Estimation of soil organic matter in mollisols based on artificial intelligence
Mollisols are a valuable natural resource, and their organic matter content can be used to evaluate soil fertility. Estimating and monitoring the soil organic matter (SOM) content of Mollisols is of great importance. This study employed an artificial intelligence method, based on deep neural network...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-12-01
|
| Series: | Ecological Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1574954125002328 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Mollisols are a valuable natural resource, and their organic matter content can be used to evaluate soil fertility. Estimating and monitoring the soil organic matter (SOM) content of Mollisols is of great importance. This study employed an artificial intelligence method, based on deep neural networks (DNNs), to predict SOM content. In this method, relevant measurement values of soil nutrients, such as phosphorus, nitrogen and potassium, and the soil pH were used as input features for the model. A dataset comprising 2490 samples was used for model training and testing. These samples were obtained through soil sampling and experimental measurements. This study validated the model by setting different ratios of training and testing datasets, and the results indicated that the proposed method can estimate the SOM content with an accuracy of nearly 95%. Furthermore, the method developed in this study was compared with six traditional machine learning methods and exhibited higher accuracy. This model will serve as the basis for designing realtime non-destructive testing of SOM. |
|---|---|
| ISSN: | 1574-9541 |