Abstract
Purpose: In the context of developing interest in regenerative farming to enhance soil health and wider ecosystem service benefits, this study aimed to evaluate the potential for using a portable gamma sensor (Medusa MS-700) to survey soil health parameters in agricultural fields.
Materials and methods: A Medusa MS-700 portable gamma spectrometer was deployed to evaluate activity concentrations of 40K, 238U, 232Th and 137Cs across agricultural fields. In parallel, 90 soil samples were collected for laboratory analysis of SOM by loss on ignition (LOI), total C (TC) and N (TN), particle size, moisture content and bulk density. Linear regression was used to derive prediction models of soil parameters based upon radionuclide concentrations.
Results and discussion: Correlation between soil parameters and radionuclide concentrations showed negative associations between radionuclides and moisture content, LOI, TC and TN, and most notably for 40K (rs > − 0.8). This is likely related to (i) the diluting effect of organic matter in relation to the gamma-emitting parent material and (ii) the attenuating effect
of soil moisture upon gamma rays. Correlations with texture properties were generally less strong with coefficients < 0.7. Linear regression prediction models returned mean absolute errors (MAE) of around 1% TC and 0.1% TN.
Conclusion: Findings show the potential for using this portable gamma sensor for rapid spatial analysis of TC and TN across agricultural land units. Considering the gamma sensor as a decision support tool, we discuss the potential for rapid detection across wide spatial scales to inform targeted measures for regenerative action. From a land management perspective, prediction models from this dataset are encouraging wherein site-specific validation of sensor data is essential.
Materials and methods: A Medusa MS-700 portable gamma spectrometer was deployed to evaluate activity concentrations of 40K, 238U, 232Th and 137Cs across agricultural fields. In parallel, 90 soil samples were collected for laboratory analysis of SOM by loss on ignition (LOI), total C (TC) and N (TN), particle size, moisture content and bulk density. Linear regression was used to derive prediction models of soil parameters based upon radionuclide concentrations.
Results and discussion: Correlation between soil parameters and radionuclide concentrations showed negative associations between radionuclides and moisture content, LOI, TC and TN, and most notably for 40K (rs > − 0.8). This is likely related to (i) the diluting effect of organic matter in relation to the gamma-emitting parent material and (ii) the attenuating effect
of soil moisture upon gamma rays. Correlations with texture properties were generally less strong with coefficients < 0.7. Linear regression prediction models returned mean absolute errors (MAE) of around 1% TC and 0.1% TN.
Conclusion: Findings show the potential for using this portable gamma sensor for rapid spatial analysis of TC and TN across agricultural land units. Considering the gamma sensor as a decision support tool, we discuss the potential for rapid detection across wide spatial scales to inform targeted measures for regenerative action. From a land management perspective, prediction models from this dataset are encouraging wherein site-specific validation of sensor data is essential.
Original language | English |
---|---|
Pages (from-to) | 2556-2563 |
Journal | Journal of Soils and Sediments |
Volume | 23 |
Issue number | 6 |
Early online date | 18 Mar 2023 |
DOIs | |
Publication status | Published - Jun 2023 |
Keywords
- Portable gamma spectrometry
- Soil health
- Soil organic carbon
- Sustainable land management