Advanced Algorithm and System Development for Cassini Radio Science Tropospheric Calibration

An important error source for the calibration of tropospheric delay variations at microwave frequencies is the “retrieval error” that is due to the uncertainty in conversion from observables (sky brightness temperatures and surface measurements) to path delay. A large database of vertical water-vapor density profiles (from lidar measurements) and vertical temperature profiles (from a radio acoustic sounding system (RASS)) has been used to quantify the expected retrieval error as a function of time scale, instrumental configuration, and measurement accuracy. One key result is that a three-channel (e.g., 22.2-, 23.8-, and 31.4-GHz) water vapor radiometer (WVR) and a two-channel microwave temperature profiler (MTP) are nearly optimal and are needed in order to meet the accuracy goals for the Cassini Gravitational Wave Search Experiment (GWE). A second key result is that model-based Bayesian inversion techniques provide substantially better accuracy than do statistical retrieval methods. The estimated retrieval error for the measurement of path delay variations (using Bayesian methods and the above instrumental configuration with targeted stability) is a factor >6 smaller than the total GWE tropospheric calibration requirement at time scales >5000 s. At a 1000-s time scale, the error is a factor of ∼2 smaller than the GWE requirement. However, the retrieval error is approximately equal to the GWE requirement at a time scale of 100 s.