Characterizing the Performance of Tropospheric Airborne Meteorological Data Relay (TAMDAR) Observations Using Radiosondes and Other Aircraft Observations

Aircraft observations provide a promising means to fill in the gaps in the existing operational radiosonde network with additional in situ observations. Vertical profiles collected during aircraft ascent and decent capture the vertical structure of the troposphere, and near major airports profiles are observed with a far greater temporal density than that of radiosondes.

Although several studies have shown the quality and utility of data obtained from larger jet aircraft that participate in the global WMO Aircraft Meteorological Data Relay (AMDAR) program, less has been published about the robustness of observations available from Tropospheric Airborne Meteorological Data Relay (TAMDAR) equipped aircraft. TAMDAR is a private sector program that outfits primarily smaller passenger aircraft with a suite of instruments to measure temperature, dew point temperature, winds among other parameters. As these smaller aircraft typically fly routes of a few hundred kilometers or less and at lower altitude than AMDAR aircraft, they perform many ascents and descents over the course of a day and are often found at smaller airports without many other airborne observations, thereby filling gaps in our conventional upper-air observing network

This study provides a validation of TAMDAR observations (temperature, dew point temperature, and winds) against other in situ datasets, including radiosondes and other airborne observations, for the entirety of 2018. This study differs from previous validation studies in a number of ways, including 1) comparing to radiosondes with in-flight location data to eliminate biases associated with radiosonde drift, 2) comparing to non-TAMDAR aircraft observations [including Water Vapor Sensing System – II (WVSS-II) moisture observations], and 3) investigating TAMDAR-to-TAMDAR observation differences in order to characterize the uncertainty of the instrument independent of radiosonde errors.

Initial results show that, when adequately quality controlled, the lower-level TAMDAR observation profiles have good systematic agreement with radiosondes, with temperature and dew point biases around 0.5 K and 2 K respectively while wind bias is on the order of 3.5 m/s. Additional results characterizing the system’s hysteresis, level of random error and performance in different climate regimes will also be presented.