Estimating the effects of urban green regions in terms of diffusion. (May 2023)

• Record Type: Journal Article
• Authors: Tokuda, Eric K; de Arruda, Henrique F; Domingues, Guilherme S; Costa, Luciano da F; Shibata, Florence AS; Cesar-Jr, Roberto M; Comin, Cesar H
• Other Names: O'Clery Neave guest-editor.; Duque Juan Carlos guest-editor.; Alvanides Seraphim guest-editor.; Schwanen Tim guest-editor.
• Is Part Of: Environment & planning. Volume 50:Number 4(2023)
• Page Start: 1023
• Page End: 1038
• Publication Date: 2023-05
• Subjects: green space -- urban simulation -- complexity sciences -- network graphs; City planning -- Periodicals; Urban ecology (Sociology) -- Periodicals; 307.11605
• DOI: 10.1177/23998083221131572
• Languages: English
• ISSNs: 2399-8083

Abstract

The interaction between cities and their respective green regions corresponds to an interesting issue that has received growing attention over the last decades. These relationships have multiple natures, ranging from providing habitat for animal life to temperature and humidity dynamics. Several methods based on area, size, shape, and distance have been considered in the literature. Given that several important contributions of green regions to urban areas involve temperature, humidity, and gases exchanges, which are intrinsically related to physical diffusion, it becomes particularly interesting to simulate the diffusion of green effects over urban areas as a means of better understanding the respective influences. The present work reports a related approach. Once the green regions of a given city are automatically identified by semantic segmentation and have eventual artifacts eliminated, successive convolutions are applied as a means to obtain the unfolding of the diffusion of the green effects along time. As illustrated, the diffusion dynamics is intrinsically interesting because it can be strongly affected by the spatial distribution of the green mass. In particular, we observed that smaller green regions could substantially contribute to the diffusion. The reported approach has been illustrated with respect to the Brazilian city of Ribeirão Preto, whose small- and medium-sized green regions were found to complement in an effective manner the diffusion of the green effects as inferred from the performed simulations under specific parameter settings.