Volume 16, Issue 2 (Journal of Soil and Plant Interactions 2025)
2025, 16(2): 61-78 |
Back to browse issues page
1- Soil and Water Research Department, South Kerman Agricultural and Natural Resources Research and Education Center, Agricultural Research, Education and Extension Organization (AREEO), Jiroft , s.heydary@areeo.ac.ir
2- Soil and Water Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj
2- Soil and Water Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj
Abstract: (565 Views)
Background and Objective: The availability of micronutrients such as copper (Cu) and manganese (Mn) in high-pH calcareous soils is a key challenge for sustainable citrus production in arid and semi-arid regions. This research was conducted to predict the concentration of these elements in citrus leaves in southern Kerman and identifying the most influential soil properties affecting their availability.
Methods: Samples were collected from 40 orchards, with a focus on both soil and leaf samples taken during the spring growth season. We used standard methods to analyze soil properties and measure Cu and Mn concentrations in the leaves.
Results: High soil pH significantly hindered nutrient absorption, as the leaf Cu and Mn showed negative correlation coefficients of ‒0.41 and ‒0.33 with pH, respectively. However, the leaf Cu was positively correlated with both clay content (r = 0.32) and organic carbon (OC) content (r = 0.39), while the leaf Mn exhibited a negative correlation with soil salinity, EC (r = -0.29). Stepwise regression selected the soil pH, clay, and phosphorus as predictors for the leaf Cu (R2 = 0.36), and pH and EC as predictors for the leaf Mn (R2 = 0.28). To overcome the limitation of linearity, a multilayer perceptron artificial neural network (ANN) was developed using the Levenberg-Marquardt algorithm with a 70/30 training/testing ratio. The best ANN model for the leaf Cu, with inputs of soil pH, OC, clay, and phosphorus and 10 hidden neurons, achieved an R2 = 0.78 and RMSE = 0.88 mg kg−1 in the testing phase. For predicting leaf Mn, a network with inputs of soil pH, OC, clay, phosphorus, and EC and 11 hidden neurons achieved an R2 = 0.66 and RMSE = 5.68 mg kg−1.
Conclusion: The ANN models reduced prediction error by 42 and 38% for the leaf Cu and Mn compared to the regression model, respectively. This significant improvement demonstrates that artificial intelligence can be an effective tool for creating precise fertilization plans and ultimately enhancing the productivity of citrus orchards in these challenging environments.
Methods: Samples were collected from 40 orchards, with a focus on both soil and leaf samples taken during the spring growth season. We used standard methods to analyze soil properties and measure Cu and Mn concentrations in the leaves.
Results: High soil pH significantly hindered nutrient absorption, as the leaf Cu and Mn showed negative correlation coefficients of ‒0.41 and ‒0.33 with pH, respectively. However, the leaf Cu was positively correlated with both clay content (r = 0.32) and organic carbon (OC) content (r = 0.39), while the leaf Mn exhibited a negative correlation with soil salinity, EC (r = -0.29). Stepwise regression selected the soil pH, clay, and phosphorus as predictors for the leaf Cu (R2 = 0.36), and pH and EC as predictors for the leaf Mn (R2 = 0.28). To overcome the limitation of linearity, a multilayer perceptron artificial neural network (ANN) was developed using the Levenberg-Marquardt algorithm with a 70/30 training/testing ratio. The best ANN model for the leaf Cu, with inputs of soil pH, OC, clay, and phosphorus and 10 hidden neurons, achieved an R2 = 0.78 and RMSE = 0.88 mg kg−1 in the testing phase. For predicting leaf Mn, a network with inputs of soil pH, OC, clay, phosphorus, and EC and 11 hidden neurons achieved an R2 = 0.66 and RMSE = 5.68 mg kg−1.
Conclusion: The ANN models reduced prediction error by 42 and 38% for the leaf Cu and Mn compared to the regression model, respectively. This significant improvement demonstrates that artificial intelligence can be an effective tool for creating precise fertilization plans and ultimately enhancing the productivity of citrus orchards in these challenging environments.
Keywords: Machine learning, Soil properties, Calcareous soils, Nutrient management, Soil organic carbon.
Type of Study: Research |
Subject:
Modeling of soil-water-plant relations and root water uptake
Received: 2025/04/27 | Accepted: 2025/08/3 | Published: 2025/09/22
Received: 2025/04/27 | Accepted: 2025/08/3 | Published: 2025/09/22
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
Journal of Soil and Plant Interactions by http://jspi.iut.ac.ir is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License which allows users to read, copy, distribute and make derivative works for non-commercial purposes from the material, as long as the author of the original work is cited properly.
This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License which allows users to read, copy, distribute and make derivative works for non-commercial purposes from the material, as long as the author of the original work is cited properly.
