Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-01-13T13:10:17Z No. of bitstreams: 1 KarlaJulieteDePaivaSilva_DISSERT.pdf: 3412095 bytes, checksum: c27e612052c475a4f87b4cafbfe37fc5 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-01-24T12:16:16Z (GMT) No. of bitstreams: 1 KarlaJulieteDePaivaSilva_DISSERT.pdf: 3412095 bytes, checksum: c27e612052c475a4f87b4cafbfe37fc5 (MD5) / Made available in DSpace on 2017-01-24T12:16:16Z (GMT). No. of bitstreams: 1 KarlaJulieteDePaivaSilva_DISSERT.pdf: 3412095 bytes, checksum: c27e612052c475a4f87b4cafbfe37fc5 (MD5) Previous issue date: 2016-07-29 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / Objetivo: As comunidades de restinga s?o formadas predominantemente por esp?cies oriundas dos Dom?nios Caatinga, Cerrado, Amaz?nia e Mata Atl?ntica, que passam pelo forte filtro ambiental representado pelas condi??es estressantes caracter?sticas de ambientes costeiros. Neste estudo, propomos testar se a varia??o ambiental do litoral brasileiro ? suficiente para criar filtros adicionais para as esp?cies da restinga. Localiza??o: Toda a extens?o da costa do Brasil, Am?rica do Sul. M?todos Constru?mos uma base de dados contendo informa??es bin?rias da flora terrestre da restinga presente em 164 localidades ao longo do litoral brasileiro. Para cada uma destas localidades, obtivemos um conjunto de 41 vari?veis ambientais, incluindo vari?veis clim?ticas e ed?ficas. Usamos Modelos de Arqu?tipos de Esp?cies (MAEs) para avaliar a resposta das esp?cies da restinga ? varia??o representada pelas vari?veis abi?ticas. Tamb?m investigamos a exist?ncia de gradientes flor?sticos atrav?s de uma An?lise de Coordenadas Principais (ACoP). Usamos estas duas abordagens estat?sticas para os dados de esp?cies herb?ceas e lenhosas, separadamente. Resultados: Foram formados 4 arqu?tipos de esp?cies herb?ceas em resposta a 10 vari?veis ambientais e 10 arqu?tipos de esp?cies lenhosas em resposta a 6 vari?veis ambientais. Os arqu?tipos de esp?cies herb?ceas responderam aos gradientes ambientais mais fortemente que os arqu?tipos de esp?cies lenhosas, embora, em geral, apenas poucos arqu?tipos mostraram fortes respostas ? varia??o ambiental. Adicionalmente, as comunidades herb?ceas da restinga n?o foram estruturadas por gradientes flor?sticos significativos e as comunidades lenhosas n?o formaram gradientes flor?sticos n?tidos, sugerindo um alto n?vel de estocasticidade na forma??o da estrutura flor?stica da restinga. Principais Conclus?es: A varia??o ambiental do litoral brasileiro parece criar filtros adicionais para as esp?cies herb?ceas e lenhosas da restinga, o que ? mais importante para a distribui??o das esp?cies herb?ceas. A baixa resposta das esp?cies ?s vari?veis ambientais sugere que a varia??o flor?stica da restinga resulta mais da rela??o espacial entre a restinga e os Dom?nios flor?sticos adjacentes. / Aim The plant communities of restinga are shaped by species from Caatinga, Cerrado, Amazon and Atlantic Domains which are able to cope with stressful conditions from the Brazilian coast. We tested whether environmental variation along the Brazilian coast is strong enough to impose additional filters for the restinga species. Location The entire coast of the Brazil, South America. Methods We gathered information about species composition from 164 distinct localities along the Brazilian coast to construct a binary database of restinga flora. For each of these localities, we obtained a set of 41 environmental variables, including climate and edaphic variables. We used Species Archetype Models (SAMs) to evaluate the response of the restinga species to the variation represented by abiotic variables. We also investigated the existence of floristic gradients using a Principal Coordinates Analysis (PCoA). We use these two statistical approaches for data of herbaceous and woody species, separately. Results We found four archetypes of herbaceous species in response to 10 environmental variables, and 10 archetypes of woody species in response to 6 environmental variables. The archetypes of herbaceous responded to environmental gradients more strongly than archetypes of woody species, althout, in general, just few archetypes showed strong responses to environmental variation. Additionally, the herbaceous communities from restinga were not structured by significant floristic gradients and the woody communities did not have clear floristic gradientes, suggesting a high level of stochasticity shapping the floristic structure of restinga. Main conclusions The environmental variation along the Brazilian coast seems to create additional filters for herbaceous and woody species of restinga, which are more important to the distribution of herbaceous species. However the low response of species to the environment suggests that the floristic variation of restinga results over from a spatial relationship between the restinga and the Caatinga, Cerrado, Amazon, Atlantic and Pampa Domains.