Prediction of some soil chemical properties in tushki using empirical equations

Abstract

An empirical method depends on the multi-variables linear regression was established in this study, to estimate some soil chemical properties, and minimize the laboratory testing cost in Tushki area. Two main equations were driven from a standard laboratory soil testing results for 390 soil samples. A standard laboratory soil testing results for 50 soil samples were carried out to validate the resulted equations by using correlation test. The main equations have the first degree with multivariable form Y= a+ b<inf>1</inf> X <inf>1</inf>+b<inf>2</inf> X<inf>2</inf>+.... +b<inf>n</inf> X<inf>n</inf>,. It was possible to estimate Ca⁺⁺, Mg⁺⁺, Na⁺ and Cl^- using only Ec<inf>e</inf> with r² values of 0.75, 0.67, 0.85 and 0.82, respectively. When other cation was used in addition to Ec<inf>e</inf>, the correlation coefficient increased. It was also possible to estimate cation exchange capacity (CEC) and exchange sodium percentage (ESP ) using values of Na or Ca with r² of 0.99 and 0.82, respectively.