Spatiotemporal autocorrelation of landslides in neighborhoods of Recife, Brazil (2015–2024) based on the Moran and Lisa index

This study analyzes the spatiotemporal dynamics of landslide occurrences in neighborhoods of the municipality of Recife (PE), from 2015 to 2024, using spatial analysis techniques based on the Global Moran’s Index and Local Indicators of Spatial Association (LISA). Official data from the Recife Civil Defense, as well as landslide susceptibility maps prepared by CPRM and municipal monitoring information, were used. The analyses were performed in a Python environment, with the support of geoprocessing and spatial statistics tools. The results highlight the persistent presence of positive spatial autocorrelation of landslides throughout the historical series, with significant values of the Global Moran’s Index in most of the years analyzed. Local analysis (LISA) allowed the identification of concentrated High-High type spatial clusters, mainly in the northern and southern extremes of the municipality, highlighting neighborhoods such as Guabiraba, Dois Unidos, Nova Descoberta, COHAB and Ibura. Analysis of the frequency and persistence of hotspots revealed chronic risk areas, reinforcing the consistency between observed empirical patterns and official susceptibility maps. It is concluded that landslide occurrences in Recife do not occur randomly, but exhibit persistent spatial clustering patterns that are consistent with areas of greater landslide susceptibility and recurrent socio-environmental exposure. These findings support the development of territorially focused risk-management and mitigation policies.