Latitude of the center of the grid cell that the air quality prediction is for.

Longitude of the center of the grid cell that the air quality prediction is for.

Air quality parameter being forecast. Parameters include CO (carbon monoxide), NO2 (nitrogen dioxide) and PM25_RH35_GCC (PM2.5 - tiny inhalable particles (with diameters of 2.5 micrometers or less), which are unhealthy to breathe in).

The concentration of the air quality parameter.

The units that the air quality parameter are in. (Note that these units do not match the units used in the Allegheny County Air Quality dataset, so make sure to do any necessary conversions when making cross-dataset comparisons.)

Air Quality Index: A standard scale from 0 to 500, that makes it easier to understand how harmful a given concentration of a pollutant would be. An AQI of 0 to 50 is considered Good, 51 to 100 is Moderate, 101 to 150 is Unhealthy for Sensitive Groups, 151 to 200 us Unhealthy, 201 to 300 is Very Unhealthy, and > 300 is Hazardous. For some parameters, there will be null/empty values for some times. For instance, an AQI that requires a 24-hour average will not be calculated until Day 2 of the forecasts, so that there are 24 records over which to average.

Translates the Air Quality Index value into the official term for the level of concern (e.g., 151 to 200 is "Unhealthy").

Specifies how the AQI is calculated (as different pollutants can use rolling averages over different time spans).

A human-readable name for a place within the geographic area of the prediction. This is kind of a landmark to help users find places of interest in the table.

A string representation of the latitude in the latitude column (which has a numerical type).

A string representation of the longitude in the longitude column (which has a numerical type).

This is the date that the prediction was made (which should not be confused with the date that the prediction is for). For instance, on the first day of the year 2030, this field would have the value 2030-01-01 for all records, while the datetime_utc value will different values, indicating the different days/times that are being predicted.