Machine Learning based Classification of Lake ice and Open water from SAR Altimetry waveform parameters
Bibliographic record
Abstract
Lakes cover a significant fraction of the landscape in many northern countries and play a key role in regulating weather and climate. Lakes also have a significant impact on northern communities since the presence (or absence), extent and thickness of lake ice affect transportation (ice roads), food availability, recreational activities, and tourism in wintertime. The recent decline in in-situ observations of lake ice phenology (i.e., freeze-up and break-up dates, and ice cover duration) and lake ice thickness makes remote sensing technology a viable means for monitoring lake ice conditions. Although satellite altimetry has been used in various cryospheric studies, little work has been conducted on lake ice compared to sea ice, for example. This study was conducted at Great Slave Lake, Northwest Territories, Canada, using Sentinel-3A/B SRAL Level 2 data from June 2018 to December 2020. Reflections of radar altimeter echoes differ with properties/conditions of the target and the resulting radar returns contain information about the target surface. Hence, we explored information provided by waveforms to discriminate between open water and lake ice based on machine learning. To characterize the waveforms, five waveform parameters were extracted: Leading Edge Width (LEW), Offset Center of Gravity (OCOG) Width, Pulse Peakiness (PP), backscatter coefficient, and the maximum value of the echo power. Random Forest (RF) and Support Vector Machine (SVM) classifiers were selected to perform along-track classification of open water and lake ice. Class labelling was performed manually via visual interpretation of Sentinel-3 SRAL Level 2 waveforms, Sentinel 2 MultiSpectral Instrument (MSI) Level 1C data, and MODIS Aqua/Terra Level 1B data. Through our proposed method, we reached the highest accuracy of 91.89% (SVM) and 89.58% (RF) during the freeze-up period (November-December). Comparatively, classification performance was lower during the break-up period (late April-early June) reaching an overall accuracy of 77.19% (SVM) and 77.32% (RF). The backscatter coefficient and OCOG Width were found to be the two parameters of most importance for discriminating between ice and open water. Analysis of early results suggests that higher classification accuracies may be achieved by subdividing open water and ice into two more classes to represent leads and melting ice. In addition, pseudo LRM data from Sentinel-3 are currently being analysed and compared to results obtained with SAR data. These new results will also be presented. Keywords: SAR altimetry, lake ice, classification, waveform, machine learning
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) | 0.005 | 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".