Informal settlements present a significant urban challenge in Valparaíso, driven by socioeconomic pressures, migration, and geographical constraints. This study utilizes open geospatial data and advanced remote sensing techniques to detect, map, and analyze the temporal and spatial dynamics of these settlements between 2017 and 2022.
Satellite imagery from Sentinel-2 and Landsat-8, combined with topographic and socioeconomic data, was processed using machine learning models for land-cover classification. Random Forest emerged as the most accurate algorithm, effectively mapping slum areas and revealing patterns of expansion and contraction. The study also incorporated terrain metrics such as slope and elevation, critical in Valparaíso’s topography, to assess their influence on settlement distribution.
The results indicate that slum areas fluctuated over the study period, from 1.14 km² in 2017 to 0.83 km² in 2022, reflecting dynamic land-use patterns influenced by migration, housing market pressures, and policy decisions. The findings underscore the role of demographic and economic factors in shaping informal urban growth, exacerbated by inadequate formal housing options and socio-spatial marginalization.
Previous Article in event
Previous Article in session
Next Article in event
Next Article in session
Detecting and mapping the structure and pattern of informal settlements using open data: a case study in Valparaíso
Published:
25 March 2025
by MDPI
in International Conference on Advanced Remote Sensing (ICARS 2025)
session Urban Remote Sensing
Abstract:
Keywords: Slums; remote sensing; machine learning; land cover; land use; land change; data mining; Valparaiso
Comments on this paper
