CHLNET: A novel hybrid 1D CNN-SVR algorithm for estimating ocean surface chlorophyll-a
Bibliographic record
Abstract
Developing a unified chlorophyll-a (Chla) inversion algorithm for cross-water types is a significant challenge owing to the insufficiency of input features and training samples. Although machine learning algorithms can build a consistent model for different trophic waters, the accuracy of the inversion is dependent on the quality of the extended features. Here, we designed a novel hybrid framework called CHLNET, which combines a one-dimensional convolutional neural network (1D CNN) and support vector regression (SVR). The 1D CNN is used to extract features from the original band features, and the SVR is used to perform a fit of Chla. CHLNET is trained and tested using match-up pairs of SeaWiFS remote sensing reflectance [Rrs(λ)] in situ with Chla ranging from 0.009 mg/m³ to 138.046 mg/m³, which covers mostly ocean water types. Performance metrics in the log space of CHLNET were better than those of the state-of-the-art algorithms on the testing dataset, and CHLNET had the best overall performance with the largest cover area in the star plot. The frequency distribution of predicted Chla by CHLNET was more consistent with that of in situ Chla. While the spatial pattern was not smooth in low Chla concentration waters, CHLNET demonstrated excellent mapping ability at the global and local scales in high Chla concentration waters. Through the band-shift method, which transfers the Rrs(λ) of MERIS and MODIS-Aqua to the Rrs(λ) of SeaWiFS in the visible spectral range, CHLNET obtained better accuracy than the blended algorithm of OCx and CI on MERIS and MODIS-Aqua matchups, which validates the generalization of CHLNET on cross-sensor types. The results indicate that CHLNET avoids the drawbacks of manually constructing extended features and the need for merging water type-appropriate algorithms for Chla retrieval, as well as provides a new idea for unified Chla concentration inversion across water types. Thus, CHLNET may serve as an alternative approach for Chla inversion.
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 | 0.002 | 0.000 |
| 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 | 0.000 | 0.001 |
| Open science | 0.001 | 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".