Estimation of daytime all-sky sea surface temperature from Himawari-8 based on multilayer stacking machine learning
Bibliographic record
Abstract
The Himawari-8 satellite has the capability to rapidly retrieve sea surface temperature (SST) data at a high frequency of 10 min, demonstrating significant potential for various scientific applications. However, the presence of cloud often results in missing SST data at cloud locations, or can reduce the accuracy of retrieved SST. In contrast to the method of reconstructing missing SST data, this study focuses on exploring the potential of inverting SST under all-sky conditions. This study proposes a three-layer stacked machine learning model (TLSM), specifically designed for SST under all-sky conditions. The model integrates cloud properties into its input features to effectively account for the influence of cloud cover. Validation using 30 % match-up pairs generated an overall root mean square error (RMSE) of 0.71 °C, a Bias of −0.01 °C, and an R2 of 0.91 based on 6383 samples. For clear-sky conditions, TLSM demonstrates a noteworthy enhancement in SST inversion accuracy (R2 = 0.98, RMSE=0.35 °C) compared to the official SST product (R2 = 0.86, RMSE=0.88 °C). In cases of optically thin clouds and clouds with low cloud top pressure, TLSM exhibits commendable proficiency in the inversion of SST. The Bias and RMSE for these cloud types indicate better performance compared to the official clear-sky SST data. While including cloud samples may reduce the overall accuracy of the TLSM, it substantially enhances the spatial coverage of the inverted SST. Considering the performance for each cloud type, TLSM may serve as an alternative approach for SST retrieve under thin clouds.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| 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 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".