Every day, hundreds of thousands of observations are made of the atmosphere around the world, measuring quantities such as pressure, wind, temperature and humidity.
Current main sources of observations
To use these observations in an operational weather forecasting system we have to monitor their availability; quality control them, and process them into a form that can be used by the computer models and forecasters.
We develop and maintain software systems which can do all of these tasks for current operational data sources, and which can be adapted for new observational systems in the future.
These collect information about cloud type and present weather and are usually sited at official weather stations. The measured surface pressure at a site usually reduced to a mean sea-level pressure, to ensure consistency between sites. Many stations report pressure at the station altitude and some high-level stations compute a height at 850 hPa.
This is surface data collected from ships and buoys. While temperature and moisture are measured at a height of 1-2 m over land, height may vary for marine data. This data can also included cloud and present weather infomation. There are also some bouys that gather more climatic data about our oceans, such as temperature and salinity at various depths. Bouys used for measuing weather data are in fixed locations, research buoys are mostly tracked by GPS.
Wind data for most of the Earth comes from measurements of ocean-backscatter observed from satellites. Wind data is gathered at 10 m over land at weather stations to minismise the effects of friction, and where this is not possible adjustments are made to ensure consistency between sites. The height may vary for marine data and, again, may need to be adjusted.
We continually investigate ways to exploit satellite data, including atmospheric motion and wind reports and satellite sounding information on the temperature and composition of the atmosphere.
Weather balloons - radiosondes
A radiosonde is a unit used on weather balloons that measures various atmospheric parameters and transmits them to a fixed receiver. Radiosonde reports help us build a detailed picture of the atmosphere at that location. In some reports there may be more than 100 levels.
The radiosonde system provides measurements of wind speed and direction, temperature and dewpoint temperature. The dewpoint data provides information on moisture and are usually combined with the temperature data to provide the moisture information as relative humidity.
A special feature of the radiosonde processing is vertical averaging. The vertical averaging maps the reports on to the computer models.
Observations of temperature and winds are available either from manual or automatic reporting systems on flights. Automated reports from commercial airliners are provided through the AMDAR (Aircraft Meteorological DAta Relay) programme.
True representation of actual conditions is more of a problem with aircraft reports than some upper-air observation systems, particularly in the vertical. Planes flying in the direction of the wind may seek the core of the jet stream, which is restricted in height, thus sampling part of the atmosphere which is not fully resolved by the model (where typically the vertical resolution may be only 50 hPa).
Automated reports from commercial airliners are provided through the AMDAR (Aircraft Meteorological DAta Relay) programme. Recently, a new type of automated aircraft report has been developed in the USA, primarily for use by regional airlines, called TAMDAR (Tropospheric Airborne Meteorological DAta Report, or 'Tropospheric AMDAR'). It is envisaged that these two types of report will ultimately replace the manual reports known as AIREPs.
Observations monitoring and quality control
We monitor meteorological observations received from a variety of sources worldwide. This is primarily to maintain and improve the use of these observations in our computer assimilation system (4D-VAR), which is part of the numerical weather prediction (NWP) system.
The quantity and quality of observational data received and assimilated are checked daily, and any problems followed up. Observations are compared with short-period forecast (background) fields and observation-minus-background (o-b) statistics are used for monitoring over various time periods. On a monthly basis any poor quality data that is identified is either added to reject lists and excluded from the assimilation or corrected prior to use.