Hugulay Albuquerque Maia Padrões Espaciais E

HUGULAY ALBUQUERQUE MAIA

PADRÕES ESPACIAIS E TEMPORAIS DAS COMUNIDADES RECIFAIS NAS ILHAS DE SÃO TOMÉ E PRÍNCIPE, GOLFO DE GUINÉ, ÁFRICA

Tese apresentada ao Programa de Pós- Graduação em Ecologia, da Universidade Federal de Santa Catarina, como parte dos requisitos para a obtenção do título de Doutor em Ecologia.

Área de concentração: Ecossistemas Marinhos

Orientador: Dr. Sergio R. Floeter

Florianópolis/SC/Brasil/ 2018 ii

iii

“Sometimes it is the people no one can imagine anything of who do the things no one can imagine”

Alan Turing vi

vii

Às três mulheres mais importantes na minha vida, dedico. Magda Vany Lourenço Dias da Silva Zulmira Pinto Vaz da Costa Jesus Bonfim Helena de Assunção Pinto Vaz da Costa viii

AGRADECIMENTOS

Gostaria de deixar expresso os meus sinceros agradecimentos a todos os que condicionalmente e incondicionalmente contribuíram para esta tese. Assim sendo, eu divido esta sessão em três categorias (familiares, pessoas e instituições). Gostaria de agradecer, do fundo do meu coração, a três mulheres da minha vida, sem as quais eu não estaria navegando nesses mares onde o conhecimento são palavras de ordem, minha esposa Magda Vany Lourenço Dias da Silva, à qual pedi a permissão para fazer o meu doutorado e ela aceitou ficar longe por quatro anos e cuidando dos nossos filhos. Minha querida Mãe, Zulmira Pinto Vaz da Costa Jesus Bonfim, pelo encorajamento e suas palavras de mãe. Lembro-me que quando eu a disse que tinha sido selecionado para fazer o meu doutoramento e ela só disse “Vai estudar meu filho”. Finalmente a minha avó Helena de Assunção Pinto Vaz da Costa, mulher que acolheu-me desde os dois anos enquanto estava no velório do seu pai. A vocês eu dedico esta tese. Diversas pessoas foram peças-chave ao longo desses 48 meses, por exemplo meu orientador Dr. Sergio Ricardo Floeter, a quem eu agradeço muito pela oportunidade e sua orientação. Além disso, em diversas ocasiões fez o papel de pai. Renato Morais Araujo, meu amigo, conselheiro, colaborador e parceiro, agradeço muito pela oportunidade, desde o momento que trocamos e-mails onde demonstrei o meu interesse em fazer parte do Laboratório de Biogeografia e Macroecologia Marinha. Agradeço principalmente pelo apoio científico, suas observações e as nossas longas discussões tudo para que essa tese se tornasse realidade. Mariana Bender, pela sua boa vontade e disponibilidade em discutir vários aspectos relacionados com o assunto do terceiro capítulo desta tese. Carlos Eduardo Ferreira, pelo apoio dado durante a discussão dos meus resultados incansáveis vezes. Aos meus amigos, Alexandre Sequeira, Murilo Dias e Juan Pablo Quimbayo agradeço pelas diversas revisões nos meus manuscritos, desde o meu projeto de pesquisa até essa etapa final. A vocês gostaria de deixar claro que a minha ciência não seria o que é se não fosse vossa paciência e vontade para comigo. A todos os pescadores artesanais de São Tomé e Príncipe que gentilmente aceitaram em ceder alguns minutos para serem entrevistados por mim. A vocês todos, gostaria de dizer que eu não teria conseguido sem vossa ajuda. Também queria agradecer a todos os outros membros integrantes do LBMM que tive oportunidade de conviver e dialogar durante três anos. Obrigado. x

Algumas pessoas que incondicionalmente contribuíram na minha jornada que não poderia deixar de agradecer, meu mentor Angus Robin Gascoigne (1962–2012), lembro-me das nossas conversas sobre biodiversidade das ilhas de Golfo de Guiné, e hoje vejo o quanto isso despertou meu interesse para o mundo cientifico (você é o meu herói). Minha professora Alzira Rodrigues que ao longo da minha jornada acadêmica fez-me perceber que existe muito mais conhecimento do que apresentado na sala de aula. Além disso, se não fosse pela sua teimosia em fazer-me candidatar no PGCD (Programa de Pós-graduação Ciência para o Desenvolvimento) hoje não estaria te agradecendo por isso. Gostaria também de agradecer a todos os meus professores de PGCD, particularmente a Ester Serrão, pela contribuição desde o início da construção do meu projeto de pesquisa e também pela suas discussões em São Tomé, Janeiro de 2016 juntamente com professor Peter Wirtz, em que discutimos vários aspectos de componentes bentos da região, e Karim Erzini onde analisamos vários assuntos sobre a pesca em São Tomé e Príncipe inclusive aquilo que constitui o principal resultado do terceiro capítulo desta tese. Aos meus amigos Pedro Sequeira de Carvalho e Ibraíne Elba das Neves Baguide que embora são de área diferente [Direito] mas sempre tentaram entender o meu trabalho e sua importância para nosso país. Finalmente ao meu amigo e parceiro, Albertino Miranda pelos bons e agradáveis mergulhos realizados nos recifes de São Tomé e Príncipe. Foram diversas instituições e entidades que contribuíram elaboração e conclusão desta tese que sem o vosso apoio tanto financeiro como logístico. Aos meus dois programas de pós-graduação, a POSECO – Programa de Pós-graduação em Ecologia da Universidade Federal de Santa Catarina e ao PGCD, pela oportunidade de conhecer e ter magnificas aulas. Ao LBMM – Laboratório de Biogeografia e Macroecologia Marinha da UFSC, onde tive privilegio de desenvolver minha pesquisa ao longo desses três anos. CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior por ter suportado meus custos universitários e estadia no Brasil. IGC – Instituto Gulbenkian de Ciências suportou minha estadia em Cabo Verde. Fundação Calouste Gulbenkian, Rufford Foudation que custearam a minha pesquisa no campo. A California Academy of Sciences que através de nosso colaborador Dr. Luiz Rocha desencadeou meios que financeiro outros membros da missão de campo nas ilhas de São Tomé em 2016. USTP – Universidade de São Tomé e Príncipe pelo seu apoio logístico quando do meu campo e a Direcção Geral das Pescas de São Tomé e Príncipe. xi

Finalmente ao Atlantic diving center pelo apoio dado nos mergulhos e ajuda prestadas com pescadores das comunidades de Praia Messias Alves. Fotos da capa – L.A. Rocha

xii

xiii

RESUMO

Localizadas no Golfo da Guiné (África Ocidental), as ilhas oceânicas de São Tomé e Príncipe apresentam alto índice de endemismo e sofrem diversas ameaças antrópicas (e.g. poluição, sobrepesca). Estes fatores associados fazem com que essas ilhas sejam consideradas hotspot de biodiversidade marinha. Apesar dessa importância, a fauna marinha tropical dessa região é uma das menos conhecidas no mundo sendo pouco estudada, principalmente quanto à sua ecologia. Considerando esta falta de estudos ecológicos, esta tese teve como objetivo geral estudar as comunidades recifais das ilhas de São Tomé e Príncipe através de duas abordagens. A primeira delas testou fatores abióticos como a profundidade, a complexidade estrutural, hidrodinamismo como potenciais variáveis explanatórias para os padrões quantitativos (i.e. riqueza, abundância e biomassa) encontrados nas assembleias de peixes recifais e de comunidade bentônicas dessas ilhas. Além disso, testou-se o efeito da comunidade bentônica sobre estrutura de peixes recifais. A segunda abordagem retratou como pontos de referência associados a padrões de assembleias de peixes recifais vem mudando ao longo do tempo entre pescadores tradicionais nas ilhas de São Tomé e Príncipe. A tese está dividida em três capítulos: o primeiro capítulo descreve, pela primeira vez, as assembleias de peixes de recife e a composição bentônica da maior ilha oceânica do Golfo da Guiné (São Tomé). Nesse capítulo foi explorada a influência dos fatores locais, tais como, complexidade estrutural do ambiente, profundidade, exposição hidrodinâmica sobre a estrutura das assembleias de peixes recifais e organismos bentônicos. Constatou-se que existe uma variação espacial nas assembleias de peixes e organismos bentônicos ao longo dos recifes da ilha de São Tomé e que essa variação foi explicada principalmente pela exposição s ondas. Além disso, observou-se correlação positiva entre a densidade de espécies, abundância e biomassa com o gradiente de hidrodinamismo e profundidade. Também se observou que composição de abundância e biomassa de peixes recifais mudam em função da mudança na cobertura bentônica em São Tomé. O segundo capítulo compara as comunidades de peixes recifais e organismos bentônicos das ilhas de São Tomé e do Príncipe. Além disso, fez-se uma comparação temporal (com dados coletados com 10 anos de diferença) das assembleias de peixes recifais em São Tomé. Nesse capítulo, conclui-se que São Tomé e Príncipe apresentou alta similaridade na composição de assembleias de peixes recifais, tanto em termos, abundância quanto na biomassa. Por outro lado, a riqueza foi maior em São Tomé. A matriz de algas epilíticas dominou xiv

em ambas ilhas. A pressão de pesca explicou uma parte da variação observada nas assembleias de peixes. Não houve diferença significativa nas assembleias de peixes em São Tomé entre 2006 e 2016. Finalmente, no terceiro capítulo, foi avaliado o conhecimento ecológico local dos pescadores para descrever a composição das pescarias locais e as tendências das capturas ao longo do tempo nas ilhas de São Tomé e Príncipe. Além disso, avaliou-se também a contribuição relativa dos possíveis fatores que causam mudanças nas assembleias de peixes de recife de acordo com as percepções dos pescadores. Como resultado, detectou-se um decréscimo significativo no tamanho corporal das espécies recifais alvos da pesca ao longo do tempo (~ 50 anos). As gerações de pescadores mais velhos e mais jovens diferiram em suas percepções de declínios ao longo do tempo, o que sugere a ocorrência do “shifting baseline syndrome” (síndrome da mudança das linhas de base) nas ilhas de São Tomé e Príncipe, ou seja, o fenômeno que descreve a mudança dos padrões que resulta de cada geração nova que não possui conhecimento da condição histórica, e presumivelmente mais natural, do meio ambiente. Portanto, cada geração define o que é natural ou normal de acordo com as condições atuais e suas experiências pessoais. Sendo assim, conclui-se que os ecossistemas marinhos das ilhas de São Tomé e Príncipe passaram por fases diferentes de acordo com o estatus do seu ambiente, desde da colonização até os dias atuais.

Palavras-chave: Atlântico Tropical Oriental; África Ocidental Tropical; Ecossistemas marinhos; Estruturação de comunidades; Comunidades bentônicas; Peixes recifais; Conservação; Pesca artesanal. xv

ABSTRACT

Located in the Gulf of Guinea (West Africa), the oceanic islands of São Tomé and Príncipe have high rates of endemism and suffer from various anthropogenic threats. These factors led to these islands being considered marine biodiversity hotspots. Despite its importance, the tropical marine fauna of this region is one of the least known in the world, and only a few studies were dedicated to its ecology. Given the lack of ecological studies, this thesis aims to study the reef communities of São Tomé and Príncipe islands following two approaches. The first tries to explain reef and benthic community assemblage quantitative patterns (i.e., richness, abundance and biomass) using environmental drivers, such as depth, structural complexity, hydrodynamics and benthic communities. The second approach portrays how benchmarks associated with patterns of reef fish assemblages have been changing among traditional fishermen in the São Tomé and Príncipe islands. The thesis is divided in three chapters: the first one describes, for the first time, the reef fish assemblages and the benthic composition in São Tomé, the largest oceanic island of the Gulf of Guinea. In this chapter, we explored how the structure of reef fish assemblages and benthic organisms is influenced by local factors, such as structural environmental complexity, depth and wave exposure. We detected a spatial variation in fish assemblages and benthic organisms along the reefs of São Tomé island, which is mostly explained by wave exposure. In addition, species density, abundance and biomass were positively correlated to the gradient of hydrodynamics and depth. The abundance and biomass of the São Tomé reef fish assemblages was influenced by benthic communities. The second chapter compares the reef fish communities and benthic organisms of São Tomé and Príncipe islands. In addition, we made a time comparison with data collected 10 years apart on São Tomé reef fish assemblages. We found that fish richness varied between islands and between the two times sampled, but that abundance and biomass were similar. Regarding benthic cover, the matrix of epilithic algae was dominant on both islands. Fishing pressure explained part of the variation detected in fish assemblages. There was no significant difference between the 2006 and the 2016 fish assemblages of São Tomé. Finally, the third chapter assesses the ecological knowledge of local fishermen to describe the composition of local fisheries and catch trends in São Tomé and Príncipe. Furthermore, we evaluated fishermen perception on the relative contribution of the factors that are likely to be changing reef fish assemblages. We detected a significant decrease in the body size of target reef fish species over the last five decades. The xvi

perceptions on fish declines over time differed between older and younger fishermen, suggesting the occurrence of shifting baseline syndrome in São Tomé and Príncipe, a phenomenon that describes pattern changes in new generations due to an absence of historical knowledge, and presumably more natural environmental conditions. Therefore, each generation defines what is natural or normal according to current conditions and their personal experiences. We can thus infer that the marine ecosystems of São Tomé e Príncipe islands have passed through different environmental phases, since colonization until nowadays.

Keywords: Eastern Tropical Atlantic; Tropical West Africa; Marine ecosystems; Structuring of communities; Benthic community; Reef fish; Conservation; Artisanal fishing.

xvii

LISTA DAS FIGURAS Introdução geral Figura 1: Pesca recifal em São Tomé. (A) Caça submarina em Diogo Vaz; observe o tamanho pequeno do peixe capturado (foto de J.L. Gasparini); (B) Vista subaquática da rede de pesca e da canoa (foto de L.A. Rocha); (C) Canoas usadas para pesca com linha e anzol (foto de C.L. Sampaio); D) Mercado de peixes em São Tomé; observar o pequeno tamanho dos peixes e a presença de peixes herbívoros - família Labridae (foto de P. Wirtz)...... 32 Figura 2: Localização de São Tomé e Príncipe em relação à África continental, Golfo da Guiné e ilhas adjacentes (A e B). Dados representados a esquerda incluem, área emersa, idade das ilhas e distância entre ilhas e do continente (adaptado de Jones & Tye 1996). Ilhas de Príncipe (C) e São Tomé (D) incluindo a área emersa em cinza e submersalinha azul (isóbata de 20 e 50m de profundidade) ...... 34 Figura 3: Expedição São Tomé 2006. (A) Censos visuais subaquáticos, foram utilizados em transectos de 20 x 2 m para contar peixes (foto de L.A. Rocha); (B) a fotografia subaquática foi usada para registar a biodiversidade, comportamento e habitats (foto por S.R. Floeter); (C) foto-quadrados de 50 x 50 cm (foto de P. Wirtz); (D) coleção de peixes, primeiro registo de Centropyge aurantonotus no Atlântico leste (foto de S.R. Floeter); (E) coleta de tecidos para análises de DNA (foto de J.L. Gasparini)...... 35 Figura 4: População por localidades e áreas urbanas nas ilhas de São Tomé e Príncipe. A Ilha de São Tomé (a esquerda) tem uma concentração da população no nordeste e quase nenhuma concentração no sudoeste, enquanto na ilha do Príncipe (a direita), a concentração populacional maior é no centro da mesma. Mapa adaptado da carta altimétrica de São Tomé e Príncipe...... 37 Figura 5: Ilhas do Golfo da Guiné (incluindo Annobón, São Tomé e Príncipe –AO) receberam a maior parte de suas linhagens do Caribe – CA e Indo-Pacífico – OI (África Austral), e menos linhagens teriam vindo do Brasil – BR, leste do pacífico – EP, oeste leste do Tethys – WT e WE (adaptado de Floeter et al. 2008)...... 39

Capítulo I Figure 1: The location of the São Tomé island in the Atlantic Ocean: (A) the Gulf of Guinea represented by a grey hexagon; (B) the São Tomé island; (C) the Distribution of sampling sites across the island. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete xviii

pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL). Blue lines represented the isobaths between 20–50 m. (-) and (+) indicate level of wave exposure of sites. * = represent level of topographic complexity of sites: * = high; ** = medium and *** = low………...…...55 Figure 2: Comparative variation in total biomass of fish assemblages per trophic groups sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)………………….………………….………………..…60 Figure 3: Comparative variation in total biomass of fish assemblages per size classes sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites……………..………………..……….……61 Figure 4: Mean (± S.E.) variation in fish assemblages per exposure levels (A, D and G), topographic complexity level (B, E and H) and depth class level (C, F and I) sampled in 2006 at the São Tomé island. Different lowercase letters indicate significantly different values (Tukey post-hoc, p < 0.001)………………………………………………………..……..62 Figure 5: Redundancy analysis biplots representing the influence of abiotic factors by trophic group (A and B) and size class (C and D). Each fish figure represents the most common species observed in each level. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).………………….64 Figure 6: Comparative variation of benthic cover sampling at São Tomé island. Each point represents a photo-quadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), Other life (OTH) and Gorgonians (GOR). See Fig. 1 for the acronym of the sites…………….68 Figure 7: Comparative cover along of a depth gradient per sites in São Tomé island. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL)…...... ……....69 Figure S.1: Comparative variation in species density (A), abundance total (B) and biomass total (C) of fish assemblages sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), mean (red diamond), upper and lower quartiles, and 95% confidence intervals. Each circle xix represents a underwater visual censuses. Letters show statistical groupings (Tukey post-hoc) and boxplots with different letters are significantly different. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL)…...... ………………89 Figure S.2: Compative variation of abundance and biomass found in São Tomé island. (A) Abundance per trophic groups. (B) Biomass per trophic groups. (C) Abundance per size classes. (D) Biomass per size classes. Each point represents UVCs. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)……………..………..…...………………………….…90 Figure S.3: Variative variation in total abundance of fish assemblages per trophic groups sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)……………………………...……………………….....91 Figure S.4: Comparative variation in abundance total of fish assemblages per size classes sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites………………………………...………...... 92 Figure S.5: Percentage of benthic cover sampling at São Tomé island. Each point represents a photo-quadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), Other life (OTH) and Gorgonians (GOR)…………..…………………………………………93

Capítulo II Figure 1: The location of the São Tomé and Príncipe islands in the Eastern Tropical Atlantic: (A) the Gulf of Guinea represented by grey hexagon; (B) São Tomé and Príncipe islands; (C) Príncipe island and (D) São Tomé island; Distribution of sampling sites across the islands. Sites from Príncipe ( ): Praia da Lapa (LAP), pedra da Galé (GAL), Ilhéu Bombom (BOM), Praia Banana (BAN), Ilhéu dos Mosteiros (MOS), Ilhéu Boné do Joquei (BJO) and Ilhas Tinhosas (TIN). Sites from São Tomé data collected 10 years separately ( ) and Sites from São Tomé ( ): Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete xx

Pedras Catedral (SPC), Sete Pedras Branca (SPB) and Ilhéu das Rolas (ROL). Blue lines represented the isobaths of 20 and 50 m...... ……....112 Figure 2: Temporal comparison of fish assemblages among the two sampling islands (São Tomé in 2006 vs. 2016) and (São Tomé vs. Príncipe in 2016). Species richness (A), abundance (B) and biomass (C). Each color represents a different years by islands. Boxplots (B and C) show medians (black line), mean (red diamond) upper and lower quartiles, and 95% confidence intervals. Red lines in (A), represent the standardized number of surveys. Each point in (B and C) represents an underwater visual census…………………...…………………………………..…117 Figure 3: Variation between sites in São Tomé 2016 vs. Príncipe 2016 islands. Species density (A, D and G), abundance (B, E and H) and biomass (C, F and I) of fish assemblages sampled. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB), Praia Banana (BAN), Boné do Joquei (BJO), Ilhéu Bombom (BOM), Pedra da Galé (GAL), Praia da Lapa (LAP), Ilhéu dos Mosteiros (MOS) and Ilhas Tinhosas (TIN). Boxplots show medians (black line), mean (red diamond), upper and lower quartiles, and 95% confidence intervals. Each point represents an underwater visual ensus.………………………………………………………………....118 Figure 4: (a) Mean – S.E. functional fish relative species density (a), relative abundance (b) and relative biomass (c) from islands. Proportional contribution of functional groups to total standing species density (d), abundance (e) and biomass (f) from islands. The numbers indicate proportions of each functional group at each islands between different years. Functional groups: herbivores-detritivores ( ), macroalgae-feeders ( ), omnivores ( ), sessile invertebrate feeders ( ), mobile invertebrate feeders ( ), planktivores ( ) and piscivores ( )………………….…..121 Figure 5: Mean – S.E. size class of fish relative species density (a), relative abundance (b) and relative biomass (c) from islands. Proportional contribution of size class to total standing species density (d), abundance (e) and biomass (f) from islands. The numbers indicate proportions of each functional group at each islands between different years. Size class: 0–7 cm ( ), 8–15 cm ( ), 16–30 cm ( ), 31–50 cm ( ) and = > 50 cm ( )…………………………………………………………………...... 122 Figure 6: Non-metric multidimensional scaling of structure of biomass of functional groups (A) and structure of biomass of size classes (B) of fishes in São Tomé (Yellow) vs. Príncipe (Blue) islands in 2016. Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores xxi

(HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)………………………………………………………..124 Figure 7: Non-metric multidimensional scaling of (A) structure of biomass of functional groups and (B) structure of biomass of size classes of fishes in São Tomé island in 2006 (Orange) vs 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of each functional groups and size classes with the nMDS axes: Functional groups: herbivores- detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).…………….…………………….……...... 124 Figure 8: Percentage of benthic cover found during sampling. São Tomé island (A) and Príncipe island (B). Each point represents a photoquadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), other life (OTH) and Gorgonians (GOR). Boxplots show medians (black line), upper and lower quartiles, the colours are represent organism and structure of categories………………….………………...…...………126 Figure S.1: Non-metric multidimensional scaling of structure of abundance of functional groups (A) and structure of abundance of size classes (B) of fishes in São Tomé (Yellow vs. Príncipe (Blue) islands in 2016. Each circle is a sampled UVS and its size is proportional to its total abundance. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)…………...……………..……….…………………....135 Figure S.2: Non-metric multidimensional scaling of structure of abundance of functional groups (A) and structure of abundance of size classes (B) of fishes in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores- detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS)………………………………………..…....135 Figure S.3: Non-metric multidimensional scaling of structure of biomass of functional groups per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of functional xxii

groups with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS). Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB)...... 136 Figure S.4: Non-metric multidimensional scaling of structure of biomass of size classes per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of size classes with the nMDS axes. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB)...... 137 Figure S.5: Non-metric multidimensional scaling of structure of abundance of functional groups per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing abundance. Vectors depict the correlations of functional groups with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS). Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB)...... 138 Figure S.6: Non-metric multidimensional scaling of structure of abundance of size classes per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing abundance. Vectors depict the correlations of size classes with the nMDS axes. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras Branca (SPB)……………...... 139

Capítulo III Figure 1: Location of São Tomé and Príncipe islands (STPI) in the Atlantic Ocean: (A) Gulf of Guinea highlighted in the grey box; (B) STPI; (C) São Tomé island; and (D) Príncipe island. ( = Traditional fishing communities). ST: SCA – Santa Catarina; PBI – Praia Brita; PSP – Praia de São Pedro; PME – Praia Melão; PMA – Praia Messias Alves; PSA – Praia de Angolares; PRP – Praia de Ribeira Peixe; PPA – Praia de Porto Alegre. PR: LAP - Praia da Lapa; BUR – Praia das Burras; SAN – Santo António; ABA – Praia de Abade; SEC – Praia Seca...... 155 Figure 2: Relationship between best day’s catch (largest individual caught in kilograms) that fishers remembered and the time of fishing event (years ago) by species…………….……………………..…..….160 xxiii

Figure 3: (A) Variation between largest individual ever caught (in kilograms) across generations of fishers (years of practice). ( = a fisher’s catch; = mean). (B) Pie charts represent the proportions of causes of change indicated by fishers. Each color represents a different category of cause...... 161 Figure 4: Variation between largest individual ever caught (in kilograms) across generations of fishers (years of practice) by species. ( = a fisher’s catch; = mean)...... 162 Figure 5: Spatial distribution of the blast fishing cited by fishers in São Tomé island (A) and Príncipe island (B). Color gradient indicates the degree of impact based on the number of citations by fishers. Sites in São Tomé: SBC – Baia de Santa Catarina; FLZ – Fundão de Lagoa Azul; LAZ – Lagoa Azul; PQI – Praia Quinze; SAN – Ilhéu Santana; ANG – Praia de Angobó; SPE – Sete Pedras; MAL – Baia de Malanza; BPA – Baia de Porto Alegre; ROL – Ilhéu das Rolas; IHA – Praia Inhame; SMG – Ilhéu de São Miguel; PJU – Ponta da Juliana; CAC – Calé Calé. Sites in Príncipe: PPF – Ponta da Pedra Furada; GAL – Pedra da Galé; MOS– Ilhéus dos Mosteiros; SBR – Sete Braças; PBM – Ponta Banana; PNC – Ponta Café; BON – Ilhéu Boné de Jóquei; TIN – Ilhas Tinhosas. Numbers in parentheses represent total interviewed fishers by community…..…165 Figure S.1: Interaction network representing the number of fishermen across traditional villages and blast fishing sites in São Tomé (A) and Príncipe (B) island. Circle size is proportional to the number of fishers that mentioned blast fishing (left) and the number of citations for each site (right hand side). Numbers in parentheses represent interviewed fishers by communities and the number within black circles represent quantity of fishers what mentioned blast fishing inside the community……....…..180

Conclusão geral Figura 1: Status de qualidade do ecossistema marinho [peixes recifais] de São Tomé e Príncipe entre 1950 e 2016 e as perspectivas futuras. O índice (*) foi baseado nos dados das entrevistas realizadas com os pescadores artesanais dessas ilhas, combinados com os dados de censos visuais subaquáticos realizados entre diferentes períodos [2006 vs. 2016]. População humana de São Tomé e Príncipe ( ). Pescadores artesanais de São Tomé e Príncipe ( ). Período de maior estresse para os organismos marinhos de São Tomé e Príncipe [ínico de pesca com rede de nylon, pesca com bombas e pesca com arpão] o que caracterizou a fase II ( )...... 189 xxiv

Apêndice Figura A.1: Artigo publicado no periódico Coral Reefs, em colaboração com Hugulay A. Maia em 2017...... 191 Figura A.2: Web mídia produzida pelo Laboratório de Biogeografia e Macroecologia Marinha da UFSC, como mecanismo de divulgação científica para a comunidade internacional...... 192

xxv

LISTA DE TABELAS Capítulo I Table 1: Results of GLMM evaluating the influence of wave exposure (WE), topographic complexity (TC) and depth (DP) on the fish assemblages (Species density, abundance and biomass). Bold values indicate p < 0.05.…………………………………………..…….…..…63 Table 2: Results of PERMANOVA evaluating the influence of wave exposure (WE), topographic complexity (TC) and depth (DP) on the fish abundance and biomass between different trophic groups and size classes and benthic community. Bold values indicate p < 0.05.………………...66 Table S.1: Abundance (individuals per m2 ± S.E.) and biomass (individuals in grams per m2 ± S.E.) estimates of fish species recorded during underwater visual census (139 transects) in São Tomé island per sites and general. Trophic group: herbivores-detritivorous (HD), macroalgal-feeder (HM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), piscivorous (PS) and omnivores (OM).……….……………………………………………...94 Table S.2: Number of Underwater Visual Censuses along the same exposure levels, topographic complexity and deep class per sites…...... 105 Table S.3: Number of photo-quadrats along the same depth (Coral cover)……………………………………………………………..…..106

Capítulo II Table 1: Results of GLMM evaluating the influence of abiotic and anthropogenic factor on the fish community (Species density, abundance and biomass) at São Tomé ad Príncipe islands. Bold values indicate significant.…………………………………………………..………..119 Table S.1: Abundance and biomass estimates of fish species recorded during underwater visual census (343 transects) at São Tomé island in (2006 and 2016) and Príncipe island in (2016). Trophic group: herbivores- detritivorous (HD), macroalgal-feeder (HM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), piscivorous (PS) and omnivores (OM)………………………………………….....140 Table S.2: Number of Underwater Visual Censuses per sites in two sampling islands (São Tomé in 2006 vs. 2016) and (São Tomé vs. Príncipe in 2016).…………………………………………………………...... 145

Capítulo III Table 1: The eight main types of fishing gear used by artisanal fishers in São Tomé and Príncipe……………………………….……………....154 xxvi

Table 2: Results of GLM evaluating the influence of fishing time and fish species on the body size of the largest individual ever caught. Bold values indicate p < 0.001…………………….…………....……………….....163 Table S.1: Total artisanal fishers and number of interviews by traditional villages on São Tomé and Príncipe island.……..…………………..….183 xxvii

SUMÁRIO INTRODUÇÃO GERAL ...... 29 1. INTRODUÇÃO GERAL ...... 30 1.1. A complexidade dos ecossistemas marinhos e as escalas ... 30 1.2. Impactos antropogênicos na estruturação de comunidades marinhas (sobrepesca) ...... 31 1.3. Atlântico Tropical Oriental e suas regiões ...... 33 1.4. Golfo de Guiné, São Tomé e Príncipe, aspectos ecológicos 33 1.5. As lacunas no conhecimento ...... 39 PREMISSAS, HIPÓTESES E OBJETIVOS ...... 40 Objetivo geral ...... 41 Objetivos espécificos ...... 41 REFERÊNCIAS ...... 42 CAPÍTULO I ...... 49 Spatial patterns and drivers of fish and benthic reef communities of São Tomé island (Tropical Eastern Atlantic) ...... 50 Abstract ...... 51 Introduction ...... 52 Material and Methods ...... 54 Study area description ...... 54 Reef fish assemblages ...... 55 Reef benthic communities ...... 56 Environmental drivers of fish assemblages ...... 57 Data analysis ...... 57 Results ...... 58 Reef fish assemblages ...... 58 The influence of wave exposure, depth, and topographic complexity on fish assemblages...... 59 Benthic cover ...... 67 Influence of benthic community on fish assemblage ...... 69 Discussion ...... 70 Reef fish assemblage ...... 70 Benthic communities ...... 72 Influence of benthic community on fish assemblage ...... 74 Conclusion ...... 75 Acknowledgements ...... 75 References ...... 76 Supporting information ...... 89 CAPÍTULO II ...... 107 Reef fish community structure and benthic cover of two islands in the Gulf of Guinea: São Tomé and Príncipe ...... 108 xxviii

Abstract ...... 109 Introduction ...... 110 Material and Methods ...... 111 Study area ...... 111 Reef fish community structure ...... 112 Reef benthic communities ...... 113 Environmental and anthropogenic drivers ...... 114 Data analysis ...... 114 Results ...... 115 Reef fish structure ...... 115 Benthic community ...... 125 Discussion ...... 127 Acknowledgements ...... 129 References ...... 130 Supporting information ...... 135 CAPÍTULO III ...... 147 Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea ...... 148 Abstract ...... 149 1. Introduction ...... 151 2. Material and methods ...... 153 2.1. Study area description ...... 153 2.2. Data collection ...... 155 2.3. Data analysis ...... 156 3. Results ...... 157 3.1. Changes in targeted fish species ...... 157 3.2. Causes of change based on local knowledge ...... 163 4. Discussion ...... 166 4.1. Factors causing changes in reef fish compositions and/or catches ...... 167 Acknowledgements ...... 170 Appendix ...... 180 CONCLUSÕES GERAIS E CONSIDERAÇÕES FINAIS ...... 187 APÊNDICE...... 190 29

INTRODUÇÃO GERAL

Ecossistemas marinhos de São Tomé e Príncipe. Acima: recife de costão rochoso na ilha de São Tomé, Sete Pedras. Abaixo: recife biogênico composto por rodolitos, algas calcáreas roladas, na ilha de Príncipe. Fotos: Luiz A. P. Rocha

1. INTRODUÇÃO GERAL 1.1. A complexidade dos ecossistemas marinhos e as escalas A origem e evolução dos organismos dependem de três processos básicos que são fundamentais para determinar os padrões de distribuição atual das espécies: especiação, extinção e dispersão (Ricklefs 1987; Pyron & Wiens 2013). Esses processos interagem e influenciam a dinâmica de comunidades naturais na escala temporal geológica, e seu reflexo pode ser percebido em diferentes escalas espaciais: da local à global (Ricklefs 1987). A formação de comunidades biológicas, porém, não depende somente de eventos no tempo evolutivo, mas de processos ecológicos em menores escalas temporais que vêm sendo investigados por ecólogos há mais de 40 anos. A partir do debate acerca da eficácia da ecologia de comunidades em produzir respostas generalizáveis ou definir padrões (Lawton 1999), tornou-se clara a importância em compreender a influência de fatores ambientais locais (i.e. físico-químicos) na estruturação das comunidades (Clements 1916; Tansley 1935; Simberloff 2004). Além desses fatores, as comunidades vêm sendo afetadas, em uma escala temporal muito recente, por uma miríade de atividades humanas em escala global (e.g., alterações climáticas) e local (e.g., sobrepesca, destruição de hábitats e poluição). Entender os ecossistemas marinhos requer compreender a composição das suas comunidades e dos organismos que os compõem, o que constitui uma importante ferramenta para conservação (Ribeiro et al. 2005). A biota marinha está sujeita a vários processos físicos e estruturadores que consequentemente influenciam sua funcionalidade e serviços. Dentre os componentes dessa biota estão os peixes recifais, que compõem grande parte de sua estrutura e biomassa. Esses organismos são considerados modelos apropriados para o estudo da biota marinha, uma vez que compreendem um grande número de espécies (Kulbicki et al. 2013, Parravicini et al. 2013), são de fácil identificação visual e possuem a taxonomia bem resolvida (Eschmeyer et al. 2010). Além disso, são importantes componentes dos ciclos de matéria e fluxos de energia dos ecossistemas marinhos e possuem ligação trófica direta com muitas sociedades humanas (Holmlund & Hammer 1999). Estudos recentes têm mostrado alguns fatores abióticos e bióticos, tais como, exposição hidrodinâmica, temperatura da água, comunidade bentônica e a produtividade primária que influenciam a estrutura das comunidades de peixes recifais (Roberts & Ormond 1987, Friedlander & Parrish 1998, Ferreira et al. 2001, Fulton et al. 2005, Floeter et al. 2007). Recentemente,

31 estes fatores estruturadores das assembleias de peixes vêm sendo investigados nas ilhas oceânicas do Atlântico Sul. Por exemplo, no Arquipélago de Fernando de Noronha, Brasil, observou-se a importância da composição bentônica e hidrodinâmismo como fatores estruturadores da comunidade de peixes recifais (Krajewski & Floeter 2011). Por outro lado, Luiz et al. (2015) observaram uma influência marcante da exposição hidrodinâmica, sem muita influência da comunidade bentônica, que é de certa forma pouco diversa nestes ambientes isolados. Além dos fatores naturais, também existem atividades antropogênicas que influenciam diretamente na estrutura das comunidades marinhas, como pesca.

1.2. Impactos antropogênicos na estruturação de comunidades marinhas (sobrepesca) Desde que homem descobriu os serviços fornecidos pelo oceano, a pesca tem sido praticada em diversas escalas (O’Connor et al. 2011). Esta atividade ocorre tanto nos recifes quanto no mar aberto e a ausência ou a ineficiência de gestão dos recursos pesqueiros tem provocado desequilíbrios no ecossistema marinho (Worm et al. 2009). Nos últimos vinte anos tem-se debatido uma série de questões referentes ao efeito ecológico da pesca nos ecossistemas marinhos a nível global (Pauly 1995, Tegner & Dayton 1999). Sobretudo, tenta-se entender como as comunidades têm respondido à exploração em grande escala (Pauly & Christensen 1995). Baseando-se em dados históricos de ecossistemas costeiros, Jackson et al. (2001) constataram que as perdas de grandes predadores de topo pela pesca têm provocado mudanças nas comunidades ecológicas, pois outras espécies de nível trófico semelhante assumem as funções ecológicas das espécies perdidas, até que elas próprias se tornem o objeto da pesca. Hoje não restam dúvidas de que a pesca excessiva e outras perturbações humanas nos ecossistemas costeiros como a poluição e degradação na qualidade da água têm contribuído para a extinção local de muitos organismos marinhos (McCauley et al. 2015). Vários estudos vêm identificando o declínio de estoques pesqueiros ao redor do mundo, como o revelado por Myers & Worm (2003). Nesse estudo, foram utilizados dados de pesca industrial para demonstrar as alterações de biomassa e composição de grandes peixes predadores de topo de cadeia trófica ao longo do tempo. A partir desses dados, foi concluído que apenas 15 anos de exploração levaram à redução de 80% na biomassa destes peixes de topo de cadeia. O mesmo padrão já tinha sido observado em um estudo que demonstrou, a partir de dados de desembarque, uma mudança nas tendências de captura (Pauly et al. 1998).

Mesmo em locais onde a pesca é desenvolvida de forma artesanal (Figura 1), o declínio de espécies alvo também vem sendo relatado. Por exemplo, Bender et al. (2013) observaram que, das nove espécies de peixes recifais analisadas no estudo, sete apresentaram tendências significativas de declínio na captura a partir de informações fornecidas pelos próprios pescadores. Esse trabalho também demonstrou que as referências ambientais dos pescadores foram se alterando ao longo das diferentes gerações. Giglio et al. (2015) detectaram evidências de que a pesca artesanal em pequena escala pode reduzir a abundância de grandes peixes costeiros, levando também ao fenômeno de extinção local.

Figura 1: Pesca recifal em São Tomé. (A) Caça submarina em Diogo Vaz; observe o tamanho pequeno do peixe capturado (foto de J.L. Gasparini); (B) Vista subaquática da rede de pesca e da canoa (foto de L.A. Rocha); (C) Canoas usadas para pesca com linha e anzol (foto de C.L. Sampaio); D) Mercado de peixes em São Tomé; observar o pequeno tamanho dos peixes e a presença de peixes herbívoros - família Labridae (foto de P. Wirtz).

1.3. Atlântico Tropical Oriental e suas regiões A província biogeográfica do Atlântico Tropical Oriental ou África Ocidental Tropical é composta por três regiões principais: oeste da África continental, arquipélago de Cabo Verde e ilhas do Golfo da Guiné (Floeter et al. 2008). Somente recentemente, esta região biogeográfica foi estudada com mais detalhes em termos de sua biodiversidade marinha (Laborel, 1974; Wirtz et al. 2007; 2013; Floeter et al. 2008; Polidoro et al. 2017) e taxa de endemismo [c. 30% para peixes recifais (Floeter et al. 2008); c. 96% para gastrópodes (Peters et al. 2013; 2015; Polidoro et al. 2017); c. 21% para as raias (Polidoro et al. 2017); e c. 18% para corais duros (Polidoro et al. 2017)].

1.4. Golfo de Guiné, São Tomé e Príncipe, aspectos ecológicos A região do Golfo da Guiné (Figura 2A e B) apresenta altos índices de endemismo e possui quatro ilhas: Ano Bom, São Tomé, Príncipe e Bioko sendo esta considerada ilha continental. A ilha de Príncipe (Figura 2C) e de São Tomé (Figura 2D), constituem um estado insular [República Democrática de São Tomé e Príncipe], localizadas em latitudes equatoriais (entre os meridianos 1044’N e 0001’S, e os paralelos de 6028’E e 7028’E), frente à costa do Gabão. Devido ao alto endemismo e às ameaças de múltiplas fontes antropogênicas que vem sofrendo é considerado um hotspot da biodiversidade marinha (Roberts et al. 2002). Os primeiros estudos ictiológicos realizados em São Tomé e Príncipe e foram desenvolvidos por Baltazar Osorio em 1890, o qual identificou 124 espécies de peixes distribuídas em 54 famílias. Afonso et al. (1999) atualizaram a lista de espécies de peixes recifais dessas ilhas, identificando 185 espécies distribuídas em 64 famílias. Em 2006, uma equipe de pesquisadores brasileiros conduziu uma expedição científica nos recifes da Ilha de São Tomé (Figura 3). Em poucos dias de amostragem de campo, a pesquisa obteve resultados notáveis, como novos registos de ocorrência de peixes recifais e corais para a região (Wirtz et al. 2007; Floeter et al. 2008). Nesses estudos, foram identificadas 235 espécies de peixes costeiros, sendo que 3% são endêmicas, distribuídas em 67 famílias, sendo que as famílias com mais espécies são: Carangidae (14 espécies), Serranidae (11 espécies), Gobiidae (8 espécies) e Scombridae (8 espécies). Além disso, esses estudos apontam que a formação dos recifes das ilhas de São Tomé e Príncipe se assemelha a de ilhas oceânicas brasileiras, com estrutura

primariamente rochosa com crescimento biogênico secundário, baixa cobertura de corais e alta cobertura de macroalgas e algas calcárias (Floeter et al. 2006, Pinheiro et al. 2011, Luiz et al. 2015).

Figura 2: Localização de São Tomé e Príncipe em relação à África continental, Golfo da Guiné e ilhas adjacentes (A e B). Dados representados a esquerda incluem, área emersa, idade das ilhas e distância entre ilhas e do continente (adaptado de Jones & Tye 1996). Ilhas de Príncipe (C) e São Tomé (D) incluindo a área emersa em cinza e submersalinha azul (isóbata de 20 e 50m de profundidade).

Figura 3: Expedição São Tomé 2006. (A) Censos visuais subaquáticos, foram utilizados em transectos de 20 x 2 m para contar peixes (foto de L.A. Rocha); (B) a fotografia subaquática foi usada para registar a biodiversidade, comportamento e habitats (foto por S.R. Floeter); (C) foto-quadrados de 50 x 50 cm (foto de P. Wirtz); (D) coleção de peixes, primeiro registo de Centropyge aurantonotus no Atlântico leste (foto de S.R. Floeter); (E) coleta de tecidos para análises de DNA (foto de J.L. Gasparini).

A ilha de Príncipe é a segunda mais antiga do Golfo da Guiné, datada de 31 milhões de anos (Figura 2C) (Lee et al. 1994), possui uma área submersa três vezes maior do que a ilha de São Tomé (Figura 2D) e uma população humana cerca de 25 vezes menor como se pode observar na Figura 4 (RDSTP-INE 2010). Além disso, foi declarada Reserva da Biosfera pela UNESCO1 em 2011 (Abreu 2013). Recentemente Tuya et al. (2017) descreveram as assembleias de peixes recifais da ilha de

1United Nations Educational, Scientifc and Cultural Organization

Príncipe e observaram uma correlação positiva entre e a riqueza taxonômica com a profundidade dos recifes. No que se refere aos outros organismos marinhos dessas ilhas, por exemplo os corais, o primeiro trabalho de identificação das espécies foi realizado por Laborel em (1974). Neste trabalho foram identificados nove espécies de corais duros ocorrendo no Atlântico Oriental. Dentre todas, Montastrea cavernosa foi considerada a espécie mais abundante nos ambientes recifais. Existem cerca de 101 espécies de algas marinhas que foram descritas por Steentoft em 1967. Os moluscos [~250 espécies] e crustáceos parecem ser os grupos mais bem estudados em termos da sua biodiversidade em São Tomé e Príncipe (Wirtz 2003, 2004,Ardovini & Cossignani 2004, Afonso-Dias et al. 2005, Guerreiro 2007, Cumberlidge 2008, Wirtz& D’Acoz 2008, Carrison-Stone et al. 2013, Owen 2014, Rolán & Gori 2014, Polidoro et al. 2017) e são grupos cujo endemismo é relativamente alto nas ilhas. Em termos ecológicos, São Tomé e Príncipe é uma das áreas menos exploradas, seus organismos marinhos requerem estudos mais aprofundados (Floeter et al. 2006). Trabalhos recentes tratando de biogeografia e evolução de algumas famílias de peixes recifais incluem representantes desta região, como no caso do gênero Clepticus (Labridae). Este estudo revelou que Clepticus africanus, espécie endêmica do Golfo da Guiné, é geneticamente mais próximo de Clepticus brasiliensis, espécie da costa brasileira, do que de Clepticus parrae, a espécie caribenha (Beldade et al. 2009). As forças geológicas e climáticas que deram origem à fauna endêmica no Golfo de Guiné são assuntos de importância biogeográfica e as origens dos endêmicos ainda são alvo de muito debate. No entanto, hipóteses biogeográficas anteriores sugerem que a troca de espécies entre as regiões é em grande parte unidirecional: da mais diversa para a menos diversa (Vermeij 1991, Briggs 1995). Dessa forma, presume-se que os organismos marinhos das Ilhas do Golfo da Guiné (incluindo Ano Bom, São Tomé e Príncipe) receberam a maior parte de suas linhagens do Caribe e Indo-Pacífico (África Austral), e que menos linhagens teriam vindo do Brasil, Macaronésia e/ou se diversificado nessa região, como é ilustrado na Figura 5.

Figura 4: População por localidades e áreas urbanas nas ilhas de São Tomé e Príncipe. A Ilha de São Tomé (a esquerda) tem uma concentração da população no nordeste e quase nenhuma concentração no sudoeste, enquanto na ilha do Príncipe (a direita), a concentração populacional maior é no centro da mesma. Mapa adaptado da carta altimétrica de São Tomé e Príncipe2.

O ambiente físico das ilhas de São Tomé e de Príncipe é caracterizado por uma forte influência da corrente da Guiné (Teixeira 2002). Esta influência verifica-se especialmente ao redor da ilha do Príncipe, onde as águas de superfície costumam ter temperaturas particularmente elevadas (28–29°C) e uma salinidade relativamente baixa 0 (33–34 /00). Ao sudoeste da ilha de São Tomé pode existir um efeito de ressurgência que leva-o uma diminuição das temperaturas (até 23°C) e um aumento da salinidade na superfície (36%) (Carneiro & Carvalho 2013). Essas variações climáticas locais potencialmente influenciam a distribuição e abundância dos organismos marinhos, tornando importante

2Democratic Republic of Sao Tome and Principe- Ministry of Natural Resources, Energy and Environment, Directorate General of the Environment

a comparação das comunidades marinhas em diferentes sítios nas duas ilhas. Além disso, desde o início da colonização europeia as atividades de pesca têm sido uma prática em São Tomé e Príncipe, tanto do ponto de vista de subsistência, como do ponto de vista econômico e desportivo (UNESCO 2013). Essas atividades têm se intensificado nas últimas décadas pelos grandes navios de pesca da Europa e da Ásia, ao passo que a pesca em pequena escala nos recifes permanece como uma importante atividade econômica nesta região (Krakstad 2010, Serigstad et al. 2012). Em São Tomé e Príncipe foram realizados dois estudos com o objetivo de compreender a situação da pesca no país. O primeiro deles objetivou conhecer quais espécies eram capturadas em diferentes pontos de pesca de São Tomé (Costa 1959), enquanto o segundo apresentou algumas causas para extinção local de algumas espécies (Sestello 1970). Em alguns casos, as espécies não puderam ser identificadas como segue no excerto seguinte:

“... Nada disso seria verdade pois, se é certo que tivemos ali o enorme prazer de capturar, ao laçado, um dos peixes mais esquisitos (e teimosos) pegados até hoje por pescadores desportivos – um misto de tubarão e raia (espécie primária, possivelmente anterior ao tubarão comum, que já por si remonta a muitos milhões de anos) adornado de uma flamante roupagem que fazia lembrar a de um leopardo...”(Sestello 1970).

Muitas espécies podem ter sido alvo de sobrepesca sem serem identificadas.

Figura 5: Ilhas do Golfo da Guiné (incluindo Annobón, São Tomé e Príncipe – AO) receberam a maior parte de suas linhagens do Caribe – CA e Indo-Pacífico – OI (África Austral), e menos linhagens teriam vindo do Brasil – BR, leste do pacífico – EP, oeste leste do Tethys – WT e WE (adaptado de Floeter et al. 2008).

1.5. As lacunas no conhecimento Considerando a lacuna no conhecimento ecológico da fauna marinha destas ilhas, bem como a importância das ilhas do Golfo da Guiné como hotspot de biodiversidade marinha, estudei a estrutura de comunidades bentônicas e de peixes recifais das ilhas de São Tomé e Príncipe através de três abordagens. Primeira, os padrões espaciais de assembleias de peixes e de comunidades bentônicas ao longo da ilha de São Tomé. No primeiro capítulo. Além disso, mostrei como certas variáveis físicas (e.g. exposição às ondas, profundidade, complexidade topográfica) estruturam as assembleias de peixes recifais e comunidades bentônicas. Finalmente, também mostrei como a comunidade bentônica estrutura as assembleias de peixes nessa ilha. Segunda, padrões espaciaisdas comunidades bentônicas e de peixes recifais entre as ilhas de São Tomé vs. Príncipe. No segundo capítulo, comparei temporalmente as assembleias de peixes recifais de São Tomé entre os anos 2006 vs. 2016. Por fim, relacionei as assembleias

de peixes das duas ilhas com fatores abióticos e antropogênicos (e.g., pressão de pesca). Finalmente, a terceira abordagem, mudança de linha de base entre pescadores tradicionais. No terceiro capítulo, usei conhecimento ecológico local das comunidades de pescadores artesanal para descrever a composição das pescarias locais e as tendências das capturas através do tempo em São Tomé e Príncipe. Também mostrei os fatores responsáveis pelas tendências detectadas nas pescarias locais segundo a percepção dos pescadores e mapeio a distribuição de um desses promotores (e.g., pesca com explosivo).

PREMISSAS, HIPÓTESES E OBJETIVOS Cientes da ausência de estudos ecológicos sobre a fauna marinha das ilhas de São Tomé e Príncipe, foram definidas algumas perguntas centrais que guiaram o presente estudo a fim de identificar e entender os padrões espaciais e temporais dessas ilhas: (1) Como varia a estrutura das comunidades de peixes recifais e bentos entre diferentes sítios na ilha de São Tomé e de Príncipe? (2) Como as comunidades de peixes recifais variam em sua estrutura trófica e tamanho corporal dos espécimes entre diferentes sítios na ilha de São Tomé e no Príncipe? (3) Como a exposição das ondas, a complexidade topográfica, a profundidade e a pressão de pesca influencia na riqueza, abundância e biomassa das espécies de peixes? (4) Como a comunidade de peixes e bentos variou temporalmente na ilha de São Tomé entre 2006 e 2016? (5) Os pescadores artesanais de São Tomé e Príncipe perceberam alguma mudança na composição de peixes recifais e no ambiente marinho ao longo do tempo? Em caso afirmativo, que fatores são responsáveis pelas mudanças, segundo as suas percepções? Assim sendo, foram definidas as seguintes hipóteses: (1) As assembleias de peixes recifais e comunidades bentonicas tem uma variação ao longo das ilhas de São Tomé e Príncipe (i.e., maiores riqueza de peixes, abundância e biomassa em locais mais complexos, mas expostos e mais profundos), e está associada aos fatores ambientais, tais como exposição às ondas, complexidade topográfica e profundidade, proporcionando diferentes microhabitates ao longo das ilhas. (2) O grupo dos peixes planctívoros é mais abundante em ambas as ilhas devido à água clara, temperatura da superfície entre 27-290C, ondulações fortes. (3) A biomassa é maior na ilha de Príncipe porque a população humana dessa ilha é 25 vezes menor do que da ilha de São Tomé fazendo com que a pressão de pesca nessa ilha seja menor. (4) Com o aumento da população

41 humana em São Tomé de 2006 a 2016 espera-se que o impacto da pesca na biomassa de peixes tenha deixado um efeito mensurável. (5) Espera- se que a “síndrome de mudanças na linha de base” ocorra em São Tomé e Príncipe, assim como tem sido reportada em vários locais do mundo.

Objetivo geral Estudar as comunidades recifais das ilhas de São Tomé e Príncipe através de duas abordagens. A primeira testa fatores ambientais como a profundidade, a complexidade estrutural, hidrodinamismo e comunidades bentônicas como potenciais variáveis explanatórias para os padrões quantitativos (i.e. riqueza, abundância e biomassa) encontrados nas assembleias de peixes recifais e de comunidade bentônicas dessas ilhas. A segunda vertente retrata como pontos de referência associados a padrões de assembleias de peixes recifais vêm mudando ao longo do tempo entre pescadores tradicionais nas ilhas de São Tomé e Príncipe.

Objetivos espécificos a. Entender como a estrutura das assembleias de peixes recifais e comunidade do bentos varia entre diferentes sítios nas ilhas de São Tomé e de Príncipe; b. Saber como a comunidade de peixes recifais varia em sua estrutura trófica e tamanho corporal entre diferentes sítios na ilha de São Tomé e no Príncipe; c. Comprender como a exposição às ondas, a complexidade topográfica, profundidade e pressão de pesca influenciam a riqueza, abundância e biomassa das espécies de peixes; d. Detectar quais são os efeitos dessas variáveis nas assembleias de peixes em grupos tróficos e em classes de tamanho corporal; e. Identificar os componentes bentônicos dominantes nos recifes na ilha de São Tomé e no Príncipe (com foco em corais); f. Saber se exposição às ondas, complexidade topográfica e profundidade influenciam a composição das comunidades bentônicas na ilha de São Tomé; g. Saber se a estrutura das assembleias de peixes recifais mudam em função das mudanças na cobertura de bentos; h. Comparar espacialmente a comunidade bentônica e assembleias de peixes de recifais da ilha de São Tomé vs. Príncipe em 2016; i. Comparar temporalmente as assembleias de peixes recifais em São Tomé entre 2006 e 2016;

j. Investigar as possíveis mudanças na pesca de peixes recifais em São Tomé e Príncipe nas últimas décadas; k. Investigar as principais causas de mudanças na composição de peixes recifais alvo de pesca de acordo com a conhecimento ecológico local de pescadores; l. Espacializar a prevalência de pesca destrutiva (pesca explosiva) nos recifes das ilhas de São Tomé e Príncipe.

REFERÊNCIAS Abreu, A.D., (2013). Príncipe Island’s Biosphere Reserve (Democratic Republic of São Tomé & Príncipe): A Living Laboratory for Sustainable Development. In: Clusener-Godt RP-S& M (ed) AfriMAB Biosphere reserves in Sub-Sahan Africa: Showcasing sustainable development, First. Department of Environmental Affairs & UNESCO, South Africa. 284–302. Ardovini, R. & Cossignani, T., (2004).West African Seashells, Mostra Mondiale Malacologia, Via Adriatica nord, 240–63012 Cupra Marittma (AP) Italy. Afonso, P., Porteiro, F.M., Santos, R.S., Barreiros, J.P., Worms, J. & Wirtz, P. (1999). Coastal marine fishes of São Tome Island (Gulf of Guinea). Arquipélago, Life and Marine Sciences, 17A:65–92. Afonso-Dias, M., Pires, A., & Clark, P.F., (2008). Occurrence of Chaceon gordonae and C. sanctaehelenae (Crustacea: Brachyura: Geryonidae) off the Island of São Tomé. Marine Biodiversity Records, 1:1–2. Beldade, R., Heiser, J.B., Robertson, D.R., Gasparini, J.L., Floeter, S.R., & Bernardi, G. (2009). Historical biogeography and speciation in the creole wrasses (Labridae, Clepticus). Marine Biology, 156:679–687. Bender, M.G., Floeter, S.R. & Hanazaki, N. (2013). Do traditional fishers recognise reef fish species declines? Shifting environmental baselines in Eastern Brazil. Fisheries Management and Ecology. 20:58–67. Briggs, J.C. (1995). Global Biogeography. Developments in Paleontology and Stratigraphy, 14. Amsterdam: Elsevier. Carneiro G., & Carvalho, J. (2013). Impato das alterações climáticas sobre os ecossistemas marinhos e a pesca em São Tomé e Príncipe. In: Alterações Climáticas e suas repercussões sócio-ambientais. Aveiro, 84–111. Carrison-Stone, D., Syoc, R.V., Williams, G., & Brian, S. W., (2013). Two new species of the gorgonian inhabiting barnacle, Conopea (Crustacea, Cirripedia, Thoracica), from the Gulf of Guinea. Zookeys,

270:1–20. Costa, F.C., (1959). Estudos de pescas em S. Tomé e Príncipe. Junta de Investigações do Ultramar, Estudos de Zoologia, 219–271. Clements, F.E. (1916). Plant sucession: an analysis of the development of vegetation. Publication 242. Washington, D. C.: Carnegie Institution of Washington. Cumberlidge, N., (2008). Insular species of Afrotropical freshwater crabs (Crustacea: Decapoda: Brachyura: Potamonautidae and Potamidae) with special reference to Madagascar and the Seychelles. Contributions to Zoology, 77(2)71–81. Eschmeyer, W.N., Fricke, R., Fong, J.D. & Polack, D.A., (2010). Marine fish diversity: History of knowledge and discovery (Pisces). Zootaxa, 2525:19–50. Ferreira, C.E.L., Gonçalves, J.E.A., & Coutinho, R., (2001). Community structure of shes and habitat complexity on a tropical rocky shore. Environmental Biology of Fishes 61:353–369. Floeter, S.R., Ferreira, C.E.L., Rocha, L.A., Wirtz, P., Sampaio, C.L.S. & Gasparini, J.L. (2007). The evolution and ecology of a poorly known Atlantic biodiversity hotspot, Report to the National Geographic Society (Grant # 7937-05). Floeter, S.R., Krohling, W., Gasparini, J.L., Ferreira, C.E.L., Zalmon, I.R., (2007) Reef fish community structure on coastal islands of the southeastern Brazil: The influence of exposure and benthic cover. Environmental Biology of Fishes, 78:147–160. Floeter, S.R., Halpern, B.S. & Ferreira, C.E.L., (2006).Effects of fishing and protection on Brazilian reef fishes. Biological Conservation, 128:391–402. Floeter, S.R., Rocha, L.A., Robertson, D.R., Joyeux, J.C., Smith-Vaniz, W.F., Wirtz, P., Edwards, A.J., Barreiros, J.P., Ferreira C.E.L., Gasparini, J. L., Brito, A., Falcón, J. M., Bowen, B. W. & Bernardi, G. (2008) Atlantic reef fish biogeography and evolution. Journal of Biogeography, 35:22–47. Friedlander, A.M. & Parrish, J.D., (1998). Habitat characteristics affecting fish assemblages on a Hawaiian coral reef. Journal of Experimental Marine Bioology Ecology, 224:1–30. Fulton, C.J., Bellwood, D.R. & Wainwright, P.C., (2005). Wave energy and swimming performance shape coral reef fish assemblages. Proceedings of the Royal Society Biological Sciences, 272:827–832. Giglio, V.J., Luiz, O.J. & Gerhardinger, L.C., (2015). Depletion of marine megafauna and shifting baselines among artisanal fishers in eastern Brazil. Animal Conservation, 18:348–358.

Guerreiro, A., (2007). Moluscos de S. Tomé e Príncipe. Centro Português de Estudos dos Mamíferos Marinhos. Portugal. Holmlund, C.M. & Hammer, M., (1999). Ecosystem services generated by fish populations. Ecological Economics, 29:253–268. Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange, C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J. & Warner, R.R. (2001). Historical overfishing and the recent collapse of coastal ecosystems. Science, 293:629–637. Jones, P.J. & Tye, A., (2006). The Birds of São Tomé and príncipe, with Annobón: islands of the Gulf of Guinea. BOU Checklist Series No. 22. British Ornithologists’Union. Oxford. Laborel J. (1974). West African reefs corals: an hypothesis on their origin. Proceedings of the 2nd. International Coral Reef Symposium, 1:425– 443. Krajewski, J.P. & Floeter, S.R., (2011). Reef fish community structure of the Fernando de Noronha Archipelago (Equatorial Western Atlantic): The influence of exposure and benthic composition. Environmental Biology of Fishes, 92:25–40. Krakstad, J.O., Alvhein, O., Lopes, J.D. S. & Iwamoto, T., (2010). Surveys of the fish resources of São Tomé & Príncipe, Survey of the demersal resources. Report to the Food and Agriculture Organization (Grant # CCP/INT/003/NOR). Kulbicki, M., Parravicini, V., Bellwood, D.R., Arias-Gonzàlez, E., Chabanet, P., Floeter, S.R., Friedlander, A., McPherson, J., Myers, R.E., Vigliola, L. & Mouillot, D., (2013). Global biogeography of reef fishes: A hierarchical quantitative delineation of regions. PLoS ONE, 8(12): e81847. Lawton, J.H., (1999). Are there general laws in ecology? Oikos, 84:177– 192. Lee, D.C., Halliday, A.N., Fitton, J.G. & Poli, G., (1994). Isotopic variations with distance and time in the volcanic islands of the Cameroon line: evidence for a mantle plume origin. Earth and Planetary Science Letters, 123:119–138 Luiz, O.J., Mendes, T.C., Barneche, D.R., Ferreira, C.G.W., Noguchi, R., Villaça, R.C., Rangel, C.A., Gasparini, J.L. & Ferreira, C.E.L., (2015). Community structure of reef fishes on a remote oceanic island (St Peter and St Paul’s Archipelago, equatorial Atlantic): the relative influence of abiotic and biotic variables. Marine and Freshwater

Research, 66(8):739–749. McCauley, D.J., Pinsky, M.L., Palumbi, S.R., Estes, J.A., Joyce, F.H. & Warner, R.R. (2015). Marine defaunation: Animal loss in the global ocean. Science, 347 (6219):1255641. Myers, R.A. & Worm, B., (2003). Rapid worldwide depletion of predatory fish communities.Nature, 423:280–283. O’Connor, S., Ono. R. & Clarkson, C., (2011). Pelagic Fishing at 42,000 years before the Present and the Maritime Skills of Modern Humans. Science, 334:1117–1121. Owen, B., (2014). A new species of Haliotis (Gastropoda) from São Tomé & Príncipe Islands, Gulf of Guinea, with comparisons to other Haliotis found in the Eastern Atlantic and Mediterranean. Zootaxa, 3838(1):113–119. Osório, B., (1891). 3a Nota. Peixes maritimos das ilhas de S. Thomí, do Principe e ilheo das Rolas. Jornal de Sciências, Mathemáticas, Physicas e Naturaes. Lisboa, 2:97–139. Parravicini, V., Kulbicki, M., Bellwood, D.R., Friedlander, A.M., Arias- Gonzalez, J.E., Chabanet, P., Floeter, S.R., Myers, R., Vigliola, L., D’Agata, S. & Mouillot, D., (2013). Global patterns and predictors of tropical reef fish species richness. Ecography, 36:1254–1262. Pauly, D., (1995). Anecdotes and the shifting baseline syndrome of fisheries. Trends in ecology & evolution (Personal edition).10:430. Pauly, D. & Christensen, V., (1995). Primary production required to sustain global fisheries. Nature, 374:255–257. Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Trres, F.J., (1998). Fishing Down Marine Food Webs.Science, 279:860–863. Pinheiro, H.T., Ferreira, C.E.L., Joyeux, J.C., Santos, R.G. & Horta, P.A., (2011). Reef fish structure and distribution in a south-western Atlantic Ocean tropical island. Jouranal of Fish Biology, 79:1984–2006. Polidoro, B.A., Carpenter, K.E., Ralph, G.M., Strongin, K., Harvey, M., Arnold, R., Buchanan, J.R., Mohamed, K., Camara, A., Collette, B.B., Comeros, M.T., Bruyne, G. De, Gon, O., Harold, A.S., Harwell, H., Hulley, P.A., Iwamoto, T., Knudsen, S.W., Dieu, J. De, Linardich, C., Lindeman, K.C., Monteiro, V., Munroe, T., Nunoo, F.K.E., Pollock, C.M., Poss, S., Russell, B., Sayer, C., Sidibe, A., Smith, W., Emilie, V., De L.T., Christophe, J., (2017). The status of marine biodiversity in the Eastern Central Atlantic (West and Central Africa). Aquatic Conservation Marine and Freshwater Ecosystems, 27:1021–1034. Pyron, R.A. & Wiens, J.J., (2013). Large-scale phylogenetic analyses reveal the causes of high tropical amphibian diversity. Proceedings of the Royal Society Biological Sciences, 11: 280(1770) –20131622.

Quimbayo J.P., Floeter, S.R., Noguchi, R., Rangel, C.A., Gasparini, J.L., Sampaio, C.L. S., Ferreira, C.E.L. & Rocha, L.A., (2012). Cleaning mutualism in Santa Luzia (Cape Verde Archipelago) and São Tomé Islands, Tropical Eastern Atlantic. Marine Biodiversity Records, 5:1– 7. RDSTP – Repupblíca Democrática de São Tomé e Príncipe, Instituto Nacional de Estatística, (2010). Inquérito Demográfico e Sanitário (IDS STP 2008–2009). São Tomé e Príncipe. RDSTP – Democratic Republic of Sao Tome and Principe - Ministry of Natural Resources, Energy and Environment, Directorate General of the Environment, (2009). Fourth national report on the biodiversity 1st draft. Ribeiro, C., Almeida, A. J., Araújo, R., Biscoito, M. & Freitas, M., (2005). Fish assemblages of Cais do Carvão Bay (Madeira Island) determined by visual census technique. Journal of Fish Biology, 67:1568–1584. Ricklefs, R.E., (1987). Community diversity: relative roles of and regional processes. Science, 235:167–171. Roberts, C.M., McClean, C.J., Veron, J.E.N., Hawkins, J.P., Allen, G.R., McAllister, D. E., Mittermeier, C.G., Schueler, F.W., Spalding, M., Wells, F., Vynne, C. & Werner, T. B., (2002). Marine biodiversity hotspots and conservation priorities for tropical reefs. Science, 295:1280–1284. Rolán E., & Gori, S., (2014). New information on the marginellids of São Tomé and Príncipe, with new records and the description of four new species. Sociedad Española de Malacologia, Iberus, 32:1–25. Steentoft, M.S.L.F., (1965) A revision of the marine algae of São Tomé and Principe (Gulf of Guinea) Botanical Journal of the Linnean Society, 60:99–146. Serigstad, B., Ensrud, T. M., Olsen, M., Ostrowski, M. & Dias, Stepien, A. M., Jokiel, A. M., Schmit, B., Lipinski, M, Alvheim, O., Uche- Okeke, C., (2012). Marine Environmental Survey of bottom fauna, selected physical and chemical compounds and fisheries survey in the Joint Development Zone between Nigeria and Saõ Tomé & Principe. Report to the Food and Agriculture Organization (Grant # GCP/INT/003/NOR) Sestello, F., (1970). A pesca desportiva de barco com factor do desenvolvimento Turístico de S. Tomé e Príncipe - Notas Breves para um Melhor Conhecimento do Acanthocybium solandri (Cuvier). In Boletim Informativo.Número especial.V centenário (1470-1970):53– 63.

Simberloff, D., (2004). Community ecology: is it time to move on? American Naturalist 163:787–799. Tansley, A.G., (1935). The use and abuse of vegetational concepts and terms. Ecological Society America, 16:284–307. Tegner, M.J., Dayton, P.K., (2000). Ecosystem effects of fishing in kelp forest communities. Journal of Marine Science, 57:579–589. Teixeira, M.C.N.A., (2002). Relatório Nacional para Elaboração da Estratégias Nacional da Conservação da Biodiversidade em S. Tomé e Príncipe. Tuya F.A., Bosch N.E.A., Abreu A.D.B., Haroun R.A., (2017). Reef fish at a remote tropical island (Principe Island, Gulf of Guinea): disentangling taxonomic, functional and phylogenetic diversity patterns with depth. Mar Freshw Res Vermeij, G.J. (1991). When biotas meet: understanding biotic interchange. Science, 253: 1099–1104. Wirtz P., (2003). New records of marine invertebrates from São Tomé island (Gulf of Guinea). Journal of the Marine Biological Association of the United Kingdom, 83: 735–736. Wirtz, P., (2004). Four amphi-Atlantic shrimps new for São Tomé and Príncipe (eastern central Atlantic). Arquipélago, Life and Marine Science, 21A:83–85. Wirtz, P., Ferreira, C.E.L., Floeter, S.R., Fricke, R., Gasparini, J.L., Iwamoto, T., Rocha. L.A. & Schliewen, U.K., (2007). Costal Fishes of São Tomé and Príncipe islands, Gulf of Guinea (Eastern Atlantic Ocean) — an update. Zootaxa, 1523:1–48. Wirtz, P. & C. d’Udekem d’Acoz (2008). Crustaceans associated with Cnidaria, Bivalvia, Echinoidea and Pisces at São Tomé and Príncipe islands. Arquipélago. Life and Marine Sciences, 25:63–69. Worm, B., Hilborn, R., Baum, J.K., Branch, T., Collie, J.S., Costello, C., Fogarty, M.J., Fulton, E., Hutchings, J., 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., Watson, R., Zeller, D. (2009). Rebuilding global fisheries. Science, 325:578–585.

CAPÍTULO I

Spatial patterns and drivers of fish and benthic reef communities of São Tomé island (Tropical Eastern Atlantic)

(Submetido para Marine Ecology) Formatado de acordo as normas da revista

Maia H.A., Morais R.A., Quimbayo J.P., et al. (2018). Spatial patterns and drivers of fish and benthic reef communities of São Tomé island (Tropical Eastern Atlantic). Marine Ecology.

Spatial patterns and drivers of fish and benthic reef communities of São Tomé island (Tropical Eastern Atlantic)

Maia H.A.*1,2,3,4, Morais R.A.*5, Quimbayo J.P.*2,6, Dias M.S.7, Sampaio C.L.S.8, Horta P.A.9, Ferreira C.E.L.6, Floeter S.R.1,2‡

1 Programa de Pós-graduação em Ecologia, Universidade Federal de Santa Catarina, Florianópolis, SC 88010-970, Brasil; 2 Laboratório de Biogeografia e Macroecologia Marinha, Departamento de Ecologia e Zoologia, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, Florianópolis, SC 88010-970, Brasil; 3 Programa de Pós-graduação Ciência para o Desenvolvimento (PGCD), Instituto Gulbenkian de Ciência, Oeiras, 62780-156, Portugal; 4 Departamento de Ciências Naturais e Biologia, Universidade de São Tomé e Príncipe, Bairro Quinta de Santo António, São Tomé e Príncipe. 5 College of Science and Engineering, James Cook University, Townsville, QLD 4811, Australia; 6 Laboratório de Ecologia e Conservação de Ambientes Recifais, Universidade Federal Fluminense, Niterói, RJ 24001-970, Brasil. 7 Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, Brasil; 8 Departamento de Engenharia de Pesca, Unidade de Ensino Penedo, Universidade Federal de Alagoas, Maceió, AL, Brasil 9 Laboratório de Ficologia, Departamento de Botânica, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, 88040-970, Florianópolis, SC, Brasil. ______*First co-authorship

‡Correspondence author: Sergio R. Floeter, Tel: +55 48 3721 5521; Fax +55 48 3721 5156; E-mail: [email protected]

Target Journal: Marine Ecology

Abstract Marine communities can vary across spatial scales due to changes in habitat structure, wave exposure, depth, and anthropogenic activities. Here, we characterize for the first time the reef fish assemblage and benthic community in six sites around São Tomé island in the Tropical Eastern Atlantic region. We performed 139 underwater visual census (40 m2) and 112 photo-quadrats (0.25 m2) to characterize São Tomé’s reef fish assemblages and benthic communities. Planktivores were the most important trophic group, both in terms of fish abundance and biomass. Regarding size classes, small fishes (0–7 cm) peaked in abundance, whereas biomass was mainly concentrated in 8–15 and 16–30 cm body size classes. About 30% of the total benthic cover was dominated by turf algae. Montastrea cavernosa was the most abundant coral (46% of the coral cover). Significant effects of all predictors were detected, but wave exposure was the most important factor influencing the reef fish structure. Benthic cover also influenced the composition of fish species both in terms of abundance and biomass. Overall, our results show that small planktivores are the dominant trophic group in the fish community and that the relatively low biomass of medium and large species probably reflects the long term fishing pressure in São Tomé island.

Keywords: Gulf of Guinea, abundance, biomass, species density, abiotic factors

Introduction Several studies have evaluated the role of environmental variables in shaping tropical reef fish assemblages (e.g., Crane et al., 2017; Jones & Syms, 1998; Luckhurst & Luckhurst, 1978). Important drivers highlighted so far include wave exposure (e.g., Friedlander et al., 2002; Fulton & Bellwood, 2004; Fulton et al., 2005), water temperature (e.g., Floeter et al., 2005; Parravinici et al., 2013), depth (e.g., Brokovich et al., 2008; Friedlander & Parrish, 1998), habitat topography and complexity (e.g., García-Charton et al., 2004; Roberts & Ormond, 1987; Wilson et al., 2007), benthic community cover (Ferreira et al., 2001; Friedlander & Parrish, 1998; García-Charton et al., 2004) and interactions among species (Wisz et al., 2013). For example, it is generally acknowledged that shallow, sheltered and highly complex coral habitats host large abundances of reef fishes (Dustan et al., 2013; Gratwicke & Speight, 2005), while deeper areas in isolated places often contain high fish biomass (Friedlander et al., 2010; 2016; Khalil et al., 2017; Luiz et al., 2015). Moreover, the interplay between hydrodynamics and fish swimming ability (with its associated feeding performance) enables certain functional groups, like planktivores, to be more abundant in high wave-exposure areas (Bellwood et al., 2002; Floeter et al., 2007; McGehee, 1994; Wainwright et al., 2002). In the last decade, studies focused on oceanic islands have revealed remarkable features, such as unusual high fish biomass and peculiar benthic communities (e.g., Friedlander & DeMartini, 2002; Friedlander et al., 2014; Sandin et al., 2008; Stevenson et al., 2007). The high abundance and biomass of large predators and herbivorous species in these isolated systems contrast with more accessible reef systems (Friedlander et al., 2003; 2010; Graham & McClanahan, 2013; Morais et al., 2017; Williams et al., 2011). These observations have provided the basis for posterior large-scale studies showing that human-related variables explain fish assemblages biomass around the world (Bellwood et al., 2012; Cinner et al., 2012; 2013; Mora et al., 2001; Williams et al., 2015). However, most studies addressing the structure of fish assemblages in tropical oceanic islands have been performed in the Indo- Pacific (Sandin et al., 2008; Williams et al., 2015), the Southwestern Atlantic (Krajewski & Floeter, 2011; Longo et al., 2015; Luiz et al., 2015; Pinheiro et al., 2011) or some into the Tropical Eastern Pacific (e.g., Edgar et al., 2011; Friedlander et al., 2012; Quimbayo et al., 2017). In comparison, fish assemblages from the Tropical Eastern Atlantic (TEA)

53 remain poorly explored apart from scant species checklists (Afonso et al., 1999; Wirtz et al., 2007, one study in oil platforms (Friedlander et al., 2014) and one relating fish communities with a depth gradient (Tuya et al., 2017). It is, therefore, important to quantitatively describe the structure of reef fish and benthic communities, to understand how they vary as a function local scale variables, and estimate to what extent, if any, humans are driving fish abundance and biomass, especially in these poorly known sites of the TEA. Oceanic island ecosystems in the Tropical Eastern Atlantic (TEA) include the Cape Verde Archipelago and the islands from the Gulf of Guinea: Annobón, São Tomé and Príncipe (Floeter et al., 2008). Although local endemism is relatively low for marine organisms in each of the Gulf of Guinea islands, the regional endemism level is high (Floeter et al., 2008; Hachich et al., 2015), a phenomenon presumably linked to the geographic isolation of the TEA from the other Atlantic reef areas (e.g., ~3,500 km from the Brazil and ~8,696 km from the Caribbean; Floeter et al., 2008), as well as a history of isolation and reconnection with the Indo- Pacific in the evolutionary timescale (Cowman et al., 2017). In addition to the Benguela Current that limits the movements of tropical species from the Indian Ocean, cold waters from the Northeastern Atlantic also limits the geographic range of tropical species (Almada et al., 2013; Floeter et al., 2008). A growing population (Belhabib, 2015), with high levels of local dependence on fisheries for food (Carneiro, 2011), and the emergence of foreign industrial fisheries vessels (Belhabib, 2015), contribute to the Gulf of Guinea being listed as one of the most threatened marine hotspots of the world (Roberts et al., 2002). These issues remain poorly addressed so far, since no Marine Protected Areas or effective fisheries management exist at these islands. Also, very little is known on how marine reef assemblages of the Gulf of Guinea islands are distributed along environmental gradients, hence hindering further progress towards marine conservation. In terms of benthic community the TEA also has low species richness of important groups such as hard corals (11 species) and high dominance of turf algae. Two coral species are endemics (18%; Polidoro et al. 2017). However, virtually nothing is known about the ecology of the benthic community in the region. The only descriptive study remains that of Laborel (1974). Thus, from our knowledge, this is the first quantitative study of the structure benthic communities in the Gulf of Guinea. Our objective here was to describe for the first time the reef fish assemblages and benthic composition of the São Tomé island (hence STI),

the largest oceanic island in the Gulf of Guinea. We were specifically interested to answer: (1) How do species fish density, abundance, and biomass vary among sites in STI? (2) How do fish assemblages vary in their trophic and body size structure among sites in STI? (3) Do wave exposure (WE), topographic complexity (TC) and depth (DP) influence fish species richness, abundance and biomass and, if so, to what extent? (4) What are the effects of these variables in the fish assemblages across trophic groups and body size classes? (5) What are the dominant benthic components in STI reefs (with a focus on corals)? (6) How do benthic communities vary among sites? (7) Do WE, TC, and DP influence the composition of benthic communities in STI? (8) Do the structure of fish assemblages change as a function of changes in coral cover? To address these questions, we combined underwater visual surveys of reef fishes and benthic communities in different sites and depth strata of STI.

Material and Methods Study area description São Tomé island is located along the Cameroon Volcanic Line in the Gulf of Guinea, Western Africa (0°20'94''N; 6°62'10''E; Figure 1). It was formed by volcanic activity around 13 million years ago (Caldeira, 2002; Lee et al., 1994). São Tomé island has a land area of 857 km2, but its steep underwater relief results in a shallow platform of about 732 km2 (Hachich et al., 2015; 2016). Underwater, São Tomé seascape is mainly dominated by volcanic rocky reefs with limited coral growth (Laborel, 1974; Quimbayo et al., 2012). Despite of its relatively old age, São Tomé has low marine endemism levels due to high oceanographic connectivity between the Gulf of Guinea islands and the coast (Floeter et al., 2008; Wirtz, 2003). The most influential of these currents are the Benguela Current (coming from South) and Gulf of Guinea Current (coming from North), which converge in São Tomé to form the South Equatorial current (Measey et al., 2007). The sea surface temperature in São Tomé ranges between 26–29°C during September to May, and decrease to approximately 25°C during June to August (Horemans et al., 1994). Cool waters from the Benguela Current are normally restricted to layers below 20–30 m and can reach temperatures of 20 °C (Laborel, 1974). We collected underwater data on reef fish and benthic communities in six sites around the island (Figure 1C). These sites were: (1) “Diogo Vaz” (DVA), which presents a small terrace with underdeveloped coral formations above sand and high concentration of gorgonians; (2) “Cabras Kia” (KIA), which encompasses a sandy bottom

55 with some sparse hard coral growth and high density of sea urchins; (3) “Ilhéu Santana” (SAN), which is composed by large rocky boulders with some level of coral overgrowth; (4) “Sete Pedras – Catedral” (SPC), primarily rocky slopes with hard coral and gorgonian growth; (5) “Sete Pedras – Pedra Branca” (STB), a submerged rocky outcrop rising steeply from sand and dominated by Tubastrea sp. growing on the vertical walls; and (6) “Ilhéu das Rolas” (ROL), presenting a small terrace with underdeveloped coral formations and the bottom dominated by a rhodolith bank. Among all sites, SPC and SPB are the only ones located less than 10 km from a river discharge.

Figure 1: The location of the São Tomé island in the Atlantic Ocean: (A) the Gulf of Guinea represented by a grey hexagon; (B) the São Tomé island; (C) the Distribution of sampling sites across the island. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL). Blue lines represented the isobaths between 20–50 m. (-) and (+) indicate level of wave exposure of sites. * = represent level of topographic complexity of sites: * = high; ** = medium and *** = low.

Reef fish assemblages Underwater visual surveys along transects of 40 m2 (20 x 2 m) were used to estimate reef fish species density, abundance and biomass. This method consists on identifying, counting and estimating the size

(total length in cm) of all fish species observed both in the water column (in the lay-out phase) and on the bottom (in the returning phase, Ferreira et al., 2001; 2004; Floeter et al., 2007). We performed 139 UVSs between 5 and 30 m of depth (see Table S.2). We visually estimated the total length (TL, in cm) of each specimens and then their weight using the allometric length-weight conversion W= a * TLb, where parameters a and b are species-specific constant, and W is the weight in grams. Length-weight parameters were gathered for each species from FishBase (Froese & Pauly, 2016). In cases where species coefficients were not available, we used coefficients of congeneric species that are presumably phylogenetically or morphologically similar. Assemblage structure was evaluated in terms of trophic and body size groups. These are highly relevant and commonly assessed fish functional traits that reflect physiological and energetic processes. We classified all observed reef fish species according to their trophic group (seven classes) following Mouillot et al. (2014), and to body size (five classes) following Kulbicki et al. (2015). The fish trophic groups considered were herbivore-detritivores (HD: feed upon turf and filamentous algae and/or detritus), macroalgae-feeders (HM: feed upon large fleshy algae and/or seagrass), omnivores (OM: feed from both vegetal and animal material), sessile invertebrate feeders (IS: feed mainly in corals, sponges or ascidians), mobile invertebrate feeders (IM: feed on mobile benthic prey, such as crabs, snails and polychaetes), planktivores (PK: feed on zooplankton in the water column) and piscivores (PS: feed on fish or cephalopods). The body size classes used were <7 cm, between 8–15 cm, between 16–30 cm, between 31–50 cm, and larger than 50 cm.

Reef benthic communities We estimated the cover (%) of each component of reef benthic communities using photo-quadrats. We distributed a mean of 19 (between seven to 35) photo-quadrats uniformly along the same depth contours and at the same sites where the visual censuses were conducted, summing up a total of 112 photo-quadrats (Table S.3). Photos were taken from a distance of ∼80 cm from the substratum, and each photo-quadrat corresponded to an area of 50×50 cm (e.g., Krajewski & Floeter, 2011). Each photo-quadrat was analyzed using the software Coral Point Count v3.6 (CPCe) with Excel extensions (Kohler & Gill, 2006). Thirty points were randomly positioned over each image and the organism below each point was identified as the “turf” (the epilithic algal matrix), calcareous algae, sponges, macroalgae, sand, coral, zoanthidae, gorgonians, and

57 other organisms (e.g., Echinoidea and Crustacea; Bell & Barnes, 2001; Littler & Littler, 1984; Steneck & Dethier, 1994). All corals observed were identified according to Laborel (1974) to the lowest taxonomic level possible.

Environmental drivers of fish assemblages We investigated the effect of depth (DP), topographic complexity (TC) and wave exposure (WE) on reef fish assemblages in São Tomé island. We classified DP in three levels: shallow (4–8 m), medium (9–16 m) and deep (17–30 m). We classified TC in three categories according to Ferreira et al. (2001): high (large boulders and holes >1 m in size and depth), medium (predominance of small boulders and holes <1 m in size and depth) and low (few and small benthic organisms and predominance of epiphytic algae). Finally, we classified WE in four levels: high (unsheltered site, unprotected from currents and dominant waves); medium-high (unsheltered site, unprotected from currents, but protected from dominant waves); low-medium (site sheltered by physical barriers such as islets, or protected against the main current or dominant waves); and low (site sheltered by physical barriers, protected against current and wave dominant).

Data analysis To assess whether species density (number of species per 40 m2), abundance and biomass of reef fishes varied with wave exposure levels (high, medium-high, low-medium and low, Table S.2), topographic complexity levels (high, medium and low, Table S.2) and depth strata (shallow, medium and deep, Table S.2), we used Generalized Linear Models (GLMs) with Gamma distribution. We considered this distribution because all variables are positive and have continuous values (Zuur et al., 2009). Tukey tests were used as post-hoc tests to determine the differences between groups. We carried out this analysis using the function glht within of the package multcomp (Hothorn et al. 2008). To test the drivers of reef fish species density, abundance and total biomass, we employed three Generalized Linear Mixed Model (GLMMs) using wave exposure, topographic complexity and depth as covariates (fixed factors), and sites as random variables (i.e., random- intercept model). For all these models, we used a Gamma distribution. Additionally, we evaluated if fish abundance and biomass structure in terms of trophic groups and size classes were affected by wave exposure, topographic complexity, depth and site. To do so, we used a Permutational

Multivariate Analyses of Variance (PERMANOVA; with 999 permutations) based on a Bray-Curtis dissimilarity matrix. We used another PERMANOVA based on a Bray-Curtis dissimilarity matrix to explore variations of the different group of benthic assemblages and the abiotic factors considered above. We used a Redundancy Analysis (RDA) to graphically display the influence of the abiotic factors (detected by PERMANOVA) on fish abundance and biomass in São Tomé. To test the influence of benthic cover on reef fish assemblage of São Tomé island, we used the Mantel analysis with 999 permutations to evaluate the concordance between fish and benthic cover. Benthic cover data represent a relative abundance index (%) and did not need to be standardized. We applied the Bray–Curtis Index for abundance and biomass to generate the dissimilarity matrices for both fish and benthic cover. All statistical analyses were performed in R software version 3.2.4 (R Core Team 2016), most of them using the vegan package (Oksanen et al., 2015).

Results

Reef fish assemblages We recorded 43,018 individuals belonging to 66 species from 30 families in the 139 UVCs (Table S1). The most representative families were Gobiidae and Serranidae (six species each), followed by Labridae, Muraenidae and Pomancentridae (five species each), Balistidae, Holocentridae and Lutjanidae (three species each; Table S1). Fifteen species represented 95% of the total abundance, whereas only seven species represented 95% of total biomass. The most important species in terms of abundance and biomass were Paranthias furcifer, Chromis multilineata, Myripristis jacobus, Holocentrus adscensionis and Heteropriacanthus cruentatus (Table S1). The reef fish assemblages varied spatially among the studied sites in STI (Figure S1). For example, mean species richness per transect (40 m2) was highest in KIA 16.35 ± 0.64 (Mean ± S.E.), followed by SPC, SPB and ROL (~12.36 ± 0.57, each site), whereas DVA and SAN has the lowest species density 10.54 ± 0.26 (Figure S1A). Fish abundance was highest in KIA (14.26 ± 0.78 ind/m2), followed by SPB (12.34 ± 0.51 ind/m2), SAN (7.32 ± 0.55 ind/m2), DVA (5.21 ± 0.37 ind/m2), SPC (3.56 ± 0.25 ind/m2), and ROL (3.57 ± 0.20 ind/m2; Figure S1B). Fish biomass was highest in SPB (359.86 ±15.66 g/m2), followed by KIA (344.23 ±

39.42 g/m2), SAN (173.73 ±16.30 g/m2), SPC (141.37 ± 13.62 g/m2), DVA (96.75 ± 6.94 g/m2), and ROL (58.85 ± 5.29 g/m2; Figure S1C). Planktivores were the most abundant trophic group, summing 84% of all individuals and an average abundance of 6.52 ± 0.57 ind/m2 (Figure S2A). Each of the other trophic groups contributed less than 15% of the total abundance. This pattern of abundance per trophic groups was observed across all sites (Figure S3). Similarly, biomass was concentrated in planktivores, summing 60% of the total biomass and with an average biomass of 111.71 ± 18.52 g/m2 (Figure 2, and S2B), followed by mobile invertebrate feeders (29.71 ± 5.69 g/m2) and herbivores/detritivores (27.05 ± 5.68 g/m2). The contribution of planktivores and mobile invertebrate feeders to total biomass was high across all sites, whereas herbivores/detritivores and the other trophic groups varied considerably (Figure 2). The smallest size class (0–7 cm) encompassed the highest abundance, summing 71% of all individuals (5.50 ± 0.45 ind/m2; Figure S2C), followed by fishes 8–15 cm that summed 27% of all individuals (mean: 2.08 ± 0.31 ind/m2), whereas other classes comprehended less than 10% of the total abundance. Biomass, however, was concentrated in the 8–15 cm size class that summed 68% of total biomass (125.40 ± 21.04 g/m2; Figure 3, and S2D), followed by the 16–30 cm size class (16%) and by the 0–7 cm size class (6%). The overall pattern of biomass per size classes was similar across sites, although larger size classes (16–30 cm or larger) were more variable than smaller ones (Figure 3, S4).

The influence of wave exposure, depth, and topographic complexity on fish assemblages Wave exposure was the most important factor that structured reef fish communities in São Tomé island (Table 1; Figure 4A, D, G). For instance, species density, abundance and biomass increased with high wave exposure level (Tukey test: p < 0.001; Figure 4A, D, G). Topographic complexity varied only for abundance (Table 1; Figure 4E). In terms of depth, abundance and biomass were highest in deeper sites (Tukey test: p < 0.001; Figure 4F, I). No clear trend was found for species density and biomass in general (Figure 4B, H).

Figure 2: Comparative variation in total biomass of fish assemblages per trophic groups sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Figure 3: Comparative variation in total biomass of fish assemblages per size classes sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites.

Figure 4: Mean (± S.E.) variation in fish assemblages per exposure levels (A, D and G), topographic complexity level (B, E and H) and depth class level (C, F and I) sampled in 2006 at the São Tomé island. Different lowercase letters indicate significantly different values (Tukey post-hoc, p < 0.001).

Table 1: Results of GLMM evaluating the influence of wave exposure (WE), topographic complexity (TC) and depth (DP) on the fish assemblages (Species density, abundance and biomass). Bold values indicate p < 0.05.

Fish assemblages df F p-value Species density DP 2 6.384 0.041

TC 2 0.01 0.920

WE 3 37.82 <0.001 Total bundance DP 2 42.54 <0.001 TC 2 57.72 <0.001 WE 3 50.00 <0.001 Total biomass DP 2 9.00 0.011 TC 2 5.40 0.067 WE 3 10.04 0.018

Overall, we detected significant effects of all predictors on reef fish structure, being wave exposure the most important factor that influenced the abundance per trophic group and size class (Table 2). For instance, 18% of the variation in abundance and 8% in biomass between trophic groups were explained by wave exposure (Table 2). In contrast, much less variance was explained by topographic complexity and depth. In terms of abundance per size class, wave exposure explained 15% of the variation observed. No single variable explained more than 19% of total variance and residual variance was very high (69% to 74%, respectively for trophic groups and size classes, Table 2). The abundance of planktivorous was positively correlated with sites most exposed to hydrodynamics, while mobile invertebrate feeders, omnivores and piscivores were negatively correlated to exposure. Herbivores-detritivores and macroalgae-feeders were positively correlated with topographic complexity and negatively with depth. On the other hand, sessile invertebrate feeders exhibited a positive effect associated to depth (Figure 5A). In terms of biomass, wave exposure was positively correlated with piscivores and negatively with mobile invertebrate feeders, omnivores and macroalgae-feeders. The herbivores-

detritivores were positively associated to topographic complexity (Figure 5B). Species with small and medium size (8–30 cm) presented high abundance in sites with high wave exposure, whereas the medium-large size (31–50 cm) were negatively associated to depth (i.e., were associated to shallow sites; Figure 5C). In terms of biomass, very small (0–7 cm) and medium (16–30 cm) species were negatively associated with depth, very large species (> 50 cm) were positively associated to topographic complexity and small species (8–15 cm) occupied mainly shallow areas with low wave exposure (Figure 5D).

Figure 5: Redundancy analysis biplots representing the influence of abiotic factors by trophic group (A and B) and size class (C and D). Each fish figure represents the most common species observed in each level. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile

invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Table 2: Results of PERMANOVA evaluating the influence of wave exposure (WE), topographic complexity (TC) and depth (DP) on the fish abundance and biomass between different trophic groups and size classes and benthic community. Bold values indicate p < 0.05.

Fish Abundance Biomass

Source df F R2 p-value df F R2 p-value Trophic group WE 3 10.98 0.18 0.001 3 4.65 0.08 0.001 TC 2 2.77 0.03 0.018 2 3.8 0.04 0.001 DP 2 2.46 0.02 0.04 2 3.69 0.04 0.001 Residual 131 0.75 131 0.81 Size classes WE 3 8.88 0.15 0.001 3 3.75 0.07 0.001 TC 2 1.75 0.02 0.078 3 2.27 0.02 0.011 DP 2 2.77 0.03 0.015 2 4.13 0.05 0.001 Residual 131 0.78 131 0.84

Table 2: Cont:

Benthic Benthic cover community

Source df F R2 p-value

WE 3 8.97 0.18 0.001 TC 2 4 0.27 0.005 DP 2 6 0.1 0.001

Residual 104 0.7

Benthic cover Overall, São Tomé island’s reef benthic communities were dominated by turf algae (associated biota that forms the epilithic algal matrix), with an average cover of 30% (± 2.42), followed by calcareous algae (28 ± 2.28%), sand (14 ± 2.41%), macroalgae (10 ± 1.50%), corals (8 ± 1.21%), sponges (7 ± 1.05%), zoanthids (3 ± 0.68%), gorgonians (0.36 ± 0.12%) and others (1.88 ± 0.51%). However, this order varied among sites (Figure 6). About one-third of the total variation on the benthic composition was associated with topographic complexity (nested within site; Table 2). Wave exposure (18%) and depth (8%) were less important factors influencing changes on the benthic community (Table 2). Montastraea cavernosa was the most abundant hard coral, being particularly abundant in areas below 17 m of depth, although in Diogo Vaz this species was also abundant between 9 and 16 m (3 to 12% cover). The cup coral Tubastraea sp. dominated shallow depths in SPC and SAN, varying in cover from six to 12%. Other relatively common coral species were Porites branneri attaining up to 16% of the coral cover at SPB; Siderastrea radians that varied from 0.2 and 3% of cover in all depths of DVA and KIA; and finally Schizoculina fissipara and Porites porites which had very small cover and were restricted to ROL, one in mid- depths and the other in the deeper depth stratum (Figure 7).

Figure 6: Comparative variation of benthic cover sampling at São Tomé island. Each point represents a photo-quadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), Other life (OTH) and Gorgonians (GOR). See Fig. 1 for the acronym of the sites.

Figure 7: Comparative cover along of a depth gradient per sites in São Tomé island. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL).

Influence of benthic community on fish assemblage We observed a relatively weak yet significant relationship between benthic communities and the composition of fish species, both in terms of abundance (Mantel R = 0.061, p = 0.019) and biomass (Mantel R = 0.102, p = 0.001). This indicates that changes in benthic community promote changes in reef fish species composition.

Discussion The fish and benthic communities in all sites around São Tomé island were mainly influenced by wave exposure, as expected, since this abiotic factor could reduce topography complexity and increase the concentration of available nutrients (i.e. food), which could favor the concentration of high abundances of fish, but limit the development of benthic species due to hydrodynamic forcing. The low biomass of top predators and large herbivores, as well as the dominance of planktivores in all sites around of the São Tomé island, suggest that fish community is influenced by human activities, such as fishing, which is expected since the main source of animal protein of this country comes from fishing activities (RDSTP, 2009; Maia et al., 2018). Besides, the low richness of hard corals and high cover of turf reflects the poor regional species pool, which is associated to level of isolation of this biogeographical province.

Reef fish assemblage Overall, wave exposure was the most important factor structuring the reef fish assemblages in STI, especially patterns of abundance per trophic group and size classes, whereas depth and topographic complexity were less important. This result could be associated with increasing availability of food, since the constant action of waves and wind removing the first layers of water column, allows the rising of cold waters rich in nutrients (Gove et al., 2016), and therefore favor planktivorous species, considered the base of the food chain at isolated habitats (Luiz et al., 2015; Quimbayo et al., 2017). These patterns are similar to those observed in other islands such as Kingman Atoll (Friedlander et al. 2010), Northwestern Hawaiian islands (Fukunaga et al., 2016) in the central Pacific, and St Peter and St Paul’s Archipelago in the mid-Atlantic Ridge (Luiz et al., 2015). Probably all these oceanic islands are physical barriers changing the deep currents and favoring high abundance and biomass of fish (Gove et al., 2016). The effects of topographic complexity on species richness abundance and biomass have been demonstrated in a range of ecosystems, including reefs (e.g. Friedlander et al., 2003), seagrass (e.g. Heck & Wetstone, 1977) and kelp (e.g. Russell, 1977) beds. In São Tomé island species richness and biomass of fishes was not correlated with topographic complexity, which is different from findings in other islands (i.e. a strong relationship between topographic complexity and richness, abundance and biomass of reef fishes; Friedlander et al., 2003; Longo et al., 2015; Luiz et al., 2015; Pinheiro et al., 2011; Quimbayo et al., 2017). 70

The lack of relationship between topographic complexity and assemblage metrics (species richness and biomass) could be related to the smamm differences among the classes of complexity (low, medium and high) when compared with biogenic coral reefs in other regions (Gratwicke & Speight, 2005; Komyakova et al., 2013). Moreover, complexity could be consequence of the human influence in certain areas, since it is known that grenades have been historically used for fishing, thus reducing topographic complexity in STI (Maia et al., 2018). Relationships with depth are also commonly reported. In the neighboring Príncipe island, for instance, Tuya et al. (2017) found that the taxonomic and functional diversity of reef fishes decreased monotonically with depth, however they did not explored abundance or biomass patterns in relation to depth. Overall, there is a consensus that richness and abundance increase toward high depth, but they peak at distinct positions along this gradient (Brokovich et al., 2008; Consoli et al., 2016; Garcia-Sais, 2010; Parker & Ross 1986). Results found here agree with all these studies on reef fish, but the peak was observed at the deepest class (i.e., 17–32 m). The observed pattern in the present study is likely related to the concentration of species density, fish abundance and biomass in deeper areas where the fishing pressure is lower, especially in sites under high human impact such as STI (e.g. fishing pressure and destructive fishing; Maia et al., 2018). The dominance of planktivores in São Tomé island (84.0% total abundance) was consistent with general pattern observed in other oceanic islands (Krajewski & Floeter 2011;Longo et al., 2015; Luiz et al., 2015; Pinheiro et al., 2011; Quimbayo et al., 2017), oil platforms (Friedlander et al., 2014), and rocky reefs (Truong et al., 2017). This result could be associated to transparency of water and upwelling, which may favor food detection, high plankton abundance, and therefore favor plankton feeders, such as Paranthias furcifer and Chromis multilineata. The absence and low contribution of piscivores and large species, such as groupers, sharks, snappers and jacks, to the abundance and biomass could be related to fishing pressure exerted by the human populations on the island, which depends intensively on this resource as animal protein source (Maia et al., 2018). Particularly, oceanic sharks, which has been almost extinct in the island, are the first fishing targets (Ferretti et al., 2010; Graham et al., 2010; Myers et al., 2007). Local extinction of top predators has also been reported in St Paul’s Rocks an oceanic island in the Atlantic Ocean (Luiz & Edwards, 2011). These results support urgent management tools and conservation plans in STI.

The high biomass of herbivores-detritivores was associated with the presence of parrotfish Sparisoma choati, which is a large-bodied species frequently observed in the UVCs, however due to the lack of top predators this species is being target of fishing, which can change it current abundance. On the other hand, the low contribution of macroalgae-feeder, sessile invertebrate feeders, mobile invertebrate feeders could be related with low regional richness of this trophic groups in the TEA province (Kulbicki et al., 2013). All these results suggest that STI is under high human impact compared other pristine islands where top predators are more abundant (Longo et al., 2015; Quimbayo et al., 2017; Sandin et al., 2008). Fish smaller than 15 cm dominated the reef of São Tomé island (98% total abundance and 68% of total biomass). These results partially corroborate those found in other oceanic islands such as, Malpelo, Trindade Island, and St Paul’s Rocks (Luiz et al., 2015; Pinheiro et al., 2011; Quimbayo et al., 2017). However, these islands have large biomass of individuals >30 cm, due to high concentration of large predators and relatively low fishing pressure, whereas the highest biomass in São Tomé island was between 8–15 cm. Our results reinforce that fish assemblage in STI is under low “prey release effect” (Dulvy et al., 2004). This phenomenon is usually observed in reefs with overexploitation of top predatory species, such as sharks and large groupers, in which populations of smaller piscivorous fishes, such as sea basses, increase in abundance (Anderson et al., 2014).

Benthic communities Turf and calcareous coralline algae were the overall dominant benthic component of São Tomé island’s reefs, with 30% and 28%, respectively, of total cover. This pattern is similar to the observed in other oceanic islands of the South Atlantic, such as Trindade Island (Turf: 30– 40%; calcareous algae: 16–25%; Pinheiro et al., 2011), Rocas Atoll (Turf: 51%; calcareous algae: 33%; Longo et al., 2015) and St. Peter and St. Paul’s Archipelago (Turf: 20–70%; calcareous algae: 10–25%; Luiz et al., 2015). Overall, turf algae (the main component of the epilithic algal matrix) are the dominant primary producers on most tropical and many temperate coasts, playing key roles in the ecology of subtidal shores (Adey & Goertemiller 1987; Airoldi, 2003; Underwood et al., 1991). They occupy substantial areas of rock substrate, both on surface cover and biomass basis (Copertino et al., 2005). Turf-dominated communities are expanding throughout the world, mainly due to a decline in water 72

73 quality around urbanized coasts, particularly due to increases in ambient nutrient concentrations and sediment loads (Airoldi, 2003; Benedetti- Cecchi et al., 2001; Eriksson et al., 2002; Gorgula & Connell 2004). Coralline algae are also important bulk producers and consolidators of the reef in most tropical and subtropical areas. They are particularly important in coral reef ecosystems where they play a major role in the carbon and carbonate budgets (Littler & Littler 1988; Payri 1995; 1997; Steneck, 1997). Coral cover was generally low in our samples, although there are some localized areas of high cover. This pattern is also typical of other South Atlantic reefs, such as the ones in the Brazilian southeastern coast (e.g. Floeter et al., 2007) and oceanic islands (e.g. Longo et al., 2015; Pinheiro et al., 2011). Laborel (1974) noted that the species poor and homogeneous coral assemblages of the Eastern Atlantic are generally unable to fuse and form large reef structures. Although isolation from more biodiverse regions (i.e. Indo-Pacific and Caribbean; Nunes et al., 2008) could be invoked to explain this pattern, the existence of coral reefs along the Brazilian coast (e.g., Castro & Pires, 2001; Leão et al., 2003) make evident that this might not be the complete picture. It is interesting to note, some of the main reef builders in Brazil are from the genus Mussismillia, which is absent from the Eastern Atlantic (Laborel, 1974). We found abundant cover of the cup coral Tubastrea sp. in shallow reefs of Santana and Sete Pedras. Species from this genus are alien either in the Caribbean and Brazil (Creed et al., 2017), but have become particularly invasive in Brazilian reefs (Capel et al., 2017; De Paula & Creed, 2004; Silva et al., 2011). Tubastrea sp. display biological traits that enhance their potential of population growth, such as aggressiveness in competition for space and an asexual reproduction mode (Capel et al., 2014). Considering the biogeographical context of the coral fauna of West Africa, Laborel (1974) concluded that Tubastraea sp. must be a relatively recent invader, maybe coming from the Indo Pacific. In addition, Friedlander et al. (2014) reported a mean of 47% of Tubastraea cover in oil rigs investigated off Gabon, where it was the most abundant taxon. They also considered it likely a relatively recent invader to the Gulf of Guinea. The absence of older records of this highly conspicuous genus, and the current doubts regarding identification lead us to conservatively categorize Tubastraea sp. as cryptogenic in the São Tomé island. Gorgonians had one of the lowest benthic cover around São Tomé’s reefs. This is probably a simple reflex of gorgonians’ body plan and verticalized growth. The islands in the Gulf of Guinea are possibly

the only tropical oceanic islands that harbour substantial populations of gorgonians in the South Atlantic. Moreover, gorgonians appear to be increasingly common from shallow to the deeper waters in São Tomé island. Wirtz and D'Udekem D'Acoz (2008) reported large gorgonians forming forests below 30 m deep in Príncipe and São Tomé islands. Morais & Maia (2016) noted that these forests, with substantial composition also from black corals, extended to mesophotic reefs down to at least 50 m depth, but certainly much deeper (Authors pers. obs.). Depth and wave exposure are all important drivers of change in the cover of reef benthic communities (Friedlander et al., 2013; 2014; 2017). Topographic complexity is normally evaluated as either correlated or as a property of benthic communities and not a potential causal agent to them (e.g. Wilson et al., 2007). Nevertheless, in our study, most of variation on species composition was explained by wave exposure and topographic complexity, with depth being a minor driver (Table 2). Wave exposure brings, at the same time, both physical and nutritional consequences to marine organisms. From one perspective, wave-swept habitats preclude the colonization by organisms that do not hold adequate morphology to withstand the energetic water movements (Denny et al., 1985; Horn et al., 1982). On the other hand, organisms with some level of morphologic plasticity can develop in different growth forms in response to water movements (e.g. massive instead of branching Milleporine corals in the Southwestern Atlantic; Souza, 2017). From a nutritional perspective, water currents bring food for many sessile benthic organisms in the form of plankton and dissolved organic matter (Gili & Coma 1998).

Influence of benthic community on fish assemblage Changes in benthic community cover promote changes in fish composition both in terms of abundance and biomass, the later showing the strongest variation change. Similar relationships between fish assemblages and benthic cover have been found in other oceanic islands having low coral cover (about 20%, Krajewski & Floeter 2011). Coral cover is an important proxy for habitat complexity (Bergman et al., 2000; Garpe & Öhman 2003; Jones & Syms, 1998; Munday, 2000) as they provide food and shelter, hence fostering the growth and persistence of fish individuals. Our analysis support a close link between fish species composition and benthic cover, which can be explained by the numerous documented interactions between fish and corals (Fautin & Allen, 1997; Litsios et al., 2012; Ricciardi et al., 2010). Such connection between two 74

75 important ecosystems compartments suggest that changes on benthic cover due to the rising of global temperature (e.g., coral bleaching) could have some influence on reef fishes.

Conclusion This work not only offers an unprecedented window into fish assemblage structure of one of the least studied islands in the world, but also tested the importance of exposure, depth and topographic complexity as drivers of reef community structuring. We concluded that: (1) reef fish assemblage, in term of richness, abundance and biomass, vary significantly along sites even on small spatial scales. Small fishes peaked in abundance, whereas the biomass was mainly concentrated in medium (8–15 and 16–30 cm) body size classes. The biomass of top predators is relatively low; (2) the exposure level was the most important abiotic factor in structuring of the reef fish assemblages in STI; (3) the benthic cover vary among sites and the turf (epilithic algal matrix) was the most important benthic group; (4) the hard coral Montastrea cavernosa was the most important species comprising 46% of coral cover; (5) the exposure, topographic complexity and depth influenced the benthic cover and (6) the reef fish assemblage change as a function of benthic cover. Taken together, these results suggest an urgent need for conservation efforts through the establishment of marine reserves along the island.

Acknowledgements We thank National Geographic Society for the Grant # 7937-05, Rufford Foundation for the Grant #18424-1, and Fundação Calouste Gulbenkian (Portugal) (P-141933) from the grants received. CNPq grants to SRF and CELF over the years. We thank CAPES (Brazil) for financial support. Direcção Geral das Pescas de São Tomé e Príncipe for permits and logistics, Peter Wirtz for assisting in the processing of the photo- quadrats. We also immensely thank all members of the National Geographic expedition at São Tomé 2006, especially João L. Gasparini, Angus Gascoigne, Luiz A. Rocha, Flavia Nunes and Nancy Knowton. Thanks to Jean Louis Testori and all the crew of Club Maxel, Alberto Miranda and all the crew from Atlantic Diving Center, Nora Rizzo, and the people of São Tomé e Príncipe for wonderful times.

References Adey, W. H. & Goertemiller, T. (1987). Coral reef algal turfs: master producers in nutrient poor seas. Phycologia, 26, 374–86. https://doi.org/10.2216/i0031-8884-26-3-374.1 Airoldi, L. (2003). The effects of sedimentation on rocky shore assemblages. Oceanography and Marine Biology: an Annual Review, 41, 161–236. Anderson, A.B., Bonaldo, R.M., Barneche, D.R., Hackradt, C.W., Félix- Hackradt, F.C., García-Chartón, J.A., & Floeter S.R. (2014). Recovery of grouper assemblages indicates effectiveness of a marine protected area in Southern Brazil. Marine Ecology Progress Series, 514, 207–215. https://doi.org/10.3354/meps11032 Anderson, A.B., Salas, E.M., Rocha, L.A. & Floeter, S.R. (2017). The recent colonization of south Brazil by the Azores chromis Chromis limbata. Journal of Fish Biology, 91, 558–573. https://doi.org/10.1111/jfb.13363 Afonso, P., Porteiro, F.M., Santos, R.S., Barreiros, J.P., Worms, J., & Wirtz, P. (1999). Coastal Marine fishes of São Tomé Island (Gulf of Guinea). Arquipélago Life and Marine Sciences, 17, 65–92. Almada, V.C., Toledo, J.S., Brito, A., Levy, A., Floeter, S.R., Robalo, J.I., Martins J., & Almada, F. (2013). Complex origins of the Lusitania biogeographic province and northeastern Atlantic fishes. Frontiers of Biogeography, 5, 3–11. Badgley, C., Smiley, T.M., Terry, R., Davis, E.B., DeSantis, L.R.G., Fox, D.L., Hopkins, S.S.B., Jezkova, T., Matocq, M.D., Matzke, N., McGuire, J.L., Mulch, A., Riddle, B.R., Roth, V.L., Samuels, J.X., Stromberg, C.A.E., & Yanites, B.J. (2017). Biodiversity and Topographic Complexity: Modern and Geohistorical Perspectives. Trends in Ecology & Evolution, 32, 211–226. Bell, J., & Galzin, R. (1984). Influence of live coral cover on coral-reef fish communities. Marine Ecology Progress Series, 15, 265–274. Bell, J.J., & Barnes, D.K.A. (2001). Sponge morphological diversity: A qualitative predictor of species diversity? Aquatic Conservation: Marine and Freshwater Ecosystems, 11, 109–121. Bellwood, D.R., Wainwright, P.C., Fulton C.J., & Hoey, A. (2002). Assembly rules and functional groups at global biogeographical scales. Functional Ecology, 16, 557–562. Bellwood, D.R, Hoey, A. S., & Hughes, T.P. (2012). Human activity selectively impacts the ecosystem roles of parrotfishes on coral reefs. The Royal Society B, 279, 1621–1629. 76

Bergman, K.C., Öhman, M.C., & Svensson, S. (2000). Influence of habitat structure on Pomacentrus sulfureus, a Western Indian Ocean reef fish. Environmental Biology of Fishes, 59, 243–252. Benedetti-Cecchi, L., Pannacciulli, F., Bulleri, F., Moschella, P. S., Airoldi, L., Relini, G. & Cinelli, F. P. (2001). Predicting the consequences of anthropogenic disturbance: large-scale effects of loss of canopy algae on rocky shores. Marine Ecology Progress Series, 214, 137–50. Brewer, T.D., Cinner, J.E., Green, A., & Pressey, R.L. (2013). Effects of human population density and proximity to markets on coral reef fishes vulnerable to extinction by fishing. Conservation Biology, 27, 443–452. Bruno, J.F., & Bertness, M.D., (2001). Habitat modification and facilitation in benthic marine communities. In Bertness, M.D., Gaines, S.D., & Hay, M.E., (Eds.), Marine community ecology (pp. 201–218). Sinauer Associates: Sunderland. Caldeira, R.J.M.M. (2002). Petrology of ultramafic nodules from São Tomé island, Cameroon Volcanic Line (oceanic sector). Journal of African Earth Sciences, 34, 231–246. Carneiro, G. (2011). They come, they fish, and they go: fisheries agreements with Cape Verde and São Tomé e Príncipe. Marine Fisheries Review, 73, 1–25. Castro, C.B. & Pires, D.O. (2001). Brazilian coral reefs: What we already know and what is still missing. Bulletin of Marine Science, 69, 357– 371. Capel, K.C.C., Migotto, A.E., Zilberberg, C., & Kitahara, M. V. (2014). Another tool towards invasion? Polyp “bail-out” in Tubastraea coccinea. Coral Reefs, 33, 1165–1165. Capel, K.C.C., Toonen, R.J., Rachid, C.T.C.C., Creed, J.C., Kitahara, M.V., Forsman, Z., & Zilberberg, C. (2017). Clone wars: asexual reproduction dominates in the invasive range of Tubastraea spp. (Anthozoa: Scleractinia) in the South-Atlantic Ocean. PeerJ, 5:e3873 Cinner, J.E., Mcclanahan, T.R., Macneil, M.A, Graham, N.A.J., Daw, T.M., Mukminin, A., Feary, D.A., Rabearisoa, A.L., Wamukota, A., Jiddawi, N., & Campbell, S.J. (2012). Comanagement of coral reef social-ecological systems. PNAS, 109, 5219–5222. Cinner, J.E., Graham, N.J., Huchery, C., & Macneil, M.A. (2013). Global effects of local human population density and distance to markets on the condition of coral reef fisheries. Conservation Biology, 27, 453–458.

Consoli, P., Esposito, V., Battaglia, P., Altobelli, C., Perzia, P., Romeo, T., Canese, S., & Andaloro, F. (2016). Fish distribution and habitat complexity on Banks of the Strait of Sicily (Central Mediterranean Sea) from Remotely-Operated Vehicle (ROV) Explorations. PLoS ONE, 11(12), e0167809. https://doi.org/10.1371/journal.pone.0167809 Copertino, M. S., Connell, S. D. & Cheshire, A. (2005). Prevalence and production of turf-forming algae on a temperate subtidal coast. Phycologia, 43, 241–8. Cowman, P.F., Parravicini, V., Kulbicki, M. & Floeter, S.R. (2017). The biogeography of tropical reef fishes: endemism and provinciality through time. Biological Reviews, 92, 2112–2130. Creed, J.C., Fenner, D., Sammarco, P., Cairns, S., Junqueira, A.O.R., Cruz, I., Marcelo, C., Mantelatto, C., Le, R.J.M., & Oigman- pszczol, S. (2017). The invasion of the azooxanthellate coral Tubastraea (Scleractinia: Dendrophylliidae) throughout the world: history, pathways and vectors. Biological Invasions, 19, 283–305. Crane, N.L., Paddack, M., Nelson, P., Abelson, A., Precoda, K., Rulmal, J.J., & Bernardi, G. (2017). Atoll-scale patterns in coral reef community structure: human signatures on Ulithi Atoll, Micronesia. PLoS ONE, 12(5), e0177083. https://doi.org/10.1371/journal.pone.0177083 De Paula, A.F., & Creed J.C., (2004). Two species of the coral Tubastraea (Cnidaria, Scleractinia) in Brazil: a case of accidental introduction. Bulletin of Marine Science, 74, 175–183. Denny MW, Daniel TL, & Koehl MAR (1985). Mechanical Limits to Size in Wave-Swept Organisms. Ecological Society of America, 55, 69– 102. Dominici-Arosemena, A., & Wolff, M. (2006). Reef fish community structure in the Tropical Eastern Pacific (Panamá): living on a relatively stable rocky reef environment. Helgoland Marine Research, 60, 287–305. Dulvy, N.K., Freckleton, R.P., & Polunin, N.V.C. (2004). Coral reef cascades and the indirect effects of predator removal by exploitation. Ecology Letters, 7, 410–416. Dustan, P., Doherty, O., & Pardede, S. (2013). Digital reef rugosity estimates coral reef habitat complexity. PLoS ONE, 8(2), e57386. https://doi.org/10.1371/journal.pone.0057386 Edgar, G.J., Banks, S.A., Bessudo, S., Cortés, J., Guzmán, H.M., Henderson, S., Martinez, C., Rivera, F., Soler, G., Ruiz, D., & 78

Zapata, F.A. (2011). Variation in reef fish and invertebrate communities with level of protection from fishing across the Eastern Tropical. Global Ecology and Biogeography, 20, 730–743. Eriksson, B. K., Johansson, G. & Snoeijs, P. (2002). Long-term changes in the macroalgal vegetation of the inner Gullmar Fjord, Swedish Skagerrak coast. Journal of Phycology, 38, 284–96. Fautin, D.G., & Allen, G.R. (1997). Anemonefishes and their host sea anemones. Perth: Western Australian Museum. Ferreira, C.E.L., Floeter, S.R., Gasparini, J.L., Ferreira, B.P., & Joyeux J.C. (2004). Trophic structure patterns of Brazilian reef fishes: a latitudinal comparison. Journal of Biogeography, 31, 1093–1106. Ferreira, C.E.L., Gonçalves, J.E.A., & Coutinho, R. (2001). Community structure of shes and habitat complexity on a tropical rocky shore. Environmental Biology of Fishes, 61, 353–369. Ferretti F, Worm B, Britten G.L, Heithaus MR & Lotze HK (2010). Patterns and ecosystem consequences of shark declines in the ocean. Ecology Letters, 13, 1055–1071. Floeter, S.R., Guimarães, R. Z. P., Rocha, L. A, Ferreira, CEL, Rangel, C.A., & Gasparini, J. L. (2001). Geographic variation in reef-fish assemblages along the Brazilian coast. Global Ecology and Biogeography, 10, 423–431. Floeter, S.R., Behrens, M. D., Ferreira, C. E. L., Paddack, M. J., & Horn, M. H. (2005). Geographical gradients of marine herbivorous fishes: patterns and processes. Marine Biology, 147, 1435–1447. Floeter, S.R., Krohling, W., Gasparini, J.L., Ferreira, C.E.L., & Zalmon, I.R. (2007). Reef fish community structure on coastal islands of the southeastern Brazil: the influence of exposure and benthic cover. Environmental Biology of Fishes, 78, 147–160. Floeter, S.R., Rocha, L.A., Robertson, D.R., Joyeux, J.C., Smith-Vaniz, W.F., Wirtz, P., Edwards, A.J., Barreiros, J.P., Ferreira, C.E.L., Gasparini, J.L., Brito, A., Falcón, J.M., Bowen, B.W., & Bernardi, G. (2008). Atlantic reef fish biogeography and evolution. Journal of Biogeography, 35, 22–47. Friedlander, A.M, & DeMartini, E.E. (2002). Contrasts in density, size, and biomass of reef fishes between the northwestern and the main Hawaiian islands: The effects of fishing down apex predators. Marine Ecology Progress Series, 230, 253–264. Friedlander, A.M., Brown, E.K., Jokiel, P.L., Smith, W.R., & Rodgers, K.S. (2003). Effects of habitat, wave exposure, and marine protected area status on coral reef fish assemblages in the Hawaiian archipelago. Coral Reefs, 22, 291–305.

Friedlander, A.M., Sandin, S.A., DeMartini, E.E., & Sala, E. (2010). Spatial patterns of the structure of reef fish assemblages at a pristine atoll in the central Pacific. Marine Ecology Progress Series, 410, 219–231. Friedlander, A.M, Ballesteros, E., Beets, J., Berkenpas, E., Gaymer, C.F., Gorny, M., & Sala, E. (2013). Effects of isolation and fishing on the marine ecosystems of Easter Island and Salas y Gómez, Chile. Aquatic Conservation: Marine and Freshwater Ecosystems, 531, 515–531. Friedlander, A.M., Ballesteros, E., Fay, M., & Sala, E. (2014). Marine communities on oil platforms in Gabon, West Africa: high biodiversity oases in a low biodiversity environment. PLoS ONE, 9(8): e103709. https://doi.org/10.1371/journal.pone.0103709 Friedlander, A.M., Obura, D., Aumeeruddy, R., Ballesteros, E., Church, J., Cebrian, E., & Sala, E. (2014). Coexistence of Low Coral Cover and High Fish Biomass a Farquhar Atoll, Seychelles. PLoS ONE, 9(1), e87359. https://doi.org/10.1371/journal.pone.0087359 Friedlander, A.M., Ballesteros, E., Clemente, S., Gonc, E.J., Estep, A., Rose, P., & Sala, E. (2017). Contrasts in the marine ecosystem of two Macaronesian islands : A comparison between the remote Selvagens Reserve and Madeira Island. PLoS ONE, 12(11): e0187935. https://doi.org/10.1371/journal.pone.0187935 Froese, R., & Pauly, D. (2016). FishBase. World Wide Web electronic publication. Available: www.fishbase.org. Accessed 2016 Jun 13. Fulton, C. J., & Bellwood, D. R. (2004). Wave exposure, swimming performance, and the structure of tropical and temperate reef fish assemblages. Marine Biology, 144, 429–437. Fulton, C.J., Bellwood, D.R., & Wainwright, P.C. (2005). Wave energy and swimming performance shape coral reef fish assemblages. The Royal Society B, 272, 827–832. Garpe, K.C., & Öhman, M.C. (2003). Coral and fish distribution patterns in Mafia Island marine park, Tanzania: fish habitat interactions. Hydrobiologia, 498, 191–211. García-Charton, J. A., Pérez-Ruzafa, Á., Sánchez-Jerez, P., Bayle- Sempere, J. T., Reñ ones, O. & Moreno, D. (2004). Multi-scale spatial heterogeneity, habitat structure, and the effect of marine reserves on Western Mediterranean rocky reef fish assemblages. Marine Biology, 144, 161–182. Gili, J.M., & Coma, R. (1998). Benthic suspension feeders: their paramount role in littoral marine food webs. Trends in Ecology & 80

Evolution, 13, 316–321. Goatley, C.H.R., & Bellwood, D.R. (2011). The Roles of Dimensionality, Canopies and Complexity in Ecosystem Monitoring. PLoS ONE, 6(11), e27307. https://doi.org/10.1371/journal.pone.0027307. Gove, J.M, McManus, M.A., Neuheimer, A.B, Polovina, J.J, Drazen, J.C, Smith, C.R,Merrifield, M.A., Friedlander, A.M, Ehses, J.S, Young, C., Dillon, A.K, &Williams, G. (2016). Near-island biological hotspots in barren ocean basins. Nature communications, 7, 1–34. Gorgula, S. K. & Connell, S. D. (2004). Expansive covers of turf-forming algae on human-dominated coast: the relative effects of increasing nutrient and sediment loads. Marine Biology, 145, 613–9. Jones, C.G., Lawton, J.H., & Shachak, M., (1994). Organisms as ecosystem engineers. Oikos, 69, 373–386. Graham, N.A.J., Spalding, M.D., Sheppard, C.R.C. (2010). Reef shark declines in remote atolls highlight the need for multi-faceted conservation action. Aquatic Conservation: Marine and Freshwater Ecosystems, 548, 543–548. Graham, N.A.J., Bellwood, D.R., Cinner, J.E., Hughes, T.P., Norström, A.V., &Nyström, M. (2013). Managing resilience to reverse phase shifts in coral reefs. Frontiers in Ecology and the Environment, 11, 541–548. Graham, N.A.J., &Mcclanahan, T.R. (2013). The last call for marine wilderness? Bioscience, 63, 397–402. Gratwicke, B., & Speight, M.R. (2005). The relationship between fish species richness, abundance and habitat complexity in a range of shallow tropical marine habitats. Journal of Fish Biology, 66, 650– 667. Hachich, N.F., Bonsall, M.B., Arraut, E.M., Barneche, D.R., Lewinsohn, T.M., & Floeter, S.R. (2015). Island biogeography: patterns of marine shallow-water organisms in the Atlantic Ocean. Journal Biogeography, 45, 1871–1882. Hachich, N.F., Bonsall, M.B., Arraut, E.M., Barneche, D.R., Lewinsohn, T.M., & Floeter, S.R. (2016). Marine island biogeography. Response to comment on ‘Island biogeography: patterns of marine shallow-water organisms’. Journal of Biogeography, 43, 2517– 2519. Heyns-Veale, E., Bernard, A.T., Parker, D., Richoux, N.B., Langlois, T., Harvey, E.S., &Gotz, A. (2016). Depth and habitat determine community structure of South Africa’s warm-temperate reef fish. Marine Biology, 163, 1–17. Heck, K.L.J., &Wetstone, G.S., (1977). Habitat complexity and

invertebrate species richness and abundance in tropical seagrass. Journal of Biogeography, 4, 135–142. Hothorn, T., Bretz, F., & Peter, W. (2008). Simultaneous inference in general parametric models. Biometrical Journal, 20, 346–363. Horn, H.S., J.T. Bonner, W. Dohle, M.J. Katz, M. A. R. Koehl, H. Meinhardt, R.A. Raff, W.E. Reif, S. C. Stearns, & R. Strathmann. (1982). Adaptive aspects of development. Pages 215-235 in J. T. Bonner, editor. Evolution and development. Dahlem Konferenzen 1982. Springer-Verlag, New York, New York, USA. Jones, G.P., &Syms, C. (1998). Disturbance, habitat structure and the ecology of fishes on coral reefs. Australian Journal of Ecology, 23, 287–297. Jones, P., &Tye, A. (1996). The Birds of São Tomé and Príncipe, with Annobón: islands of the Gulf of Guinea. (B Ornithologists, Ed.). Oxford. Jones. G.P., & Syms, G. (1998). Disturbance, habitat structure and the ecology of fishes on coral reefs. Australian Journal of Ecology, 23, 287–297. Khalil, M.T., Bouwmeester, J., & Berumen, M.L. (2017). Spatial variation in coral reef fish and benthic communities in the central Saudi Arabian Red Sea. PeerJ, 5, e3410. https://doi.org/10.7717/peerj.3410 Komyakova, V., Munday, P.L., &Jones, G.P. (2013). Relative importance of coral cover, habitat complexity and diversity in determining the structure of reef fish communities. PLoS ONE, 8(12), e83178. https://doi.org/10.1371/journal.pone.0083178 Krajewski, J.P., & Floeter, S.R. (2011). Reef fish community structure of the Fernando de Noronha Archipelago (Equatorial Western Atlantic): the influence of exposure and benthic composition. Environmental Biology of Fishes, 92, 25–40. Krakstad, J.O., Alvheim, O., & Lopes, J.S.D. (2010).Surveys of the resources of São Tomé & Príncipe. Survey of the demersal resources 10 May–20 May 2010. Institute of Marine Research. Technical Report. Available at http://hdl.handle.net/11250/107339. (Accessed 20/09/2015). Kulbicki, M., Parravicini, V., Bellwood, D.R. Arias-Gonzàlez, E., Chabanet, P., Floeter, S.R., Friedlander, A., McPherson, J., Myers, R.E., Vigliola, L., & Mouillot, D.,(2013). Global biogeography of reef fishes: a hierarchical quantitative delineation of regions. PLoS ONE, 8(12), e81847. https://doi.org/10.1371/journal.pone.0081847 82

Kulbicki, M., Parravicini, Valeriano, & Mouillot, D. (2015). Patterns and processes in reef fish body size. In: Mora, Camilo, (Ed.) Ecology of Fishes on Coral Reefs. Cambridge University Press, Cambridge, UK, pp. 104–115. Klumpp, D.W, &Mckinnon A.D. (1989). Temporal and spatial patterns in primary production of a coral-reef epilithic algal community. Journal of Experimental Marine Biology and Ecology, 131, 1–22. Laborel, J. (1974). West African reefs corals: an hypothesis on their origin. Proceedings of the 2nd. International Coral Reef Symposium, 1, 425–443. Leão, Z.M., Kikuchi, R., & Testa, V. (2003). Corals and coral reefs of Brazil. Latin American Coral Reefs (ed. by J. Cortés), pp. 9–52. Elsevier, Amsterdam. Lee, D.C., Halliday, A.N., Fitton, J.G., & Poli, G. (1994). Isotopic variations with distance and time in the volcanic islands of the Cameroon line: evidence for a mantle plume origin. Earth Planet. Earth and Planetary Science Letters, 123, 119–138. Legendre, P., &Legendre, L.F. (2012). Numerical ecology (Vol. 24). Elsevier. Litsios, G., Sims, C.A., Wuest, R.O., Pearman, P.B., Zimmermann, N.E., & Salamin, N. (2012). Mutualism with sea anemones triggered the adaptive radiation of clownfishes. BMC Evolutionary Biology, 12– 212. https://doi.org/10.1186/1471-2148-12-212 Littler, M.M, &Littler D.S. (1984). Relationships between macroalgal functional form groups and substrata stability in a subtropical rocky-intertidal system. Journal of Experimental Marine Biology and Ecology, 74, 13–34. Littler, M.M. & Littler, D.S. (1988). Structure and role of algae in tropical reef communities. In Lembi, C.A. & Waaland, J.R. [Eds.] Algae and Human Affairs. Vol. 28. Cambridge University Press, Cambridge, pp. 29–56. Longo, G.O, Morais, R.A, Martins, C.D.L, Mendes, T.C, Aued, A.W, Cândido, D.V., Oliveira, J.C., Nunes, L.T, Fontoura, L., Sissini, M.N, Teschima, M.M., Silva, M.B, Ramlov, F., Gouvea, L.P, Ferreira, C.E.L., Segal, B., Horta P.A, & Floeter S.R. (2015). Between habitat variation of benthic cover, reef fish assemblage and feeding pressure on the benthos at the only atoll in south Atlantic: Rocas atoll, NE Brazil. PLoS ONE, 10(6), e0127176. https://doi.org/10.1371/journal.pone.0127176 Luckhurst, E., &Luckhurst, K. (1978). Analysis of the influence of substrate variables on coral reef fish communities. Marine Biology,

323, 317–323. Luiz, O.J., & Edwards A.J. (2011). Extinction of a shark population in the Archipelago of Saint Paul’s Rocks (equatorial Atlantic) inferred from the historical record. Biology Conservation, 144, 2873–2881. Luiz, O.J., Mendes, T.C., Barneche, D.R., Ferreira, C.G.W., Noguchi, R., Villaça, R.C., Rangel, C.A., Gasparini, J.L., & Ferreira, C.E.L. (2015). Community structure of reef fishes on a remote oceanic island (St Peter and St Paul’s Archipelago, equatorial. Atlantic): the relative influence of abiotic and biotic variables. Marine andFreshwater Research, 66, 739–749. MacArthur, R.H. & Wilson, E.O. (1967). The theory of island biogeography. University Press, Princeton. Maia, H.A., Morais, R.A., Siqueira, A.C., Hanazaki, N., Floeter, S.R., & Bender, M.G., (2018). Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea. Ocean and Coastal Management. https://doi.org/10.1016/ j.ocecoaman.2018.01.006 Measey, J. G., Vences, M., Drewes, R.C., Chiari, Y., Melo, M., & Bourles, B. (2007). Freshwater paths across the ocean: Molecular phylogeny of the frog Ptychadena newtoni gives insights into amphibian colonization of oceanic islands. Journal Biogeography, 34, 7–20. Myers, R.A., Baum, J.K., Shepherd, T.D., Powers, S.P., & Peterson, C.H. (2007). Cascading effects of the loss of apex predatory sharks from a coastal ocean. Science, 315, 1846–1850. McGehee, M.A. (1994). Correspondence between assemblages of coral reef fishes and gradients of water motion, depth and substrate size off Puerto Rico. Marine Ecology Progress Series, 105, 243–255. Mouillot, D., Villéger, S., Parravicini, V., Kulbicki, M., Arias-Gonzales, J.E., Bender, M.G., Chabanet, P., Floeter, S.R., Friedlander, A., Vigliola, L., & Bellwood, D.R. (2014). Functional over-redundancy and high functional vulnerability in global fish faunas of tropical reefs. Proceedings of the National Academy of Sciences of the United States of America, 111, 13757–13762. Mora, C., Andrefouet, S., Costello, M.J., Kranenburg, C., Rollo, A., Veron, J., Gaston, K.J., & Myers, R.A. (2006). Coral reefs and the global network of marine protected areas. Science, 312, 1750–1751. Mora, C, Aburto-Oropeza, O, Ayala, B.A., Ayotte, P.M., Banks, S., Bauman, A.G., Beger, M., Bessudo, S., Booth, D.J., Brokovich, E., Brooks, A., Chabanet, P., Cinner, J.E., Cortés, J., Cruz-Motta, J.J., Cupul, M.A., Demartini, E.E., Edgar, G.J., Feary, D.A., Ferse, 84

S.C.A., Friedlander, A.M., Gaston, K.J., Gough, C., Graham, N.A.J., Green, A., Guzman, H., Hardt, M., Kulbicki, M., Letourneur, Y., López Pérez, A., Loreau, M., Loya, Y., Martinez, C., Mascareñas-Osorio, I., Morove, T., Nadon, M.O., Nakamura, Y., Paredes, G., Polunin, N.V.C., Pratchett, M.S., Reyes, B.H., Rivera, F., Sala, E., Sandin, S.A., Soler, G., Stuart-Smith, R., Tessier, E., Tittensor, D.P., Tupper, M., Usseglio, P., Vigliola, L., Wantiez, L., Williams, I., Wilson, S.K., & Zapata, F.A. (2011). Global human footprint on the linkage between biodiversity and ecosystem functioning in reef fishes. PLoS BIOLOGY, 9(4), e1000606. https://doi.org/10.1371/journal.pbio.1000606 Morais, R.A., & Maia H.A. (2016). Lush underwater forests in mesophotic reefs of the Gulf of Guinea. Coral Reefs, 36, 95–95. Mumby, P.J., Steneck, R.S., & Hastings, A. (2013). Evidence for and against the existence of alternate attractors on coral reefs. Oikos, 122, 481–49. Munday, P.L. (2000). Interactions between habitat use and patterns of abundance in coral dwelling fishes of the genus Gobiodon. Environmental Biology of Fishes, 58, 355–369. Newman, M.J.H., Paredes, G.A., Sala, E., & Jackson, J.B.C. (2006). Structure of Caribbean coral reef communities across a large gradient of fish biomass. Ecology Letters, 9, 1216–1227. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, H.H., & Wagner, H. (2015). vegan: Community Ecology Package. R package version 2.3-4. : https://CRAN.R- project.org/package=vegan. Parravicini, V., Kulbicki, M., Bellwood, D.R., Friedlander, A.M., Chabanet, P., Floeter, S.R., Myers, R., Vigliola, L., Agata, S.D., & Mouillot, D. (2013). Global patterns and predictors of tropical reef fish species richness. Ecography, 37, 1254–1262. Payri, C.E. (1995). Production carbonatée de quelques algues calcifiées sur un récif corallien en Polynésie française. Bulletin de la Société Géologique de France, 166, 74–84. Payri, C. E. (1997). Hydrolithon reinboldii rhodolith distribution, growth and carbon production of a French polynesian reef. Proc. 8th Int. Coral Reef Symposium, 1, 755–60. Pinheiro, H.T., Ferreira, C.E.L., Joyeux, J.C., Santos, R.G., & Horta P.A. (2011).Reef fish structure and distribution in a south western Atlantic Ocean tropical island. Journal Fish Biology, 79, 1984– 2006.

Polidoro, B.A., Carpenter, K.E., Ralph, G.M., Strongin, K., Harvey, M., Arnold, R., Buchanan, J.R., Mohamed, K., Camara, A., Collette, B.B., Comeros, M.T., Bruyne, G.D, Gon, O., Harold, A.S., Harwell, H., Hulley, P.A., Iwamoto, T., Knudsen, S.W., Dieu, J. D., Linardich, C., Lindeman, K.C., Monteiro, V., Munroe, T., Nunoo, F.K.E., Pollock, C.M., Poss, S., Russell, B., Sayer, C., Sidibe, A., Smith, W., Emilie, V., De L.T., & Christophe, J., (2017). The status of marine biodiversity in the Eastern Central Atlantic (West and Central Africa). Aquatic Conservation: Marine and Freshwater Ecosystems, 27, 1021–1034. Quimbayo, J.P., Floeter, S.R., Noguchi, R., Rangel, C.A., Gasparini, J.L., Sampai, C.L.S., Ferreira, C.E.L., & Rocha, L.A. (2012). Cleaning mutualism in Santa Luzia (Cape Verde Archipelago) and São Tomé islands, Tropical Eastern Atlantic. Marine Biodiversity Records, 5, 1–7. Quimbayo, J.P., Mendes, T.C., Kulbicki, M., Floeter, S.R., & Zapata, F.A. (2017) Unusual reef fish biomass and functional richness in a remote island in the Tropical Eastern Pacific. Environmental Biology of Fishes, 100, 149–162. RDSTP (República Democrática de São Tomé e Príncipe), Relatório nacional de avaliação da estratégia de implementação das Maurícias (2009). São Tomé e Príncipe. Technical Report. Available at https://sustainabledevelopment.un.org/content/documents/1300Sao _Tome_and_Principe-MSI-NAR2009.pdf, Accessed 28 June 2017. Ricciardi, F., Boyer, M., & Ollerton, J. (2010). Assemblage and interaction structure of the anemonefish-anemone mutualism across the Manado region of Sulawesi. Indonesia Environmental Biology of Fishes, 87, 333–347. Roff, G, & Mumby, P.J. (2012). Global disparity in the resilience of coral reefs. Trends in Ecology & Evolution, 27, 404–413. Russell, B.C., (1977). Population and standing crop estimates for rocky reef fishes of North-Eastern New Zealand, New Zealand. New Zealand Journal of Marine and Freshwater Research, 11, 23–36. Sandin, S.A., Smith, J.E., Demartini, E.E., Dinsdale, E.A., Donner, S.D., Friedlander, A.M., Konotchick, T., Malay, M., Maragos, J.E., Obura, D., Pantos, O., Paulay, G., Richie, M., Rohwer, F., Schroeder, R.E., Walsh, S., Jackson, J.B.C., Knowlton, N., & Sala E. (2008). Baselines and degradation of coral reefs in the Northern Line Islands. PLoS ONE, 3(2):e1548. doi:/10.1371/journal.pone.0001548. 86

Scheffer, M., Carpenter, S., Foley, J.A., Folke, C., & Walker, B. (2001). Catastrophic shifts inecosystems. Nature, 413, 591–596. Serigstad, B., Ensrud, T.M, Olsen, M, Ostrowski, M, &Dias, J. (2012). Marine environmentalsurvey of bottom fauna, selected physical and chemical compounds and fisheriessurvey in the Joint Development Zone between Nigeria an Sao Thome & Principe. Institute of Marine Research. Technical Report. Available at http://hdl.handle.net/11250/107442. (Accessed 20/09/2015). Silva, A.G., Lima, R.P., Gomes, A.N., Fleury, B.G., & Creed, J.C. (2011). Expansion of the invasive corals tubastraea coccinea and tubastraea tagusensis into the tamoios ecological station marine protected area, Brazil. Aquatic Invasions, 6, 105–110. Souza, J.N., Nunes, F.L.D., Zilberberg, C., Sanchez, J.A., Migotto, A.E., Hoeksema, B.W., Serrano, X.M., Baker, A.C., & Lindner, A. (2017). Contrasting patterns of connectivity among endemic and widespread fire coral species (Millepora spp.) in the tropical Southwestern Atlantic. Coral Reefs, 36, 701–716. Steneck, R.S. (1986). The ecology of coralline algal crust: convergent patterns and adaptative strategies. Annual Review of Ecology and Systematics, 17, 273–303. Steneck, R.S., &Dethier, M.N. (1994). A functional group approach to the structure of algal dominated communities. Oikos, 69, 476–498. Steneck, R.S. (1997). Crustose corallines, other algal functional groups and sediments: complex interactions along reef productivity gradients. Proc. 8th Int. Coral Reef Symposium, 1, 695–700. Szmant, A.M. (2002). Nutrient enrichment on coral reefs: Is it a major cause of coral reef decline? Estuaries and Coasts, 25, 743–766. Tuya, F.A., Bosch, N.E.A., Abreu, A.D.B., & Haroun, R.A., (2017). Reef fish at a remote tropical island (Principe Island, Gulf of Guinea): disentangling taxonomic, functional and phylogenetic diversity patterns with depth. Marine and Freshwater Research. https://doi.org/10.1071/MF17233. Trebilco, R., Baum, J.K., Salomon, A.K., & Dulvy, N.K. (2013). Ecosystem ecology: Size-based constraints on the pyramids of life. Trends in Ecology & Evolution, 28, 423–431. Truong, L., Suthers, I.M., Cruz, D.O., & Smith, J.A. (2017). Plankton supports the majority of fish biomass on temperate rocky reefs. Marine Biology, 73, 1–12. Underwood, A.J., Kingsford, M.J. & Andrew, N. L. (1991). Patterns in shallow subtidal marine assemblages along the coast of New South Wales. Australian Journal of Ecology, 6, 231–49.

Wainwright, P.C., Bellwood, D.R., &Westneat, M.W. (2002). Ecomorphology of locomotion in labrid fishes. Environmental Biology of Fishes, 65, 47–62. Wilding, T.A., Palmer, E.J.L., & Polunin N.V.C. (2010). Comparison of three methods for quantifying topographic complexity on rocky shores. Marine Environmental Research, 69, 143–151. Williams, I.D., Baum, J.K., Heenan, A., Hanson, K.M., Nadon, M.O., & Brainard, R.E. (2015). Human, oceanographic and habitat drivers of central and Western Pacific coral reeffish assemblages. PLoS ONE, 10(4), e0120516. https://doi:10.1371/journal.pone.0120516. Williams, G.J., Smith, J.E., Conklin, E.J., Gove, J.M., Sala, E., & Sandin, S.A. (2013). Benthic communities at two remote Pacific coral reefs: effects of reef habitat, depth, and wave energy gradients on spatial patterns. PeerJ, 81, 1–26. Wilson, S.K., Graham, N.A.J., & Polunin, N.V.C. (2007). Appraisal of visual assessments of habitat complexity and benthic composition on coral reefs. Marine Biology, 151, 1069–1076. Wirtz, P. (2003). New records of marine invertebrates from São Tomé island (Gulf of Guinea). Journal of the Marine Biological Association of the United Kingdom, 83, 735–736. Wirtz, P., Ferreira, C.E.L., Floeter, S.R., Fricke, R., Gasparini, J.L., Iwamoto, T., Rocha, L.A., &Sampaio, C.L.S. (2007). Coastal fishes of São Tomé and Principe islands, Gulf of Guinea (Eastern Atlantic Ocean) — an update. Zootaxa, 1523, 1–48. Wisz, M.S., Pottier, J., Kissling, W.D., Pellissier, L., Lenoir, J., Damgaard, C.F., Dormann, C.F., Forchhammer, M.C., Grytnes, J.A., Guisan, A., Heikkinen, R.K., Hoye, T.T., Kühn, I., Luoto, M., Maiorano, L., Nilsson, M.C., Normand, S., Öckinger, E., Schmidt, N.M., Termansen, M., Timmermann, A., Wardle, D.A., Aastrup, P., & Svenning, J.C. (2013). The role of biotic interactions in shaping distributions and realised assemblages of species: Implications for species distribution modelling. Biological Reviews, 88, 15–30. Zuur, A., Ieno, E.N., Walker, N., Saveliev, A.A., & Smith, G.M. (2009). Statistics for biology and health mixed effects models and extensions in ecology with R. Springer, Netherlands.

Supporting information

Figure S.1: Comparative variation in species density (A), abundance total (B) and biomass total (C) of fish assemblages sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), mean (red diamond), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. Letters show statistical groupings (Tukey post-hoc) and boxplots with different letters are significantly different. Sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras catedral (SPC), Sete pedras branca (SPB) and Ilhéu das rolas (ROL).

Figure S.2: Compative variation of abundance and biomass found in São Tomé island. (A) Abundance per trophic groups. (B) Biomass per trophic groups. (C) Abundance per size classes. (D) Biomass per size classes. Each point represents UVCs. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Figure S.3: Variative variation in total abundance of fish assemblages per trophic groups sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites. Trophic groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Figure S.4: Comparative variation in abundance total of fish assemblages per size classes sampled in 2006 at the São Tomé island. Boxplots denote medians (black line), upper and lower quartiles, and 95% confidence intervals. Each circle represents a underwater visual censuses. See Fig. 1 for the acronym of the sites.

Figure S.5: Percentage of benthic cover sampling at São Tomé island. Each point represents a photo-quadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), Other life (OTH) and Gorgonians (GOR).

Table S.1: Abundance (individuals per m2 ± S.E.) and biomass (individuals in grams per m2 ± S.E.) estimates of fish species recorded during underwater visual census (139 transects) in São Tomé island per sites and general. Trophic group: herbivores- detritivorous (HD), macroalgal-feeder (HM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), piscivorous (PS) and omnivores (OM).

DVA KIA SAN Family & Species DIET Abundance biomass Abundance biomass Abundance biomass

Acanthuridae Acanthurus monroviae** HD 0.01 ± 0.00 0.11 ± 0.10 0.06 ± 0.02 5.86 ± 5.72 0.05 ± 0.02 2.08 ± 0.73 Prionurus biafraensis** HD 0.05 ± 0.02 14.14 ± 7.99 0.00 ± 0.00 <0.01 0.03 ± 0.02 2.17 ± 1.19 Apogonidae Apogon affinis*** PK <0.01 <0.01 0.57 ± 0.41 1.16 ± 0.83 <0.01 <0.01 Apogon imberbis**** PK <0.01 <0.01 0.81 ± 0.29 1.65 ± 0.60 <0.01 <0.01 Aulostomidae Aulostomus strigosus** PS 0.18 ± 0.04 2.16 ± 0.54 0.06 ± 0.01 0.08 ± 0.05 0.01 ± 0.00 <0.01 Balistidae Balistes capriscus*** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Balistes punctatus** IM 0.01 ± 0.00 0.36 ± 0.23 <0.01 0.13 ± 0.12 <0.01 0.01 ± 0.01 Canthidermis sufflamen*** PK <0.01 <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 Blenniidae Ophioblennius sp* HD 0.03 ± 0.01 0.09 ± 0.04 0.04 ± 0.01 0.06 ± 0.01 0.08 ± 0.02 0.47 ± 0.29

Bothidae Bothus lunatus*** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Carangidae Caranx crysos**** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Caranx latus*** IM 0.01 ± 0.01 0.13 ± 0.11 <0.01 <0.01 <0.01 <0.01 Centracanthidae Spicara nigricauda** PK 0.03 ± 0.03 0.07 ± 0.07 0.02 ± 0.02 1.15 ± 1.15 0.55 ± 0.33 32.87 ± 21.13 Chaetodontidae Chaetodon robustus** IM <0.01 <0.01 0.01 ± 0.00 0.29 ± 0.20 0.01 ± 0.00 0.59 ± 0.29 Prognathodes marcellae** IM 0.02 ± 0.01 0.07 ± 0.03 <0.01 <0.01 <0.01 <0.01 Cirrhitidae Cirrhitus atlanticus** IM 0.01 ± 0.00 0.27 ± 0.08 0.01 ± 0.00 0.15 ± 0.06 0.04 ± 0.01 0.55 ± 0.16 Diodontidae Diodon holocanthus***** IM <0.01 <0.01 0.01 ± 0.00 0.57 ± 0.25 <0.01 <0.01 Gobiidae Gnatholepis thompsoni*** OM <0.01 <0.01 0.03 ± 0.01 0.05 ± 0.01 0.06 ± 0.02 0.09 ± 0.03 Gobius rubropunctatus** IM <0.01 <0.01 0.25 ± 0.03 1.04 ± 0.25 0.02 ± 0.01 0.04 ± 0.01 Gorogobius nigricinctus** IM <0.01 <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 gorogobius stevcici* IM <0.01 <0.01 <0.01 <0.01 0.04 ± 0.02 0.07 ± 0.03 Wheelerigobius wirtzi** IM 0.00 ± 0.00 <0.01 0.05 ± 0.02 0.08 ± 0.03 0.05 ± 0.02 0.09 ± 0.03

Didogobius amicuscaridis* IM <0.01 0.01 ± 0.00 <0.01 <0.01 <0.01 <0.01 Haemulidae Pomadasys incisus** IM 0.13 ± 0.04 8.94 ± 3.06 <0.01 <0.01 <0.01 <0.01 Holocentridae Holocentrus adscensionis*** IM <0.01 <0.01 0.30 ± 0.07 4.32 ± 1.05 <0.01 <0.01 223.65 ± Myripristis jacobus*** PK 0.22 ± 0.10 11.46 ± 4.66 5.05 ± 1.18 0.38 ± 0.15 20.00 ± 6.81 73.27 Sargocentron hastatum** IM 0.01 ± 0.00 0.87 ± 0.27 0.52 ± 0.45 23.15 ± 23.04 0.05 ± 0.03 2.39 ± 1.69 Kyphosidae Kyphosus spp*** HD 0.01 ± 0.01 2.59 ± 1.54 <0.01 <0.01 <0.01 0.79 ± 0.79 Labridae Bodianus pulchellus*** IM 0.1 ± 0.03 5.40 ± 1.66 0.02 ± 0.01 0.66 ± 0.34 <0.01 <0.01 Bodianus speciosus** IM 0.01 ± 0.00 0.46 ± 0.32 0.02 ± 0.01 0.20 ± 0.10 0.01 ± 0.01 0.26 ± 0.17 Clepticus africanus* PK 0.02 ± 0.01 2.44 ± 1.21 <0.01 <0.01 <0.01 <0.01 Coris atlantica** IM <0.01 0.06 ± 0.04 <0.01 0.03 ± 0.03 <0.01 0.03 ± 0.03 Thalassoma newtoni* PK 0.07 ± 0.02 0.41 ± 0.12 0.51 ± 0.07 2.22 ± 0.81 0.14 ± 0.03 0.57 ± 0.20 Lethrinidae Lethrinus atlanticus** IM <0.01 <0.01 <0.01 0.17 ± 0.17 0.07 ± 0.04 4.06 ± 2.30 Lutjanidae Apsilus fuscus** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Lutjanus fulgens** IM 0.03 ± 0.03 0.18 ± 0.11 0.03 ± 0.03 1.23 ± 1.23 <0.01 <0.01 96

Lutjanus goreensis** IM 0.01 ± 0.00 0.59 ± 0.35 0.02 ± 0.01 1.75 ± 0.74 <0.01 0.55 ± 0.55 Monacanthidae Aluterus scriptus***** PK 0.02 ± 0.01 4.61 ± 2.74 <0.01 <0.01 <0.01 <0.01 Cantherhines pullus*** IM 0.03 ± 0.01 0.64 ± 0.30 0.02 ± 0.00 0.61 ± 0.24 0.02 ± 0.01 0.57 ± 0.21 Mullidae Mulloidichthys IM 0.01 ± 0.01 0.47 ± 0.31 0.13 ± 0.06 4.18 ± 3.00 0.06 ± 0.05 3.20 ± 3.07 martinicus*** Pseudupeneus prayensis** IM 0.08 ± 0.05 0.14 ± 0.07 0.10 ± 0.02 2.75 ± 2.20 0.02 ± 0.01 0.06 ± 0.04 Muraenidae Channomuraena PS <0.01 0.03 ± 0.03 <0.01 0.13 ± 0.09 <0.01 <0.01 vittata***** Enchelycore nigricans*** IM <0.01 0.06 ± 0.04 0.01 ± 0.00 0.73 ± 0.26 <0.01 0.04 ± 0.04 Gymnothorax vicinus*** PS <0.01 <0.01 0.01 ± 0.01 20.22 ± 13.66 <0.01 1.40 ± 1.40 Muraena melanotis*** PS <0.01 0.09 ± 0.04 0.01 ± 0.00 8.74 ± 5.50 <0.01 6.39 ± 2.63 Muraena robusta*** PS <0.01 0.02 ± 0.02 <0.01 0.25 ± 0.14 <0.01 <0.01 Ophichthidae Myrichthys pardalis** IM <0.01 <0.01 <0.01 0.01 ± 0.01 <0.01 0.01 ± 0.01 Ostraciidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Acanthostracion IS <0.01 0.05 ± 0.05 <0.01 <0.01 <0.01 <0.01 notacanthus*** Pomacanthidae Holacanthus africanus** IS <0.01 0.06 ± 0.06 <0.01 0.02 ± 0.01 0.01 ± 0.00 0.18 ± 0.11

Pomacentridae Abudefduf saxatilis*** OM 0.01 ± 0.01 0.1 ± 0.05 <0.01 <0.01 0.02 ± 0.01 0.83 ± 0.54 Abudefduf taurus*** OM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Chromis multilineata*** PK 0.39 ± 0.15 1.31 ± 0.58 3.05 ± 0.82 3.97 ± 1.04 0.47 ± 0.18 0.85 ± 0.41 Microspathodon frontatus** HD 0.04 ± 0.02 2.13 ± 0.72 <0.01 <0.01 0.02 ± 0.01 0.71 ± 0.41 Stegastes imbricatus** HD 0.11 ± 0.02 0.26 ± 0.04 <0.01 <0.01 0.01 ± 0.01 0.02 ± 0.01 Priacanthidae Heteropriacanthus IM 0.07 ± 0.01 10.9 ± 2.76 0.03 ± 0.01 2.34 ± 1.36 <0.01 0.13 ± 0.13 cruentatus***** Scaridae Scarus hoefleri** HD 0.03 ± 0.01 5.52 ± 1.83 <0.01 <0.01 0.01 ± 0.00 0.42 ± 0.28 Sparisoma choati** HD 0.03 ± 0.01 2.91 ± 0.91 0.02 ± 0.01 3.46 ± 3.15 0.01 ± 0.00 1.05 ± 0.64

Serranidae Cephalopholis nigri** IM <0.01 <0.01 <0.01 0.01 ± 0.00 <0.01 <0.01 Cephalopholis taeniops**** PS <0.01 0.03 ± 0.03 0.04 ± 0.01 1.03 ± 0.60 0.01 ± 0.00 0.09 ± 0.08 Epinephelus adscensionis*** IM <0.01 0.12 ± 0.12 <0.01 0.21 ± 0.17 <0.01 0.25 ± 0.18 Paranthias furcifer*** PK 3.36 ± 0.69 15.85 ± 3.87 2.29 ± 0.51 23.28 ± 6.64 4.95 ± 1.23 89.5 ± 29.11 Rypticussubbifrenatus*** IM <0.01 0.23 ± 0.10 0.03 ± 0.01 1.86 ± 0.46 <0.01 0.05 ± 0.05 Serranus pulcher* PS <0.01 <0.01 0.02 ± 0.01 0.04 ± 0.01 0.06 ± 0.02 0.12 ± 0.05 Synodontidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Synodus synodus*** PS <0.01 <0.01 0.03 ± 0.01 0.44 ± 0.23 <0.01 <0.01 98

Tetraodontidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Canthigaster supramacula** IM 0.05 ± 0.01 0.41 ± 0.15 0.09 ± 0.02 0.34 ± 0.07 0.05 ± 0.01 0.18 ± 0.04 14.26 ± 344.23 ± 5.21 ± 0.37 96.75 ± 6.94 7.32 ± 0.55 0.78 39.42 173.37 ± 16.30

SPC SPB ROL DIET Family & Species Abundance biomass Abundance biomass Abundance biomass Acanthuridae Acanthurus monroviae** HD 0.01 ± 0.01 2.08 ± 1.88 0.26 ± 0.08 91.92 ± 30.7 0.03 ± 0.01 0.69 ± 0.36 Prionurus biafraensis** HD 0.51 ± 0.45 54.29 ± 36.1 0.01 ± 0.01 1.52 ± 1.52 0 .00 ± 0.00 <0.01 Apogonidae Apogon affinis*** PK <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Apogon imberbis**** PK <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 0.03 ± 0.02 Aulostomidae Aulostomus strigosus** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Balistidae Balistes capriscus*** IM <0.01 <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 Balistes punctatus** IM 0.01 ± 0.00 0.17 ± 0.16 <0.01 0.14 ± 0.14 0.01 ± 0.01 0.19 ± 0.11 Canthidermis sufflamen*** PK <0.01 <0.01 <0.01 1.21 ± 1.21 0.02 ± 0.01 0.04 ± 0.02

100

Blenniidae Ophioblennius sp* HD 0.14 ± 0.03 0.21 ± 0.04 0.05 ± 0.02 0.07 ± 0.02 0.09 ± 0.02 0.20 ± 0.06 Bothidae Bothus lunatus*** PS <0.01 <0.01 <0.01 <0.01 0.02 ± 0.01 0.15 ± 0.14 Carangidae Caranx crysos**** PS <0.01 <0.01 <0.01 <0.01 <0.01 0.61 ± 0.55 Caranx latus*** IM <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 0.02 ± 0.01 Centracanthidae Spicara nigricauda** PK <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Chaetodontidae Chaetodon robustus** IM <0.01 <0.01 <0.01 0.4 ± 0.40 0.01 ± 0.00 0.14 ± 0.12 Prognathodes marcellae** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Cirrhitidae Cirrhitus atlanticus** IM <0.01 0.13 ± 0.13 0.03 ± 0.01 0.69 ± 0.27 0.01 ± 0.01 0.26 ± 0.20 Diodontidae Diodon holocanthus***** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Gobiidae Gnatholepis thompsoni*** OM <0.01 <0.01 0.01 ± 0.01 0.01 ± 0.01 <0.01 <0.01 Gobius rubropunctatus** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Gorogobius nigricinctus** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

100

101

gorogobius stevcici* IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Wheelerigobius wirtzi** IM 0.04 ± 0.02 0.06 ± 0.03 0.01 ± 0.01 0.02 ± 0.01 0.05 ± 0.01 0.07 ± 0.02 Didogobius amicuscaridis* IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Haemulidae Pomadasys incisus** IM <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 0.89 ± 0.89 Holocentridae Holocentrus adscensionis*** IM 0.12 ± 0.06 17.85 ± 9.38 0.15 ± 0.05 20.85 ± 7.94 0.11 ± 0.03 3.7 ± 0.88 Myripristis jacobus*** PK 0.12 ± 0.04 11.15 ± 4.29 0.68 ± 0.27 60.73 ± 26.64 0.07 ± 0.04 5.20 ± 3.69 Sargocentron hastatum** IM 0.02 ± 0.01 0.82 ± 0.47 <0.01 <0.01 <0.01 <0.01 Kyphosidae Kyphosus spp*** HD 0.05 ± 0.03 13.35 ± 9.56 <0.01 <0.01 <0.01 <0.01 Labridae Bodianus pulchellus*** IM <0.01 <0.01 <0.01 <0.01 <0.01 0.28 ± 0.20 Bodianus speciosus** IM <0.01 0.73 ± 0.73 0.01 ± 0.00 2.02 ± 1.33 0.01 ± 0.01 0.03 ± 0.03 Clepticus africanus* PK 0.01 ± 0.01 3.28 ± 2.91 0.08 ± 0.02 21.62 ± 7.19 <0.01 <0.01 Coris atlantica** IM <0.01 <0.01 <0.01 0.06 ± 0.06 <0.01 0.07 ± 0.07 Thalassoma newtoni* PK 0.27 ± 0.08 1.38 ± 0.49 0.19 ± 0.04 1.12 ± 0.47 0.26 ± 0.04 1.09 ± 0.73 Lethrinidae Lethrinus atlanticus** IM <0.01 <0.01 <0.01 <0.01 <0.01 0.01 ± 0.00 Lutjanidae

102

Apsilus fuscus** IM <0.01 <0.01 <0.01 0.22 ± 0.22 <0.01 <0.01 Lutjanus fulgens** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Lutjanus goreensis** IM <0.01 <0.01 0.01 ± 0.00 25.28 ± 23.61 0.01 ± 0.00 0.25 ± 0.11 Monacanthidae Aluterus scriptus***** PK <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Cantherhines pullus*** IM 0.02 ± 0.01 1.44 ± 0.74 0.04 ± 0.01 3.05 ± 1.26 0.05 ± 0.01 2.24 ± 0.54 Mullidae Mulloidichthys martinicus*** IM 0.01 ± 0.01 0.18 ± 0.16 0.04 ± 0.04 2.55 ± 2.55 0.08 ± 0.06 5.53 ± 4.18 Pseudupeneus prayensis** IM <0.01 <0.01 <0.01 <0.01 0.08 ± 0.03 0.15 ± 0.06 Muraenidae Channomuraena vittata***** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Enchelycore nigricans*** IM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Gymnothorax vicinus*** PS <0.01 <0.01 <0.01 <0.01 0.01 ± 0.00 8.47 ± 5.66 Muraena melanotis*** PS 0.02 ± 0.01 0.70 ± 0.32 0.01 ± 0.01 15.93 ± 8.39 <0.01 2.90 ± 1.98 Muraena robusta*** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ophichthidae Myrichthys pardalis** IM <0.01 <0.01 <0.01 0.03 ± 0.03 <0.01 <0.01 Ostraciidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Acanthostracion IS <0.01 <0.01 <0.01 <0.01 <0.01 0.01 ± 0.01 notacanthus*** Pomacanthidae 102

103

Holacanthus africanus** IS 0.01 ± 0.01 1.00 ± 0.45 0.04 ± 0.01 2.24 ± 1.07 0.02 ± 0.01 0.28 ± 0.15 Pomacentridae Abudefduf saxatilis*** OM <0.01 <0.01 <0.01 <0.01 0.02 ± 0.01 0.67 ± 0.66 Abudefduf taurus*** OM <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Chromis multilineata*** PK 0.86 ± 0.30 1.81 ± 0.90 2.61 ± 1.04 3.32 ± 1.32 2.02 ± 0.53 14.74 ± 7.54 Microspathodon frontatus** HD 0.06 ± 0.02 4.29 ± 1.18 0.01 ± 0.00 0.47 ± 0.34 Stegastes imbricatus** HD 0.33 ± 0.05 0.75 ± 0.12 0.15 ± 0.02 0.34 ± 0.05 0.07 ± 0.01 0.16 ± 0.03 Priacanthidae Heteropriacanthus IM 0.01 ± 0.01 2.52 ± 1.46 0.01 ± 0.00 1.2 ± 0.82 <0.01 0.50 ± 0.37 cruentatus***** Scaridae Scarus hoefleri** HD <0.01 0.13 ± 0.13 <0.01 0.11 ± 0.11 0.02 ± 0.01 3.07 ± 1.27 Sparisoma choati** HD 0.07 ± 0.01 16.69 ± 3.53 0.03 ± 0.02 6.46 ± 3.30 0.02 ± 0.01 3.27 ± 1.54

Serranidae Cephalopholis nigri** IM <0.01 <0.01 <0.01 <0.01 0.02 ± 0.01 0.03 ± 0.01 Cephalopholis taeniops**** PS 0.02 ± 0.01 0.31 ± 0.21 0.01 ± 0.00 0.52 ± 0.43 0.05 ± 0.03 0.22 ± 0.15 Epinephelus adscensionis*** IM 0.01 ± 0.00 0.96 ± 0.58 0.02 ± 0.01 3.19 ± 1.27 <0.01 0.04 ± 0.04 Paranthias furcifer*** PK 0.81 ± 0.39 4.98 ± 1.77 7.85 ± 1.26 89.78 ± 21.7 0.20 ± 0.08 1.69 ± 1.24 Rypticussubbifrenatus*** IM <0.01 <0.01 <0.01 0.62 ± 0.51 <0.01 0.14 ± 0.10 Serranus pulcher* PS <0.01 <0.01 <0.01 2.02 ± 2.02 <0.01 0.01 ± 0.01 Synodontidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

104

Synodus synodus*** PS <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Tetraodontidae <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Canthigaster supramacula** IM 0.03 ± 0.01 0.12 ± 0.05 0.01 ± 0.01 0.05 ± 0.03 0.15 ± 0.10 0.79 ± 0.38 141.37 ± 359.86 ± 12.34 ± 0.51 3.57 ± 0.20 58.85 ± 5.29 3.56 ± 0.25 13.62 15.66

Geographic Range: Local endemic of São Tomé e Príncipe - Gulf of Guinea (*), Regional endemic - Eastern Atlantic (**), Atlantic Ocean (‡), Atlantic and Mediterranean Sea (₴) and Circumglobal (+).

104

105

Table S2: Number of Underwater Visual Censuses along the same exposure levels, topographic complexity and deep class per sites.

Eposure Sites low low-medium high-medium hegh Digo Vaz (DVA) 18 21 0 0 Ilhéu das Cabras (KIA) 0 0 0 28 Ilhéu Santana (SAN) 0 0 22 5 Sete Pedras Catedral (SPC) 0 0 10 1 Sete Pedras Branca (SPB) 0 0 13 0 Ilhéu das Rolas (ROL) 0 8 13 0 Total 18 29 58 34 Topographic complexity Sites low medium high Digo Vaz (DVA) 0 10 29 Ilhéu das Cabras (KIA) 14 14 0 Ilhéu Santana (SAN) 13 14 0 Sete Pedras Catedral (SPC) 0 11 0 Sete Pedras Branca (SPB) 0 7 6 Ilhéu das Rolas (ROL) 19 2 0 Total 46 58 35 Deep class Sites Shallow medium deep Diogo Vaz (DVA) 21 11 7 Ilhéu das Cabras (KIA) 0 14 14 Ilhéu Santana (SAN) 5 8 14 Sete Pedras Catedral (SPC) 1 10 0 Sete Pedras Branca (SPB) 0 4 9 Ilhéu das Rolas (ROL) 0 13 8 Total 27 60 52

106

Table S3: Number of photo-quadrats along the same depth (Coral cover)

Deep class Sites Shallow medium deep Diogo Vaz (DVA) 14 13 8 Ilhéu das Cabras (KIA) 0 12 0 Ilhéu Santana (SAN) 17 7 7 Sete Pedras Catedral (SPC) 6 0 0 Sete Pedras Branca (SPB) 0 0 7 Ilhéu das Rolas (ROL) 0 7 7 Total 37 39 29

107

CAPÍTULO II

Reef fish community structure and benthic cover of two islands in the Gulf of Guinea: São Tomé and Príncipe

(Em preparação para Environmental Biology of Fishes) Formatado de acordo as normas da revista

Maia H.A., Morais R.A., Quimbayo J.P., et al. (2018). Reef fish community structure and benthic cover of two islands in the Gulf of Guinea: São Tomé and Príncipe. Environmental Biology of Fishes.

108

Reef fish community structure and benthic cover of two islands in the Gulf of Guinea: São Tomé and Príncipe

Maia H.A.1,2,3, Morais R.A2,4, Quimbayo J.P.1, Rocha, L.A.5, Ferreira C.E.L.6, Floeter S.R.1*

1 Laboratório de Biogeografia e Macroecologia Marinha, Departamento de Ecologia e Zoologia, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, Florianópolis, SC 88010-970, Brasil;

2 Programa de Pós-graduação Ciência para o Desenvolvimento (PGCD), Instituto de Gulbenkian de Ciência, Oeiras, 62780-156, Portugal;

3 Departamento de Ciências Naturais e Biologia, Universidade de São Tomé e Príncipe, Bairro Quinta de Santo António, São Tomé and Príncipe.

4 College of Science and Engineering, James Cook University, Townsville, QLD 4811, Australia;

5 Section of Ichthyology, California Academy of Sciences, San Francisco, CA, USA;

6 Laboratório de Ecologia e Conservação de Ambientes Recifais, Universidade Federal Fluminense, Niterói, RJ 24001-970, Brasil.

______

*Correspondence author: Sergio R. Floeter, Tel: +55 48 3721 5521; Fax +55 48 3721 5156; E-mail: [email protected]

Target Journal: Environmental Biology of Fishes

109

Abstract

Located in the Gulf of Guinea (West Africa), the oceanic islands of São Tomé and Príncipe are one of the least studied marine ecosystems in the world, even though they are considered hotspots of marine biodiversity. Príncipe island is less populated, allowing for comparisons of marine ecosystem health between two distinct levels of anthropogenic influence. Here, we compare for the first time fish and benthic reef communities of São Tomé and Principe to evaluate the potential impact of higher human population in these islands. In addition, we provide a temporal comparison (with data collected in a 10-year interval) of the reef fish assemblages in São Tomé. We collected a total of 292 underwater visual censuses (40 m2) to characterize reef fish communities and 608 photo- quadrats (0.25 m2) to characterize the benthic communities. Reef fish species richness varied significantly between São Tomé and Príncipe islands. Mean fish abundance was similar between islands and in the 10- year interval, around 7 ind/m2. Biomass was higher in Príncipe (244.38 g/m2) than in São Tomé in 2016 (221.91), while in São Tomé 2006 we found the lowest mean (179.61). In terms of trophic groups, the planktivores dominated the two islands, both for (80-84%, proportion of abundance) and (48-62%, proportion of biomass). Benthic cover was dominated by algal turf with 30% cover in São Tomé and 27% to Príncipe. Corals comprised 8% of the benthic cover in São Tomé and 12% in Príncipe. Montastraea cavernosa was the most abundant hard coral in both islands. Our results suggest similarities of the benthic and reef fish communities in both islands, despite the higher biomass found in Príncipe. Additionally, we did not found a significant temporal variation between the reef fish communities in São Tomé island between the 10- year interval.

Keywords: Tropical Eastern Atlantic, Africa, oceanic islands, marine hotpots, Biosphere Reserve, fishing pressure.

110

Introduction The Tropical Eastern Atlantic (TEA) marine province extends from northwest Africa to Angola and includes four oceanic island-groups (Cape Verde archipelago and the Gulf of Guinea islands of Príncipe, São Tomé and Annobon; Floeter et al. 2008). Only recently, this biogeographic region has been studied in more detail regarding its marine biodiversity (Wirtz et al. 2007; 2013; Floeter et al. 2008; Polidoro et al. 2017) and rate of endemism [c. 30% for reef fishes (Floeter et al. 2008); c. 96% for cone snails (Peters et al. 2013; 2015; Polidoro et al. 2017); c. 21% for batoids; c. 18% reef-building corals (Polidoro et al. 2017)]. Regarding community structure of reef fishes only two studies have been published to date, one about oil platforms off Gabon (Friedlander et al. 2014) and another on depth gradient at Principe island (Tuya et al. 2017). No quantitative study of the benthic community structure is known to date on the TEA. São Tomé and Príncipe are part of the volcanic islands of the Cameroon line situated between 255 and 220 km off the coast of Africa, respectively (Lee et al. 1994). São Tomé has a larger area (895 km2) and human population (c.179,814) than Príncipe (area: 139 km2; population c. 7,542) (Lee et al. 1994; Caldeira 2002). Agriculture and fishing are the third most important sector of the economy of the archipelago, representing c. 20% of Gross domestic product, employs 30% of the active population (ADB 2012). Príncipe island is largely vegetated with tropical forest and has several small islets of varying sizes (Ilhéu Bom- Bom, Pedra da Galé, Ilhéus dos Mosteiros, Tinhosas, Boné do Jóquei). In 2011, Príncipe was declared a Biosphere Reserve by UNESCO (Abreu 2012) based solely on its terrestrial fauna and flora and diversity of habitats. The human population number is correlated with fish biomass in many studies around the world (Mora et al. 2011; Williams et al. 2015). Fisheries have had a strong impact in reef fishes along history (reviewed by Roberts, 2007). Since its discovery in 1470 and further colonization 20 years later, São Tomé and Príncipe have had a long history of fishing (Hodges and Mewitt 1988; Maia et al. 2018a). Intensive overfishing of many prized nearshore species, like groupers and snappers, has been a concern throughout the region for decades (Hodges and Mewitt 1988; Horemans et al. 1994; Belhabib 2015; Maia et al. 2018). In contrast to most locations in the region, Príncipe island is less populated, allowing for comparisons of marine ecosystem health between two distinct levels

111

of anthropogenic influence, while still within the same biogeographic region. Based on the differences between the islands described above, here we compare fish and benthic reef communities of São Tomé and Príncipe, as well as fishing pressure estimates to evaluate potential differences between them. In addition, we provide a temporal comparison (with data collected in a 10-year interval) of the reef fish assemblages in São Tomé. We expect that Príncipe communities will be in better shape (more large- bodied fish and more coral cover) than the more populated island of São Tomé. We also expect that data from 2006 will show more biomass of fishes than that of 2016, taking into account that the number of fishermen doubled in the island (c. 2000 to ~4000).

Material and Methods

Study area The study was performed at São Tomé and Príncipe (Democratic Republic of São Tomé and Príncipe), both oceanic islands located near the equator in the Gulf of Guinea, West Africa (Figure 1A and B). Data were collected in two expeditions, the first in February 2006 in São Tomé island (Figure 1D) and the second expedition in January 2016 in São Tomé and Príncipe islands (Figure 1D and C). Sampled sites in São Tomé were: “DiogoVaz” (DVA), “Cabras Kia” (KIA), “Ilhéu Santana” (SAN), “Sete Pedras – Catedral” (SPC), “Sete Pedras – Pedra Branca” (STB) and “Ilhéu das Rolas” (ROL) (see description of each site in Maia et al. 2018b). In Príncipe island were: “Praia da Lapa” (LAP), “Pedra da Galé” (GAL), “Ilhéu Bombom” (BOM), “Praia Banana” (BAN), “Ilhéu dos Mosteiros” (MOS), “Ilhéu Boné do Jóquei” (BJO) and “Tinhosa Grande” (TIN). TIN, SPC and SPB are dominated by seabird’s colonies. Overall, for both islands, all sites had a similar habitat structure mostly composed by rock reefs, sand patches and a few biogenic reefs (i.e., rhodoliths, coralline algae, turf, macroalgae, sponges and soft and hard corals). All sites represent local artisanal fishers fishing areas, both for line fishing and spearfishing. In addition, these sites have been targeted by destructive fishing (blast fishing) (Maia et al. 2018a).

112

Figure 1: The location of the São Tomé and Príncipe islands in the Eastern Tropical Atlantic: (A) the Gulf of Guinea represented by grey hexagon; (B) São Tomé and Príncipe islands; (C) Príncipe island and (D) São Tomé island; Distribution of sampling sites across the islands. Sites from Príncipe ( ): Praia da Lapa (LAP), pedra da Galé (GAL), Ilhéu Bombom (BOM), Praia Banana (BAN), Ilhéu dos Mosteiros (MOS), Ilhéu Boné do Joquei (BJO) and Ilhas Tinhosas (TIN). Sites from São Tomé data collected 10 years separately ( ) and Sites from São Tomé ( ): Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Catedral (SPC), Sete Pedras Branca (SPB) and Ilhéu das Rolas (ROL). Blue lines represented the isobaths of 20 and 50 m.

Reef fish community structure Reef fishes were surveyed along 342 transects (São Tomé 2006, n=139 transects; São Tomé 2016, n=51; Príncipe 2016, n=153) distributed

113

over 13 sites (six in São Tomé and seven in Príncipe, see Table supplementary 2). Sites were selected to represent the broad environmental features of these islands. Underwater visual surveys along transects of 40 m2 (20 x 2 m) were used to estimate reef fish species density, abundance and biomass, while free or scuba diving. UVCs were performed in areas haphazardly chosen by the diver. This method consists on identifying, counting and estimating the size (total length in cm) of all fish species observed both in the water column (in the lay-out phase) and on the bottom (in the returning phase, Floeter et al., 2007; Morais et al. 2017). The UVCs were performed in areas between 5 and 30 m of depth. Furthermore, during the UVCs, the diver also collected some physical parameters, such as, depth and topographic complexity (three classes: low, medium and high following Ferreira et al. 2001). The body mass of fishes were estimated for each individual fish using the allometric length-weight conversion W= a * TLb, where parameters a and b are species-specific constant, and W is the weight in grams. Length-weight parameters were gathered for each species from FishBase (Frose and Pauly 2016). In cases where species coefficients were not available, we used coefficients of congeneric species that were phylogenetically or morphologically similar. All fish species were classified according to their functional group (seven classes) and body size (five classes) based in proprieties of life-history traits following Mouillot et al. (2014) and Kulbicki et al. (2015). The trophic groups considered were herbivore-detritivores (HD: feed upon turf and filamentous algae and/or detritus); macroalgae-feeders (HM: large fleshy algae and/or seagrass); omnivores (OM: both vegetal and animal material); sessile invertebrate feeders (IS: corals, sponges, ascidians); mobile invertebrate feeders (IM: benthic prey, such as crabs and mobile mollusks); planktivores (PK: small organism in the water column) and piscivores (FC: fish and cephalopods). The body size classes used were <7 cm, between 8–15 cm, between 16–30 cm, between 31–50 cm, and larger than 50 cm.

Reef benthic communities

Benthic cover was assessed through 608 photo-quadrats were analysed (112 to São Tomé - 2006 and 496 to Príncipe - 2016). Quadrats were displayed uniformly along the same depth contours and at the same sites where the fish visual censuses were conducted. Photos were taken from a distance of ∼80 cm from the substratum, and each photo-quadrat

114

corresponds an area of 50×50 cm (e.g. Krajewski and Floeter 2011). Photographs were analysed using the software photoQuad v7.6 (MCR) with Excel extensions (Trygonis and Sini 2012). Thirty points were randomly positioned over each image and the organism below each point was identified as the “turf” (the epilithic algal matrix), calcareous algae, sponges, macroalgae, sand, coral, zoanthidae, gorgonians, and other organisms (e.g., Echinoidea and Crustacea) (Bell and Barnes 2001; Littler and Littler 1984; Steneck and Dethier 1994). All corals observed were identified according to Laborel (1974) to the lowest taxonomic level possible.

Environmental and anthropogenic drivers We investigated the effect of depth (DP), topographic complexity (TC), wave exposure (WE), and fishing and canoe pressure on reef fish structure in São Tomé and Príncipe. The WE was classified in four levels: high (unsheltered site, unprotected from currents and dominant waves); medium-high (unsheltered site, unprotected from currents, but protected from dominant waves); low-medium (site sheltered by physical barriers, e.g., islets, and protected against the current or dominant waves); and low (site sheltered by physical barriers, protected against current and wave dominant). The TC was classified in three categories according to Ferreira et al. (1997): high (large boulders and holes >1 m in size and depth), medium (predominance of small boulders and holes <1 m in size and depth) and low (few and small benthic organisms and predominance of epiphytic algae). Depth was also classified in three levels: as shallow (4– 8 m), medium (9–16 m) and deep (17–32 m). Finally, fishing pressure (Fp) was estimated for each sites using Fp= (PP/PS)*100, where PP is sum of the artisanal fishers observed fishing in each place during the samplings and PS is sum of the fishers of the artisanal communities near the sites with capacity of maximum displacement of the fishing between 12 m to 22 km. Concerning Canoe pressure (Cp) was estimated for each sites using Cp= (PC/CA)*100 where PC is sum of the canoe of artisanal fishers observed in each place during the samplings and CA is sum of the canoe of artisanal communities near the sites.

Data analysis To evaluate the structure of the reef fish community between the islands (spatial) and temporally in São Tomé, richness, mean abundances and biomass were calculated and compared. In addition, we also calculated the means of species densities, abundances and mean biomass

115

across all trophic groups and size classes. The number of transects varied among islands between years (i.e., from 51 to 153, mean ± SD =114.3 ± 55.3). To avoid overrepresentation of highly sampled islands, the mean of species density, abundance and biomass values from all transects were used as the independent observations. Total species density, abundance and biomass among islands that differed in sample size were compared using a randomization procedure (bootstrap). After this, data were modeled using a gamma-distributed generalized linear mixed model (GLMM). The gamma-distribution was chosen because it adequately represents the data: continuous values with a large frequency of small and small frequency of large values. Random intercepts for each locality and site (nested random effects; Bunnefeld and Phillimore 2012) were added. This GLMM was employed because of the study’s hierarchical sampling design and aimed at assessing and quantifying which spatial scale introduced more variability to species density, abundance and biomass (site, locality or residual) and predicting average values of species density, abundance and biomass per islands. The covariates used were all measured at the sites among islands scale and included mean depth, mean wave exposure, and mean topographic complexity, anthropogenic effect (fishing and canoe pressure) variables. Depth and topographic complexity were obtained for each transect and averaged for the site-scale. To explore patterns in islands segregation according to functional group and size class structure, non-metric multidimensional scaling (nMDS) over Bray–Curtis dissimilarity matrices were calculated from transformed data. To investigate how much of the variance in (abundance and biomass) of functional and size structure could be explained by islands or locality an analysis of similarities (ANOSIM) were performed using non-rarefied data normalized employing the package vegan in R (www.r-project.org; Oksanen et al. 2017) on the square-root transformed data. Finally, the influence of covariates on both functional and size class patterns of assemblage structure was assessed by correlating the nMDS axes with covariates using the function envfit and its permutation test of significance in the package vegan in R.

Results

Reef fish structure Considering both islands, a total of 10,853 individuals belonging to 74 fish species were surveyed during this study (Table S1). Within trophic groups, mobile invertebrate feeders were the most species rich (39 species, 51% of surveyed species), followed by planktivores (12 species,

116

16%), piscivores (9 species, 12%), herbivore-detritivores (8 species, 10%), omnivores (5 species, 6%) macroalgae-feeders (2 species, 3%) and sessile invertebrate feeders (2 species, 3%). The most important species both in terms of abundance and biomass were: Chromis multilineata, Paranthias furcifer, Thalassoma newtoni, Myripristis jacobus, Wheelerigobius wirtzi, Stegastes imbricatus, Gnatholepis thompsoni, Mulloidichthys martinicus and Microspathodon frontatus (Table S1). Total cumulated species richness varied between São Tomé and Príncipe island (in 2016), but not temporarily between São Tomé island in 2006 vs. 2016 (Figure 2A). However, some variation occurred in species density (per 40 m2) among sites of each island, both spatially (São Tomé vs. Príncipe in 2016) and temporally (São Tomé in 2006 and 2016; Figure 3A, D and G). Mean total abundance of both islands was (~7.73 ± 0.38 ind/m2, Figure 2B), while mean abundance per site varied from ~7 to 15 ind/m2 (Figure 3A, E and H). Mean total biomass was higher at Príncipe island in 2016 (244.38 ± 12.53 g/m2), followed by São Tomé in 2016 (221.91 ± 14.83g/m2) and São Tomé island in 2006 with the lowest mean (179.61 ± 14.16 g/m2; Figure 2C). They also varied between different sites along the islands (Figure 3C, F and I). Wave exposure and fishing pressure significantly influenced the species density, abundance and biomass of fish species in São Tomé and Príncipe islands (Table 1), while depth influenced only total abundance.

117

Figure 2: Temporal comparison of fish assemblages among the two sampling islands (São Tomé in 2006 vs. 2016) and (São Tomé vs. Príncipe in 2016). Species richness (A), abundance (B) and biomass (C). Each color represents a different years by islands. Boxplots (B and C) show medians (black line), mean (red diamond) upper and lower quartiles, and 95% confidence intervals. Red lines in (A), represent the standardized number of surveys. Each point in (B and C) represents an underwater visual census.

118

Figure 3: Variation between sites in São Tomé 2016 vs. Príncipe 2016 islands. Species density (A, D and G), abundance (B, E and H) and biomass (C, F and I) of fish assemblages sampled. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB), Praia Banana (BAN), Boné do Joquei (BJO), Ilhéu Bombom (BOM), Pedra da Galé (GAL), Praia da Lapa (LAP), Ilhéu dos Mosteiros (MOS) and Ilhas Tinhosas (TIN). Boxplots show medians (black line), mean (red diamond), upper and lower quartiles, and 95% confidence intervals. Each point represents an underwater visual census.

119

Table 1: Results of GLMM evaluating the influence of abiotic and anthropogenic factor on the fish community (Species density, abundance and biomass) at São Tomé ad Príncipe islands. Bold values indicate significant.

Fish structure df F p-value

Species density Depth 1 0.2 0.652 Topographic complexity 1 2.53 0.111 Wave exposure 1 9.12 0.002 Canoe pressure 1 2.08 0.148 Fishing pressure 1 4.72 0.029

Total abundance Depth 1 6.83 0.008 Topographic complexity 1 0.76 0.383 Wave exposure 1 4.42 0.035 Canoe pressure 1 2.28 0.13 Fishing pressure 1 4.34 0.033

Total biomass Depth 1 2.2 0.137 Topographic complexity 1 2.58 0.107 Wave exposure 1 8.35 0.003 Canoe pressure 1 2.06 0.151 Fishing pressure 1 4.34 0.036

Planktivores were the most abundant trophic group in both São Tomé and Príncipe and between years in São Tomé, summing up 70% of all individuals in São Tomé 2006, 68% in São Tomé 2016 and 60% in Príncipe 2016 (Figure 4B and E). The other trophic groups contributed each with less than 30 to 40% of the total abundance. Species density was mainly led by three major functional groups in both islands, mobile invertebrate feeders (31–36%), followed planktivores (28–32%) and herbivores-detritivores (14–23%; Figure 4A and D). In relation to total biomass, it was concentrated in macroalgae-feeders, summing up 40% in São Tomé 2006, 36% in São Tomé 2016 and 42% in Príncipe 2016, followed by planktivores (32% - São Tomé 2006, 21% - São Tomé 2016

120

and 19% - Príncipe 2016) and herbivores/detritivores, mobile invertebrate feeders and piscivores, all with similar proportions (Figure 4C and F). The smallest size class (0–7 cm) encompassed the highest abundance, totaling 71% in São Tomé 2006, (46%) to São Tomé 2016 and (35%) to the Príncipe 2016 of all individuals (Figure 5B and E), followed by fishes 8– 15 cm that summed up (32% - São Tomé 2006, 40% - São Tomé 2016 and 42% - Príncipe 2016) of all individuals. Other classes comprehended less than 20% of the total abundance. Biomass, however, was concentrated in the fishes of 50 cm of size or over both in São Tomé 2006 and 2016 summing up 58% and 43% of the total biomass. In relation of Príncipe 2016, the biomass was mainly concentrated in fishes of sizes between 16– 30 cm that summed up 36% of the total biomass (Figure 5C and F).

121

Figure 4: (a) Mean – S.E. functional fish relative species density (a), relative abundance (b) and relative biomass (c) from islands. Proportional contribution of functional groups to total standing species density (d), abundance (e) and biomass (f) from islands. The numbers indicate proportions of each functional group at each islands between different years. Functional groups: herbivores-detritivores ( ), macroalgae-feeders ( ), omnivores ( ), sessile invertebrate feeders ( ), mobile invertebrate feeders ( ), planktivores ( ) and piscivores ( ).

122

Figure 5: Mean – S.E. size class of fish relative species density (a), relative abundance (b) and relative biomass (c) from islands. Proportional contribution of size class to total standing species density (d), abundance (e) and biomass (f) from islands. The numbers indicate proportions of each functional group at each islands between different years. Size class: 0–7 cm ( ), 8–15 cm ( ), 16–30 cm ( ), 31–50 cm ( ) and = > 50 cm ( ).

123

Overall, the trophic and size class structure of reef fishes was similar in São Tomé and Príncipe islands. However, we observed a higher biomass in São Tomé vs. Príncipe in 2016, which tended to occupy opposite areas of both functional and size class in the nMDS biplots (Figure 6A and B). Planktivores and mobile invertebrate feeders were positively related to high biomass in São Tomé island in 2016, whereas macroalgae-feeders, herbivores-detritivores and sessile invertebrate feeders were negatively related to high biomass in the nMDS biplots (Figure 6A). For size classes, fishes between 16–30 cm were positively related to high biomass in Príncipe island in 2016, whereas fishes between 8–15 cm and >50 cm were negatively related to high biomass in the nMDS biplots (Figure 6B). In terms of abundance of structure functional and size class of reef fish, also observed that high abundance of São Tomé vs Príncipe in 2016 tended to occupy opposite areas in nMDS biplots (Figure S.1). In temporal terms, the higher biomass of São Tomé in 2006 vs. 2016 appears to occupy the same areas of the functional nMDS biplot (Figure 7A), while, size classes tended to occupy opposite areas (Figure 7B). Planktivores and mobile invertebrate feeders were positively related to the higher biomass in São Tomé island in 2006 and 2016, whereas macroalgae-feeders, herbivores-detritivores and sessile invertebrate feeders were negatively related to biomass in the nMDS biplot (Figure 7A). Concerning the structure of biomass in size classes between São Tomé 2006 vs. 2016, fish between 8–15 cm in size was positively related to high biomass in São Tomé island in 2006, whereas fish between 16–30 cm were negatively related to higher biomass in nMDS biplot (Figure 7B). In addition, we observed a positive relationship between higher biomass of fishes in the transects of São Tomé 2016 with size over 30 cm and a negative relation with fish smaller than 7 cm (Figure 7B). In terms of the functional structure of fish abundances, we do not observed differences in the transects of São Tomé in 2006 and 2016, while in terms of size classes, they appear to occupy the opposite areas of the biplot (Figure S2). The dissimilarity in the functional structure and size classes of reef fishes were stronger when comparing site by site between 2006 vs. 2016 in São Tomé island (Figure S. 3, 4, 5 and 6). For example, higher biomass in DVA in 2016 are associated with piscivores, whereas in 2006 it seems that higher biomass are associated with herbivores-detritivores, planktivores and mobile invertebrate feeders (Figure S. 3). Concerning the class size structure in the same site, we observed that higher biomass in 2006 are associated with small fishes (0–7 cm), while in 2016 they are associated with larger fish (>31 cm; Figure S. 4).

124

Figure 6: Non-metric multidimensional scaling of structure of biomass of functional groups (A) and structure of biomass of size classes (B) of fishes in São Tomé (Yellow) vs. Príncipe (Blue) islands in 2016. Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Figure 7: Non-metric multidimensional scaling of (A) structure of biomass of functional groups and (B) structure of biomass of size classes of fishes in São Tomé island in 2006 (Orange) vs 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of each functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

125

Benthic community Overall, turf dominated in both island, with a mean of 30 % (± 2.42) in São Tomé and 27 % (± 0.43) in Príncipe. This component of the benthic community was followed by calcareous algae in São Tomé (28 ± 2.28%) and in Príncipe (20 ± 0.43%), sand in São Tomé (14 ± 2.41%) and in Príncipe (9 ± 0.47%). Other benthic components were (in order of importance): macroalgae in São Tomé (10 ± 1.50%) and in Príncipe (8 ± 0.27%), corals in São Tomé (8 ± 1.21%) and in Príncipe (12 ± 0.60%), sponges in São Tomé (7 ± 1.05%) and in Príncipe (16 ± 0.71%), zoanthids in São Tomé (3 ± 0.68%) and in Príncipe (6 ± 0.32%), and gorgonians in São Tomé (0.36 ± 0.12%) and in Príncipe (0.13 ± 0.01%) (Figure 8). Four species dominated the coral cover in both São Tomé and Príncipe reefs. Montastraea cavernosa was the most abundant hard coral in both islands (46% in São Tomé and 78% in Príncipe of the total coral cover), followed by Tubastraea sp. in São Tomé with 29% but only 2% in Príncipe. Porites branneri was the third most abundant coral species in São Tomé with 15% but only 1% in Príncipe, followed by Siderastrea radians with 9% in São Tomé and 17% in Príncipe. Schizoculina africana, Porites porites, Siderastrea siderea, Stylaster blatteus and Favia gravida contributed with less than 1% in both islands.

126

Figure 8: Percentage of benthic cover found during sampling. São Tomé island (A) and Príncipe island (B). Each point represents a photoquadrat. Benthic categories: Turf (TUR), Coralline Algae (CAL), Sponges (SPO), Macroalgae (MAC), Sand (SAN), Coral (COR), Zoanthids (ZOA), other life (OTH) and Gorgonians (GOR). Boxplots show medians (black line), upper and lower quartiles, the colours are represent organism and structure of categories.

126

127

Discussion We observed an unexpected high similarity of the fish community between São Tomé and Príncipe (both in general terms as well as in trophic and size class structure). In the same way, the benthic cover was similar among the two islands. Given that the human population are much lower in Principe we expected less fishing pressure there, but it seems that fishing effort is equivalent in both São Tomé and Príncipe what is reflected in the similar biomass and size structure between them. We also expected more differences in the fish and benthic structure between the 10-year intervals of the samples. But very few differences were detected. One explanation may be related with "regime shifts" (i.e. conspicuous jumps from one rather stable condition to another) (Scheffer et al. 2001). It seems that we have distinct phases in the São Tomé and Príncipe marine ecosystem: I) the pristine pre-colonization, II) than the colonization, than III) a steep decline of fisheries with the start of the nylon nets for fishing around 1970, and IV) a relatively stable phase from 2000 until today (Maia et al. 2018a).The lack of substantial differences between the two islands could also be associated with the relatively small distance between them, a distance that can be easily traversed by the larval phases of most fishes and benthic organisms (Luiz et al. 2013). Wave exposure, depth and fishing pressure proved to be important drivers of the reef fish community structure and standing biomass stock. Wave exposure has been suggested as one of the key physical factors influencing community structure in shallow reef habitats in various locations around the world including oceanic islands (Friedlander 2003; Floeter et al. 2007; Krajewski and Floeter 2011; Luiz et al. 2015; Andradi-Brown et al. 2016; Crane et al. 2017). Evidence from a wide range of intertidal and subtidal marine habitats has indicated that concurrent changes in wave exposure and the distribution and abundance of species are evident for both mobile and cryptobenthic species (e.g. Lewis 1968; Denny 1988; Luiz et al. 2015). In sessile organisms, this relationship has often been attributed to differences in the ability of each species to withstand the water movements produced by the crash and surge of breaking waves. Such water movements have been found to inflict physical damage through impact and abrasion (Shanks and Wright 1986; Bodkin et al. 1987), affect rates of growth (Dennison and Barnes 1988; Trussell 2002), and increase rates of mortality through the detachment and removal of individuals from habitats (Ebeling et al. 1985; McQuaid and Lindsay 2000). Sessile organisms susceptible to higher energy hydrodynamics often display functional characteristics to counteract these effects, including morphologies that reduce drag (Denny

128

1994; Denny and Gaylord 1996), simpler branching in algal thalli (Gaylord et al. 1994; Friedland and Denny 1995), and robust colony architectures in calcareous algae and corals (Done 1983). Consequently, the functional morphology of resident sessile organisms has often been correlated with levels of wave exposure, particularly in high wave energy habitats (Denny 1994; Vogel 1994). Recent studies on a highly mobile group of organisms like reef fishes, have indicated that their distribution patterns across a range of wave exposures may be directly related to their swimming abilities (Fulton et al. 2001; Floeter et al. 2007). Strong correlations were found to exist between the swimming performance of labrid fishes (family Labridae) and their distribution and abundance amongst coral reef habitats of differing wave exposure, both within (Fulton et al. 2001) and among reefs (Bellwood and Wainwright 2001). These patterns applied to both adults and juveniles (Fulton and Bellwood 2002), and were consistent for several tropical localities over global biogeographical scales (Bellwood et al. 2002). In the Gulf of Guinea islands, more specifically in São Tomé, physical disturbance from waves is the primary natural mechanism controlling coral reef community structure (Maia et al. 2018b). In this study, the degree of wave exposure was shown to have a positive relationship with several measures of species density, abundance and biomass of fish organization in this island. This relationship suggests that habitats most exposed to highest wave energies maintain larger fish populations with greater, such as, abundance and biomass of fish. We found evidence for depth effects on the distribution of abundance of reef fish from São Tomé and Príncipe islands. This pattern was also observed in studies conducted in other regions (Sheppard et al. 1992; Friedlander 2010; Andradi-Brown et al. 2016; Friedlander 2003). Depth is a metric that can be incorporated into species distribution models easily, so has considerable utility in spatial planning. For example, Tuya et al. (2017), in the Príncipe island found evidence for depth effects on the distribution of functional traits of nearshore reef fish. In this study, they observed a positive association between planktivorous species and depth, whereas a negative association was between herbivorous fish species and depth. In addition, in the São Tomé island (Maia et al. 2018b) were observed a positive relationship between species density, abundance and biomass of fish reef fish and depth in this island. Depth and wave exposure are not mutually exclusive and the application of both is likely more useful than each individually for management planning purposes. Fishing pressure is important anthropogenic driver of change in the structure of reef fish (Grigg 1994; Pardede and Campbell 2006). Since

129

species have distinct sensitivities, differences in abundance distributions are expected under stress, namely with some species that share the same features decreasing in abundance, such as keystone species (e.g., shark and groupers) while others remain stable or even increase such as, the planktivores (Friedlander et al. 1998; McClanahan et al. 2006). In our study, a part of variation in species density, abundance and biomass composition was explained by fishing pressure (Table 1). Overfishing has negative consequences to marine organisms. Many studies nonetheless provide correlative evidence for the negative impacts of fishing pressure on apex predator abundance (Friedlander et al. 1998; Stallings 2009; Williams et al. 2008; 2011; 2015; Richards et al. 2012). Presently, São Tomé and Príncipe have a population of approximately 190,000 people (183,000 to São Tomé and 7,000 to Príncipe) (RGPH, 2012). Although the human population on the island of Principe is about twenty times smaller than the population of São Tomé, there is a constant flow of fishers from São Tomé to Príncipe (Maia et al. 2018a). In addition, fishing is the sector, representing c. 3% of Gross domestic product, and employs 30% of the active population along with agriculture in the islands. Overall, our results suggest higher similarity in the composition of the benthic and reef fish communities in both islands. Additionally, we also concluded that there was no significant temporal variation between the reef fish communities on the São Tomé island between a 10-year interval.

Acknowledgements We thank Rufford Foundation (#18424-1) and Fundação Calouste Gulbenkian (Portugal) (P-141933) from the grants received. We thank CAPES (Brazil) for scholarships and CNPq for financial support over the years. Direcção geral das Pescas de São Tomé e Príncipe for permits and logistics, Peter Wirtz for helping identifying the benthic components. All member of the California Academy of Sciences expedition to Príncipe in 2016. Roça Belo Monte. All member of the National Geographic expedition at São Tomé 2006, in especial João L. Gasparini, Angus Gascoigne. Thanks Jean Louis and all the crew of Club Maxel, Alberto Miranda of Atlantic Diving Center, Nora Rizzo, Flavia Nunes, Nancy Knowton, and the people of São Tomé e Príncipe for wonderful times.

130

References Abreu AD (2013) Príncipe Island’s Biosphere Reserve (Democratic Republic of São Tomé & Príncipe): A Living Laboratory for Sustainable Development. In: Clusener-Godt RP-S& M (ed) AfriMAB Biosphere reserves in Sub-Sahan Africa: Showcasing sustainable development, First. Department of Environmental Affairs & UNESCO, South Africa. 284 – 302. Andradi-Brown D, Gress E, Wright G, et al (2016) Reef Fish Community Biomass and Trophic Structure Changes across Shallow to Upper- Mesophotic Reefs in the Mesoamerican Barrier Reef, Caribbean. PLoS One 11:e0156641. doi: 10.1371/journal.pone.0156641 A.D.B. (African Development Banck), Ddevelopment Centre of the Organisation for Co-Operation and Development, United Nations Development Programme, United Nations Economic Commission for Africa (2012) African Economic Outlook. www.africaneconomicoutlook.org/po. Acessado em 02/01/2018. Bell JJ, Barnes DKA (2001) Sponge morphological diversity: A qualitative predictor of species diversity? Aquatic Conservation: Mar Fres Ecos 11:109–121. Bellwood DR, Wainwright PC (2001) Locomotion in labrid fishes: implications for habitat use and cross-shelf biogeography on the Great Barrier Reef. Coral Reefs 20:139–150. Bellwood DR, Wainwright PC, Fulton CJ, et al (2002) Assembly rules and functional groups at global biogeographical scales. Funct Ecol 16:557–562. Belhabib D (2015) Fisheries of Sao Tome and Principe, a catch reconstruction (1950-2010). Fisheries Centre Working Paper. Working Paper #2015 - 6786. University of British Columbia, Vancouver, BC, V6T 1Z4, Canada. Bodkin JL, Van Blaricom GR, Jameson RJ (1987) Mortalities of kelp forest fishes associated with large oceanic waves off central California, 1982–1983. Environ Biol Fishes 18:73–76. Bunnefeld N, Phillimore AB (2012) Island, archipelago and taxon effects: mixed models as a means of dealing with the imperfect design of nature’s experiments. Ecography 35:15–22. Caldeira RJMM (2002) Petrology of ultramafic nodules from São Tomé island, Cameroon Volcanic Line (oceanic sector). J Afr Earth Sci 34:231–246. Campbell SJ, Pardede ST (2006) Reef fish structure and cascading effects in response to artisanal fishing pressure. Fish Res 79:75–83. Crane NL, Paddack M, Nelson P, et al (2017) Atoll-scale patterns in coral

131

reef community structure: Human signatures on Ulithi Atoll, Micronesia. PLoS One 12: e0177083. doi:10.1371/journal.pone.0177083 Dennison WC, Barnes DJ (1988) Effect of water motion on coral photosynthesis and calcification. J Exp Mar Biol Ecol 115:67–77. Denny MW (1988) Biology and the mechanics of the wave-swept environment. Princeton University Press, Princeton. Denny MW (1994) Roles of hydrodynamics in the study of life on wave- swept shores. In: Wainwright PC, Reilly SM (eds) Ecological morphology: integrative organismal biology. University of Chicago Press, Chicago, pp 169–204. Denny M, Gaylord B (1996) Why the urchin lost its spines— hydrodynamic forces and survivorship in three echinoids. J Exp Biol 199:717–729. Done TJ (1983) Coral zonation: its nature and significance. In: Barnes DJ (ed) Perspectives on coral reefs. Clouston, Manuka, pp 107–147. Ebeling AW, Laur DR, Rowley RJ (1985) Severe storm disturbances and reversal of community structure in a southern California kelp forest. Mar Biol 84:287–294. Ferreira CEL, Gonçalves JEA, Coutinho R (2001) Community structure of shes and habitat complexity on a tropical rocky shore. Environ Biol Fishes 61:353–369. Ferreira CEL, Floeter SR, Gasparini JL, et al (2004) Trophic structure patterns of Brazilian reef fishes: a latitudinal comparison. J Biogeogr 31:1093–1106. Floeter SR, Krohling W, Gasparini JL, et al (2007) Reef fish community structure on coastal islands of the southeastern Brazil: The influence of exposure and benthic cover. Environ Bio Fish 78:147–160. Friedland MT, Denny MW (1995) Surviving hydrodynamic forces in a wave-swept environment: consequences of morphology in the feather boa kelp, Egregia menziesii (Turner). J Exp Mar Biol Ecol 190:109–133. Friedlander AM, Brown EK, Jokiel PL, et al (2003) Effects of habitat, wave exposure, and marine protected area status on coral reef fish assemblages in the Hawaiian archipelago. Coral Reefs 22:291–305. Froese R, Pauly D (2016) FishBase. World Wide Web electronic publication. Available: www.fishbase.org. Accessed 2016 Jun 13. Floeter SR, Krohling W, Gasparini JL, et al (2007) Reef fish community structure on coastal islands of the southeastern Brazil: the influence of exposure and benthic cover. Env Biol Fish 78:147–160. Floeter SR, Rocha LA, Robertson DR, et al (2008) Atlantic reef fish

132

biogeography and evolution. J Biogeogr 35:22–47. Fulton CJ, Bellwood DR, Wainwright PC (2001) The relationship between swimming ability and habitat use in wrasses (Labridae). Mar Biol 139:25–33. Fulton CJ, Bellwood DR (2002) Ontogenetic habitat use in labrid fishes: an ecomorphological perspective. Mar Ecol Prog Ser 236:255–262. Gaylord B, Blanchette CA, Denny MW (1994) Mechanical consequences of size in wave-swept algae. Ecol Monogr 64:287–313. Grigg RW (1994) Effects of sewage discharge, fishing pressure and habitat complexity on coral ecosystems and reef fishes in Hawaii. Mar Ecol Prog Ser 103:25–34. Hodges T, Mewitt M (1988) São Tomé and Príncipe from plantation colony to microstate. United States of America. Horemans B, Gallène J, Njock J (1994). Revista setorial da pesca artesanal a São Tomé e Príncipe. Programa de desenvolvimento integrado das pescas artesanais na África Ocidental (DIPA). Kulbicki M, Parravicini, Valeriano, et al (2015) Patterns and processes in reef fish body size. In: Mora, Camilo, (Ed.) Ecology of Fishes on Coral Reefs. Cambridge University Press, Cambridge, UK, pp. 104–115. Krajewski JP, Floeter SR (2011) Reef fish community structure of the Fernando de Noronha Archipelago (Equatorial Western Atlantic): The influence of exposure and benthic composition. Environ Biol Fishes 92:25–40. Laborel J (1974) West African reefs corals: an hypothesis on their origin. Proceedings of the 2nd. International Coral Reef Symposium 1:425–443. Lee DC, Halliday AN, Fitton JG, et al (1994) Isotopic variations with distance and time in the volcanic islands of the Cameroon line: evidence for a mantle plume origin. Earth Planet Sci Lett, 123:119– 138. Lewis JR (1968) Water movements and their role in rocky shore ecology. Sarsia 34:13–36. Littler MM, Littler DS (1984) Relationships between macroalgal functional form groups and substrata stability in a subtropical rocky-intertidal system. Mar Biol Ecol 74:13–34. Luiz OJ, Allen AP, Robertson DR, et al (2013) Adult and larval traits as determinants of geographic range size among tropical reef fishes. PNAS, 110:16498–16502. Luiz OJ, Mendes TC, Barneche DR, et al (2015) Community structure of reef fishes on a remote oceanic island (St Peter and St Paul’s

133

Archipelago, equatorial Atlantic): the relative influence of abiotic and biotic variables. Mar Freswater Res 66:739–749. Maia HA, Morais RA, Siqueira AC, et al (2018a) Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea. Ocean and Coastal Management. https://doi.org/10.1016/ j.ocecoaman.2018.01.006 Maia HA, Morais RA, Quimbayo JP, et al (2018b) Spatial patterns and drivers of fish and benthic reef communities of São Tomé island (Tropical Eastern Atlantic). Marine Ecology. Submitted – Cap 1. Morais RA, Ferreira CEL, Floeter SR (2017) Spatial patterns of fish standing biomass across Brazilian reefs, Southwestern Atlantic. J Fish Biol 91:1642–1667. Mouillot D, Villeger S, Parravicini V et al (2014) Functional overredundancy and high functional vulnerability in global fish faunas on tropical reefs. Proc Natl Acad Sci 38:13757–13762. Oksanen J, Blanchet FG, Kindt R, et al (2015) vegan: Community Ecology Package. R package version 2.3-4. : https://CRAN.R- project.org/package=vegan. Peters H, O'Leary BC, Hawkins JP, et al (2013) Conus: First comprehensive conservation Red List assessment of a marine gastropod mollusc genus. PLoS One 8: e83353. doi:10.1371/journal.pone.0083353 Peters H, O'Leary BC, Hawkins JP, et al (2015) Identifying species at extinction risk using global models of anthropogenic impact. Glob Change Biol 21:618–628. Polidoro BA, Carpenter KE, Ralph GM, et al (2017) The status of marine biodiversity in the Eastern Central Atlantic (West and Central Africa). Aquat Conserv Mar Freshw Ecosyst 27:1021–1034. RGPH, (2012) Atlas Demográfico do Recenciamento Geral da População Humana. Instituto Nacional de Estatística. São Tomé e Príncipe. Richards BL, Williams ID, Vetter OJ, et al (2012) Environmental factors affecting large-bodied coral reef fish assemblages in the Mariana Archipelago. PloS One, 7: e31374. doi:10.1371/journal.pone.0031374 Roberts CM (2007) The Unnatural History of the Sea. Island Press, Washington, DC. Scheffer M, Carpenter S, Foley JA, et al (2001) Catastrophic shifts in ecosystems. Nature 413:591–596. Shanks AL, Wright WG (1986) Adding teeth to wave action: the destructive effects of wave-borne rocks on intertidal organisms.

134

Oecologia 69:420–428. Steneck RS, Dethier MN (1994) A functional group approach to the structure of algal dominated communities. Oikos 69:476–498. Stallings CD (2009) Fishery-Independent Data Reveal Negative Effect of Human Population Density on Caribbean Predatory Fish Communities. PLoS One 4: e5333. doi:10.1371/journal.pone.0005333. Tuya FA, Bosch NEA, Abreu ADB, et al (2017) Reef fish at a remote tropical island (Principe Island, Gulf of Guinea): disentangling taxonomic, functional and phylogenetic diversity patterns with depth. Mar Freswater Res, doi.org/10.1071/MF17233. Trussell GC (2002) Evidence of countergradient variation in the growth of an intertidal snail in response to water velocity. Mar Ecol Prog Ser 243:123–131. Trygonis V, Sini M (2012) photoQuad: a dedicated seabed image processing software, and a comparative error analysis of four photoquadrat methods. Mar Biol Ecol 424:99–108. Vogel S (1994) Life in moving fluids: the physical biology of flow, 2nd edn. Princeton University Press, Princeton. Williams ID, Walsh WJ, Schroeder RE, et al (2008) Assessing the importance of fishing impacts on Hawaiian coral reef fish assemblages along regional-scale human population gradients. Environ Conserv 35:261–272. Williams ID, Richards BL, Sandin SA, et al (2011) Differences in reef fish assemblages between populated and remote reefs spanning multiple archipelagos across the Central and Western Pacific. 2011. doi:10.1155/2011/826234. Williams ID, Baum JK, Heenan A, et al (2015). Human, oceanographic and habitat drivers of central and western Pacific coral reef fish assemblages. PLoS One 10, e0120516. Wirtz P, Ferreira CEL, Floeter SR, et al (2007) Coastal fishes of São Tomé and Principe islands, Gulf of Guinea (Eastern Atlantic Ocean) – an update. Zootaxa 48:1–48. Wirtz P, Brito A, Falcon JM, et al (2013) The coastal fishes of the Cape Verde Islands – new records and an annotated checklist. Spixiana 36:113–142.

135

Supporting information

Figure S.1: Non-metric multidimensional scaling of structure of abundance of functional groups (A) and structure of abundance of size classes (B) of fishes in São Tomé (Yellow vs. Príncipe (Blue) islands in 2016. Each circle is a sampled UVS and its size is proportional to its total abundance. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

Figure S.2: Non-metric multidimensional scaling of structure of abundance of functional groups (A) and structure of abundance of size classes (B) of fishes in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing. Vectors depict the correlations of functional groups and size classes with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS).

136

Figure S.3: Non-metric multidimensional scaling of structure of biomass of functional groups per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of functional groups with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS). Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB).

137

Figure S.4: Non-metric multidimensional scaling of structure of biomass of size classes per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing biomass. Vectors depict the correlations of size classes with the nMDS axes. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB).

138

Figure S.5: Non-metric multidimensional scaling of structure of abundance of functional groups per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing abundance. Vectors depict the correlations of functional groups with the nMDS axes: Functional groups: herbivores-detritivores (HD), macroalgae-feeder (HM), omnivores (OM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), and piscivores (PS). Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete Pedras Branca (SPB).

139

Figure S.6: Non-metric multidimensional scaling of structure of abundance of size classes per sites in São Tomé island in 2006 (Orange) vs. 2016 (Yellow). Each circle is a sampled UVS and its size is proportional to its total standing abundance. Vectors depict the correlations of size classes with the nMDS axes. Diferentes sites: Diogo Vaz (DVA), Ilhéu das Cabras (KIA), Ilhéu Santana (SAN), Sete pedras Branca (SPB).

Table S.1: Abundance and biomass estimates of fish species recorded during underwater visual census (343 transects) at São Tomé island in (2006 and 2016) and Príncipe island in (2016). Trophic group: herbivores-detritivorous (HD), macroalgal-feeder (HM), sessile invertebrate feeders (IS), mobile invertebrate feeders (IM), planktivores (PK), piscivorous (PS) and omnivores (OM).

Abundance Abundance Abundance Biomass STI Biomass STI Biomass PRI Trophic STI 2006 STI 2016 PRI 2016 Family Species 2006 (individuals 2016 (individuals 2016 (individuals Group (individuals (individuals (individuals per m2) ± s.e. per m2) ± s.e. per m2) ± s.e. per m2) ± s.e. per m2) ± s.e. per m2) ± s.e. 1110.66 ± 1344.82 ± Pomacentridae Abudefduf hoefleri OM 15.00 ± 2.01 22.54 ± 5.50 101.33 303.97 Pomacentridae Abudefduf saxatilis OM 4.30 ± 0.42 124.32 ± 21.74

Pomacentridae Abudefduf taurus OM 1.00 ± 0.00 2.41 ± 0.00

Ostraciidae Acanthostracion notacanthus IS 2.00 ± 0.11 6.5 ± 0.40 42.47 ± 3.52 439.54 ± 4.12

2337.36 ± Acanthuridae Acanthurus monroviae HD 4.41 ± 0.68 1.96 ± 0.20 4.96 ± 0.95 896.57 ± 282.44 213.18 ± 28.50 635.80 Monacanthidae Aluterus scriptus PK 2.88 ± 0.42 798.68 ± 126.45

160.00 ± Apogonidae Apogon affinis PK 325.58 ± 38.91 19.12 Apogonidae Apogonimberbis PK 36.88 ± 7.16 3.88 ± 0.44 7.75 ± 0.83 75.04 ± 14.59 20.23 ± 2.76 12.72 ± 1.61 Apogonidae Apogonpseudomaculatus PK 2.00 ± 0.09 5.37 ± 0.56 11.59 ± 0.39 29.38 ± 4.81

Lutjanidae Apsilus fuscus IM 1.00 ± 0.00 12.00 ± 0.70 38.21 ± 4.48 79.29 ± 4.64

Aulostomidae Aulostomusstrigosus PS 4.73 ± 0.65 1.36 ± 0.06 1.53 ± 0.12 47.58 ± 10.20 327.84 ± 39.57 100.51 ± 9.30 Balistidae Balistes capriscus IM 2.00 ± 0.00 5.46 ± 0.00

Balistidae Balistes punctatus IM 2.05 ± 0.20 1.00 ± 0.00 1.33 ± 0.09 60.38 ± 8.81 152.74 ± 11.27 231.78 ± 32.12 Labridae Bodianus pulchellus IM 3.54 ± 0.48 4.45 ± 0.69 1.45 ± 0.10 184.31 ± 38.89 61.41 ± 7.09 156.14 ± 9.37

140

141

Labridae Bodianus speciosus IM 1.78 ± 0.17 1.00 ± 0.00 2.06 ± 0.22 82.15 ± 15.66 9.23 ± 0.94 22.45 ± 6.63 Bothidae Bothus guibei IM 1.00 ± 0.00 55.36 ± 0.00

Bothidae Bothus lunatus PS 3.25 ± 0.14 32.25 ± 5.17

Monacanthidae Cantherhines pullus OM 1.84 ± 0.14 2.20 ± 0.19 1.29 ± 0.07 76.18 ± 11.14 136.01 ± 10.49 135.56 ± 9.87 Balistidae Canthidermis sufflamen PK 2.85 ± 0.14 95.97 ± 24.86

Tetraodontidae Canthigaster supramacula IM 4.70 ± 1.05 3.31 ± 0.28 2.72 ± 0.27 24.52 ± 4.87 8.71 ± 0.71 5.64 ± 0.77 Carangidae Carangoides bartholomaei IM 4.50 ± 0.40 1.00 ± 0.00 1719.14 ± 99.22 2056.86 ± 60.73

Carangidae Caranx crysos PS 1.00 ± 0.00 257.22 ± 23.45

Carangidae Caranx latus IM 4.42 ± 0.42 31.42 ± 6.50

Serranidae Cephalopholis nigri IM 1.81 ± 0.15 1.11 ± 0.04 1.44 ± 0.10 3.18 ± 0.28 246.85 ± 20.75 144.91 ± 24.93 Serranidae Cephalopholis taeniops PS 2.47 ± 0.37 1.24 ± 0.06 1.34 ± 0.07 36.98 ± 11.27 119.27 ± 2.26 191.47 ± 22.89 Chaetodontidae Chaetodon robustus IM 1.64 ± 0.06 1.70 ± 0.05 1.00 ± 0.00 92.11 ± 9.57 52.76 ± 2.90 52.46 ± 0.00 Muraenidae Channomuraena vittata PS 1.00 ± 0.00 1.00 ± 0.00 1.00 ± 0.00 61.21 ± 1.49 223.74 ± 0.00 818.58 ± 33.41 102.96 ± 115.20 ± Pomacentridae Chromis multilineata PK 98.33 ± 14.47 275.34 ± 83.17 512.31 ± 89.22 680.29 ± 169.14 15.28 16.84 Cirrhitidae Cirrhitus atlanticus IM 1.75 ± 0.16 1.07 ± 0.03 1.05 ± 0.03 34.72 ± 3.99 165.30 ± 28.76 88.82 ± 16.00 2806.64 ± Labridae Clepticus africanus PK 3.07 ± 0.44 4.60 ± 0.39 11.00 ± 1.25 588.88 ± 87.07 184.75 ± 20.12 490.09 Labridae Coris atlantica IM 1.28 ± 0.05 1.16 ± 0.04 2.45 ± 0.41 36.07 ± 2.26 19.15 ± 1.48 53.80 ± 8.49 Gobiidae Didogobius amicuscaridis IM 3.50 ± 0.05 5.51 ± 0.09

Diodontidae Diodon holocanthus IM 1.20 ± 0.04 1.00 ± 0.00 126.59 ± 4.81 210.30 ± 0.00

10109.21 ± Diodontidae Diodon hystrix IM 1.00 ± 0.00 524.51

Muraenidae Enchelycorenigricans IM 1.27 ± 0.07 1.00 ± 0.00 1.00 ± 0.00 86.91 ± 5.75 895.70 ± 58.67 513.11 ± 37.25 1114.43 ± Serranidae Epinephelus adscensionis IM 1.29 ± 0.07 1.00 ± 0.00 1.00 ± 0.00 165.68 ± 16.12 1188.13 ± 36.00 119.18 Gobiidae Gnatholepis thompsoni OM 4.45 ± 0.41 4.96 ± 0.52 13.55 ± 1.40 6.73 ± 0.63 3.03 ± 0.94 3.81 ± 0.60 Gobiidae Gobius rubropunctatus IM 6.71 ± 0.56 4.31 ± 0.47 2.06 ± 0.16 15.25 ± 2.58 29.10 ± 3.67 5.05 ± 0.57 Gobiidae Gorogobius nigricinctus IM 2.00 ± 0.09 2.00 ± 0.16 2.08 ± 0.22 3.14 ± 0.15 1.48 ± 0.01 3.44 0.62 Gobiidae Gorogobiusstevcici IM 5.44 ± 0.61 8.57 ± 0.96

3474.75 ± Muraenidae Gymnothoraxvicinus PS 2.33 ± 0.31 1.00 ± 0.00 1.00 ± 0.00 991.85 ± 41.75 636.55 ± 44.59 482.79 Congridae Heterocongerlongissimus PK 15.00 ± 00 506.62 ± 0.00

Priacanthidae Heteropriacanthuscruentatus IM 2.67 ± 0.27 1.43 ± 0.11 1.96 ± 0.11 391.43 ± 47.13 245.96 ± 19.27 498.73 ± 43.23 Pomacanthidae Holacanthusafricanus IS 1.51 ± 0.11 1.5 ± 0.13 1.38 ± 0.07 52.15 ± 10.75 114.79 ± 10.27 213.97 ± 40.93 Holocentridae Holocentrusadscensionis IM 7.44 ± 0.99 3.86 ± 0.91 2.46 ± 3.06 355.27 ± 82.03 686.42 ± 143.06 571.28 ± 102.84 1196.85 ± 6156.82 ± Kyphosidae Kyphosusspp HM 4.11 ± 0.42 11.00 ± 1.53 133.36 787.12 Labrisomidae Labrisomus nuchipinnis IM 2.00 ± 0.00 97.45 ± 0.00

Lethrinidae Lethrinusatlanticus IM 11.00 ± 1.26 2.00 ± 0.17 5.77 ± 1.25 653.72 ± 82.83 287.36 ± 47.65 357.66 ± 70.86 1384.76 ± Lutjanidae Lutjanus agennes IM 1.00 ± 0.00 3.25 ± 0.26 1370.03 ± 40.34 140.28 Lutjanidae Lutjanusendecacanthus IM 53.00 ± 4.71 5.70 ± 0.46 605.94 ± 66.75 117.04 ± 18.73

Lutjanidae Lutjanusfulgens IM 15.00 ± 1.94 1.00 ± 0.00 275.84 ± 56.15 273.43 ± 0.00

5138.87 ± 1914.71 ± Lutjanidae Lutjanusgoreensis IM 1.94 ± 0.22 5.25 ± 0.36 4.33 ± 1.23 885.84 ± 308.23 395.39 456.73 Balistidae Melichthysniger HD 1.00 ± 0.00 510.74 ± 13.68

142

143

Pomacentridae Microspathodon frontatus HD 3.32 ± 0.25 4.37 ± 0.39 5.72 ± 0.69 182.95 ± 20.66 654.22 ± 90.16 866.64 ± 168.31 1359.86 ± 3436.76 ± Mullidae Mulloidichthys martinicus IM 13.20 ± 1.99 16.14 ± 3.28 15.79 ± 2.63 621.58 ± 123.21 297.52 718.90 1031.70 ± Muraenidae Muraena melanotis PS 1.37 ± 0.09 1.00 ± 0.00 1.11 ± 0.04 407.85 ± 25.06 390.75 ± 45.45 149.74 Muraenidae Muraena robusta PS 1.50 ± 0.05 1.00 ± 0.00 77.30 ± 3.82 2415.94 ± 0.00

Ophichthidae Myrichthys pardalis IM 2.00 ± 0.22 1.00 ± 0.00 9.08 ± 0.52 8.99 ± 0.00

3308.05 ± 2331.60 ± 1701.83 ± Holocentridae Myripristis jacobus PK 68.72 ± 14.51 38.67 ± 9.11 23.86 ± 3.74 995.15 534.66 354.84 Blenniidae Ophioblenniussp HD 4.13 ± 0.33 2.15 ± 0.17 3.38 ± 0.32 15.83 ± 6.51 20.17 ± 3.23 13.85 ± 1.59 101.52 ± 1129.99 ± 1507.22 ± Serranidae Paranthias furcifer PK 29.02 ± 5.31 38.36 ± 6.86 326.84 ± 59.85 16.00 322.28 457.08 Haemulidae Pomadasysincisus IM 18.90 ± 1.18 1335.16 ± 98.14

1390.47 ± 1042.64 ± 1230.37 ± Acanthuridae Prionurusbiafraensis HM 12.81 ± 4.25 2.21 ± 0.23 4.28 ± 0.90 334.01 156.99 241.00 Chaetodontidae Prognathodesmarcellae IM 3.00 ± 0.23 2.00 ± 0.00 11.61 ± 0.93 105.53 ± 0.00

Mullidae Pseudupeneusprayensis IM 5.59 ± 1.15 3.80 ± 0.41 1.00 ± 0.00 59.26 ± 34.94 253.72 ± 31.06 119.97 ± 0.00 Serranidae Rypticus sp IM 1.74 ± 0.13 1.51 ± 0.09 1.21 ± 0.06 108.79 ± 9.03 644.16 ± 43.68 317.33 ± 22.81 1081.37 ± Holocentridae Sargocentronhastatum IM 23.75 ± 10.49 1.71 ± 0.14 1.00 ± 0.00 182.21 ± 17.41 154.16 ± 7.75 543.54 Scaridae Scarus hoefleri HD 1.76 ± 0.14 1.00 ± 00 1.64 ± 0.12 273.59 ± 40.25 1727.84 ± 0.00 519.06 ± 72.64 Scorpaenidae Scorpaenodesafricanus HD 1.00 ± 0.00 0.03 ± 0.00

Serranidae Serranuspulcher IM 4.09 ± 0.50 1.00 ± 0.00 4.14 ± 0.48 58.12 ± 25.28 2.03 ± 0.00 4.36 ± 0.63 Scaridae Sparisomachoati HD 2.08 ± 0.14 2.03 ± 0.16 1.78 ± 0.14 152.73 ± 26.91 202.35 ± 32.50 581.55 ± 77.46 5271.52 ± 3393.83 ± 3024.36 ± Sparidae Spicaranigricauda PK 93.28 ± 11.76 56.66 ± 4.35 43.83 ± 2.75 811.92 291.70 205.36

Pomacentridae Stegastes imbricatus HD 6.68 ± 0.56 4.12 ± 0.49 5.18 ± 0.48 15.12 ± 1.27 33.87 ± 4.63 53.81 ± 15.78 Synodontidae Synodus synodus PS 2.14 ± 0.16 1.00 ± 0.00 35.06 ± 5.10 67.15 ± 7.03

Labridae Thalassoma newtoni PK 9.17 ± 1.08 5.01 ± 0.71 9.37 ± 1.11 42.08 ± 11.15 24.74 ± 3.16 33.49 ± 6.23 Gobiidae Wheelerigobius maltzani IM 1.81 ± 0.12 3.00 ± 0.30 0.32 ± 0.02 0.40 ± 0.04

Gobiidae Wheelerigobius wirtzi IM 4.06 ± 0.40 4.51 ± 0.51 13.20 ± 1.30 6.40 ± 0.64 1.46 ± 0.16 3.60 ± 0.38 Mean total 20.65 ± 7.66 13.21 ± 5.28 15.77 ± 5.79 479.42 ± 280.48 423.47 ± 169.46 543.26 ± 229.20

144

145

Table S2: number of Underwater Visual Censuses per sites in two sampling islands (São Tomé in 2006 vs. 2016) and (São Tomé vs. Príncipe in 2016).

Locality Sites UVC

Príncipe_2016 Praia Banana (BAN) 13

Príncipe_2016 Ilhéu Boné do Joquei (BJO) 18

Príncipe_2016 Ilhéu Bombom (BOM) 24

Príncipe_2016 Pedra da Galé (GAL) 31

Príncipe_2016 Praia da Lapa (LAP) 26

Príncipe_2016 Ilhéu dos Mosteiros (MOS) 23

Príncipe_2016 Ilhas Tinhosas (TIN) 18

S. Tomé_2006 Diogo Vaz (DVA) 39

S. Tomé_2006 Ilhéu das Cabras (KIA) 28

S. Tomé_2006 Ilhéu das Rolas (ROL) 21

S. Tomé_2006 Ilhéu Santana (SAN) 27

S. Tomé_2006 Sete Pedras Branca (SPB) 13

S. Tomé_2006 Sete Pedras Catedral (SPC) 11

S. Tomé_2016 Diogo Vaz (DVA) 12

S. Tomé_2016 Ilhéu das Cabras (KIA) 12

S. Tomé_2016 Ilhéu Santana (SAN) 15

S. Tomé_2016 Sete Pedras Branca (SPB) 12

Total 343

146

147

CAPÍTULO III

Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea

(Publicado na Ocean and Coastal Management) Formatado de acordo as normas da revista

Maia, H.A., Morais, R.A., Siqueira, A.C. et al. (2018). Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea. Ocean and Coastal Management.

148

Shifting baselines among traditional fishers in São Tomé and Príncipe islands, Gulf of Guinea

Maia H.A.1,2,3,4, Morais R.A.5,6, Siqueira A.C. 5,6, N. Hanazaki1, Floeter S.R.1,2, Bender M.G.2,7

1 Programa de Pós-graduação em Ecologia, Universidade Federal de Santa Catarina, Florianópolis, SC 88040-900, Brazil.

2 Laboratório de Biogeografia e Macroecologia Marinha, Departamento de Ecologia e Zoologia, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, Florianópolis, SC 88040-900, Brazil.

3 Programa de Pós-graduação Ciência para o Desenvolvimento (PGCD), Instituto Gulbenkian de Ciência, Oeiras, 62780-156, Portugal.

4 Departamento de Ciências Naturais e Biologia, Universidade de São Tomé e Príncipe, Bairro Quinta de Santo António, São Tomé and Príncipe.

5 College of Science and Engineering, James Cook University, Townsville, QLD 4811, Australia.

6 ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia.

7 Departamento de Ecologia e Evolução, Universidade Federal de Santa Maria, Santa Maria, RS 97105-900, Brazil.

______*Correspondence author: Hugulay A. Maia, [email protected] Target Journal: Ocean and Coastal Management

148

149

Abstract

Local ecological knowledge has filled baseline gaps in conservation biology, providing important information that has contributed to resource management policies both on land and at sea. Marine ecosystems are globally threatened by overfishing, yet we know little on if and how fishers perceive changes in fisheries composition through time. This is particularly important in developing nations where people rely on fishery resources as their main source of food and income. We interviewed 178 artisanal fishers to collect information regarding their perceptions on the trends and composition of reef fisheries in São Tomé and Príncipe islands, a marine biodiversity hotspot. In addition, we investigated the relative contribution of the possible factors causing changes in these reef fish assemblages according to fishers’ perceptions. Of six reef fish species assessed, five exhibited significant declining catch trends. We found a declining trendin individual body size for targeted species based on reports from older (mean± S.E. = 43.3 ± 2.6 kg) and younger (21.0 ± 0.7 kg) fishers’ generations. Generations also differed in their perceptions of declines over time, all of the very experienced fishers reported decline, while only one-third of inexperienced fishers did so. The main causes for fish catch changes identified by experienced fishers (>40 years of fishing practice) were the increasing number of fishers (25%), destructive fishing practices (mainly blast fishing) (18%) and industrial fishing (29%). Our results suggest the occurrence of the shifting baseline syndrome phenomena among traditional fishers and provide baseline information for the conservation and management of São Tomé and Principe marine ecosystems.

150

Keywords: Overfishing, Blast fishing, Gulf of Guinea, Western Africa, Artisanal Fisheries, Tropical Fisheries

150

151

1. Introduction Ever since humans began collecting seafood along coastal areas, there have been impacts and changes on marine diversity (McCauley et al., 2015). Anthropogenic impacts have escalated through time and driven profound changes to marine systems, especially in the past few centuries (Jackson, 1997; Jackson et al., 2001; Halpern et al., 2008). The increase in human impacts is associated with population growth, and the augmented need for space and a multitude of resources, including marine ones. The industrialization era marked the expansion of fishing activities across the oceans, upscaling the impacts of human exploitation from local to global (Jackson et al., 2001; Roberts, 2007). Consequently, few marine environments have been left as pristine, i.e. locations that maintain relatively untouched biodiversity. Most of these pristine environments are located in remote, unpopulated and/or strictly protected areas (Edgar et al., 2014; Maire et al., 2016). While the majority of marine environments are heavily impacted by human activities (e.g. Jackson et al., 2001; Bellwood et al., 2004) there is still the need for a proper baseline knowledge of the past natural conditions of the oceans to better assess the magnitude of these impacts. Fisheries have direct effects on fish populations, causing declines in overall abundance and reductions in average fish body size (Jennings and Kaiser, 1998; Myers and Worm, 2003; Jennings and Blanchard, 2004). Additionally, fisheries can have indirect effects that include impacts on marine habitats from destructive fishing practices such as bottom trawling and blast fishing (Roberts, 2007). While bottom trawling is mostly practiced in temperate zones (Dayton et al., 1995), blast fishing (using dynamite, grenades or other explosive mixtures) is often employed on tropical reefs (Ruddle 1996; Jennings and Kaiser, 1998; DeMartini et al. 1999). This activity has been particularly prevalent in low-income countries with declining fish catches, especially in Southeast Asia and Africa (Alcala and Gomez, 1987; Edinger et al., 1998; Ruddle 1996; Pet- Soede and Erdman, 1998; Fox et al. 2003; Wells 2009, Morais and Maia, 2016). Although blast fishing causes and consequences are well known (Alcala and Gomez, 1987; Fox et al. 2003; Wells, 2009), we still have a poor knowledge on how pervasive is this practice among artisanal fisheries. Ethnographic methods are used in several scientific domains (e.g. medicine, anthropology, agriculture) and have also been applied in conservation biology to expand knowledge on past environmental conditions (Shackeroff and Campbell, 2007; Ternes et al., 2016; Braga et

152

al., 2017; Lima et al., 2017; Turvey et al., 2017).The use of ethnographic methods such as interviews and participant observation enables the collection of data about local ecological knowledge (LEK) (Gadgil et al., 1993; Colding, 1998; Johannes, 1998; Berkes, 1999). The LEK of individuals and communities can reveal important insights regarding marine ecosystems (Ruddle, 1994a; 1994b; and 1994c; Silvano and Begossi, 2010). For example, at a local scale, the assessment of LEK from fishers can reveal valuable information about life history characteristics of fish species (Ruddle, 1993). In addition, LEK can provide important information for local resources management and policy design. The assessment of local ecological knowledge has also been critical for the investigation of the shifting baseline syndrome (SBS; Pauly, 1995). This phenomenon describes how the baseline information of past conditions biases the characterization of present environmental states. In fisheries science, detecting the SBS depends on comparisons of present day conditions with the most pristine state as possible (i.e. before human expansion) (Pauly, 1995; Pandolfi et al., 2003). In the last 20 years, marine scientists have conducted several studies to demonstrate how our baseline shifted and its implications for conservation.Those studies have reported declines in the number of fished species, associated with declines in fish abundance (McClanahan et al., 2007; Mora et al., 2011; Sandin et al., 2008), fish sizes (DeMartini et al., 2008; Kittinger et al., 2011; McClenachan, 2008) and biomass (Friedlander and DeMartini, 2002; Graham and McClanahan, 2013; McClanahan et al., 2007). Additionally, some studies reported changes in species’ reproductive potential (DeMartini and Smith, 2015) and in food web structure (Sandin et al., 2005; Sandin and Zgliczynki, 2015; Sala and Sugihara, 2005; Trebilco et al., 2013; Williams et al. 2011) as well as declines in reef resilience (Sandin et al., 2005; McClanahan, 2015). The perception of different fishers generations has also been used to describe changes in fisheries through time and to assess the occurrence of the SBS (e.g. Turvey et al., 2010; Baisre, 2013; Hanazaki et al., 2013; Bender et al., 2014; Plumeridge and Roberts, 2017; Sáenz-Arroyo et al., 2005). These studies allow us to make comparisons to better understand how fisheries have changed through time and how fishers perceived these gradual changes throughout generations. Despite being a marine hotspot (Roberts et al., 2002), São Tomé and Príncipe (STP), located in the Gulf of Guinea, Western Africa, has been understudied in many aspects of marine ecology, including the local trends in artisanal fisheries. Yet, its reef fish diversity has been relatively well studied in STP in the last decade, with 234 recorded species, 3% 152

153

being endemic (Wirtz et al., 2007; Floeter et al., 2008). In the present study, we assessed the LEK of fishing communities to describe local fisheries composition and catch trends through time in STP. We also investigated the drivers of the detected trends in local fisheries and mapped the distribution of one of these drivers. Specifically, we intended to: (1) investigate changes in fisheries species composition, maximum catches and maximum body size of fish caught in STP in the past decades; (2) investigate the main causes of changes according to the LEK of fishers; and (3) to spatialize the prevalence of destructive fishing (blast fishing) in STP. Results from this work will contribute to establish a baseline for future management strategies for marine ecosystems in STP.

2. Material and methods

2.1. Study area description São Tomé and Principe is an archipelago with two main islands (São Tomé island – ST and Príncipe island - PR) located in the Gulf of Guinea, West Africa (Fig. 1). The Democratic Republic of São Tomé and Príncipe is a Small Island Developing State (SIDS) (Frynas et al., 2003; Briguglio, 1995; Monnereau and Failler, 2014) that regained independence in 1975, after five centuries of Portuguese colonization. Presently, São Tomé and Príncipe have a population of approximately 190,000 people (RGPH, 2012). Fishing contribute with more than 80% of animal protein consumed by the population (RDSTP, 2009). Attempts to order local fisheries were first implemented by Portugal, through the Serviço de Capitania dos Portos (literally, port captaincy service) in the 1950s. At that time, a Chefe de Praia (literally, chief of the beach) was appointed to supervise fishing activities on each district (São Tomé and Príncipe has six districts, five in ST and one in PR) along the 30 fishing communities that existed in the two main islands (Hodges and Newitt, 1988). Today, both industrial and artisanal fishing occur in São Tomé and Príncipe. Industrial fishing is practiced in the Exclusive Economic Zone (EEZ) of São Tomé and Príncipe, mainly by the European Union, Japan, Taiwan and China (vessels ranging from 160 to 290 feet) (Frynas et al., 2003; Krakstad, 2010; Carneiro, 2011; Serigstad et al., 2012). The present work focuses on artisanal fishing, which takes place mainly in traditional fishing villages (Fig. 1C-D). These villages hosted 2,428 active fishers in 2010 (88% in ST and 12% in PR) (DGAP-RDSTP, 2010). Most artisanal fishers of São Tomé and Príncipe are descendants of the Angolar people that occupy especially the western and eastern portions of São Tomé island (Ceita, 1991). Fishing is mainly practiced by

154

men using wooden canoes while the fish is marketed by women called palayês (literally, saleswomen) (Ceita, 1991). The traditional Santomean canoes (or Dongos, Figueiredo, 1966) are made from a single trunk of local trees (normally Ceiba pentandra, but also other species). Large Dongos can reach 30 feet long, but on average measure 15 to 20 feet (Horemans et al., 1994) and are operated with paddles by one to three men. Only more recently (after the decades of 1970 or 1980), did Dongos start to be equipped with outboard motors of four to 40 Hp. Motors, lines, hooks and the Dongos themselves, which rarely last more than five years, constitute a heavy money investment for the fishers (Costa, 1959; Hodges and Newitt, 1988). The eight main types of fishing gear used by artisanal fishers in São Tomé and Príncipe are described in Table 1.

Table 1: The eight main types of fishing gear used by artisanal fishers in São Tomé and Príncipe. Fishing gear Description This technique is applied to four types of fishing in the São Tomé and Line Príncipe islands: Corrico, palangre de fundo, and bolo [the fishermen pull the lines with their hands, it is usually practiced on top of the canoe, but sometimes it is observed that it is practiced by women and children from the coast]. The diameters of the lines vary according to the target species. In the São Tomé and Príncipe islands there are three main variants of fishing nets: Surrounding net (length between 100-2000 meters, height Fishing net from two to 25 meters and 13/29mm mesh), drift net (length between 1000-1500 meters, height ~870 meters and 210/6mm mesh) and bottom set gill net (length ~300 meters, height ~4 meters and 13mm mesh). Voador panhan Is the most widespread fishing gear, and consists in capturing flying- fish (family Exocoetidae) with the use of floating traps made of herbs or grass (Costa, 1959, Horemans et al., 1994). Spearfishing Is practiced by the younger fishers and involves the capture of mainly reef organisms (e.g. octopuses, cuttlefish, morays and rays). Casting net Is an almost extinct fishing gear, being practiced almost exclusively by fishers over 70 years old. Pingue/palangre Is a composite of hundreds of hooks hanging over the seabed that de fundo targets bottom species such as the Bluespotted seabream (Pagrus caeruleostictus) and the flying gurnard (Dactylopterus volitans). Corrico The Corrico gear uses bait (e.g. Euthynnus spp. and Decapterus spp.) towed by a motorized canoe and targets mainly sharks. Candieiro is practiced at night, with the use of light to attract and Candieiro capture fish.

154

155

São Tomé and Príncipe has a warm equatorial oceanic climate that receives seasonally the influence of dry and cool southern winds (Horemans et al., 1994). This result in two main seasons in the country: a dry season, from June to August, and a rainy season that occurs in the remaining months.

Figure 1: Location of São Tomé and Príncipe islands (STPI) in the Atlantic Ocean: (A) Gulf of Guinea highlighted in the grey box; (B) STPI; (C) São Tomé island; and (D) Príncipe island. ( = Traditional fishing communities). ST: SCA – Santa Catarina; PBI – Praia Brita; PSP – Praia de São Pedro; PME – Praia Melão; PMA – Praia Messias Alves; PSA – Praia de Angolares; PRP – Praia de Ribeira Peixe; PPA – Praia de Porto Alegre. PR: LAP - Praia da Lapa; BUR – Praia das Burras; SAN – Santo António; ABA – Praia de Abade; SEC – Praia Seca.

2.2. Data collection We interviewed fishers in both São Tomé and Príncipe islands (Fig. 1C and D) from December 2015 to March 2016 using a structured questionnaire. Fishers were intentionally approached at the beach (Silvano et al., 2006; Bender et al., 2013) as they were leaving for fishing or repairing Dongos or nets. Before each interview, we informed fishers about the objectives of the study and asked for their consent to take part

156

on the research. Interviews were recorded only after verbal consent since many fishers are unable to write or sign their own names. Interviews were mostly conducted in Portuguese but in some cases were held in local languages (Santome and Angolar) by the first author. To maximize the expression of personal experiences, each fisher was interviewed individually. Retired fishers indicated by other peers were visited in their households. The structured questionnaire consisted of closed and open questions, where the interviewees had the opportunity to elaborate on answers as desired (see questions in Appendix). The first part of the questionnaire was used to determine the sociological characteristics of fishers (name, birth place, age, fishing gear and fishing experience in years). The second part consisted of displaying fish species images to the interviewees and requesting them to assign the name of those they recognized (based on Silvano et al., 2006). Following each fish species identification, they were questioned on: 1) The average size of individuals caught in the present; 2) the largest individual ever caught (in kilograms); 3) when and where those largest fish were caught; 4) fishing gear used; and 5) the best day’s catch, i.e., the greatest catch, in kilograms, that they remembered landing for each species. These questions were repeated for every species identified by fishers. The last part of the questionnaire comprised questions on their environmental perceptions: 1) past and present catch composition, 2) trends in catch over time, and 3) impacts on fish abundance in the region. To assess catch trends over time, interviewees were asked specifically to compare the current fisheries situation with that when they started fishing and to define potential trends over time. We used photographic images from six commercially targeted fish species in STP islands (Costa, 1959; EST, 1982): the Dungat grouper, Epinephelus goreensis (Valenciennes, 1830); the Great barracuda, Sphyraena barracuda (Edwards, 1771); theWhite grouper, Epinephelus aeneus (Geoffroy St. Hilaire 1817); the African red snapper, Lutjanus agennes (Bleeker, 1863); the Gorean snapper, Lutjanus goreensis (Valenciennes, 1830); and the Bluespotted seabream, Pagrus caeruleostictus (Valenciennes, 1830). We also questioned fishers on other large-sized fish species caught during their fishing practice. 2.3. Data analysis Fishers were categorized into four groups according to their time devoted to fishing activities: inexperienced (≤15 years of practice, 53 fishers); intermediate (16–30 years of practice, 48 fishers); experienced

156

157

(31–40 years of practice, 37 fishers); and very experienced (>40 years of practice, 40 fishers). It is important to highlight that this classification is solely based on time devoted to fishing, in years, and brings no qualitative judgement about the real experience of each fishers. We acknowledge that this classification assumes that all fishers spend the same proportion of their year devoted to fishing, and that all fishers are able to acquire experience (in the sense of "knowledge") at the same rates. Nevertheless, this classification allowed us to quantify trends in fish catches, as well as the differences in overall perceptions of these trends across fishers’ generations.A simple linear regression was used to evaluate how the best day’s catch reported by each respondent varied according to the time of that fishing event (in years before present), for each species. To assess whether maximum body size of fish caught in STP in the past decades varied with fishers’ experience (inexperienced, intermediate, experienced and very experienced), we applied a generalized linear model – GLM – with a Gamma distribution. To assess whether possibly detected changes were due to changes in target species that attained different sizes rather than changes in size per se, we modelled how maximum body size of captured fish responded to both experience level and fish species. To detect differences among groups, we used a post-hoc Tukey test built upon the GLM, with the function “glht” within the package multcomp (Hothorn et al., 2008). Those fishers who described changes in catches over time were also questioned regarding the possible cause of changes. We further quantified and grouped the main causes of change mentioned in categories of problems, or drivers of change. The analyses of the fishing impact were conducted by island. Given the repetitive accounts on the use of blast fishing in the region (Morais and Maia, 2016), we included this topic in our interviews. Specifically, we asked if 1) they had ever heard about blast fishing occurring in their region and, if so, 2) if they had ever witnessed blast fishing and, if so 3) if they had ever practiced blast fishing. To uncover the most affected sites from blast fishing, we further asked interviewees the fishing grounds in which blast fishing has been used and spatialized this distribution in a map for each island.

3. Results

3.1. Changes in targeted fish species We interviewed 178 (~7.3% of 2,428) active fishers (see table 1, supplementary material). All were men, with ages ranging from 16 to 92 years old (average=43 years). In Príncipe island we interviewed fishers

158

from five traditional villages: Praia da Lapa (n=3), Praia das Burras (n=25), Santo António (n=1), Praia de Abade (n=12) and Praia Seca (n=1) (Figure 1D); and in São Tomé island we interviewed fishers from eight villages: Santa Catarina (n=35), Praia Brita (n=7), Praia de São Pedro (n=3), Praia Melão (n=7), Praia Messias Alves (n=25), Praia de Angolares (n=12), Praia de Ribeira Peixe (including the communities of Yó-Grande and Praia Pesqueira) (n=17) and Praia de Porto Alegre (including the community of Malanza) (n=30) (Fig. 1C). Among the interviewed fishers, 81% (145) reported changes in fisheries composition and/or catches in São Tomé and Príncipe over time, while18% (33) reported no change. The vast majority of these 145 fishers reported declines in fish sizes and catch trends (90%). The proportion of fishers that reported a decline in catches over time differed according to experience categories: all of the very experienced fishers reported a decline, compared to 75% of experienced and 50% of intermediately experienced, and to only 35% of inexperienced fishers. Only inexperienced or intermediately experienced fishers reported no change in fisheries composition over time, whereas only inexperienced fishers reported an increase in the captures. Regarding changes in fisheries composition, 76% (134) of the fishers mentioned species that were not targeted in the past but that have become economically important over time. Among these species, the most cited were: sharks (Nurse shark, Ginglymostoma cirratum; Blue shark, Prionace glauca; Sand tiger shark, Carcharias taurus and Scalloped hammerhead, Sphyrna lewini), West African goatfish, Pseudupeneus prayensis; Broadbanded moray, Channomuraena vittata; Stout moray, Muraena robusta; Creole-fish, Paranthias furcifer; Flat needlefish, Ablennes hians; Atlantic agujon needlefish, Tylosurus acus rafale; Cornetfish, Fistularia tabacaria; Balao halfbeak, Hemiramphus balao; Blackbar soldierfish, Myripristis jacobus; Senegalese rockfish, Scorpaena laevis; Swallowtail seaperch, Anthias anthias; Tripletail, Lobotes surinamensis; Bogue, Boops boops; Chubs, Kyphosus spp.;African sicklefish,Drepane africana; Cape Verde gregory, Stegastes imbricatus;Blackbar hogfish, Bodianus speciosus. Sete Pedras, Ilhéu das Rolas, Ilhéu Santana, Lagoa Azul and Ilhéu das Cabras were sites repeatedly mentioned as good fishing spots in the past that are now considered overexploited by experienced and very experienced fishers for São Tomé. In Príncipe island, the sites most frequently referred to as overexploited in the present were Ilhéu Bombom, Ilhéu dos Mosteiros, Tinhosas and Pedra da Galé. The relationship between the largest fish individuals ever caught by fishers against time of fishing event (years ago) revealed changes in 158

159

fish catch over time (Fig. 2). When looking at each species, the results suggest that best catches have occurred in the past for the White grouper E. aeneus (slope = -0.71; Adjusted R2 = 0.43; p< 0.001; Fig. 2A) and the Dungat grouper E. goreensis (slope = -0.68; Adjusted R2 = 0.41; p< 0.001; Fig. 2B). For the Great barracuda S. barracuda (slope = -0.43; Adjusted R2 = 0.11; p = 0.02; Fig. 2D), the African red snapper L. agennes (slope = -0.26; Adjusted R2 = 0.08; p = 0.03; Fig. 2E) and the Gorean snapper L. goreensis (slope = -0.51; Adjusted R2 = 0.16; p< 0.001; Fig. 2C), a significant but weak negative relationship was found. For P. caeruleostictus the relationship was not significant (slope = -0.14; Adjusted R2 = 0.24; p = 0.22; Fig. 2F). The body size of the largest individual ever caught differed among groups of fishers with different levels of experience, but was not affected by species identity (Table 2; see also Fig. 3A).Very experienced fishers have caught larger fish than all other categories (Tukey test: p< 0.01; mean ± S.E., 43.3 ± 2.6 kg, Fig. 3A). The largest fish caught by experienced and intermediate level fishers had a similar size (26.7 ± 0.8 kg and 26.5 ± 1.1 kg respectively; Fig. 3A), yet both had caught larger fish compared to inexperienced fishers (21.0 ± 0.7 kg; Tukey test: p< 0.001; Figure 3A). For each species, the results for the largest individual caught mirrored those of the largest fish caught (Fig. 4).There was a clear increase in the largest individual caught from inexperienced to very experienced fishers for almost all species analyzed (Fig. 4A-E). The exception was the Bluespotted seabream, P. caeruleostictus, which only the most experienced had caught larger individuals compared to other categories (Fig. 4F).

Figure 2: Relationship between best day’s catch (largest individual caught in kilograms) that fishers remembered and the time of fishing event (years ago) by species.

160

161

Figure 4: Variation between largest individual ever caught (in kilograms) across generations of fishers (years of practice) by species. ( = a fisher’s catch; = mean).

162

163

Table 2: Results of GLM evaluating the influence of fishing time and fish species on the body size of the largest individual ever caught. Bold values indicate p < 0.001.

Source Estimate Std. Error t value p-value (Intercept) 0.066 0.035 1.85 0.0663 Fishing time: very experienced -0.015 0.004 -3.50 0.0006 Fishing time: inexperienced 0.009 0.005 1.82 0.0704 Fishing time: Intermediate 0.004 0.004 0.09 0.9258 Fish species: E. aeneus -0.019 0.035 -0.54 0.5836 Fish species: E. goreensis -0.030 0.036 -0.82 0.4090 Fish species: S. barracuda -0.031 0.035 -0.89 0.3718 Fish species: L. agennes -0.028 0.035 -0.81 0.4195 Fish species: L. goreensis -0.029 0.035 -0.83 0.4060 Fish species: P. caeruleostictus -0.007 0.044 -0.15 0.8741

3.2. Causes of change based on local knowledge Fishers identified seven main causes for changes in local fisheries: 1) increases in the number of fishers (reported by 45; 31% fishers); 2) destructive fishing practice (blast fishing, 32; 22%); 3) presence of industrial fishing (51; 35%); 4) climate change (12; 8%); 5) presence of whales (1; 1%); 6) pollution with chemicals (2; 1%); and 7) black magic (a cultural practice evocating supernatural powers, mentioned by two fishers; 1%). The perceptions of those causes varied among fishers of different experience categories. For example, the proportion of fishers that cited blast fishing as a cause of change in fish catches decreased from very experienced (28%) to experienced (23%), intermediate (22%) and inexperienced fishers (12%) (Fig. 3B). The cause “increase in the number of fishers” had a similar pattern of decline from very experienced (37%) to inexperienced (28%). Contrarily, the perception of industrial fishing as a driver of change in local fisheries increased from very experienced fishers (25%) to inexperienced (44%) (Fig. 3B). Among 136 fishers interviewed in São Tomé island, 68 (50%) affirmed that they have never heard about blast fishing, 19 (14%) fishers have heard but have never witnessed its use, and 49 (36%) had witnessed

164

and knew fishers that practiced this type of fishing. In Príncipe, among 42 interviewed fishers, 15 (36%) declared they have never heard of blast fishing, 13 (31%) have heard but have never seen it, and 14 (33%) declared themselves as witnesses of this practice. None of the fishers declared himself as a blast fisher. Fourteen sites were mentioned as being targeted by blast fishing in São Tomé island, three sites (Ilhéu Santana, Sete Pedras and Ilhéu das Rolas) being the most cited (Figure 5). Figure S1 details community reports for each site. The sites were reported in six of the eight communities where we interviewed fishers. Fishers from the community of Santa Catarina mentioned six blasted sites, followed by the fishers from Porto Alegre who cited five sites. In the other communities, less than three sites were mentioned. Fishers from two nearby communities (Praia de Ribeira Peixe and Praia de Angolares) presented several citations of Sete Pedras (SPE) site. At Príncipe island the sites with the highest citations were Ilhéu dos Mosteiros and Pedra da Galé (Fig. 5B). Fishers from Praia Abade cited seven spots while fishers from Praia das Burras and Praia da Lapa mentioned four spots (Fig. S1B).

165

Figure 5: Spatial distribution of the blast fishing cited by fishers in São Tomé island (A) and Príncipe island (B). Color gradient indicates the degree of impact based on the number of citations by fishers. Sites in São Tomé: SBC – Baia de Santa Catarina; FLZ – Fundão de Lagoa Azul; LAZ – Lagoa Azul; PQI – Praia Quinze; SAN – Ilhéu Santana; ANG – Praia de Angobó; SPE – Sete Pedras; MAL – Baia de Malanza; BPA – Baia de Porto Alegre; ROL – Ilhéu das Rolas; IHA – Praia Inhame; SMG – Ilhéu de São Miguel; PJU – Ponta da Juliana; CAC – Calé Calé. Sites in Príncipe: PPF – Ponta da Pedra Furada; GAL – Pedra da Galé; MOS– Ilhéus dos Mosteiros; SBR – Sete Braças; PBM – Ponta Banana; PNC – Ponta Café; BON – Ilhéu Boné de Jóquei; TIN – Ilhas Tinhosas. Numbers in parentheses represent total interviewed fishers by community.

166

4. Discussion We observed a strong negative relationship between the maximum size of fish caught by fishers and time before present for two of the six species evaluated. For three other species we also observed a significant, albeit weak, negative relationship. Also, fishers with more than 40 years of fishing practice reported having caught larger fish than less- experienced fishers. This pattern was also observed when looking separately at species. For instance, the size of the largest S. barracuda and E. aeneus caught by fishers was smaller for the inexperienced when compared with the intermediate and experienced classes. This suggests a potential decline in the catch trends of São Tomé and Príncipe’s artisanal fisheries. A related similar pattern was observed for the perceptions of fishers regarding fishing activities over time: while all of the very experienced fishers declared that fisheries have declined, only 35% of the inexperienced fishers have recognized that pattern. Fishers who were not able to perceive changes in fisheries catch and composition were either inexperienced or intermediately experienced. Furthermore, all fishers that reported an increase in catches were inexperienced ones. These results collectively suggest that fishers in São Tomé and Príncipe might have been experiencing the Shifting Baselines Syndrome (Pauly, 1995). The declining trend in artisanal fisheries as reported by fishers in São Tomé and Príncipe has also been reported in many other localities around the world, including continental Africa (Baum and Myers, 2004; Dulvy and Polunin, 2004;Sáenz-Arroyo et al., 2005a; 2005b; Bunce et al., 2008; Pinnegar and Engelhard, 2008; Bender et al., 2014). The single artisanal fisheries statistics available for São Tomé and Príncipe encompasses landings of coastal species from 1961 to 1978 (EST, 1982). In this report, there is an apparent decline pattern for traditional fisheries over the period. Nevertheless, 17 years can be considered, in most cases, an insufficient time to reliably detect trends in fisheries or fishing stocks, especially in tropical fisheries targeting multiple species (Pauly and Murphy, 1982). Similar to fisheries declining trends, the Shifting Baselines Syndrome has been detected by other studies around the world (e.g. Sáenz-Arroyo et al., 2005a,b; Ainsworth et al., 2008; Bunce et al., 2008; Bender et al., 2014; Giglio et al. 2015), with clear signs of changes in the perception of fishers through generations. For example, Sáenz-Arroyo et al. (2005a,b) identified that older fishers in Mexico's Gulf of California recognized five times more species and four times more fishing sites that suffered severe

167

declines in catches. In the southwestern Atlantic, Bender et al. (2014) showed that older fishers not only have caught larger individuals of specific fishes compared to younger ones, but also reported that the catches of large fish occurred around four decades ago. In this same study, fishers reported a steep declining trend in the catches of five out of nine reef fish species assessed. In Africa, Bunce et al. (2008) noted marked differences in fishers’ perceptions of size and number of fish reported by three successive generations of fishers from an island. In this study, the median number of species reported by the oldest group of fishers was twice as high compared to the youngest. For São Tomé and Príncipe, the majority of experienced fishers reported a strong decline in local fisheries. On the other hand, only one-third of inexperienced fishers reported declines over time (with some reports of increasing captures). Furthermore, some inexperienced fishers reported having never caught some of the studied species, in contrast with experienced fishers that recognized all species. Other species not included in our analyses but mentioned by the interviewees indicate the occurrence of changes in fisheries composition. Based on the accounts, sharks were not commonly caught in São Tomé and Príncipe before 1970 because they were laborious to capture (requiring between six to ten hours of effort to remove them from the water) and had no commercial value. Despite the relative isolation of the country, since 1975 there has been an increased access to fishing technologies (nylon lines, nets and boat motors) that eventually replaced the traditional fishing gears (Hodges and Mewitt, 1988). Moreover, since the independence of the country, international organizations for fisheries have provided localfishers with equipment and technical support in the form of workshops (Hodges and Mewitt, 1988; Carneiro, 2011). These new technologies and access to new knowledge may have greatly contributed to the increase of fishing pressure on the islands, and to the overexploitation of some species, resulting in the changes perceived by fishers.

4.1. Factors causing changes in reef fish compositions and/or catches Because of their accumulated empirical knowledge, fishers represent a rich source of information about habitats and behaviors of fish species (Raychaudhuri, 1980; Ruddle, 1993; Silvano et al., 2006; Silvano and Begossi, 2010). In our study, artisanal fishers pointed out seven factors as the main causes of change in fisheries catches and species composition, of which the most cited were: industrial fishing, increases

168

in the number of fishers, blast fishing, and climate change. STP does not have a national industrial fishing fleet, and the fleets operating in the country’s Economic Exclusive Zone are mainly from Japan, Taiwan, China and the European Union (Belhabib, 2015). According to 32 fishers, although the amount of fish caught by industrial fleets is high compared to artisanal fisheries, large captures are not the main damage generated for artisanal fishers by industrial fisheries. In their perception, the major damage caused by industrial fishing fleets is the abandonment of ships that brings, over time, impacts to the marine environments like oil-spills. From 1955 to 2010, the number of artisanal fishers in São Tomé and Príncipe increased approximately 116% from 1,127 to 2,428 fishers (Costa, 1955; DGAP-RDSTP, 2010). Moreover, this time period also encompassed a huge technological revolution that resulted in more efficient and profitable fisheries all around the world, and STP was no exception. For example, fishers from two communities in our study (Malanza and Angolares) mentioned that nylon lines were scarce in the country before the independence (1975). According to them, before that time they employed hand-made lines from local plants in a laborious process that limited the amount of time they could spend at sea. Thus, technological improvements such as the provision of better quality nets, lines and hooks, and motorized boats, coupled with an increasing demand from a growing population and an increase in the number of fishers likely resulted in increased catches. In fact, a catch reconstruction estimated that artisanal fisheries captures increased by more than 300% from 1950 to 2002, from around 1,500t to more than 4,700t (Belhabib, 2015; FAO, 2017). More than a third of the interviewed fishers in São Tomé and Príncipe islands had witnessed blast fishing. Blast fishing not only indiscriminately kills fish, but it also inflicts heavy damage to the reef structure (Fox and Caldwell, 2006). As such, it can have detrimental effects to fishing activities, if the targeted stocks depend on the reef matrix for their life cycles (McManus et al., 1997; Fox and Erdmann, 2000). In 2001, the Santomean government established a legislation banning all fishing activities that have strong negative direct or indirect impacts on aquatic and coastal environments (e.g. fishing with explosives, grenades, toxic products and fishing with homemade bombs; RDSTP, 2001). The main sites where fishers reported having heard of blast fishing in both São Tomé and Principe are also some of the most sensitive and biologically important of the islands. For instance, in São Tomé island, the sites that present the highest coral cover are Ilhéu Santana, Sete

169

Pedras and Ilhéu das Rolas (Floeter et al., 2007). These are also the most cited places for blast fishing occurrences. This raises a strong concern about the future of coral communities in São Tomé, and suggests that these places should be important targets for conservation efforts. Similarly, Pedra da Galé and Ilhéu dos Mosteiros in Príncipe island, and Lagoa Azul in São Tomé island are areas of concentration of gorgonian and black coral forests (Wirtz & D’acoz, 2008; Morais and Maia, 2016). These places were repeatedly cited by fishers as affected by blast fishing. Fortunately, some of these sites harbor gorgonian and black coral forests in mesophotic depths still far from the reach of blast fishers (Morais and Maia, 2016). Climate change was also recognized as an important driver of change in fisheries catches and species composition by interviewed fishers. Interviewed fishers mentioned climate change in the context of a recent shortening of the dry season. According to them, the Gravana, as the dry season is designated, has begun later and finished earlier in recent times. Fishers also mentioned weaker winds and warmer ocean waters associated to this purported change and that they result in interferences, especially in fish reproductive season. A possible topic for further investigations is to evaluate if these climate change perceptions actually scale with meteorological and oceanographic data and, if so, whether reproductive cycles of target species could be affected. Local Ecological Knowledge of fishers has been applied before in the context of identifying the reproductive season of fish species (e.g. Gerhardinger et al. 2006; Silvano et al., 2006; Silvano and Begossi, 2010; Ferreira et al. 2014). This type of LEK could be allied with meteorological and fisheries research to yield future insights on the possible effects of climate change in fish reproduction and catch rates. The Gulf of Guinea, including São Tomé and Príncipe, has been considered as one of the 18 global hotspots for the conservation of marine biodiversity (Roberts et al., 2002). However, so far, São Tomé and Príncipe host no marine protected areas. The fishers themselves recognize that the problem of overexploitation of marine resources could be addressed, at least partially, through the creation of marine reserves. According to one of them: “regardless of [fishing activities] being our livelihood it is also a cultural issue”. Our study showed that most fishers perceive a declining trend in fishing catches and changes in fisheries species composition in the last decades in São Tomé and Principe. We also found evidence of the occurrence of the shifting baseline syndrome among fishers’ generations. The pattern of change identified by local fishers in STP corroborates the widespread degradation pattern noted for

170

marine environments across the world (Jackson et al., 2001; Worm et al., 2006). We emphasize the importance of using local ecological knowledge, not only to detect baseline changes in marine habitats, but also to identify causes of change to better manage marine resources.We thus recommend: 1) enforcement of the existing laws prohibiting destructive fishing practices (e.g. fishing with explosives, grenades, toxic products and bombs); 2) tracing the origin of the products used for the illegal practices and controlling its use; 3) the creation and enforcement of no-take marine protected areas around both islands; 4) a more strict control of industrial fishing (this could include on-board fisheries observers, setting catch quotas, setting enforcement systems for illegal industrial ships; and 5) the update and maintenance of statistical reports of artisanal fisheries. These actions would help to prevent further declines in STP fisheries and support the maintenance of resources for future generations of artisanal fishers in the islands.

Acknowledgements This study benefitted from permits from the Department of Natural Sciences and Biology of the Universidade de São Tomé e Príncipe and the government of São Tomé and Príncipe, through the Direcção Geral de Pescas. We acknowledge funding from The Rufford Foundation (Grant No.18424-1), California Academy of Sciences, Fundação Calouste Gulbenkian (Portugal) (P-141933) from the grants received and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES (Brazil). Logistical support was provided by partner institutions at Príncipe (Roça Belo Monte Resort) and São Tomé (Direcção de Pescas, Universidade de São Tomé and local NGOs, Mar Ambiente e Pesca Artesanal - MARAPA and Associação de biólogos Santomense - ABS). We also thank Alberto Miranda of Atlantic Diving Center, Instituto Nacional de Estatística de Portugal, Dr. Cristina Brito and the people of São Tomé and Príncipe (traditional fishing villages) for wonderful times. Finally, we thank Nivaldo Peroni, Thiago Cesar Lima Silveira and Teresa Cerveira Borges for reviewing earlier drafts.

171

References Ainsworth, C.H., Pitcher, T.J., Rotinsulu, C., 2008. Evidence of fishery depletions and shifting cognitive baselines in Eastern Indonesia. Biol. Cons. 141, 848–859. http://dx.doi.org/10.1016/j.biocon.2008.01.006. Alcala, A.C., Gomez, E.D, 1987. Dynamiting coral reefs for fish: a resource destructive fishing method. In: Salvat, B. (ed.), Human impacts on coral reefs: facts and recommendations. Antenne Museum EPHE, French Polynesia, 51–60. Baisre, J.A., 2013. Shifting baselines and the extinction of the Caribbean monk seal. Cons. Biol. 27, 927–935. http://dx.doi.org/10.1111/cobi.12107. Baum, J.K., Myers, R.A., 2004. Shifting baselines and the decline of pelagic sharks in the Gulf of Mexico. Ecol. Lett. 7,135–145. Belhabib, D., 2015. Fisheries of Sao Tome and Principe, a catch reconstruction (1950-2010). Fisheries Centre Working Paper. Working Paper #2015 - 6786. University of British Columbia, Vancouver, BC, V6T 1Z4, Canada. Bellwood, D.R., Hughes, T.P., Folke, C., et al., 2004. Confronting the coral reef crisis. Nature 429, 927–833. http://dx.doi.org/10.1038/nature02691. Bender, M.G., Floeter, S.R., Hanazaki, N., 2013. Do traditional fishers recognise reef fish species declines? Shifting environmental baselines in Eastern Brazil. Fish. Manag. Ecol. 20, 58–67. http://dx.doi.org/10.1111/fme.12006. Bender, M.G., Machado, G.R., Silva, P.J., Floeter, S.R., Monteiro-Netto, C., Luiz, O. J., Ferreira, C. E. L., 2014. Local ecological knowledge and scientific data reveal overexploitation by multigear artisanal fisheries in the southwestern Atlantic. PLoS ONE 9, e110332. http://dx.doi.org/10.1371/journal.pone.0110332. Berkes, F., 1999. Sacred ecology. Traditional ecological knowledge and resource management. Taylor and Francis, Philadelphia and London, UK. Braga H.O., Azeiteiro U.M., Oliveira H.M.F., Pardal, M.A., 2017. Evaluating fishermen’s conservation attitudes and local ecological knowledge of the European sardine (Sardina pilchardus), Peniche, Portugal. J. Ethnobiol. Ethnomed. 13, 1–12. http://dx.doi.org/10.1186/s13002-017-0154-y. Briguglio, L., 1995. Small Island Developing States and their economic vulnerabilities. World Dev. 23, 1615–1632.

172

Bunce, M., Rodwell, L.D., Gibb, R., Mee, L., 2008. Shifting baselines in fishers’ perceptions of island reef fishery degradation. Ocean. Coast. Manag. 51, 285–302. Carneiro, G., 2011. They come, they fish, and they go: EC fisheries agreements with Cape Verde and São Tomé e Príncipe. Mar. Fish. Ver. 73, 1–25. Ceita, M.N., 1991. Ensaio para uma reconstrução Histórico- Antropológica dos Angolares de S. Tomé, CEA/ISCTE, Portugal. Colding, J., 1998. Analysis of hunting options by the use of general food taboos. Ecol. Mod. 110, 5–17. Costa, F.C., 1959. Estudos de pescas em S. Tomé e Príncipe. Expedition report, Ministério Junta Investigação Ultramarino, 8, 219–271. Dayton, P.K, Thrush, S.F, Agardy, M.T, Hofman, R.J., 1995. Environmental effects of marine fishing. Aquat. Conserv. 5, 205– 232. DeMartini, E.W., Mundy, B.C, 1999. Status of nearshore and comercial fishing and impacts on biodiversity in the Tropical Insular Pacific. In Eladredge, L. Maragos, J., Holthus, P., Takeuchi, H., (eds.), Marine and Coastal Biodiversity in the Tropical Island Pacific region. Pacific Science Association. East-West Center, pp.339–355. DeMartini, E.E., Friedlander, A.M., Sandin, S.A., Sala, E., 2008. Differences in fish-assemblage structure between fished and unfished atolls in the northern Line Islands, central Pacific. Mar. Ecol. Prog. Ser. 365, 199–215. DeMartini, E.E., Smith, J.E., 2015. Effects of fishing on the fishes and habitat of coral reefs. In Mora, C. (ed.), Ecology of Fishes on Coral Reefs. Cambridge University Press, pp. 135–144. DGAP-RDSTP (Direção Geral da Agricultura e Pesca da República Democrática de São Tomé e Príncipe), 2010. Boletim Estatístico dos dados de captura estimados das principais espécies da pesca artesanal de São Tomé e Príncipe de 2001 a 2010. Direção Geral de Pesca. Technical Report. Dulvy, N.K., Polunin, N.V.C., 2004. Using informal knowledge to infer human-induced rarity of a conspicuous reef fish. Anim. Cons. 7, 365–374. 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., Försterra, G., Galván, 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

173

marine protected areas with five key features. Nature. 506, 216–20. Edinger, E.N., Jompa, J., Limmon, G.V., Widjatmoko, W., Risk, M.J., 1998. Reef degradation and coral biodiversity in Indonesia: Effects of land-bassed pollution, destructive fishing practices and changes over Time. Mar. Pollut. Bull. 36, 617–630. EST (Estatística de São Tomé), 1980. Estatísticas das Pescas. Ministério do Plano, Direção de Estatística, Lisboa. Technical Report. FAO (Food and Agricultural Organization of the United Nations), 2017. The Statistical Programme of Work. http://www.fao.org/faostat/en/#data/AO (Accessed 04/07/2017). Figueiredo, R., 1966. Análise do sector de pesca no Ultramar. Subsídios para o seu estudo. In: Boletin Gereral Ultramarino, agosto/setembro. Lisboa, pp.144–159. Floeter, S.R., Rocha, L.A., Robertson, D.R., Joyeux, J.C., Smith-Vaniz, W.F., Wirtz, P., Edwards, A.J., Barreiros, J.P., Ferreira, C.E.L., Gasparini, J.L., Brito, A., Falcón, J.M., Bowen, B.W., Bernardi, G., 2008. Atlantic reef fish biogeography and evolution. J. Biogeogr. 35, 22–47. Fox, H., Caldwell, R., 2006. Recovery from blast fishing on coral reefs: A tale of two scales. Ecol. Appl. 16, 1631–1635. Fox, H.E., Erdmann, M.V., 2000. Fish yields from blast fishing in Indonesia. Coral Reefs. 19, 114–114. http://dx.doi.org/10.1007/s003380000094. Fox, H.E., Pet, J.S., Dahuri, R., Caldwell, R.L., 2003. Recovery in rubble fields: Long-term impacts of blast fishing. Mar. Poll. Bull. 46, 1024–1031. Ferreira H.M., Reuss-strenzel, G.M., Alves, J.A., Schiavetti, A., 2014. Local ecological knowledge of the artisanal fishers on Epinephelus itajara (Lichtenstein, 1822) (Teleostei: Epinephelidae) on Ilheus Coast, Bahia, Brazil. J. Ethnobiol. Ethnomed. 10, 1–15. Friedlander, A.M., DeMartini, E.E., 2002. Contrasts in density, size, and biomass of reef fishes between the northwestern and the main Hawaiian islands: The effects of fishing down apex predators. Mar. Ecol. Prog. Ser. 230, 253–264. Frynas, J.G., Wood, G., Oliveira, R.M.S.S., 2003. Bussiness and politics in São Tomé e Príncipe: from cocoa monoculture to petro-state. Afr. Aff. 102, 51–80. https://doi.org/10.1093/afraf/adg002. Gadgil, M., Berkes, F., Folke, C., 1993. Indigenous knowledge for biodiversity conservation. J. Hum. Ecol. 22,151–156. Gerhardinger, L.C., Marenzi, R.C., Bertoncini, A.A., Medeiros, R.P.,

174

Hostim-Silva, M., 2006. Local Ecological Knowledge on the Goliath Grouper Epinephelus itajara (Teleostei: Serranidae) in Southern Brazil. Neotrop. Ichthyol. 24,441–450. Giglio, V.J., Luiz, O.J., Gerhardinger, L.C., 2015. Depletion of marine megafauna and shifting baselines among artisanal fishers in eastern Brazil. Anim. Cons. 18, 348–358. Graham, N.A.J., McClanahan, T.R., 2013. The Last Call for Marine Wilderness? Bioscience. 63, 397–402. Halpern, B.S., Walbridge, S., Selkoe, K.A., 2008. A Global map of human impact on marine ecosystems. Science. 319, 948–953. http://dx.doi.org/10.1126/science.1149345. Hanazaki, N., Herbst, D.F., Marques, M.S., Vandebroek, I., 2013. Evidence of the shifting baseline syndrome in ethnobotanical research. J. Ethnobiol. Ethnomed. 9, 75. http://dx.doi.org/10.1186/1746-4269-9-75. Hodges, T., Mewitt, M., 1988. São Tomé and Príncipe from plantation colony to microstate. United States of America. Horemans, B., Gallène, J., Njock, J., 1994. Revista setorial da pesca artesanal a São Tomé e Príncipe. Programa de desenvolvimento integrado das pescas artesanais na África Ocidental (DIPA). Hothorn, T., Bretz, F., Westfall, P., 2008. Simultaneous inference in general parametric models. Biom. J. 50, 346–363. Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bruce, J., Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange1, C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J., Warner, R.R., 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science, 293, 629–637. http://dx.doi.org/10.1126/science.1059199. Jackson, J.B.C., 1997. Reefs since Columbus. Coral Reefs, 16, 23–32. Jennings, S., Blanchard, J.L., 2004. Fish abundance with no fishing: predictions based on macroecological theory. J. Anim. Ecol. 73, 632–642. Jennings, S., Kaiser, M.J., 1998. The effects of fishing on marine ecosystems. Adv. Mar. Biol. 34, 201–212. Johannes, R. E. 1998. The case for data-less marine resource management: examples from tropical nearshore fisheries. Trends. Ecol. Evol. 13, 243–246. Kittinger, J.N., Pandolfi, J.M., Blodgett, J.H., Hunt, T.L., Jiang, H., McClenachan, L.E., Schultz, J.K., Wilcox, B.A., 2011. Historical

175

Reconstruction Reveals Recovery in Hawaiian Coral Reefs. PLoS ONE 6(10): e25460. doi:10.1371/journal.pone.0025460. Krakstad, J., Alvheim, O., Lopes, J.D.S., 2010. Surveys of the fish resources of São Tomé & Príncipe survey of the demersal resources. Institute of Marine Research, Cruise Report. http://hdl.handle.net/11250/107339. Lima, M.S.P., Oliveira, J.E.L., Nóbrega, M.F., Lopes P. F. M., 2017. The use of local ecological knowledge as a complementary approach to understand the temporal and spatial patterns of fishery resources distribution. J. Ethnobiol. Ethnomed. 13, 30–42. Maire, E., Cinner, J., Velez, L., Huchery, C., Mora, C., Dagata, S., Vigliola, L., Wantiez, L., Kulbicki, M., Mouillot, D., 2016. How accessible are coral reefs to people? A global assessment based on travel time. Ecol. Lett. 19, 351–360. McCauley, D. J., Pinsky, M. L., Palumbi, S. R., Estes, J. A., Joyce, F. H., Warner, R. R., 2015. Marine defaunation: Animal loss in the global ocean. Science. http://dx.doi.org/10.1126/science.1255641. 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. Bull. Ecol. Soc. Am. 17, 1055–1067. McClenachan, L., 2009. Documenting loss of large trophy fish from the florida keys with historical photographs. Conserv. Biol. 23,636– 643. McClanahan, T.R., 2014. Resilience in reef fish communities. In Mora C (ed.) Ecology of fishes on coral reefs. Cambridge University Press. McManus, J.W., Reyes, R.B., Nañola, C.L., 1997. Effects of some destructive fishing methods on coral cover and potential rates of recovery. Environ. Manage. 21, 69–78. Morais, R.A., Maia, H.A., 2016. Lush underwater forests in mesophotic reefs of the Gulf of Guinea. Coral Reefs. 36, 95–95. http://dx.doi.org/10.1007/s00338-016-1523. Mora, C., Aburto-Oropeza, O., Ayala Bocos, A., Ayotte, P.M., Banks, S., Bauman, A.G., Beger, M., Bessudo, S., Booth, D.J., Brokovich, E., Brooks, A., Chabanet, P., Cinner, J.E., Cortés, J., Cruz-Motta, J.J., Cupul Magaña, A., DeMartini, E.E., Edgar, G.J., Feary, D.A., Ferse, S.C.A., Friedlander, A.M., Gaston, K.J., Gough, C., Graham, N.A.J., Green, A., Guzman, H., Hardt, M., Kulbicki, M., Letourneur, Y., López Pérez, A., Loreau, M., Loya, Y., Martinez, C., Mascareñas-Osorio, I., Morove, T., Nadon, M.O., Nakamura, Y., Paredes, G., Polunin, N.V.C., Pratchett, M.S., Reyes Bonilla, H.,

176

Rivera, F., Sala, E., Sandin, S.A., Soler, G., Stuart-Smith, R., Tessier, E., Tittensor, D.P., Tupper, M., Usseglio, P., Vigliola, L., Wantiez, L., Williams, I., Wilson, S. K., Zapata, F.A., 2011. Global human footprint on the linkage between biodiversity and ecosystem functioning in reef fishes. PLoS Biol., 9,e1000606. Monnereau, I., Failler, P., 2014. Unlocking the full potential of the blue economy: Are African Small Island Developing States ready to embrace the opportunities? United Nations Economic Commission for Africa. Technical Report. Myers, R., Worm, B., 2003. Rapid worlwide depletion of predatory fish communities. Nature. 243, 280–283. Pauly, D., Murphy, G.I., 1982. Theory and management of tropical fisheries. ICLARM/CSIRO, Manilla, Philippines. Pauly, D., 1995. Anecdotes and the shifting baseline syndrome of fisheries. Trend. Ecol. Evol.10, 430. http://dx.doi.org/10.1016/S0169-5347(00)89171-5. Pandolfi, J.M., Bradbury, R.H., Sala, E., Hughes, T.P., Bjorndal, K.A., Cooke, R.G., McArdle, D., McClenachan, L., Newman. M.J., Paredes, G., Warner, R.R., Jackson, J.B., 2003. Global trajectories of the long-term decline of coral reef ecosystems. Science. 301, 955–958. Pet-Soede, L., Erdman, M.V., 1998. Blast Fishing in southwest Sulawesi, Indonesia. Naga, 21, 4–9. Pinnegar, J.K., Engelhard, G.H., 2008. The “shifting baseline” phenomenon: A global perspective. Rev. Fish. Biol. Fish. 18, 1–16. Plumeridge, A.A., Roberts, C.M., 2017. Conservation targets in marine protected area management suffer from shifting baseline syndrome: A case study on the Dogger Bank. Mar. Poll. Bull. 116, 395–404. http://dx.doi.org/10.1016/j.marpolbul.2017.01.012. Raychaudhuri, B., 1980. The moon and the net: Study of a transient community of fishers at Jambudwip, Anthropological Survey of India, Calcutta. RDSTP (República Democrática de São Tomé e Príncipe), 2001. Lei de pesca e recurso hauliêutico, São Tomé e Príncipe. LEI N.°09/2001. Available at http://extwprlegs1.fao.org/docs/pdf/sao72016.pdf. (Accessed 28/11/2016). RDSTP (República Democrática de São Tomé e Príncipe), 2009. Relatório nacional de avaliação da estratégia de implementação das Maurícias, São Tomé e Príncipe. Technical Report. Available at https://sustainabledevelopment.un.org/content/documents/1300

177

Sao_Tome_and_Principe-MSI-NAR2009.pdf. (Accessed 28/06/2017). RGPH, 2012. Atlas Demográfico do Recenciamento Geral da População Humana. Instituto Nacional de Estatística. São Tomé e Príncipe. Roberts, C.M., 2007. The Unnatural History of the Sea. Island Press, Washington, DC. Roberts, C.M., McClean, C.J., Veron, J.E., Hawkins, J.P., Allen, G.R., McAllister, D.E., Mittermeier, C.G., Schueler, F.W., Spalding, M., Wells, F., Vynne, C., Werner, T.B., 2002. Marine biodiversity hotspots and conservation priorities for tropical reefs. Science. 295, 1280–1284. http://dx.doi.org/10.1126/science.1067728. Ruddle, K., 1993. The Transmission of traditional ecological knowledge. Traditional Ecological Knowledge: Concepts and Cases, 17–31. Ruddle, K., 1994a. Local knowledge in the future management of inshore tropical marine resources and environments, Nature and Resources, 30, 28–37. Ruddle, K., 1994b. A Guide to the Literature Traditional Community- Based Fisheries Management Systems in the Tropics of the Asia- Pacific Region, Fisheries Technical paper, FAO, Rome. Ruddle, K., 1994c. Local knowledge in the folk management of fisheries and coastal marine environments, n C.L. Dyer and J.E. McGoodwin (eds), Folk Management in the World Fisheries, Nowot, Colorado, University Press of Colorado, pp. 161-206. Ruddle, K., 1996. Geography and human ecology of reef fisheries. In “Reef Fisheries” (N.V.C. Polunin and C.M. Roberts, eds), pp. 137- 160. Chapman and Hall, London. Sáenz-Arroyo, A., Roberts, C.M., Torres, J., Cariño-Olvera, M., Enríquez-Andrade, R.R., 2005a. Rapidly shifting environmental baselines among fishers of the Gulf of California. Proc. Biol. Sci. 272, 1957–1962. http://dx.doi.org/10.1098/rspb.2005.3175. Sáenz-Arroyo, A., Roberts, C.M., Torres, J., Cariño-Olvera, M., 2005b. Using fisher’s anecdotes, naturalist’s observations and grey literature to reassess marine species at risk: the case of the Gulf grouper in the Gulf of California, Mexico. Fish. Fish. 6,121–133. Sala, E., Sugihara, G., 2005. Food web theory provides guidelines for marine conservation. In A B, U S, J D, et al. (ed.) Aquatic food webs: An ecosystem approach. Oxford University Press. Sandin, S.A., Smith, J.E., DeMartini, E.E., Dinsdale, E.A., Donner, S.D., Friedlander, A.M., Konotchick, T., Malay, M., Maragos, J.E.,

178

Obura, D., Pantos, O., Paulay, G., Richie, M., Rohwer, F., Schroeder, R.E., Walsh, S., Jackson, J.B.C., Knowlton, N., Sala, E., 2008. Baselines and degradation of coral reefs in the Northern Line Islands. PLoS ONE, 3 (2):e1548 Sandin, S.A., Zgliczynski, B.J., 2014. Inverted trophic pyramids. In Mora C (ed) Ecology of fishes on coral reefs. Cambridge University Press. Serigstad, B., Ensrud, T.M., Olsen, M., 2012. Marine environmental survey of bottom fauna, selected physical and chemical compounds and fisheries survey in the Joint Development Zone between Nigeria and São Tomé & Príncipe. Cruise Report, Available at https://brage.bibsys.no/xmlui/bitstream/handle/11250/107442/2 011405%20JDZ.pdf?sequence=1&isAllowed=y, (Accessed 28/11/2016). Silvano, R., MacCord, P.F.L., Lima, R.V., Begossi, A., 2006. When does this fish spawn? Fishers’s local knowledge of migration and reproduction of Brazilian coastal fishes. Env. Biol. Fish., 76, 371– 386. http://dx.doi.org/10.1007/s10641-006-9043-2. Silvano, R.A.M., Begossi, A., 2010. What can be learned from fishers? An integrated survey of fishers’ local ecological knowledge and bluefish (Pomatomus saltatrix) biology on the Brazilian coast. Hydrobiologia. 637, 3–18. http://dx.doi.org/10.1007/s10750-009- 9979-2. Shackeroff, J.M., Campbell, L.M., 2007. Traditional ecological knowledge in conservation research: problems and prospects for their constructive engagement. Conservat. Soc. 5, 343–360. Ternes, M.L.F., Gerhardinger, L.C., Schiavetti, A., 2016. Seahorses in focus: local ecological knowledge of seahorse-watching operators in a tropical estuary. J. Ethnobiol. Ethnomed.12, 1–12. Turvey, S.T., Barrett, L.A., Yujiang, H., Lei, Z., Xinqiao, Z., Xianyan, W., Yadong, H., Kaiya, Z., Hart, T., Ding, W., 2010. Rapidly shifting baselines in yangtze fishing communities and local memory of extinct species. Cons. Biol. 24, 778–787. http://dx.doi.org/10.1111/j.1523-1739.2009.01395.x. Turvey, S.T., Bryant, J.V., Duncan, C., Wong, M.H., Guan, Z., Fei, H., Ma, C., Hong, X., Nash, H.C., Chan, B.P., Xu, Y., Fan, P., 2017. How many remnant gibbon populations are left on Hainan? Testing the use of local ecological knowledge to detect cryptic threatened primates. Am. J. Primatol. 79, 1–13. Trebilco, R., Baum, J.K., Salomon, A.K., Dulvy, N.K., 2013. Ecosystems

179

ecology: Size based constraints on the pyramids of life. Trends Ecol. Evol. 28: 423-431 Wells, S., 2009. Dynamite fishing in northern Tanzania – pervasive, problematic and yet preventable. Mar. Poll. Bull. 58, 20–23. Wirtz, P., Ferreira, C.E.L., Floeter, S.R., Fricke, R., Gasparini, J.L. Iwamoto, T., Rocha, L.A., Sampaio, C.L., Schliewen, U.K., 2007. Coastal Fishes of São Tomé and Príncipe islands, Gulf of Guinea (Eastern Atlantic Ocean) – an update. Zootaxa. 1523, 1–48. Wirtz, O., D’acoz, C. D., 2008. Crustaceans associated with Cnidaria, Bivalvia, Echinoidea and pisces at São Tomé and Príncipe islands. Arquipélago: Life and Marine Sciences, 25, 63–69. Williams, I.D., Richards, B.L., Sandin, S.A., Baum, J.K., Schroeder, R.E., Nadon, M.O., Zagliczynski, B., Craig, P., McIlwain, J.L., Brainard, R.E., 2011. Differences in reef fish assemblages between populated and remote reefs spanning multiple archipelagos across the central and western Pacific. J Mar Biol. 2011, 1–14. Worm, B., Barbier, E.B., Beaumont, N., Duffy, J.E., Folke, C., Halpern, B.S., Jackson, J.B., Lotze, H.K., Micheli, F., Palumbi, S.R., Sala, E., Selkoe, K.A., Stachowicz, J.J., Watson. R., 2006. Impacts of biodiversity loss on ocean ecosystem services. Science. 314, 787– 90. http://dx.doi.org/10.1126/science.1132294.

Appendix

Figure S.1: Interaction network representing the number of fishermen across traditional villages and blast fishing sites in São Tomé (A) and Príncipe (B) island. Circle size is proportional to the number of fishers that mentioned blast fishing (left) and the number of citations for each site (right hand side). Numbers in parentheses represent interviewed fishers by communities and the number within black circles represent quantity of fishers what mentioned blast fishing inside the community.

180

181

Formulary

1. Date/Island

2. Age

3. Fishing time

4. Fishing art

5. Catch:

a) Do you know any of these fish? (show photos)

( ) not ( ) yes: what is the name? (fill in spreadsheet)

b) What's the biggest fish you've ever caught?

c) When?

d) Where was he caught?

e) How was he caught?

f) What was the highest amount (kg / individuals) of fish of

this species that you captured in a single day of fishing?

g) What year did this happen?

N° Yes/Not Name Biggest When? Where? How? Best Obs. catch day’ s catch

182

6. Shifs in target species:

a) Do you know of any species that were formerly

discarded during fishing or used only as bait, but which

have now become the target of fishers?

( ) Yes ( ) Not

b) Do you know any spots in this region that was

considered a good fishing ground because of the

abundance and productivity of fish, but which is

currently overexploited? ( ) Yes ( ) Not

7. Fishing impact:

a) Which of these fish do you most fish today?

b) And there are 40/30/20/10 years?

c) Which of these species

8. How do you compare the fishing situation today compared to

when you started fishing?

183

Table S.1: Total artisanal fishers and number of interviews by traditional villages on São Tomé and Príncipe island.

Total artisanal Number of Traditional villages (São Tomé island) fishers interviews

Praia São João (São Pedro) 42 3

Praia Cruz 242

Praia Gamboa 108

Praia Lochinha 183

Praia de Pantufo 135

Praia Cova Água 59 25

Praia Messias Alves 78

Praia de Plano de Água Izé 22

Praia Almocherife 28

Praia de Ribeira Afonso 49

Praia Quinze 43

Praia de Micoló 87

Praia de Fernão Dias 23

Praia de Mouro Peixe 176

Praia Água Tomá 133

Praia de Bairro Benga 226

Praia de Bairro Rosema 121

Praia Brita 87 7

184

Praia de Santa Catarina 216 35

Praia Melão 190 7

Praia de Angra Toldo 24

Praia de Angolares 137 12

Pria Yô Grande 32 2

Praia Pesqueira 22 8

Praia de Ribeira Peixe 24 7

Praia de Monte Mário 34

Praia de Malanza 50 14

Praia de Porto Alegre 98 16

Total artisanal Number of Traditional villages (Príncipe island) fishers interviews

Praia Seabra 12

Praia de São Pedro 32

Praia de Capitania 18

Praia Abelha 14

Praia Lama 5

Praia de Sundy 68

Praia Abade 37 12

Praia das Burra 50 25

Praia Campanha 18

Praia Lapa 19 3

Praia Seca 16 1

185

Praia Yola 11

Praia de Esprainha (Santo António) 6 1

Praia de São João 3

Praia Caixão 13

186

187

CONCLUSÕES GERAIS E CONSIDERAÇÕES FINAIS As ilhas oceânicas são consideradas laboratórios naturais porque possuem uma grande variedade de fauna e flora ímpares, com muitas espécies ainda totalmente desconhecidas pela ciência, como é o caso das ilhas de Golfo de Guiné (São Tomé e Príncipe). Do ponto de vista ecológico, pouco se sabe sobre padrões de estruturação da comunidade dos organismos marinhos em ilhas oceânicas, particularmente as ilhas localizadas no Atlântico Equatorial, como São Tomé e Príncipe. Diante desse cenário, e com os resultados apresentados ao longo dos três capítulos apresentados nesta tese, concluiu-se que: (1) Tanto as assembleias de peixes recifais quanto as comunidades bentônicas apresentam variações espaciais ao longo das ilhas. Por exemplo, a riqueza, abundância e biomassa de peixes diferiram significativamente entre os locais tanto ao longo da ilha de São Tomé quanto na ilha de Príncipe. Em relação aos organismos bentônicos, conclui-se que foram dominados por turf e algas calcárias; enquanto que os corais e as gorgônias foram menos representativos. (2) A exposição às ondas (hidrodinamismo), a profundidade e a pressão de pesca determinaram o padrão de distribuição de organismos marinhos de São Tomé e Príncipe. Por exemplo, a densidade, a abundância e a biomassa de peixes tiveram valores mais altos em locais mais expostos as ondas e também em áreas mais profundas. (3) Não houve mudanças significativas nas assembleias de peixes recifais na ilha de São Tomé no intervalo temporal de dez anos, o que sugere que as assembleias conseguem se manter mesmo em ambientes com fortes distúrbios antropogênicos, tais como pressão de pesca. Finalmente, detectou-se ocorrência de “shifting baseline syndrome” entre pescadores tradicionais nas ilhas de São Tomé e Príncipe. Ou seja, observou-se uma tendência decrescente no tamanho corporal individual para espécies de peixes ao longo do tempo com base em relatos de quatro gerações de pescadores, desde os mais velhos e até os mais jovens. Além disso, percebeu-se que as gerações também diferiram em suas percepções de declínios tanto no tamanho maximo quanto na abundância de peixes ao longo do tempo (cerca de 40 anos), com a maioria dos pescadores inexperientes que relataram não haver queda nas pescarias. Assim sendo, podemos sugerir que os ecossistemas marinhos das ilhas de São Tomé e Príncipe passaram por fases diferentes de acordo com o status do seu ambiente (Figura 1). Fase 0) período pré-colonização, onde os ambientes estavam mais próximos ao estado prístino, já que

188

ambas as ilhas eram desabitadas; Fase I) da colonização até próximo a 1950. Nesta fase, pensa-se que os ambientes marinhos sofreram algum declínio em termos de biomassa de peixes, porém como compreende um período de 450 anos, esse declínio não foi bem documentado, portanto não sabemos a sua magnitute; Fase II) Fase iniciada em 1970, caraterizada por um declínio mais acentuado nas pescarias que coicide com o início da utilização de redes de nylon, pesca destrutiva e pesca de arpão; Fase III) uma fase relativamente estável de 2006 até os dias atuais, onde detectamos poucas diferenças nas assembleias de peixes; Fase IV) uma previsão de como o ambiente marinho de São Tomé e Príncipe poderá ser no futuro. Acredita-se que se as medidas de conservação forem levadas em consideração, tais como, a criação de áreas marinhas protegidas da pesca, regulamentação da pesca e fiscalização como tem sido reportado em outros locais3, é provável que possamos retornar ao início da fase II ou final da fase I.

3 D’agata, S., Mouillot, D., Wantiez, L., Friedlander, A.M., Kulbicki, M. & Vigliola, L. (2016). Marine reserves lag behind wilderness in the conservation of key functional roles. Nat. Commun. 7:12000 doi: 10.1038/ncomms12000

189

Figura 1: Status de qualidade do ecossistema marinho [peixes recifais] de São Tomé e Príncipe entre 1950 e 2016 e as perspectivas futuras. O índice (*) foi baseado nos dados das entrevistas realizadas com os pescadores artesanais dessas ilhas, combinados com os dados de censos visuais subaquáticos realizados entre diferentes períodos [2006 vs. 2016]. População humana de São Tomé e Príncipe ( ). Pescadores artesanais de São Tomé e Príncipe ( ). Período de maior estresse para os organismos marinhos de São Tomé e Príncipe [ínico de pesca com rede de nylon, pesca com bombas e pesca com arpão] o que caracterizou a fase II ( ).

190

Com os resultados alcançados nesta tese, acredita-se que será possível construir uma linha de referência para compreensão da biodiversidade marinha de São Tomé e Príncipe no contexto ecológico. Essa linha de referência irá permitir futuras comparações em escala geográfica mais ampla, bem como o monitoramento de alterações futuras na biota dessas ilhas. Além disso, espera-se que com os resultados alcançados seja possível contribuir e ajudar na seleção de áreas com atributos especiais que devem receber esforços de conservação em São Tomé e Príncipe. Um melhor conhecimento da diversidade, abundância, biomassa e da singularidade da fauna local é essencial para qualquer governo entender a necessidade urgente de esforços de conservação. No caso específico de São Tomé e Príncipe acreditava-se que a ilha de Príncipe, por ter uma população residente cerca de 25 vezes menor que a ilha de São Tomé, poderia estar em melhor estado de conservação (do ambiente marinho), ou seja, mais próximo ao seu estado natural. No entanto, observou-se que os padrões das comunidades de peixes e de organismos bentônicos foram similares entre elas, ambas dominadas por peixes de tamanho pequenos. A biomassa entre elas foi relativamente similar. No entanto, ao longo de cada ilha existem certas especificidades. Por exemplo, Sete Pedras e ilhéu Santana em São Tomé apresentaram uma cobertura alta de Tubastrea spp. supostamente um invasor no Atlântico. Na ilha de Príncipe os locais tais como, Ilhéu Bombom, Praia da Lapa e ilheu dos Mosterios poderiam ser locais com alto potencial para criação de areas marinhas protegidas devido a diversidade de corais e de espécies de peixes. Em São Tomé, ilhéu das Rolas e Sete Pedras apresentaram caraterísticas semelhantes. Todos funcionam como berçario, ou seja, locais de recrutamento para diversas espéices. No futuro próximo, precisamos conhecer melhor as comunidades recifais de profundidades mesofóticas tanto em São Tomé como na ilha de Príncipe (entre os 30 e os 60 m), já que potencialmente são tão ou mais exuberantes do que as comunidades de águas rasas o que sugere ser zona de refugio para diversas espécies (ver Morais & Maia, 2017 – Apêndice).

191

APÊNDICE

Figura A.1: Artigo publicado no periódico Coral Reefs, em colaboração com Hugulay A. Maia em 2017.

192

Figura A.2: Web mídia produzida pelo Laboratório de Biogeografia e Macroecologia Marinha da UFSC, como mecanismo de divulgação científica para a comunidade internacional. Disponível em: https://www.youtube.com/watch?v=2OkfKNRyZZE&feature=youtu.be.

Laboratório de Biogeografia e Macroecologia Marinha – UFSC