Assessing land use type impacts on soil quality: Application of multivariate statistical and expert opinion-followed indicator screening approaches

Abstract

Knowledge on the influence of land use practices on soil quality is crucial to develop soil-specific management practices. In order to quantify soil quality index of different land use types in Hintalo Wajerat district of northern Ethiopia, twenty-three biochemical and physical soil attributes were measured at 0–30 cm depth from cultivated land (CL), grass land (GL) and shrub land (SL). Principal component analysis (PCA), discriminant analysis (DA) and expert opinion (EO) were used for minimum data sets (MDS) selection, and six soil quality indices were generated using linear (LS) and non-linear (NLS) scoring methods. One-way analysis of variance (ANOVA) was used to identify members of total data set. Results indicated a significant (p < 0.05) difference in soil quality index among the land use types with a definite trend in the following increasing order of SL > GL > CL, reflecting the need for improved managements in CL soils. The estimated soil quality index using NLS indicator transformation and DA indicator selection approach (SQI_NLS_DA) best distinguished soil functions related to varying land use types. The findings of this study indicated the significant influence of land use practices on soil quality and SQI_NLS_DA is suggested for future soil quality studies in semi-arid tropical soils.