Satellite heat sensing (VIIRS on NOAA-20) picked up 9 detections with fire radiative power at or above 20 MW over two days inside the watched boundary of Parque Nacional Serra das Confusões in Brazil, the hottest at 57.8 MW. A later pass on 29 September found 4 more hot detections in the same area, which is the independent check this finding needed. Fire density inside the boundary ran about 1.584 times the surrounding 30 km neighbourhood (4.662 versus 2.942 hot detections per deg²).
What this is not: it is not evidence of forest clearing, and it does not say who lit anything or why. The sensor cannot separate a wildfire from managed, agricultural or legally prescribed burning. The season is running under a strong El Niño (ONI 1.8), a pattern associated with drier, more fire-prone conditions in this region. The watched area is a bounding box, not the park's legal polygon, so some detections may sit outside the park.
What would change our mind: mapping against the legal boundary placing the heat outside the park, a burn permit or management record covering these dates, or later passes and imagery showing a brief, contained burn scar.
Written by AI, then checked by a separate reviewer before publication. Every claim points at the evidence below.