Quantitative Study On The Process Of Seasonal Frozen Soil Freezing And Thawing During Snowmelt Period

The northern slope of Tianshan Mountains River basin is rich of snow in Xinjiang. Because of the presence of a large number of snow melting and the incompletely melted seasonal frozen soil, the snowmelt floods are often caused in this area in spring. In order to reduce and avoid unnecessary losses, a set of completing and accurating snowmelt flood forecast and warn pattern which is suitable for this area and this season and it should be established. The research team has already established a short-term snowmelt hydrological forecasting pattern for small and medium-scale in arid and semiarid regions. But the forecasting flood peak discharge is often less than the observed value. That mainly due to the permafrost module of model is not perfect. However, the quantitative study of seasonal frozen soil with the mechanism can provide strong support for the improvement of the module. The primary object of this thesis is the freezing and thawing process of seasonal frozen soil during the period of snow melting in the JunTangHu river basin on the northern slope of Tianshan Mountains. And a closed small watershed in the middle of the study area is the experimental field. The freezing and thawing process of soil during the whole observation period from March 20, 2014 to November 21, 2013, were analyzed and discussed in this thesis. As well the characteristics of the freezing thawing process of seasonal frozen soil during the snowmelt period from March 1, 2014 to March 20, 2014.After that, the factors that were used to quantitatively characterize the freezing and thawing process of the seasonal frozen soil during the thawing period were selected. On this basis, the field observation experiment and indoor quantitative simulation experiment of the typical research area were combined, and then the required variables and methods for modeling were selected and optimized respectively. At last, a regression model of freezing depth of different situations and stages of freezing snow was established by using multiple linear regressions finally. The main results of this thesis are as follows:(1)The soil in the experimental field of the study area is silty loam, and its freezing temperature is between-0.5 ℃ ~-1.6 ℃.(2)Through the whole observation period, soil freezing period lasted nearly four months; melting period lasted for nearly half a month. And the soil freezing and thawing process can be divided into five stages, including unstable, fast and stable freeze, and rapid melting period, melt anaphase.(3) It is shown that the freezing progress of soil is in a single direction in the study area, and it freezes from the surface downwards to the bottom. While the thawing process is progressing in two directions at the same time, which from the surface towards to the deep and from the deep to the surface.(4)The freezing depth process of seasonal frozen soil can be divided into three stages during snowmelt period, which the same as change process of the air temperature, soil temperature and humidity, and the three stages are basic on time in sync. It illustrates that, in the snowmelt period, the change of the seasonal frozen soil freezing depth and the change of air temperature, soil temperature and soil moisture content are closely related. The freezing depth of seasonal frozen soil is a reflection of the combined effect of these three. There is a great correlation between freezing depth and the negative accumulated temperature. So frozen depth changes can be used to quantitatively study the seasonal frozen soil freezing and thawing process.(5) Stephen equation does not suitable for the frozen depth calculation. So eventually, the final multiple linear regression was chosen, after that, soil temperature, soil moisture content are selected as self variable. At last, four multivariate linear regression equation are set up, which suitable for different situations, different stages frozen depth changes. Then according to the observation data of inside and outside to test the model, and their simulation results are satisfactory, but they still need further improve and revise to meet the practical requirements.