Enhancing soil profile analysis with soil spectral libraries and laboratory hyperspectral imaging
Bibliographic record
Abstract
• Hyperspectral imaging technique provides a powerful tool for the detailed and dynamic analysis of SOC. • Continuum-removal method was used to select representative samples from SSL. • PLSR is neither robust nor recommended for fine-scale SOC mapping with hyperspectral imaging. • Local model development using a random forest algorithm was suggested because it performs a reasonable SOC map in profile. Soil visible-near-infrared (vis–NIR) spectroscopy offers a rapid, uncontaminated, and cost-efficient method for estimating physicochemical properties such as soil organic carbon (SOC). The development of soil spectral libraries (SSLs) and localized modeling methods has significantly improved the selection of appropriate modeling sets from SSLs for soil analysis. Nevertheless, most studies assume that the SSLs sufficiently cover the target samples for prediction. This study challenges this assumption by investigating the feasibility of using an SSL to predict SOC accurately in a local area when the dataset to be predicted (156,800 samples) vastly exceeds the SSL capacity (3755 samples). We utilized 1-meter-deep whole-soil profile and employed spectral similarity and continuum-removal (SS-CR) calculation to construct a Local dataset from the SSL, with a Global subset serving as a baseline for comparison. The effectiveness of partial least-squares regression (PLSR) and random forest (RF) algorithms in establishing quantitative relationships between spectra and SOC content was evaluated. Our results demonstrated that the Local model, with significantly fewer samples (1116), achieved higher predictive accuracy than the Global model. Both Global ( R 2 = 0.80, RMSE = 0.74 %) and Local ( R 2 = 0.83, RMSE = 0.75 %) models, developed using the RF algorithm, not only exhibited excellent accuracy but also enabled detailed and cost-effective characterization of the spatial distribution of SOC. Thus, leveraging SSLs enhances the cost-efficiency and predictive capacity of vis–NIR spectral analysis, particularly in handling large datasets at a local scale, underscoring the value of localized approaches in soil spectroscopy.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".