An improved correction method for high quality wind and temperature observations derived from Mode-S EHS

Upper air atmospheric wind and temperature information is crucial for numerical weather prediction and nowcasting. The current observation systems which are exploited to collect this information are radiosonde, aircraft, wind profilers, Doppler radar and satellites. A novel method to measure wind and temperature is related to tracking and ranging by an enhanced surveillance (EHS) Air Traffic Control (ATC) radar. This EHS radar interrogates in a selective mode (Mode-S) all aircraft in sight on which the aircraft replies with a message containing for example magnetic heading, airspeed and Mach number. From this information wind and temperature can be inferred. Since meteorological information is not directly measured in this way, preprocessing is necessary to obtain atmospheric information with adequate quality. Temperature is deduced from the Mach number and airspeed. The wind vector is deduced from the difference between the ground track vector and the orientation and speed of the aircraft relative to the air. The ground track is observed accurately but the aircraft orientation contains systematic errors and preprocessing steps are essential (de Haan, 2011). In this report the correction method is revisited taking into account the time dependency of the heading correction and, additionally, airspeed adjustment. The method developed in de Haan (2011) was limited by the fact that a heading correction value could only be estimated for aircraft regularly landing at Amsterdam Airport Schiphol. Using external sources of wind information, a heading and airspeed correction can be estimated for all aircraft when enough comparison can be generated. A dynamic heading correction database is created using the operational numerical weather prediction model from the European Center for Middle range Weather Forecast (ECMWF) and is cross validated with heading corrections derived using AMDAR wind information. This increased the number of observations by a factor of four in the Amsterdam Flight Information Region (FIR). The north-south and east-west component of wind speed observation is improved by 2 to 5% in standard deviation when both heading and airspeed corrections are applied. Next to the heading correction a flight phase dependent (ascending, descending, or constant level) temperature adjustment is developed by comparing the derived temperature against ECMWF. As it turns out, this adjustment has for some aircraft a time dependent character and thus also for this reason a dynamic temperature database has been developed. The resulting temperature bias is closer to zero, and standard deviation between model temperature and observation is reduced by 50% for all aircraft all together. For Mode-S EHS observations from AMDAR aircraft, the derived temperature observations show no improvement in standard deviation. Above 800 hPa, Mode-S EHS derived and AMDAR wind observations are comparable. Below 800 hPa, AMDAR wind observations exhibit an increase in standard deviation from 2 m/s at 800hPa to 2.5 m/s at 1000hPa when compared to ECMWF. Airspeed and heading corrected Mode-S EHS derived wind observations show to have a constant standard deviation (2 m/s) compared to ECMWF.