Land Suitability | Hadi Shokati

This project integrates advanced multi-criteria decision-making (MCDM) techniques with machine learning to assess land suitability for wheat cultivation, providing actionable insights for sustainable agricultural planning and food security.

Overview

Land suitability assessment is critical for modern agriculture, involving the evaluation of soil properties, climate, topography, hydrology, and socio-economic factors. This study provides a comprehensive framework for determining optimal land use for wheat cultivation. By combining multiple MCDM approaches with Random Forest machine learning, we offer a robust methodology for agricultural decision-making that balances productivity and environmental sustainability.

Research Objectives

This study aims to:

Determine land suitability for wheat cultivation in western Iran by integrating MCDM and machine learning methods
Compare advanced MCDM techniques including TOPSIS, GRA, and SAW for land evaluation
Identify the most important factors influencing wheat cultivation suitability in the study area
Develop spatial prediction models using Random Forest and digital soil mapping techniques

Methodology

Study Area

The Gavshan region (5,341 hectares) in western Iran.

Data Collection

Soil Sampling and Analysis:

70 soil profiles selected using stratified random sampling based on geomorphology
Analysis of 10 soil properties and wheat yield measurements
Laboratory analysis: organic carbon (OC), pH, electrical conductivity (EC), cation exchange capacity (CEC), calcium carbonate equivalent (CCE), particle size distribution, available phosphorus and potassium

Field Data Collection:

Wheat yield sampling from 1 m² plots at each profile location
Harvest date: August 23, 2019
Uniform management practices across the study area

Multi-Criteria Decision Making (MCDM) Methods

Three advanced MCDM techniques were employed:

1. TOPSIS (Technique for Order Preference by Similarity to Ideal Solution):

2. GRA (Gray Relational Analysis):

3. SAW (Simple Additive Weighting):

Attribute Weighting:

Shannon’s entropy method used to determine objective weights
Reduces subjective influence of decision-makers
Accounts for information content and uncertainty in data

Machine Learning Modeling

Random Forest (RF) Algorithm:

Ensemble learning for spatial prediction of land suitability values
Environmental covariates from multiple sources:
- Terrain features from 10 m digital elevation model (slope, LS factor, topographic wetness, aspect, curvature)
- Landsat 8 OLI spectral bands and vegetation indices (30 m resolution)
- Geomorphological map (categorical data)

Comprehensive methodological framework showing the integration of field sampling, soil analysis, MCDM techniques, and machine learning modeling for land suitability assessment.

Key Findings

Critical Factors for Wheat Cultivation

Shannon’s entropy weighting revealed the most important attributes:

Slope (0.439 weight): Strongest influence on wheat suitability
Organic Carbon (OC) (0.135 weight): Key soil fertility indicator
Cation Exchange Capacity (CEC) (0.108 weight): Critical for nutrient dynamics

Other factors (in order of importance):

Calcium carbonate equivalent (CCE): 0.090
Electrical conductivity (EC): 0.086
Gravel content: 0.077
Soil thickness: 0.054
Soil texture: 0.010

Key Observations:

TOPSIS and GRA methods showed stronger correlation with wheat yield than SAW
TOPSIS efficiently determined land suitability across most study areas
SAW method, while simple, did not capture differences in land suitability as effectively

Random Forest Spatial Modeling

RF algorithm successfully mapped land suitability with high accuracy:

Method	RMSE	MAE	R²
SAW	0.061	0.003	0.88
GRA	0.072	0.005	0.83
TOPSIS	0.081	0.017	0.80

Spatial Distribution of Suitability

Good Suitability Areas (central region):

Characteristics: Slope < 5%, yield > 1.4 t/ha, CEC > 15 cmol+/kg, OC > 1%
Coverage: 33% (TOPSIS), 20% (GRA), 6% (SAW)
Geomorphological units: Pi111, Pi211, Pl111

Moderate Suitability Areas (central, southern, eastern regions):

Characteristics: Slope 5-10%, yield 1-1.4 t/ha, CEC 10-15 cmol+/kg, OC 0.5-1%
Coverage: 16% (TOPSIS), 22% (GRA), 29% (SAW)
Geomorphological units: Pi121, Pi212, Pl211

Poor Suitability Areas:

High slope areas (>10%)
Low OC and CEC values
Predominantly in hilly and mountainous terrain

Spatial Autocorrelation Analysis

Analysis confirmed the importance of key factors:

Slope: Strong negative spatial autocorrelation with wheat yield in high-slope areas (Hi111, Hi121, Hi131, Hi211, Mo111, Mo121)

OC and CEC: Positive spatial autocorrelation with yield in areas with high organic carbon and cation exchange capacity

MCDM Methods: Positive spatial autocorrelation with yields across the study area, with TOPSIS showing the strongest correlation

Practical Applications

Agricultural Planning Sustainable Land Management

Policy and Decision Support

(Nabiollahi et al., 2024)

Land suitability assessment, as an important process in modern agriculture, involves the evaluation of numerous aspects such as soil properties, climate, relief, hydrology and socio-economic aspects. The aim of this study was to evaluate the suitability of soils for wheat cultivation in the Gavshan region, Iran, as the country is facing the task of becoming self-sufficient in wheat. Various methods were used to evaluate the land, such as multi-criteria decision-making (MCDM), which is proving to be important for land use planning. MCDM and machine learning (ML) are useful for decision-making processes because they use complicated spatial data and methods that are widely available. Using a geomorphological map, seventy soil profiles were selected and described, and ten soil properties and wheat yields were determined. Three MCDM approaches, including the technique of preference ordering by similarity to the ideal solution (TOPSIS), gray relational analysis (GRA), and simple additive weighting (SAW), were used and evaluated. The criteria weights were extracted using Shannon’s entropy method. Random forest (RF) model and auxiliary variables (remote sensing data, terrain data, and geomorphological maps) were used to represent the land suitability values. Spatial autocorrelation analysis as a statistical method was applied to analyze the spatial variability of the spatial data. Slope, CEC (cation exchange capacity), and OC (organic carbon) were the most important factors for wheat cultivation. The spatial autocorrelation between the key criteria (slope, CEC, and OC) and wheat yield confirmed these results. These results also showed a significant correlation between the land suitability values of TOPSIS, GRA, and SAW and wheat yield (0.74, 0.72, and 0.57, respectively). The spatial distribution of land suitability values showed that the areas classified as good according to TOPSIS and GRA were larger than those classified as moderate and weak according to the SAW approach. These results were also confirmed by the autocorrelation of the MCDM techniques with wheat yield. In addition, the RF model showed its effectiveness in processing complex spatial data and improved the accuracy of land suitability assessment. In this study, by integrating advanced MCDM techniques and ML, an applicable land evaluation approach for wheat cultivation was proposed, which can improve the accuracy of land suitability and be useful for considering sustainability principles in land management.