全球变化知识资源中心

The thermal regime of the active layer temperature (ALT) is a key variable with which to monitor permafrost changes and to improve the precision of simulations and predictions of land surface processes. The dynamics of the active layer thermal regime can differ substantially under various land surface types and climatic conditions. The proper simulation of these different processes is essential for accurately predicting the changes in water cycles and ecosystems under a warming climate scenario. In this paper, an artificial neural network (ANN) forecasting model system was developed using only two accessible parameters, air and ground surface temperatures, to predict and simulate the ALT thermal regime. The model results show that the ANN model has better real-time prediction capability than other physics-based models and performs well at simulating and forecasting variations in soil temperature with a step size of 12days in permafrost regions on the Qinghai-Tibetan Plateau. The influence of an increase in air temperature on the ALT thermal regime was more intense during the thawing process than during the freezing process, and this influence decreased with an increase in soil depth.