Favelas 4D: Scalable methods for morphology analysis of informal settlements using terrestrial laser scanning data

TL;DR

This paper introduces a scalable, automated methodology using terrestrial LiDAR data to analyze the complex morphology of informal settlements, providing detailed maps that can inform urban planning and safety assessments.

Contribution

It presents a novel scalable approach for morphological analysis of informal settlements using terrestrial LiDAR, applicable at both global and local scales.

Findings

Revealed meaningful morphological differences and similarities across streets.

Created high-resolution morphological maps for urban planning.

Demonstrated the scalability and automation of the method.

Abstract

One billion people live in informal settlements worldwide. The complex and multilayered spaces that characterize this unplanned form of urbanization pose a challenge to traditional approaches to mapping and morphological analysis. This study proposes a methodology to study the morphological properties of informal settlements based on terrestrial LiDAR (Light Detection and Ranging) data collected in Rocinha, the largest favela in Rio de Janeiro, Brazil. Our analysis operates at two resolutions, including a \emph{global} analysis focused on comparing different streets of the favela to one another, and a \emph{local} analysis unpacking the variation of morphological metrics within streets. We show that our methodology reveals meaningful differences and commonalities both in terms of the global morphological characteristics across streets and their local distributions. Finally, we create morphological maps at high spatial resolution from LiDAR data, which can inform urban planning assessments of concerns related to crowding, structural safety, air quality, and accessibility in the favela. The methods for this study are automated and can be easily scaled to analyze entire informal settlements, leveraging the increasing availability of inexpensive LiDAR scanners on portable devices such as cellphones.