AI2 updates its HiRO-ACE climate emulator to produce 3 km global weather fields with improved extreme-event accuracy at substantially lower cost
Research publication Provisional 86% confidence first seen
The Allen Institute for AI released HiRO-ACEv2, an update to its HiRO-ACE climate emulator that improves downscaling of coarse climate simulations to 3 km global weather fields. The update adds a downscaling model to better predict rare extreme events (e.g., rainfall, wind, and pressure) and reduces the estimated cost of generating global 3 km climate data to about 1/300th of X-SHiELD, while enabling longer high-resolution datasets.