Seasonal frozen soils accounts for 54% of Chinese total acreage,mostly distributed in Northeast China where snowing is frequent. Soil moisture under snow cover may interact with temperature, both affecting many processes of biogeochemistry. However, few studies have focused on soil moisture and temperature interaction under snow covers. Therefore, this study investigated the soil moisture and temperature characteristics under seasonal frozen soils in field experiments. The experiment was set up in an experiment field of Northeast Agricultural University, Harbin, China. It was located in south Songnen Plain. During November 8 of 2013-April 28, 2014, four treatments were designed including bare land, natural snow cover land, snow compacted land, and snow thickened land. Each treatment was repeated three times. In bare land, snow was removed by hand. In snow compacted land, snow density was manually increased to 0.256 g/cm3. In the snow thickened land, snow depth was increased but snow density was similar with natural snow cover land. During the experiment, soil temperature was measured by thermometers at 5, 10, 15 and 20 cm soil depth, and soil moisture was measured by time domain reflectometery at same depth. Soil moisture data were divided into two parts: one for model establishment and the other for model validation. Relationship between soil moisture and temperature was described by exponential models. Fractal dimensions were used to evaluate complexity of soil moisture and temperature. The results showed that: 1) The whole experiment duration could be divided into rapid freezing stage, stable freezing stage, and melting stage; On March 1, the freezing depth reached the highest value of 118 cm; 2) Soil moisture and temperature could be well described by exponential models with determination coefficient about 0.9 and relative errors less than 5% for the model establishment and relative error less than 3% for the model validation; 3) The complexity of soil moisture and temperature interaction was simpler during freezing phase than the melting phase since the relative errors of the former were less than that of the latter; The relative errors decreased with increasing snow density and thickness during the freezing phase, but the contrast was observed during the melting phase, indicating the interaction of soil temperature and moisture is greatly affected by freezing-thawing process; 3) During melting stage, soil moisture was 14.31%, 15.9%, and 16.91% for natural snow, snow compacted, and snow thickened treatments, respectively, and soil temperature ranged -5.9-5.3, -3.6-6.9 and -3.1-3.8℃ for natural snow, snow compacted, and snow thickened treatments, respectively, and the interaction between soil moisture and temperature weakened; The complexity of soil temperature and moisture interaction increased with snow cover; and 4) The fractal dimension of soil temperature under different treatments lied between 1.4149-1.6019 during the freezing phase and 1.4621-1.5775 during the melting phase, revealing a downward trend with snow thickness and compactness during the freezing phase but an opposite trend during the melting phase; The fractal dimension of soil moisture ranged from 1.5047 to 1.5973 and from 1.547 to 1.6607 during the freezing and melting phases, respectively, with a similar trend with soil temperature changes among different snow cover treatments. The results are helpful in predicting soil temperature and moisture in sowing stage, and revealing soil moisture and temperature dynamics.%为了研究季节性冻土区土壤水热时空分异特征，揭示冻融过程中的水热相互作用机理及其复杂性。在松嫩平原黑土区，以野外实测试验数据为基础，分别建立裸地、自然降雪、积雪压实和积雪加厚覆盖处理条件下20 cm土层土壤含水率和温度的耦合模型，并通过对比模型的预测效果研究土壤水热变化的复杂程度，采用差异性分析和基于小波变换的分形理论方法定量研究含水率与温度序列的变异波动性和分维指标，进而验证不同覆盖处理条件对于土壤水热空间变异复杂性的影响。结果表明：冻结期，积雪阻碍了环境因素对于土壤水热迁移过程的影响，土壤含水率和温度的耦合效果较好，并且预测值与实测值能够较好的吻合，裸地、自然降雪、积雪压实、积雪加厚处理条件下的相对误差分别为0.42%、0.31%、0.13%和0.06%，随着积雪厚度的增加和密度的增大，含水率和温度的差异性减弱，复杂程度逐渐降低；融化期，积雪覆盖区的融雪水入渗抑制了土壤温度稳定提升，含水率出现骤然升高的现象，自然降雪、积雪压实、积雪加厚条件下的含水率变化分别为14.31%、15.90%和16.91%，土壤温度变化范围分别为−5.9～5.3、−3.6～6.9和−3.1～3.8℃，二者的互作效应减弱，并且随着积雪覆盖量的增加，土壤的水热时空迁移复杂程度逐渐增强。同时，采用基于小波变换的分形理论研究土壤含水率和温度的时间序列复杂性精度较高、结果可靠。该研究对于揭示冻土区土壤水热迁移动态规律，准确预测春播期土壤温度和墒情，合理高效地利用松嫩平原的土壤水资源具有重要的理论价值和现实意义。
展开▼
