Skip to content

Getting Started With riweather

import matplotlib.pyplot as plt

import riweather

%matplotlib inline

Suppose you need weather data at a certain site. To list all weather stations in order of their distance to that site, use riweather.rank_stations().

my_lat, my_lon = 39.98, -105.13
ranked_stations = riweather.rank_stations(my_lat, my_lon)

print("Number of stations returned:", ranked_stations.shape[0])
ranked_stations.head()

Number of stations returned: 2851
name distance latitude longitude years quality
usaf_id
720534 ERIE MUNICIPAL AIRPORT 7971.794695 40.017 -105.050 [2014, 2015, 2016, 2017, 2018, 2019, 2020, 202... [low, high, high, high, high, high, high, high...
724699 ATCT ROCKY MOUNTAIN METRO 8951.899139 39.900 -105.117 [2006, 2007, 2008, 2009, 2010, 2011, 2012, 201... [medium, medium, medium, medium, high, high, h...
720533 BOULDER MUNICIPAL AIRPORT 9477.070760 40.033 -105.217 [2010, 2011, 2012, 2013, 2014, 2015, 2016, 201... [medium, high, high, high, high, high, high, h...
720538 VANCE BRAND AIRPORT 21002.247034 40.167 -105.167 [2010, 2011, 2012, 2013, 2014, 2015, 2016, 201... [low, high, high, high, high, high, high, high...
725650 DENVER INTERNATIONAL AIRPORT 43134.270016 39.847 -104.656 [2005, 2006, 2007, 2008, 2009, 2010, 2011, 201... [high, high, high, high, high, high, high, hig...

This gives us the name and USAF identifiers of all ISD weather stations in the United States, along with their latitudes and longitudes and their distances (in meters) to the target site. The “years” column is a list of years for which there is available weather data from that station. The “quality” column contains the quality of the data file for the corresponding year—this is based on the number of observations in the file.

You probably don’t want all three thousand weather stations that are returned, so let’s be more efficient and limit our search to stations that have data for 2022 and stations that are within 20 km.

ranked_stations = riweather.rank_stations(
    my_lat,
    my_lon,
    year=2022,
    max_distance_m=20000,
)
ranked_stations
name distance latitude longitude years quality
usaf_id
720534 ERIE MUNICIPAL AIRPORT 7971.794695 40.017 -105.050 [2014, 2015, 2016, 2017, 2018, 2019, 2020, 202... [low, high, high, high, high, high, high, high...
724699 ATCT ROCKY MOUNTAIN METRO 8951.899139 39.900 -105.117 [2006, 2007, 2008, 2009, 2010, 2011, 2012, 201... [medium, medium, medium, medium, high, high, h...
720533 BOULDER MUNICIPAL AIRPORT 9477.070760 40.033 -105.217 [2010, 2011, 2012, 2013, 2014, 2015, 2016, 201... [medium, high, high, high, high, high, high, h...

Select the closest station, Erie Municipal Airport, using riweather.select_station().

station = riweather.select_station(ranked_stations, rank=0)
station

Station("720534")

Let’s see how complete we can expect the data to be when we pull it by looking at the quality report.

station.quality_report()
usaf_id wban_id year quality jan feb mar apr may jun jul aug sep oct nov dec count n_zero_months
0 720534 00161 2014 low 0 0 0 0 0 0 2191 2172 1869 2240 2160 2174 12806 6
1 720534 00161 2015 high 2232 1981 2226 2117 2244 2035 2232 2233 2084 2217 2151 2232 25984 0
2 720534 00161 2016 high 2240 2094 2239 2168 2255 2147 2203 2129 2173 2211 2164 2150 26173 0
3 720534 00161 2017 high 2143 1961 2213 2163 2061 2167 2204 2095 2172 2198 2888 2242 26507 0
4 720534 00161 2018 high 2219 2033 2272 2198 2523 2191 2292 2388 2125 2298 2217 2284 27040 0
5 720534 00161 2019 high 2279 2084 2255 2286 2432 2288 1780 1810 2088 2226 2169 2239 25936 0
6 720534 00161 2020 high 2056 2093 2234 2167 2232 2145 2096 2026 2161 2187 2160 2202 25759 0
7 720534 00161 2021 high 2227 2027 2245 2159 2232 2174 2168 2243 2098 2235 2145 2208 26161 0
8 720534 00161 2022 high 2237 2020 2152 2049 2232 2168 2250 2241 2160 2231 2169 2187 26096 0
9 720534 00161 2023 high 575 0 0 0 0 0 0 0 0 0 0 0 575 0

The data for 2022 is "high" quality, with more than 2,000 observations in each month. Assuming they are evenly spaced, that corresponds to one observation roughly every 20 minutes.

Now we can use the Station.fetch_temp_data() method to get temperature data in degrees Fahrenheit.

ts = station.fetch_temp_data(2022, value="temperature", scale="F")
ts.head()

2022-01-01 00:00:00+00:00    24.2620
2022-01-01 01:00:00+00:00    20.1995
2022-01-01 02:00:00+00:00    17.9525
2022-01-01 03:00:00+00:00    16.7690
2022-01-01 04:00:00+00:00    16.3715
Freq: H, Name: tempF, dtype: float64

It’s important to note that the timestamps are always in UTC. The temperature data is hourly by default, but riweather can resample it to a different frequency as well.

ts_30min = station.fetch_temp_data(
    2022,
    value="temperature",
    scale="F",
    period="30min",
)
ts_30min.head()

2022-01-01 00:00:00+00:00    26.0780
2022-01-01 00:30:00+00:00    23.3540
2022-01-01 01:00:00+00:00    21.0470
2022-01-01 01:30:00+00:00    19.3520
2022-01-01 02:00:00+00:00    18.2975
Freq: 30T, Name: tempF, dtype: float64

ts_30min.plot(
    title=f"Air temperature at {station.name.title()}, 2022",
    ylabel="Air temperature (degrees F)",
)
plt.show()

We can verify the location of our site and the nearest weather stations by plotting them on a map with riweather.plot_stations().

riweather.plot_stations(my_lat, my_lon, ranked_stations, distance_unit="mi")
Make this Notebook Trusted to load map: File -> Trust Notebook

Only have a ZIP code for your site and not a precise latitude and longitude? No problem. You can use riweather.zcta_to_lat_lon() to get an approximate location.

my_lat, my_lon = riweather.zcta_to_lat_lon("80027")
print(f"{my_lat:.3f}", f"{my_lon:.3f}")

39.951 -105.164


ranked_stations = riweather.rank_stations(my_lat, my_lon, max_distance_m=20000)
riweather.plot_stations(my_lat, my_lon, ranked_stations, distance_unit="mi")
Make this Notebook Trusted to load map: File -> Trust Notebook