Ai2 发布 HiRO-ACEv2，以约三百分之一成本生成全球 3 公里气候数据

🤗 模型 | 📄 论文 | 📓 教程笔记本 | 🌎 演示

🤗 Models | 📄 Paper | 📓 Tutorial notebooks | 🌎 Demo

我们推出 HiRO-ACEv2，这是我们高分辨率输出 Ai2 气候模拟器（HiRO-ACE）系统的最新更新。HiROv2 能够高精度地重现极其罕见的降雨、风和气压事件，而整个系统可以生成全球 3 公里气候数据，成本约为 X-SHiELD 的 1/300。X-SHiELD 是一个高分辨率全球大气模型，能够显式模拟风暴和其他小尺度天气。

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.

于 2026 年早些时候推出的 HiRO-ACE 包含两部分：ACE，一个以较粗的 100 公里分辨率运行的 AI 模型；以及 HiRO，一个将这些粗分辨率输出锐化为精细 3 公里细节的模型。HiRO-ACEv2 最大的变化是新的降尺度模型 HiROv2，它现在可以 1）同时预测多个高分辨率场，而不仅仅是降雨，因此能够捕捉范围更广的极端天气；2）在预测极端事件方面的能力有所提升。

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.

HiRO-ACEv2 的主要关注点是很好地模拟极端事件。我们对 HiROv2 训练方式的改动提升了它重现最强降雨、风和气压值的能力。当输入粗化的 X-SHiELD 数据时，它与目标模拟的高分辨率统计量匹配误差在 3% 以内，即使对于百万个数据点中才出现一次的极罕见事件也是如此。它还能准确估算热带气旋的总能量，包括训练中从未见过的风暴，并生成具有逼真风眼和雨带飓风，不过风眼周围的风和气压仍比 X-SHiELD 中略显不对称。此外，HiROv2 能逼真地生成沿山坡下泄的强风。

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.

我们还为该系统的 ACE 训练了一个更新版本。在为期 10 年的模拟中，其飓风、平均气候模式以及降雨、风和气压的整体分布都与 X-SHiELD 高度吻合。这表明 ACE 的粗分辨率模拟是生成长期高分辨率气候数据的坚实基础。

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.

将 HiROv2 和 ACE 结合，我们生成了整整一年的全球 3 公里数据。其统计量与 X-SHiELD 密切匹配，尽管 HiRO 是基于 ACE 自行运行的模拟而非真实的 X-SHiELD 数据工作的。在 10 年运行中，HiRO-ACEv2 的飓风能量与同期 X-SHiELD 相当，该运行的快照还显示出各种其他极端事件，与 X-SHiELD 中的事件相似。

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.

北大西洋一个强飓风的示例，生成于 10 年 HiRO-ACEv2 模拟期间。

An example of a strong hurricane in the North Atlantic, generated during the 10-year HiRO-ACEv2 simulation.

总体而言，HiRO-ACEv2 为研究人员提供了一种灵活且低成本的方式，可为 10 年或更长时间的极端事件生成多个版本的高分辨率降雨、风和气压数据。用它生成全球高分辨率数据的成本大约比运行 X-SHiELD 低 300 倍。如果只需要一个区域，节省的成本可超过一百倍，具体取决于区域大小。我们还在开发一个精简版 HiROv2，有望使其速度再提升十倍以上。

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.

我们未来还有工作要做：因为 HiRO 是从 X-SHiELD 学习的，所以它只能与为匹配 X-SHiELD 而调优的 ACE 版本配合使用。这意味着它目前还无法为其他气候生成高分辨率数据，例如更温暖的未来气候，或 ERA5 等数据集中的历史记录。接下来，我们计划在新的 X-SHiELD 数据集上训练，其中包括一个模拟全球升温 4°C 的数据集，以便该系统能够处理不同的气候。我们还计划增加近地表温度以及雪和冰等输出，这将带来更多用途，例如积雪预测。

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.

可视化并使用 HiRO 输出

Visualizing and using HiRO outputs

ACE 数据可视化工具中下加利福尼亚上空模拟的 10 米风速：主地图上为 100 公里分辨率的 ACE2，插图中为 3 公里分辨率的 HiRO-ACEv2。

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.

ACE 数据可视化工具中东北太平洋上空模拟的风暴。主地图显示 100 公里分辨率的 10 米风速；3 公里分辨率的 HiRO-ACEv2 插图显示了风暴内部更精细的结构。

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.

为了让用户无需运行任何推理就能轻松查看 HiRO 生成的结果，我们构建了一个网页演示，将 HiROv1 和 HiROv2 预生成的降水、风和海平面气压输出与它们训练所基于的模拟 X-SHiELD 并排展示。该演示补充了我们技术报告中的定量比较：它提供了对输出外观的快速视觉检查，并让用户放大到自己感兴趣的区域来判断那里的表现。

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.

如需更深入的分析，模型可供下载，我们的教程 notebook 会逐步介绍如何运行它们。

For more in-depth analysis, the models are available to download, and our tutorial notebooks walk through how to run them.

注意转载授权 · 阅读原文。中文正文为 AI 翻译，请以原文为准。

注意转载授权 · 阅读原文