From OpenStreetMap Wiki
Jump to navigation Jump to search

Tagging: Meteorologic Station

Meteorological monitoring networks are often operated by government agencies, but there is also an increasing number of stations operated by private agencies or even by hobby-meteorologists. The focus here lies on government monitoring networks.

Locating stations

Government stations can often be found near government facilities such as schools, government and infrastructure facilities, airports etc. WMO recommends to conduct temperature measurements above cutted grass with no close significant thermal radiation sources (sealed ground such as concrete or asphalt, glass facades, heating facilities, cars and trucks) nearby. Often stations can be found at open spots with grass surface a bit away from other facilities. Often stations have a fence around them, which is visible on the aerial imagery.

It is sufficient to mark the station as point matching the Stevenson screen, which is visible as big white spot on the aerial image.

Mandatory Tags

These tags are essential to define this station as monitoring the weather.

Important Tags

If known, these tags should definitely being added as they add valuable information.

  • name=* - stations normally have the name of the city/town or facility (airport) where they are located, this name should be always in the country's language and alphabet (e.g. Палић for Serbian station 13067)
  • int_name=* - provide a transliteration/romanization for name=* if necessary (e.g. Palić for Палић), in addition add a proper name:lang_translit tag to specify the transliteration scheme used (see also Multilingual_names)
  • ref:wmo=* - WMO associates a unique id to each registered station (e.g. 11035). To each country a range of ids is allocated. This id makes it possible to map transmitted weather data to a particular station.
  • ref=* - if no ref:wmo is available, other identifiers used to associate data with the station.
  • ref:icao=* - ICAO associates a unique identifier to each airport (e.g. LOWW). The first two characters identify the region and often also the country. If the station is located at an airport this identifier should be added since METARs contain this identifier and this allows to map them to a particular station.
  • ele=* - elevation information helps in performing reducing pressure values to mean sea level for comparability (see Mean sea level pressure for details).
  • source=* - data sources differ in quality, especially in precision of coordinates, mentioning the source helps to judge the quality of the information. Source could be e.g.,, etc.

Note on callsigns: Some agencies use callsigns for their non-airport stations that mimic the ICAO codes. These should NOT be tagged with ref:icao=*. Instead another key ref:*=* should be used, if these callsigns need to be added. Often this is not necessary as they have only local importance.

Note on elevation information: Often station listings mention two different types of elevation information (WMO calls it Hp - elevation of the station - and Hha - ground elevation or altitude of aerodrome). Often the barometer is located within the Stevenson screen so the two values deviate only by few meters (as temperature is measured 2m above ground). Sometimes the barometer is somewhere else e.g. on the roof of a building, so there could be more deviation. For ele=* always the ground elevation (or Hha) should be used as this corresponds to the normal usage on OSM. However, for Mean sea level pressure the station elevation (or Hp) is required, so for this value the key ele:barometer=* could be used.

Note on sources: Information may come from different sources. For example, you may find the location of a station's Stevenson's Screen on Bing, get name, and ref:wmo from WMO, measured elevation with your GPS-device and got information about monitored variables from the national agency. While it should not get too complicated with source tags it may be worthwhile to identify the main sources of information. For this example source:position=Bing,, source:ele=survey, could be used. If you have one main source of information (e.g. national agency) just use source=* and add important source:*=* (e.g. for position) accordingly.

Helpful Tags

  • operator=* - mentioning the operating agency helps in classifying stations, abbreviations should be used
  • fixme=* - indicating necessary improvements, if data is missing or if position is not precise etc. (e.g. fixme=position if position is not exact).

Normally Unnecessary Tags

Generally, only as much information as is necessary should be added.

  • is_in=* - boundary information is nowadays well developed in OSM and so the software is able to check in which administrative regions a node lies. Some stations are not located within national borders (e.g. at sea platforms) and they should not be tagged with country names. Instead, the key operator=* can be used to specify which agency operates this station.
  • local or agency-specific identifiers (unless no ref:wmo is available) - the ref:wmo=* is globally unique and should be sufficient for stations that are participating in global data exchange. Sometimes the wmo-id changes, data centers may need to keep track of these historical changes for data harmonization purposes. However, since OSM does not contain historic information this has to be kept elsewhere (and of course there are also OSM history files).


Weather monitoring stations normally monitor more than one parameter at once. There are radar stations, dedicated stations for monitoring precipitation (often operated by hydrological agencies and often measuring also temperature for diffentiating between rain and snow), radiosonde locations and normal weather stations (which often measure temperature, humidity, pressure, precipitation, wind and sun duration). Standard monitored parameters may differ between countries. An overview over instruments of a common U.S. airport station can be found at wikipedia.

Atmospheric Pressure

Cloud Base

  • weather:ceilometer=yes/no - a ceilometer is used to monitor the height of a cloud base (it is also possible to monitor aerosol concentration). This instrument is often used at airports.


  • weather:humidity_sensor=yes/no - to measure relative humidity, often capacitive or resistive sensors that are cheaper than psychrometers (see below) are used in private weather stations. Psychrometers are often used in official monitoring stations.
  • weather:psychrometer=yes/no - a psychrometer is used to monitor dewpoint temperature and compute relative humidity, mostly located within the Stevenson screen; most sophisticated automated solutions use chilled mirror dewpoint psychrometers, cheaper humidity sensors include capacitive or resistive sensors (see above).


  • weather:lightning_detector=yes/no - a lightning detector monitors lightning activity by observing the arrival of electro-magnetic waves produced by lightning at the detectors and calculating runtime differences to compute locations. Lightning detectors are often separate from normal weather stations and are also often operated by different agencies (e.g. ALDIS in Austria is a joint venture of the Austrian power transmission grid provider and the Austrian electrotechnical association).


  • weather:ombrometer=yes/no - an ombrometer monitors liquid precipitation and has the form of a bucket. There exist also dedicated ombrometers often operated by hydrological agencies that measure temperature in addition to distinguish between snow and rain.
  • weather:precipitation_indicator=yes/no - stations often use infrared beams to determine if there is currently falling precipitation. Some instruments may also be able to distinguish between types of preciptiation. Precipitation indicators can be necessary in cases where precipitation amount is too low to be recognized by ombrometers.
  • weather:radar=yes/no - weather radar is used to monitor the areal and vertical distribution as well as the type (rain, drizzle, snow, hail) of precipitation.
  • weather:snow_height_sensor=yes/no - automated snow height sensors operate with ultrasound.

Solar/Direct/Atmospheric radation (Sun Duration)

  • weather:pyranometer=yes/no - a pyranometer is used to monitor solar irradiance. This instrument is often mounted on a pole or tower. Sunhine duration can be computed from irradiance values (exceeding 120 W/m2).
  • weather:pyrgeometer=yes/no - a pyrgeometer is used to monitor atmospheric infra-red radiation. It measures downward long wave radiation and so a radiation balance between surface and atmosphere can be computed.
  • weather:pyrheliometer=yes/no - a pyrheliometer is used to monitor direct beam solar irradiance. This instrument is often mounted on a pole or tower. Using this instrument for monitoring sunshine duration is more exact than computing it from pyranometer values (see above).

Air/Ground/Soil Temperature

Vertical Layering of Atmosphere

  • weather:radiosonde=yes/no - radiosondes are basically weather balloons to monitor the vertical layering of the atmosphere. Radiosonde locations are often at airports as the vertical structure there is important for landing or starting aircraft. Radiosondes monitor different variables at once (such as temperature, pressure, wind speed/direction). Often they are tracked through GPS or radar. Radar for tracking radiosondes may look like precipitation radar on aerial imagery (like the radiosonde radar at Wien-Hohe Warte).



  • weather:anemometer=yes/no - to monitor wind speed and wind direction (not necessary to differentiate between wind speed and wind direction as the professional standard are now sonic anemometers which monitor both variables by design). Often located on a pole or tower as wind is monitored 10 m above ground.

Note on instrument names: There are two ways of indicating the equipment of a weather station: naming the instruments or naming the variables that are monitored. Both ways have drawbacks as there are different ways of measuring a variable, but this leads to more choices and possible inconsistency. However, naming instruments has some advantages. First, with naming variables (e.g. precipitation) there is no way to distinguish between precipitation indicators, ombrometers and also precipiation radars. Second, naming instruments captures also quality differences as psychrometers are more exact than capacitive or resistive humidity sensors.

Station Data


Weather monitoring stations with set ref:wmo=* in OSM compared to WMO Volume A and percentage of stations tagged with fixme=*. Only countries where the station number is greater than 0 are listed.

Stats have moved here.

Data Sources

Country Code Country Name Name of national organisation Station page Station file License Notes
AUT Austria ZAMG Link Link CC-BY AT permission by mail from MK
BGR Bulgaria NIMH Link unknown only map image without additional information
CHE Switzerland MeteoSwiss Link Link unknown web map and list without wmo ids, KML file SwissMetNet.kml with detailed but HTML-encapsulated data
CHZ Czech Republic CHMI Link unknown interactive map without wmo id
FIN Finland FMI Link unknown web list with wmo id
HRV Croatia DHMZ main: Link automatic: Link main: Link automatic: Link unknown only google map, but also xml that contains station information, no wmo id
ROU Romania INMH unknown no station inventory found
SVK Slovakia SHMU Link Link unknown map of stations and PDF with wmo id, coordinates and elevation
SVN Slovenia ARSO Link unknown interactive atlas showing locations of meteorological stations (see Monitoring Points > Meteorological stations), no export, no wmo-ids, relevant stations may be categorized as glavna meteorološka postaja
TUR Turkey DMI Link unknown web list by provinces with wmo-ids and (sometimes fake) icao-ids
Source Name Station page Station file License Notes
WMO Pub Nr. 9 Volume A Link Link unknown
OurAirports Airport Identifiers Link Link Public Domain does not include WMO information


Effective queries can be made with the help of overpass turbo, which uses the Overpass API. Data can be exported to various formats (GeoJSON, GPX, KML, etc.).

Below you find a sample query that returns all weather monitoring stations within a given area (e.g. Slovenia). To change the area name change the text after nominatimArea:.

<osm-script output="xml" timeout="25">
  <id-query {{nominatimArea:Slovenia}} into="area"/>
    <query type="node">
      <has-kv k="man_made" v="monitoring_station"/>
      <has-kv k="monitoring:weather" v="yes"/>
      <area-query from="area"/>
  <print mode="meta"/>
  <recurse type="down"/>
  <print mode="meta" order="quadtile"/>

Another example returns all weather monitoring stations in Turkey that have a ref:wmo. To change to stations that have NO ref:wmo add a modv="not" before regv=".".

<osm-script output="xml" timeout="25">
  <id-query {{nominatimArea:Turkey}} into="area"/>
    <query type="node">
      <has-kv k="man_made" v="monitoring_station"/>
      <has-kv k="monitoring:weather" v="yes"/>
      <has-kv k="ref:wmo" regv="."/>
      <area-query from="area"/>
  <print mode="meta"/>
  <recurse type="down"/>
  <print mode="meta" order="quadtile"/>

Due to performance reasons it is better to use the command line to query for larger datasets. Here wget or curl can be used.

wget -O output.osm --post-file=query.oapi ""

curl --output output.osm --data @query.oapi ""

Both examples should yield the same results. The output is written to output.osm and the file query.oapi contains an Overpass API Query.

Return a list of worldwide station - country relations in the way that the station id is followed by the country id and so on.

out ids;

The script first queries for all weather monitoring stations that have a ref:wmo and saves it to the set stats. Next, for each element of stats, it is determined in which areas this element is contained and the element of the country boundary is selected. This result (ids only) is printed out together with the input node from stats.

Keeping track of changes

The following script returns all weather monitoring stations that have been changed since the specified timestamp. The results could then be loaded into JOSM for review.

out meta;


Licensing status of station data is currently unknown. Inquiry at WMO has been made.

Important here is WMO Resolution 40, which defines essential and additional data and products. Essential data has to be made freely available by WMO members. I think station listings fall into the category of essential data, but this has to be confirmed.