Meteorology
Meteorological vocabulary, weather concepts and how some of the parameters relate to each other.
Meteorological vocabulary
Time series
In our case, a time series is a series of measurements listed in time order. Usually there is a regular time period between each measurement.
Station
A weather station is a platform or a place, on land or sea, with instruments to measure weather parameters. A MET Norway station often consists of one weather station unit containing several instruments, or several separate instruments installed in close proximity to each other. A station can be situated on land, on a ship, on a buoy or on an airplane.
Metadata
Information about data, but not the content of the data itself. The location of a meteorological station and the quality of a measurement are both examples of metadata found in Frost. The data itself are the weather observation values, such as the measured air temperature or the wind speed. You can also read the Wikipedia definition of metadata.
Meteorological concepts
What are parameterids, stationids, stationnames etc...
Some request parameters rely on having or getting internal MET Norway information. Most of our weather observations are supplied by stationary (automatic) weather stations installed and maintained by MET Norway. The stations usually, but not always, contain multiple instruments to measure weather parameters such as air temperature, wind speed and direction, humidity, pressure etc. There may be more than one sensor measuring each weather parameter and they may be installed at different heights (see chapter on levels) and/or send the observation data to our central collection system at different times of the day (see chapters on timeresolutions and timeoffset).
Each station (see definition of a station in the Vocabulary) is registered in our internal system with a station name (stationnames), a short version of this name (stationshortnames), an internal ID number (stationids) and ID number(s) as defined by the World Meteorological Organization (WMO) (stationalternateids). Each weather parameter is also registered in our system with a name (elementid), an ID number (parameterid) and a definition.
To find station IDs and other information about a station/weather parameter, you can make a request containing only time and for example a location using a place name you know (set incobs to false to receive only information and not actual observation data). You will get information, including station names and IDs, plus elementids and parameterids, of matching stations. If a location name does not get you what you want, you can try to use a geographical search using for example nearest or polygon.
Standard height of measurements
Historically, some meteorological measurements should be done at a specific height above the ground as a world-wide standard set by the World Meteorological Organization (WMO). For example: air temperature is measured at 2m and wind is measured at 10m above the ground. These heights are called default or standard levels and you will find that most of the measurements in our archive have been done at a standard level. In the past, the height of the measurements was not always stored along with the measurements because it was assumed to be at the standard level.
However, sometimes it is not physically or economically possible to install an instrument at the standard height above the ground. There are also other reasons to measure at non-standard heights, for example if the measurements are done to explore a local weather phenomenon and/or they are part of a special project.
In the frost API response you will find the level of the time series in the id section of the header. The level is the height of the measurement in meters above ground (or above the water level, if you are looking at observations from e.g. a buoy).
Time offset
The time offset is the observation time relative to midnight. Some weather elements are historically measured and aggregated at specific times during the day. For example, standard sampling time for snow observations is at 06 UTC. The corresponding time offset is given as PT6H. The mean daily value of a weather element with the PT6H offset is calculated between 06 UTC on one day and 06 UTC on the next day. You will find this information in the API response.
Some examples of typical time offsets:| element_id | timeResolution | Preferred offset | Why |
|---|---|---|---|
| max(air_temperature P1D) | P1D | PT18H | Standard sampling period for temperature extremes is 19-18 UTC |
| max(air_temperature PT12H) | PT12H | PT6H | This offset indicates our standard timeseries, where the first calculation is performed at 06 UTC |
| min(air_temperature P1D) | P1D | PT18H | Standard sampling period for temperature extremes is 19-18 UTC |
| min(air_temperature PT12H) | PT12H | PT6H | This offset indicates our standard timeseries, where the first calculation is performed at 06 UTC |
| snow_coverage_type | P1D | PT6H | Standard sampling time for snow observations is 06 UTC |
| sum(precipitation_amount P1D) | P1D | PT6H | Standard sampling period for precipitation is 07-06 UTC |
| sum(precipitation_amount P1M) | P1M | PT6H | Standard sampling period for precipitation is 07-06 UTC |
| surface_snow_thickness | P1D | PT6H | Standard sampling time for snow observations is 06 UTC |
| Other elements | Other resolutions | PT0H | For cases not listed above, PT0H is always the preferred offset when multiple offsets are returned. |
Relationship between time, timeoffset, and timeresolution
timeresolution is the period between each data value, i.e. how often a weather phenomenon is measured. "timeresolution": "P1D" means one data value once a day.
timeoffset is observation time relative to midnight. "timeoffset": "PT6H" means that the daily data value has observation time 06:00 UTC. If “timeresolution”: “PT12H”, “timeoffset”: “PT6H”, it means that the first data value has observation time 06:00 UTC and the second at 18:00 UTC.
Example 1: Observations with timeresolution shorter than 1 dayObservations with shorter timeresolution than 1 day, are generally stored with date and time in the data storage. The time is then equal to the observation time, and no timeoffset should be added to time to get the original observation time.
Example 2: Observations with timeresolution 1 day or longer
Observations with daily or longer time resolution can be stored without a time in the data storage. If a time is found in the data storage, then API will return this time, if not it will return the first midnight on the given date.
NOTE: If time is different from midnight, time is equal to observation time, and timeoffset should not be added.
Example 2.1: Instantaneous observations
An instantaneous observation represents a time point, not a period. E.g. an instantaneous temperature observation vs. an average temperature over a longer period of time. Instantaneous observations with timeresolution ≥ 1 day are stored with date and time in the data storage, unlike other observations with timeresolution ≥ 1 day. The time is equal to the observation time, and timeoffset should not be added to time.
Example 2.2: Diurnal (daily) data
Diurnal data represent a 24 hour period on a specific date. They are stored with date only in the data storage with a time of the form
If timeoffset = PT0H, the observation value represents the period from midnight to midnight on the calendar date, i.e. from
If timeoffset != PT0H, the observation represents the last 24 hr, at the time after midnight equal to timeoffset. E.g. elementId = sum(precipitation_amount P1D) has timeoffset PT6H. This means that the observation period lasts continuously from the previous date at 06 UTC to current date 06 UTC. time
Example 2.3: Monthly data
Monthly data are aggregated data. They are stored with observation time as
If timeoffset = 0, the monthly value represents the calendar month, midnight to midnight. The time is of the form
If timeoffset != 0, the monthly value represents the period from p1 to p2 where p1 and p2 are midnight + timeOffset on the last day of the previous and current month respectively.
As an example, element ID sum(precipitation_amount P1M) has timeOffset PT6H. This means that the observation period lasts from
Exception: Old monthly minimum air temperature, min(air_temperature P1M)em> from 1894-1937 were calculated using the average minimum temperature of 2 days, and is hence based on the period from
Example 2.4: Seasonal data
Seasonal data represent the periods of spring (1 March - 31 May), summer (1 June - 31 Aug.), autumn (1 Sep. - 20 Nov.), winter (1 Dec. - 28/29 Feb.), winter half year (1 Oct. - 31 March), and summer half year (1 April - 30 Sep.).
In this case, time is set to the start date of the seasonal period, e.g. for spring: time =
timeoffset follows the same rules as for monthly data.
Example 2.5: Annual data
Annual data cover the calendar year. Time and timeoffset for annual data follow the same conventions as for seasonal and monthly data.