Ocorrência dos pinguins-de-Magalhães Spheniscus magellanicus (Forster, 1781) no litoral entre os estados de Santa Catarina e Rio de Janeiro sob padrões de produtividade primária na plataforma continental sul e sudeste do Brasil
Carregando...
Data
2022-01-14
Autores
Santos, Natália Beatriz de Mendonça
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Endêmico da Patagônia, o pinguim-de-Magalhães (Spheniscus magellanicus) migra para o norte durante o inverno austral, alcançando o litoral brasileiro, até o Rio de Janeiro (21ºS). Na mesma época, a anchoíta (Engraulis anchoita), pequeno peixe pelágico endêmico da mesma região, até Cabo de São Tomé (22ºS), também migra setentrionalmente, a fim de se reproduzir. Fortes evidências indicam que S. magellanicus segue anualmente os cardumes de E. anchoita, já que esse pescado é um dos principais componentes de sua dieta. O padrão de massas d’água, acaba por interferir nos padrões de produtividade primária (PP) ao longo de toda a plataforma continental, apresentando índices distintos, a depender da localidade e época do ano. Esse acontecimento certamente interfere na distribuição dos cardumes de anchoítas e, consequentemente, na distribuição dos pinguins. Ao atingir a costa do Brasil, severamente debilitados devido ao imenso esforço físico despendido no ato migratório, esses animais acabam encalhando em praias. O Projeto de Monitoramento de Praias da Bacia de Santos, mantido pela PETROBRAS, atua no monitoramento dessas ocorrências. O presente estudo compilou dados sobre as ocorrências dos pinguins-de-Magalhães no litoral entre os Estados de Santa Catarina e do Rio de Janeiro, de 2015 a 2020 e correlacionou tais incidências com os índices de PP registradas ao longo da plataforma continental do Oceano Atlântico Sudoeste, para o mesmo período. O processamento dos dados foi realizado no software Python, por meio do modelo de aprendizagem Gradient Boosting Regressor. Um total de 25.813 ocorrências foram registradas, a maioria oriunda de Santa Catarina (n = 15.751). O Paraná registrou a maior taxa de ocorrência geral (40,18 pinguins/km) e por ano (7,44 pinguins/km/ano). O ano de maior incidência foi 2018 (n = 12.044). A maioria dos animais ocorreu durante o inverno (n = 22.572), a maior parte em setembro (n = 8.112). De janeiro a maio somente 0,3% do total foi contabilizado (n = 74). A maioria eram de juvenis (n = 21.708) e 91,5% já estavam mortos no momento do resgate (n = 23.626). A feature mais significativa para as ocorrências em Santa Catarina foi o “setor 1” (lag = 4 meses); para o Paraná foi o “setor 10” (lag = 0 meses); para São Paulo foi o “setor 8” (lag = 3 meses) e para o Rio de Janeiro foi o “setor 5” (lag = 3 meses). Em todos os setores mais significativos para cada estado, os índices de PP são mais elevados antes do inverno, exceto para o Paraná. O pico de PP coincide com a época de cuidados parentais e muda da plumagem dos pinguins. Possivelmente, a predação é extremamente reduzida e nesta época os estoques de anchoíta se elevam, voltando a serem predados durante o inverno, conferindo uma dinâmica presa-predador de Lotka-Volterra. Foi realizada ainda uma previsão de ocorrências para 2021, cujo resultado foi coerente com o padrão já esperado, sendo o modelo útil para a conservação da espécie. A inclusão de variáveis ambientais no modelo (temperatura e salinidade) e extensão do conjunto de dados futuros deverão melhorar o treinamento do modelo, bem como melhor interpretação de cada feature nas ocorrências de cada estado.
Endemic to the Patagonia, the Magellanic penguin (Spheniscus magellanicus) migrates northward during the austral winter, reaching the Brazilian coast up to the Rio de Janeiro (21ºS). At the same time, the anchoita (Engraulis anchoita), a small pelagic fish endemic to the same region up to the Cabo de São Tomé (22ºS), also migrates northward in order to reproduce. Strong evidence suggests that S. magellanicus annually follows the schools of E. anchoita, as this fish is one of the main components of its diet. The pattern of water masses ends up interfering with the primary productivity (PP) patterns along the entire continental shelf, presenting different rates depending on the location and time of year. This event certainly interferes in the distribution of schools of anchoites and, consequently, in the distribution of penguins. Upon reaching the coast of Brazil, severely weakened due to the immense physical effort expended in the migratory act, these animals end up running aground on the beaches. The Santos Basin Beach Monitoring Project, maintained by PETROBRAS, monitors these events. The present study compiled data on occurrences of Magellanic penguins on the coast between the States of Santa Catarina and Rio de Janeiro, from 2015 to 2020 and correlated such incidences with PP rates recorded along the continental shelf of the Atlantic Ocean Southwest, for the same period. Data processing was performed in Python software, using the Gradient Boosting Regressor learning model. A total of 25,813 occurrences was registered, most from Santa Catarina (n = 15,751). Paraná recorded the highest overall occurrence rate (40.18 penguins-km) and per year (7.44 penguins/km/year). The year with the highest incidence was 2018 (n = 12,044). Most animals occurred during the local winter (n = 22,572), most in September (n = 8,112). From January to May only 0.3% of the total was accounted for (n = 74). Most were juveniles (n = 21,708) and 91.5% were already dead at the time of rescue (n = 23,626). The most significant feature for the occurrences in Santa Catarina was “sector 1” (lag = 4 months); for Paraná it was “sector 10” (lag = 0 months); for São Paulo it was “sector 8” (lag = 3 months) and for Rio de Janeiro it was “sector 5” (lag = 3 months). In each one of the states’ most significant sectors, the PP indexes are higher before winter, except for Paraná. The PP peak coincides with the time of parenting and changes in penguin plumage. Possibly, predation is extremely reduced and, at this time, the anchoite stocks increase, returning to be predated during the winter, giving a Lotka-Volterra prey-predator dynamic. A prediction of occurrences for 2021 was also carried out, the result of which was consistent with the expected pattern, being the useful model for the conservation of the species. The inclusion of environmental variables (temperature and salinity) and the extension of the future database must improve the model’s training and also the interpretation of each feature in the occurrences of each state.
Endemic to the Patagonia, the Magellanic penguin (Spheniscus magellanicus) migrates northward during the austral winter, reaching the Brazilian coast up to the Rio de Janeiro (21ºS). At the same time, the anchoita (Engraulis anchoita), a small pelagic fish endemic to the same region up to the Cabo de São Tomé (22ºS), also migrates northward in order to reproduce. Strong evidence suggests that S. magellanicus annually follows the schools of E. anchoita, as this fish is one of the main components of its diet. The pattern of water masses ends up interfering with the primary productivity (PP) patterns along the entire continental shelf, presenting different rates depending on the location and time of year. This event certainly interferes in the distribution of schools of anchoites and, consequently, in the distribution of penguins. Upon reaching the coast of Brazil, severely weakened due to the immense physical effort expended in the migratory act, these animals end up running aground on the beaches. The Santos Basin Beach Monitoring Project, maintained by PETROBRAS, monitors these events. The present study compiled data on occurrences of Magellanic penguins on the coast between the States of Santa Catarina and Rio de Janeiro, from 2015 to 2020 and correlated such incidences with PP rates recorded along the continental shelf of the Atlantic Ocean Southwest, for the same period. Data processing was performed in Python software, using the Gradient Boosting Regressor learning model. A total of 25,813 occurrences was registered, most from Santa Catarina (n = 15,751). Paraná recorded the highest overall occurrence rate (40.18 penguins-km) and per year (7.44 penguins/km/year). The year with the highest incidence was 2018 (n = 12,044). Most animals occurred during the local winter (n = 22,572), most in September (n = 8,112). From January to May only 0.3% of the total was accounted for (n = 74). Most were juveniles (n = 21,708) and 91.5% were already dead at the time of rescue (n = 23,626). The most significant feature for the occurrences in Santa Catarina was “sector 1” (lag = 4 months); for Paraná it was “sector 10” (lag = 0 months); for São Paulo it was “sector 8” (lag = 3 months) and for Rio de Janeiro it was “sector 5” (lag = 3 months). In each one of the states’ most significant sectors, the PP indexes are higher before winter, except for Paraná. The PP peak coincides with the time of parenting and changes in penguin plumage. Possibly, predation is extremely reduced and, at this time, the anchoite stocks increase, returning to be predated during the winter, giving a Lotka-Volterra prey-predator dynamic. A prediction of occurrences for 2021 was also carried out, the result of which was consistent with the expected pattern, being the useful model for the conservation of the species. The inclusion of environmental variables (temperature and salinity) and the extension of the future database must improve the model’s training and also the interpretation of each feature in the occurrences of each state.
Descrição
Palavras-chave
Pinguins, Migração de aves, Produtividade primária, Plataforma continental, Python (Linguagem de programação de computador), Penguins, Bird migration, Primary productivity, Continental shelf, Python (Computer program language)