Dissemin is shutting down on January 1st, 2025

Links

Tools

Export citation

Search in Google Scholar

Morphological metrics and unsupervised neural networks to analyse urban sprawl and intercity commuting

This paper is available in a repository.
This paper is available in a repository.

Full text: Download

Question mark in circle
Preprint: policy unknown
Question mark in circle
Postprint: policy unknown
Question mark in circle
Published version: policy unknown

Abstract

This article focuses on methods and techniques to investigate urban sprawl and intercity commuting regarding the city of São José dos Campos and cities of the Paraíba Valley region, southeast of Brazil. Upon basis of a spatial-relational database containing both its census districts and data derived from origin-destination interviews, which have been applied through stratified sampling, an assessment of its inter-and intra-regional mobility dynamics was carried out, aiming to characterize the profile of commuting regarding not only house-to-work trips, but also trips for the purposes of study, shopping, services, leisure and recreation in secondary residences, and their respective frequencies. This assessment has been conducted by means of unsupervised neural networks, meant to jointly analyse socioeconomic patterns of the population in relation to the trips purpose and frequencies, so as to recognize clusters of similar trip profiles regarding typologies (destination, transport modes, purpose) and intensity (frequencies) in São José dos Campos. These investigations have been supported by urban morphology analyses through landscape metrics, obtained from the delimitation of main urban centres and urban sprawl nuclei extracted from medium-and high-resolution satellite imagery.