Abstract: Surface heat and water vapour fluxes have an important and extensive influence on global climate change and atmospheric circulation, while Monin-Obukhov(M-O) similarity function plays an important role in calculating surface layer heat and water vapour fluxes and it is an essential factor in estimation model of atmospheric optical turbulence. Therefore, it is necessary and urgent to propose a reasonable, scientific and effective similarity function. The meteorological datasets at the 35 m tower in the western suburbs of Hefei was analysed and a new set of M-O similarity functions are fitted by use of non-linear least squares method, which shows good consistency with the previous similarity functions. At the same time, the functions were applied to estimate atmospheric optical turbulence, the estimated values have a good agreement with the measured values. The analysis results show that under stable and very unstable conditions, the functions are similar, however, this similarity breaks down weakly unstable conditions.