A Novel Method for Soil Organic Matter Determination by Using an Artificial Olfactory System.

Area of Science:

• Agricultural Science
• Environmental Science
• Sensor Technology

Background:

• Soil organic matter (SOM) is crucial for soil fertility and nutrient content.
• Accurate and efficient methods for measuring SOM are essential for sustainable agriculture.
• Traditional methods for SOM analysis can be time-consuming and resource-intensive.

Purpose of the Study:

• To design and evaluate a novel method for measuring soil organic matter (SOM) using an artificial olfactory system (AOS).
• To compare the predictive performance of different regression models for SOM estimation.
• To establish a new, potentially more efficient, approach for assessing soil organic matter.

Main Methods:

• An artificial olfactory system (AOS) with 10 gas sensors at varying temperatures was employed to capture soil gases.
• Key features were extracted from sensor response curves, including maximum value, mean differential coefficient, response area, and transient value.
• Regression models, specifically Back-Propagation Neural Network (BPNN), Support Vector Regression (SVR), and Partial Least Squares Regression (PLSR), were developed for SOM prediction.

Main Results:

• Support Vector Regression (SVR) achieved the highest prediction accuracy for SOM, with a coefficient of determination (R²) of 0.895 and a ratio of performance to deviation (RPD) of 3.003.
• Back-Propagation Neural Network (BPNN) and Partial Least Squares Regression (PLSR) also showed good predictive capabilities, with R² values of 0.880 and 0.808, respectively.
• The developed AOS method, particularly with SVR, proved effective in accurately predicting soil organic matter content.

Conclusions:

• The artificial olfactory system (AOS) provides a promising new methodology for the rapid and accurate assessment of soil organic matter (SOM).
• Support Vector Regression (SVR) is identified as the most effective model for predicting SOM using the AOS data.
• This research offers innovative tools for soil analysis, contributing to improved soil fertility management and agricultural practices.