It is worth highlighting that, for the first time to the best of our knowledge, vertical profiles of atmospheric parameters and $C_n^2$ were measured at Lhasa, south of the Tibetan Plateau, using balloon-borne radiosondes. Based on the measurements, two new statistical models (Lhasa HMN and Lhasa Dewan) for estimating turbulence strength are proposed. Attention has been paid to evaluate the reliability of the two models to reconstruct vertical profiles of $C_n^2$ from a statistical perspective. The statistical analyses presenting the Lhasa HMN model are accompanied with lower bias, root mean square error (RMSE), and bias-corrected RMSE ($\sigma$) than those of the Lhasa Dewan model, which implies the Lhasa HMN model can better reveal the nature of turbulence characteristics of Lhasa influenced by unique local weather conditions. In addition, the comparison between the Lhasa HMN model and measurements in calculating integrated astroclimatic parameters is carried out, and the result suggests that the performance of the Lhasa HMN model is reliable and satisfactory. The new reliable $C_n^2$ model offers new insight into the characteristics of optical turbulence at Lhasa and provides support for pursuing astronomical site selection in the Tibetan Plateau.