Spatiotemporal prediction of Nitrogen Dioxide in Madrid city using ConvLSTM

These datasets are part of the following work (Iskandaryan, D., Ramos, F., & Trilles, S. (2022). Comparison of nitrogen dioxide predictions during a pandemic and non-pandemic scenario in the city of Madrid using a convolutional LSTM network. International Journal of Computational Intelligence and Applications, 21(02), 2250014). The datasets contain hourly nitrogen dioxide and meteorological data (ultraviolet radiation, wind speed, wind direction, temperature, relative humidity, barometric pressure, solar irradiance, and precipitation) of Madrid from January to June of  2019 and 2020,  and latitude and longitude of control stations. The data were obtained from the Open Data portal of the Madrid City Council. There are 24 air quality control stations and 26 meteorologic control stations. Using ArcGIS Pro software grid was created (Top -4,486,449.725263 m;  Bottom - 4,466,449.725263 m; Left - 434,215.234430 m; Right - 451,215.234430 m). The value of each cell includes the values of NO2 and meteorological attributes obtained from assigned stations at a certain time. The value of the cell with no stations was assigned as zero. Generated grid was exported as Comma Separated Values (CSV) files. Overall, 4344 and 4368 CSV files were generated every hour during the first six months of 2019 and 2020, respectively.