Soil Profile Data and Minimum Dataset Analysis for Relative Soil-Condition Assessment in Southwestern Nigeria
Description
This dataset contains the cleaned soil-profile observations and analytical outputs supporting the manuscript “Can a Reduced Soil Quality Index Represent Soil Condition? A Data-Driven Minimum Dataset Approach for Tropical Soils.” The dataset comprises 41 soil-horizon observations from nine pedons representing three study locations in southwestern Nigeria. The accompanying analysis includes the standardized analytical dataset, principal component analysis, minimum dataset selection, PCA-informed indicator weights, relative indicator scores, and reduced soil-condition index calculations. Eighteen non-redundant soil indicators were initially evaluated. Seven principal components with eigenvalues greater than one explained 78.97% of the standardized variance. The final minimum dataset comprised soil organic carbon, available phosphorus, clay, pH, magnesium, calcium and iron. The dataset is provided to support transparency, reproducibility and independent reuse of the analyses reported in the associated manuscript.
Files
Steps to reproduce
Soil samples were collected from three study locations in southwestern Nigeria, representing nine soil pedons and 41 sampled soil horizons. Soil profiles were described in the field, and samples were collected from identified genetic horizons following standard soil-profile sampling procedures. The samples were air-dried, gently disaggregated, and passed through a 2-mm sieve before laboratory analysis. Soil bulk density, particle-size distribution, soil pH, electrical conductivity, soil organic carbon, total nitrogen, available phosphorus, exchangeable bases, exchangeable acidity, cation-exchange properties, and selected micronutrients were determined using standard soil analytical procedures. The resulting laboratory observations were compiled at horizon level. The dataset was checked for completeness and consistency before analysis. Variables that were mathematically derived or compositionally redundant were excluded from the initial multivariate analysis. Eighteen non-redundant soil indicators were retained: bulk density, clay, silt, pH, electrical conductivity, organic carbon, total nitrogen, available phosphorus, potassium, sodium, calcium, magnesium, exchangeable aluminium, exchangeable hydrogen, iron, manganese, zinc and copper. The 18 variables were standardized to zero mean and unit variance and subjected to principal component analysis (PCA). Sampling adequacy was evaluated using the Kaiser-Meyer-Olkin measure and Bartlett’s test of sphericity. Principal components with eigenvalues greater than 1 were retained. Seven components satisfied this criterion and together explained 78.97% of the standardized variance. A minimum dataset was then established using the PCA results together with the objective of retaining non-redundant indicators that represented major dimensions of soil variation. The final seven indicators were soil organic carbon, available phosphorus, clay, pH, magnesium, calcium and iron. PCA-informed weights were calculated for the selected indicators, and each indicator was transformed into a relative unitless score using the direction of its expected relationship with soil condition. The weighted soil-condition index was calculated as the sum of the weighted indicator scores. An equal-weight formulation was also calculated as a sensitivity analysis. Pearson and Spearman correlation coefficients were used to evaluate the agreement between the PCA-weighted and equal-weight formulations. The accompanying Excel workbook contains the cleaned analytical dataset and the analytical outputs, while the CSV file provides the cleaned data and calculated soil-condition index values for independent inspection and reuse.
Institutions
- Federal University Oye EkitiEkiti State, Ikole-Ekiti