环境空气质量预测数据（城市）

天融环境大模型，基于全国高密度大气污染监测网络、ERA5大气再分析数据及多源异构数据资源，依托华为盘古科学计算大模型技术底座，预测输出未来7日城市环境空气质量状况。
实现预测周期由3日延长至7日；时空分辨率提升至3公里×3公里；PM2.5/PM10预测准确率提升幅度超过10个百分点；推理时长由10小时缩减至25分钟，可广泛应用于环境管理与执法、城市规划与建设、工业与交通、公众服务等领域。
时间范围及分辨率：未来7日城市环境空气质量小时数据
空间范围及分辨率：全国超300个地级及以上城市
指标参数：PM2.5、PM10、SO2、NO2、CO、O3浓度、空气质量指数（AQI）、空气质量等级、首要污染物等
数据提供方式：可直接提供数据集，也支持通过接口获取，也可存储为库表形式，满足多样化应用需求。

The Tianrong Environmental Large Model is built upon the national high-density air pollution monitoring network, ERA5 atmospheric reanalysis dataset, and multi-source heterogeneous data resources, and leverages Huawei Pangu Scientific Computing Large Model as its technical backbone to predict and output urban ambient air quality conditions for the next 7 days.
It has achieved the following performance optimizations: extended the prediction cycle from 3 days to 7 days; improved the spatiotemporal resolution to 3 km × 3 km; increased the prediction accuracy of PM2.5/PM10 by more than 10 percentage points; reduced the inference time from 10 hours to 25 minutes. This model can be widely applied to multiple fields including environmental management and law enforcement, urban planning and construction, industry and transportation, and public services.
Temporal scope and resolution: Hourly urban ambient air quality data for the next 7 days.
Spatial scope and resolution: Over 300 prefecture-level and above cities across China.
Indicator parameters: Concentrations of PM2.5, PM10, SO2, NO2, CO, O3, Air Quality Index (AQI), air quality grade, primary pollutants, and other relevant indicators.
Data provision approaches: The dataset can be directly provided, obtained via application programming interfaces (APIs), or stored in the form of database tables to meet diverse application requirements.