Air pollution datasets

Processed air pollution datasets originally obtained from the open data portal of the Madrid City Hall. The pollutants include:

Fine particulate matter: PM2.5

Coarse particulate matter: PM10

Ozone: O3

Nitrogen monoxide: NO

Nitrogen dioxide: NO2

Nitrogen oxides: NOx

Sulfur dioxide: SO2

Carbon monoxide: CO

Toluene: TOL

Benzene: BEN

Ethylbenzene: EBE

The period covered goes from the 1st of January 2010 to the 30th of April 2022. Each pollutant is recorded by a variable number of sensors, between 6 and 24 of them (additional information here). They cover Madrid city and surroundings (see this map). Additional information about the pollutants can be found here.

The specific datasets in HDF5 format are:

01h_flat_raw.h5: Hourly raw data, one table per pollutant. The first column is the timestamp, which are in time-since-epoch. 108049 rows and between 7 and 25 columns.

01h_35x30_norm_linear_J0.0.h5: Hourly mesh-grid data, normalized and linearly interpolated. Single table of shape: (108049, 35, 30, 11).

01h_35x30_norm_nearest_J0.0.h5: Hourly mesh-grid data, normalized and nearest-neighbors interpolated. Single table of shape: (108049, 35, 30, 11).

01h_35x30_raw_linear_J0.0.h5: Hourly mesh-grid data, linearly interpolated. Single table of shape: (108049, 35, 30, 11).

01h_35x30_raw_nearest_J0.0.h5: Hourly mesh-grid data, nearest-neighbors interpolated. Single table of shape: (108049, 35, 30, 11).

01h_35x30_stand_linear_J0.0.h5: Hourly mesh-grid data, standardized and linearly interpolated. Single table of shape: (108049, 35, 30, 11).

01h_35x30_stand_nearest_J0.0.h5: Hourly mesh-grid data, standardized and nearest-neighbors interpolated. Single table of shape: (108049, 35, 30, 11).

This datasets are prepared to work with the framework published at https://github.com/iipr/air-quality