Abstract: Soil hydrothermal conditions affect plant growth, species distribution, and community stability. Here, a temperaturecontrolling experiment was performed using Carex duriuscula, a dominant species in the Mu Us bottomland. There were four soil temperature conditions, including control, warming Ⅰ, warming Ⅱ, and cooling. Plant height, spike number, and the contents of chlorophyll, malondialdehyde and soluble sugar of C. duriusculawere measured periodically. The results showed that: (1) Plant height was positively correlated with soil temperature before August, but negatively correlated with soil temperature after August.  (2) The contents of malondialdehyde and soluble sugar of C. duriuscula were positively correlated with soil temperature from the end of August to the beginning of September and negatively correlated with soil temperature from midSeptember to October. Warming or cooling treatment inhibited chlorophyll synthesis. Malondialdehyde and soluble sugar could be used as indicators of the effects of soil temperature on plant growth. (3) Regression models with soil temperature as an independent variable and plant growth parameters or physiological parameters as dependent variables were constructed. The exponential model of plant spike number and soil temperature could be used to predict the effects of soil temperature on the growth of C. duriuscula due to the high fitting degree of the model.

Key words: winter wheat, chlorophyll content, hyperspectral remote sensing, stationary wavelet transform, estimation model.