No abstract

No abstract

Metacommunity models predict that species richness and composition patterns in communities are determined predominantly by environmental selection and dispersal, with speciation and drift playing a lesser role. In the tropics, our understanding about these processes comes almost purely from empirical data; there is lack of formal confrontation between models and data. Here, we evaluated if a metacommunity simulation model designed for riverine networks could predict insect diversity patterns observed in a tropical stream network within the Atlantic Forest biome. Our individual‐oriented model simulated biological processes, including dispersal of adults, oviposition, dispersal of larvae, colonisation and mortality. We used this model to simulate communities considering three assembly scenarios (niche, niche–stochastic and stochastic) and three flight dispersal ability groups (low, medium and high). To parameterise the model, we used data from 97 riffle communities, distributed along a stream network, and biological data from the literature. For the high dispersal ability group, none of the assembly scenarios significantly predicted the observed richness and community dissimilarity. The niche–stochastic assembly scenario, however, provided a better approximation of observed richness and community dissimilarity for the medium and low dispersal ability groups. Our results show that deterministic and stochastic processes combine to better explain observed aquatic insect diversity patterns. Although simulations were limited to similar explanatory power to studies using pattern‐oriented approaches, our findings highlight the promise of a more predictive use of metacommunity simulation models.

RESUMO O controle biológico no Brasil vem crescendo devido aos problemas gerados pelo uso indiscriminado de inseticidas químicos. A Musca domestica representa o maior problema em granjas avícolas devido às condições favoráveis para seu crescimento populacional. Sendo assim, foram realizadas capturas de dípteros muscoides em um aviário na região de Montes Claros, MG, usando armadilhas contendo isca química e captura por busca direta e, destas moscas, foram isolados e identificados fungos residentes nestes insetos. Os fungos isolados foram duas espécies do gênero Aspergillus sp., um do gênero Memnoniella sp., Scopulariopsis sp., Paecilomyces sp e um fungo da família Moniliaceae. Foram também requeridos junto ao CENARGEM/EMBRAPA as espécies de fungos entomopatogênicos Beauveria bassiana CG 470 e CG 472; Metarhizium anisopliae CG 34 e CG 312 e o Paecilomyces sp. CG 301. As espécies selecionadas para os bioensaios foram um Aspergillus sp., Memnoniella sp. e os Metarhizium anisopliae CG 34 e CG 312 por terem boa esporulação. Os fungos Aspergillus sp. e o Memnoniella sp. não apresentaram capacidade entomopatogênica, os fungos M. anisopliae CG 34 e CG 312 foram bastante eficientes em controlar a emergência dos adultos de M. domestica, mostrando-se bons agentes de controle biológico.