What evidence exists on birds and mammals' biodiversity in the Brazilian Atlantic Forest (BAF) agricultural ecosystems? A systematic map protocol.

Background: Brazil has one of the planet's greatest biodiversity, with over 20% of the world's total species. The Brazilian Atlantic Forest (BAF) spans 17 Brazilian states, making it the third-largest biome in Brazil. The BAF is composed of a range of ecological formations, with climatic conditions and landscape diversity that directly contribute to the different structures of the forest. The fragmentation of the original habitats, mainly due to anthropogenic activities, is one of the main causes of biodiversity loss, causing the decline and habitat loss for several species, including birds and mammals. While there has been extensive research on species diversity in forest fragments or protected areas, there is a lack of basic research in agricultural ecosystems. Only 1.5% of the studies on bird occurrence data in the BAF were reported from pasture habitats and 1.4% from exotic tree plantations. To address this gap, the present systematic map protocol aims to carry out a bibliographic survey on the presence of birds and mammals in agricultural landscapes and its adjacent areas of natural vegetation to describe the prevalence of species across different (semi-)natural and anthropogenic habitat types. Collecting this data is important to support environmental management policies to preserve biodiversity in these areas. Methods: We will conduct a systematic literature review on the biodiversity of birds and mammals in agricultural landscapes within the Brazilian Atlantic Forest as well as adjacent areas of natural vegetation. Our search will cover the following databases, without limiting the year of publication: Web of Knowledge, Scopus, and PubMed. We will also include grey literature such as dissertations and theses, performing the search at the "Biblioteca Digital Brasileira" database. The results will be screened for relevance based on predefined criteria. The screening process will take place in two stages: firstly, the articles will be screened by title and abstract, and then the eligible articles will be screened in full text. Only articles that meet the eligibility criteria will proceed to data extraction. The extracted data will provide the elements to build a systematic map. [ABSTRACT FROM AUTHOR]