Projected heatwave hazards in an eastern Amazonian metropolis under contrasting CMIP6 scenarios.

Heatwaves (HWs) are among the most impactful climate extremes affecting tropical urban environments, yet local‑scale assessments of their future characteristics remain scarce in the Amazon. Here, we quantify projected changes in key HW indicators for the Belém urban core, eastern Amazon, using a station‑referenced framework based on a carefully evaluated and bias‑corrected ensemble of CMIP6 global models. HWs are identified from daily maximum air temperature (Tmax) using a smoothed day‑of‑year 95th percentile threshold calculated over the 1994-2023 baseline period. Changes in annual frequency, event duration, and thermal intensity are assessed for near‑future (2025-2049) and far‑future (2050-2074) periods under four Shared Socioeconomic Pathways (SSP1‑2.6, SSP2‑4.5, SSP3‑7.0, and SSP5‑8.5). A multi‑metric performance evaluation is first applied to select the best‑performing CMIP6 models for Tmax over Belém, followed by bias correction using Quantile Delta Mapping calibrated against local observations. Results show a clear, scenario‑dependent intensification of HW hazards. In the current climate, 126 HW events were detected, with pronounced interannual variability and a marked increase during the last decade. Near‑future changes are modest and variable across scenarios. In contrast, far‑future projections under the high‑emission SSP5‑8.5 pathway reveal a substantial transformation of the HW regime: annual frequency more than doubles relative to the baseline, and the 99th percentile of event duration increases from approximately 19 to 65 days (Δ = +46 days), with a 14.5% probability that any given HW will exceed the current extreme threshold. Peak Tmax during HWs increases from 37.0 °C to 37.55 °C (Δ = +0.55 °C), thus indicating an upper-tail amplification. Under the strong mitigation pathway SSP1‑2.6, changes in HW duration and intensity remain statistically indistinguishable from present‑day conditions. While this study does not assess impacts or adaptation capacity, the results provide a hazard‑based, scenario‑dependent characterization of future heatwave behavior in an Amazonian urban context, offering a quantitative foundation for subsequent assessments of climate risk in tropical cities.