Meteorological data from ground stations
The processing of observed station weather into the MCYFS involves four steps:
Data acquisition from weather stations
The selection of stations is limited to those stations that regularly collect data and can supply data in near real time. Relevant meta data of stations includes station number, station name, latitude, longitude and altitude. This data is available in the object STATIONS.
Currently, data acquisition and processing applies to two regional windows: Europe and China. Mainly examples from Europe are shown in this documentation.
Some of the historic meteorological data were purchased directly from National Meteorological Services. Others were acquired via the . As data are obtained from a variety of different sources, considerable pre-processing was necessary to convert them into a standard format. Around 1992 the historic meteorological data represented approximately 380 stations in the EU, Switzerland, Poland and Slovenia with data from 1949 to 1991 (Burrill and Vossen, 1992). Later the historic sets have been extended with stations in Eastern Europe, western Russia, Maghreb and Turkey. The historic data were converted into consistent units and checked on realistic values. The database was also scanned for inconsistencies, such as successive days with the same value for a variable, or minimum temperatures higher than maximum temperatures (Burrill and Vossen, 1992).
From 1991 to present, meteorological data is received in near-real-time from open data sources and from contracted providers like ECOMET or national or regional meteorological services. Sources include the WMO GTS network, NOAA data access points, regional and national meteorological services, and the access points for non-essential WMO reports. The data arrives either in standardized encoded formats as defined by the WMO or ICAO, or in proprietary formats as used by individual providers. It is first decoded and converted into a generalized structure, including unit and time zone conversions, alignment of reference periods and - where needed - the assignment of station-id. Basic first level data sanity checks are applied. In a next step, the data is converted into the input-format as required by the QUACKME software package. The temporal resolution of the data ranges from 1-hourly to 24-hourly, depending on the parameter. QUACKME is applying data quality checks, calculates derived parameters and daily aggregates and writes the data into the file formats as expected by the EFAS and MCYFS downstream processes.
In recent years, the earlier archives (1975-2004) of Scandinavia and eastern Europe have been enriched. In 2016 data from around 300 Chinese stations have been acquired starting a new service for this region.
The stations, stored in object STATIONS holds over 10221 stations distributed over 40 countries in Europe and neighboring countries. Over 5100 of these stations provide weather data in near real time. All weather data is stored in the stations weather database (daily data in object WEATHER_OBS_STATION and 3- and 6-hourly precipitation in object WEATHER_OBS_STATION_RAIN).
Raw station data is collected from various sources:
- (essential data and data licensed by restrictions)
- European National Meteorological Institutes (NMI) (licensed)
- Various regional networks in Europe
For transmission and international exchange, the internationally exchanged station reports are encoded in formats standardized and maintained by and International Civil Aviation Organizaton (ICAO)
Observations as provided directly by National Meteorological Institutes or regional authorities come from secondary networks and are provided in proprietary formats.
Meteorological stations selected in priority are those located in the agricultural zones and equally distributed over the mainland (instead of over islands - for Portugal, Spain or Greece in particular). In particular, for western Russia (western of Urals) the main areas covered are the agricultural districts.
In the case of China roughly 300 stations were selected meeting the following criteria:
- Near real time delivery
- A 20-years archive
- Located in the main agricultural areas
- Covering the elements: precipitation, minimum and maximum temperature, humidity and wind speed
The basic indicators that are received from weather stations include:
- Sum of precipitation
- 2m air temperature
- Maximum of 2m air temperature
- Minimum of 2m air temperature
- Downward directed solar radiation measured at earth's surface (global radiation)
- Duration of sunshine
- Total cloud cover
- Water vapour pressure
- Relative humidity
- 10m mean wind speed
- Snow depth
For WMO SYNOP FM 12 and BUFR FM 94 bulletins, WMO defines regional regulations to consider time zones and national coding practices. The extent of reported parameters and the report frequency differs per country and is for ECOMET member countries affected as well by license restrictions.
The METAR code is standardized through the ICAO. In Europe and China, the WMO-maintained codes SYNOP FM12 and BUFR FM 94 provide higher accuracy for the various parameters and more detail. In these regions, METAR provides only temperature, dew point, visibility, cloud amount and wind speed and is reported in coarser increments for the various parameters. The reporting frequency is determined by the individual flight operation of each airport. Nevertheless, METAR reports are used as well, mostly to fill spatial gaps in areas with less WMO stations.
Observation data from several European regional meteorological networks became available after 2015. The data from regional networks are mostly not available in the standard meteorological formats, but have to be collected and converted individually. The quality of the data is determined by the installed sensors and the siting of the stations. The reported parameters and frequency differ by network.
The following table summarizes basic information on the availability and reporting regulations from the various observing station data sources:
|Parameter||Reference periods of reports as defined by WMO||WMO formats BUFR or SYNOP (*)||METAR (**)||Regional networks|
|Sum of precipitation||24-hourly sum, 12-hourly sums, 6-hourly sum, 1-hourly sum reported, depending on region WMO-region and local regulations||Europe: 06 UTC: past 24 hours / 00 UTC and 12 UTC: past 6 hours / 06 UTC and 18 UTC: past 12 hours / 1-hourly -- China: Reports 00 UTC for past 24 hours, some stations report 21 UTC for previous 24 hours (***)||Not reported in Europe and China||Individual, mostly 1-hourly|
|2 m air temperature||Instantaneous value||Reported with 0.1 K accuracy||Reported as full degrees||Mostly 0.1 K accuracy|
|Maximum 2 m air temperature||Maximum of continuous measurement during reference period (****)||Europe and China: reported 18 UTC||Not reported in Europe and China||Individual|
|Minimum 2 m air temperature||Minimum of continuous measurement during reference period (****)||Europe and China: reported 06 UTC||Not reported in Europe and China||Individual|
|Downward directed surface solar radiation (global radiation)||Sum accumulated over past 24 hours, sum past 1 hour||Available for some European countries at 00 UTC, 1-hourly||Not reported in Europe and China||Individual definition, mostly 1-hourly|
|Duration of sunshine||Sum accumulated over past 24 hours||Most European countries report at 06 UTC||Not reported in Europe and China||Individual, mostly 1-hourly|
|Total cloud cover||Instantaneous value||Octas 0-8||5 stages, only clouds up to a height of 5000 feet over ground reported||Not reported|
|Measures for the humidity of the air at 2 m above ground: dew point, water vapour pressure and relative humidity||Instantaneous value of dew point temperature reported (*****)||Reported with 0.1 K accuracy||Reported as full degrees||to be derived from other humidity parameters like relative humidity and air temperature|
|10 m mean wind speed||Mean over past 10 minutes||Meters per second||Mostly full knots, occasionally less accuracy during low wind situations||Individual definition|
|Snow depth||Instantaneous value, increasing automatization of measurement||When a station reports snow depth, it is done in Europe by 06 UTC, in China by 00 UTC||not reported||Not reported|
(*) Main synoptic hours are 00, 06, 12, 18 UTC. Intermediate synoptic hours are 03, 09, 15, 21 UTC.
For most European countries, 1-hourly data is used as well.
(**) The report frequency is determined by the airport's schedule and can be as often as 20 minutes. The frequency of reports can change over daytime, weekday, and season.
(***) In BUFR, several countries do not provide the reference period during dry conditions in the FM94 code, supposedly by accident. In this case, it is assumed that the WMO definitions for the reference period are applied.
(****) Europe: Covers past 12 hours. China: Covers past 24 hours.
(*****) Other thermodynamical measures for the humidity of air can be calculated from dew point and air temperature.
Data quality check
The software package Quality Checks Meteorological Data (QUACKME) as developed by the JRC is the main processing tool for completing and quality evaluation of actual meteorological data which is used as input for agro-meteorological models. The data processing workflow with quality control and aggregation can be described as follows.
Near real-time pre-processing (1-hourly reports with extended information at intermediate and main synoptic hours, irregular reports)
- Near real-time collection of reports from the various data sources.
- Decoding of the WMO and ICAO standard formats with dedicated decoder software (FMDecode). Reports in other formats from regional, secondary networks are translated into a uniform structure using individual proprietary converters.
- The data is converted into a generalized structure, including the conversion towards UTC, standard units, the alignment of reference periods and the calculation of derived parameters. Basic sanity checks are applied.
Preparation of QUACKME input data
- Generate a csv with all available observation data for the period of 24 hours (07 UTC - 06 UTC next day) for the European region and for China, respectively. The format of the csv is described in the (pdf)
- When data from a station are found to be erroneous for a longer time, the station can be listed on a so-called blacklist, either by parameter or for the whole report. Observations from blacklisted station-parameter combinations are not written into the csv. The blacklist is manually checked every three months.If the messages are considered trustworthy again, the station-parameter combination is removed from the blacklist.
- Generate csv with location specific, near real time forecasts for the same stations and period as the data in the observation-csv. The format of this csv is as well described in the QUACKME Technical Guide.
For a number of weather elements, QUACKME compares the observed values with near-real-time forecast values. The forecast is used as reference for the reasonable range of possible values. The forecasts are obtained through a technique called MOS (Model Output Statistics). Meteorological forecast models, e.g. the ECMWF model, compute the physical status of the atmosphere on a grid, and the results represent the expected situation per grid box. The MOS forecast is using statistical relationships between the observations of a particular station and historic model forecasts for surrounding grid points. Each observing location has its own statistics. In this way, the local conditions at the weather station can be modelled much more accurately. QUACKME is using the individual location forecasts to define time- and location-dependent thresholds for the trustworthiness of station reports, for the elements air temperature (including minimum and maximum), dew point (applies to all derived measures for the humidity of the air), precipitation, and wind speed, respectively. That way, the thresholds consider season, climatology and even the actual weather pattern. A welcome side effect is the high spatial consistency of the statistical MOS approach and therefore of the thresholds. Individual MOS forecasts is used for almost all stations (approx. 5000, state January 2021).
Running the QUACKME modules and interactive data quality checks by the meteorologist
- This does not apply for precipitation, i.e. for consecutive reports of 0 mm. This rather typical reporting bug is not found when quality checks are applied on to the data of the very day. Due to the mostly “patchy” pattern of precipitation events quality checks accept dry stations in between. To find stations that report consecutively 0 mm several weeks of history need to be considered, see retrospective checks.
- Correct automatically obvious errors detected while performing these checks;
- Automatically fill gaps in the database through interpolation based on time consistency criteria;
- Flag dubious observations which cannot be corrected automatically;
- Write all automatic corrections and flagged dubious observations to a log file;
- Have the flagged observations checked and, if necessary, corrected by a trained meteorologist; when a correction is done, the derived parameters are recalculated and the data are written back to the database.
Dedicated trained and qualified meteorologists go through the dubious observation values that are flagged as such by the QUACKME automatic pre-checking program. An interactive system for the visualization of meteorological data is used to graphically visualize and analyze additional information such as:
- Station observation data
- Satellite images
- Precipitation Radar data
- Analysis and short range forecasts computed by physical models of the atmosphere
- Short range forecasts for weather station locations
This additional data is used by the analyst to decide on either confirmation or rejection of the observed values.
Conversion to daily values
Once the database has been filled following the method described above, data are aggregated to daily values. This includes the indicators as summarized in the following table:
|Parameter||Aggregation||Reference period Europe||Reference period China|
|Total cloud cover (N)||Daily mean||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC, 21 UTC, 00 UTC next day||18 UTC prev., 21 UTC prev., 00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC|
|Duration of sunshine (Msun)||24-hourly sum||00–24 UTC||Not available|
|Downward directed surface solar radiation (global radiation) (Mrad)||24-hourly sum||00-24 UTC||Not available|
|Minimum 2m air temperature (Tn)||Lowest value of continuous reference period (*)||18 previous day -06 UTC||06 UTC previous day – 06 UTC|
|Maximum 2m air temperature (Tx)||Highest value of continuous reference period (**)||06-18 UTC||18 UTC previous day – 18 UTC|
|Water vapour pressure (e)||Daily mean||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC, 21 UTC, 00 UTC next day||18 UTC prev., 21 UTC prev., 00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC|
|10m mean wind speed (ff10)||Daily mean||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC, 21 UTC, 00 UTC next day||18 UTC prev., 21 UTC prev., 00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC|
|Sum of precipitation (RRR)||24-hourly sum||Mostly 06 UTC until 06 UTC next morning|| Mostly 00 UTC – 00 UTC next day (indicator 2).|
For some stations 21 UTC previous day – 21 UTC (indicator 6)
|2m air temperature (TT)||03-hourly instantaneous values during daytime||06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC|
|Relative humidity (RH)||03-hourly instantaneous values during daytime||06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC|
|State of soil||Instantaneous value (***)||00 UTC following day|
|Water vapour pressure deficit (vpd)||Daily mean||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC, 21 UTC, 00 UTC next day||18 UTC prev., 21 UTC prev., 00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC|
|Slope of saturation vapour pressure vs. temperature curve slope||Daily mean||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC, 21 UTC, 00 UTC next day||18 UTC prev., 21 UTC prev., 00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC|
|Total cloud cover (N)||Daytime mean||06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC|
|Low or (when no low clouds) medium clouds (Nh)||Daytime mean||06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC||Not available|
|Calculated sunshine duration (Csun)||24-hourly sum||To be calculated by QUACKME, 0-24 UTC of the day specified||To be calculated by QUACKME, 18 UTC previous day - 18 UTC of the day specified|
|Highest possible global radiation at clear sky (Crad)||24-hourly sum||To be calculated by QUACKME, 0-24 UTC of the day specified||To be calculated by QUACKME, 18 UTC previous day - 18 UTC of the day specified|
|Potential evapotranspiration (ETP)||24-hourly sum||To be calculated by QUACKME, 0-24 UTC of the day specified||To be calculated by QUACKME, 18 UTC previous day - 18 UTC of the day specified|
|Visibility (VV)||Daytime mean||06 UTC, 09 UTC, 12 UTC, 15 UTC, 18 UTC||00 UTC, 03 UTC, 06 UTC, 09 UTC, 12 UTC|
|Snow depth||Instantaneous value||06 UTC||00 UTC|
(*)When no minimum is reported but hourly instantaneous temperatures QUACKME estimates the minimum from the hourly local early morning values, see
(**)When no maximum is reported but hourly instantaneous temperatures QUACKME estimates the maximum from the hourly local afternoon values, see (pdf)
(***)Code, for translation see BUFR documentation.
Information on the way the daily element values are constructed/defined is stored in the object WEATHER_OBS_STATION_INFO. Currently this is only done for precipitation e.g. period definition of the daily rainfall sum. Codes are:
- 0 = real observation 06 - 06 UTC next day
- 1 = period 06 - 06 UTC next day, short range forecast has been used to cover the complete period
- 2 = real observation 00 UTC - 24 UTC
- 3 = real observation 03 UTC - 03 UTC next day
- 4 = real observation 12 UTC previous day - 12 UTC
- 5 = real observation 18 UTC previous day - 18 UTC
- 6 = real observation 21 UTC previous day - 21 UTC
More information on the 3- and 6-hourly precipitation data are stored in object WEATHER_OBS_STATION_RAIN (column IDFLAG). Codes are:
- 1 = Changed by meteorologist (not applicable)
- 2 = Automaticaly corrected (not applicable)
- 3 = Observation
- 4 = Linear interpolation from observations
- 5 = Interpolated via MOS from observations
- 6 = MOS analyses (not managed yet)
Finally, meta data of all stations is checked once a year.
Retrospective checks and blacklisting of suspect stations
Some suspicious station reports are only detectable by checking time series of several weeks. Continuous reports of 0 mm precipitation (instead of a "precipitation not observed" flag) do not stick out in daily rainfall sums, but only by investigating the station's reports over a longer period. Global radiation and cloud cover have a high spatial volatility, and continuous observation or encoding errors at a certain station become more explicit when looking into several weeks of station reports.
For all European stations, the QUACKME output of the past 40 days is inspected each week through time series checks. Provided a station reported on more than half of the tested days, the reports are checked, consulting ECMWF model analysis and short range forecasts for model grid points surrounding the station of request.
The checks are set up as follows:
A station is flagged as suspicious for precipitation when suspicious consecutive zero rainfall reports are detection. Criteria are
- The observed precipitation sum is 0 mm whan aggregated over the whole checked period.
- the ECMWF model near real time forecasts during the checked period included at least 10 wet days. A day is considered being wet when more than 0.5 mm precipitation is forecasted by the model's near real time foreast.
For each day of the investigated period, the station’s maximum possible daily solar radiation sum is calculated, based on its latitude, the time of year, and using a standard atmospheric optical depth. A station is flagged as being suspicious for radiation when:
- There are at least 10 days with observed solar radiation exceeding 110 % of the maximum possible amount of solar radiation.
- There are at least 10 days on which the observed solar radiation remained below 10 % of the maximum possible daily sum of solar radiation.
- There are at least 10 days with observed radiation of 0 MJ m-2 day-1 whilst the ECMWF short-range forecast analysed solar radiation exceeding 0 MJ m-2 day-1.
- The total sum of observed solar radiation is less than 25 % of the maximum possible radiation sum for the period, whilst the sum of the model's short-range forecasts for the parameter exceeded 25 % of the maximum possible daily sum of solar radiation. Naturally, the maximum possible radiation period's is only summed up from days with observations being available.
Mean daytime cloudiness
A station is flagged as being suspicious for cloudiness when:
- A difference of more than 2.5 octa between the daily mean of observed total cloud cover and the daily mean of ECMWF model analysis and short-range forecast for total cloud is found for all days of the investigated period.
- The reported instantaneous cloudiness was always higher than 4.0 octa whilst the model analysis and short-range forecasted for at least three time steps (hours) in the period a total cloud cover of less then 3.0 octa.
- For all time steps in the period more than 5 octa total cloud cover was reported.
Duration of sunshine
For each day in the investigated period, the maximum day length is calculated based on the day of the year and the station latitude. Dividing the observed sunshine duration for a day by the calculated day length gives the relative sunshine.
A station is flagged as being suspicious for sunshine duration when:
- For more than 10 days in the period, the observed duration of sunshine is more than 110% of the calculated day length.
OR Depending on the dominant season during the period:
- Summer: highest relative sunshine value is less than 30% (i.e. the station is always cloudy).
- Winter: the lowest relative sunshine value is more than 70% (i.e. the station is always sunny)
- Spring/autumn: highest relative sunshine value is less than 30% (see summer check) OR the lowest relative sunshine value is more than 70% (see winter check).
The dominant season is determined as the season with the largest number of days in the investigated period. When 50% of the investigated days are winter/summer days, the dominant season will be winter/summer.
When the process flags stations as suspect a final manual inspection by a meteorologist follows. If the time series of the station are found to be wrong the following actions are executed:
- The station is added to a blacklist: the station is immediately excluded from the operational station list.
- The erroneous time series are deleted from the objects WEATHER_OBS_STATION_RAIN and WEATHER_OBS_STATION. The erroneous values are flagged (object WEATHER_OBS_STATION_RAIN, column TYPE) or deleted (object WEATHER_OBS_STATION and WEATHER_OBS_STATION_INFO) and deleted values are saved in separate objects (WEATHER_OBS_STATION_ERRORS and WEATHER_OBS_STATION_INFO_ERR).
- All affected grid cells (object WEATHER_OBS_GRID) and regions (object WEATHER_OBS_REGIONCOVER) are reprocessed at regular time intervals. This also includes the crop simulation results.
Every three months, by the end of the quarter, each station on the blacklist is verified. Afterwards it is decided if stations can return to the operational work flow. Falsely blocked data is back-ordered, added and reprocessed.
Station data availability
Each month an overview is created showing the delivered number of stations per country. Information is also added on sudden changes and follow-up actions. Example monthly overview
Every day, the newly produced data files are compared with those of the previous day. If the number of delivered values for individual countries and parameters decreases significantly, an alert is sent by email is sent to the project team. The threshold above which a decline in the number of values delivered is considered critical depends exponentially on the number of values in the country.
The number of values flagged by the weak, heavy and threshold checks of QUACKME are monitored on a daily and on a monthly basis. Stations that are flagged particularly frequently are identified and the cause can be analysed separately. As a result, stations can be blacklisted or an improvement of the QUACKME checks can be suggested.
The following maps illustrate the available stations (red 0-20% - green 80-100%) for the main elements in a recent year 2019. The main elements (maximum temperature, minimum temperature, precipitation, sun shine, cloud cover, wind speed and vapor pressure) have a good spatial spread over Europe with a relative high spatial density in western and central Europe. Availability of measured radiation is mainly limited to western and central Europe.
|maximum temperature||minimum temperature||precipitation|
|global radiation||sunshine||cloud cover|
|wind speed 10m||vapor pressure||snow depth|
The following graph shows the increase of observations for the main elements between 1975 and 2019. Most elements have at least 600,000 annual observations which equals over more than 1600 stations in case they would have a complete temporal coverage. However, most stations have temporal gaps and therefore the number of reporting stations is much higher. Since 2004 the number of observations increased up-till a level of around 1,500,000 reported by more than 4500 stations. During the recent years also observations of radiation related elements increased drastically. This is especially true for cloud cover and sunshine. Prior to 1995 these elements have a relative low number of observations meaning that the global radiation of these years, required in MCYFS, is mainly based on the daily temperature range, see Calculation of advanced parameters.
In general the station density and available data in the monitored areas is considered sufficiently high for the purpose of the project.
Ingestion into the database
After the station weather data passed all checks, daily weather data is exported to a fixed formatted ASCII file (s-file) containing the data of a single day that can be imported in the object WEATHER_OBS_STATION. In the near real time situation a s-file is delivered one day later. For example in the afternoon of day 31 March 2016 the following file is generated: s20160330.dat.
|Format ASCII s*.dat file (daily station weather)|
* Codes for state of soil: 0 = surface of ground dry, without cracks or appreciable amount of dust or loose sand, 1 = surface of ground moist
The 3-hourly rainfall data is exported to a plain ASCII file (rrr3h_*.txt file) containing the data of one 3-hourly time step within one single day. This data can be imported in the object WEATHER_OBS_STATION_RAIN. In the near real time service each day 8 rrr3h_*.txt files are generated at once containing data of 8 3-hourly time steps:
- 09 UTC (06-09 UTC of previous day)
- 12 UTC (09-12 UTC of previous day)
- 15 UTC (12-15 UTC of previous day)
- 18 UTC (15-18 UTC of previous day)
- 21 UTC (18-21 UTC of previous day)
- 00 UTC (21-00 UTC of previous day)
- 03 UTC (00-03 UTC of present day)
- 06 UTC (03-06 UTC of present day)
For example in the afternoon of day 31 March 2016 the following files are generated: rrr3h_2016033009.txt, rrr3h_2016033012.txt, rrr3h_2016033015.txt, rrr3h_2016033018.txt, rrr3h_2016033021.txt, rrr3h_2016033100.txt, rrr3h_2016033103.txt and rrr3h_2016033106.txt.
|Format ASCII rrr3h_*.txt file (3-hourly station rainfall)|
The 6-hourly rainfall data is exported to a plain ASCII file (rrr_*.txt file) containing the data of one 6-hourly time step within one single day. This data can be imported in the object WEATHER_OBS_STATION_RAIN. In the near real time service each day 4 rrr_*.txt files are generated at once containing data of 4 6-hourly time steps:
- 12 UTC (06-12 UTC of previous day)
- 18 UTC (12-18 UTC of previous day)
- 00 UTC (18-00 UTC of previous day)
- 06 UTC (00-06 UTC of present day)
For example in the afternoon of day 31 March 2016 the following files are generated: rrr_2016033012.txt, rrr_2016033018.txt, rrr_2016033100.txt and rrr_2016033106.txt.
|Format ASCII rrr_*.txt file (6-hourly station rainfall)|
Calculation of advanced parameters
Global radiation is the daily sum of incoming solar radiation that reaches the earth surface. It is mainly composed of wavelengths between 0.3 μm and 3 μm. Approximately half of the incoming radiation with wavelengths between 0.4 and 0.7 μm is Photosynthetically Active Radiation (PAR). Global radiation is the driving variable in the growth-determining CO2 assimilation process and thus crop growth models are sensitive to radiation data (van Diepen, 1992).
A major problem is the scarcity of measured global radiation. In cases where no direct observations are available it must be derived from sunshine duration, cloud cover and/or temperature, on the basis of statistical relationships. If measured global radiation is missing, it is based on one of three formulae (Ångström-Prescott, Supit-Van Kappel, and Hargreaves), depending on the availability of meteorological parameters. An important component in these formulae is the amount of Angot radiation which is the extraterrestrial radiation integrated over the day at certain latitude on a certain day. The calculation of the Angot radiation and the three different formulae are described by Supit et al. (1994) and van der Goot (1998a).
The principle component of all three formulae is the extraterrestrial radiation, or Angot radiation. In fact, all of the three formulae estimate the fraction of Angot radiation actually received at the earth surface. The Angot radiation is calculated as:
The following hierarchical method is used to calculate global radiation for each station (Supit and van Kappel, 1998) in case measured global radiation is missing:
If sunshine duration is available, global radiation is calculated using the equation postulated by Ångström (1924) and modified by Prescott (1940). The two constants in this equation depend on the geographic location.
Supit-Van Kappel formula
When neither measured radiation nor sunshine duration are available, but minimum and maximum temperature and daytime cloud cover are known, the Supit-Van Kappel formula is used. This is an extension of the Hargreaves formula (Supit, 1994). The regression coefficients depend on the geographic location.
When only the minimum and maximum temperatures are known the equation of Hargreaves et al. (1985) is used. The regression coefficients depend on the geographic location.
Any one of the above three methods has an additional upper limit. The maximum calculated global radiation is set to Angot radiation, corrected for atmospheric transmissivity, by multiplying the Angot value with the sum of the Angstrom A and B coefficients.
Deriving Ångström-Prescott, Supit-Van Kappel, and Hargreaves regression constants
The main problem with the application of the Ångström-Prescott, Supit-Van Kappel, and Hargreaves formulae is the quality of the regression constants. Studies by Supit (1994), Supit and van Kappel (1998) and van Kappel and Supit (1998) showed no relationship between latitude and the coefficients for Europe, although such a relation is frequently used to estimate these regression constants. Initially in MCYFS regression constants of Supit and van Kappel (1998) and van Kappel and Supit (1998) for Europe were used. They obtained sets of regression constants for the formulae for as many weather stations as possible, with a geographic distribution that corresponds to the area of interest for the MCYFS. As a result, a set of 256 reference stations was identified for which a relevant set of measured radiation data and other parameters in the formulae existed. For these stations regression constants were calculated based on measured radiation data for the three formulae mentioned above.
In 2012 the regression coefficients of these solar radiation models for Europe were updated using a new set of weather station data (temperature, sunshine and cloudcover) and an alternative training data set: 6 years (2005-2010) of the down-welling surface shortwave radiation flux (DSSF) 30-minutes product derived from Meteosat Second Generation satellite data by the Land Surface Analysis Satellite Applications Facility (LSA SAF) (Bojanowski et al.,2013). For each solar radiation model a set of weather stations was selected having sufficient observations of either sunshine duration, or cloud cover/temperature or only temperature (minimum and maximum) to perform a regression analysis. Results are stored in object STATION_REFERENCE_COEFFICIENTS (CGMS14SYS).
Station archive data for China did not include measured radiation nor sunshine. Therefore radiation was derived from other observed elements namely cloud cover and minimum and maximum temperature. The Hargreaves and Supit-VanKappel models have been trained using modelled radiation by Tang et al., 2013. The 50yrRad database of Tang et al., 2013 containing ‘modelled’ radiation data for 716 CMA stations, has demonstrated its superior performance over previous estimates of locally calibrated Angstrom-Prescott models. While radiation is based on the Hargreaves or Supit-VanKappel models, coefficients of the Angstrom method are still required to calculate net outgoing long wave radiation within the potential evapotranspiration calculation. For determining Angstrom coefficients only the 50yrRad archive was used. Since no sunshine duration data is available, an alternative was sought. Transmissivity was derived by dividing the measured solar radiation at the ground by the solar radiation at the top of the atmosphere. By selecting only the period between day of year 150 and 200 (during mid-summer) the transmissivity is almost constant and can be linked to the Angstrom coefficients.
The program SupitConstants uses this set of data (via the view SUPIT_REFERENCE_STATIONS, CGMS14SYS), consisting of latitude, longitude, altitude and calculated regression constants, to derive the regression constants for all stations in the MCYFS. Interpolation of the regression constants of the reference stations to other stations is based on a distance weighted average of the three nearest stations. This process is carried out once, unless the set of reference stations changes or when new stations are added or when meta data of stations change.
|Interpolation of regression constants|
| Data of the reference stations, consisting of latitude, longitude, altitude and the regression constants, is being used for the derivation of the regression constants for the set of stations used for the interpolation of the daily meteorological data. This is a process that only has to be carried out once, unless the set of reference stations changes or when new stations are added or when meta data of stations change. Once the regression constants have been established for the operational set of stations, the global radiation estimation can proceed using any one of the formulae.
The interpolation of the regression constants is based on a simple distance weighted average of the three nearest stations. For each of the three sets of constants (Ångström-Prescott, Supit-Van Kappel, and Hargreaves) a subset is created from the complete set of reference stations, by selecting only those stations that have the regression coefficients for the desired method. This subset of stations is then sorted based on distance to the station for which the regression coefficients are being calculated. This sorting process is also subject to an altitude threshold test i.e. if the altitude difference between the target station and a reference station is greater than a set threshold the reference station is rejected in favour of the next nearest reference station. Depending on a distance threshold, the nearest one, two or three stations are then used to calculate the regression constants. If the threshold tests exclude all stations, the nearest station will be used, regardless of the distance. The altitude threshold value is 200 m; the distance threshold is 200 km.
The distance weighted average method used, is based on the relative distance of the reference stations to the station of interest.
Assume the distances d0, d1 and d2 to be the distances to the three nearest reference stations, and w0, w1 and w2 the weights to be used in the calculation. As an example, assume that d1 is 2*d0, then w1 will be w0/2. More general, w1 = w0*d0/d1. Similarly, w2 = w0*d0/d2. Furthermore, the sum of the weights should be 1, so w0+w1+w2 = 1. From the above, the following relation can be established:
Interpolated regression constants are written in the temporary object SUPIT_CONSTANTS (CGMS14SYS) and copied to object STATIONS (CGMS14SYS). After the regression constants have been established for all stations, global radiation can be calculated by using any one of the above formulae. Finally, the derived daily global radiation of each station is written into object WEATHER_OBS_STATION_CALCULATED (see flowchart).
Daily meteorological station data collected from stations does not contain potential evapotranspiration by crop, wet soils and open water. Potential crop evapotranspiration (ET0) is calculated by the Penman-Monteith equation while potential evapotranspiration of wet soils (ES0) and open water (E0) is calculated by the Penman equation.
|Calculation of potential evapotranspiration|
++ Penman-Monteith ++
Daily meteorological station data collected from stations does not contain potential crop evapotranspiration. This parameter is calculated by the Penman-Monteith equation (Allen et all., 1998). In general, the evapotranspiration from a reference surface, the so-called reference crop evapotranspiration or reference evapotranspiration can be described by the FAO‑Penman-Monteith:
Next, the different components of this formula are calculated. As the magnitude of the day or ten-day soil heat flux (G) beneath the grass reference surface is relatively small, it is ignored.
The net radiation (Rn) is the difference between the incoming net shortwave radiation (Rns) and the outgoing net longwave radiation (Rnl). The net shortwave radiation (Rns) is calculated as follows:
The outgoing net longwave radiation (Rnl) is calculated as follows. First clear-sky radiation (Rso) is derived:
Then, the outgoing net longwave radiation (Rnl) is calculated:
The psychrometric constant is corrected for atmospheric pressure:
Next, saturated-vapour-pressure is calculated for both the minimum and maximum temperature and averaged afterwards:
Finally, the slope of the saturation vapour pressure curve is determined (first the minimum and maximum temperature are averaged to obtain the average temperature):
The Penman-Monteith algorithm is valid only for a reference canopy (ET0) and therefore it is not used to calculate the reference values for bare soil and open water (ES0, E0). The background is that the Penman-Monteith model is basically a surface energy balance where the net solar radiation is partitioned over latent and sensible heat fluxes (ignoring the soil heat flux). To estimate this partitioning, the method links between the surface and air temperature. However, the assumptions underlying the model are valid only when the surface where this partitioning takes place is the same for the latent and sensible heat fluxes. For a crop canopy this assumption is valid because the leaves of the canopy form the surface where both latent heat flux (through stomata) and sensible heat flux (through leaf temperature) are partitioned.
For a soil, this principle does not work because when the soil is drying the evaporation front will quickly disappear below the surface and therefore the assumption that the partitioning surface is the same does not hold anymore. For water surfaces, the assumptions underlying Penman-Monteith do not hold because there is no direct relationship between the temperature of the water surface and the net incoming radiation as radiation is absorbed by the water column and the temperature of the water surface is co-determined by other factors (mixing, etc.). Only for a very shallow layer of water (1 cm) the Penman-Monteith methodology could be applied. For bare soil and open water the Penman model is preferred. Although it partially suffers from the same problems, it is calibrated somewhat better for open water and bare soil based on its empirical wind function.
Finally, in crop simulation models the open water evaporation and bare soil evaporation only play a minor role (pre-sowing conditions and flooded rice at early stages), it is not worth investing much effort in improved estimates of the reference values.
Evapotranspiration from a wet bare soil surface (ES0) and from a water surface (E0) is calculated with the Penman formula (Penman, 1948). Only the albedo and surface roughness differs for these two types of evapotranspiration as explained below:
The net absorbed radiation depends on incoming global radiation, net outgoing long-wave radiation, the latent heat and the reflection coefficient of the considered surface (albedo). For ES0 and ET0 albedo values of 0.15 and 0.20 are used respectively. The evaporative demand is determined by humidity, wind speed and surface roughness. For a free water surface and for the wet bare soil (E0, ES0) a surface roughness value of 0.5 is used. For a more detailed description of the underlying formulae we refer to Supit et al. (1994) and van der Goot (1997).
Calculated E0, ES0, and ET0 are stored in object WEATHER_OBS_STATION_CALCULATED.
Messages to the Project Management Board
Information on successfull completion of the various processing steps is sent to the Project Management Board (PMB).
|List of signals communicated to the Project Management Board (PMB) in connection to the processing of observations from ground weather stations.|