Spatiotemporal dynamics of soil organic carbon at 30 m-resolution in Europe: responses to land cover stability and conversion
Bibliographic record
Abstract
High-resolution soil organic carbon (SOC) mapping can provide more detailed information on SOC distribution, enabling accurate assessment of the impact of small-scale land cover changes on SOC content within a region. This study provides the first high-resolution (30 m) spatiotemporal SOC maps for 23 European countries. A Voting Regressor (VR) model combining Random Forest (RF), Categorical Boosting (CatBoost), and Light Gradient Boosting Machine (LightGBM) was used with soil data from the Land Use/Cover Area frame statistical Survey (LUCAS) soil data and environmental factors to predict SOC. This study revealed the differential impacts of land cover conversion types on SOC content, thereby addressing a knowledge gap in quantitatively assessing the carbon effects of land cover changes at the mesoscale regional level. The VR model demonstrated good predictive accuracy, with an R 2 ranging from 0.57 to 0.75 and an average of 0.64. It showed a statistically significant improvement over the RF (ΔR 2 = +0.11), CatBoost (ΔR 2 = +0.10), and LightGBM (ΔR 2 = +0.12) models. Furthermore, clay content was identified as the most influential variable in all models. SOC levels exhibited a highly skewed distribution, with the vast majority of areas below 60.00 g kg −1 , while isolated extreme values above 300.00 g kg −1 were observed in regions such as northern UK, Ireland, the Scandinavian Peninsula, and the Alps. Despite an overall increasing SOC trend, regions such as north-western Ireland and the UK, southern Italy and Spain, north-western Greece, and northern Scandinavia saw continued SOC losses. Wetlands, grasslands, and forests—especially wetlands—demonstrated high carbon sequestration capacity, while agricultural land had the lowest SOC levels. Land cover conversions significantly influenced SOC: All land cover transitions or stable conditions resulted in statistically significant changes in SOC. Conversion from forest to agriculture caused the largest SOC loss (median decrease of 13.22 g kg −1 , P < 0.001), while land abandonment to grassland (increase of 11.14 g kg −1 , P < 0.001) and other conversions to grassland (increase of 19.25 g kg −1 , P < 0.001) both led to significant SOC gains.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".