EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HyAQMet : Hourly Air Quality and Meteorology Dataset for Dhaka

Adity Barua, Mrinal Basak Shuvo, Sharika Ali, Asad Ullah Khan, Asif Mohammed Saad

Geoscience Data Journal · 2027-01 · 14 (1)

DOI: 10.1002/gdj3.70111 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Geoscience Data Journal, online erschienen am 05.11.2026. Verfasst von Adity Barua und 4 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a unified environmental dataset combining meteorological drivers and atmospheric composition variables from multiple independent sources for Dhaka, Bangladesh, spanning a 6‐year period (2017–2022) with 52,584 continuous hourly records across 42 spatiotemporally harmonised variables. Ground‐based air quality monitoring in this heavily polluted megacity is sparse, making multi‐source atmospheric reanalysis data highly practical. To address data fragmentation across disparate platforms in incompatible formats, this dataset integrates four independent products: ERA5‐Land (0.1°, hourly), ERA5 Single Levels (0.25°, hourly), Copernicus Atmosphere Monitoring Service (CAMS) EAC4 (0.75°, 3‐hourly) and NASA MERRA‐2 (0.5° 0.625°, hourly). The collection and processing workflow involved source‐appropriate spatial extraction centred on Dhaka, temporal alignment to Bangladesh Standard Time, linear interpolation to a uniform hourly grid, unit standardisation and solar radiation flux correction. To eliminate multi‐collinearity, variance inflation factor screening filtered 26 candidate MERRA‐2 variables down to 13 physically distinct predictors. Furthermore, seasonal labels based on the local 6‐season calendar and cyclical trigonometric time encodings are embedded to support predictive modelling. Technical validation demonstrates the internal consistency of the dataset through cross‐source meteorological comparisons, particulate matter scaling, and boundary layer dynamics co‐validation. This open‐access dataset lowers the entry barrier for multidisciplinary urban climate studies. Its potential value, use and reuse include training machine learning architectures for pollutant forecasting, establishing seasonal air quality climatology, characterising reanalysis product biases and assessing population exposure risks. Extraction footprints range from approximately 20 to 83 km across the four source products, so HyAQMet represents a regional background signal rather than intra‐urban spatial gradients and should not be used as a surrogate for point‐source or city‐block‐scale pollution dynamics. The complete dataset is freely available in CSV, Parquet, and JSON formats via Zenodo at https://doi.org/10.5281/zenodo.20705735 .

Zitieren

APA

Barua, A., Shuvo, M. B., Ali, S., Khan, A. U., & Saad, A. M. (2027). HyAQMet : Hourly Air Quality and Meteorology Dataset for Dhaka. Geoscience Data Journal, 14(1). https://doi.org/10.1002/gdj3.70111

Vancouver

Barua A, Shuvo MB, Ali S, Khan AU, Saad AM. HyAQMet : Hourly Air Quality and Meteorology Dataset for Dhaka. Geoscience Data Journal. 2027;14(1). doi:10.1002/gdj3.70111

BibTeX
@article{Barua2027,
  author = {Barua, Adity and Shuvo, Mrinal Basak and Ali, Sharika and Khan, Asad Ullah and Saad, Asif Mohammed},
  title = {HyAQMet : Hourly Air Quality and Meteorology Dataset for Dhaka},
  journal = {Geoscience Data Journal},
  year = {2027},
  volume = {14},
  number = {1},
  doi = {10.1002/gdj3.70111}
}
Kostenlos starten →