An improved evaporation duct RSHMU model and prediction performance analysis

Aiming the problem that the RSHMU evaporation duct model is too sensitive and can’t be used conveniently under the stable condition, some improved methods introduced BH91 expression, CB05 expression and SHEBA07 expression are given in this paper. The sensitivity analysis results under different wind speed and relative humitivity show that changing the profile function does not improve the sensitivity of evaporation duct height diagnosis. Based on the revised method given by Paulus, an improved prediction method called P-RSHMU evaporation duct model is presented which effectively solves the unreasonable diagnosis results under the stable condition compared with the original model. Compared with the measured data of gradient tower in Bohai sea, it is shown that the average error of the modified model is −1.02 m and the root mean square error is 1.49 m in stable conditions. The modified model performs better than the original RSHMU model significantly.