Made available in DSpace on 2016-06-02T19:31:53Z (GMT). No. of bitstreams: 1 3492.pdf: 2323425 bytes, checksum: c5d62a190e73b4125711e80147b1bbfb (MD5) Previous issue date: 2011-02-04 / Universidade Federal de Minas Gerais / There is much variation in patterns of edge influence (EI) within a region due to factors such as vegetation structure, adjacent land use, orientation and edge contrast, and assessment of this variation is important for conservational planning. We studied EI on aspects of microclimate and vegetation in seven cerrado fragments in South-Eastern Brazil, sampling both savanna and forest vegetation. We located transects perpendicular to 14 cerrado edges adjacent to pastures, highways, sugarcane, silviculture, urban area, and forest roads. When significant, distance of EI varied from 15 to 80 m for temperature, 2 to 50 m for air moisture, 2 to 20 m for maximum vegetation height, 2 m for wind speed, 0 to 100 m for canopy cover, 0 to 20 m for exotic grass biomass, and 0 to 10 m for native grass biomass. The largest EI was observed in areas adjacent to pastures, but factors such as orientation, vegetations structure and land use were not sufficient to explain the observed variability. For conservation purposes, we recommend that buffers should be created between cerrado and the adjacent land use, especially in fragments adjacent to pastures and higways, in order to reduce EI. / Existe ampla variação nos padrões de influência de borda (IB) dentro de uma região, devido a fatores como estrutura da vegetação, uso da terra adjacente, orientação e contraste da borda. Entender esta variação é importante para o planejamento direcionado à conservação. Nós estudamos IB sobre aspectos de microclima e vegetação em sete fragmentos de cerrado no Sudeste do Brasil, amostrando áreas tanto de savana quanto de floresta. Nós delimitamos transectos perpendiculares a 14 bordas de cerrado adjacentes a pastagens, estradas, cana-de-açucar, silvicultura, área urbana e aceiros. Quando significativa, a distância de influência de borda variou de 15 a 80 m para temperatura, 2 a 50 m para umidade do ar, 2 m para velocidade do vento, 2 a 20 m para altura máxima de vegetação, 0 a 100 m para cobertura de dossel, 0 a 20 m para biomassa de gramíneas exóticas e 0 a 10 m para biomassa de gramíneas nativas. A maior IB foi observada em áreas adjacentes a pastages, mas fatores como orientação, estrutura da vegetação e uso da terra não foram suficientes para explicar a variabilidade observada. Para propósitos de conservação visando reduzir a IB, nós recomendamos a criação de zonas de amortecimento entre o cerrado e áreas adjacentes com outros usos da terra, especialmente em fragmentos adjacentes a pastagens e estradas.

The world-wide intensification of agriculture has led to a decline in species richness due to land use change, isolation, and fragmentation of natural and semi-natural habitats in agricultural and forestry landscapes. As a consequence, there is a current landscape management focus on the importance of green infrastructure to mitigate biodiversity decline and preserve ecosystem functions e.g. pollination services and pest control. Even though intensification in agriculture has been ongoing for several hundreds of years, remnant habitats from earlier management practices may still be remaining with a surprisingly high plant richness. Preserving these habitats could help conserving plant species richness in agricultural landscapes, as well as other organisms that are dependent on plants for food and shelter. In this thesis I focus on two small remnant habitats; midfield islets and borders between managed forest and crop field in southeastern Sweden. In the past, both habitats were included in the grazing system and therefore often still have remnant population of grassland specialist species left today. I have used these two remnant habitats as model habitats to investigate the effect of landscape factors and local factors on species richness of plants, flower morphologies and plants with fleshy fruits. Additively, I analysed the effect of surrounding landscape and local openness on the functions; pollination success, biological pest control of aphids and seed predation on midfield islets. One of my studies showed that spatial distribution and size of the habitat affected plant species richness. Larger habitat size and higher connectivity between habitats increased species richness of plants in the habitats. Openness of the habitats was shown to be an important factor to increase species richness and richness of flower morphologies, both on midfield islets and in forest borders. Even though midfield islets had the highest species and morphology richness, both habitat types are needed for habitat complementary as forest borders have more plants with fleshy fruits and a higher richness of plant species that flowers in spring/early summer. It was also shown that a more complex forest border, not just with gaps in the canopy, but also with high variation in tree stem sizes increases plant species richness in the field layer. The conclusion is that by managing small remnant habitats to remain or become more semi-open and complex in their structure, would increase species richness of plants, grassland specialist species, and flower morphologies. It would also increase some ecosystem functions as seed predation and biologic pest control of aphids are more effective close to trees. If both midfield islets and forest borders would be managed to be semi-open, the area and connectivity of semi-open habitat would increase in the agricultural landscape, which may also improve pollination success as the connectivity between populations has a possibility to increase. Grassland specialist species are clearly abundant in the small remnant habitats. As the decline of semi-natural grasslands is causing a decline in grassland specialists’ species, not only plants, I recommend that small remnant habitats are included in conservation and management plans and strategies to improve habitat availability and connectivity for grassland species in agricultural landscapes. / <p>Research funder Ekoklim. Project:4339602.</p><p>At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 4: Manuscript.</p>