Satellite fire sensors (VIIRS on NOAA-20) logged 16 detections with fire radiative power of at least 20 MW over two days inside the watched boundary of Parque Nacional Do Araguaia, the strongest reaching 89.4 MW. A later pass on 28 September found 55 hot detections again, with a peak of 371.8 MW — the independent revisit that confirms this is not a one-off sensor artefact. Detection density inside the area ran about 2.787 times that of the surrounding 30 km ring. Conditions are El Niño (ONI 1.8), which tends to dry fuels across this part of Brazil.
What this is not: heat alone does not say who lit it or why. The rule cannot separate a wildfire from managed pasture burning, agricultural fire or a legal prescribed burn, and it says nothing about forest clearing. No nearby natural-hazard event is on record. The watched area is a bounding box, not the park's legal polygon, so some detections may fall outside the protected boundary.
What would change our mind: burn-scar mapping showing the fire footprint outside the legal polygon, a documented prescribed-burn authorisation, or later passes showing no scar where the heat was recorded.
Written by AI, then checked by a separate reviewer before publication. Every claim points at the evidence below.