Machine learning models for predicting soil particle size fractions from routine soil analyses in Quebec
Bibliographic record
Abstract
Abstract Soil texture and particle size distribution are important soil properties for understanding and interpreting the multiple processes and interactions in soil agrosystems. However, soil particle size analysis is rarely included in routine soil laboratory analyses due to cost but can be derived from other soil physico‐chemical properties and from more powerful regressions using machine learning techniques. The performance of multiple linear regression and four machine learning algorithms (MLAs)—K‐nearest neighbor (KNN), Random Forest, extra‐gradient boosting (XGBoost), and multilayer neural network (NeuralNetwork)—was compared to predict particle size fractions of clay, sand, and silt using routine analyses of soil pH, cation exchange capacity, Mehlich‐3 extracted elements, and density of dried sieved soil from 8,364 soil samples distributed across Quebec. Particle size fractions were predicted as compositional data using isometric log ratio transformation. The XGBoost model performed best, with RMSE values of 7, 10, and 12% and R 2 values of .77, .57, and .73 for the prediction of clay, silt, and sand fractions, respectively. Feature importance classification varied from one model to another, but the best predictors remained the same regardless of the model used. Sieved soil density measured in the laboratory and Mehlich‐3 Mg, K, Fe, and Zn ranked as better predictors of particle size fractions. Predicting soil particle sizes using routine soil physico‐chemical properties and MLAs appears an option for incomplete legacy datasets and a promising economic alternative to current methods carried out in commercial laboratories.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".