Soil Organic Carbon Measurements Influences FT-NIR Model Training in Calcareous Soils of Saskatchewan
Bibliographic record
Abstract
This study compares acid digestion and temperature ramping methods for obtaining soil organic carbon (SOC) reference data to train Fourier Transform Near Infrared (FT-NIR) models in carbonate-rich Saskatchewan agricultural soils. FT-NIR spectra were measured on soil samples (n = 431) from carbonate-rich Dark Brown Chernozem soil, with quantification of inorganic and organic carbon. Spectra were transformed using continuous wavelet transform and analyzed using cubist regression tree models. Models were built using a 70:30 train-test split validation approach. Spectral feature selection, wavelet scale, model and hyperparameter optimization were conducted using 5-fold cross-validation analysis on the training dataset. All validation metrics were calculated using the testing dataset. The temperature ramping method identified outliers with soil inorganic carbon (SIC) greater than 1.5%, which were not detected using the acid digestion method. SOC and SIC prediction accuracy was higher using temperature ramping data (Coefficient of determination: R2 = 0.66 and 0.63, Lin's concordance: ccc = 0.78 and 0.77) compared to acid digestion data (R2 = 0.44 and 0.42, ccc = 0.64 and 0.62). Total carbon (TC) prediction accuracy was similar for both methods (R2 = 0.58, ccc = 0.71). Removing samples with high carbonate (SIC > 1.5%) improved SOC and TC prediction accuracy using temperature ramping data (R2 = 0.70, ccc = 0.81 for SOC; R2 = 0.64; ccc = 0.75 for TC), but not when using acid digestion method. This study suggests that high carbonate content may negatively affects SOC model accuracy, especially when relying upon acid digestion methods for reference SOC data.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.009 | 0.003 |
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 source (direct Gemma or distilled Codex), 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".