SOIL Studies

SOIL Studies

2021, Vol 10, Num, 1     (Pages: 068-077)

Determination of Regression Models Between Thermal Properties of Soils and Some Physical and Chemical Properties

İmanverdi EKBERLİ 1 ,Coşkun GÜLSER 1 ,Orhan DENGİZ 1

1 Ondokuz Mayıs Üniversitesi, Ziraat Fakültesi Toprak Bilimi ve Bitki Besleme Bölümü, Samsun/Türkiye

DOI: 10.21657/topraksu.885688
Viewed: 1371
 - 
Downloaded : 595

The formation of the temperature field in the soil and the realization of heat transfer are related to the change of the thermal properties of the soils. Thermal properties vary depending on climate factors and soil properties. In this study, regression models were created between the thermal properties of soils such as volumetric heat capacity, thermal diffusivity and thermal conductivity coefficients and some physical and chemical soil properties (EC, OM, Clay, Silt, Sand, Db, θ ) that could be determined more easily experimentally. The statistical significance level (p = 0.001) and regression coefficient (R2 = 0.76) of the regression model made between the volumetric heat capacity of soils and the properties of EC, OM, Clay, Silt, Sand and Db were determined to be high. Adding volumetric moisture content (θ ) to independent variables increased the performance of the models (R2 = 0.77-0.99); the statistical significance level (p = 0.000) and the regression coefficient of the regression model including θ, EC, OM, Clay, Silt, Sand, Db properties were found to be very high. The thermal diffusion coefficient of soils and the regression models between EC, OM, Clay, Sand, Silt, Db, θ properties were determined as R2 = 0.76 and 0.79 (p = 0.000 and 0.002). The expression of regression models with polynomials including the square, square root and product of soil properties increased the performance of the models; the model including Clay, θ, Clay2, θ, √OM , (EC·Db), √Clay , OM, OM2, Db soil properties showed high level of statistical significance (p = 0.001) and was characterized by a higher regression coefficient (R2 = 0.90). The performance of the regression model between the thermal conductivity coefficient and Clay, Silt, Sand, Db, θ soil properties is high (R2 = 0.71; p = 0.001); the performance of the model among the parameters θ, Sand, Clay, Silt, Db, EC2, OM2, Db2, θ2 , EC, √OM, was determined to be very high (R2 = 0.91; p = 0.012). It seems possible that the regression models obtained can be applied in estimating the thermal properties of soils.


Keywords : Volumetric heat capacity, Thermal diffusion coefficient, Thermal conductivity coefficient, Soil properties, Regression model