Peak-fitting based prediction of soil temperature according to soil moisture content in solar greenhouse

Abstract: Soil temperature and moisture directly impact on the root growth and seed germination of crops in greenhouse in cold area. Decreasing of soil moisture content would be beneficial for increasing soil temperature in greenhouse during cold situations. Considering low soil temperature that restricts activities of microorganisms in soil, optimizing of soil moisture content is essential to keep balance between soil temperature and activity of soil microorganisms. The objective of this research was to determine the authentic relationship between soil temperature and moisture content in greenhouse by establishing and validating of a soil temperature variation with soil moisture content equation using peak fitting analysis. All the works were carried out in a typical solar greenhouse from Dec 20, 2013 to Feb 20, 2014. We designed four test zones (2 m×2 m each) in the central part of the greenhouse, namely zone I (irrigated 10 h ago), zone II (irrigated 8 d ago), zone III (irrigated 33 d ago), and zone IV (control). To monitor the soil temperature, three soil moisture/temperature sensors were placed in soil of each test zone with 0, 15 and 30cm depth, respectively. For control zone, a soil moisture/temperature sensor was placed on the soil surface (0 cm). We tested the moisture content and temperature variations of different soil depth. According to a Single Peak Fitting Module (PEM), we obtained the Extreme Fitting Function of the diurnal change of soil temperature. The relationship between temperature-variation equation and the moisture content was determined by calculating the area under temperature-variation curve and analyzing coefficients of the equation. The extreme function which has specific physical meaning of parameters was named as Estimated Extreme function. The reliability and validity of Peak Fitting Method were confirmed by error analysis of certain moisture content in different weather condition on the surface of the control plot. We investigated the change of soil temperature under different moisture content in greenhouse by using Peak Fitting Method. The results showed that the condition of strong evaporation can increase soil temperature and resulted in enhancing of unsaturated hydraulic conductivity of soil and the suction on wet evaporate face, leading to increase of liquid water level to the dry-wet surface. Using this method to simulate the temperature variation curve, the minimum matching degree was 0.9590. Based on the calculation of enveloping area of temperature curve, we obtained the average soil temperature and found out that decrease of soil moisture content caused an increase of average temperature of soil in both day and night. According to the coefficients of fitting equation established by the data from sunny day, we found that amplitude, center and width factors decreased while offset substantially increased along with the decrease of soil moisture content on the soil surface. However, offset, center and width factors increased while amplitude decreased in soil at 15 cm depth, and all four factors increased in the soil at 30 cm depth when the soil moisture content decreased. These results indicated that drop of soil moisture content can result in decrease of the regenerative capacity of soil and content of liquid water, and down shift of dry-wet surface. Validation results showed that the minimal discrepancy of temperature was 0.71℃, implying that the four factors of Estimated Extreme function were good enough to calculate the change of soil temperature. The method developed in this work is able to improve the efficiency by decreasing the measurement of soil temperature and convert the discrete data points into a continuous function.