 1 week fail to explain, for
instance, the low genetic differentiation observed between even moderately distant
southern reefs (e.g. Puerto Lobos) and the entire corridor.
The spatial scale of connectivity suggested by CBOM’s outputs is in general
agreement with the spatial scale of average genetic similarities among pairs of
individuals. For a pre-competency period of 2 weeks, the estimated mean lineal distance
of dispersal from the CBOM ranged between 50 and 110 km depending on the site of
origin ( X higher dispersion case = 82.8,
X
; lower dispersion case = 78.8 km, Table 1).
Similarly, the spatial threshold where genetic similarity among individuals disappears
was estimated at 88.1 km (Fig. 6). Therefore, both genetic and CBOM spatial scales of
connectivity are in agreement suggesting that, on average, demographic and genetic
connectivity decreases when the spatial scale is >100 km. Our findings are consistent
with larval dispersal distances (range 30-100 km) estimated for mussels through
elemental fingerprints (Becker et al. 2007), rate of range expansion (McQuaid & Phillips
2000), and genetic markers (Gilg & Hilbish 2003). Also, mean larval dispersal distances
for a variety of fish and invertebrates with pelagic larvae have been estimated to fall
within 25 and 150 km (reviewed by Palumbi 2003), and the acorn barnacle Balanus
glandula, with a minimum planktonic larval duration of two weeks, has an average
dispersal of ~70 km (Sotka & Palumbi 2006).
Our estimation of genetic connectivity by direct methods (Bayesian assignment
methods) supports the existence of two subtly differentiated genetic clusters throughout
the sampled area along the eastern side of the NGC (Fig. 7 and 8). These stocks could be
108
described as a southern stock, including San Jorge Island, Puerto Lobos, and
Desemboque de los Seris, with a comparatively reduced gene flow towards a northern
downstream stock represented by a section of the corridor, including San Francisquito,
Los Tanques, and La Cholla. Migration rates larger than one migrant per generation -and
consequently Hardy-Weinberg equilibrium- are needed within each cluster (but not
between them) to maintain nearly identical allelic frequencies (Lowe & Allendorf 2010)
and explain the identification of two distinct genetic clusters through Bayesian methods.
It is noteworthy that sampled S. calcifer individuals from San Francisquito and San Jorge
Island, located close to each other (<10 km), showed the highest levels of genetic
structure observed among the sampled localities (Table 3). Hence, larvae exchange
between these sites appears to be restricted compared to other sites. Larvae produced at
San Francisquito might be dispersed following a nearshore pathway, while larvae from
the island might be following an offshore dispersal as suggested by the CBOM (Fig. 2).
Likewise, in spite of San Jorge Island appearing as a minor source for the northwest sites
of the corridor (La Cholla, Sandy Beach and Las Conchas) in the CBOM, which
contradicts the spatial location of the northern genetic cluster identified, the CBOM also
suggested a lack of dispersal between the island and Los Tanques, located 25 km to the
north. This latter observation is consistent with the location of the boundary between the
two genetic clusters identified. These observations suggest that the dynamics of larvae
dispersal in nearshore coastal waters over a small geographic scale is complex, and the
enhanced CBOM might reflect in some instances, but not in others, the true dynamics of
larvae dispersal. Similarly, weak genetic structure observed in three coastal marine
109
species along the west coast of California and Baja California were found ecologically
meaningful when correlated with other environmental (e.g. temperature) and ecological
parameters (e.g. kelp distribution) instead of oceanographic pathways (Selkoe et al.
2010).
In summary, since we estimated low levels of genetic structure among sampled
localities, without the development of a spatially explicit CBOM we would have been
unable to estimate a relative degree of importance for upstream sources contributing into
the corridor. Subtle gene structure in marine invertebrate populations have been reported
(Palumbi 2003, Kenchington et al. 2006, Selkoe et al. 2010). However, as noted by Lowe
& Allendorf (2010), most genetic studies provide little or inconclusive information on
demographic connectivity, the type of connectivity that is most useful for fisheries
management. It is in these types of situations when CBOMs become an important tool to
complement population genetic studies and vice versa, helping to visualize the relative
strength of connectivity among sites, as in this study. For example, after coupling both
analyses we were able to estimate the relative contribution of larvae between Puerto
Lobos and Desemboque de los Seris. Whereas protecting Puerto Lobos may be more
relevant for fisheries management purposes, protecting both sites could be critical if the
long-term conservation of evolutionary processes and adaptive potential is the main
interest. Because estimations on gene structure are highly sensitive to rare long-distance
dispersal, it is suggested that CBOMs should include those oceanographic situations
where particles could be dispersed the most in order to best estimate the longest tail of
110
particle distributions. Addressing these parameters would be even more important if a
higher spatial scale, such as the entire NGC, is the subject of analysis.
Temporal and spatial variation on spat recruitment
In this study, we observed a high consistency between predicted outputs from the
CBOM and observed spat (<5 mm in shell height) recruitment on artificial collectors for
the whole corridor for 2 and 3 weeks (Fig. 11). These significant correlations can be
explained only if the reproductive timing throughout all localities is highly synchronized
and enough time (2-3 weeks) is allowed for larvae from upstream sources to reach the
corridor. For the corridor, a gradient of higher values of both predicted particles and
observed densities of S. calcifer juveniles (<100 mm in shell height) were previously
reported for the northern sites of La Cholla and Sandy Beach, while lower abundances
were estimated for the central areas (Las Conchas and Los Tanques) (Cudney-Bueno et
al. 2009). Our results underscore this previously reported gradient between central and
northern sites, and also indicate that San Jorge Island and San Francisquito are sites
where spat recruitment is significantly higher in comparison with other sites. In both
years, San Jorge Island showed the highest spat recruitment, accounting for 35% and
45% of the total recruited spat in the corridor (Fig. 9). This higher availability of larvae
might explain the significantly higher densities of S. calcifer individuals (both juveniles
and adults) measured around the island (up to 1.6 scallops m-2), in comparison to central
and northwest sites (<0.1 scallops m-2) (Cudney-Bueno et al. 2009, Martínez-Tovar
2010). However, spat abundance may not be correlated with juvenile densities because
111
post-dispersal density dependent factors can ultimately shape juvenile densities with
unexpected results (Orensanz et al. 2006). Indeed, Cudney-Bueno et al. (2009) showed
little change in density of juveniles in San Jorge Island after reserve establishment
Nonetheless, we successfully applied a methodology to collect S. calcifer spat
which could be used to perform further studies to test the origin of the spat through of
elemental fingerprinting or population genetic analyses to address this open question
(Becker et al. 2007, Werner et al. 2007), or to implement stock enhancement programs
through massive collection and seeding of spat.
Implications for Fisheries Management and Conservation
Spondylus calcifer is a sessile broadcast spawner that relies on external
fertilization and is highly dependent upon the proximity of spawning individuals. We
documented the existence of commercial beds throughout the eastern side of the NGC. In
this region, the three northernmost sites (La Cholla, Los Tanques, and San Francisquito)
showed strong signs of a recent reduction in population size, most likely attributable to
fishing pressure. Due to the higher proximity of these fishing beds to the fishing town of
PP, these sites might be exposed to comparatively higher levels of fishing pressure
compared to San Jorge Island, Puerto Lobos, and Desemboque de los Seris. Thus, the
reproductive success of the species would likely benefit from having a high density of
adults to diminish any depensatory density effect on reproductive success due to
overfishing. Although spat recruitment levels may fluctuate highly between consecutive
years, higher adult densities are often correlated with higher spat recruitment levels in
112
most scallop fisheries (Orensanz et al. 2006). In the case of S. calcifer from the study
area, growth overfishing problems that may occur when juvenile animals are overfished,
may be neglected because animals are fished above the minimum reproductive size
(Cudney-Bueno & Rowell 2008). Therefore, to increase the reproductive fitness of the
species and reduce potential depensatory effect on fertilization success, conservation and
management efforts should focus on season closures and protecting key beds of S.
calcifer from fishing mortality. The current management plan for S. calcifer in the
corridor follows this advise recommending that fishing sites be closed to harvesting when
rock scallop density is <5 scallops 100 m-2, (Martínez-Tovar 2010).
Cudney-Bueno et al. (2009) previously suggested that San Jorge Island could be
acting as a key source for larval export to adjacent areas in the corridor, while direct
influence from southern sources was neglected. Our findings provide further evidences in
support of the idea that San Jorge Island is a larval source for downstream beds.
However, our results provide additional insights for the management and conservation of
this species, suggesting that S. calcifer beds within the corridor could also share multiple
sources of larvae south of San Jorge Island, including San Francisquito, Puerto Lobos,
and likely Desemboque de los Seris. Consequently, larvae recruitment within the corridor
may not be entirely due to local production and could be linked to a larger spatial scale of
larval inputs, including subpopulations from at least (but not limited to) Puerto Lobos
(~100 km south of the corridor). When marine reserves were informally implemented in
the area, Cudney-Bueno et al. (2009) suggested compensatory effects as a plausible
explanation of the lack of increase in density of juveniles of S. calcifer in San Jorge
113
Island, even after years of protection. Our results suggest that the lack of enhanced
densities could be attributed to larger spatial scale processes involving stock-recruitment
relationships, driven by larvae production in remote areas (e.g. Puerto Lobos located
~100 km south of San Jorge Island) and temporal fluctuations on larvae advections.
Whether the lack of increase in density of S. calcifer in San Jorge Island is due to density
compensatory effects or to a significant dependence of larvae availability on southern
sources requires further study. Despite the low connectivity predicted by the CBOM, the
lack of genetic structure and the presence of allele frequency similarities between Puerto
Lobos and Desemboque de los Seris suggests that larval flows from southern sources
towards downstream beds is significant enough to dilute any geographic differences.
Within the corridor, San Jorge Island, Sandy Beach, and Las Conchas were
formerly proposed as marine reserves (Cudney-Bueno & Basurto 2009, Cudney-Bueno et
al. 2009). According to the CBOM it is likely that the island is acting as a key component
for larval export towards sites in the northwestern extreme of the corridor, such as El
Borrascoso (~80 km northwest of San Jorge Island) and La Cholla (Fig 1b & 4b), rather
than towards central sites. As for the other proposed reserves (Sandy Beach and Las
Conchas), their main contribution is also predicted to be towards western sites, but given
the short spatial proximity (<10 km) between them, the contribution from Las Conchas
into Sandy Beach is unlikely (see Fig. 2b,c). Rather, these near sites could be considered
as a single unit as suggested by the spatial scale estimated in this study. In addition, this
study also suggests that Puerto Lobos could act as an alternative site for the establishment
of marine reserves. This site holds important connectivity attributes (major contribution
114
of larvae and lower genetic differentiation with other sites within the corridor), which
make this site an interesting candidate for the establishment of marine reserves. However,
other factors need to be considered such as site specific reproductive output and densities
of adults.
It should also be noted that closing or reducing the fishing area will initially
reduce the availability of harvestable scallops, which, by shifting harvesting pressure
elsewhere, could lead to negative or unpredictable consequences on other stocks (Hilborn
et al. 2004). The impact of fishing effort re-allocation or a change in targeted species
should always be considered for the proper design of marine reserves (individually or
reserve networks) in the NGC and elsewhere. Therefore, the spatial structure of fishery
stocks and the spatio-temporal distribution of fishing activities must be considered for
siting marine reserves (Gilg & Hilbish 2003, Hilborn et al. 2004, Cudney-Bueno et al.
2009). If San Jorge Island, Puerto Lobos, San Francisquito or any other combination of
sites are to be selected as sites for marine reserves, the likely movement of fishing
pressure and the willingness of the fishing sector towards such spatially explicit
management tool must be addressed. It should be noted that in the case of S. calcifer,
permits to access this fishery are extremely limited, and the fishers with permits do not
have legal access to harvest the southern sites of this corridor. In this study, it was
noteworthy that there were relatively low differences on allele frequencies among
sampled localities, suggesting the absence of a strong barrier to migration. Consequently,
the siting of marine reserves at upstream sites would likely benefit downstream
subpopulations. Moreover, the spatial scale of demographic and genetic connectivity as
115
estimated through the CBOM and genetic analysis (~100 km) should be used as a spatial
reference for siting both upstream and downstream marine reserves in the study area.
CBOM constraints and future work
The accuracy of CBOM’s outputs could be further improved by addressing key
biological and physical factors. On the biological side, the inclusion of differences in
reproductive timing among localities and larvae attenuation due to natural mortality
would allow the model to simulate larvae delivery variations from genetically different
localities. Furthermore, larvae swimming capabilities, ontogenic depth preferences, and
differential growth rates were not addressed in this study but could be integrated as
information on these topics is made available. Also, the effect of differential fishing
pressure could be incorporated into the model to further predict the consequences of
fishing mortality upon connectivity patterns, which might affect the delivery of larvae
from genetically different sources. Finally, future genetic work could verify instances of
rare long-distance dispersal, as suggested by the CBOM outputs (e.g. between Puerto
Lobos and El Borrascoso separated by ~300 km) and the incorporation into the analyses
of other sites located in Baja California and the Midriff Archipelago Region to address if
there is indeed no connection between these sites and those found on the mainland coast.
As it was mentioned before, at the spatial resolution of the present study, CBOM would
be improved by the inclusion of extreme meteorological and oceanographic events
(hurricanes, ENSO, gyres, etc) that might represent the effect of rare long-distance
116
dispersers over medium and large temporal scales which would help to explain some of
the observed constrains between CBOM and population genetic studies.
The value of combining genetic and demographic methods for understanding the
different spatial and temporal scales of connectivity has been recently highlighted
(Hedgecock et al. 2007, Lowe & Allendorf 2010, Selkoe et al. 2010). Particularly,
indirect demographic methods (CBOM) allowed us to quantify the relative contribution
from explicit locations and identify the most likely sources of larvae that contribute
significantly to local recruitment, affecting population growth of fishing beds in the PP
corridor (i.e. demographic connectivity). On the other hand, the use of indirect and direct
genetic methods made it possible to verify the cases in which such larvae dispersal events
actually really occurred over recent time and shed some light on the magnitude of gene
flow. The results presented in this work provide novel information for the design and
siting of marine reserves as a fishery management tool in the NGC region.
Acknowledgements
We wish to thank A. Cinti, W. Shaw, T. Pfister, M. Lavin, and P. Turk-Boyer for
reviewing and providing insightful comments, and A. Macias-Duarte for his assistance
with statistical analysis. We are grateful to W. Ludt, D. Manjon, J. Hall and R. LoaizaVillanueva for their field and laboratory support. We appreciate the logistic support
provided by the cooperative Buzos de Puerto Punta Peñasco. We collected individuals
under the permit #SGPA/DGVS 01349/08 issued to the Centro Intercultural de Estudios
de Desiertos y Oceanos, A.C. by the Secretaria del Medio Ambiente y Recursos Naturales
117
(SEMARNAT) in Mexico. This work was funded by The David and Lucile Packard
Foundation, The Nature Conservancy and the Conservancy’s RJ KOSE Grant Program,
and the Wallace Research Foundation. This is a scientific contribution of the PANGAS
Project (www.pangas.arizona.edu).
118
Literature cited
Aiken CM, Navarrete SA, Castillo MI, Castilla JC (2007) Along-shore larval dispersal
kernels in a numerical ocean model of the central Chilean coast. Mar Ecol Prog
Ser 339:13-24
Amos W, Hoffman JI, Frodsham A, Zhang L, Best S, Hill AVS (2007) Automated
binning of microsatellite alleles: problems and solutions. Mol Ecol Notes 7:10–14
Arnold WS, Marelli D, Bray CP, Harrison MM (1998) Recruitment if bay scallops
Argopecten irradians in Floridan Gulf of Mexico waters: scales of coherence.
Mar Ecol Prog Ser 170:143-157
Backhaus JO (1985) A three-dimensional model for the simulation of the shelf sea
dynamics. Dtsch Hydrogr Z 38:165-187
Beaumont A (2006) Genetics. In: Shumway SE, Parsons GJ (eds) Scallops: Biology,
Ecology and Aquaculture, Vol 35. Elsevier, Amsterdam, p 543-594
Becker BJ, Levin LA, Fodrie FJ, McMillan PA (2007) Complex larval connectivity
patterns among marine invertebrate populations. PNAS 27:3267-3272
Beukers-Stewart BD, Vause BJ, Mosley MWJ, Rossetti HL, Brand AR (2005) Benefits
of closed area protection for a population of scallops. Mar Ecol Prog Ser 298:189204
Carr S, Capet X, McWilliams J, Pennington T, Chavez F (2008) The influence of diel
vertical migration on zooplankton transport and recruitment in an upwelling
region: estimates from a coupled behavioral-physical model. Fish Oceanogr 17:115
Carrillo L, Palacios-Hernández E (2002) Seasonal Evolution of the Geostrophic
Circulation in the Northern Gulf of California. Est Coast Shelf Sci 54:157-173
119
CONANP (2009) Mapa de Áreas Naturales Protegidas. Accessed 20 June 2009.
http://www.conanp.gob.mx/mapa_anp.html
Cornuet JM, Luikart G (1996) Description and power analysis of two tests for detecting
recent population bottlenecks from allele frequency data. Genetics 144:2001-2014
Crowder LB, Lyman SJ, Figueira WF, Priddy J (2000) Source-sink population dynamics
and the problem of siting marine reserves. Bull Mar Sci 66:799-820
Cudney-Bueno R, Basurto X (2009) Lack of cross-scale linkages reduces robustness of
community-based fisheries management. PLoS ONE 4:e6253.
doi:6210.1371/journal.pone.0006253
Cudney-Bueno R, Lavín MF, Marinone SG, Raimondi PT, Shaw WW (2009) Rapid
effects of marine reserves via larval dispersal. PLoS ONE 4(1):e4140.
doi:4110.1371/journal.pone.0004140
Cudney-Bueno R, Rowell K (2008) Establishing a baseline for management of the rock
scallop Spondylus calcifer (Carpenter 1857): Growth and reproduction in the
Upper Gulf of California, Mexico. J Shellfish Res 27:625-632
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals
using the software structure: a simulation study. Mol Ecol 14:2611-2620
Falush D, Stephens M, Pritchard JK (2003) Inference of population structure using
multilocus genotype data: Linked loci and correlated allele frequencies. Genetics
164:1567-1587
Fogarty MJ, Botsford LW (2007) Population connectivity and spatial management of
marine fisheries. Oceanography 20
Gell FR, Roberts CM (2003) Benefits beyond boundaries: the fishery effects of marine
reserves. Trends Ecol Evol 18:448-455
Gilg MR, Hilbish TJ (2003) The geography of marine dispersal: coupling genetics with
fine-scale physical oceanography. Ecology 84:2989-2998
120
Goudet J (1995) FSTAT (Version 1.2): A computer program to calculate F-Statistics. J
Hered 86:485-486
Guillot G (2008) Inference of structure in subdivided populations at low levels of genetic
differentiation--the correlated allele frequencies model revisited. Bioinformatics
24:2222-2228
Guillot G, Estoup A, Mortier F, Cosson JF (2005) A spatial statistical model for
landscape genetics. Genetics 170:1261-1280
Guillot G, Santos F, Estoup A (2008) Analyzing georeferenced population genetics data
with Geneland: a new algorithm to deal with null alleles and a friendly graphical
user interface. Bioinformatics 24:1406-1407
Halpern BS (2003) The impact of marine reserves: Do reserves work and does reserves
size matters? Ecol Applications 13:117-137
Hedgecock D, Barber PH, Edmands S (2007) Genetic approaches to measuring
connectivity. Oceanography 20:70-79
Hedrick PW (2005) A standardized genetic differentiation measure. Evolution 59:16331638
Hellberg ME, Burton RS, Neigel JE, Palumbi SR (2002) Genetic assessment of
connectivity among marine populations. Bull Mar Sci 70:273-290
Hilborn R, Stokes K, Maguire J-J, Smith T and others (2004) When can marine reserves
improve fisheries management? Ocean Coast Manage 47:197-205
Hubisz MJ, Falush D, Stephens M, Pritchard JK (2009) Inferring weak population
structure with the assistance of sample group information. Mol Ecol Resources
9:1322-1332
Jakobsson M, Rosenberg NA (2007) CLUMPP: a cluster matching and permutation
program for dealing with label switching and multimodality in analysis of
population structure. Bioinformatics 23:1801-1806
121
Jones GP, Srinivasan M, Almany GR (2007) Population connectivity and conservation of
marine biodiversity. Oceanography 20:100-111
Jost L (2008) Gst and its relatives do not measure differentiation. Mol Ecol 17:4015-4026
Kaplan DM (2006) Alongshore advection and marine reserves: consequences for
modeling and management. Mar Ecol Prog Ser 309:11-24
Kenchington EL, Patwary MU, Zouros E, Bird CJ (2006) Genetic differentiation in
relation to marine landscape in a broadcast-spawning bivalve mollusc
(Placopecten magellanicus). Mol Ecol 15:1781-1796
Latch EK, Dharmarajan G, Glaubitz JC, Rhodes OE (2006) Relative performance of
Bayesian clustering software for inferring population substructure and individual
assignment at low levels of population differentiation. Conservation Genet 7:295302
Lester SE, Halpern BS, Grorud-Colvert K, Lubchenco J and others (2009) Biological
effects within no-take marine reserves: a global synthesis. Mar Ecol Prog Ser
384:33-46
Lipcius RN, Crowder LB, Morgan LE (2005) Metapopulation structure and marine
reserves. In: Norse E, Crowder L (eds) Marine Conservation Biology: The
Science of Maintaining the Sea's Biodiversity, Vol 19. Island Press, p 328-345
Lowe WH, Allendorf FW (2010) What can genetic tell us about population connectivity?
Mol Ecol 19:3038-3051
Marinone SG (2003) A three-dimensional model of the mean and seasonal circulation of
the Gulf of California. J Geophys Res 108:1-25
Marinone SG (2008) On the three-dimensional numerical modeling of the deep
circulation around Angel de la Guarda Island in the Gulf of California. Est Coast
Shelf Sci 80 430–434
122
Marinone SG, Ulloa MJ, Pares-Sierra A, Lavin MF, Cudney-Bueno R (2008)
Connectivity in the northern Gulf of California from particle tracking in a threedimensional numerical model. J Mar Sys 71:149-158
Martínez-Tovar I (2010) Estimación de la densidad y tasa de aprovechamiento del callo
de escarlopa, Spondylus calcifer en Puerto Peñasco, Sonora, Centro Intercultural
de Estudios de Desiertos y Océanos, Puerto Peñasco
McQuaid CD, Phillips TE (2000) Limited wind-driven dispersal of intertidal mussel
larvae: in situ evidence from the plankton and the spread of the invasive species
Mytilus galloprovincialis in South Africa. Mar Ecol Prog Ser 201:211-220
Moreno-Báez M, Orr BJ, Cudney-Bueno R, Shaw WW (2010) Using Fishers' local
knowledge to aid management at regional scales: Spatial distribution of smallscale fisheries in the Northern Gulf of California, Mexico. Bull Mar Sci 86:339353
Moreno C, Rojo M, Torre J (2008) Diagnóstico Socioeconómico de la Pesca Artesanal en
la Región del Norte del Golfo de California. Report No. PANGAS project,
Guaymas
Munguía-Vega A, Soria G, Pfister T, Cudney-Bueno R (2010) Isolation and
characterization of microsatellite loci in the rock scallop (Spondylus calcifer)
(Bivalvia: Spondylidae) from the Northern Gulf of California, Mexico.
Conservation Genet Resour:doi: 10.1007/s12686-12009-19141-12685
Neff BD, Fraser BA (2010) A program to compare genetic differentiation statistics across
loci using resampling of individuals and loci. Mol Ecol Resources 10:546-550
Orensanz JM, Parma AM, Turk T, Valero J (2006) Population, Dynamics and
Management of Natural Scallops. In: Shumway SE, Parsons GJ (eds) Scallops:
Biology, Ecology and Aquaculture, Vol 35. Elsevier, Amsterdam, p 765-868
Palumbi SR (2003) Population genetics, demographic connectivity, and the design of
marine reserves. Ecol Applications 13:146-158
123
Peakall R, Smouse PE (2006) Genalex 6: genetic analysis in Excel. Population genetic
software for teaching and research. Mol Ecol Notes 6:288-295
Pelc RA, Baskett ML, Tanci T, Gaines SD, Warner RR (2009) Quantifying larval export
from South African marine reserves. Mar Ecol Prog Ser 394:65-78
Pelc RA, Warner RR, Gaines SD, Paris CB (2010) Detecting larval export from marine
reserves. PNAS doi:10.1073/pnas.0907368107
Pollnac R, Christie P, Cinner JE, Dalton T and others (2010) Marine reserves as linked
social-ecological systems. PNAS doi:10.1073/pnas.0908266107
Poutiers JM (1995) Bivalvos. In: Fischer W, Krupp F, Schneider W, Sommer C,
Carpenter KE, Niem VH (eds) Guía FAO para la identificación de especies para
los fines de la pesca: Pacífico centro oriental, Vol 1: Plantas e Invertebrados.
Organizaciones de las Naciones Unidas para la Agricultura y la Alimentación,
Roma, p 646
Pritchard JK, Stephens M, Donnelly P (2000) Inference of Population Structure Using
Multilocus Genotype Data. Genetics 155:945-959
Rasmussen LL, Cornuelle BD, Levin LA, Largier JL, Di Lorenzo E (2008) Effects of
small-scale features and local wind forcing on tracer dispersion and estimates of
population connectivity in a regional scale circulation model. J Geophys Res
114:C01012
Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223–225
Rosenberg NA (2004) Distruct: a program for the graphical display of population
structure (p 137-138). Mol Ecol Notes 4:137–138
Rossi RE, Mulla DJ, Journel AG, Franz EH (1992) Geostatistical T-tools for modeling
and interpreting ecological spatial dependence. Ecol Monogr 62:277-314
124
Selkoe KA, Watson JR, White C, Horin TB and others (2010) Taking the chaos out of
genetic patchiness: seascape genetics reveals ecological and oceanographic
drivers of genetic patterns in three temperate reef species. Mol Ecol 19:3708-3726
SEMARNAT (2001) Norma Oficial Mexicana NOM-059-ECOL-2001. Protección
ambiental -Especies nativas de México de flora y fauna silvestres- Categorias de
riesgo y especificaciones para su inclusión, exclusión o cambio. Lista de especies
en Riesgo. Accessed 22 April 2010
http://www.semarnat.gob.mx/leyesynormas/normas/Pages/normasoficialesmexica
nasvigentes.aspx
Siegel DA, Kinlan BP, Gaylord B, Gaines SD (2003) Lagrangian descriptions of marine
larval dispersion. Mar Ecol Prog Ser 260:83-96
Siegel DA, Mitarai S, Costello CJ, Gaines SD, Kendall BE, Warner RR, Winters KB
(2008) The stochastic nature of larval connectivity among nearshore marine
populations. PNAS 105:8974-8979
Skoglund C, Mulliner DK (1996) The genus Spondylus (Spondylus calcifer) of the
Panamic Province. The Festivus 28:93-107
Smouse PE, Peakall R (1999) Spatial autocorrelation analysis of individual multiallele
and multilocus genetic structure. Heredity 82:561–573
Soria G, Tordecillas-Guillen J, Cudney-Bueno R, Shaw W (2010) Spawning induction,
fecundity estimation, and larval culture of Spondylus calcifer (Carpenter, 1857)
(Bivalvia: Spondylidae). J Shellfish Res 29:143-149
Sotka EE, Palumbi SR (2006) The use of genetic clines to estimate dispersal distances of
marine larvae. Ecology 87:1094–1103
Villalejo-Fuerte M, Arellano-Martínez M, Ceballos-Vázquez BP, García-Domínguez F
(2002) Reproductive cycle of Spondylus calcifer Carpenter, 1857 (Bivalvia:
Spondylidae) in the "Bahia de Loreto" National Park, Gulf of California, Mexico.
J Shellfish Res 21:103-108
125
Villalejo-Fuerte M, Muñetón-Gómez MS (2002) Tópicos sobre la biología de la almeja
burra Spondylus calcifer (Carpenter, 1857). Hidrobiológica 12:79-81
Watson JR, Mitarai S, Siegel DA, Caselle JE, Dong C, McWilliams JC (2010) Realized
and potential larval connectivity in the Southern California Bight. Mar Ecol Prog
Ser 401:31-48
Werner FE, Cowen RK, Pars CB (2007) Coupled biological and physical models: Present
capabilities and necessary developments for future studies of population
connectivity. Oceanography 20:54-69
Wright S (1978) Evolution and the genetics of populations. University of Chicago Press,
Chicago, USA.
126
Tables
Table B.1: Lineal distance (km) traveled by particles released at different sites. Higher
dispersion case (HD): position at high tide of passive particles released at spring tide, and
lower dispersion case (LD): position at low tide of active particles released at neap tide.
Mean (SD) values
Site
1 week
2 weeks
3 weeks
HD
56.6
(8.3)
LD
53.9
(10.7)
HD
67.8
(8.4)
LD
66.4
(7.4)
HD
78.9
(10.1)
LD
77.0
(8.9)
Sandy Beach
58.0
(9.5)
62.1
(4.9)
72.3
(4.9)
68.0
(8.6)
83.7
(6.6)
77.9
(9.5)
Las Conchas
60.9
(12.3)
58.2
(13.6)
80.2
(13.5)
77.3
12.1)
89.4
(13.8)
86.4
(11.6)
Los Tanques
70.2
(10.6)
67.9
(11.4)
93.4
(10.1)
90.0
(10.3)
101.6
(10.0)
97.9
(9.6)
San Jorge Island
44.7
(27.7)
30.3
(25.5)
47.8
(34.7)
56.0
(29.5)
54.1
(32.6)
60.1
(28.5)
San Francisquito
60.5
(19.8)
53.4
(15.4)
104.8
(22.8)
101.0
(26.3)
118.0
(18.7)
113.2
(23.2)
Puerto Lobos
74.3
(13.3)
77.0
(14.2)
135.2
(34.6)
117.5
(28.9)
165.6
(31.3)
157.2
(34.8)
Puerto Libertad
65.7
(21.5)
55.7
(20.7)
109.5
(25.8)
110.9
(22.7)
140.4
(37.5)
141.5
(38.4)
Las Cuevitas
21.3
(8.6)
30.6
(13.2)
71.6
(30.8)
56.0
(20.3)
116.5
(37.1)
84.1
(33.1)
Desemboque de los Seris
22.6
(9.6)
22.2
(9.5)
44.9
(20.0)
44.5
(23.2)
76.9
(35.7)
62.9
(35.9)
82.75
78.76
102.51
95.82
La Cholla
GRAND MEAN
53.5
51.1
127
Table B.2. Genetic variation among sampled localities. Sample size (N), mean (+SE)
numbers of alleles (NA), effective alleles (NE), and observed (HO) and expected (HE)
heterozygosities. Last two columns show the p values from BOTTLENECK according to
a Sign test and a Wilcoxon sign rank test for the stepwise mutation model (above) and the
two-phase mutation model (below).
Population
N
NA
(+SE)
NE
(+SE)
HO
(+SE)
HE
(+SE)
Sign
test
Wilcoxon
test
La Cholla
32
9.444
(1.788)
4.202
(1.145)
0.504
(0.095)
0.583
(0.100)
0.006
(0.029)
0.003
(0.004)
Los
Tanques
32
9.778
(1.839)
4.857
(1.263)
0.51
(0.096)1
0.630
(0.097)
0.004
(0.035)
0.003
(0.009)
San Jorge
Island
18
8.333
(1.555)
4.737
(1.386)
0.545
(0.093)
0.629
(0.096)
0.058
(0.406)
0.039
(0.156)
San
Francisquito
32
9.111
(1.874)
4.710
(1.199)
0.514
(0.101)
0.579
(0.118)
0.004
(0.024)
0.003
(0.009)
Puerto lobos
30
9.556
(1.804)
4.363
(1.089)
0.568
(0.100)
0.611
(0.098)
0.055
(0.038)
0.019
(0.027)
Desemboque
de los Seris
32
9.667
(1.650)
4.897
(1.043)
0.593
(0.092)
0.668
(0.093)
0.033
(0.311)
0.000
(0.179)
Table B.3. Genetic differentiation between pairs of localities: Mean (95% CI) Fst (above diagonal) and Gst' (below diagonal).
LCH: La Cholla, LTA: Los Tanques, SJO: San Jorge Island, PLO: Puerto Lobos, and DDS: Desemboque de los Seris.
LCH
LTA
SJO
SFR
PLO
DDS
0.027
(-0.006, 0.083)
0.024
(-0.012, 0.082)
0.033
(-0.001, 0.087)
0.032
(-0.002, 0.079)
0.037
(-0.003, 0.110)
0.02
(-0.017, 0.080)
0.038
(-0.002, 0.104)
0.03
(-0.003, 0.085)
0.044
(-0.003, 0.129)
0.026
(-0.003, 0.076)
0.031
(-0.002, 0.089)
0.04
(-0.006, 0.114)
0.031
(-0.004, 0.087)
Fst
0.031
(-0.006, 0.088)
LCH
0.06
(-0.014, 0.168)
SJO
0.06
(-0.013, 0.173)
0.072
(-0.007, 0.199)
SFR
0.048
(-0.026, 0.176)
0.039
(-0.025, 0.138)
0.086
(-0.004, 0.228)
PLO
0.074
(0.000, 0.175)
0.081
(0.000, 0.207)
0.061
(-0.008, 0.171)
0.082
(-0.005, 0.230)
DDS
0.07
(-0.004, 0.186)
0.063
(-0.006, 0.166)
0.069
(-0.009, 0.183)
0.064
(-0.006, 0.171)
Gst’
LTA
0.029
(-0.004, 0.084)
0.069
(-0.005, 0.202)
128
129
Figures
Fig. B.1. The Northern Gulf of California a), spatial units of analysis (gray solid lines),
release sites (red cruxes), genetic sample collection sites (arrows), and fishing beds
(green zones). SLI: San Lorenzo Island and SEI: San Esteban Island. Panel b) the main
study area of Puerto Peñasco (PP) corridor and former marine reserves (blue areas), and
selected spatial units of analysis (gray solid lines) EBO: EL Borrascoso, LCH: La Cholla,
SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques. SJO: San Jorge Island, and
SFR: San Francisquito. Larvae collection sites are depicted with black stars. Panel c)
southern fishing beds.
130
Fig. B.2. Final position of particles from the coupled biological-oceanographic model for
Spondylus calcifer a) higher dispersion case (position at high tide of passive particles
released at spring tide): outputs at 1, 2, and 3 weeks, b) output at 2 weeks for higher
dispersion case, and c) lower dispersion case (position at low tide of active particles
released at neap tide): output at 2 weeks. Color at each release site matches its particle’s
colors. EBO: El Borrascoso, LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas,
LTA: Los Tanques, SJO: San Jorge Island, SFR: San Francisquito, PLO: Puerto Lobos,
PLI: Puerto Libertad, LCU: Las Cuevitas, and DDS: Desemboque de los Seris.
131
DDS
SFR
PLO
DDS
n= 34
50
PLI
n= 52
LCU
n= 39
n= 101
PLI
La Cholla
n= 86
LCU
La Cholla
n= 119
SJO
LCN
La Cholla
75
LTA
LCH
SBE
DDS
PLI
c) Three weeks
LCU
SFR
PLO
SJO
LTA
SBE
LCN
LCH
DDS
PLI
b) Two weeks
LCU
PLO
SFR
SJO
LCN
LTA
SBE
100
LCH
a) One week
25
0
100
Sandy beach
Sandy beach
Sandy beach
n= 92
75
n= 94
n= 27
n= 15
n=26
n= 18
50
25
0
Relative abundance (%)
100
Las Conchas
Las Conchas
n= 89
75
n= 129
Las Conchas
n=68
n= 17
n=66
n= 33
50
25
0
100
Los Tanques
75
50
Los Tanques
Los Tanques
n=80
n= 94
n= 29
n= 128
n= 63
n= 54
25
0
100
San Jorge Island
75
50
San Jorge Island
San Jorge Island
n= 31
n= 92
n= 22
n= 51
n= 73
n= 106
25
0
100
San Francisquito
75
San Francisquito
San Francisquito
n= 267
n= 183
n= 77
n= 238
n= 104
n= 68
50
25
PLO
SFR
SJO
LTA
LCN
LCH
SBE
DDS
PLI
LCU
PLO
SFR
LTA
SJO
LCN
LCH
SBE
DDS
LCU
PLI
PLO
SFR
SJO
LTA
LCN
LCH
SBE
0
Releasing sites
Fig. B.3: Relative abundance of particles for each spatial unit of analysis at 1, 2, and 3
weeks. Gray bars: lower dispersion case (position at low tide of active particles released
at neap tide). Black bars: higher dispersion case (position at high tide of passive particles
released at spring tide). LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA:
Los Tanques, SJO: San Jorge Island, SFR: San Francisquito, PLO: Puerto Lobos, PLI:
Puerto Libertad, LCU: Las Cuevitas, and DDS: Desemboque de los Seris.
132
a) One week
c) Three weeks
b) Two weeks
100
Vaquita Reserve
Relative abundance (%)
75
Vaquita Reserve
Vaquita Reserve
n= 131
n= 299
n= 747
n= 56
n= 314
n= 958
50
25
0
100
El Borrascoso
El Borrascoso
n= 1199
n= 746
n= 549
n= 1057
n= 687
n= 403
75
El Borrascoso
50
PLI
LCU
SFR
PLO
LTA
SJO
LCN
SBE
LCH
VAQ
EBO
PLI
LCU
SFR
PLO
SJO
LTA
LCN
LCH
SBE
VAQ
PLI
LCU
SFR
PLO
LTA
SJO
LCN
SBE
LCH
EBO
VAQ
0
EBO
25
Releasing sites
Fig. B.4: Relative abundance of particles for each area at 1, 2, and 3 weeks for the
downstream areas of Vaquita Refuge and El Borrascoso. Gray bars: lower dispersion case
(position at low tide of active particles released at neap tide). Black bars: higher
dispersion case (position at high tide of passive particles released at spring tide). VAQ:
Vaquita Refuge, EBO: EL Borrascoso, LCH: La Cholla, SBE: Sandy Beach, LCN: Las
Conchas, LTA: Los Tanques, SJO: San Jorge Island, SFR: San Francisquito, PLO: Puerto
Lobos, PLI: Puerto Libertad, and LCU: Las Cuevitas.
133
b) Two weeks
a) One week
c) Three weeks
100
Relative abundance (%)
75
Puerto Lobos
n=132
Puerto Lobos
n= 226
n= 203
Puerto Lobos
n= 125
n= 136
n= 226
50
25
0
100
Puerto Libertad
Puerto Libertad
75
n=567
n=300
n= 489
n= 557
Puerto Libertad
n= 71
n= 82
50
ISL
Releasing sites
Fig. B.5: Relative abundance of particles for each area at 1, 2, and 3 weeks for the
upstream areas of Puerto Lobos and Puerto Libertad. Gray bars: lower dispersion case
(position at low tide of active particles released at neap tide). Black bars: higher
dispersion case (position at high tide of passive particles released at spring tide). PLO:
Puerto Lobos, PLI: Puerto Libertad, LCU: Las Cuevitas, ISE: San Esteban Island, DDE:
Desemboque de los Seris, IPA: Patos Island, ITIn: Tiburón Island (north), IDA: El Dátil
Island.
IDA
ITW
IPA
ITIn
ISE
DDS
PLI
LCU
IDA
ISL
ITIn
ITW
ISE
IPA
LCU
DDS
PLI
PLO
ISL
IDA
ITIn
ITW
ISE
IPA
DDS
PLI
LCU
PLO
0
PLO
25
134
Autocorrelation
coefficient (r)
0.015
0.010
0.005
0.000
-0.005
-0.010
-0.015
50
100
150
200
Distance class (km)
250
Fig. B.6. Spatial autocorrelation coefficient (r) among individuals of Spondylus calcifer.
The genetic similarity between pairs of individuals within each distance class is measured
by r. Positive values indicate individuals are genetically more similar than expected by
random. Bars represent 95% CI. Dashed lines represent upper (U) and lower (L)
confidence limits bound the 95% CI about the null hypothesis of no spatial structure for
the combined data set as determined by permutation. When r = 0, distance class length =
88.1 km.
135
b)
a)
0.6
LCH
LTA
0.5
31º N
0.3
Gulf of California
PLO
0.2
Density
0.4
SFR
SJO
30º N
0.0
0.1
DDS
1
2
3
4 5
6
Number of genetic
clusters
114º W
113º W
Fig. B.7. GENELAND’s clustering algorithm: a) posterior distribution of the number of
distinct genetic clusters, b) Sampling localities (red dots) and the assignments to the two
genetic clusters (green and gray, respectively). LCH: La Cholla, LTA: Los Tanques, SJO:
San Jorge Island, PLO: Puerto Lobos, and DDS: Desemboque de los Seris.
a)
Mean of Ln probability of data
136
-4800
-5000
-5200
-5400
-5600
-5800
-6000
-6200
-6400
0
1
2
4
3
5
6
K
b) 1
0.5
0
DDS
PLO
SJO
SFR
LTA
LCH
Fig. B.8. STRUCTURE’ clustering algorithm: a) Mean and standard deviation of ln
probability of data for K 1-6, b) Bar plot showing the mean individual assignment
probabilities among 10 independent replicates of K = 2 (red: southern cluster and green:
northern cluster). DDS: Desemboque de los Seris, PLO: Puerto Lobos, SJO: San Jorge
Island, SFR: San Francisquito, LTA: Los Tanques, and LCH: La Cholla.
Mean number of spats site
-1
137
2500
2000
Agust 2007 (N = 13182)
August 2008 (N = 15023)
1500
1000
500
0
LCH
SBE
LCN
LTA
SJO
SFR
Sites
Fig. B.9. Mean number (SD) of Spondylus calcifer spat recruited on artificial collectors
pooled per site and year. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA:
Los Tanques, SJO: San Jorge Island and SFR: San Francisquito.
Mean number of spats collector
-1
138
2007
1200
1-m
3-m
2008
5-m
a
7-m
a
1-m
3-m
5-m
7-m
1000
a
800
aa
a,b
a
600
aa
b
a
a
SBE
LCN
LTA
SJO
a
SFR
LCH
SBE
a
b
a,b
b
b
b
LCH
a
a
b
a
200
0
a,b
a
400
b
LCN
LTA
SJO
SFR
Sites
Fig. B.10. Mean number (SD) of Spondylus calcifer spat recruited per collector at
different depth and year. Different letters indicate significant different values (Tukey’s p
< 0.05) between depths after one-way ANOVAs for each site and year. LCH: La Cholla,
SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and
SFR: San Francisquito.
139
0.30
Wi lower dispersion
Wi higher dispersion
Oi 2007
a) one week
40
30
0.20
20
10
0.00
0
0.30
40
b) two weeks
30
0.20
20
0.10
0.00
0.30
10
c) three weeks
0
40
Observed values (Oi )
Predicted values (Wi )
0.10
30
0.20
20
0.10
10
0
0.00
LCH SBE LCN LTA SJO SFR
Sites
Fig. B.11. Correlation between relative abundances of predicted values (Wi) and observed
relative values of Spondylus calcifer spat recruited on artificial collectors (Oi) at a) 1
week, b) 2 weeks, and c) 3 weeks for lower dispersion case (position at low tide of active
particles released at neap tide) and higher dispersion case (position at high tide of passive
particles released at spring tide). LCH: La Cholla, SBE: Sandy Beach, LCN: Las
Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR: San Francisquito.
140
APPENDIX C: RECRUITMENT OF CATARINA SCALLOP, ARGOPECTEN
VENTRICOSUS, LARVAE ON ARTIFICIAL COLLECTORS ALONG A
GEOGRAPHICAL AREA IN THE NORTHERN GULF OF CALIFORNIA
TO BE SUBMMITED TO THE JOURNAL OF EXPERIMETNAL MARINE
BIOLOGY AND ECOLOGY
Gaspar Soriaa∗, Iván Martínez-Tovarb & Alberto Macias-Duartea
a
School of Natural Resources and the Environment, University of Arizona. Biosciences
East 325D. Tucson, Arizona 85721, USA
b
Centro Intercultural de Estudios de Desiertos y Océanos. Apar. Postal 53. Puerto
Peñasco, Sonora 83550, México
Keywords Scallop, Argopecten ventricosus, Larvae, Collector, Spat recruitment, Gulf of
California
∗
corresponding author: [email protected]
141
Recruitment of catarina scallop, Argopecten ventricosus, larvae on artificial
collectors along a geographical area in the Northern Gulf of California
Gaspar Soria, Iván Martínez-Tovar & Alberto Macias-Duarte
Abstract
We evaluated the spatial and temporal availability of catarina scallops,
Argopecten ventricosus, larvae along an environmental gradient in Puerto Peñasco,
Sonora, Mexico. We deployed artificial collectors (Netlon nets: 200 x 40 cm; mesh
opening = 7x12 cm) in 6 sites from June 2007 to August 2008. We placed temperature
loggers to record bottom temperature, and measured sea surface water salinity,
temperature, and dissolved oxygen. We used a repeated measures analysis of variance to
evaluate differences in the number of recruited spat between months, sites, and depths,
and analyzed the presence of multimodal spat shell size frequency distributions. Bottom
temperature ranged between 11.32-32.65 oC and followed similar patterns among all
sites. Salinity values ranged between 34.3-36.0 g l-1. Dissolved oxygen levels remained
>80% throughout the year. When all sites are pooled, spat recruitment significantly
differed between months (p < 0.001). At each site, spat recruitment was significantly
different each month (p < 0.001). We did not find significant differences between months
(p = 0.143) at each depth. Overall, spat recruitment in Puerto Peñasco was negatively
correlated with seawater temperature. Higher spat recruitment abundances ( x = 1631.6
spat collector-1, SD = 820.1, n = 112,580) were observed in February when sea water
temperature was ∼15 ºC. Spat recruitment extended throughout spring until the beginning
142
of summer .We estimated multimodal shell size frequency distributions characterized by
small (shell height range = 1.39-7.19 mm), intermediate (range = 4.07-18.94 mm) and
large scallops (range = 7.21-19.08 mm). The proportion of variability explained by these
distributions varied markedly. The presence of multicohort distributions suggests that
spat collectors could be retrieved and replaced earlier than 2 months, particularly during
warmer months (April-June) when abundance of larger size scallops (>10 mm) is higher.
Observed patterns of spat recruitment suggest that the species has a protracted spawning
activity, with an intense activity throughout cold and temperate months. The lack of
recruitment in October suggests a brief resting period at the end of the summer. The
natural collection of A. ventricosus spat on artificial collectors in the area of Puerto
Peñasco can be successfully performed over a protracted period (December-June), with
an intense activity throughout cold and temperate months. The presence of diverse
recruitment patterns suggest that large scale processes and local site conditions may be
playing significant roles as drivers of the observed patterns. Our results complement the
information available concerning the recruitment of A. ventricosus spat on commercial
collectors, thus extending the geographical region where natural collection can be
successful. This information can be used to promote further activities such as
aquaculture, stock enhancement and repopulation programs.
143
Introduction
The Gulf of California and the Pacific side of the Baja California Peninsula,
Northwest Mexico (Fig. 1), are highly productive marine environments where industrial,
small-scale fisheries and aquaculture activities take place (Cisneros-Mata, 2010). In this
region, the catarina scallop Argopecten ventricosus (Sowerby, 1842) is an economically
important resource (Félix-Pico, 2006; González-Anativia, 2001). Of approximately 20
bivalve species harvested in the region, A. ventricosus landings accounted for 50% of the
total captures from 1986-2001 (Carta Nacional Pesquera, 2004). The species is harvested
by small-scale hookah-diving fishers primarily for its adductor muscle (locally know as
“callo”), which is mainly exported to the United States of America (Félix-Pico, 2006).
The species is also cultured but in much less quantity than harvested naturally. For
instance, the largest record of landings of A. ventricosus totaled 15,800 Mt in 2008 (Fig.
2a) (FAO, 2010), while aquaculture production peaked in 2001 with 127 Mt (Fig 2b)
(Carta Nacional Pesquera, 2006; FAO, 2010).
Despite its relevance, the production of both fisheries and aquaculture activities
has been characterized by marked fluctuations over the last decades (Fig. 2), and many
scallop beds have been overexploited (Félix-Pico, 2006). Because of the stochastic nature
on the fishery’s captures of A. ventricosus (a common phenomenon also seen in other
scallops fisheries elsewhere) (Félix-Pico, 2006; Orensanz et al., 2006), the aquaculture
production of this species could be an appropriate strategy to complement fisheries’
landings or even increase the natural production of scallops (Félix-Pico, 2006; Rangel-
144
Dávalos, 1990; Uriarte et al., 2001). Furthermore, the development of aquaculture and
conservation initiatives (e.g. stock enhancement and repopulation programs) relies
greatly on the availability of larvae being produced either under hatchery conditions or
collected from natural environments (Arnold et al., 1998; Félix-Pico, 2006; Narvarte et
al., 2001; Pouliot et al., 1995; Thouzeau, 1991; Uriarte et al., 2001).
Argopecten ventricosus inhabits fine and coarse sandy bottoms along both side of
the Baja California Peninsula and the east coast of the Gulf of California to the Peruvian
coast (Peña, 2001). A. ventricosus is a functional hermaphrodite species and broadcast
spawner and can reach 90 mm of maximal shell height (Peña, 2001). The sexual maturity,
when 50% of the scallops exhibit mature gonads, occurs at an age of <4 months (mean
shell height = 20 mm) (Cruz et al., 2000). Scallops at maturing and spawning stages can
be present all year-round, however the reproductive patterns vary according to local
environmental conditions (Baqueiro-Cárdenas and Aranda, 2000; Luna-González et al.,
2000). Depending on rearing temperatures and food availability, spawned scallops can
mature and spawn again within <4 weeks (Félix-Pico, 2006; Monsalvo-Spencer et al.,
1997). Under hatchery conditions, scallop larvae can reach the pediveliger stage within
14-18 days depending on water quality parameters (Félix-Pico, 2006). At sea, artificial
substrates (like plastic meshes, plastic plates, and ropes), were used successfully to
collect spat (post-larvae) with marked variations between locations and seasons (FélixPico, 2006; Félix-Pico et al., 1997; Maeda-Martínez et al., 1993). Several techniques
have been applied to culture A. ventricosus for commercial harvest and these techniques
vary according to local conditions. These methods include both bottom or suspension
145
culture techniques (Félix-Pico, 2006; Maeda-Martínez et al., 2001). Under various culture
techniques, catarina scallops can reach commercial size (shell height 60 mm) in <1 year
(Avendaño et al., 2001; Félix-Pico, 2006; Maeda-Martínez et al., 2000).
Whereas commercial fisheries and aquaculture activities along both margins of
the Baja California Peninsula are well documented (Félix-Pico, 2006; Narvarte et al.,
2001), the status of the fishery and aquaculture initiatives along the east coast of the
Northern Gulf of California is less documented or constrained to fewer sites (Félix-Pico,
2006). Puerto Peñasco is an important fishing town on the northeastern side of the Gulf of
California, where the species represents an important economic influx for local fishers
when it is available in the natural environment (Martinez-Tovar, pers. obs). Nonetheless,
landings of this scallop species have also fluctuated markedly over recent years in this
region (Fig. 2a), with the most recent and intense fishing pulses observed in 2002 and
2009 (Martinez-Tovar, pers. obs). The 2009 fishing pulse lasted about 7 months, involved
approximately 100 fishers, and was localized over a single bed, accounting for ∼7000 Mt
(Martinez-Tovar, pers. obs.). In spite of the species’ importance, there are no records of
to aquaculture initiatives fro A. ventricosus in the Puerto Peñasco area. In addition,
information on the natural availability of scallop spat (and the factors affecting it) is
unavailable.
In order to understand the spatial and temporal availability of larvae along a
geographical area in Puerto Peñasco, we deployed artificial collectors (Netlon nets) from
June 2007 to August 2008. The aim of this study is to provide information about the
natural availability of A. ventricosus spat (optimum site locations, timing, depth
146
variances, and intensity of settlement), needed for the development of aquaculture and
conservation initiatives this important fishing community in the Northern Gulf of
California.
Material and Methods
Temporal and spatial variation of larvae abundance
We estimated recruitment of A. ventricosus spat on artificial collectors deployed
in 6 sites along a geographical area in Puerto Peñasco, covering ~75 km of coast. These
sites included La Cholla (31°20'N-113°38'W), Sandy Beach (31°19'N-113°36'W) in the
NW sector, Las Conchas (31°16'N-113°26'W) and Los Tanques (31°14'N-113°19W) in
the center, San Jorge Island (31°0'N-113°14'W), and the southern site of San Francisquito
(30°55'N-113°7'W) (Fig. 1b). At each site, we deployed three vertical lines, each one
consisting of a polyethylene-rope (Ø= 0.8 cm) tied at one end to a screw-type anchor
(length = 75 cm), and the other end to a polystyrene-buoy (buoyancy = 5 kg) (Fig. 3). We
tied collectors to the rope at 1, 3, 5, and 7 m starting from the anchor. Each collector was
constructed with polyethylene Netlon® (200 x 40 cm; mesh opening = 7x12 cm) which
in turn is enclosed inside a plastic-bag (60 x 40 cm; mesh opening = 0.8x1 cm) (Fig. 3).
This outer retains any spat detached by strong currents (Shumway and Parsons, 2006).
We deployed the first batch of collectors on June 22nd, 2007, and replaced them with new
ones every 2 months to prevent the built-up of fouling on the collectors. We ended the
field collection of spats on August 24th, 2008.
147
We programmed and placed temperature loggers (Maxim-Dallas; ibutton 1-Wire
model DS1922L) in each site to record bottom temperature (oC) every 4 h. On each field
trip, we used a hand-sensor (Yellow Spring Incorporated, YSI 85) to measure surface sea
water salinity (g l-1), dissolved oxygen (%), and temperature.
Laboratory procedure and statistical analyses
We detached spat by washing and sieving each collector through 250-µm nylon
mesh. We transferred spat to a plastic pan to be counted and measured. We used the
program Image ProPlus 4.0 (Media Cybernetics) to measure shell height (distance from
the umbo the opposite shell margin) based on digital images. We used a repeated
measures analysis of variance (longitudinal study) to test significant differences in the
number of recruited spat between months, sites and depths. We applied a natural
logarithmic transformation [i.e. ln (x+1)] to the numbers of spat per collector to meet
normality and homogeneity of variances. Because main factors as well as their
interactions were significant, we performed individual one-way ANOVAs to compare
spat recruitment at different depths for each site and month separately. When significant
differences in depth were observed, we performed a post hoc Tukey’s test. We visually
examined spat shell size frequency distributions at each site and month and observed
multimodality in most cases. To simplify the representation and interpretation of the data
we performed the analysis pooling all depths together. We assumed that these
distributions were actually the mixture of >1 normal distributions (i.e. presence of
multiple spat cohorts). We used the package “mixtools” (Benaglia et al., 2009) in “R”
148
software v2.10.1 for Mac® (R Development Core Team, 2009) to fit mixture
distributions to the size dataset at each site and month. We fitted mixture distributions
with 2-4 components, and chose the mixture distribution with the lowest negative loglikelihood. We estimated mean, standard deviation and proportion for each component of
the modeled mixture distributions. In all statistical tests we considered differences were
significant when α <0.05.
Results
Sea water parameters
Sea bottom temperature fluctuated markedly between seasons. Across all sites,
sea temperature reached the lowest mean values in January (temperature range = 14.3815.29 oC) and the highest mean values in August (range = 30.34-31.33 oC) (Fig 4).
Furthermore, in August San Jorge Island and San Francisquito showed lower mean
temperature values (30.34 and 30.92 oC, respectively) than the other sites (range = 31.2031.33 oC). Similarly, in January, San Jorge Island showed the higher mean temperature
value (15.29 oC) in comparison to the other sites (range 14.38-14.91 oC). The lowest
temperature (11.32 oC) was recorded for La Cholla in January 2008 and the highest
(32.65 oC) for Los Tanques in September 2007 (Fig 4). We observed maximum salinity
values throughout summer months (> 35.5 g l-1) and lower values in winter (<35 g l-1).
Dissolved oxygen levels, expressed as percentage of saturation, remained >80 %
throughout the year with higher concentration values in summer and winter months
(Table 1).
149
Spat recruitment
A total of 308,590 spat recruited throughout the study period. When all sites are
pooled together, spat recruitment varied markedly between months. For instance, in
August (2007 and 2008) and October recruitment was either low or null while during
winter and spring seasons recruitment was considerable higher (Fig. 5). Therefore, we did
not take into account these months (August and October). After excluding these months,
spat recruitment significantly differed between months (repeated measures 3-way
ANOVA; F 1, 62 = 14.65, p < 0.001). In addition, spat recruitment at each site was
significantly different each month (repeated measures 3-way ANOVA; F 5, 62 = 20.39, p <
0.001). At each depth we did not find significant differences between months (repeated
measures 3-way ANOVA; F1, 65 = 2.19, p = 0.143).
Spat recruitment along the study area showed highly variable patterns at each site.
In most sites, higher spat recruitments were observed throughout winter (JanuaryFebruary) when sea bottom temperatures were at their lowest values. In addition, spat
recruitment extended throughout spring until the beginning of summer when sea water
temperature was increasing. For instance, at the northernmost site of La Cholla, we
observed a very similar mean spat recruitment from February ( x = 1410 spat collector-1,
SD = 431.1) to June ( x = 1601.7 spat collector-1, SD = 705.8), while Sandy Beach
showed a single peak in April ( x = 1900.4 spat collector-1, SD = 485.1) followed by
minor values in June ( x = 1173.7 spat collector-1, SD = 391.4) (Fig. 5). For the central
site of Las Conchas, we estimated spat recruitment peaks in February ( x = 1848.3 spat
collector-1, SD = 681.2) and April ( x = 1603.8 spat collector-1, SD = 470.8). For Los
150
Tanques we observed a single peak in February ( x = 1404.3 spat collector-1; SD = 455.6)
(Fig. 5). San Jorge Island showed the highest spat recruitment in February ( x = 2233.5
spat collector-1, SD = 657.8) and lower values in June ( x = 1269; SD = 593.2 spat
collector-1) and December ( x = 1321.3 spat collector-1, SD = 738.7). The southernmost
site of San Francisquito showed a single and vigorous recruitment in February ( x =
2555.4 spat collector-1; SD = 842.9) (Fig. 5).
Spat recruitment in the study area also showed different patterns according to the
location site and deployment depth. For example in December, at San Jorge Island and
the southernmost site of San Francisquito, we found lower spat recruitment abundances
near the bottom (Fig 6a). Contrary to this pattern, in February spat recruitment tended to
be higher near the bottom in La Cholla, Las Conchas, Los Tanques, and San Jorge Island
(Fig. 6b). In April, an inverse relationship between depth and spat recruitment was
observed between Las Conchas, with higher spat recruitment near the bottom, and San
Jorge Island, with lower spat recruitment near the bottom (Fig. 6c). In June, spat
recruitment was lower near the bottom at La Cholla, San Jorge and San Francisquito,
while for Las Conchas and Los Tanques we observed the inverse pattern (Fig. 6d).
Spat shell size frequency distributions
We observed multimodal shell size frequencies distributions at each site and
month (Fig. 7-10; Table 2). Generally, the size distributions exhibited a multimodal
pattern with one component ( µ 1) characterized by small scallops (mean shell height
range = 1.39-7.19 mm), followed by a second component ( µ 2) of higher sized scallops
151
(range = 4.07-18.94 mm). In most cases a third component ( µ 3) was also present (range
= 7.21-19.08 mm) (Table 2). The mixing proportion of each of these components varied
markedly between and within sites throughout seasons (Fig. 7-10; Table 2).
At the northernmost site of La Cholla, the three components estimated for
December and February (Fig. 7 & 8; Table 2) explained similar proportions of spat
recruitment. For example, in December the smaller shell height distribution component
showed a mean value of 1.58 mm (SD = 0.58), which represented 37% of the data, while
the second component ( x = 5.96 mm, SD = 2.55) accounted for 40% of the data, and the
third distribution ( x = 11.9 mm, SD = 4.46) explained 23 % of the data (Fig. 7; Table 2).
On the other hand, in April and June, the majority of the data (>85%) was explained by
the two larger size component distributions combined (Fig. 9 & 10; Table 2). Similarly,
Sandy Beach showed a distribution with three modes for December through June. For the
spawning peak identified in April (Fig. 5), the smaller component distribution showed a
mean shell height of 1.8 mm (SD = 0.7), and accounted for 14% of the data, while the
second ( x = 7.95 mm, SD = 2.79) and the third ( x = 18.90 mm, SD = 3.36) distributions
accounted for 75% and 11% of the data, respectively (Fig. 9; Table 2). For the central site
of Las Conchas, we estimated a bimodal distribution in February. Fourteen percent of the
data was explained by the one component ( x = 2.55 mm, SD = 1.02) and 86% accounted
for the second component distribution ( x = 7.91 mm, SD = 2.56). In a different pattern, in
April we estimated three modal distributions, however we were unable to fit appropriate
mixed distributions (Fig. 8 & 9; Table 2) (see Discussion section). For the spawning peak
identified in Los Tanques in February (Fig. 5), shell size frequency distributions suggest
152
a bimodal distribution with more than 97% of the data explained by the first modal
distribution (Fig. 8). Likewise La Cholla and Sandy Beach, we estimated three modal
distributions for the spat recruited at San Jorge Island for December through June. The
proportion of the third component in San Jorge Island showed a tendency to increase
from December (0.03%) to June when 65% of the data was explained by this distribution
(Fig. 7-10; Table 2). Finally, the vigorous recruitment observed in February at San
Francisquito (Fig. 5) was mainly represented by smaller scallops ( x = 6.0 mm, SD =
2.94) accounting for 76% of the data, though we argue this distribution might be
overestimated (see Discussion section) (Fig. 8; Table 2).
Discussion
Natural collection of A. ventricosus spat through artificial collectors has been well
documented in Northwest Mexico, especially in Baja California Sur (Félix-Pico, 2006;
Narvarte et al., 2001). In this study, our results complement the information available
concerning the recruitment of A. ventricosus spat on commercial collectors, and thus
extend the geographical region where natural collection can be successful.
On average, spat recruitment showed strong variations between seasons, and
between and within sites (Fig. 5 & 6). These highly fluctuating patterns in spat
recruitment are similar to the patterns described for the species (Félix-Pico, 2006;
Narvarte et al., 2001) and other scallops species elsewhere (Narvarte et al., 2001;
Orensanz et al., 2006). For A. ventricosus, Bahía Magdalena in the Pacific side of the
Baja California Peninsula (Fig. 1) has been suggested as one of the best sites to collect
153
spat naturally because of its low temperatures (range = 20-26.6 ºC) and high primary
productivity (range 1.5-5.1 mg chlorophyll-a m-3) throughout the year (Cruz et al., 2000;
Félix-Pico, 2006). Reports of spat recruitment for this bay ranged between 50-10000 spat
per collector depending on the season and to the site of collection (Félix-Pico, 2006).
Similarly, in Bahía Concepción in the west coast of the Gulf of California (Fig. 1), spat
recruitment varied between 150-25000 spat per collector, but was characterized by high
fluctuations between years (Félix-Pico, 2006). Although estimated spat abundances per
collector for the Puerto Peñasco area are generally lower than the maximum values
observed in both Bahía Magdalena and Bahía Concepción (e.g. highest value x = 3152
spat collector-1, SD = 640.08 at San Francisquito), our estimations are higher than the
ones from other regions within the Gulf of California including Bahía de Bacochibampo
(1700 spat collector-1) and Bahía de la Paz (10-650 spat collector-1) (Félix-Pico, 2006).
Spat recruitment along the Puerto Peñasco area also showed different patterns
according to the site location and deployment depth (Fig. 5 & 6). The presence of diverse
recruitment patterns suggest that large scale processes (seasonal currents, proximity to
parental stock, etc) and local site conditions (e.g. sea water parameters) may be playing a
significant role as drivers of the observed patterns, as has been observed in most
commercial scallops worldwide (Cyr et al., 2007; Narvarte, 2001; Orensanz et al., 2006;
Ruzzante and Zaixso, 1985; Thouzeau, 1991). A large-scale process likely affecting spat
recruitment in the region may be the seasonal-reversing currents that dominates the
circulation in the Northern Gulf of California. Throughout summer (from June to
September), marine currents flow from Southeast to Northwest areas, while in winter,
154
marine currents move in the opposite direction (Carrillo and Palacios-Hernández, 2002).
Even though the location of A. ventricosus beds outside the study area are unknown, the
flow of marine currents suggest that spat recruiting from December to April might be
originated from scallop beds located north of the study area. On the other hand, during
summer it is likely that the spat might be originated from southern sources. Furthermore,
during the last fishing pulse in 2009, fishers targeted a bed located between Los Tanques
and San Jorge Island within the study area (Martinez-Tovar, pers. obs.). Given the
seasonal-reversing currents, this scallop bed might contribute with larvae either to
northern and central sites or to San Jorge Island San Francisquito according to season.
Nonetheless, this requires further investigation to confirm or reject these hypotheses.
Overall, the spat recruitment along the Puerto Peñasco area was negatively
correlated with seawater temperature values. For most sites, higher spat recruitment
abundances were observed in February (Fig. 5 & 6) when sea water temperature was ∼15
ºC (Fig. 4 & 6). In addition, these spat recruitment peaks may take place when primary
productivity is relatively higher in comparison to summer months. Based on satellite
imagery, Sanchez-Velasco et al. (2009) estimated that primary productivity ranged
between 2-4 mg chlorophyll-a m-3 from December to June, with maximum values
throughout February (3.5-4 mg chlorophyll-a m-3). On the other hand, reduced levels of
primary productivity throughout July-August (<1.2 mg chlorophyll-a m-3) (SánchezVelasco et al., 2009) are coupled with higher sea water temperature values (>30 ºC; Fig.
4 & 6), which might explain the low recruitment levels observed in August and October.
155
Site selection for the deployment of collectors should vary by season because of
the strong seasonality in spat recruitment of A. ventricosus observed in the study area
(Fig. 6). For instance, in December, spat recruitment at San Jorge Island ( x = 1321 spat
collector-1 SD = 739) could be 3-5 times higher than in other sites. In February, all sites
but Sandy Beach showed significantly higher spat recruitment compared to other months.
In April, the northern sites and Sandy Beach showed spat abundances 3-15 times higher
than in other sites. In June, with the exception of San Francisquito, spat recruitment was
more similar among all collecting sites (x = 690-1600 spat collector-1). Despite the low
recruitment observed in August of both years, it is noteworthy that spat recruited at San
Jorge Island accounted for >85% of the total spat recruiting in the study area. The upper
thermal tolerance for juveniles scallops (shell height = 10 mm) was suggested to be
around 29 ºC (Maeda-Martínez et al., 2000; Sicard et al., 1999; Sicard et al., 2006).
Therefore, higher survival rates might be related to slightly lower temperatures recorded
around the island (mean temperature 30.34 – 30.42 ºC) in comparison with the other sites
(31.20-31.33 ºC).
Multimodal shell size frequencies and strong peaks of small sized scallops (mean
shell height range = 1.43-2.92 mm) characterized spat recruitment patterns in most cases
(Fig. 8-11). However, it is important to mention that in some cases the software provided
inappropriate mixture distributions. These cases were: San Francisquito in December
(Fig. 8), Sandy Beach, Los Tanques and San Francisquito in February (Fig. 9), Las
Conchas and San Jorge Island in April (Fig. 10), and Sandy Beach in June (Fig. 11). In
these cases the analysis did not differentiate the smaller spat (<3 mm in shell height size)
156
as an independent cohort, which may represent a recently recruited cohort. In this regard,
A. ventricosus spat can attain a shell height range of 5-8 mm in <1 month under hatchery
conditions (Maeda-Martínez et al., 1997), which suggest that the smaller component may
correspond to a recently recruited cohort <2 weeks. Nevertheless, we successfully fitted a
distribution to each peak in the size data in most cases. This presence of multicohort
distributions implies that spat collectors could be retrieved and replaced in <2 months,
particularly during warmer months (April-June) when larger size scallops (>10 mm in
shell height size) are present. Usually, spats are transferred to juvenile intermediate
culture devices (e.g. pearl-nets, nestier trays) when scallops’ shell sizes are ~10 mm
(Secretaria de Pesca 1994, Avendaño et al. 2001). Fewer cases in April and most cases in
June showed a third component distribution that corresponded with higher values (range
x= 14.84-19.08 mm) than the reference value used as a proxy to transfer spat to the next
rearing condition (Fig. 10-11). Also, the third modal distributions (range = 12.75-19.08)
observed in June explained between 42 and 70% of the data, which is an indicator of the
relative abundance of spat that could be ready to be transferred to further grow-out
devices.
The lack of a clear correlation between spat recruitment and depth, suggest that
local site conditions (e.g. primary productivity, DO, temperature, competitors, predators
etc.) might be responsible for such trends. Further research on this topic is needed to
address the role of local effects more conclusively.
Histological analyses of the gonads have shown that mature and spawning A.
ventricosus scallops can be found year-round in most regions in Northwest Mexico. In
157
spite of this, reproductive patterns of A. ventricosus in Northwest Mexico vary markedly
according to local conditions (Villalejo-Fuerte & Ochoa-Baez 1993, Baqueiro-Cárdenas
& Aranda 2000, Cruz et al. 2000, Luna-González et al. 2000).According to Barber and
Blake (2006), the reproductive cycle of scallops is a genetically controlled response to
environmental factors, with temperature and food availability as the most important
drivers. Thus, the reproductive pattern in a particular location will depend on the
interactions between endogenous and exogenous factors (Cruz et al. 2000, Barber &
Blake 2006). Like many other scallops species A. ventricosus has the capacity to mature
and spawn through the use of available food from the environment or, if primary
productivity is scarce, through the use of energy stored in the adductor muscle (VillalejoFuerte & Ceballos-Vásquez 1996, Luna-González et al. 2000, Barber & Blake 2006). The
interplay of such characteristics might be responsible for the observed differences on the
reproductive patterns of A. ventricosus in Northwest Mexico. For example, scallops from
La Paz, Baja California Sur (west coast of the Gulf of California) shows a protracted
spawning period from July to December (summer and fall months) with minor peaks at
the end of winter and early spring, and a clear resting period in June (end of spring)
(Luna-González et al. 2000). On the other hand, in Bahía Concepción the main spawning
season is from September to April with a peak of activity from January and February,
which coincided with the lowest temperature (16oC) recorded in the area, and another
minor peak in June (Villalejo-Fuerte & Ochoa-Baez 1993). In a different pattern, scallops
from Bahía Magdalena in the Pacific coast, spawn from March to May when water
temperatures are close to 20oC (Cruz & Ibarra 1997), whereas northern populations
158
located at Ojo de Liebre Lagoon spawn all year round with three moderate peaks (March,
August and December), without resting period (Baqueiro-Cárdenas & Aranda 2000).
Histological studies addressing the reproductive cycle of A. ventricosus from the
Northern Gulf of California have not been conducted yet. However, the observed patterns
of spat recruitment on artificial collectors in the Puerto Peñasco area suggest that the
species has a protracted spawning activity (December-August), with an intense activity
throughout cold and temperate months (Fig. 5). A reproductive pattern that might
resemble the reproductive strategy of scallops from Bahía Concepción (Villalejo-Fuerte
& Ochoa-Baez 1993). The existence of multicohort distributions (Fig. 8-11) in the
majority of the sites suggests an intense spawning activity, which might be related with
the capacity of spawned scallops to mature and spawn again within <4 weeks when
environmental conditions are appropriate (Monsalvo-Spencer et al. 1997). Lack of
recruitment in October (collectors were deployed in August) suggests a brief resting
period from the end of the summer to the beginning of the spring. High temperature (>30
ºC) and low primary productivity (<1.5 mg chlorophyll-a m-3), similar to those observed
for the study area in summer, have been suggested as plausible factors triggering oocyte
reabsorption and gonads atresia followed by a resting stage in A. ventricosus scallops
from Bahía Concepción (Cruz et al. 2000). This has been found in other scallop species
as well (Barber & Blake 2006). Therefore, scallops may experience unfavorable
conditions throughout the summer, which may trigger a period of low gametogenic
activity and lead to low spat recruitment levels. Further studies are needed to address the
159
likely role of environmental parameters as drivers of the reproductive cycle of the species
in the study area.
In this study we demonstrated that the natural collection of A. ventricosus spat in
the study area can be successfully performed over a protracted period. Also, spat
recruitment varied markedly between seasons and highly significant differences per site
were observed. This information can be used to promote further activities such as
aquaculture initiatives, stock enhancement and repopulation programs. Site selection has
been shown to be a key factor in any successful bivalve aquaculture initiative (Uribe &
Blanco 2001, Sicard et al. 2006). In this regards, there are several studies that can be used
to help guide the selection of appropriate locations for further grow-out. For instance,
under laboratory conditions, oxygen consumption and algae ingestion rates measured on
juveniles scallops (shell height size = 10 mm) reared at different temperatures (range =
16-28 ºC), showed optimum values between 19-22 ºC. These values were also correlated
with higher growth and survival rates. On the other hand, detrimental effects on these
physiological rates were observed at temperatures higher than 26 ºC with an upper lethal
thermal tolerance of 29 ºC (Sicard et al. 1999, Maeda-Martínez et al. 2000, Sicard et al.
2006). Furthermore, according to Signoret-Brailovsky et al., (1996), the species is
osmoconformist with an ample salinity tolerance (27-47 g/l), and thus salinity
concentration observed in the region might have no detrimental effects on scallops held
on grow-out setting. As for adult scallops, massive mortalities (>95%) of cultured
scallops have been associated with intense spawning events coupled with higher
temperatures (>30 ºC). These massive mortalities were likely attributable to the thermal
160
tolerance observed for the species (reviewed in Maeda-Martínez et al. (2000), and
Maeda-Martínez et al. (1997)). Therefore, given the strong seasonality of environmental
parameters that characterize the study region, these physiological references could be
used as a proxy to select the most suitable sites for further grow-out or conservation
programs in the area of Puerto Peñasco.
Finally, grow-out methods include bottom cultures, such as direct sowing
(Maeda-Martínez et al. 2001), and the culture of scallops in plastic sleeves (MaedaMartínez et al. 2000), suspended cultures in plastic cages (Nestier trays) (MaedaMartínez et al. 1997) and lanterns units (Félix-Pico 2006). Each method has been shown
to have specific advantages over the others depending on local environmental and
ecological conditions. Determining whether the A. ventricosus in the region of Puerto
Peñasco should be reared following a particular methodology requires further study.
Acknowledgements
We wish to thank A. Cinti for reviewing and providing insightful comments. We
acknowledge the financial support provided from The Nature Conservancy and the
Conservancy’s RJ KOSE Grant Program, and the Wallace Research Foundation. We
wish to thank P. Turk-Boyer, and R. Loaiza-Villanueva from Centro Intercultural de
Estudios de Desiertos y Océanos (CEDO), and J. Torre-Cosio, and M. Rojo-Amaya from
Comunidad y Biodiversidad, A. C. (COBI). We appreciate the logistic support provided
by the fishing cooperative Buzos de Puerto Punta Peñasco, especially to J. Salazar, V.
León, and A. Ramos who greatly contributed with the deployment of collectors. We
161
appreciate the logistic support provided by the D. Manjon, W. Ludt, J. Hall, M.
Ogonowski, E. Koltenuk, and A. Rife. We collected catarina scallops under the fishing
permit #DGOPA 11216.0611109. 3883 (Permiso Pesca Fomento), issued to COBI A.C.
by the Secretaria de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación
(SAGARPA), Mexico. This is a scientific contribution of the PANGAS Project
(www.pangas.arizona.edu).
162
References
Arnold, W.S., Marelli, D., Bray, C.P., Harrison, M.M., 1998. Recruitment if bay scallops
Argopecten irradians in Floridan Gulf of Mexico waters: scales of coherence.
Mar. Ecol. Prog. Ser. 170, 143-157.
Avendaño, M., Cantillanez, M., Le Pennec, M., Lodeiros, C., Freites, L., 2001. Cultivo
de pectinidos Iberoamericanos en suspensión. In: Maeda-Martínez, A.N. (Ed.),
Los Moluscos Pectínidos de Iberoamérica: Ciencia y Acuicultura. Editorial
Limunsa, México, pp. 193-211.
Baqueiro-Cárdenas, E., Aranda, D.A., 2000. A review of reproductive patterns of bivalve
mollusks from Mexico. Bull. Mar. Sci. 66, 13-27.
Barber, B.J., Blake, N.J., 2006. Reproductive Physiology. in: Shumway, S.E., Parsons,
G.J. (Eds.), Scallops: Biology Ecology and Aquaculture. Elsevier, pp. 357-416.
Benaglia, T., Chauveau, D., Hunter, D.R., Young, D.S., 2009. mixtools: An R Package
for Analyzing Finite Mixture Models. Journal of Statistical Software. 32, 1-29.
Carrillo, L., Palacios-Hernández, E., 2002. Seasonal Evolution of the Geostrophic
Circulation in the Northern Gulf of California. Est. Coast. Shelf Sci. 54, 157-173.
Carta Nacional Pesquera, 2004. Secretaria de Agricultura, Ganadería, Desarrollo Rural,
Pesca y Alimentación (SAGARPA). Diario Oficial de la Federación, México, pp.
76-189.
Carta Nacional Pesquera, 2006. Secretaria de Agricultura, Ganadería, Desarrollo Rural,
Pesca y Alimentación (SAGARPA). Diario Oficial de la Federación, México, pp.
1-128.
Cisneros-Mata, M.A., 2010. The importance of fisheries in the Gulf of California and
ecosystem-based sustainable co-management for conservation. In: Brusca, R.
(Ed.), The Gulf of California biodiversity and conservation. The University of
Arizona Press, Tucson, pp. 119-134.
163
Cruz, P., Ibarra, A.M., 1997. Larval growth and survival of two Catarina scallop
(Argopecten circularis, Sowerby, 1835) populations and their reciprocal crosses.
J. Exp. Mar. Biol. Ecol. 212, 95-110.
Cruz, P., Rodriguez-Jaramilllo, C., Ibarra, A.M., 2000. Environment and population
origin effects on first sexual maturity of catarina scallop, Argopecten ventricosus
(Sowerby II, 1842). J. Shellfish Res. 19, 89-93.
Cyr, C., Myrand, B., Cliche, G., Desrosiers, G., 2007. Weekly spat collection of sea
scallop, Placopecten magellanicus, and undesirable species as potential tool to
predict an optimal deployment period of collectors. J. Shellfish Res. 26, 10451054.
FAO, 2010. Global production statistics 1950-2008. Food and Agriculture Organization
of the United Nations - Fisheries and Aquaculture Information and Statistics
Service. http://www.fao.org/fishery/topic/16140/en
Félix-Pico, E.F., 2006. Mexico. in: Shumway, S.E., Parsons, G.J. (Eds.), Scallops:
Biology, Ecology and Aquaculture. Elsevier, Amsterdam, pp. 1337-1390.
Félix-Pico, E.F., Tripp-Quezada, A., Castro-Ortíz, J.L., Serrano-Casillas, G., GonzálezRamírez, P.G., Villalejo-Fuerte, M., Palomares-García, R., García-Domínguez, F.,
Mazón-Suástegui, J.M., Bojórquez-Verástica, G., Lopez-García, G., 1997.
Repopulation and culture of the Pacific Calico scallops in Bahía Concepción, Baja
California Sur, México. Aqua. Int., 551-563.
González-Anativia, C.R., 2001. Mercados y Comercialización de Pectínidos. In: MaedaMartínez, A.N. (Ed.), Los Moluscos Pectínidos de Iberoamérica: Ciencia y
Acuicultura. Editorial Limusa, México, pp. 451-468.
Luna-González, A., Cáceres-Martínez, J., Zúñiga-Pacheco, C., López-López, S.,
Ceballos-Vázquez, B.P., 2000. Reproductive cycle of Argopecten ventricosus
(Sowerby 1842) (Bivalvia: pectinidae) in the Rada del Puerto de Phichilingue, B.
164
C. S., México and its relation to temperature, salinity and food. J. Shellfish Res.
19, 107-112.
Maeda-Martínez, A.N., Omart, P., Mendez, L., Acosta, B., Sicard, M.T., 2000. Scallop
growout using a new bottom-culture system. Aquaculture. 189, 73-84.
Maeda-Martínez, A.N., Lombeida, P., Freites, L., Lodeiros, C.J., Sicard, M.T., 2001.
Cultivo de pectínidos en fondo y estanques. in: Maeda-Martínez, A.N. (Ed.), Los
moluscos pectínidos de Iberoamérica: Ciencia y acuicultura. Editorial Limunsa,
México, pp. 213-231.
Maeda-Martínez, A.N., Reynoso-Granados, T., Sólis-Marín, F., Leija-Tristán, A.,
Aurioles-Gamboa, D., Salinas-Zavala, C., Luch-Cota, D., Ormart, P., 1993. A
model to explain the formation of catarina scallop, Argopecten circularis
(Sowerby, 1835), beds, in Magdalena Bay, Mexico. Aqua. Fish. Manag. 24, 323339.
Maeda-Martínez, A.N., Reynoso-Granados, T., Monsalvo-Spencer, P., Sicard, M.T.,
Mazón-Suástegui, J.M., Hernandez, O., Segovia, E., Morales, R., 1997.
Suspension culture of Catarina scallop Argopecten ventricosus (=circularis)
(Sowerby II, 1842), in Bahia Magdalena, Mexico, at different densities.
Aquaculture. 158, 235-246.
Monsalvo-Spencer, P., Maeda-Martínez, A.N., Reynoso-Granados, T., 1997.
Reproductive maturity and spawning induction in the Catarina scallop Argopecten
ventricosus (=circularis) (Sowerby II, 1842). J. Shellfish Res. 16, 67-70.
Narvarte, M.A., 2001. Settlement of tehuelche scallop, Aequipecten tehuelchus D'Orb.,
larvae on artificial substrata in San Matias Gulf (Patagonia, Argentina).
Aquaculture. 196, 55-65.
Narvarte, M.A., Félix-Pico, E.F., Ysla-Chee, L.A., 2001. Asentamiento larvario de
pectínidos en colectores artificiales. In: Maeda-Martínez, A.N. (Ed.), Los
Moluscos Pectínidos de Iberoamérica: Ciencia y Acuicultura. Editorial Limusa,
165
México, pp. 173-192.
Orensanz, J.M., Parma, A.M., Turk, T., Valero, J., 2006. Population, Dynamics and
Management of Natural Scallops. in: Shumway, S.E., Parsons, G.J. (Eds.),
Scallops: Biology, Ecology and Aquaculture. Elsevier, Amsterdam, pp. 765-868.
Peña, J.B., 2001. Taxonomía, morfología, distribución y hábitat de los pectínidos
iberoamericanos. In: Maeda-Martínez, A.N. (Ed.), Los Moluscos Pectínidos de
Iberoamérica: Ciencia y acuicultura. Editorial Limusa, México, pp. 1-23.
Pouliot, F., Bourget, E., Frechette, M., 1995. Optimizing the design of giant scallop
(Placopecten magellanicus) spat collectors: field experiments. Mar. Biol. 123,
277-284.
R Development Core Team, 2009. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Viena, Austria.
Rangel-Dávalos, C., 1990. El cultivo de moluscos marinos en México. In: de la LanzaEspino, G., Arredondo-Ezpino, J.L. (Eds.), La acuicultura en México de los
conceptos a a la producción. Insituto de la Biología, Universidad Nacional
Autónoma de México, México, pp. 107-138.
Ruzzante, D.E., Zaixso, E., 1985. Settlement of Chlamys tehuelcha (D'Orb.) on artificial
collectors. Seasonal changes in spat settlement. Mar. Ecol. Prog. Ser. 26, 195-197.
Sánchez-Velasco, L., Lavín, M.F., Peguero-Icaza, M., León-Chávez, C.A., ContrerasCatala, F., Marinone, S.G., Gutiérrez-Palacios, I.V., Godínez, V.M., 2009.
Seasonal changes in larval fish assemblages in a semi-enclosed sea (Gulf of
California). Cont. Shelf Res. 29, 1697-1710.
Secretaria de Pesca, 1994. Cultivo de Almeja Catarina. Instituto Nacional de la Pesca,
Mexico.
Shumway, S., Parsons, G.J., 2006. Scallops: Biology, Ecology and Aquaculture, 2nd ed.
Elsevier.
166
Sicard, M.T., Maeda-Martínez, A.N., Ormart, P., Reynoso-Granados, T., Carvalho, L.,
1999. Optimum temperature for growth in the catarina scallop, Argopecten
ventricosus-circularis, Sowerby II, 1842. J. Shellfish Res. 18, 385-392.
Sicard, M.T., Maeda-Martínez, A.N., Lluch-Cota, S.E., Lodeiros, C., Roldán-Carrillo,
L.M., Mendoza-Alfaro, R., 2006. Frequent monitoring of temperature: an
essential requirement for site selection in bivalve aquaculture in tropical–
temperate transition zones. Aqua. Res. 37, 1040-1049.
Singnoret-Brailowski, G., Maeda-Martínez, A.N., Reynoso-Granados, T., Soto-Galera,
E., Monsalvo-Spencer, P., Valle-Meza, G., 1996. Salinity tolerance of the
Catarina scallop Argopecten ventricosus-circularis (Sowerby II, 1842). J.
Shellfish Res. 15, 623-626.
Thouzeau, G., 1991. Experimental collection of postlarvae of Pecten maximus (L.) and
other benthic macrofaunal species in the Bay of Saint-Brieuc, France. II.
Reproduction patterns and postlarval growth of five mollusc species. Aquaculture.
148, 181-200.
Uriarte, I., Rupp, G., Abarca, A., 2001. Producción de juveniles de pectínidos
Iberoamericanos bajo condiciones controladas. In: Maeda-Martínez, A.N. (Ed.),
Los Moluscos Pectínidos de Iberoamérica: Ciencia y Acuicultura. Editorial
Limusa, México, pp. 147-171.
Uribe, E., Blanco, J., 2001. Capacidad de los sistemas acuáticos para el sostenimiento del
cultivo de pectínidos: El caso de Argopecten purpuratus en la Bahía de Tongoy,
Chile. in: Maeda-Martínez, A.N. (Ed.), Los Moluscos Pectínidos de Iberoamérica:
Ciencia y Acuicultura. Editorial Limusa, México, pp. 233-248.
Villalejo-Fuerte, M., Ochoa-Baez, R.I., 1993. The reproductive cycle of the scallop
Argopecten circularis (Swerby, 1835) in relation to temperature and photoperiod,
in Bahia Concepcion, B.C.S., Mexico. Cienc. Mar. 19, 181-202.
167
Villalejo-Fuerte, M., Ceballos-Vásquez, B.P., 1996. Variación de los índices de
condición general, gonádico y de rendimiento muscular en Argopecten circularis
(Bivalvia: Pectinidae). Rev. Biol. Trop. 44, 591-594.
168
Tables
Table C.1: Sea water salinity, dissolved oxygen and temperature recorded at each collection site
in the study area of Puerto Peñasco.
La Cholla
Sandy
Beach
Las
Conchas
Los
Tanques
San Jorge
Island
San
Francisquito
35.43
35.38
35.68
35.20
34.77
35.17
34.85
35.40
35.53
35.56
35.58
35.07
34.85
34.73
34.90
ND
36.00
35.44
35.62
35.07
34.93
34.80
ND
35.40
35.80
35.46
35.36
35.10
34.92
35.22
34.40
35.50
35.20
35.00
35.30
34.88
34.48
34.30
35.20
35.00
35.58
35.10
35.18
34.67
34.35
34.52
35.50
35.40
07/24/07
08/17/07
10/19/07
12/18/07
02/28/08
04/25/08
06/27/08
08/25/08
Temperature
o
( C)
90.28
89.92
88.64
98.72
102.58
89.33
95.53
89.70
97.35
115.60
89.10
95.98
104.82
82.10
95.20
ND
97.20
110.20
90.34
94.48
96.07
67.00
ND
109.70
98.03
112.30
95.56
98.47
102.60
90.35
104.87
113.40
92.73
100.26
85.50
94.03
110.57
97.73
94.20
95.80
93.55
105.50
92.08
99.05
101.43
100.00
94.10
88.60
07/24/07
08/17/07
10/19/07
12/18/07
02/28/08
04/25/08
06/27/08
08/25/08
30.93
31.16
24.14
16.95
16.63
21.63
28.47
31.45
31.00
31.24
24.80
17.38
16.70
20.40
28.40
ND
30.86
31.10
24.84
16.92
17.17
19.93
ND
31.48
30.83
31.90
25.26
16.67
16.68
22.00
30.75
31.51
29.40
30.32
25.44
17.83
15.97
20.90
25.64
30.54
30.40
31.40
24.55
15.90
16.83
21.23
28.03
31.13
-1
Salinity (g l )
07/24/07
08/17/07
10/19/07
12/18/07
02/28/08
04/25/08
06/27/08
08/25/08
Dissolved
oxygen (%)
Table C.2: Argopecten ventricosus. Spat shell size frequency distributions at each site and month from December 2007 to June 2008. Cohorts
were identified fitting mixture distributions to the size dataset. Mean ( µ ), standard deviation (SD), and proportion (%) of data explained by
each component of the modeled mixture distributions.
Month
December
February
Site
%
µ
1
µ
2 (SD)
%
µ
2
µ
3 (SD)
%
µ
3
n
1.58 (0.58)
0.37
5.96 (2.55)
0.40
11.99 (4.46)
0.23
4860
Sandy Beach
1.43 (0.55)
0.34
7.40 (3.32)
0.64
17.68 (2.05)
0.02
5304
Las Conchas
1.63 (0.58)
0.37
4.07 (1.31)
0.14
7.21 (2.51)
0.49
6085
Los Tanques
1.39 (0.51)
0.31
5.47 (2.26)
0.45
9.43 (3.51)
0.24
3155
San Jorge Island
1.72 (0.65)
0.29
7.11 (2.81)
0.67
12.08 (4.12)
0.03
15856
San Francisquito*
5.62 (3.76)
0.99
9.19 (0.01)
0.01
--
--
2905
La Cholla
2.16 (0.89)
0.34
6.25 (2.16)
0.39
9.65 (3.26)
0.27
16924
Sandy Beach*
7.19 (3.14)
0.94
8.67 (1.11)
0.06
--
--
7273
Las Conchas
2.55 (1.02)
0.14
7.91 (2.56)
0.86
--
--
22180
Los Tanques*
5.97 (3.01)
0.97
18.41 (3.17)
0.03
--
--
16852
San Jorge Island
2.92 (0.89)
0.17
7.82 (2.72)
0.68
17.39 (3.75)
0.15
26802
San Francisquito*
6 (2.94)
0.76
14.42 (5.13)
0.24
--
--
22549
La Cholla
June
1 (SD)
La Cholla
2.04 (0.55)
0.14
7.37 (2.88)
0.69
14.84 (5.11)
0.16
19110
1.8 (0.7)
0.14
7.95 (2.79)
0.75
18.90 (3.36)
0.11
22805
Las Conchas*
7 (3.03)
0.89
18.94 (3.03)
0.11
--
--
19234
Los Tanques
2.21 (0.57)
0.15
7.40 (2.17)
0.58
17.22 (3.98)
0.27
7593
San Jorge Island*
5.48 (2.48)
0.67
14.24 (2.88)
0.19
18.92 (2.84)
0.14
6956
San Francisquito
2.35 (0.57)
0.27
11.49 (5.32)
0.72
--
--
773
La Cholla
1.83 (0.4)
0.11
5.44 (1.81)
0.19
15.38 (3.93)
0.70
19220
Sandy Beach*
4.75 (2.09)
0.36
13.76 (3.35)
0.65
--
--
14072
Las Conchas
1.98 (0.65)
0.11
6.90 (2.65)
0.32
14.90 (2.79)
0.56
10404
Los Tanques
1.83 (0.52)
0.13
4.63 (1.5)
0.19
12.75 (3.41)
0.69
8327
San Jorge Island
1.94 (0.56)
0.25
4.93 (1.88)
0.11
15.79 (4.62)
0.65
15228
San Francisquito
1.96 (0.72)
0.41
6.69 (2.95)
0.17
19.08 (3.46)
0.42
3018
Sandy Beach
April
µ
* In these cases the analysis did not differentiate the smaller spat (<3 mm in shell height size) as an independent cohort, which may represent a recently
recruited cohort.
169
170
Figures
Figure C.1. a) The Gulf California and the Baja California Peninsula, Mexico. b) The
study area of Puerto Peñasco showing spat collection sites (black) for catarina scallop,
Argopecten ventricosus. LCH: La Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA:
Los Tanques, SJO: San Jorge Island, and SFR: San Francisquito.
171
35000
800
700
600
500
400
300
200
100
0
30000
25000
20000
15000
10000
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
0
140
120
100
80
60
40
20
0
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Aquaculture production (Mt)
b)
5000
Captures (Mt) - Official
statistics Puerto Peñasco
Captures (Mt) - FAO statistics
a)
Year
Figure C.2. Catarina scallop Argopecten ventricosus. a) Annual fishery’s captures for
Northwest Mexico (solid line) (Source FAO, 2010), and Official captures declared at the
regional office of CONAPESCA in Puerto Peñasco (dashed line), Mexico. b) Annual
aquaculture production for Northwest Mexico (Source FAO, 2010). Values are expressed
in Mt of whole animal weight.
172
Sea level (low tide)
polaris
1m
Pangas
7m
Buoy
Rope (8mm)
Spat collector (Netlon)
Bag
5m
3m
0.4m
1m
Bottom
Earth-type anchor
Figure C.3. Diagram of vertical line collecting unit used at each spat collection site.
173
35
La Cholla *
Sandy Beach *
30
25
20
15
Temperature (oC)
10
35
Las Conchas
Los Tanques
30
25
20
15
10
35
San Jorge Island
San Francisquito
30
25
20
15
10
J A S O N D J F M A M J J A
J A S O N D J F M A M J J A
2007
2008
2007
2008
Time
Figure C.4. Sea bottom water temperature recorded every 4 hours at each spat collection
site along the Puerto Peñasco area from July 23rd, 2007 to August 24th, 2008. The set of
arrows in San Jorge Island panel represent the moment when collectors were replaced,
which is also representative for the other sites. We ended the field collection of spat on
August 24th, 2008. *Temperature loggers were lost for the periods were temperature data
is missing.
174
Mean number of spat collector-1
35
Sandy Beach
3000
30
2500
25
2000
20
1500
15
1000
10
500
5
0
0
3500
Las Conchas
35
Los Tanques
3000
30
2500
25
2000
20
1500
15
1000
10
500
5
0
0
3500
San Francisquito
San Jorge Island
Temperature (oC)
La Cholla
3500
35
3000
30
2500
25
2000
20
1500
15
1000
10
500
5
0
0
A
2007
O
D
F
A
2008
J
A
A
O
2007
D
F
A
2008
J
A
Time
Figure C.5. Monthly spat recruitment per collector (black bars) and monthly mean bottom
temperature (oC) (gray lines) at each collection site. Vertical lines represent standard
deviation.
175
4000
a) December
a
3000
a
a
2000
1000
aa
b
b
0
4000
b) February
a
a
Mean number of spat collector
-1
3000
a
a
a
a
2000
a
a,b
a
a,b
b
a,b
b
b
b
b
1000
0
4000
c) April
3000
a
2000
a
a
b
1000
aa
b
a
0
4000
d) June
a
3000
a
2000
a
a
a
a
b
1000
a
a
a
b
b
a
a
b
b
a
b
a
0
LCH SBE LCN LTA SJO SFR
Sites
Figure C.6. Mean number of spat recruited per collector at different depths 1 m ( ), 3 m
( ), 5 m ( ), and 7 m ( ) at each collection site at different months. Different letters
indicate significantly different values (Tukeys’s p < 0.05) between depths after one-way
ANOVAs for each site and month. Vertical lines represent standard deviation. LCH: La
Cholla, SBE: Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge
Island, and SFR: San Francisquito.
176
Frequency (%)
DECEMBER
La Cholla
n = 4860
Sandy Beach
n = 5303
Las Conchas
n = 6085
Los Tanques
n = 3155
San Jorge Island
n = 15856
San Francisquito∗
n = 2905
Shell height (mm)
Figure C.7. Modal analysis of size frequency distributions of Argopecten ventricosus spat
recruited on artificial collectors in December 2007 at different sites in Puerto Peñasco.
Cohorts were fitting mixture distributions to the size dataset. LCH: La Cholla, SBE:
Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR:
San Francisquito. * In these cases the analysis did not differentiate the smaller spat (<3
mm in shell height size) as an independent cohort, which may represent a recently
recruited cohort.
177
Frequency (%)
FEBRUARY
La Cholla
n = 16924
Sandy Beach∗
n =7273
Las Conchas
n = 22180
Los Tanques∗
n = 16852
San Jorge Island
n = 26802
San Francisquito∗
n = 22549
Shell height (mm)
Figure C.8. Modal analysis of size frequency distributions of Argopecten ventricosus spat
recruited on artificial collectors in February 2007 at different sites in Puerto Peñasco.
Cohorts were fitting mixture distributions to the size dataset. LCH: La Cholla, SBE:
Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR:
San Francisquito. * In these cases the analysis did not differentiate the smaller spat (<3
mm in shell height size) as an independent cohort, which may represent a recently
recruited cohort.
178
Frequency (%)
APRIL
La Cholla
n = 19110
Sandy Beach
n = 22805
Las Conchas∗
n = 19234
Los Tanques
n = 7593
San Jorge Island∗
n = 6956
San Francisquito
n = 773
Shell height (mm)
Figure C.9. Modal analysis of size frequency distributions of Argopecten ventricosus spat
recruited on artificial collectors in April 2007 at different sites in Puerto Peñasco. Cohorts
were fitting mixture distributions to the size dataset. LCH: La Cholla, SBE: Sandy Beach,
LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR: San
Francisquito. * In these cases the analysis did not differentiate the smaller spat (<3 mm in
shell height size) as an independent cohort, which may represent a recently recruited
cohort.
179
Frequency (%)
JUNE
La Cholla
n = 19220
Sandy Beach∗
n = 14072
Las Conchas
n = 10404
Los Tanques
n = 8327
San Jorge Island
n = 15228
San Francisquito
n = 3018
Shell height (mm)
Figure C.10. Modal analysis of size frequency distributions of Argopecten ventricosus
spat recruited on artificial collectors in June 2007 at different sites in Puerto Peñasco.
Cohorts were fitting mixture distributions to the size dataset. LCH: La Cholla, SBE:
Sandy Beach, LCN: Las Conchas, LTA: Los Tanques, SJO: San Jorge Island, and SFR:
San Francisquito. * In these cases the analysis did not differentiate the smaller spat (<3
mm in shell height size) as an independent cohort, which may represent a recently
recruited cohort.
180
APPENDIX D: CAPTACIÓN DE SEMILLAS DE BIVALVOS (POSTLARVAS) EN
LA ZONA DE BAHÍA DE KINO, SONORA, MÉXICO
Guía de trabajo
Captación de semillas de bivalvos (postlarvas) en la zona de
Bahía de Kino, Sonora, México
Protocolo elaborado por:
Gaspar Soria. Universidad de Arizona, Escuela de Recursos Naturales, BioSciences E
325D. Tucson, Arizona. 85721. USA Tel/fax. +1 (520) 621 7256.
[email protected]
Ivan Martinez. Centro Intercultural de Estudios de Desiertos y Oceanos, AC. (CEDO)
Apat. Postal #53. Puerto Peniasco, Sonora. CP 83550, México. Tel/fax. +52 (638) 3820113. [email protected]
181
Introducción
La captación de juveniles (semillas) de bivalvos comerciales mediante el uso de
sustratos artificiales o la recolección directa desde bancos naturales se realiza desde hace
ya muchos años en diferentes países. En México, existe mucha información sobre
captación de semillas en el golfo de California, principalmente para las especies como la
mano de león (Nodipecten subnodosus), la almeja catarina (Argopecten ventricosus) y el
callo de árbol (Pteria sterna).
Otra forma de producción de semillas es mediante el cultivo de individuos
adultos, su inducción al desove y posterior cultivo de las larvas y juveniles en tanques
especiales. Ejemplo de ello son las tareas se desarrollan en el Centro Reproductor de
Especies Marinas del Estado de Sonora (CREMES) en Bahía de Kino con el ostión
(Crassotrea gigas) y el callo de riñón (Atrina tuberculosa), callo de hacha (Pinna
rugosa).
¿Por qué queremos captar semillas de bivalvos?
Existen diversos motivos por los cuales es conveniente realizar captación de
semillas. Por ejemplo, en algunas situaciones es de interés captar semillas en grandes
cantidades para colocarlas en sistemas de engorda hasta que los individuos alcancen la
talla comercial adecuada para su venta. En nuestro caso estamos interesados en
comprender cuando las diferentes especies de bivalvos se reproducen, donde se depositan
o reclutan y en que época del año. También estamos interesados en comprobar modelos
de circulación de corrientes en la zona de bahía de Kino (Isla Tiburón, Isla San Pedro
Mártir etc.). Mediante la captación de semillas queremos estudiar los patrones de
circulación del agua que existen en el golfo de California y comprender de donde vienen
y hacia donde van los juveniles de bivalvos comerciales.
182
Ciclo de vida de bivalvos
El ciclo de vida de la mayoría de las especies de bivalvos se divide en dos fases,
una fase fija donde los individuos juveniles y adultos no se mueven (o se mueven muy
poquito) y viven fijos al sustrato como la escarlopa y una fase planctónica o móvil
durante el inicio de su vida. Esta última fase se la conoce como desarrollo larvario o fase
planctónica (Fig. 1). Algunas especies de bivalvos son hermafroditas como la mano de
león o la almeja catarina. Esto significa que un individuo tiene los dos sexos (macho y
hembra) y que por lo tanto liberara los dos productos sexuales (huevos y esperma). Otras
especies como el callo de escarlopa o el callo de hacha presentan los sexos separados.
Generalmente la fertilización del huevo se produce en el medio ambiente. Es aquí donde
se desarrolla una larva que puede vivir hasta 30 días dependiendo de las condiciones de
temperatura y alimento. Como larvas solo se mantienen durante un tiempo hasta que
llegan a una talla adecuada para asentarse. Es justo en este momento donde la larva
desarrolla un pie móvil para buscar un sustrato y se las denomina larvas competentes. Si
la larva encuentra un sustrato a gusto ahí se quedara y dejara de nadar. Si por el contrario
no le ayuda el sustrato, las larvas pueden resuspenderse y volver a nadar. Al proceso de
búsqueda de sustrato y fijación lo llamamos comúnmente asentamiento larvario. Las
larvas asentadas recientemente se las denomina semillas (Fig. 1). En este proyecto se
colocaran colectores a diferentes profundidades para captar larvas competentes, es decir
aquellas que están buscando sustrato para asentarse.
La duración de la fase larvaria hasta el desarrollo de larvas competentes es muy
variable y depende de las condiciones del mar (temperatura, alimento) y de las
características de la especie. Por ejemplo las larvas de callo de hacha duran 21 días
mientras que las larvas de escarlopa pueden llegar a ser competentes en 15 días (Fig. 1).
183
Asentamiento
Semillas
Larva
pediveliger o
competente
Fase
Fija
Colector
Larva “D”
Fase
Planctónica
Adultos
huevos
esperma
Larva trocófora
Figura E.1. Ciclo de vida de bivalvos comerciales como el callo de escarlopa, callo de
riñón y madreperla.
Materiales y métodos
Diseño de colectores
Para este proyecto se utilizaran colectores artificiales de plástico usados
comúnmente para la captación de larvas competentes de especies como la almeja
catarina, mano de león o voladora (Fig. 2). El nombre comercial es red de netlon y su
color es azul. Sobre estos colectores también asentaran otras especies como hachas,
escarlopa, madreperla y callo de árbol. Todas estas especies cuando asientan son de
forma y color exactamente igual a los individuos adultos pero el tamaño será mucho
menor (a penas uno o dos milímetros).
Estas redes colectoras deben llevar por fuera otra red o bolsa de color verde. El
propósito de esta bolsa externa es para evitar perder semillas por efecto del las corrientes.
Las larvas competentes pueden atravesar la bolsa y asentarse en los colectores sin ningún
problema.
184
En nuestro estudio elaboraremos líneas de colectores para poder evaluar la
presencia de larvas a diferentes profundidades tal como se muestra en la figura 2. Las
diferentes profundidades así como el tipo de fondeo o ancla serán definidas de acuerdo al
sitio de estudio.
Nivel del mar en bajamar
polaris
2-3m
Pangas
Boya
Línea madre (8mm)
7m
Colector (Netlon)
Bolsa protectora
13-15m
5m
3m
0.6-0.8m
1m
Fondo
Ancla
Figura E.2. Diagrama de una línea de colectores con 8 redes colectoras de netlon.
Características técnicas:
Línea madre: 6 u 8 mm
Boya: de 5L de capacidad (Fig. 3a).
Redes colectoras: largo= 2 m, deben enrollarse sobre si mismas para formar un
bollo de 40 cm de largo (Fig. 3a).
Bolsa verde: 40 x 60 cm. Apertura de malla= 2mm (Fig. 3b)
Anclas: barras de hierro de 75 cm de largo de las que se enroscan en el fondo
185
La línea madre se amarrara a las anclas mediante nudos tipo cochi y un as de guía
con un remate.
Las bolsas (conteniendo un colector cada una) se amarraran de a pares a las líneas
madres mediante un nudo cochi. Las bolsas serán atadas cada 2 metros a partir del fondo
y hasta los 7 u 9 m de profundidad, dependiendo de las condiciones del lugar (Fig. 3 a-d).
Una vez reemplazadas las líneas pueden ser transportadas al aire libre, esto
ayudara a que las semillas se sequen y sea más fácil su extracción de cada colector.
Manejo y procesado
Los colectores serán reemplazados cada 2 meses, momento en que serán traídos a
laboratorio para la limpieza, clasificación y fijación del material biológico en alcohol.
Todo el material biológico será almacenado en alcohol etílico (común que se compra en
las farmacias) al 70% en frascos de vidrio o plástico con tapa a rosca. El cerrado debe ser
hermético para evitar la descomposición de las muestras y perdida de alcohol.
Para el procesado de los colectores se debe disponer de una mesa de trabajo de 2
m de largo por 60 cm de ancho y agua corriente (Fig. 4 y 5).
Luego bajo lupa estereoscópica las semillas serán clasificadas, contabilizadas y
medidas (Fig. 6).
El material biológico de cada colector deberá ser rotulado con la siguiente
información:
Fecha, Lugar, Profundidad y número de línea.
Las muestras deben ser guardadas en lugar fresco y seco.
186
a)
b)
d)
c)
c)
d)
Figura E.3: Diferentes etapas de armado de colectores y vista parcial de los colectores
instalados. a) Boyas, b) Redes colectoras, c) redes colectoras dentro de bolsas, d) líneas
colectoras, e) amarre de una línea a un ancla de hierro y f) vista parcial desde debajo de
una línea colectora.
187
Figura E.4. Red colectora con semillas de hachas.
Figura E.5: Procesado para extracción de las semillas pegadas a la bolsa (verde) y a la red
colectora (azul).
Resultados esperados y potencialidad del estudio
Los resultados obtenidos en este estudio contribuirán a mejorar los modelos
oceanográficos de dispersión y conectividad desarrollados para la región del Norte del
Golfo de California, específicamente para la zona comprendida entre Isla San Pedro
Mártir y Puerto Peñasco.
Este estudio contribuirá con información necesaria para el desarrollo de planes y
herramientas de manejo pesquero, incluyendo el diseño de áreas pesqueras protegidas.
188
Este trabajo proveerá información básica para la captación de juveniles de
bivalvos comerciales, información necesaria para el desarrollo de actividades productivas
y de repoblamiento de bancos naturales en la región.
P. mazatlanica
S. princes
S calcifer
P. sterna
1 mm
a)
E. vogdesi
c)
1 mm
b)
A. purpuratus
1 mm
Pinnidae
d)
1 cm
Figure E.6. Diferentes especies de bivalvos de importancia comercial que puede ser
encontrados en colectores artificiales. a) Madre perla (P. mazatlanica) y callo de árbol
(Pteria sterna), b) Almeja mechuda (S. princeps) y callo de escarlopa (S. calcifer), c)
almeja voladora (E. vogdesi) y almeja catarina (A. ventricosus), y d) hachas (Pinnidae).
Literatura
Los Moluscos Pectínidos de Iberoamerica. Ciencia y Acuicultura (2001). MaedaMartínez Ed. Editorial Limusa. 235pp
Félix-Pico E. (2006) Mexico. En: Shumway SE, Parsons GJ (Eds). Scallops:
Biology, Ecology and Aquaculture. Elsevier, Amsterdam, p 1137-1390
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement