Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ LAReferencia - Red F...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
versions View all 1 versions
addClaim

Impacto da resolução espacial na modelagem da expansão urbana da região metropolitana de São Paulo

Authors: Massabki, José Augusto Rodrigues;

Impacto da resolução espacial na modelagem da expansão urbana da região metropolitana de São Paulo

Abstract

The objective of this work was to analyze the impacts of the spatial resolution related to the urban expansion modeling of the Metropolitan Region of São Paulo (MRSP) by means of Cellular Automata (CA) combined with Artificial Neural Networks. Data regarding the urbanized perimeter in the periods of 1881, 1905, 1929, 1949, 1974 and 2005 were used for the construction of a series of spatial models. A grid of regular cells was conceived to represent the study area, whose spatial resolution was based on cells of 1,000 by 1,000 meters, 800 by 800 meters and 600 by 600 meters. The models also took into account the combination of variables representing the initial state of the cell (urban or non-urban), the number of neighboring cells classified as urban, and the number of neighboring cells classified as non-urban. The results showed that the variation on the spatial resolution did not result in a significant impact on the performance of the developed models, since they were quite similar across the different structures and grid sizes explored. There were some highlights, for example, the Grid of 1000 by 1000 meters, based on the state of the cell and the number of urban neighbors, with 90.09% of global correct predictions; the Grid of 800 by 800 meters, based on the state of the cell, the number of urban neighbors and the number of non-urban neighbors, with 90.25% of global correct predictions; and finally, the Grid of 600 by 600 meters, based on the state of the cell and number of urban neighbors, with 90.14% of global correct predictions. In the sequence, a forecast of the territorial occupation was simulated for year 2030 considering those highlighted models. The results showed some patterns of urban expansion oriented by infrastructures, as well as processes of urban occupation in inappropriate areas. In sum, this study demonstrated that an apparent increase in the spatial resolution of the grid does not produce positive effects in relation to an increase in the performance of the spatial models developed in this work. Therefore, the Grid of 1000 by 1000 meters can serve as a tool for urban planning processes according to the methodology used. It should also be noted that this grid ends up making the computational processing lighter, due to the smaller amount of data involved.

O objetivo deste trabalho foi analisar os impactos da resolução espacial na modelagem da expansão urbana da Região Metropolitana de São Paulo (RMSP) por meio de autômatos celulares (CA, do inglês: Cellular Automata) combinados com redes neurais artificiais. Para tanto, foram utilizados dados do perímetro urbanizado referentes aos períodos de 1881, 1905, 1929, 1949, 1974 e 2005 para a construção de uma série de modelos espaciais. Uma estrutura de células regulares (grid) foi concebida para se representar a área de estudo, cuja resolução espacial se baseou em células de 1.000 por 1.000 metros, 800 por 800 metros e 600 por 600 metros. Os modelos também levaram em consideração a combinação de variáveis representando o estado inicial da célula (urbana ou não urbana), o número de células vizinhas classificadas como urbanas e o número de células vizinhas classificadas como não urbanas. Os resultados mostraram que a variação da resolução espacial não proporcionou impacto significativo no desempenho das modelagens desenvolvidas, visto que os desempenhos obtidos a partir das diferentes estruturas e tamanhos de grid foram bastante similares. Há os destaques para o Grid de 1000 por 1000 metros, baseado no estado e no número de vizinhos urbanos, com 90,09% de acertos global; para o Grid de 800 por 800 metros, baseado no estado, no número de vizinhos urbanos e no número de vizinhos não urbanos, com 90,25% de acertos global; e finalmente, para o Grid de 600 por 600 metros, baseado no estado e no número de vizinhos urbanos, com 90,14% de acertos global. Na sequência, uma previsão de ocupação do território para 2030 foi avaliada para esses modelos em destaque, observando-se tanto padrões de expansão urbana orientados por infraestruturas, como processos de ocupação urbana em áreas impróprias. Em síntese, esse estudo demonstrou que um aparente aumento na resolução espacial do grid não produz efeitos positivos em relação a um aumento no desempenho dos modelos espaciais desenvolvidos neste trabalho. Dessa forma, o Grid de 1000 por 1000 metros pode servir como ferramenta nos processos de planejamento urbano conforme a metodologia empregada. Cabe ressaltar ainda que esse grid acaba tornando o processamento computacional mais leve, em função da menor quantidade de dados envolvidos.

Pós-graduação em Engenharia Civil e Ambiental - FEB

Country
Brazil
Keywords

Redes neurais artificiais, Modelagem espacial, Autômatos celulares

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green