Consistent Multi-Mission Measures of Inland Water Algal Bloom Spatial Extent Using MERIS, MODIS and OLCI
Bibliographic record
Abstract
Envisat’s MERIS and its successor Sentinel OLCI have proven invaluable for documenting algal bloom conditions in coastal and inland waters. Observations over turbid eutrophic waters, in particular, have benefited from the band at 708 nm, which captures the reflectance peak associated with intense algal blooms and is key to line-height algorithms such as the Maximum Chlorophyll Index (MCI). With the MERIS mission ending in early 2012 and OLCI launched in 2016, however, time-series studies relying on these two sensors have to contend with an observation gap spanning four years. Alternate sensors, such as MODIS Aqua, offering neither the same spectral band configuration nor consistent spatial resolution, present challenges in ensuring continuity in derived bloom products. This study explores a neural network (NN) solution to fill the observation gap between MERIS and OLCI with MODIS Aqua data, delivering consistent algal bloom spatial extent products from 2002 to 2020 using these three sensors. With 14 bands of MODIS level 2 partially atmospherically corrected spectral reflectance as the NN input, the missing MERIS/OLCI band at 708 nm required for the MCI is simulated. The resulting NN-derived MODIS MCI (NNMCI) is shown to be in good agreement with MERIS and OLCI MCI in 2011 and 2017 respectively over the western basin of Lake Erie (R2 = 0.84, RMSE = 0.0032). To overcome the impact of MODIS sensor saturation over bright water targets, which otherwise renders pixels unusable for bloom detection using R-NIR wavebands, a variant NN model is employed which uses the 9 MODIS bands with the lowest probability of saturation to simulate the MCI. This variant NN predicts MCI with only a small increase in uncertainty (R2 = 0.73, RMSE = 0.005) allowing reliable estimates of bloom conditions in those previously unreported pixels. The NNMCI is shown to be robust when applied beyond the initial training dataset on Lake Erie, and when re-trained on different geographic areas (Lake Winnipeg and Lake of the Woods). Despite differences in spatial, temporal, and spectral resolution, MODIS algal bloom presence/absence was correctly classified in >92% of cases and bloom spatial extent derived within 25% uncertainty, allowing the application to the 2012–2015 time period to form a continuous and consistent multi-mission monitoring dataset from 2002 to 2020.
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".