Numerical Mapping and Modeling Permafrost Thermal Dynamics across the Qinghai-Tibet Engineering Corridor, China Integrated with Remote Sensing

• Published in: Remote Sensing
• Main Authors: Guoan Yin, Hao Zheng, Fujun Niu, Jing Luo, Zhanju Lin, Minghao Liu
• Format: Text
• Language: English
• Published: Multidisciplinary Digital Publishing Institute 2018
• Subjects: numerical model; climate change; permafrost degradation; remote sensing; Qinghai–Tibet Plateau; Active layer thickness; permafrost
• Online Access: https://doi.org/10.3390/rs10122069

Permafrost thermal conditions across the Qinghai–Tibet Engineering Corridor (QTEC) is of growing interest due to infrastructure development. Most modeling of the permafrost thermal regime has been conducted at coarser spatial resolution, which is not suitable for engineering construction in a warming climate. Here we model the spatial permafrost thermal dynamics across the QTEC from the 2010 to the 2060 using the ground thermal model. Soil properties are defined based on field measurements and ecosystem types. The climate forcing datasets are synthesized from MODIS-LST products and the reanalysis product of near-surface air temperature. The climate projections are based on long-term observations of air temperature across the QTEC. The comparison of model results to field measurements demonstrates a satisfactory agreement for the purpose of permafrost thermal modeling. The results indicate a discontinuous permafrost distribution in the QTEC. Mean annual ground temperatures (MAGT) are lowest (<−2.0 °C) for the high mountains. In most upland plains, MAGTs range from −2.0 °C to 0 °C. For high mountains, the average active-layer thickness (ALT) is less than 2.0 m, while the river valley features ALT of more than 4.0 m. For upland plains, the modeled ALTs generally range from 3.0 m to 4.0 m. The simulated results for the future 50 years suggest that 12.0%~20.2% of the permafrost region will be involved in degradation, with an MAGT increase of 0.4 °C~2.3 °C, and the ALT increasing by 0.4 m~7.3 m. The results of this study are useful for the infrastructure development, although there are still several improvements in detailed forcing datasets and a locally realistic model.