| Energy changes, water vapor exchange etc. in the land surface will affect the local weather, climate through the land-atmosphere interaction. Soil temperature which is playing an important part in the energy balance of land surface influences the interaction between earth and atmosphere significantly. Based on the above mechanism, we mainly carried out the following research work about the soil temperature algorithm:1. In this dissertation, we not only reviewed the three most important soil thermodynamic parameters:soil thermal capacity, soil thermal conductivity and soil thermal diffusivity, but also introduced three theories of soil thermal transport: thermal conduction, thermal convection and coupled thermal conduction-convection process. A large number of methods to estimate soil temperature have been developed. There are two typical algorithms. One is assuming that soil is vertically heterogeneous and took into account only thermal conduction process. The other is considering the vertical heterogeneity and taking into account the thermal convection effect on soil temperature estimate which is induced by the movement of liquid water in soil. We also focused on the analysis of the coupled thermal conduction-convection equations by using the Harmonic Method.2. Theoretical comparison between the two methods called Phase algorithm and Amplitude algorithm, which are based on the traditional heat conduction equation, and the Coupled conduction-convection algorithm based on the coupled conduction-convection equation were conducted. The results showed that the Phase algorithm overestimates the Amplitude of the modeled soil temperature, furthermore, the Amplitude algorithm overestimates the phase of the modeled soil temperature. So, we give our great attention to the coupled conduction-convection method is this dissertation.3. In order to verify the ability of simulating soil temperature of these three algorithms, we used two different soil temperature data sets collected at meadow soil in Litang site over Tibetan Plateau during 25 August to 24 September in 2006 and at a bare dessert soil site located at the Huazhaizi desert in Zhangye, Gansu Province during 10 July to 15 August in 2011. We used the three methods mentioned above to calculate the soil thermal diffusivity k, and the soil water flux density W which is obtained by the coupled conduction-convection algorithm. Then we modeled soil temperature at different depths according the obtained k and W. The results show thatPhase algorithm overestimates the soil temperature amplitude, but the Amplitude algorithm overestimates the phase compared with the results estimated by the Coupled conduction-convection algorithm which are in satisfactory agreement with measured data. For instance, the Phase method overestimates the soil temperature amplitude about 0.96℃, the Amplitude method overestimates the soil temperature phase about 0.45h compared to the observed soil temperature at the depth of 0.10m in Tibetan Plateau meadow soil. Fortunately, the coupled conduction-convection method only overestimates the phase 0.21h and the soil temperature amplitude about 0.79℃. Overall, the field experiment data sets verify the theoretical results, and the Coupled conduction-convection algorithm is more satisfying in simulating the soil temperature. |
