Answers are grounded in this paper and cite their sources.
Extreme rainfall events are major hydroclimatic hazards in Brazil, causing floods, landslides, infrastructure damage, and agricultural losses across distinct climatic regions. This study develops a Bayesian spatio‐temporal Log‐Gaussian Cox Process (LGCP) framework to model the occurrence of extreme precipitation using CHIRPS daily rainfall data. The results reveal clear differences between moderate‐extreme and ultra‐extreme rainfall regimes. Lower thresholds are strongly influenced by seasonal forcing associated with the South American Monsoon System, whereas the most severe events exhibit stronger temporal dependence, episodic clustering, and more localized spatial organization, indicating increasingly persistent spatio‐temporal structures in the latent risk field as precipitation intensity increases. Regional decompositions further show substantial heterogeneity in trends, seasonality, and persistence across Brazil, reflecting the influence of distinct hydroclimatic regimes and atmospheric processes. The estimated latent spatial fields identify persistent high‐risk corridors over the Amazon basin, southeastern coastal Brazil, and southern subtropical regions. Overall, the findings suggest that extreme rainfall occurrence in Brazil is governed by multiple interacting regional processes whose persistence and spatial organization may evolve under climate variability and climate change.
No takes yet. Share an insight, caveat, or question.
Recacho et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: