Priscilla — 2025

Hurricane Priscilla formed from a monsoon trough off the southern Mexican coast and became a tropical depression at 1200 UTC on 4 October 2025 about 260 n mi south-southwest of Manzanillo, Mexico. It moved generally northwestward and strengthened to a tropical storm later that day, became a hurricane at 0000 UTC 6 October, and reached its peak intensity on 7 October while a few hundred nautical miles south-southwest of the southern tip of the Baja California Peninsula. Priscilla weakened steadily after 7 October, became a post-tropical remnant low at 0000 UTC 11 October, and dissipated by 0600 UTC 12 October. Priscilla remained offshore and did not make a direct landfall on mainland Mexico or the Baja California Peninsula. Tropical storm watches were issued for stretches of the mainland (Punta San Telmo to Punta Mita and portions near Manzanillo) and for parts of Baja California Sur (Cabo San Lucas to Santa Fe and later Santa Fe to Cabo San Lázaro), but the system stayed primarily over water and those watches were later discontinued. The storm’s maximum sustained winds peaked at 100 kt (115 mph) at 1800 UTC on 7 October, making Priscilla a Category 3 major hurricane at its strongest. The best-track minimum central pressure at peak intensity was 954 mb, based on a wind–pressure relationship used in the analysis. Priscilla produced heavy rainfall along the southwestern coast of Mexico while remaining offshore. Notable storm-total rainfall amounts included 13.24 inches (336.2 mm) at Socorro Island, 10.57 inches (268.4 mm) at Zihuatanejo in Guerrero, and 10.55 inches (268.0 mm) at Melchor Ocampo in Michoacán. Localized flooding and rough surf were reported near Puerto Vallarta and in Baja California Sur; runoff and remnant moisture also contributed to flash flooding across northwestern Mexico, Baja California, and into the southwestern United States after the system weakened. No deaths were reported in association with Priscilla. Reported impacts were localized flooding, rough surf damage near Puerto Vallarta, and in Baja California Sur where nine homes were affected and 309 people were temporarily displaced. Forecasts and watches had good lead time: NHC genesis, track, and intensity forecasts for Priscilla were generally accurate, with official track errors well below recent 5‑year means and intensity errors below the 5‑year means; an experimental Google DeepMind ensemble often outperformed other guidance on track.