Ai2 Introduces HiRO-ACEv2 for High-Resolution Climate and Extreme Weather Simulation

Ai2 Introduces HiRO-ACEv2 for High-Resolution Climate and Extreme Weather Simulation The Allen Institute for AI has released HiRO-ACEv2, an updated climate emulation system capable of generating global 3 km resolution weather data at roughly 1/300th the cost of the X-SHiELD physics model. The update enhances the downscaling model to predict multiple fields simultaneously and accurately reproduce rare extreme weather events. HiRO-ACEv2 couples ACE at 100 km resolution with the updated HiROv2 downscaler to produce 3 km data at about 1/300th the cost of X-SHiELD. HiROv2 downscales multiple atmospheric fields—rain, wind, and pressure—matching high-resolution statistics to within 3% even for one-in-a-million extreme events. A streamlined HiROv2 variant under development could increase speed by more than ten times. Current models are constrained to X-SHiELD-tuned data and cannot yet simulate different historical or warmer climates, such as ERA5 or a 4°C warming scenario. System Architecture and Extreme Weather Modeling HiRO-ACE, introduced earlier in 2026, combines ACE, which simulates coarse weather at 100 km, and HiRO, which sharpens output to a 3 km grid. The updated HiROv2 downscales multiple high-resolution fields at once—rain, wind, and sea-level pressure—rather than precipitation alone. Training updates allow it to match target X-SHiELD statistics within 3%, accurately simulate mountain slope winds, capture total energy of unseen tropical cyclones, and depict hurricane eyes and rain bands, despite slightly lopsided winds and pressure near the eye. Long-Term Simulations and Cost Efficiency When driven by a newly trained version of ACE, HiRO-ACEv2 produced a full year of global 3 km data that closely aligns with X-SHiELD metrics. Over a 10-year simulation, hurricane energies and extreme event snapshots remained comparable to baseline simulations. Operating globally costs approximately 300 times less than running X-SHiELD, with regional downscaling savings exceeding that factor by more than a hundred times depending on area size. A streamlined version in development aims to increase processing speeds tenfold. Current Limitations and Future Plans Because HiRO is trained exclusively on X-SHiELD, it is currently limited to ACE configurations tuned to match that model and cannot generate fine-resolution data for historical reanalysis like ERA5 or alternative future climate scenarios. Future work will expand training across datasets representing a 4°C warmer world, as well as incorporate variables such as near-surface temperature, snow, and ice to enable applications like snowpack modeling.

HiRO-ACEv2：以三百分之一成本模拟3公里分辨率极端天气 艾伦人工智能研究所（Ai2）发布HiRO-ACEv2气候仿真系统，通过将100公里分辨率粗尺度AI模型与全新3公里降尺度模型相结合，以约为传统高分辨率大气模型X-SHiELD三百分之一的成本，精确复现极端降雨、强风及气压事件。 HiRO-ACEv2结合了100公里分辨率模型ACE与负责降尺度的HiROv2，可同时预测降水、风速和气压等多个高分辨率要素。 在输入粗化数据时，HiROv2模拟高分辨率统计数据的误差在3%以内，即便对百万分之一发生概率的罕见极端事件依然有效。 全系统生成全球3公里气候数据的成本约为X-SHiELD的1/300，若仅针对局部区域计算，成本降幅还可增加上百倍。 当前模型局限于配合X-SHiELD微调的ACE版本运行，未来计划训练包括升温4°C情景在内的数据集并引入近地表气温等输出。 系统架构与极端天气模拟能力 HiRO-ACE系统包含两部分：在100公里分辨率下运行的ACE基础模型，以及将粗尺度输出细化至3公里分辨率的HiRO降尺度模型。HiROv2主要升级在于能够同时预测风、降水及气压等多个高分辨率要素，且在罕见极端事件模拟上提升显著。测试显示其统计结果与X-SHiELD的差距在3%以内，不仅能精准估算未参与训练的热带气旋能量，还能生成具有风眼和雨带的飓风形态以及山坡下坡强风，不过风眼周围的风场与气压分布较X-SHiELD稍显不对称。 长期模拟一致性与计算成本优势 在更新版ACE的基础上，结合HiROv2生成的一整年全球3公里数据在各项统计特征上与X-SHiELD高度吻合。在长达10年的连续模拟中，HiRO-ACEv2呈现的飓风总能量及各类极端事件特征均与目标模拟相当。成本方面，该系统生成全球3公里气候数据的开销仅为X-SHiELD的约1/300，若仅需特定区域数据，成本节约可再提升上百倍。团队还在研发精简版本，有望在现有基础上再提速十倍以上。 现有局限与后续计划 由于HiRO模型完全基于X-SHiELD进行训练，目前只能与针对该环境微调后的ACE协同工作，尚不能直接生成ERA5历史数据或不同气候情景的高分辨率数据。研发团队下一步计划接入模拟升温4°C等新气候情景的X-SHiELD数据集，并扩展近地表气温、冰雪等预测变量，以支持积雪预测等更多应用场景。相关模型权重、代码教程及用于对比检验的Web演示现已公开。

🤗 Models | 📄 Paper | 📓 Tutorial notebooks | 🌎 Demo We introduce HiRO-ACEv2 , the newest update to our High Resolution Output Ai2 Climate Emulator (HiRO-ACE) system. HiROv2 can reproduce even extremely rare rain, wind, and pressure events with high accuracy, while the full system can generate global 3 km climate data at about 1/300th the cost of X-SHiELD, a high-resolution global atmospheric model that explicitly simulates storms and other small-scale weather. HiRO-ACE, introduced earlier in 2026 , has two parts: ACE, an AI model that operates at a coarser 100 km resolution, and HiRO, a model that sharpens those coarse outputs into fine 3 km detail. The biggest change in HiRO-ACEv2 is the new downscaling model, HiROv2, which can now 1) predict several high-resolution fields at once instead of just rainfall, so it can capture a much wider range of extreme weather, and 2) has improved skill in predicting extreme events. Our main focus with HiRO-ACEv2 was to simulate extreme events well. Changes to how we train HiROv2 improved its ability to reproduce the most intense rain, wind, and pressure values. When given coarsened X-SHiELD data, it matches the target simulation's high-resolution statistics to within 3%, even for events so rare they occur only once in a million data points. It also accurately estimates the total energy of tropical cyclones, including storms it never saw during training, and generates hurricanes with realistic eyes and rain bands, though winds and pressure around the eye remain somewhat more lopsided than in X-SHiELD. Additionally, HiROv2 realistically generates strong winds flowing down mountain slopes. We also trained an updated version of ACE for this system. Over a 10-year simulation, its hurricanes, average climate patterns, and overall distributions of rain, wind, and pressure all line up well with X-SHiELD. That tells us ACE's coarse simulations are a solid foundation for producing long-term high-resolution climate data. Putting HiROv2 and ACE together, we generated a full year of global 3 km data. Its statistics closely match X-SHiELD's, even though HiRO was working from a simulation ACE ran on its own rather than from real X-SHiELD data. Across a 10-year run, the energy of HiRO-ACEv2's hurricanes is comparable to X-SHiELD's over the same period, and snapshots from that run show a variety of other extreme events that resemble those in X-SHiELD. An example of a strong hurricane in the North Atlantic, generated during the 10-year HiRO-ACEv2 simulation. Overall, HiRO-ACEv2 offers researchers a flexible and inexpensive way to generate many versions of high-resolution rain, wind, and pressure data for extreme events over 10 years or more. Producing global high-resolution data with it costs roughly 300 times less than running X-SHiELD. If only one region is needed, the savings can be more than a hundred times greater, depending on the size of the area. We're also working on a streamlined version of HiROv2 that could make it more than ten times faster still. We still have future work ahead of us: because HiRO learned from X-SHiELD, it only works with the version of ACE tuned to match X-SHiELD. That means it can't yet produce high-resolution data for other climates, such as a warmer future or the historical record in datasets like ERA5. Next, we plan to train on new X-SHiELD datasets, including one simulating a world 4°C warmer, so the system can handle different climates. We also plan to add outputs like near-surface temperature and snow and ice, which would allow for even more uses, such as snowpack prediction. Visualizing and using HiRO outputs Simulated 10 m wind speed over Baja California in the ACE data visualizer: ACE2 at 100 km resolution on the main map and HiRO-ACEv2 at 3 km in the inset. A simulated storm over the northeastern Pacific in the ACE data visualizer. The main map shows 10 m wind speed at 100 km resolution; the 3 km HiRO-ACEv2 inset shows finer structure within the storm. To make it easy to see what HiRO produces without running any inference, we built a web demo that shows pre-generated precipitation, wind, and sea-level pressure output from HiROv1 and HiROv2 side by side with X-SHiELD, the simulation they were trained on. The demo complements the quantitative comparisons in our tech report: it offers a quick visual check of how the outputs look, and lets users zoom in on their own region of interest to judge performance there. For more in-depth analysis, the models are available to download , and our tutorial notebooks walk through how to run them.