NCDC Global Summary of the Day Dataset


The US National Climatic Data Center has been collecting weather data at stations around the globe since 1929. In particular, the Global Summary of the Day contains samples of surface weather data like rainfall, temperature, and so on.


The data stored in this table is denormalized from the original data, so that each row contains complete information about the sample, rather than requiring a join to on station_number to get State, latitude, longitude, etc. The source data was imported in May, 2010 from http://www.ncdc.noaa.gov/oa/ncdc.html. For a complete interpretation of these fields please see that site, in particular the GSOD description and the Station History.

Field name Type Description
station_number integer
wban_number integer
year integer
month integer
day integer
mean_temp float Farenheit
num_mean_temp_samples integer Farenheit
mean_dew_point float millibars
num_mean_dew_point_samples integer
mean_sealevel_pressure float millibars
num_mean_sealevel_pressure_samples integer
mean_station_pressure float millibars
num_mean_station_pressure_samples integer
mean_visibility float miles
num_mean_visibility_samples integer
mean_wind_speed float knots
num_mean_wind_speed_samples integer
max_sustained_wind_speed float knots
max_gust_wind_speed float knots
max_temperature float Fahrenheit
max_temperature_explicit boolean
min_temperature float Fahrenheit
min_temperature_explicit boolean
total_precipitation float inches
snow_depth float inches
fog boolean false means no fog OR unknown
rain boolean false means no rain OR unknown
snow boolean false means no snow OR unknown
hail boolean false means no hail OR unknown
thunder boolean false means no thunder OR unknown
tornado boolean false means no tornado OR unknown