​The objective of this study is to identify salt accumulation on the topsoil of arid and semiarid regions. To achieve this goal, thirty-four topsoil samples were analyzed for salt accumulation indicators such as ECe and water leachable ions from the soil. The data were evaluated statistically using Pearson correlation analysis, principal component analysis (PCA), stepwise regression analysis (SWR), and clustering analysis utilizing a K-means technique, which included soil chemical markers as well as spatial salinity behavior. High collinearity (condition number 2160) exists within variables in the obtained data matrix (43 samples; 13 chemical predictors), necessitating a more flexible K-means approach for soil grouping. Pearson analysis indicated that Cl− positively contributed (R2 0.9781) to the total salinity, which was attributed to its ionic conductance and high concentration among leached ions. The PCA analysis indicated the high colinearity among variables, especially those that contribute to salinity like Cl−, Na+, and Ca2+. Based on PCA analysis, the 43 samples were clustered into five groups: three groups have high salinity (15–37 mS/m) but with unequal proportions of high-conductance ions (Cl−, Na+, Ca2+) and two groups of low salinity (1–11 mS/m) but with unequal proportions of low-conductance ions (Fe and SO42−). Stepwise regression analysis was helpful in removing unnecessary chemical variables and predicting the total salinity (mS/cm) from Cl− (0.076), SO42− (0.047), and Na+ (0.022) concentrations (meq/L). The spatial analysis reveals that the salinity effect is heterogeneous and is derived from the original source of the parent material or impacted by the anthropogenic effects.​