Detalhes bibliográficos
| Ano de defesa: |
2021 |
| Autor(a) principal: |
CROVADOR JUNIOR, SIDNEI ANTONIO
 |
| Orientador(a): |
Figueiredo Filho, Afonso
|
| Banca de defesa: |
Não Informado pela instituição |
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Universidade Estadual do Centro-Oeste
|
| Programa de Pós-Graduação: |
Programa de Pós-Graduação em Ciências Florestais (Mestrado)
|
| Departamento: |
Unicentro::Departamento de Ciências Florestais
|
| País: |
Brasil
|
| Palavras-chave em Português: |
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
http://tede.unicentro.br:8080/jspui/handle/jspui/2078
|
Resumo: |
Recently, the concern with the quality of the landscape has been a topic that has gained greater attention, and studies on the quality of the urban environment can be considered vital tools for subsidizing planning, as they provide information that helps to promote quality life for the citizens. By the direct method, this research aims to determine the Landscape Visual Quality (LVQ) of wooded and non-wooded sidewalks in Curitiba, Paraná state, Brazil. As a study model, twenty-four streets (wooded and non-wooded) were chosen from the four Residential Zones of the city, in which cross-sectional photographs were collected, taken in both directions of the streets, in the three segments more representative and at two different angles (A1 and A2), hemispheric and variable photographs of trees (DBH, total height, and crown diameter) and of the urban structure (sidewalk width, width of the front setback and height of buildings), for qualify and analyze the landscape of the sidewalks. The collection was repeated in the four seasons of the year. From the variables obtained in the images (percentage of vegetation, sky, and built environment), the LVQ of each image was determined, adding the rates after assigning weights to the variables. With the LVQ data, a Cluster Analysis was performed to establish three classes of the visual quality of the landscape, defining them as low, medium, and high. An analysis of the differences between zones, seasons, and the photographs' angles was carried out through a Random Block Design in a factorial scheme. Also, with the values of LVQ and other variables measured “in loco”, adjustments were made to multiple linear regression models to estimate the visual quality of the landscape, applying the Stepwise procedure to select the best variables. The evaluated landscapes on streets with tree-lined sidewalks, 26.4% had Landscape Visual Quality classified as High (H), 50.0% as Medium (M), and 23.6% as Low (L). On the other hand, in the streets with non-tree sidewalks, 23.6% had LVQ classified as High (H), 41.7% as Medium (M), and 34.7% as Low (L). By Tukey's test observed that Residential Zone 4 (RZ4) differs statistically from the others for tree-lined streets, with the lowest average. Elseways, on roads with non-wooded sidewalks, Residential Zone 1 (RZ1) differed significantly from the others, with a higher standard. As for the seasons of the year, for streets with tree-lined sidewalks, the winter season showed a significant difference in LVQ about the other seasons of the year (p-value<0.01). As for non-tree streets, it could be noted that autumn had the highest average. Regarding the angle of the photographs, differences were found in the tree-lined streets, according to the angles of the photographs and the residential areas. As for modeling, the results suggest that Generalized Additive Models have a greater predictive capacity for tree-lined street data. The models generated that showed a better performance is composed of variables related both tree elements and urban infrastructure, and the Summer model showed superior statistics (R² = 0.59). For non-tree streets, machine learning presented the best estimates, and the Spring station model showed the best results (R² =0.35). |
