According to the monitoring data from the Embassy of the United States, there was on average 39 micrograms of PM2.5 particles per cubic meter to be found in the air in Beijing during 2023. The air quality has improved considerably since 2013.

Reasons for air pollution in Beijing

China’s capital city Beijing is one of the most populous cities in China with over 20 million inhabitants. Over the past 20 years, Beijing’s GDP has increased tenfold. With the significant growth of vehicles and energy consumption in the country, Beijing’s air quality is under great pressure from the economic development. In the past, the city had a high level of coal consumption. Especially in winter, in which coal consumption increased due to heating, the air quality could get extremely bad on the days without wind. In spring, the wind from the north would bring sand from Mongolian deserts, resulting in severe sandstorms in Beijing. The bad air quality also affected the air visibility and threatened people’s health. On days with very bad air quality, people wearing masks for protection can be seen on the streets in the city.

Methods to improve air quality in Beijing

Over the past years, the government has implemented various methods to improve the air quality in Northern China. Sandstorms, which were quite common 15 years ago, are now rarely seen in Beijing’s spring thanks to afforestation projects on China’s northern borders. The license-plate lottery system was introduced in Beijing to restrict the growth of private vehicles. Large trucks were not allowed to enter certain areas in Beijing. Above all, the coal consumption in Beijing has been restricted by shutting down industrial sites and improving heating systems. Beijing’s efforts to improve air quality has also been highly praised by the UN as a successful model for other cities. However, there is also criticism pointing out that the improvement of Beijing’s air quality is based on the sacrifice of surrounding provinces (including Hebei), as many factories were moved from Beijing to other regions. Besides air pollution, there are other environmental problems like water pollution that China is facing. The industrial transformation is the key to China’s environmental improvement.

China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei Region data was reported at 44.000 mcg/Cub m in May 2018. This records a decrease from the previous number of 52.000 mcg/Cub m for Apr 2018. China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei Region data is updated monthly, averaging 68.000 mcg/Cub m from Mar 2013 (Median) to May 2018, with 61 observations. The data reached an all-time high of 151.000 mcg/Cub m in Feb 2014 and a record low of 37.000 mcg/Cub m in Aug 2016. China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei Region data remains active status in CEIC and is reported by China National Environmental Monitoring Centre. The data is categorized under China Premium Database’s Environmental Protection – Table CN.EPJ: Air Quality: PM2.5 Concentration: Region.

📘 Overview

This dataset provides hourly air-quality measurements for 50 major global cities over a continuous 15-day period, including pollutant concentrations, meteorological conditions, geographical metadata, and an engineered AQI index.

All values are synthetically generated using historically consistent pollutant patterns and statistical ranges, allowing researchers and ML practitioners to work with realistic air-quality trends without licensing restrictions or data-collection barriers.

This dataset is ideal for time-series modeling, forecasting, environmental analytics, and machine-learning experimentation.

🧭 Cities Included

Covers all major regions:

North America — New York, Los Angeles, Toronto

Europe — London, Paris, Berlin, Zurich

Asia — Delhi, Tokyo, Seoul, Beijing, Singapore

Middle East — Dubai, Riyadh, Doha

Africa — Lagos, Cairo, Nairobi

Oceania — Sydney, Melbourne, Auckland

South America — São Paulo, Buenos Aires

🧱 Dataset Structure

Each hourly record includes:

Air Pollutants

PM2.5 (µg/m³)

PM10 (µg/m³)

NO₂ (ppb)

SO₂ (ppb)

O₃ (ppb)

CO (ppm)

Weather Features

Temperature (°C)

Humidity (%)

Wind Speed (m/s)

Location Metadata

City

Country

Latitude

Longitude

Other

Timestamp (ISO-8601)

AQI (Computed index)

🧹 Data Quality & Formatting

No missing values — 100% complete

Numeric values rounded to 3 decimals

Clean column names (snake_case)

Consistent hourly frequency

Fully ML-ready

📊 Example Use Cases

✔ AQI forecasting (LSTM, GRU, Transformers) ✔ Multivariate time-series modeling ✔ Clustering cities by pollution patterns ✔ Environmental trend visualization ✔ Weather–pollution correlation studies ✔ Anomaly detection (peak pollution events)

🧪 Data Generation Method (Provenance)

This dataset is synthetically generated using realistic pollutant behavior patterns based on historical studies and open-source environmental datasets.

Modeling steps included:

City-specific pollutant baseline ranges

Randomized variation using Gaussian noise

Temporal patterns using sinusoidal diurnal cycles (morning & evening peaks)

Weather-pollution correlation rules (e.g., low wind → higher PM)

AQI computed using standard US-EPA breakpoints

All numeric values standardized to 3-decimal precision

This ensures that although synthetic, the dataset follows realistic environmental dynamics.

📁 File Information

global_air_quality_50_cities.csv

Rows: 18,000+

Columns: 16

Format: UTF-8 CSV

China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei and Vicinity (55 City) data was reported at 36.000 mcg/Cub m in Sep 2018. This records an increase from the previous number of 35.000 mcg/Cub m for Aug 2018. China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei and Vicinity (55 City) data is updated monthly, averaging 38.000 mcg/Cub m from Jun 2018 (Median) to Sep 2018, with 4 observations. The data reached an all-time high of 42.000 mcg/Cub m in Jun 2018 and a record low of 35.000 mcg/Cub m in Aug 2018. China Air Quality: PM2.5 Concentration: Monthly Average: Beijing-Tianjin-Hebei and Vicinity (55 City) data remains active status in CEIC and is reported by China National Environmental Monitoring Centre. The data is categorized under China Premium Database’s Environmental Protection – Table CN.EPJ: Air Quality: PM2.5 Concentration: Region.

The global average concentration of particulate matter (PM2.5) was **** micrograms per cubic meter of air (μg/m3) in 2022. This not only represented a decline of ** percent from 2010 levels but was also the lowest concentration of this pollutant recorded since the turn of the century. Despite the significant decline, PM2.5 levels in 2022 were still **** times above WHO air quality guidelines. China cleans up its air China has been a key driver of the global decline in air pollution in recent years, with annual PM2.5 levels in the country falling by more than ** percent since 2014 – the year the Chinese government declared war on air pollution. The country’s capital city has seen considerable improvements in its air quality, with average PM2.5 levels in Beijing more than halving over the past decade. Improved air quality monitoring networks, low emissions zones, and the increased uptake of electric vehicles are just some of the ways the country has tackled air pollution. Air pollution's impact on health Governments around the world have been battling air pollution not only because of its environmental impacts, but also due to its adverse effects on human health. In fact, air pollution is the second-leading cause of death worldwide, with more than ***** million deaths attributed to this risk factor each year. Heart disease, strokes, and lung cancer are just some of the causes of premature deaths related to air pollution.

Dataset

This dataset was created by Giridhar Vasu

Released under CC0: Public Domain

Contents

Air Quality: PM2.5 Concentration: Annually Average: Beijing data was reported at 33.000 mcg/Cub m in 2021. This records a decrease from the previous number of 38.000 mcg/Cub m for 2020. Air Quality: PM2.5 Concentration: Annually Average: Beijing data is updated yearly, averaging 58.000 mcg/Cub m from Dec 2013 (Median) to 2021, with 9 observations. The data reached an all-time high of 89.500 mcg/Cub m in 2013 and a record low of 33.000 mcg/Cub m in 2021. Air Quality: PM2.5 Concentration: Annually Average: Beijing data remains active status in CEIC and is reported by Beijing Municipal Bureau of Statistics. The data is categorized under China Premium Database’s Environmental Protection – Table CN.EPK: Air Quality: PM2.5 Concentration: Prefecture Level City.

Air Quality Forecast: Contaminant Concentration: Ozone: China: Beijing data was reported at 43.467 mcg/Cub m in 22 May 2025. This records a decrease from the previous number of 47.887 mcg/Cub m for 21 May 2025. Air Quality Forecast: Contaminant Concentration: Ozone: China: Beijing data is updated daily, averaging 27.147 mcg/Cub m from Oct 2019 (Median) to 22 May 2025, with 2038 observations. The data reached an all-time high of 147.614 mcg/Cub m in 24 Jul 2022 and a record low of 0.285 mcg/Cub m in 09 Dec 2021. Air Quality Forecast: Contaminant Concentration: Ozone: China: Beijing data remains active status in CEIC and is reported by CEIC Data. The data is categorized under Global Database’s China – Table CAMS.AQF: Air Quality Forecast: Contaminant Concentration: Ozone: by Cities. [COVID-19-IMPACT]

Accurate forecasting of PM2.5 concentrations serves as a critical tool for mitigating air pollution. This study introduces a novel hybrid prediction model, termed MIC-CEEMDAN-CNN-BiGRU, for short-term forecasting of PM2.5 concentrations using a 24-hour historical data window. Utilizing the Maximal Information Coefficient (MIC) for feature selection, the model integrates Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Convolutional Neural Network (CNN), and Bidirectional Recurrent Gated Neural Network (BiGRU) to optimize predictive accuracy. We used 2016 PM2.5 monitoring data from Beijing, China as the empirical basis of this study and compared the model with several deep learning frameworks. RNN, LSTM, GRU, and other hybrid models based on GRU, respectively. The experimental results show that the prediction results of the hybrid model proposed in this question are more accurate than those of other models, and the R2 of the hybrid model proposed in this paper improves the R2 by nearly 5 percentage points compared with that of the single model; reduces the MAE by nearly 5 percentage points; and reduces the RMSE by nearly 11 percentage points. The results show that the hybrid prediction model proposed in this study is more accurate than other models in predicting PM2.5.

BackgroundPrevious studies have shown that air pollution affects the incidence of ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI) differently. However, limited studies have examined the impact of air pollution on the mortality of these acute myocardial infarction (AMI) subtypes.MethodsUsing AMI hospitalization data from Beijing (2013–2019), we applied a time-stratified case-crossover design with conditional Poisson regression models to evaluate associations between short-term exposure to six pollutants (PM2.5, PM10, SO2, NO2, CO, and O3) and daily in-hospital mortality for overall AMI, STEMI, and NSTEMI. Subgroup analyses based on demographics, comorbidities, and coronary artery disease (CAD) history were conducted to identify vulnerable populations. Additionally, a retrospective case–control analysis with multivariable logistic regression involved all AMI admission cases, was conducted to explore whether the association between air pollution exposure and in-hospital AMI mortality is independent of other mortality risk factors.ResultsDuring the study period, there were 149,632 AMI admissions, with 10,983 in-hospital deaths (4,361 STEMI and 4,299 NSTEMI). Elevated levels of PM2.5, PM10, SO2, NO2, and CO on admission day were significantly associated with increased in-hospital mortality for overall AMI and NSTEMI, but not for STEMI. The effect of pollutants on NSTEMI mortality was greater in patients with old myocardial infarction (OMI) or percutaneous coronary intervention/coronary artery bypass grafting (PCI/CABG) history. In case–control analysis with multivariable logistic regression, increased pollutants concentration remained significantly associated with in-hospital NSTEMI mortality after adjusting for other mortality risk factors.ConclusionShort-term exposure to PM2.5, PM10, SO2, NO2, and CO increases the risk of in-hospital AMI mortality, particularly for NSTEMI. Individuals with CAD history require more protective measures due to the vulnerability to air pollution.