Hierarchical partitioning of fish diversity and scale-dependent environmental effects in tropical coastal ecosystems.
Mar Environ Res
; 148: 26-38, 2019 Jun.
Article
en En
| MEDLINE
| ID: mdl-31077965
The spatial structure of the fish diversity and site-scale and landscape-scale environmental effects were investigated across hierarchical levels in tropical coastal ecosystems. Total diversity (γ) was hierarchically partitioned into α and ß components using both the additive and multiplicative methods. A model selection based on the AICc was applied to generalized linear mixed models relating diversity measures to environmental variables and including random effects for hierarchical levels and season. Short-term seasonal effects were negligible. Spatial effects were more relevant at the site level and negligible at the subregion level, due to the high spatial heterogeneity and the natural pooling of ecosystems, respectively. Site-scale environmental effects were more relevant at the subregion level, with eutrophic conditions (continental influence) favoring the species richness (α and γ) and higher absence of species (ßA) in oligotrophic conditions (marine influence). At the system level, the positive effect of the distance from the ocean on γ and higher ßA in oligotrophic conditions reinforced the positive continental influence on fish diversity. Environmental homogenization processes were most likely associated with the negative effect of the pasture cover on α at the system level, and γ and ßA at the site level. The negative effect of the forest cover on the later diversity measure evidenced its relevance to maintain richer but more similar assemblages, whereas the positive continental influence was most likely due to the loss of stenohaline marine species. This study evidenced that disentangling spatial, land use, and marine vs. continental effects on diversity components is critical to understand the primary determinants of the fish diversity in tropical coastal ecosystems.
Palabras clave
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Asunto principal:
Monitoreo del Ambiente
/
Biodiversidad
/
Peces
/
Modelos Biológicos
Tipo de estudio:
Prognostic_studies
Límite:
Animals
País/Región como asunto:
America do sul
/
Brasil
Idioma:
En
Revista:
Mar Environ Res
Asunto de la revista:
BIOLOGIA
/
SAUDE AMBIENTAL
/
TOXICOLOGIA
Año:
2019
Tipo del documento:
Article
País de afiliación:
Brasil
Pais de publicación:
Reino Unido