Abstract

Exploring hydrothermal behavior of soil of the Qinghai-Tibet Plateau is essential to uncover the mechanism of frequent freeze-thaw disasters triggered by warming climate. This study applies a hydrothermal coupled model for freeze-thaw soil to describe hydrothermal behavior, and proposes dynamic inverse modelling approach driven by multi-source time-series data to determine time-dependent hydrothermal parameters, where backpropagation neural network, genetic algorithm and orthogonal design were integrated. The proposed approach can achieve the precise determination of six time-dependent hydrothermal parameters including isothermal and thermal hydraulic conductivities for liquid water and water vapor, thermal conductivity, and volumetric heat capacity, which comprehensively reflect the coupling characteristics of soil influenced by complex external environment. Selecting a typical research area in seasonally frozen ground of the QTP, the proposed method was utilized to explore the hydrothermal properties during freezing period and evolution of thermo-hydraulic behavior within seasonally freeze-thaw layer under climate warming. The results indicated that it is suitable for the proposed approach to reasonably obtain the time-dependent hydrothermal parameters. With the 2.15 °C-increase in air temperature after 50 years, the most significant increment of 88.5% for surface heat flux induces the conspicuous rise of soil temperature within 1.1 m and evident shrink of 30-days for freezing duration. The most striking enhancement of soil temperature appeared at the depth of 0.15 m, which is up to 1.8 °C. However, the obviously decreasing hydraulic conductivities caused a weak reduction in water content within the depth of 0.70 m. Long-term warming climate ultimately leads to the thickening seasonally freeze-thaw layer, which is unfavorable for permafrost stability and slope stability. This study provides guidance for clarifying the mechanism of freeze-thaw disasters on the QTP under warming and humidifying climate.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call