A machine learning-based framework for spatio-temporal extension and filling of SMOS surface soil moisture observations over Canada
Bibliographic record
Abstract
Advancements in global satellite missions have revolutionized the assessment of Surface Soil Moisture (SSM) at global to local scales. However, spatio-temporal data discontinuities in specific regions remain a challenge. This study proposes a Machine Learning (ML)-based framework to extend the Soil Moisture Ocean Salinity (SMOS) SSM product both in the spatial and temporal domains, over Canada. In the first phase of the proposed framework, ML models based on Random Forest (RF) and Convolutional Neural Networks (CNN) are trained and validated with SMOS SSM as target and SSM-relevant climatic variables and geophysical variables, obtained from fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis data (ERA5), for the 2011–2020 period, as predictors. Developed models, when tested on unseen data for the years 2021–2022, suggest slightly better performance of the RF model compared to CNN, with root mean square error (RMSE) of 0.033 and 0.056 respectively; prediction biases mostly noted for regions with large inter-annual variability. The spatial filling of SSM for grid cells that were excluded during the training process, with similar land types as those in the SMOS training data, yields reasonable performance, with RF (RMSE = 0.013) performing better than CNN (RMSE = 0.064). In the second phase, the RF model is selected to extend the SMOS dataset for the 2008–2010 period. The temporal correlation between extended SMOS and ASCAT (Advanced Scatterometer) SSM demonstrates a reasonable association, with correlation coefficient exceeding 0.6. Additionally, spatial correlation analysis reveals similar patterns between the two datasets, with smaller values for the summer season owing to the importance of local processes on SSM during this period. However, extending SMOS SSM spatially for surface types that were not included in the training process such as peatlands, remains a challenge warranting additional studies. The developed framework is robust and can address spatio-temporal discontinuities in other SSM products.
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 | 0.000 |
| 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) | 0.000 | 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".