Ontogeny of Similarity and Disparity in Length-at-Age of Coral Reef Fishes in
the Family Acanthuridae Between Marine Protected and Fished Areas
by
Mathias Tjarko Cramer
Bachelor of Arts Mount Allison University 2017
A thesis submitted to the Department of Ocean Engineering and Marine Sciences at the Florida Institute of Technology in partial fulfillment of the requirements for the degree of
Master of Science in Biological Sciences Marine Biology
Melbourne, Florida May 2019
We the undersigned committee hereby recommend that the attached document be accepted as fulfilling in part the requirements for the degree of Master of Science in Biological Sciences
“Ontogeny of Similarity and Disparity in Length-at-Age of Coral Reef Fishes in the Family Acanthuridae Between Marine Protected and Fished Areas” a thesis by Mathias Tjarko Cramer
Ralph G. Turingan, Ph.D., Committee Chairperson Professor of Biological Sciences and Program Chair Department of Ocean Engineering and Marine Sciences
Jonathan Shenker, Ph.D., Committee Member Associate Professor of Biological Sciences Department of Ocean Engineering and Marine Sciences
Kelli Hunsucker, Ph.D., Outside Member Assistant Professor of Oceanography Department of Ocean Engineering and Marine Sciences
Richard Aronson, Ph.D. Professor and Department Head Department of Ocean Engineering and Marine Sciences
Abstract
Ontogeny of Similarity and Disparity in Length-at-Age of Coral Reef Fishes in the Family Acanthuridae Between Marine Protected and Fished Areas by Mathias Tjarko Cramer
Chairperson of Advisory Committee: Ralph G. Turingan, Ph.D.
Drastic reductions in the harvestable biomass of exploited fish populations on coral reefs and other marine coastal ecosystems have been hypothesized to be a consequence of astronomical and unrelenting fishing pressures, particularly in the coastal small-scale and sustenance fisheries of the world. It has recently been proposed that the causal relationship between fishing pressure and fish biomass is rooted in the negative effects of size-selective fishing mortality, which consequently instigates directional shifts in phenotype (i.e., fishing induced evolution). To mitigate the effects of unsustainable fishing pressures, parcels of traditional fishing grounds have been designated as Marine Protected Areas
(MPAs). To date, the assessment of the efficacy of MPAs as fisheries management tools has been primarily limited to visual census surveys, largely because the “no-take” designation of MPAs precludes repeated sampling. As such, this investigation capitalized on a rare opportunity to perform an approved static sampling excursion in 2015 on four acanthurid species (Acanthuridae,
Teleostei) between five pairs of MPAs and adjacent fished areas (AFAs) within a network of coral reefs along the coast of the Zambales Province, Philippines to introduce the novel use of a “back-calculation” technique into the field of MPA assessment. For every specimen, standard length, a proxy for fitness, was measured and otoliths extracted, after which ageing was performed in iii
collaboration with the Age and Growth Laboratory of the Florida Fish and
Wildlife Commission Fish and Wildlife Research Institute. Back-calculated lengths-at-age were examined for spatial (MPA vs. AFA) and temporal (inter- annual) similarities and disparities throughout the observed life history. Analyses revealed five major findings. First, newly settled A. nigrofuscus were phenotypically similar at ages one and two. Second, A. nigrofuscus transformed into significantly different length-at-age morphs between ages three and six, in which MPA populations were predominantly larger than AFA populations. Third, this phenotypic divergence disappeared towards the latter life history stages.
Fourth, the observed life histories of the other acanthurid species were dominated by spatial phenotypic homogeneity. Fifth, all species displayed prolonged temporal growth. These observed patterns may result from (1) a release from fishing pressures within MPAs, (2) spatiotemporal variations in food and habitat resources between MPAs and AFAs and (3) density-dependent forces that could promote phenotypic change through phenotypic plasticity and, potentially, subsequent genotypic adaptation. Altogether, four major conclusions are drawn from this investigation. First, the results presented here confirm the success of this methodology, promoting the MFBC model as a bio-mathematical tool capable of accurately reconstructing the size-related life histories of coral reef fishes and thus conferring a novel solution to the logistic and legal restrictions placed on repeated sampling within protected areas. Second, analyses determined that the sampled
MPAs are gradually restoring historical body sizes in exploited regions, both within and outside of their borders. Third, patterns of phenotypic convergence and divergence were variable, including within the same genus (Ctenochaetus), iv
suggesting that these phenotypic responses are species-specific. As a result, assessments should not make assumptions that the phenotypic responses of a single or select few species are representative of all other species contained with the protected area. Fourth, the results of this investigation perfectly example the caution that needs to be taken when interpreting the findings of static spatial analyses that were performed in attempt to assess dynamic tools. A lack of phenotypic divergence between MPAs and AFAs has historically been interpreted as a failure of the MPA to instigate phenotypic enlargements, which is disproven in this investigation (i.e. C. binotatus, C. striatus and Z. scopas). As a result, investigators should refrain from using the results of independent static spatial analyses as a proxy of the biophysical performance of MPAs. Although these findings offer critical insight and a comprehensive provision of MPA functionality within the Northwest Philippine region of the Coral Triangle, the ubiquity of the observed phenotypic responses across other taxa and regions is unknown. Whilst further spatiotemporal analyses are required across other regions and taxa of differential exploitation and life histories, this investigation adds to the growing body of evidence that MPAs can provide positive conservation and fishery benefits. Continued studies identifying the positive outcomes of MPAs will be critical in garnering support for the establishment of MPAs in the coming decades.
v
Table of Contents
Abstract ...... iii
Keywords ...... vii
List of Figures ...... viii
List of Tables...... xii
List of Abbreviations...... xiii
Acknowledgments ...... xiv
Dedication ...... xv
Introduction ...... 1
Materials and Methods ...... 8
Target Species ...... 8
Sampling Locations and Permits ...... 8
Length and Otolith Dissections ...... 11
Modified Fry Back-Calculation Model ...... 11
Statistical Analyses ...... 12
Results ...... 17
Acanthurus nigrofuscus ...... 17
Ctenochaetus binotatus ...... 30
Ctenochaetus striatus ...... 42
Zebrasoma scopas ...... 51
Discussion ...... 61
References ...... 72 vi
Keywords
Acanthuridae Acanthurus nigrofuscus Back-Calculation Body Length Conservation Coral Reef Fishes Ctenochaetus binotatus Ctenochaetus striatus Fisheries-Induced Evolution Fishing Marine Protected Area Marine Reserve Ontogeny Phenotype Phenotypic Plasticity Philippines Surgeonfishes Zebrasoma scopas
vii
List of Figures
Figure 1. Photographs of the target species: Acanthurus nigrofuscus,
Ctenochaetus binotatus, Ctenochaetus striatus, Zebrasoma scopas
...... 9
Figure 2. The five MPA and AFA sites at which sampling occurred ...... 10
Figure 3. The left sagittal otolith of a three-year-old Ctenochaetus striatus
specimen with annotated annuli rings ...... 14
Figure 4. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 1 cohorts between 2003 and 2014 ...... 20
Figure 5. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 2 cohorts between 2004 and 2014 ...... 21
Figure 6. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 3 cohorts between 2005 and 2014 ...... 22
Figure 7. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 4 cohorts between 2006 and 2014...... 23
Figure 8. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 5 cohorts between 2007 and 2014 ...... 24
Figure 9. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 6 cohorts between 2008 and 2014 ...... 25
Figure 10. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 7 cohorts between 2009 and 2014 ...... 26
Figure 11. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 8 cohorts between 2010 and 2014 ...... 27
viii
Figure 12. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 9 cohorts between 2011 and 2014 ...... 28
Figure 13. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA
age 10 cohorts between 2012 and 2014 ...... 29
Figure 14. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 1 cohorts between 2003 and 2014 ...... 32
Figure 15. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 2 cohorts between 2004 and 2014 ...... 33
Figure 16. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 3 cohorts between 2005 and 2014 ...... 34
Figure 17. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 4 cohorts between 2006 and 2014 ...... 35
Figure 18. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 5 cohorts between 2007 and 2014 ...... 36
Figure 19. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 6 cohorts between 2008 and 2014 ...... 37
Figure 20. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 7 cohorts between 2009 and 2014 ...... 38
Figure 21. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 8 cohorts between 2010 and 2014 ...... 39
Figure 22. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 9 cohorts between 2011 and 2014 ...... 40
Figure 23. LOESS regressions of the Ctenochaetus binotatus MPA and AFA
age 10 cohorts between 2012 and 2014 ...... 41 ix
Figure 24. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 1 cohorts between 2006 and 2014 ...... 44
Figure 25. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 2 cohorts between 2007 and 2014 ...... 45
Figure 26. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 3 cohorts between 2008 and 2014 ...... 46
Figure 27. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 4 cohorts between 2009 and 2014 ...... 47
Figure 28. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 5 cohorts between 2010 and 2014 ...... 48
Figure 29. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 6 cohorts between 2011 and 2014 ...... 49
Figure 30. LOESS regressions of the Ctenochaetus striatus MPA and AFA
age 7 cohorts between 2012 and 2014 ...... 50
Figure 31. LOESS regressions of the Zebrasoma scopas MPA and AFA age 1
cohorts between 2005 and 2014 ...... 53
Figure 32. LOESS regressions of the Zebrasoma scopas MPA and AFA age 2
cohorts between 2006 and 2014 ...... 54
Figure 33. LOESS regressions of the Zebrasoma scopas MPA and AFA age 3
cohorts between 2007 and 2014 ...... 55
Figure 34. LOESS regressions of the Zebrasoma scopas MPA and AFA age 4
cohorts between 2008 and 2014 ...... 56
Figure 35. LOESS regressions of the Zebrasoma scopas MPA and AFA age 5
cohorts between 2009 and 2014 ...... 57 x
Figure 36. LOESS regressions of the Zebrasoma scopas MPA and AFA age 6
cohorts between 2010 and 2014 ...... 58
Figure 37. LOESS regressions of the Zebrasoma scopas MPA and AFA age 7
cohorts between 2011 and 2014 ...... 59
Figure 38. LOESS regressions of the Zebrasoma scopas MPA and AFA age 8
cohorts between 2012 and 2014 ...... 60
xi
List of Tables
Table 1. Comparison of the number of fish (i.e., sample size) between the
original stocks of fish used for otolith ageing and the pooled fish
(i.e., cohort) that resulted from the back-calculation methodology
...... 15
Table 2. Acanthurus nigrofuscus mean lengths (x̅ ) and standard errors (σx̅ )
for each age cohort separated by management status, along with
Two-Way ANOVA F and p-values for both Status and Year effects
...... 19
Table 3. Ctenochaetus binotatus mean lengths (x̅ ) and standard errors (σx̅ )
for each age cohort separated by management status, along with
Two-Way ANOVA F and p-values for both Status and Year effects
...... 31
Table 4. Ctenochaetus striatus mean lengths (x̅ ) and standard errors (σx̅ ) for
each age cohort separated by management status, along with Two-
Way ANOVA F and p-values for both Status and Year effects ....43
Table 5. Zebrasoma scopas mean lengths (x̅ ) and standard errors (σx̅ ) for
each age cohort separated by management status, along with Two-
Way ANOVA F and p-values for both Status and Year effects ....52
xii
List of Abbreviations
AFA: Adjacent Fished Area
FIE: Fisheries-Induced Evolution
MFBC: Modified Fry Back-Calculation
MPA: Marine Protected Area
MR: Marine Reserve
LOESS: Locally Estimated Scatterplot Smoothing
xiii
Acknowledgements
I would like to acknowledge the Florida Institute of Technology, particularly the Department of Ocean Engineering and Marine Sciences and the
Center for Corrosion and Biofouling Control for their financial support through
Graduate Student Assistantships.
I would like to acknowledge the U.S. Student Fulbright Fellowship, a collaborative program between the United States Department of State, the
Institute of International Education and the Philippine-American Educational
Foundation, which partially funded this investigation.
I would like to acknowledge the Department of Agriculture of the
Philippines, the Bureau of Fisheries and Aquatic Resources of the Philippines and the local governmental units of Candelaria, Masinloc and Santa Cruz for granting permission to conduct sampling excursions between the fore-mentioned Marine
Protected Areas and Adjacent Fished Areas.
I would like to thank the chairperson of my advisory committee, Dr. Ralph
G. Turingan, for his immense amount of mentorship and guidance that has undoubtedly molded me into the marine biologist that I am today. I would like to extend my thanks to Dr. Jonathan Shenker and Dr. Kelli Hunsucker, whose collaboration transformed this investigation into a focused yet well-rounded piece of scientific literature. I would also like to extend my thanks to the department head, Dr. Richard Aronson, and administrative assistant, Dee Dee van Horn, for their direction, counseling and never-ending willingness to help. Lastly, I would like to thank Dr. Robbie Fidler for granting me permission to use a dataset that he invested an incredible amount of time and effort into collecting. xiv
Dedication
I dedicate this thesis to my family. Whilst no words can do justice to the magnitude of thanks that I owe you all, I will try my best. To Ma and Pa, thank you for the unconditional love and support that you have blessed me, without which I undoubtedly would not have achieved this lifelong dream. To Michi and
Alexito, thank you for your unwavering friendship and companionship that helped me persevere through the struggles of graduate school. To Omi, thank you for your endless inspiration. To Masha, thank you for being the strong and loving person that you are, I would not have been able to accomplish this feat without you. I will forever be grateful to you all.
xv
1. Introduction
The exploitation of marine environments has long been an anthropogenic tradition, with conception dating back to the earliest of civilizations (Braudel
2001). Human success at marine exploitation had always been hindered by an inability to control the biophysical processes of the marine environment, but the birth of the industrial revolution broke down this barrier. Technological advancements that eased the financial resources and manpower required for adequate exploitation resulted in a proliferation of fisheries that, subsequently, damaged marine environments worldwide. Unfortunately, the proverb “out of sight, out of mind” has never proven truer. Whilst terrestrial environments were the first to receive legal protection due to undisputable anthropogenic destruction, the damage imposed on marine environments remained largely unknown until the mid-to-late 20th century, evidenced by the collapse of the Atlantic Northwest cod fishery in 1992 (Myers et al. 1997).
Rates of marine exploitation have since increased, drastically reducing the amount of harvestable biomass in fisheries worldwide and threatening one of humans’ most critical food markets (FAO 2016; Worm et al. 2009; Bell et al.
2017). Decades of research have determined that populational collapses can be instigated by various factors, but a primary catalyst is alteration of fish phenotype by directional fishing pressure. For example, typical fishing practices are size- selective, preferentially targeting larger individuals and thus enforcing a directional survival advantage for smaller individuals (Swain et al. 2007;
Birkeland and Dayton 2005; Uusi-Heikkila et al. 2008). Subsequent directional shifts in body size ultimately diminish individual fitness and population 1
sustainability through their respective effects on reproduction and ontogeny
(Calder 1985; Reiss 1989; Rochet 2000; Swain et al. 2007; Heino and Dieckmann
2009). Whilst larger body sizes confer greater survival probability, reproductive output, and offspring viability (Barneche et al. 2018; Lewin et al. 2006), smaller body sizes are highly correlated with the opposite, including diminished egg production, egg size, overall fecundity (Bobko and Berkeley 2004; Berkeley et al.
2004; Birkeland and Dayton 2005), size-at-hatch, feeding rate, and viability
(Haugen and Vollestad 2001; Walsh et al. 2005). Collectively, these alterations are referred to as fisheries-induced evolution (FIE) (Trippel 1995; Heino and
Godo 2002, Law 2007), and they have become a primary concern of fisheries management efforts.
In response, Marine Protected Areas (MPAs) and, more specifically,
Marine Reserves (MRs) have become popular fisheries management and biodiversity conservation tools (Kelleher and Kenchington 1991; Edgar et al.
2007). MPAs are defined as geographical marine areas that have been designated by a governing authority to strengthen the long-term sustainability of the contained natural resources (Claudet 2011). MRs, on the other hand, are specific types of MPAs in which all exploitative activities are strictly prohibited. To allow for uniform terminology with past studies, all protected areas will simply be referred to as MPAs from hereon. The concept of an MPA was in fact an accidental discovery, in which North Sea fisheries witnessed a restoration of stock and body sizes after a prolonged interruption during World War II (Gulland
1974). Since then, the hypothetical function of an MPA has been to serve as a refuge in which subsets of exploited populations can be shielded from the harmful 2
fishing practices that have been well documented to produce FIE (Gell and
Roberts 2003). In theory, MPAs should allow for the restoration of historical body-size distributions by eliminating size-selective fishing mortality and resuming ecological forces that naturally promote larger body-sizes (Trexler and
Travis 2000; Edeline et al. 2007; Fidler et al. 2018). In turn, it is expected that protected populations will subsidize adjacent fished areas (AFAs) through larval seeding and post-settlement spillover (Russ 2002; Harrison et al. 2012).
Whilst many studies have found increased biomass, biodiversity, and size- structures within MPAs compared to AFAs (Lester et al. 2009; Fidler et al. 2017;
Babcock et al. 2010; Russ and Alcala 2011), definitive assessments of MPA functionality require extensive repeated sampling to map the phenotypic responses of populations to protection. Unfortunately, the “no take” status of
MPAs precludes the conduct of repeated sampling to examine longitudinal trends in the biology of exploited stocks inside and outside these protected sites. As a result, current understandings of the capacity of MPAs to achieve their stated biological goals remain largely obscure. To overcome the challenges of repeated sampling within MPAs, this study introduces the application of a traditional fisheries tool known as a “back-calculation” to procure length-at-age life history data from fish collected between MPAs and AFAs during an approved static sampling excursion (see section 2.2).
The term “back-calculation” was most eloquently defined by Francis
(1990) as “a technique that uses a set of measurements made on a fish at one time to infer its length at an earlier time or times.” The origin of this technique is believed to have stemmed from Reibish (1899) who first used otoliths for age 3
determination, a method that was immediately transformed by Lea (1910) to provide the first account of using scales to reconstruct growth patterns. This theory then underwent extrapolation via Hickling’s (1933) research that applied the technique to otoliths, which has since become the standard method (Secor et al. 1995; Casselman 1990). Biometric advancements in the first half of the 20th century, followed by the onset of the digital revolution in the late 1950s, instigated a proliferation of back-calculation models (Francis 1990; Vigliola and
Meekan 2009). Recent meta-analytic investigations that comparatively validated these models determined that the recently developed Modified Fry Back-
Calculation (MFBC) model was optimal (Vigliola et al. 2000), producing accurate and precise outputs that had been validated with longitudinal data (Wilson et al.
2009). Recent validation coupled with the technique’s ability to ameliorate sampling restrictions transform the MFBC model into a seemingly perfect solution to the difficulties experienced during within-MPA repeated sampling, a notion that is further supported by Vigliola and Meekan (2009) who state that “in many cases, the use of back-calculation techniques is not a choice, it is a necessity.”
In addition to easing data collection, back-calculations provide invaluable information on the spatiotemporal dynamics and life history patterns of a population. Traditionally, a primary function of back-calculations has been to ease the attainment of length-at-age data in order to construct growth curves, which have been used extensively throughout the 20th century in the fields of aquaculture and fisheries management. More recently, these back-calculated growth curves have been integrated into mixed-effects models to identify intrinsic and extrinsic 4
drivers of growth (Martino et al. 2018). When a generational approach is implemented, these growth curves can additionally provide insight into shifting demographics and phenotypes (Taylor and McIlwain 2010).
These traditional back-calculation approaches are easily adopted for analyses of MPA functionality, most obviously because growth models are highly indicative of a population’s specific biophysical scenario. As a result, analyses of age-specific growth curves between MPA and AFA populations can provide critical insight into their respective, generational responses to protection.
Ultimately, this information not only allows for a more accurate examination of
MPA performance, but it also enhances our ability to inform fisheries managers of potential fallbacks and possible renovations to management techniques. This practical application is pursued in this current case study, in which a rare opportunity to perform a static sampling between a network of no-take MPAs and
AFAs in the Philippines was used to introduce the novel application of a back- calculation model to investigate disparities in the phenotypic ontogenies of conspecifics. Specifically, length-at-age data obtained through otolith-based back- calculations of four species within the Acanthuridae family were used to construct age-specific growth models, allowing for in-depth spatiotemporal analyses of the species’ regional phenotypic responses to protection.
The Acanthuridae family was chosen for numerous reasons, the first and foremost being their critical anthropogenic importance. In Southeast Asia alone, an approximate 350 million people live within a 50-kilometer radius of the coastline (Clifton et al. 2010), of which an approximate 120 million rely directly on marine resources for economic and dietary stability (Moberg and Folke 1999, 5
Dutton 2005, Cullen 2007). Broader investigations estimate that near shore fisheries, of which acanthurids comprise a major component, sustain the primary source of dietary protein for an approximate one billion people in Asia (Clifton et al. 2010). Second, acanthurids are primarily classified as grazers and detrivores that feed on the epithilic algal matrix and thereby perform the critical ecological process of mitigating macroalgal colonization on coral reefs (Wilson and
Bellwood 1997; Green and Bellwood 2009; Nash et al. 2013). As a result, severe
FIE could interfere with the group’s capacity to adequately mitigate algal growth and consequently harm the health of coral reef ecosystems. In fact, recent investigations have witnessed this exact outcome in regions where populations of parrotfish have been heavily harvested (Bellwood et al. 2012). Third, and perhaps most shockingly, is that there remains a major gap in our current knowledge of acanthurid species’ populational health in Southeast Asia. These concerns are intensified in the wake of analyses that have determined, via the application of the
IUCN / SSC’s Susceptibility Matrix, that many acanthurid species exhibit extrinsic as well as intrinsic factors that make them particularly susceptible to localized extirpation (Donaldson 2007).
As fish stocks continue to decline and reliance on MPAs as a primary fisheries management tool increases, it is critical that we gain a more thorough understanding of the impacts of protection on both protected as well as harvested fish populations. Whilst an intra-familial approach has been implemented in this introductory investigation into the regional responses of acanthurid species to
MPAs in the Northwest Philippine region of the Coral Triangle, subsequent inter- familial investigations across other regions and taxa of differential exploitation 6
and life histories will be required to assess the generality of the observed phenotypic responses to protection.
7
2. Materials and Methods
2.1 Target Species
Four acanthurid species were targeted for either their dietary importance in the Coral Triangle region (1. Acanthurus nigrofuscus: Brown Surgeonfish, 2.
Ctenochaetus binotatus: Two-spot Bristletooth, 3. Ctenochaetus striatus: Lined
Bristletooth) or popularity in the aquaria industry (4. Zebrasoma scopas: Brushtail
Tang) (Figure 1).
2.2 Sampling Locations and Permits
Specimens were collected in 2015 across five strongly guarded MPAs and
AFAs that spanned three municipalities within the Zambales Province of Luzon,
Philippines: 1. Sinabacan-Malimanga MPA (Municipality of Candelaria [est.
2011; 124 ha]), 2. Hermana Menor MPA (Municipality of Santa Cruz [est. 2003;
266 ha]), 3. Bani MPA (Municipality of Masinloc [est. 2006; 50 ha]), 4. San
Salvador MPA (Municipality of Masinloc [est. 1989; 127 ha]) and 5. Taklobo
Farm (Municipality of Masinloc [est. 1989; 2 ha]) (Figure 2). All fished reef specimens were collected at least 300 meters away from MPA boundaries to ensure exposure to typical fishing pressures.
Approvals from the Department of Agriculture of the Philippines, the
Bureau of Fisheries and Aquatic Resources of the Philippines and the local governmental units of Candelaria, Masinloc and Santa Cruz were obtained.
Sampling was performed via SCUBA and spearfishing equipment, and was in accordance with the animal-handling protocol provided in the Republic of the
Philippines, Department of Agriculture Gratuitous Permit GP-0096-15 issued on
June 01, 2015. 8
Figure 1. Clockwise from top left: 1. Acanthurus nigrofuscus: Brown
Surgeonfish, 2. Ctenochaetus binotatus: Two-spot Bristletooth, 3.
Ctenochaetus striatus: Lined Bristletooth, 4. Zebrasoma scopas:
Brushtail Tang. Photographs were obtained through FishBase and
remain the property of the respective photographer: 1. Richard
Field, 2. David Cook, 3. Richard Field, 4. Jim Greenfield.
9
Figure 2. The five MPA and AFA sites at which sampling occurred. Images
adapted from Stamen Maps (maps.stamen.com/).
10
2.3 Length and Otolith Dissections
Following collection, all specimens were euthanized via ice immersion and then transported back to a laboratory at which standard length (mm) measurements were taken. Sagittal otoliths were removed and prepared for ageing using the protocol described in Fidler et al. 2018. Ageing was performed in collaboration with the Age and Growth Laboratory of the Florida Fish and
Wildlife Commission Fish and Wildlife Research Institute (FWC-FWRI), guaranteeing accuracy and precision. A comparison of age distributions that had been independently determined by multiple professional otolith readers ensured a percent error below FWC-FWRI’s standard threshold of ten percent.
2.4 Modified Fry Back-Calculation Model
The Modified Fry Back-Calculation (MFBC) model was selected for its robustness and validated capacity to procure accurate length data in spite of allometric growth between a fish’s body and otoliths (Vigliola et al. 2000; Wilson et al. 2009; Vigliola and Meekan 2009). The back-calculated length (L) of a fish at age (i) was calculated as:
퐿푖 = 푎 + exp ( ln (퐿0푃 −
[ln (퐿 – 푎) – ln (퐿 – 푎)] [ln (푅 ) – ln (푅 )] 푎) + 푐푝푡 0푃 푖 0푃 ) [ln (푅푐푝푡) – ln(푅0푃))
where Lcpt is the length of a fish at capture, Rcpt is the radius of a fish’s otolith at capture, a is the biological intercept between Lcpt and Rcpt, L0P is the length of a
11
fish at increment formation, Ri is the otolith radius at age i, and R0P is the mean radius of first annulus in the otolith.
ImageJ bundled with Java 1.8.0_172 was used to obtain incremental radial measurements, defined as the distance from the nucleus of an otolith to the end of any given annulus, from photographs of specimens’ left sagittal otoliths (Figure
3). The MFBC model was coded and applied on the R Statistics Software v. 3.5.0.
Packages required were: “FSA” and “openxlsx”. The total numbers of specimens collected within each age cohort are presented alongside the updated, back- calculated sample sizes in Table 1.
2.5 Statistical Analyses
Similarities or disparities in length-at-age were statistically analyzed via
Two-Way Analyses of Variance (Two-Way ANOVAs), which allowed for examinations of (1) age-specific differences between MPA and AFA populations
(referred to as “status effects”) as well as (2) age-specific differences between year classes (referred to as “year effects”). Given the expectation that protection would create a more advantageous biophysical environment within MPAs, it was hypothesized that MPA populations would exhibit greater body lengths-at-age than AFA populations. In addition, given the expectation that the biophysical conditions within an MPA would continue to better with prolonged protection, it was hypothesized that newer generations (more recent year classes) would exhibit greater body lengths-at-age than older generations (older year classes). Results are presented as Locally Estimated Scatterplot Smoothing (LOESS) regressions with a smoothing parameter of 0.95 and 95% confidence intervals (CIs). LOESS regression is a robust regression method that circumvents the shortcomings of 12
classical, parametric regression methods (i.e., normal distribution and independence between levels of predictor variables are rarely met by biological data). All statistics and figures were computed on the R Statistics Software v.
3.5.0. Packages required were: “car”, “userfriendlyscience” and “ggplot2”.
13
14
Figure 3. The left sagittal otolith of a three-year-old Ctenochaetus striatus. Blue dots represent first (1), second (2), and third
(3) annuli, with their respective incremental distances.
Table 1. Comparison of the sample sizes between the original stocks of fish used for otolith ageing and the pooled fish that
resulted from the Modified Fry Back-Calculation (MFBC) methodology.
Age Age Age Age Age Age Age Age Age Age Age Age Species Data Status 1 2 3 4 5 6 7 8 9 10 11 12
Original MPA 30 24 11 10 9 9 5 3 4 3 1 2 Sample Size AFA 31 25 28 21 20 18 5 1 1 6 2 1 Acanthurus nigrofuscus MFBC MPA 111 81 57 46 36 27 18 13 10 6 3 2 Sample
15 Size
AFA 159 128 103 75 54 34 16 11 10 9 3 1
Original MPA 54 23 12 5 13 8 6 3 2 5 4 2 Sample Size AFA 41 20 17 15 26 11 11 9 9 10 6 1 Ctenochaetus binotatus MFBC MPA 139 85 62 50 45 32 24 16 13 11 6 2 Sample Size AFA 176 135 115 98 83 57 46 35 26 17 7 1
Table 1. Continued.
Age Age Age Age Age Age Age Age Age Age Age Age Species Data Status 1 2 3 4 5 6 7 8 9 10 11 12
Original MPA 51 26 15 14 7 10 16 4 8 NA NA NA Sample Sizes AFA 39 57 57 42 16 15 13 3 2 NA NA NA Ctenochaetus striatus MFBC MPA 151 100 74 59 45 38 28 12 8 NA NA NA Sample Size AFA 244 205 148 91 49 33 18 5 2 NA NA NA
16
Original MPA 46 26 7 7 10 14 19 9 7 6 NA NA Sample Sizes AFA 39 22 17 13 25 31 13 13 9 6 NA NA Zebrasoma scopas MFBC MPA 151 105 79 72 65 55 41 22 13 6 NA NA Sample Size AFA 188 149 127 110 97 72 41 28 15 6 NA NA
3. Results
3.1. Acanthurus nigrofuscus
Significant disparities were found in the body lengths of MPA and AFA populations of A. nigrofuscus at five age cohorts (Two-Way ANOVA). No significant status effect was determined at ages 1 and 2 (F = 2.19, p = 0.140; F
= 0.21, p = 0.651, respectively) although an effect appeared at ages 3, 4, 5 and 6
(F = 6.55, p = 0.012; F = 9.16, p = 0.003; F = 7.65, p = 0.007; F = 9.65, p =
0.003, respectively), Figures 4 to 9. Significant disparity disappeared at ages 7,
8 and 9 (F = 2.08, p = 0.164; F = 2.68, p = 0.124; F = 4.50, p = 0.055, respectively) and reappeared at age 10 (F = 5.81, p = 0.039), Figures 10 to 13.
No significant status effect was detected at the latter age of 11 (F = 2.48, p =
0.256), as shown in Table 2.
Significant year effects were abundant throughout the observed life history of fish, from ages 1 through 8 (F = 53.32, p < 0.001; F = 81.04, p <
0.001; F = 54.28, p < 0.001; F = 31.09, p < 0.001; F = 32.23, p < 0.001; F =
22.16, p < 0.001; F = 10.64, p < 0.001; F = 6.91, p = 0.003, respectively),
Figures 4 to 11. No significant year effect was detected at the latter ages of 9,
10 and 11 (F = 3.11, p = 0.067; F = 3.27, p = 0.086; F = 4.82, p = 0.159, respectively), Figures 12 and 13. The determined year effects are observed as positive slopes in the LOESS regressions, which indicate newer generations are exhibiting greater body lengths than older generations.
Statistical analysis on age cohort 12 was bypassed because all observations belong to the same cohort. LOESS regression plotting was
17
17
additionally bypassed on age cohorts 11 and 12 due to insufficient temporal cohorts (< 3).
18
18
Table 2. Acanthurus nigrofuscus mean lengths (x̅ ) and standard errors (σx̅ ) in millimeters for each age cohort separated by
management status, along with Two-Way ANOVA F-values (F), degrees of freedom (DF) and p-values (p) for both
Status and Year effects. Significant values indicated in bold.
Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age 7 Age 8 Age 9 Age 10 Age 11 Age 12 MPA x̅ 38.55 49.57 56.90 65.36 71.46 76.58 76.53 79.47 83.27 86.84 87.60 93.00 σx̅ 1.36 1.71 1.87 2.06 2.11 2.90 2.86 2.84 2.41 2.37 1.31 4.00 AFA x̅ 36.14 48.46 57.89 64.50 70.96 73.10 70.34 70.41 74.54 81.01 83.37 78.00
19 σx̅ 1.33 1.36 1.45 1.60 2.00 2.40 3.96 3.56 3.01 3.18 8.73 NA
Difference Absolute 2.41 1.11 0.99 0.86 0.50 3.48 6.19 9.06 8.73 5.83 4.23 15.00 Percent 6.46 2.27 1.73 1.33 0.70 4.65 8.43 12.08 11.07 6.95 4.95 17.54 Status F 2.19 0.21 6.55 9.16 7.65 9.65 2.08 2.68 4.50 5.81 2.48 NA DF 1 1 1 1 1 1 1 1 1 1 1 NA p 0.140 0.651 0.012 0.003 0.007 0.003 0.164 0.124 0.055 0.039 0.256 NA Year F 53.32 81.04 54.28 31.09 32.23 22.16 10.64 6.91 3.11 3.27 4.82 NA DF 11 10 9 8 7 6 5 4 3 2 1 NA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.003 0.067 0.086 0.159 NA
20
Figure 4. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 1 cohorts between 2003 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
21
Figure 5. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 2 cohorts between 2004 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
22
Figure 6. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 3 cohorts between 2005 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
23
Figure 7. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 4 cohorts between 2006 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
24
Figure 8. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 5 cohorts between 2007 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
24
25
Figure 9. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 6 cohorts between 2008 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
25
26
Figure 10. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 7 cohorts between 2009 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
26
27
Figure 11. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 8 cohorts between 2010 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
27
28
Figure 12. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 9 cohorts between 2011 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
28
29
Figure 13. LOESS regressions of the Acanthurus nigrofuscus MPA and AFA age 10 cohorts between 2012 and 2014. Two-
Way ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects)
indicated via year numbers above the regressions.
29
3.2. Ctenochaetus binotatus
No significant disparities in the body lengths of MPA and AFA populations of C. binotatus were observed across the life history of fish, as shown in Table 3 (Two-Way ANOVA). No significant status effect was determined at ages 1 through 11 (F = 0.04, p = 0.841; F = 1.74, p = 0.189; F =
1.70, p = 0.194; F = 0.83, p = 0.363; F = 0.75, p = 0.363; F = 0.75, p = 0.389, F
= 0.48, p = 0.492, F = 0.12, p = 0.727, F = 1.90, p = 0.176; F = 1.97, p = 0.170;
F = 4.28, p = 0.050; F = 0.10, p = 0.756, respectively), Figures 14 to 23.
Significant year effects were abundant throughout the observed life history of fish, from ages 1 through 11 (F = 194.87, p < 0.001; F = 152.89, p <
0.001; F = 142.80, p < 0.001; F = 112.69, p < 0.001; F = 128.77, p < 0.001; F =
59.53, p < 0.001; F = 31.76, p < 0.001; F = 22.38, p < 0.001; F = 7.69, p =
0.001; F = 9.00, p = 0.001; F = 8.98, p = 0.015, respectively), Figures 14 to 23.
The determined year effects are observed as positive slopes in the LOESS regressions, which indicate newer generations are exhibiting greater body lengths than older generations.
Statistical analysis on age cohort 12 was bypassed because all observations belong to the same cohort. LOESS regression plotting was additionally bypassed on age cohorts 11 and 12 due to insufficient temporal cohorts (< 3).
30
30
Table 3. Ctenochaetus binotatus mean lengths (x̅ ) and standard errors (σx̅ ) in millimeters for each age cohort separated by
management status, along with Two-Way ANOVA F and p-values for both Status and Year effects. Significant
values indicated in bold.
Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age 7 Age 8 Age 9 Age 10 Age 11 Age 12 MPA x̅ 41.37 49.34 56.46 61.72 69.55 71.99 74.31 77.17 81.40 89.21 90.69 88.50 σx̅ 1.33 1.90 2.37 2.34 2.37 2.70 2.74 2.48 2.00 3.19 4.02 5.50 AFA x̅ 35.68 44.08 54.35 62.06 68.43 70.24 75.28 77.89 81.12 86.23 94.18 97.00
31 σx̅ 1.16 1.31 1.64 1.77 1.79 1.93 2.00 1.86 1.67 1.45 2.28 NA
Difference Absolute 5.69 5.25 2.12 0.33 1.12 1.75 0.96 0.72 0.28 2.98 3.49 8.50 Percent 14.78 11.24 3.82 0.54 1.62 2.47 1.29 0.93 0.34 3.39 3.77 9.16 Status F 0.04 1.74 1.70 0.83 0.75 0.48 0.12 1.90 1.97 4.28 0.10 NA DF 1 1 1 1 1 1 1 1 1 1 1 NA p 0.841 0.189 0.194 0.363 0.389 0.492 0.727 0.176 0.170 0.050 0.756 NA Year F 194.87 152.89 142.80 112.69 128.77 59.53 31.76 22.38 7.69 9.00 8.98 NA DF 11 10 9 8 7 6 5 4 3 2 1 NA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.001 0.015 NA
31
32
Figure 14. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 1 cohorts between 2003 and
2014. Two-Way ANOVA results are shown in the lower-left, with significant disparities between year classes (year
effects) indicated via year numbers above the regressions.
32
33
Figure 15. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 2 cohorts between 2004 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
33
34
Figure 16. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 3 cohorts between 2005 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
34
35
Figure 17. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 4 cohorts between 2006 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
35
36
Figure 18. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 5 cohorts between 2007 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
36
37
Figure 19. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 6 cohorts between 2008 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
37
38
Figure 20. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 7 cohorts between 2009 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
38
39
Figure 21. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 8 cohorts between 2010 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
39
40
Figure 22. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 9 cohorts between 2011 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
40
41
Figure 23. LOESS regressions of the Ctenochaetus binotatus MPA and AFA age 10 cohorts between 2012 and 2014. Two-
Way ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects)
indicated via year numbers above the regressions.
41
3.3. Ctenochaetus striatus
Significant disparities were found in the body lengths of MPA and AFA populations of C. striatus at two age cohorts (Two-Way ANOVA). No significant status effect was determined at ages 1 through 5 (F = 3.55, p =
0.060; F = 0.99, p = 0.320; F = 0.23, p = 0.634; F = 0.05, p = 0.831; F = 1.29, p
= 0.260, respectively) although an effect appeared at ages 6 and 7 (F = 4.93, p =
0.030; F = 6.22, p = 0.017, respectively) Figures 24 to 30. No significant status effect was detected at the latter age of 8 (F = 1.36, p = 0.265), as shown in
Table 4.
Significant year effects were abundant throughout the observed life history of fish, from ages 1 through 7 (F = 77.58, p < 0.001; F = 98.25, p <
0.001; F = 70.00, p < 0.001; F = 58.85, p < 0.001; F = 30.24, p < 0.001; F =
24.72, p < 0.001; F = 17.10, p < 0.001, respectively), Figures 24 to 30. No significant year effect was detected at the latter age 8 (F = 0.74, p = 0.404),
Table 4. The determined year effects are observed as positive slopes in the
LOESS regressions, which indicate newer generations are exhibiting greater body lengths than older generations.
Statistical analysis on age cohort 9 was bypassed because all observations belong to the same cohort. LOESS regression plotting was additionally bypassed on age cohorts 8 and 9 due to insufficient temporal cohorts (< 3).
42
42
Table 4. Ctenochaetus striatus mean lengths (x̅ ) and standard errors (σx̅ ) in millimeters for each age cohort separated by
management status, along with Two-Way ANOVA F and p-values for both Status and Year effects. Significant
values indicated in bold.
Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age 7 Age 8 Age 9 MPA x̅ 54.55 69.26 79.67 90.67 98.94 111.31 120.53 121.98 130.63 σx̅ 1.20 1.94 2.30 2.05 1.97 2.11 2.29 2.19 2.47 AFA x̅ 51.16 75.73 90.31 99.75 102.89 110.87 115.89 117.22 129.50
43 σx̅ 0.79 1.34 1.50 1.90 2.19 2.53 3.10 5.15 8.50 Difference Absolute 3.40 6.47 10.63 9.07 3.95 0.44 4.64 4.77 1.13 Percent 6.43 8.92 12.51 9.53 3.91 0.40 3.93 3.99 0.86 Status F 3.55 0.99 0.23 0.05 1.29 4.93 6.22 1.36 NA DF 1 1 1 1 1 1 1 1 NA p 0.060 0.320 0.634 0.831 0.260 0.030 0.017 0.265 NA Year F 77.58 98.25 70.00 58.85 30.24 24.72 17.10 0.74 NA DF 8 7 6 5 4 3 2 1 NA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.404 NA
43
44
Figure 24. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 1 cohorts between 2006 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
44
45
Figure 25. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 2 cohorts between 2007 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
45
46
Figure 26. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 3 cohorts between 2008 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
46
47
Figure 27. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 4 cohorts between 2009 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
47
48
Figure 28. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 5 cohorts between 2010 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
48
49
Figure 29. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 6 cohorts between 2011 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
49
50
Figure 30. LOESS regressions of the Ctenochaetus striatus MPA and AFA age 7 cohorts between 2012 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
50
3.4. Zebrasoma scopas
No significant disparities in the body lengths of MPA and AFA populations of Z. scopas were observed across the observed life history of fish, as shown in Table 5 (Two-Way ANOVA). No significant status effect was determined at ages 1 through 9 (F = 1.31, p = 0.253; F = 1.94, p = 0.165; F =
1.21, p = 0.272; F = 0.25, p = 0.616 F = 0.09, p = 0.766; F = 0.06, p = 0.813, F
= 0.41, p = 0.524, F = 0.12, p = 0.733, F = 0.06, p = 0.814, respectively),
Figures 31 to 38.
Significant year effects were abundant throughout the observed life history of fish, from ages 1 through 8 (F = 121.04, p < 0.001; F = 92.20, p <
0.001; F = 66.66, p < 0.001; F = 61.34, p < 0.001; F = 56.49, p < 0.001; F =
31.81, p < 0.001; F = 13.29, p < 0.001, F = 8.67, p = 0.001, respectively),
Figures 31 to 38. No significant year effect was detected at the latter age of 9 (F
= 3.81, p = 0.063), as shown in Table 5. The determined year effects are observed as positive slopes in the LOESS regressions, which indicate newer generations are exhibiting greater body lengths than older generations.
Statistical analysis on age cohort 10 was bypassed because all observations belong to the same cohort. LOESS regression plotting was additionally bypassed on age cohorts 9 and 10 due to insufficient temporal cohorts (< 3).
51
51
Table 5. Zebrasoma scopas mean lengths (x̅ ) and standard errors (σx̅ ) in millimeters for each age cohort separated by
management status, along with Two-Way ANOVA F and p-values for both Status and Year effects. Significant
values indicated in bold.
Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age 7 Age 8 Age 9 Age 10 MPA x̅ 38.26 50.17 57.51 68.74 78.13 85.80 91.88 96.50 101.37 107.00 σx̅ 0.93 1.43 1.36 1.56 1.58 1.61 1.68 1.97 2.17 3.34 AFA x̅ 37.08 50.82 62.45 72.67 82.32 88.83 91.78 96.29 101.51 108.83
52 σx̅ 0.79 1.19 1.36 1.42 1.50 1.61 1.80 1.82 2.21 1.64
Difference Absolute 1.18 0.65 4.94 3.93 4.20 3.03 0.11 0.21 0.14 1.83 Percent 3.12 1.28 8.24 5.56 5.23 3.47 0.12 0.21 0.14 1.70 Status F 1.31 1.94 1.21 0.25 0.09 0.06 0.41 0.12 0.06 NA DF 1 1 1 1 1 1 1 1 1 NA p 0.253 0.165 0.272 0.616 0.766 0.813 0.524 0.733 0.814 NA Year F 121.04 92.20 66.66 61.34 56.49 31.81 13.29 8.67 3.81 NA DF 9 8 7 6 5 4 3 2 1 NA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.063 NA
52
53
Figure 31. LOESS regressions of the Zebrasoma scopas MPA and AFA age 1 cohorts between 2005 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
53
54
Figure 32. LOESS regressions of the Zebrasoma scopas MPA and AFA age 2 cohorts between 2006 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
54
55
Figure 33. LOESS regressions of the Zebrasoma scopas MPA and AFA age 3 cohorts between 2007 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
55
56
Figure 34. LOESS regressions of the Zebrasoma scopas MPA and AFA age 4 cohorts between 2008 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
56
57
Figure 35. LOESS regressions of the Zebrasoma scopas MPA and AFA age 5 cohorts between 2009 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
57
58
Figure 36. LOESS regressions of the Zebrasoma scopas MPA and AFA age 6 cohorts between 2010 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
58
59
Figure 37. LOESS regressions of the Zebrasoma scopas MPA and AFA age 7 cohorts between 2011 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
59
60
Figure 38. LOESS regressions of the Zebrasoma scopas MPA and AFA age 8 cohorts between 2012 and 2014. Two-Way
ANOVA results are shown in the lower-left, with significant disparities between year classes (year effects) indicated
via year numbers above the regressions.
60
4. Discussion
As our reliance on MPAs as a primary fisheries management tool increases, it becomes critical that we develop a more thorough understanding of the spatiotemporal variation in the impacts of MPAs on the dynamics of exploited fish populations. For decades, assessments of the efficacy of MPAs in meeting their management goals have relied on visual census surveys of fish abundance and diversity as well as habitat quality within MPAs and AFAs. The major findings of such studies have typically identified increased biomass, density, biodiversity, and body sizes within protected areas (Lester et al. 2009) as well as beneficial shifts in populational demography (Fidler et al. 2017), with benefits being extrapolated onto surrounding fished areas via processes of connectivity (Russ and Alcala 2011). However, strictly static investigations are inadequate at assessing the long-term responses of populations to MPA establishment.
This study introduces, for the first time in the field of MPA assessment, the conventional fisheries technique known as a “back-calculation” to determine spatiotemporal variations in length-at-age of acanthurid
(Acanthuridae, Teleostei) fishes between MPAs and AFAs. The use of back- calculated length-at-age is invaluable as it overcomes the restrictions placed on repeated sampling within MPAs whilst drastically increasing sample size per cohort, thereby allowing an in-depth examination of the populations’ longitudinal phenotypic responses to protection. As previously stated, the phenotypic characteristic examined was body length, a critical determinant of
61
61
fitness in fishes (Birkeland and Dayton 2005) that, as a result, has been purported to be an outstanding metric of the contribution of MPAs to the sustainability of fisheries resources.
Altogether, five major findings are drawn from the results of this study.
First, newly settled A. nigrofuscus were phenotypically similar at ages one and two. Second, A. nigrofuscus transformed into significantly different length-at- age morphs between ages three and six, in which MPA populations were predominantly larger than AFA populations. Third, this phenotypic divergence disappeared towards the latter life history stages. Fourth, the observed life histories of the other acanthurid species were dominated by spatial phenotypic homogeneity. Fifth, all species displayed prolonged temporal growth.
In discussion of our spatial analyses, it is conceivable that the patterns of convergence (similarity) and divergence (disparity) in length-at-age reflect the complexity of the main and interactive effects of factors that underlie the impacts that MPAs have on the biology, physiology, and ecology of exploited fish populations on coral reefs. That being said, the general dominance of phenotypic homogeneity between MPAs and AFAs promotes two plausible hypotheses: (1) protection is having little discernible effect on phenotype or (2) the effects produced by protection are not apparent through static spatial analyses alone. The observed abundance of significant year effects, however, inherently revokes the first hypothesis. On the other hand, the inadequacy of static spatial analyses at assessing MPAs has already been confirmed.
62
62
Phenotypic homogeneities during the early life history stages are typically identified as products of larval connectivity (Russ 2002; Harrison et al.
2012), a fitting explanation given that all four acanthurid species investigated here are broadcast spawners (Robertson 1983). This explanatory hypothesis is further supported alongside the widely accepted notion that coral reef connectivity is primarily established and maintained through pelagic larval dispersal (Jones et al. 2009, Planes et al. 2009; Christie et al. 2010). Given the expectation that elevated rates of larval connectivity would reduce site-specific self-recruitment and subsequently promote genetic panmixia between connected reefs, all subsequent instances of phenotypic divergence would be best described as products of post-settlement processes including phenotypic plasticity, the ability of a single genotype to produce various phenotypes in response to differing extrinsic factors (Price et al. 2003; Fusco and Minelli
2010). For example, differential exposure to size-selective fishing efforts and, as a result, the differential density-dependent scenarios produced between
MPAs and AFAs suggest that large-bodied individuals inhabiting protected areas will experience heightened survival probabilities whilst similar sized individuals inhabiting AFAs will experience heightened mortality probabilities.
The likelihood of this dichotomy is emphasized for species of the Acanthuridae family, whose characteristically fast growth rates place them well within the target range of fishermen from an early age (Choat and Axe 1996). In fact, this explanatory hypothesis is further supported by the observed significant status
63
63
effects that almost exclusively determined the presence of greater body sizes within MPAs compared to AFAs (Tables 2 and 4).
On the other hand, post-settlement connectivity via spillover has long been considered insufficient to maintain populational connectivity due to the characteristic site fidelity of coral resident fish species. As a result, whilst early phenotypic homogeneity can be expected due to larval dispersal, a lack of post- settlement connectivity would promote a gradual upsurge of status effects.
Whilst this was the case for the mid-life history of A. nigrofuscus, it certainly was not the case for C. binotatus, C. striatus and Z. scopas. A lack of phenotypic divergence at the latter life history stages could be a result of reduced sample sizes, but an investigation into these atypically prolonged phenotypic homogeneities identified another two potential causes: 1) insufficient threshold distance used to separate MPA and AFA sampling sites relative to the species’ home ranges and 2) insufficient MPA sizes relative to the species’ home ranges. A comparison between Z. scopas’, C. binotatus’ and
C. striatus’ home ranges (approximately 1.0km2, 0.3km2 and 0.3km2, respectively) and the minimum threshold distance of 300 meters promotes the plausibility that the three species’ post-settlement patterns of movement overlapped between sampling sites. The likelihood of this explanatory hypothesis is emphasized for Z. scopas, whose home range roughly equaled the mean MPA size of 1.14km2. Ultimately, such a situation would enable
“protected” and “fished” populations to simultaneously benefit from the
64
64
advantageous biophysical conditions produced by protection, thereby eliminating any expectation of phenotypic divergence.
Whilst these spatial analyses have provided insight into the relative health of acanthurid populations between MPAs and AFAs, our understanding of the long-term populational effects that MPA-establishment has had will remain largely obscured until our static vision transcends into the temporal dimension. This was achieved via the successful application of the MFBC model, which procured over a decade’s worth of valuable length-at-age data. A. nigrofuscus, C. binotatus, C. striatus and Z. scopas all exhibited immediately significant year effects that were maintained until the latter ages of eight, eleven, seven and eight, respectively. These significant year effects were directionally positive, observed as positive slopes in the LOESS regressions, thereby indicating that, within the significant age cohorts, newer generations were significantly larger than older generations. The loss of significant year effects at the latter life history stages is simply believed to be a result of reduced sample sizes.
Conveniently, a discussion of the hypothesized catalytic mechanisms behind these year effects resumes our discussion of status effects, in which the observed phenotypic divergences were best described as products of phenotypic plasticity and the differing directional extrinsic pressures experienced between
MPAs and AFAs. However, if such directional pressures were maintained across generations, then phenotypic shifts that were previously instigated through individual plasticity could become genetically engrained through
65
65
genotypic adaptations (Dingemanse and Wolf 2013). Whilst definitive answers undoubtedly require genetic analyses, the multigenerational patterns of growth and hinted predispositions manifested within the observed year effects are highly indicative of such genotypic adaptations occurring at a fine scale.
The magnitude of these observed restorations would be best described with a comparison to historic body size distributions; however, a lack of such records precludes our ability to perform such examinations. That being said, a superficial alternative can be completed via a comparison of maximum observed body lengths. The maximum observed total body lengths for A. nigrofuscus, C. binotatus, C. striatus and Z. scopas were identified in the 1990s as being 21.0cm, 22.0cm, 26.0cm and 20.0cm, respectively (Sommer et al.
1996). Whilst the typical long-term goals of MPAs have been to restore to near pristine biological conditions, which would require reversions that extend well beyond the 1990s, these maximum body size records provide a good baseline from which we can judge regional progress. In comparison to the above- mentioned records, the maximum observed total lengths obtained during this sampling excursion were 15.7cm, 16.5cm, 21.9cm and 15.9cm for A. nigrofuscus, C. binotatus, C. striatus and Z. scopas, respectively. These discrepancies show that restoration is far from complete and, given the per annum rates of phenotypic restoration calculated in this investigation, estimates suggest that A. nigrofuscus, C. binotatus, C. striatus and Z. scopas will only achieve complete restoration to the mean phenotype of the 1990s in another
10.79, 8.35, 6.37 and 7.78 years, respectively.
66
66
That being said, our spatiotemporal findings suggest that the five sampled MPAs are operating efficiently and gradually progressing towards this long-term goal, with the results being best described as follows: individuals inhabiting MPAs are likely to experience phenotypic enlargement, promoted through (1) an elimination of size-selective fishing efforts, (2) a resumption of ecological forces that naturally champion larger body sizes and (3) elevated biomass, which consequently increases the availability of food items (Edgar et al. 2014, McClanahan et al. 2007). Oppositely, individuals inhabiting AFAs are likely to experience phenotypic reductions due to (1) explicit exposure to directional fishing pressures and (2) fishing-induced reductions in biomass and food availability (Uusi-Heikkila et al. 2008). These polarizations in phenotype are manifested as status effects, but ultimately lead to drastic shifts in the reproductive potentials of these populations. Whilst fished populations will have to cope with depleted energy budgets and the physiological constraints of reduced body sizes, which are associated with reduced egg production, egg size, overall fecundity, size-at-hatch, feeding rate, and viability, protected populations will experience the opposite (Bobko and Berkeley 2004; Berkeley et al. 2004; Birkeland and Dayton 2005; Haugen and Vollestad 2001; Walsh et al. 2005). As a result, the observed polarization in phenotype transforms into a polarization in fitness, which will inherently promote the gradual predominance of genotypes from protected populations, thus producing the observed year effects. However, if home ranges were to overlap between MPAs and AFAs, as was proposed for the three acanthurid species that displayed prolonged
67
67
phenotypic homogeneity, then both MPA and AFA populations would simultaneously benefit from the biophysical conditions produced by protection and concurrently experience genotypic adaptations for phenotypic enlargement.
In summary, a review of these findings determined that the sampled
MPAs are making gradual and significant progress towards their stated conservation goals, ultimately reversing FIE towards historic body-size distributions. That being said, these interpretations required inclusion of a temporal analysis and would have otherwise been impossible to construct if solely presented with an independent spatial analysis. For example, a lack of phenotypic divergence between MPAs and AFAs has historically been interpreted as a failure of the MPA to instigate phenotypic enlargements. If this perspective were implemented here, a predominant lack of status effect across the life histories of C. binotatus, C. striatus and Z. scopas would suggest that the sampled MPAs are failing, which evidently is not the case. As a result, investigators should refrain from using the results of independent static spatial analyses as a proxy of the biophysical performance of MPAs. If such assessments are considered, investigators should reserve expectations of phenotypic divergence between MPAs and AFAs until extrinsic and intrinsic factors associated with the onset of status effects have been examined for that species-specific case study. Furthermore, the observed variation in patterns of phenotypic convergence and divergence between MPAs and AFAs across the four acanthurid species emphasize a lack of intra-familial generality in phenotypic response to protection. In fact, variable trends in the occurrence of
68
68
status effects within the same genus (Ctenochaetus), suggest that these phenotypic responses are likely to be species-specific. As a result, assessments should not make assumptions that the phenotypic responses of a single or select few species are representative of all other species contained with the protected area.
In light of the limitations discussed above, a review of limitations specific to this study’s novel methodology is undoubtedly warranted. First and foremost, is the question of whether the observed effects are historically accurate or mere artifacts of the back-calculation model. This critique is heightened when the output of a back-calculation is visualized via LOESS regressions with a high span parameter, which determines the degree of smoothing. High span values reduce finer-scale temporal fluctuations and, as a result, operate effectively in investigations into broader and larger-scale trends, such as the phenotypic shifts examined in this study. Nonetheless, heightened smoothing will oftentimes instigate skepticism among viewers due to a lack of finer-scale natural variation. It is vital that critics distinguish between their dislike for the method by which these models’ empirical outputs were visualized and their supposed dislike for the actual empirical outputs of the back-calculation model. The earlier argument can be easily resolved by decreasing the span value or by proceeding with an alternative method of representation (i.e. linear model, generalized linear model, etc.). The latter argument, as to whether the observed effects are simply artifacts of the back- calculation model, requires a more comprehensive review of the model’s
69
69
validation. Wilson et al. (2009) used tagging and longitudinal in-aquaria growth experiments on Gobids to validate five modern and prominently used back- calculation models, all of which produced moderately biased outputs except for the MFBC model. The MFBC model exhibited an incredible robustness in the procurement of accurate and precise length-at-age data despite exposure to differing intrinsic and extrinsic factors that affected somatic growth.
Furthermore, critiques that the proportional relationship between the growth of an individual’s body and otolith may fluctuate throughout the life history and consequently render back-calculations ineffective have been debunked following validation from Vigliola et al. (2000), who confirmed the MFBC model’s capacity to accommodate such shifts in proportionality. Altogether, it was uniformly determined that the MFBC model operated as the most efficient and conservative model for the back-calculation of body lengths from annuli distances, hence our selection of the MFBC model for this investigation.
In conclusion, this study was the first to implement a back-calculation model into the field of MPA assessment. The results presented here confirm the success of this methodology, promoting the MFBC model as a bio- mathematical tool capable of accurately reconstructing the size-related life histories of coral reef fishes and thus conferring a novel solution to the logistic and legal restrictions placed on repeated sampling within protected areas.
Analyses of the longitudinal data procured via the application of the back- calculation model determined that the sampled MPAs are gradually restoring historical body sizes in exploited regions, both within and outside of their
70
70
borders, thus ameliorating any uncertainty on the capacity of these MPAs to achieve their species-specific management and conservation goals. That being said, the intra-familial variability in these patterns of phenotypic convergence and divergence suggests that phenotypic responses are likely species-specific.
As a result, assessments should not make assumptions that the phenotypic responses of a single or select few species are representative of all other species contained with a protected area. Furthermore, this investigation perfectly examples the inadequacy of static spatial analyses when used to assess the performance of dynamic tools. Whilst the general lack of phenotypic disparity between MPAs and AFAs in C. binotatus, C. striatus and Z. scopas would have historically been interpreted as a failure of the MPAs to instigate phenotypic enlargements, inclusion of a temporal analysis completely reversed this interpretation. As a result, investigators should refrain from using the results of independent static spatial analyses as a proxy of the biophysical performance of
MPAs. Lastly, although these findings offer critical insight and a comprehensive provision of MPA functionality within the Northwest Philippine region of the Coral Triangle, the ubiquity of the observed phenotypic responses across other taxa and regions is unknown. Whilst further spatiotemporal analyses are required across other regions and taxa of differential exploitation and life histories, this investigation adds to the growing body of evidence that
MPAs can provide positive conservation and fishery benefits. Continued studies identifying the positive outcomes of MPAs will be critical in garnering support for the establishment of MPAs in the coming decades.
71
71
References
Abesamis, R.A., & Russ, G.R. (2005). Density-dependent spillover from a
marine reserve: long term evidence. Ecological Applications, 15(5),
1798-1812.
Babcock, R.C., Shears, N.T., Alcala, A.C., Barrett, N.S., Edgar, G.J., Lafferty,
K.D., McClanahan, T.R., & Russ, G.R. (2010). Decadal trends in marine
reserves reveal differential rates of change in direct and indirect effects.
Proceedings of the National Academy of Sciences, 107(43), 18256-
18261.
Barneche, D.R., Robertson, D.R., White, C.R., & Marshall, D.J. (2018). Fish
reproductive-energy output increases disproportionately with body size.
Science, 360(6389), 642-645.
Bell, J.D., Watson, R.A., & Ye, Y. (2017). Global fishing capacity and fishing
effort from 1950 to 2012. Fish and Fisheries, 18(3), 489-505.
Bellwood, D.R., Hoey, A.S., & Hughes, T.P. (2012). Human activity selectively
impacts the ecosystem roles of parrotfishes on coral reefs. Proceedings
of the Royal Society of London B, 279(1733), 1621-1629.
Berkeley, S.A., Chapman, C., & Sogard, S.M. (2004). Maternal age as a
determinant of larval growth and survival in a marine fish, Sebastes
melanops. Ecology, 85(5), 1258-1264.
Birkeland, C., & Dayton, P.K. (2005). The importance in fishery management
of leaving the big ones. Trends in Ecology & Evolution, 20(7), 356-358.
72
72
Bobko, S.J., & Berkeley, S.A. (2004). Maturity, ovarian cycle, fecundity, and
age-specific parturition of black rockfish (Sebastes melanops). Fishery
Bulletin, 102(3), 418-429.
Braudel, F. (2001). Memory and the Mediterranean. New York, NY, USA:
Knopf.
Calder, W.A.I. (1985). Size and metabolism in natural systems. Ecosystem
Theory for Biological Oceanography, 65-75.
Casselman, J.M. (1990). Growth and relative size of calcified structures of fish.
Transactions of the American Fisheries Society, 119(4), 673-688.
Choat, J.H., & Axe, L.M. (1996). Growth and longevity in acanthurid fishes; an
analysis of otolith increments. Marine Ecology Progress Series, 134, 15-
26.
Christie, M.R., Tissot, B.N., Albins, M.A., Beets, J.P., Jia, Y., Ortiz, D.M.,
Thompson, S.E., & Hixon, M.A. (2010) Larval connectivity in an
effective network of marine protected areas. PLoS ONE, 5(12): e15715.
Claudet, J. (2011). Marine protected areas: A multidisciplinary approach.
Cambridge, UK: Cambridge University Press.
Clifton, J., Unsworth, R.K.F., & Smith, D.J. (2010). Marine research and
conservation the Coral Triangle: the Wakatobi national park. Retrieved
from:
https://www.researchgate.net/publication/259495449_Marine_Research
_and_Conservation_in_the_Coral_Triangle_the_Wakatobi_National_Pa
rk
73
73
Cullen, L.C. (2007). Marine resource dependence, resource use patterns and
identification of economic performance criteria within a small island
community: Kaledupa, Indonesia. University of Essex, England.
Dingemanse, N.J., & Wolf, M. (2013). Between-individual differences in
behavioural plasticity within populations: causes and consequences.
Animal Behaviour, 85(5), 1031-1039.
Donaldson, T.J. (2007). Extinction susceptibility of reef fishes in spawning
aggregations. Gulf and Caribbean Fisheries Institute, 59, 379-386.
Dutton, I.M. (2005). If only fish could vote: the enduring challenges of coastal
and marine resources management in post-reformasi Indonesia. In
Resosudarmo, B.P. (Eds.), The politics and economics of Indonesia’s
natural resources (162-178). Singapore: ISEAS Publications.
Edeline, R., Carlson, S.M., Stige, L.C., Winfield, I.J., Fletcher, J.M., James,
J.B., Haugen, T.O., Vollestad, L.A., & Stenseth, N.C. (2007). Trait
changes in a harvested population are driven by a dynamic tug-of-war
between natural and harvest selection. Proceedings of the National
Academy of Sciences, 104(40), 15799-15804.
Edgar, G.J., Russ, G.R., & Babcock, R.C. (2007). Marine protected areas.
Marine Ecology, 534-565.
74
74
Edgar, G.J., Stuart-Smith, R.D., Willis, T.J., Kininmonth, S., Baker, S.C.,
Banks, S., Barrett, N.S., Becerro, M.A., Bernard, A.T.F., Berkhout, J.,
Buxton, C.D., Campbell, S.J., Cooper, A.T., Davey, M., Edgar, S.C.,
Forsterra, G., Galvan, D.E., Irigoyen, A.J., Kushner, D.J., Moura, R.,
Parnell, P.E., Shears, N.T., Soler, G., Strain, E.M.A., & Thomson, R.J.
(2014). Global conservation outcomes depend on marine protected areas
with five key features. Nature, 506, 216-220.
FAO. (2016). The state of world fisheries and aquaculture, 2016. Rome, Italy:
FAO United Nations.
Fidler, R.Y., Carroll, J., Rynerson, K.W., Matthews, D.F., & Turingan, R.G.
(2018). Coral reef fishes exhibit beneficial phenotypes inside marine
protected areas. PLoS ONE, 13(2): e0193426.
Fidler, R.Y., Turingan, R.G., White, A.T., & Alava, M.N.R. (2017). Reef-wide
beneficial shifts in fish population structure following establishment of
marine protected areas in Philippine coral reefs. Marine Ecology
Progress Series, 570, 187-202.
Francis, R. (1990). Back-calculation of fish length: a critical review. Journal of
Fish Biology, 36, 883-902.
Fusco, G., & Minelli, A. (2010). Phenotypic plasticity in development and
evolution: facts and concepts. Philosophical Transactions of the Royal
Society B, 365, 547-556.
75
75
Gell, F.R., & Roberts, C.M. (2003). Benefits beyond boundaries: the fishery
effects of marine reserves. Trends in Ecology & Evolution, 18(9), 448-
455.
Green, A.L., & Bellwood, D.R. (2009). Monitoring functional groups of
herbivorous reef fishes as indicators of coral reef resilience. Gland,
Switzerland: IUCN.
Green, A.L., Maypa, A.P., Almany, G.R., Rhodes, K.L., Weeks, R., Abesamis,
R.A., Gleason, M.G., Mumby, P.J., & White, A.T. (2015). Larval
dispersal and movement patterns of coral reef fishes, and implications
for marine reserve network design. Biological Reviews, 90, 1215-1247.
Gulland, J.A. (1974). The management of marine fisheries. Seattle, WA, USA:
University of Washington Press.
Harrison, H.B., Williamson, D.H., Evans, R.D., Almany, G.R., Thorrold, S.R.,
Russ, G.R., Feldheim, K.A., van Herweden, L., Planes, S., Srinivasan,
M., Berumen, M.L., & Jones, G.P. (2012). Larval export from marine
reserves and the recruitment benefit for fish and fisheries. Current
Biology, 22, 1023-1028.
Haugen, T.O., & Vollestad, L.A. (2001). A century of life-history evolution in
grayling. Genetica, 112(1), 475-491.
Heino, M., & Dieckmann, U. (2009). Fisheries-induced evolution. In:
Encyclopedia of Life Sciences (ELS). John Wiley & Sons, Ltd:
Chinchester, UK. p. 1-7.
76
76
Heino, M., & Godo, O.R. (2002). Fisheries-induced selection pressures in the
context of sustainable fisheries. Bulletin of Marine Sciences, 70(2), 639-
656.
Hickling, C.F. (1933). The natural history of the hake. Part IV. Age-
determination and the growth rate. Fish. Invest. Min. Agric. Food Fish,
2(13), 120.
Jones, G.P., Almany, G.R., Sale, P.F., Steneck, R.S., van Oppen, M.J.H., &
Willis, B.L. (2009). Larval retention and connectivity among
populations of corals and reef fishes: history, advances and challenges.
Coral Reefs, 28(2), 307-325.
Kelleher, G., & Kenchington, R. (1991). Guidelines for establishing marine
protected areas. Gland, Switzerland: IUCN.
Law, R. (2007). Fisheries-induced evolution: Present status and future
directions. Marine Ecology Progress Series, 335, 271-277.
Lea, E. (1910). Contributions to the methodics in herring-investigations. ICES
Journal of Marine Science, 1(53), 7-33.
Lester, S.E., Halpern, B.S., Grorud-Colvert, K., Lubchenco, J., Ruttenberg, B.I.,
Gaines, S.D., Airame, S., & Warner, R.R. (2009). Biological effects
within no-take marine reserves: a global synthesis. Marine Ecology
Progress Series, 384, 33-46.
Lewin, W.C., Arlinghaus, R., & Mehner, T. (2006). Documented and potential
biological impacts of recreational fishing: insights for management and
conservation. Reviews in Fisheries Science, 14(4), 305-367.
77
77
Martino, J.C., Fowler, A.J., Doubleday, Z.A., Grammer, G.L., & Gillanders,
B.M. (2019). Using otolith chronologies to understand long-term trends
and extrinsic drivers of growth in fisheries. Ecosphere, 10(1), e02553.
McClanahan, T.R., Graham, N.A.J., Calnan, J.M., MacNeil, M.A. (2007).
Toward pristine biomass: reef fish recovery in coral reef marine
protected areas in Kenya. Ecological Applications, 17(4), 1055-1067.
Moberg, F., & Folke, C. (1999). Ecological goods and services of coral reef
ecosystems. Ecological Economics, 29(2), 215-233.
Myers, R.A., Hutchings, J.A., & Barrowman, N.J. (1997). Why do fish stocks
collapse? The example of cod in Atlantic Canada. Ecological
Applications, 7(1), 91-106.
Nash, K.L., Graham, N.A.J., & Bellwood, D.R. (2013). Fish foraging patterns,
vulnerability to fishing, and implications for the management of
ecosystem function across scales. Ecological Applications, 23(7), 1632-
1644.
Planes, S., Jones, G.P., & Thorrold, S.R. (2009). Larval dispersal connects fish
populations in a network of marine protected areas. Proceedings of the
National Academy of Sciences, 106(14), 5693-5697.
Price, T.D., Qvarnstrom, A., & Irwin, D.E. (2003). The role of phenotypic
plasticity in driving genetic evolution. Proceedings of the Royal Society
of London B, 270, 1433-1440.
78
78
Reibish, J. (1899). Uber die einzahl bei Pleuronectes platessa und die
alterbestimmung dieser form aus den otolithen. Wissenschaftliche
Meeresuntersuchung, 4, 233-248
Reiss, M.J. (1989). The allometry of growth and reproduction. Cambridge, UK:
Cambridge University Press.
Robertson, D.R. (1983). On the spawning behavior and spawning cycles of
eight surgeonfishes (Acanthuridae) from the Indo-Pacific.
Environmental Biology of Fishes, 9(3/4), 193-223.
Rochet, M.J. (2000). A comparative approach to life-history strategies and
tactics among four orders of teleost fish. ICES Journal of Marine
Science, 57(2), 228-239.
Russ, G.R. (2002). Yet another review of marine reserves as reef fishery
management tools. In Sale, P.F. (Eds.), Coral reef fishes: dynamics and
diversity in a complex ecosystem (421-444). Burlington, MA, USA:
Elsevier.
Russ, G.R., & Alcala, A.C. (2011). Enhanced biodiversity beyond marine
reserve boundaries: the cup spillith over. Ecological Applications, 21(1),
241-250.
Secor, D.H., Trice, T.M., & Hornick, H.T. (1995). Validation of otolith-based
ageing and a comparison of otolith and scale-based ageing in mark-
recaptured Chesapeake Bay striped bass, Morone saxatilis. Fishery
Bulletin 93(1), 186-190.
79
79
Sommer, C., Schneider, W., & Poutiers, J.M. (1996). FAO species identification
field guider for fishery purposes. The living marine resources of
Somalia. Rome, Italy: FAO.
Swain, D.P., Sinclair, A.F., & Hanson, M.J. (2007). Evolutionary response to
size-selective mortality in an exploited fish population. Proceedings of
the Royal Society of London B, 274(1613), 1015-1022.
Taylor, B.M., & McIlwain, J.L. (2010). Beyond abundance and biomass: effects
of marine protected areas on demography of a highly exploited reef fish.
Marine Ecology Progress Series, 411, 243-258.
Trexler, J.C., & Travis, J. (2000). Can marine protected areas restore and
conserve stock attributes of reef fishes? Bulletin of Marine Science,
66(3), 853-873.
Trippel, E.A. (1995). Age at maturity as a stress indicator in fisheries.
BioScience, 45(11), 759-771.
Uusi-Heikkila, S., Wolter, C., Klefoth, T., & Arlinghaus, R. (2008). A
behavioral perspective on fishing-induced evolution. Trends in Ecology
& Evolution, 23(8), 419-421.
Vigliola, L., Harmelin-Vivien, M., & Meekan, M.G. (2000). Comparison of
techniques of back-calculation of growth and settlement marks from the
otoliths of three species of Diplodus from the Mediterranean Sea.
Canadian Journal of Fisheries and Aquatic Sciences, 57(6), 1291-1299.
80
80
Vigliola, L., & Meekan, M.G. (2009). The back-calculation of fish growth from
otoliths. In: Green, B.S., Mapstone, B.D., Carlos, G., & Begg, G.A.
(Eds.), Tropical fish otoliths: information for assessment, management
and ecology (174-211). Dordrecht, Netherlands: Springer.
Walsh, M.R., Munch, S.B., Chiba, S., & Conover, D.O. (2005). Maladaptive
changes in multiple traits caused by fishing: impediments to population
recovery. Ecology Letters, 9(2), 142-148.
Wilson, S., & Bellwood, D.R. (1997). Cryptic dietary component of territorial
damselfishes (Pomacentridae, Laboroidei). Marine Ecology Progress
Series, 153, 299-310.
Wilson, J.A., Vigliola, L., & Meekan, M.G. (2009). The back-calculation of
size and growth from otoliths: validation and comparison of models at
an individual level. Journal of Experimental Marine Biology and
Ecology, 368(1), 9-21.
Worm, B., Hilborn, R., Baum, J.K., Branch, T.A., Collie, J.S., Costello, C.,
Fogarty, M.J., Fulton, E.A., Hutchings, J.A., Jennings, S., Jensen, O.P.,
Lotze, H.K., Mace, P.M., McClanahan, T.R., Minto, C., Palumbi, S.R.,
Parma, A.M., Ricard, D., Rosenberg, A.A., Watson, R., & Zeller, D.
(2009). Rebuilding global fisheries. Science, 325(5940), 578-585.
81
81