Pedotransfer models for predicting bulk density values from measured soil properties

The purpose of this study was to determine the relationships between bulk density and some soil properties, and to develop a regression model to estimate bulk density values. In this study, generally fine textured 78 soil samples collected from the Samsun district were used. According to the results; except silt, pH, exchangeable natrium and lime content, the other soil properties were highly correlated with bulk density value (r=0.45-0.77). The direct effects of some soil properties on bulk density of the soils were obtained in the following order; silt+clay content, sand content, lime content, exchangeable Ca+Mg content, permanent wilting point and field capacity, cation exchange capacity and organic matter content. The Indirect effect of the sail properties were generally obtained in the fallowing order; silt+clay content, sand content, clay content and exchangeable Ca+Mg capacity. Then, the data set was statistically evaluated and PTF were generated by multiple linear regression method to estimate bulk density. In order to compare the measured bulk density values by traditional method and predicted bulk density values with generated PTF, determination coefficient (r2) was used, and the highest r2 was obtained as 0.67. The results of this research pointed out that generated pedotransfer models can be used with the aim of predicting bulk density in Samsun District