Validation of Mode‐S Meteorological Routine Air Report aircraft observations

Abstract

The success of mesoscale data assimilation depends on the availability of three‐dimensional observations with high spatial and temporal resolution. This paper describes an example of such observations, available through Mode‐S air traffic control system composed of ground radar and transponders on board the aircraft. The meteorological information is provided by interrogation of a dedicated meteorological data register, called Meteorological Routine Air Report (MRAR). MRAR provides direct measurements of temperature and wind, but is only returned by a small fraction of aircraft. The quality of Mode‐S MRAR data, collected at the Ljubljana Airport, Slovenia, is assessed by its comparison with AMDAR and high‐resolution radiosonde data sets, which enable high‐ and low‐level validation, respectively. The need for temporal smoothing of raw Mode‐S MRAR data is also studied. The standard deviation of differences between smoothed Mode‐S MRAR and AMDAR is 0.35°C for temperature, 0.8 m/s for wind speed and below 10 degrees for wind direction. The differences with respect to radiosondes are larger, with standard deviations of approximately 1.7°C, 3 m/s and 25 degrees for temperature, wind speed and wind direction, respectively. It is concluded that both wind and temperature observations from Mode‐S MRAR are accurate and therefore potentially very useful for data assimilation in numerical weather prediction models.