Abstract: Refractivity from clutter (RFC) refers to techniques that estimate the near real-time, regional and non-uniform tropospheric refractivity from radar clutter returns. Radar parameters, sea(ground) surface characteristics and spatio-temporal statistics of refractivity affect performance of RFC. In order to obtain better performance of RFC for evaporation duct in China sea areas, RFC estimation performance was computed in term of a root mean square error (RMSE) of estimation over prior statistics distribution when a radar frequency-antenna height combination is given.The estimation was achieved by particle swarm optimization (PSO) including adaptive objective function.The prior distribution in this paper were provided by advanced refractive effects prediction system (AREPS) belonging to air and sea warfare systems center in San Diego, USA.Through studying RFC estimation performance at different months of China different sea regions, the best possible combination of radar frequency-antenna height was summarized. The results would lay an important reference on ocean environment inspecting, design and application of radio system emplaced in China offshore.