Applicability Analysis of River Ice Thickness Prediction Based on an Improved One‐Dimensional Thermodynamic Model

ABSTRACT Ice thickness is a key parameter reflecting the structural stability of ice covers and the magnitude of ice loads. Therefore, accurately predicting river ice thickness is essential for the safety assessment of bridge engineering and the formulation of emergency plans. To enhance the accuracy of ice‐thickness prediction, this study proposes an improved one‐dimensional thermodynamic model based on the traditional formulation, which explicitly accounts for the effects of longwave radiation flux and turbulent fluxes. To prove the reliability of the presented model, comparative analyses were conducted using in situ observations from the Songhua River against the Ashton model, the Ding model and the conventional one‐dimensional thermodynamic model. The results indicate that, compared with the measured ice thickness, the average absolute errors of the predicted ice thicknesses by the Ashton's thermodynamic model, Ding's thermodynamic model, traditional one‐dimensional thermodynamic model and the improved model are 4.72, 4.69, 21.46 and 3 cm, respectively, demonstrating that the proposed model achieves the highest prediction accuracy. The improved model demonstrates superior applicability, primarily due to the incorporation of the energy balance equation and heat conduction equation, which comprehensively consider the bidirectional heat transfer between the upper and lower boundaries of the ice layer. Overall, the findings of this study provide a robust methodological foundation for quantitatively describing river‐ice evolution in cold regions and offer scientific support for the safety assessment and emergency management of hydraulic and bridge structures in icy environments.