Modeling and Digital Mapping of Soil Quality Index using Interpolation and Machine Learning

Articles in Press

Document Type : Research Paper

Authors

1 Soil and Water Engineering Department,, Agriculture Faculty, Ilam University

2 Soil and Water Department, Agriculture Faculty, Ilam University, Ilam, Iran

3 Soil and Water Research Institute, Agricultural Research, Education and Extension Organization, Karaj, Iran.

4 PhD in Soil Resource Management, Department of Soil Science, Faculty of Agriculture, University of Tehran, Karaj. Iran

Abstract

Soil quality assessment is a key tool for evaluating the sustainability of agricultural and natural resource systems. This study aimed to model and digitally map the soil quality index (SQI) using machine learning algorithms and spatial interpolation methods in Ilam Province, western Iran. A total of 76 surface soil samples (0-20 cm) were collected using the conditioned Latin hypercube sampling (cLHS) approach, and 10 physicochemical properties were measured. Principal component analysis (PCA) was applied to identify the minimum data set (MDS), from which five variables were selected. Soil quality was assessed using the Integrated Quality Index (IQI) and Nemoro Quality Index (NQI) for both the total data set (TDS) and minimum data set (TDS) and MDS under linear and nonlinear scoring methods. Spatial prediction of SQI was performed using ordinary kriging (MDS) under linear and nonlinear scoring functions. Spatial prediction of SQI was performed using ordinary kriging (OK) and inverse distance weighting (IDW), as well as two machine-learning models: random forest (RF) and k-Nearest Neighbors (k-NN). Results indicated that the soils of the study area fall within a low quality class due to low organic carbon and available phosphorus, along with high pH, sand content, and bulk density. Among interpolation methods, OK outperformed IDW, exhibiting higher accuracy and lower prediction error. RF showed the highest performance for most SQI computations, whereas k-NN performed best for NQITDS. Variable importance analysis revealed that topographic factors were the most influential predictors of soil quality.

Keywords

Main Subjects

Iranian Journal of Soil and Water Research

Articles in Press, Accepted Manuscript
Available Online from 18 April 2026

Files

Share

How to cite

Statistics